Lapres 5 Senin

LAPORAN RESMI
PRAKTIKUM PROSES KIMIA
Materi :
HIDROLISA PATI
Oleh:
Nama NIM
Gema Adil Guspiani 21030114120040

Sekar Ayu Septianis 21030114120025
Siti Aghnia Salsabilla P 21030114130138
Laboratorium Proses Kimia
Jurusan Teknik Kimia Fakultas Teknik
Universitas Diponegoro
Semarang
2016
LAPORAN RESMI
PRAKTIKUM PROSES KIMIA
Materi :
HIDROLISA PATI
Oleh:
Nama NIM
Gema Adil Guspiani 21030114120040

Sekar Ayu Septianis 21030114120025
Siti Aghnia Salsabilla P 21030114130138
Laboratorium Proses Kimia
Jurusan Teknik Kimia Fakultas Teknik
Universitas Diponegoro
Semarang
2016
i
HIDROLISA PATI
HALAMAN PENGESAHAN
Laporan resmi praktikum proses kimia materi Hidrolisa Pati yang disusun oleh
Kelompok : 5 Senin
Anggota Kelompok : Gema Adil Guspiani NIM: 21030114120040
Sekar Ayu Septianis NIM: 21030114120025
Siti Aghnia Salsabilla P NIM: 21030114130138
Telah diterima dan disetujui oleh Dr. Siswo Sumardiono, ST, MT selaku dosen
pembimbing materi hidrolisa pati
Hari : Jumat
Tanggal : 20 Mei 2016
Semarang , 20 Mei 2016
Dosen Pembimbing Asisten Pembimbing
Dr. Siswo Sumardiono, ST, MT Ilham Dwiyanto Emzar

NIP 197509152000121001 NIM 2103011212008
ii
LABORATORIUM PROSES KIMIA 2016

HIDROLISA PATI
INTISARI
Pati adalah karbohidrat dengan komponen utama amilosa dan amilopektin. Amilosa
tersusunatas satuan glukosa yang saling berikatan dengan ikatan 1,4 -D glukosa sedangkan
amilopektin merupakan polisakarida yang tersusun atas ikatan 1,4 -D glukosa dan rantai
cabang 1,6 -D Untuk meningkatkan nilai ekonomi tapioka, diperlukan modifikasi dengan
cara hidrolisa. Hidrolisa adalah proses dekomposisi kimia dengan menggunakan air untuk
memisahkan ikatan kimia dari substansinya. Hidrolisa pati merupakan proses pemecahan
molekul amilum menjadi bagian-bagian penyusunnya yang lebih sederhana seperti dekstrin,
isomaltosa, maltosa, dan glukosa. Hidrolisa pati terjadi antara suatu reaktan pati dengan
reaktan air. Bahan yang digunakan adalah tapioka dengan variabel jenis katalis HCl 0,1 N
dan H2SO4 0,1 N. Analisa hasil menggunakan titrasi balik dengan titran glukosa standar.
Percobaan diawali dengan menghitung densitas pati dan katalis serta pembuatan glukosa
standar. Selanjutnya dilakukan tahap penentuan kadar pati awal dengan memasukkan tapioka
dan katalis sesuai variabel ke dalam labu leher tiga dan dipanaskan pada 700C selama 1 jam
lalu analisa hasil. Tahap terakhir adalah hidrolisa pati dengan memasukkan tapioka, katalis,
dan aquadest sesuai variabel ke dalam labu leher tiga dan dipanaskan pada 700C dan analisa
hasil setiap 5 menit sebanyak 5 kali.
Pada percobaan variabel 1 dengan katalis HCl 0,1 N menunjukkan konversi pati yang
lebih besar dibandingkan variabel 2 dengan katalis H2SO4 0,1 N. Hal ini disebabkan karena
aktivitas katalitik ion hidrogen lebih tinggi pada HCl dibandingkan dengan H2SO4. variabel
dengan katalis HCl menunjukkan konstanta kecepatan reaksi yang lebih tinggi dibandingkan
variabel 2 dengan katalis H2SO4. Konstanta kecepatan reaksi hidrolisis tapioka dengan katalis
HCl adalah 0,169 menit-1 sedangkan dengan katalis H2SO4 adalah 0,027 menit-1.
Dapat disimpulkan bahwa proses hidrolisa dengan katalis HCl lebih cepat untuk
mengubah polisakarida pati menjadi monosakarida jika dibandingkan dengan katalis H2SO4.
Penelitian lebih lanjut dibutuhkan dengan menggunakan katalis asam lain untuk mengetahui
pengaruh terhadap konversi pati dan konstanta kecepatan reaksi. Selain itu, percobaan
hidrolisa pati dikembangkan dengan menggunakan metode lain untuk mengetahui perbedaan
pati termodifikasi yang dihasilkan.
Kata Kunci: Hidrolisa, Pati, Katalisator
iii

HIDROLISA PATI
SUMMARY
Starch is a carbohydrate with the major components amylose and amylopectin.

Amylose arange by glucose units are linked to each other by bonding 1,4-D glucose while
amylopectin is a polysaccharide composed of bonds 1,4-D glucose and branch chain 1,6-
D glucose. The amylose content of 17%, while tapioca amylopectin by 83%. To increase the
economic value of tapioca, the necessary modifications by means of hydrolysis. Hydrolysis is
a chemical decomposition process using water to separate the chemical bonds of the substance.
Hydrolysis of starch is a starch molecule solving process into its constituent parts becomes
simpler form such as dextrin, isomaltose, maltose and glucose. Starch hydrolysis occurs
between reactant starch with water. Materials were used are tapioca with variable types of
catalysts 0.1 N HCl and H2SO4 0,1N. Analysis of the results were used titration with standard
glucose titrant. Trial began by calculating the density of the catalyst as well as the manufacture
of starch and glucose standard. Furthermore, the initial stage of determined the starch content
by entered the appropriate catalyst tapioca and variables into a three-neck flask and heated at
700C for 1 hour and then analyzed the results. The last step was to include tapioca starch
hydrolysis, catalyst, and distilled water corresponding variables into three neck flask and
heated at 700C and the result was analyzed every 5 minutes on 5 times.
In the experiment with variable 1 with 0.5 N HCl catalyst showed conversion of starch
which was greater than the variable 2 with 0.5 N H2SO4 catalyst. This was because the
catalytic activity of hydrogen ions was higher in comparison with H2SO4 HCl. variable with
HCl catalyst shows the reaction rate constants are higher than variable 2 with H2SO4 catalyst.
Hydrolysis reaction rate constants tapioca with HCl catalyst is 0.169 min-1 while with H2SO4
catalyst was 0,027 min-1
It can be concluded that the process of hydrolysis with HCl catalyst more quickly to
changing starch polysaccharides into monosaccharides compared with H2SO4 catalyst.
Further research is needed using another acid catalyst to determine the effect of the conversion
of starch and reaction rate constants. In addition, the starch hydrolysis experiments were
developed by using other methods to determine differences in modified starches are produced.
Keyword: Hydrolysis, starch, catalyst.
iv

HIDROLISA PATI
PRAKATA
Puji syukur di panjatkan kepada Tuhan Yang Maha Esa berkat karuniaNya
laporan resmi praktikum proses kimia ini dapat diselesaikan dengan lancar sesuai
harapan. Terselesaikannya laporan ini tidak lepas dari bantuan berbagai pihak, oleh
karena itu ucapan terima kasih juga di sampaikan kepada Bapak Dr.Siswo Sumardiono,
ST, MT selaku dosen pengampu materi Hidrolisa pati, Roynaldy Daud selaku
koordinator asisten Laboratorium Proses Kimia , Ilham Dwiyanto dan Ihdina selaku
asisten pembimbing materi Hidrolisa pati, dan seluruh teman-teman yang telah
membantu baik dalam segi waktu maupun motivasi.
Laporan resmi praktikum proses kimia ini berisi materi hidrolisa pati. Hidrolisa
merupakan reaksi pengikatan gugus hidroksil (-OH) oleh suatu senyawa. Gugus OH
dapat diperoleh dari senyawa air. Hidrolisis dapat digolongkan menjadi hidrolisis
murni, hidrolisis katalis asam, hidrolisis katalis basa, hidrolisis gabungan alkali dengan
air dan hidrolisis dengan katalis enzim. Sedangkan berdaasarkan fase reaksi yang
terjadi diklasifikasikan menjadi hidrolisis fase cair dan hidrolisis fase uap.Hidrolisis
pati terjadi antara suatu reaktan pati dengan reaktan air. Reaksi ini adalah orde satu,
karena reaktan air yang dibuat berlebih, sehingga perubahan reaktan dapat. diabaikan.
Laporan resmi ini telah disusun sebaik mungkin, namun disadari masih banyak
kekurangan dalam penyusunan laporan resmi ini. Oleh karena itu, kritik dan saran dari
berbagai pihak diperlukan untuk laporan yang lebih baik. Akhir kata, semoga laporan
resmi ini dapat bermanfaat bagi semua pihak yang membutuhkan.
Semarang, Mei 2016
Penyusun

HIDROLISA PATI
DAFTAR ISI
HALAMAN JUDUL............................................................................................. i
HALAMAN PENGESAHAN ............................................................................... ii
RINGKASAN ....................................................................................................... iii
SUMMARY .......................................................................................................... iv
PRAKATA ............................................................................................................ v
DAFTAR ISI ......................................................................................................... vi
DAFTAR GAMBAR ............................................................................................ viii
DAFTAR TABEL ................................................................................................. ix
BAB I PENDAHULUAN ..................................................................................... 1
1.1 Latar Belakang ...................................................................................... 1
1.2 Rumusan Masalah ................................................................................. 1
1.3 Tujuan Percobaan .................................................................................. 2
1.4 Manfaat Percobaan ................................................................................ 2
BAB II LANDASAN TEORI ............................................................................... 3
2.1 Pengertian Pati ....................................................................................... 3
2.2 Hidrolisa Pati ........................................................................................ 3
2.3 Modifikasi Pati....................................................................................... 4
2.4 Variabel-variabel yang Berpengaruh ..................................................... 5
BAB III PELAKSANAAN PERCOBAAN .......................................................... 8
3.1 Rancangan Praktikum ............................................................................ 8
3.1.1 Skema Rancangan Praktikum ...................................................... 8
3.1.2 Variabel Operasi .......................................................................... 8
3.2 Gambar Alat........................................................................................... 8
3.3 Bahan dan Alat yang Digunakan ........................................................... 9
vi

HIDROLISA PATI
3.3.1 Bahan ........................................................................................... 9

3.3.2 Alat .............................................................................................. 9
3.4 Prosedur Percobaan .............................................................................. 9
BAB IV HASIL PERCOBAAN DAN PEMBAHASAN ..................................... 13
4.1 Pengaruh Jenis Katalis Terhadap Konversi (XA) Hidrolisa Pati .......... 13
4.2 Pengaruh Jenis Katalis Terhadap Konstanta Kecepatan Reaksi (k) Hidrolisa
Pati ............................................................................................................. 14
BAB V PENUTUP ................................................................................................ 16

5.1 Kesimpulan ............................................................................................ 16
5.2 Saran ...................................................................................................... 16
DAFTAR PUSTAKA ........................................................................................... 17
LAMPIRAN
- Data Percobaan (Laporan Sementara)
- Lembar Perhitungan
- Referensi
LEMBAR ASISTENSI
vii

HIDROLISA PATI
DAFTAR GAMBAR
Gambar 3.1 Skema Rancangan Percobaan............................................................ 8

Gambar 3.2 Rangkaian alat hidrolisis .................................................................. 9
Gambar 4.1 Pengaruh jenis katalis terhadap konversi hidrolisa pati .................... 14
Gambar 4.2 Pengaruh jenis katalis terhadap konstanta kecepatan reaksi hidrolisa
pati ..................................................................................................... 15
viii

HIDROLISA PATI
DAFTAR TABEL
Tabel 4.1 Hasil data hidrolisa pati katalis HCl 0,1 N ........................................... 13
Tabel 4.2 Hasil data hidrolisa pati katalis H2SO4 0,1 N........................................ 13
ix

HIDROLISA PATI 1
BAB I
PENDAHULUAN
1.1 Latar Belakang

Pati dan juga produk turunannya merupakan bahan yang multiguna dan
banyak digunakan pada berbagai industri antara lain pada minuman, makanan
yang diproses, kertas, makanan ternak, farmasi dan bahan kimia serta industri
nonpangan seperti tekstil, detergent, kemasan dan sebagainya. Dalam industri
makanan pembentuk gel dan encapsulating agent. Dalam industri kertas
digunakan sebagai zat aadtive seperti wet-end untuk surface size dan coating
binder, bahan perekat, dan glass fiber sizing (Chiu & Solarek, 2009).
Berbagai varian pati didasarkan pada perbedaan struktural, kandungan
amilosa, amilopketin, protein dan lipid. Secara umum kandungan pati yang
utama yaitu polimer anhidroglukosa meliputi amilosa dan amilopketin,
keduanya diikat dengan ikatan (1,4) dalam segmen linear; serta ikatan (1,6)
di titik percabangan. Amilopektin merupakan kandungan utama pati, berkisar
70-80% dan berpengaruh pada physiochemical serta cita rasa pati (Dona., et all.
2010)
Pada reaksi hidrolisa biasanya dilakukan dengan menggunakan
katalisator asam seperti HCl (asam klorida). Bahan yang digunakan untuk proses
hidrolisis adalah pati. Di indonesia banyak dijumpai tanaman yang
menghasilkan pati. Pati yang dihasilkan akan meningkat seiiring dengan
konsentrasi asam serta waktu tinggal yang meningkat . karena selama proses
berlangsung tidak ada gula yang terdegradasi (Jing Bian.,et all. 2014). Tanaman-
tanaman itu seperti seperti padi, jagung, ketela pohon, umbi-umbian, aren dan
sebagainya.
1.2 Rumusan Masalah

1. Bagaimana pengaruh perbedaan katalis terhadap reaksi hidrolisa pati?
2. Bagaimana menghitung konstanta kecepatan reaksi dan menganalisa
pengaruh perbedaan katalis terhadap konstanta kecepatan reaksi?

HIDROLISA PATI 2
1.3 Tujuan Percobaan

1. Mempelajari pengaruh perbedaan katalis terhadap reaksi hidrolisa pati
2. Menghitung konstanta kecepatan reaksi dan menganalisa pengaruh
perbedaan katalis terhadap konstanta kecepatan reaksi
1.4 Manfaat Percobaan

1. Mahasiswa dapat mengetahui pengaruh perbedaan katalis terhadap reaksi
hidrolisa pati
2. Mahasiswa dapat menghitung konstanta kecepatan reaksi dan
menganalisa pengaruh perbedaan katalis terhadap konstanta kecepatan
reaksi

HIDROLISA PATI 3
BAB II
TINJAUAN PUSTAKA
2.1 Pengertian Pati

Pati secara luas digunakan dalam industri makanan sebagai bahan utama
untuk penyusunan dan pengolahan produk. Pati termasuk dalam polisakarida
yang merupakan polimer glukosa, yang terdiri atas amilosa dan amilopektin.
Komponen dalam pati menentukan struktur mikronya sertarasio keberadaan
amilosa dan amilopektin (C. Hernandez., et all. 2014). Amilosa merupakan
bagian polimer linier dengan ikatan -(1,4) unit glukosa yang meruapakan rantai
linear. Derajat polimerisasi (DP) amilosa berkisar antara 5006.000 unit
glukosa,bergantung pada sumbernya. Adapun amilopektin merupakan polimer
-(1,4) unit glukosa dengan rantai samping -(1,6) unit glukosa. Ikatan -(1,6)
unit glukosa ini jumlahnya sangat sedikit dalam suatu molekul pati, berkisar
antara 45%. Namun, jumlah molekul dengan rantai cabang, yaitu amilopektin,
sangat banyak dengan DP berkisar antara 105 dan 3x106 unit glukosa dan
merupakan komponen utama yang dapat mempengaruhi physiochemical dan cita
rasa dari pati. Amilosa dapat terdegradasi dengan ezim amilase yang merupakan
enzim utama yang terlibat dalam reaksi hidrolisis. Amilase sangat penting dalam
reaksi biologis seperti fermentasi, perkecambahan atau pencernaan namun
banyak juga yang digunakan dalam industri (Georges Tawil., et all. 2012).
2.2 Hidrolisa Pati

Pemecahan struktur pati merupakan proses yang kompleks yang
melibatkan katalis maupun enzim yang digunakan. Kondisi operasi saat
melakukan pemecahan pati juga harus sesuai (Anthony, 2010). Hidrolisa
merupakan reaksi pengikatan gugus hidroksil (-OH) oleh suatu senyawa. Gugus
OH dapat diperoleh dari senyawa air. Hidrolisis dapat digolongkan menjadi
hidrolisis murni, hidrolisis katalis asam, hidrolisis katalis basa, hidrolisis
gabungan alkali dengan air dan hidrolisis dengan katalis enzim. Sedangkan
berdaasarkan fase reaksi yang terjadi diklasifikasikan menjadi hidrolisis fase cair

HIDROLISA PATI 4
dan hidrolisis fase uap.Hidrolisis pati terjadi antara suatu reaktan pati dengan
reaktan air. Reaksi ini adalah orde satu, karena reaktan air yang dibuat berlebih,
sehingga perubahan reaktan dapat. diabaikan. Reaksi hidrolisis pati dapat
dilakukan menggunakan katalisator H+ yang dapat diambil dari asam. Reaksi
yang terjadi pada hidrolisis pati adalah sebagai berikut :
(C6H10O5)x + H2O x C6H12O
Berdasarkan teori kecepatan reaksi :
- rA = k . Cpati . Cair (1)
Karena volume air cukup besar, maka dapat dianggap konsentrasi air selama
perubahan reaksi sama dengan k, dengan besarnya k :
k = k . Cair (2)
sehingga persamaan 1 dapat ditulis sebagai berikut
-rA = k. C
Pati dari persamaan kecepatan reaksi ini, reaksi hidroisis merupakan reaksi orde
satu. Jika harga rA = dCA/dt maka persamaan 2 menjadi :

= (3)

= (4)

Apabila CA = CA0 (1-XA) dan diselesaikan dengan integral dan batas kondisi t1,
CA0 dan t2 : CA akan diperoleh persamaan :
1
0 = 2 (5)

0
= (2 1) (6)

1
(1) = (2 1) (7)
Dimana xA = konversi reaksi setelah satu detik. Persamaan 7 dapat

diselesaikan dengan menggunakan pendekatan regresi y = mx + c, dengan dan
1
= (1) dan x = t2.
2.3 Modifikasi Pati

Pati asli pada umumnya memiliki struktur granular, tidak larut air, dan
dalam bentuk ini digunakan hanya dalam beberapa aplikasi spesifik yang

HIDROLISA PATI 5
terbatas. Modifikasi adalah pati yang gugus hidroksinya telah mengalami

perubahan. Pati memiliki sifat tidak dapat digunakan secara langsung dan oleh
karena itu harus dimodifikasi secara kimia atau fisik untuk meningkatkan sifat
positif dan mengurangi sifat yang tidak diinginkan. Pati biasanya digunakn
untuk produk makanan, bahan perekat dan glass fiber sizing. Selain itu juga
ditambahkan dalam plastik unutk mempercepat proses degradasi. Modifikasi
secra kimia umunya meliputi esterifikasi, etherifikasi, hidrolisis, oksidasi dan
cross-linking (Chiu & Solarek, 2009). Pati yang telah termodifikasiakan
mengalami perubahan sifat yang dapat disesuaikan untuk keperluan-keperluan
tertentu. Akan tetapi sama seperti pati alami, pati termodifikasi bersifat tidak
larut dalam air dingin (Koswara, 2009).
2.4 Variabel yang Berpengaruh

Variabel - variabel yang berpengaruh dalam reaksi hidrolisa pati meliputi :
1. Katalisator
Hampir sama semua reaksi hidrolisa membutuhkan katalisator untuk
mempercepat jalannya reaksi. Katalisator yang dipakai dapat berupa enzim
atau asam karena kinerjanya lebih cepat. Asam yang dipakai beraneka
jenisnya mulai dari HCl, H2SO4 sampai HNO3. Yang mempengaruhi
kecapatan reaksi adalah konsentrasi ion H+, bukan jenis asamnya. Jika
dibandingkan antara reaksi hidrolisis dengan katalis H2SO4 atau dengan
HCl hasil yang didapatkan lebih baik dengan menggunakan H2SO4 karena
memiliki zeta potensial yang lebih tinggi dan memiliki kerapatan yang
kecil sehingga mudah berdistribusi (Benxi Wei., et all. 2014). Meskipun
demikian, didalam industri umumnya diakai asam klorida (HCl).
Pemilihan ini didasarkan atas sifat garam yang terbentuk pada penetralan
tidak menimbulkan gangguan apa-apa selain rasa asin jika konsentrasinya
tinggi. Oleh karena itu, konsentrasi asam dalam air penghidrolisa ditekan
sekecil mungkin. Umumnya dipergunakan larutan asam yang mempunya
konsentrasi asam yang lebih tinggi daripada pembuatan sirup. Hidrolisa
pada tekanan 1 atm memerlukan asam yang jauh lebih pekat. Secara

HIDROLISA PATI 6
keseluruhan, hidrolisa pati dengan katalis asam merupakan proses yang

kompleks yang mendasari perubahan mikrostruktur pati dalam jangka
waktu jam hingga hari (Utrilla Coello.,et all. 2014)
2. Suhu, tekanan, dan pH
Pengaruh suhu terhadap kecepatan reaksi mengikuti persamaan
Arrhenius, dimana semakin tinggi suhu maka semakin cepat laju
reaksinya. Untuk mencapai konversi tertentu, diperlukan waktu sekitar 3
jam untuk menghidrolisa pati ketela rambat pada suhu 100 C. Tetapi jika
suhunya dinaikkan hingga 135 C, konversi yang sama dapat dicapai
dalam waktu 40 menit (Ayodeji Ayoola.., et all, 2013). Hidrolisis pati
gandum dan jagung dengan katalisator H2SO4 memerlukan suhu 160 C.
Karena panas reaksi mendekati nol dan reaksi berjalan dalam fase cair
maka suhu dan tekanan tidak banyak mempengaruhi keseimbangan.
Hidrolisis dapat dilakukan pada pH 4 untuk suhu 105C dan pH
5 untuk suhu 150C, batas maksimum pH adalah 6,5 dan batas suhu
yaitu 700C (Erik Ollson., et all,2013). Untuk menghindari cross-linking
pada proses hidrolisa pati, dapat dilakukan dengan memilih pH dan suhu
reaksi yang cocok (Caroline Menzel., et all, 2013).
3. Pencampuran (pengadukan)
Supaya zat pereaksi dapat saling bertumbukan dengan sebaik-
baiknya perlu adanya pencampuran. Untuk proses Batch, hal ini dapat
dicapai dengan bantuan pengaduk atau alat pengocok. Apabila prosesnya
berupa proses alir (kontinyu), maka pecampuran dilakukan dengan cara
mengatur aliran didalam reaktor supaya terbentuk olakan.
4. Perbandingan zat pereaksi
Jika salah satu zat pereaksi dibuat berlebihan jumlahnya maka
keseimbanga n dapat bergeser kearah kanan dengan baik. Oleh karena itu,
suspensi pati yang kadarnya rendah memberi hasil yang lebih baik
dibandingkan dengan yang kadarnya tinggi. Bila kadar suspensi pati
diturunkan dari 40% menjadi 20% atau 1% maka konversi akan bertambah
dari 80% menjadi 87 atau 99 % . Pada permukaan, kadar suspensi pati

HIDROLISA PATI 7
yang tinggi sehingga molekul-molekul zat pereaksi akan sulit bergerak.

Untuk menghasilkan glukosa biasanya dipergunakan suspensi pati sekitar
20%

HIDROLISA PATI 8
BAB III
METODOLOGI PERCOBAAN
3.1 Rancangan Praktikum

3.1.1 Skema Rancangan Praktikum
Gambar 3.1 Skema Rancangan Praktikum

3.1.2 Variabel Operasi
a. Variabel berubah
- HCl 0,1 N
- H2SO4 0,1
b. Variabel tetap
- Jenis tepung : Tepung Tapioka
- Suhu operasi : 650
- Volume total : 450 ml
- Perbandingan volume pati dan volume aquadest 1:16
3.2. Gambar Alat Utama

Keterangan:
1. Magnetic stirer + heater
2. Waterbath
3. Labu leher tiga
4. Termometer

HIDROLISA PATI 9
5. Pendingin balik
6. Klem
7. Statif
Gambar 3.2 Rangkaian alat hidrolisis
3.3 Bahan dan Alat yang digunakan

3.3.1 Bahan
1. Glukosa anhidrit 6. Indikator MB
2. Tepung Tapioka 7. Fehling A
3. NaOH 8. Fehling B
4. HCl 9. Aquadest
5. H2SO4
3.3.2 Alat
1. Gelas ukur 5. Buret
2. Termometer 6. Labu leher tiga
3. Erlenmeyer 7. Labu takar
4. Statif dan klem
3.4 Prosedur percobaan

1. Persiapan awal
a. Menghitung densitas pati
Ke dalam gelas ukur, 5 ml aquades dimasukkan 1 gram pati, catat
penambahan volume.
b. Menghitung densitas HCl
Timbang berat picnometer kosong (m1), masukkan HCl ke dalam
picnometer yang telah diketahui volumenya (v), timbang beratnya
(m2), hitung densitas HCl

HIDROLISA PATI 10

=

1 2
/24 =

c. Membuat glukosa standar
Glukosa anhidrit sebanyak 2 gram dilarutkan dalam 1000 ml aquadest
2. Penentuan kadar pati
a. Standarisasi larutan fehling
5 ml Fehling A + 5 ml Fehling B + 15 ml glukosa standar, dipanaskan
sampai mendidih. Setelah mendidih ditambahkan 3 tetes MB,
kemudian larutan dititrasi dengan glukosa standard hingga warna
berubah menjadi merah bata. Catat volume titran (F) yang diperlukan,
proses titrasi dilakukan dalam keadaan mendidih (diatas kompor)
b. Penentuan kadar pati awal
Untuk variabel 1, sebanyak 32,66 gram pati, 5,81 ml katalis HCl dan
418,08 ml aquadest dimasukkan ke dalam labu leher tiga dan
dipanaskan hingga suhu 90oC, selama 1 jam. Setelah itu larutan
didinginkan, diencerkan dengan aquades sampai 500 ml lalu diambil 20
ml dan dinetralkan dengan NaOH (PH = 7). Larutan diambil 5 ml
diencerkan sampai 100 ml, diambil 5 ml. Ke dalam Erlenmeyer
dimasukkan 5 ml larutan + 5 ml Fehling A + 5 ml fehling B + 15 ml
glukosa standard, kemudian dipanaskan sampai mendidih. Lalu
ditambahkan 3 tetes indikator MB. Kemudian larutan dititrasi dengan
glukosa standard sehingga berubah warna menjadi warna merah bata.
Catat volum titran yang dibutuhkan (M). Yang perlu diperhatikan,
proses titrasi dilakukan dalam keadaan mendidih diatas kompor.
Lakukan hal yang sama untuk variabel lain
3. Hidrolisa pati
Sebanyak 32,66 gram pati, 5,81 ml katalis HCl dan 418,08 ml
aquadest dimasukkan ke dalam labu leher tiga dan dipanaskan hingga suhu
90oC, anggap sebagai t0 diambil sampel sebanyak 20 ml. Kemudian sampel
dinetralkan dengan NaOH (PH = 7). Larutan diambil 5 ml diencerkan

HIDROLISA PATI 11
sampai 100 ml, diambil 5 ml. Kedalam Erlenmeyer dimasukkan, kemudian

dipanaskan sampai mendidih. Lalu ditambahkan3 tetes indikator Mkan 5 ml
larutan +5 ml Fehling A + 5 ml fehling B + 15 ml glukosa
standard.Kemudian larutan dititrasi dengan glukosa standard sehingga
berubah warna menjadi warna merah bata. Catat V titran yang dibutuhkan
(M). Yang perlu diperhatikan, proses titrasi dilakukan dalam keadaan
mendidih diatas kompor. Pengambilan sampel dilakukan setiap selang
waktu 5 menit sebanyak 5 kali yaitu 20 menit. (t0=menit ke-0 ,t1=menit ke-
5, t2=menit ke-10, t3=menit ke-15, t4=menit ke-20). Lakukan hal yang sama
untuk variabel 2
Rumus penentuan kadar pati awal =

500 100
( ) 0,9
0 = 5

Dimana
N = 0,002 gr/ml
W = berat pati
4. Perhitungan kebutuhan reagen
a. Menghitung kebutuhan HCl

=
1000
Dimana :
kadar HCl = 0,25 untuk 25% dan 0,37 untuk 37%
grek HCl =1
b. Menghitung kebutuhan pati

% =
+ +
=
= ( )
=

HIDROLISA PATI 12
5. Prosedur titrasi
a. 5 ml fehling A + 5 ml fehling B + 5 ml glukosa standar (jika ada
hasil hidrolisa, prosedur diatas ditambah 5 ml sampel hasil hidrolisa)
b. Dipanaskan sampai mendidih
c. 100 detik dari mendidih ditambah 3 tetes indikator MB
d. 2 menit kemudian dititrasi dengan glukosa standar, catat volume
titran (titrasi maksimal dijalankan 1 menit )
Catatan : titrasi dilakukan diatas kompor dalam keadaan mendidih

HIDROLISA PATI 13
BAB IV
HASIL PERCOBAAN DAN PEMBAHASAN
4.1 Pengaruh Jenis Katalis Terhadap Konversi (XA) Hidrolisa Pati

Dari hasil percobaan hidrolisa pati pada berbagai variabel jenis katalis dan
waktu pengambilan sampel pada tabel 4.1 dan tabel 4.2, dapat diketahui
pengaruh jenis katalis terhadap konversi hidrolisa pati yang disajikan pada
gambar 4.1.
Tabel 4.1 Hidrolisa Pati katalis HCl 0,1 N

t (menit) XA -ln(1-XA)
0 0,7706 1,4722
1 0,7711 1,4744
2 0,7796 1,5123
3 0,7881 1,5516
4 0,7923 1,5716
k = 0,169 menit-1
Tabel 4.2 Hidrolisa Pati katalis H2SO4 0,1 N
t (menit) XA -ln(1-XA)
0 0,5390 0,7743
1 0,7012 1,2079
2 0,7676 1,4590
3 0,7676 1,4590
4 0,7759 1,4950
k = 0,027 menit-1

HIDROLISA PATI 14
0,9
0,8
0,7
0,6
0,5
Xa
0,4 HCl
0,3
H2SO4
0,2
0,1
0
0 1 2 3 4 5
t (menit)
Gambar 4.1 Pengaruh Jenis Katalis Terhadap Konversi Hidrolisa Pati

Dengan menjalankan reaksi hidrolisis pada konsentrasi air yang
berlebihan maka reaksi dapat dibawa ke reaksi orde satu. Gambar 4.1 dapat
terlihat bahwa konversi akan meningkat jika waktu semakin lama. Hal ini
disebabkan jika waktu bertambah maka kesempatan bertumbukan antara zat-zat
yang bereaksi akan semakin besar.
Berdasarkan Gambar 4.1, variabel 1 dengan katalis HCl 0,1 N
menunjukkan konversi pati yang lebih besar dibandingkan variabel 2 dengan
katalis H2SO4 0,1 N. Hal ini disebabkan karena aktivitas katalitik ion hidrogen
lebih tinggi pada HCl dibandingkan dengan H2SO4. Selain itu, H2SO4 dapat
membentuk garam-garam dengan beberapa senyawa non karbohidrat yang dapat
menghambat proses hidrolisa. Sedangkan garam-garam yang terbentuk oleh HCl
tidak menghambat proses hidrolisa. Oleh karena itu, konversi pati menggunakan
katalis HCl lebih besar dibandingkan dengan katalis H2SO4.
4.2 Pengaruh Jenis Katalis Terhadap Konstanta Kecepatan Reaksi (k)

Hidrolisa Pati
Pada percobaan hidrolisa pati yang dilakukan pada berbagai variabel
jenis katalis dan waktu pengambilan sampel yang disajikan gambar 4.2, dapat
diketahui pengaruh jenis katalis terhadap nilai konstanta kecepatan reaksi pada
proses hidrolisa pati.

HIDROLISA PATI 15
1,5
-ln (1-Xa)
1
HCl
0,5 H2SO4
0
0 1 2 3 4 5
t
Gambar 4.2 Pengaruh Jenis Katalis Terhadap Konstanta Kecepatan Reaksi

Hidrolisa Pati
Berdasarkan Gambar 4.2, variabel dengan katalis HCl menunjukkan
konstanta kecepatan reaksi yang lebih tinggi dibandingkan variabel 2 dengan
katalis H2SO4. Konstanta kecepatan reaksi hidrolisis tapioka dengan katalis HCl
adalah 0,169 menit-1 sedangkan dengan katalis H2SO4 adalah 0,027 menit-1. Pada
nilai konsentrasi asam yang sama, untuk katalis asam sulfat menghasilkan berat
glukosa yang lebih kecil. Dalam hal ini, yang mempengaruhi kecepatan reaksi
adalah jumlah H+ pada H2SO4 jumlah H+ lebih banyak daripada HCl, H+ dari
asam akan berikatan dengan OH- dari air membentuk gula reduksi. Untuk OH-
yang sama pada H2SO4 terdapat sisa H+ yang tidak bereaksi sehingga
menyebabkan glukosa yang dihasilkan pada H2SO4 lebih sedikit. (Enny K, dkk.
2012)
Untuk menurunkan energi aktivasi (menurunkan suhu reaksi) dan
mempercepat jalannya reaksi hidrolisis pati dibutuhkan suatu katalis. Secara
mikro, mekanisme kerja katalis dapat dijelaskan sebagai terjadinya tumbukan
antar elektron yang mengakibatkan adanya perubahan konfigurasi elektron
sehingga didapat unsur baru yang pada akhirnya menghasilkan zat (senyawa)
baru.

HIDROLISA PATI 16
BAB V
PENUTUP
5.1 Kesimpulan
Reaksi hidrolisis pati dalam percobaan ini mengikuti reaksi orde satu. Hal
tersebut dapat dilihat dari grafik yang berupa garis lurus. Hidrolisis pati dengan
HCl 0,1 N menghasilkan konversi maksimum 0,7923, lebih besar
dibandingkan dengan katalis H2SO4 0,1 N dengan konversi maksimum 0,7759.
Hal tersebut dikarenakan aktivitas katalitik ion hidrogen lebih tinggi pada HCl
dibandingkan dengan H2SO4. Konstanta kecepatan reaksi hidrolisa pati dengan
katalis HCl 0,1 N sebesar 0,169 menit-1, lebih besar dibandingkan dengan
katalis H2SO4 0,1 N sebesar 0,027 menit-1. Dapat disimpulkan bahwa proses
hidrolisa dengan katalis H2SO4 lebih lambat untuk mengubah polisakarida pati
menjadi monosakarida jika dibandingkan dengan katalis HCl.
5.2 Saran
Penelitian lebih lanjut dibutuhkan dengan menggunakan katalis asam
lain untuk mengetahui pengaruh terhadap konversi pati dan konstanta
kecepatan reaksi. Selain itu, percobaan hidrolisa pati dikembangkan dengan
menggunakan metode lain untuk mengetahui perbedaan pati termodifikasi
yang dihasilkan.

HIDROLISA PATI 17
DAFTAR PUSTAKA
Ayoola, Ayodeji., et al. 2013. Optimum Hydrolysis conditions of cassava starch

for glucose production. International Journal of Advanced Research in IT
and Engineering. Nigeria
Bian, Jing., et al. 2014. Microwave-assisted acid hydrolysis to produce
xylooligosaccharides from sugarcane bagasse hemicelluloses. Beijing Key
Laboratory of Lignocellulosic Chemistry, College of Materials Science
and Technology, Beijing Forestry University, Beijing. China.
Chiu, C.-w., & Solarek, D. 2009. Modification of starch. Starch: Chemistry and
Technology, Third Edition ISBN: 978-0-12-746275-2.
Coello, Utrilla., et al. 2014. Acid hydrolysis of native corn starch: Morphology,
crystallinity,rheological and thermal properties.Universidad Autnoma
Metropolitana-Iztapalapa, Departamento de Ingeniera de Procesos e
Hidrulica, Apartado Postal. Mexico
Dona, A. C., Pages, G., & Kuchel, P. W. 2010. Digestion of starch:In vivo andin
vitro kinetic models used to characterise. Carbohydrate Polymers 80 (2010)
599617
Dona, Anthony C., et all. 2010. Digestion of starch: In vivo and in vitro kinetic
models used to characterise oligosaccharide or glucose release. School of
Molecular & Microbial Biosciences, University of Sydney, Sydney.
Australia.
Kriswiyanti, Enny., et all. 2012. Pengaruh jenis dan konsentrasi asam terhadap
kinetika reaksi hidrolisis pelepah pisang. 11 (2), 73-77.
Jaimes, C Hernandez., et all. 2014. Acid hydrolysis of native corn starch:
Morphology, crystallinity,rheological and thermal properties. Universidad
Autnoma Metropolitana-Iztapalapa, Departamento de Ingeniera de
Procesos e Hidrulica, Apartado Postal. Mexico
Koswara, S. 2009. Teknologi Modifikasi Pati. ebookpangan.com.
Levenspiel. O., Chemical Reaction Engineering 2nd ed, Mc. Graw Hill Book
Kogakusha Ltd, Tokyo, 1970

HIDROLISA PATI 18
Menzel, Caroline., et all. 2013. The effect of pH on hydrolysis, cross-linking and

barrier properties ofstarch barriers containing citric acid. Department of
Engineering and Chemical Sciences, Karlstad University. Sweden.
Ollson, Erik., et all. 2013. The effect of pH on hydrolysis, cross-linking and barrier
properties ofstarch barriers containing citric acid. Department of
Engineering and Chemical Sciences, Karlstad University. Sweden.
Tawil, George., et all. 2012. Hydrolysis of concentrated raw starch: A new very
efficient amylase from Anoxybacillus flavothermus. Unit Biopolymres,
Interactions, Assemblages, Institut National de la Recherche Agronomique,
Rue de la Graudire. France.
Wei, Benzi., et al. 2013. Effect on pHs on Dispersity of Maize Starch Nanocrystals
in Aqueous Medium. The State Key Laboratory of Food Sci ence and
Technology. China.

LAPORAN SEMENTARA
PRAKTIKUM PROSES KIMA
Materi:
Hidrolisa Pati
Group : 5 / Senin
Nama : Gema Adil Guspiani NIM : 21030114120040
Sekar Ayu Septianis NIM : 21030114120025
Siti Aghnia Salsabilla P NIM : 21030114130138
LABORATORIUM PROSES KIMIA

TEKNIK KIMIA FAKULTAS TEKNIK
UNIVERSITAS DIPONEGORO
SEMARANG
I. TUJUAN PERCOBAAN
1. Mempelajari pengaruh perbedaan katalis terhadap reaksi hidrolisa pati
2. Menghitung konstanta kecepatan reaksi dan menganalisa pengaruh
perbedaan katalis terhadap konstanta kecepatan reaksi
II. PERCOBAAN
2.1 Bahan
1. Glukosa anhidrit
2. Tepung Tapioka
3. NaOH
4. HCl
5. H2SO4
6. Indikator MB
7. Fehling A
8. Fehling B Gambar Rangkaian Alat Hidrolisa Pati
9. Aquadest
2.2 Alat
1. Gelas ukur 5. Buret
2. Termometer 6. Labu leher tiga
3. Erlenmeyer 7. Labu takar
4. Statif dan klem
2.3 Cara Kerja
1. Persiapan awal
a. Menghitung densitas pati
Ke dalam gelas ukur, 5 ml aquades dimasukkan 1 gram pati, catat
penambahan volume.
b. Menghitung densitas HCl
Timbang berat picnometer kosong (m1), masukkan HCl ke dalam
picnometer yang telah diketahui volumenya (v), timbang beratnya
(m2), hitung densitas HCl

=

1 2
/24 =

c. Membuat glukosa standar
Glukosa anhidrit sebanyak 2 gram dilarutkan dalam 1000 ml aquadest
2. Penentuan kadar pati
a. Standarisasi larutan fehling
5 ml Fehling A + 5 ml Fehling B + 15 ml glukosa standar, dipanaskan
sampai mendidih. Setelah mendidih ditambahkan 3 tetes MB,
kemudian larutan dititrasi dengan glukosa standard hingga warna
berubah menjadi merah bata. Catat volume titran (F) yang diperlukan,
proses titrasi dilakukan dalam keadaan mendidih (diatas kompor)
b. Penentuan kadar pati awal
Untuk variabel 1, sebanyak 32,66 gram pati, 5,81 ml katalis HCl dan
418,08 ml aquadest dimasukkan ke dalam labu leher tiga dan
dipanaskan hingga suhu 90oC, selama 1 jam. Setelah itu larutan
didinginkan, diencerkan dengan aquades sampai 500 ml lalu diambil 20
ml dan dinetralkan dengan NaOH (PH = 7). Larutan diambil 5 ml
diencerkan sampai 100 ml, diambil 5 ml. Ke dalam Erlenmeyer
dimasukkan 5 ml larutan + 5 ml Fehling A + 5 ml fehling B + 15 ml
glukosa standard, kemudian dipanaskan sampai mendidih. Lalu
ditambahkan 3 tetes indikator MB. Kemudian larutan dititrasi dengan
glukosa standard sehingga berubah warna menjadi warna merah bata.
Catat volum titran yang dibutuhkan (M). Yang perlu diperhatikan,
proses titrasi dilakukan dalam keadaan mendidih diatas kompor.
Lakukan hal yang sama untuk variabel lain
3. Hidrolisa pati
Sebanyak 32,66 gram pati, 5,81 ml katalis HCl dan 418,08 ml aquadest
dimasukkan ke dalam labu leher tiga dan dipanaskan hingga suhu 90oC, anggap
sebagai t0 diambil sampel sebanyak 20 ml. Kemudian sampel dinetralkan
dengan NaOH (PH = 7). Larutan diambil 5 ml diencerkan sampai 100 ml,
diambil 5 ml. Kedalam Erlenmeyer dimasukkan, kemudian dipanaskan sampai
mendidih. Lalu ditambahkan3 tetes indikator Mkan 5 ml larutan +5 ml Fehling
A + 5 ml fehling B + 15 ml glukosa standard.Kemudian larutan dititrasi dengan
glukosa standard sehingga berubah warna menjadi warna merah bata. Catat V
titran yang dibutuhkan (M). Yang perlu diperhatikan, proses titrasi dilakukan
dalam keadaan mendidih diatas kompor. Pengambilan sampel dilakukan setiap
selang waktu 5 menit sebanyak 5 kali yaitu 20 menit. (t0=menit ke-0 ,t1=menit
ke-5, t2=menit ke-10, t3=menit ke-15, t4=menit ke-20). Lakukan hal yang sama
untuk variabel 2
Rumus penentuan kadar pati awal =

500 100
( ) 0,9
0 = 5

Dimana
N = 0,002 gr/ml
W = berat pati
4. Perhitungan kebutuhan reagen
a. Menghitung kebutuhan HCl

=
1000
Dimana :
kadar HCl = 0,25 untuk 25% dan 0,37 untuk 37%
grek HCl =1
b. Menghitung kebutuhan pati

% =
+ +
=
= ( )
=
5. Prosedur titrasi
a. 5 ml fehling A + 5 ml fehling B + 5 ml glukosa standar (jika ada hasil
hidrolisa, prosedur diatas ditambah 5 ml sampel hasil hidrolisa)
b. Dipanaskan sampai mendidih
c. 100 detik dari mendidih ditambah 3 tetes indikator MB
d. 2 menit kemudian dititrasi dengan glukosa standar, catat volume titran
(titrasi maksimal dijalankan 1 menit )
Catatan : titrasi dilakukan diatas kompor dalam keadaan mendidih 2.4 Hasil
Percobaan
2.4 Hasil Percobaan
A. Perhitungan Reagen
a. Kebutuhan katalis
- HCl
Densitas HCl
Massa piknometer kosong = 16,47 gram
Massa piknometer isi HCl = 44,72 gram
Massa HCl = 44,72 16,47 = 28,25 gram
Volume piknometer = 25 ml

=

28,25
= = 1,13 /
25
Kebutuhan HCl
. . 1000. .
=
.
1,13. . 1000.1.0,25
0,1 =
36,5 . 450
= 5,81
- H2SO4
Densitas HCl
Massa piknometer isi H2SO4 = 53,95
Massa H2SO4 = 53,95 16,47 = 37,48 gram

=

28,25
H2SO4 = = 1,56/
25
Kebutuhan H2SO4
. . 1000. .
H2SO4 =
.
1,56. . 1000.2.0,0,97
0,1 =
98 .450
H2SO4 = 1,46
b. Kebutuhan Massa Pati dan Volume Air
- Densitas Pati
Massa pati = 1 gram
V = 0,8 ml

=

1
= = 1,25/
0,8
- Kebutuhan Pati dan Aquadest untuk katalis HCl
Pati : Aquadest = 1 :16
Volume total = 450 ml
= + +
450 = 5,81 + 16 +
450 = 5,81 + 17
4505,81
= 17
= 26, 13
=
= 1,2526,13
= 32,66
= 16
= 16 26,13
= 418,08
- Kebutuhan Pati dan Aquadest untuk katalis H2SO4
= 24 + +
450 = 1,46 + 16 +
450 = 1,46 + 17
4501,46
= = 26,38
17
=
= 1,2526,38
= 32,98
= 16
= 16 26,38
= 422,08
B. Penentuan Kadar Pati Awal
- Standarisasi larutan fehling
F = 19 ml
- Penentuan Kadar Pati Awal
M HCl = 4 ml
M H2SO4 = 4 ml
C. Hidrolisa Pati
Katalis HCl Katalis H2SO4
t (menit) M (ml) t (menit) M (ml)

(x) (x)
0 2,5 0 7
5 2,4 5 3,5
10 2,3 10 2
15 2,1 15 2
20 2 20 1,8
PRAKTIKAN MENGETAHUI
ASISTEN
Gema, Sekar, Salsa Ilham Dwiyanto Emzar

HIDROLISA PATI
LEMBAR PERHITUNGAN
A. Perhitungan Reagen
a. Kebutuhan katalis
- HCl
Densitas HCl
Massa piknometer isi HCl = 44,72 gram
Massa HCl = 44,72 16,47 = 28,25 gram

=

28,25
= = 1,13 /
25
Kebutuhan HCl
. . 1000. .
=
.
1,13. . 1000.1.0,25
0,1 =
36,5 . 450
= 5,81
- H2SO4
Densitas HCl
Massa piknometer isi H2SO4 = 53,95
Massa H2SO4 = 53,95 16,47 = 37,48 gram

=

28,25
H2SO4 = = 1,56/
25

HIDROLISA PATI
Kebutuhan H2SO4
. . 1000. .
H2SO4 =
.
1,56. . 1000.2.0,0,97
0,1 =
98 .450
H2SO4 = 1,46
b. Kebutuhan Massa Pati dan Volume Air

- Densitas Pati
Massa pati = 1 gram
V = 0,8 ml

=

1
= = 1,25/
0,8
- Kebutuhan Pati dan Aquadest untuk katalis HCl

= + +
450 = 5,81 + 16 +
450 = 5,81 + 17
4505,81
= 17
= 26, 13
=
= 1,2526,13
= 32,66
= 16
= 16 26,13
= 418,08

HIDROLISA PATI
- Kebutuhan Pati dan Aquadest untuk katalis H2SO4

= 24 + +
450 = 1,46 + 16 +
450 = 1,46 + 17
4501,46
=
17
= 26,38
=
= 1,2526,38
= 32,98
= 16
= 16 26,38
= 422,08
B. Perhitungan Kadar Pati Awal
a. Standarisasi Larutan Fehling dengan glukosa standar
F = 19 ml
b. Kadar pati awal
500 100
( ) 0,9
= 350 5

- HCl
500 100
(19 4)0,002/ 0,9
= 350 5
32,66
= 0,0236
- H2SO4
500 100
(19 3,5)0,002/ 0,9
= 350 5
32,99

HIDROLISA PATI
= 0,0241
c. Hidrolisa Pati
- Perhitungan pati terhidrolisa
500 100
( ) 0,9
= 350 5

=
0
- Penentuan harga konstanta laju reaksi

= .

=

1

= .
0 0
0
= . +

= 0 (1 )
1
= . +
(1 )
1
=
(1 )
= +
Katalis HCl
0 = 0,0236
500 100
( ) 0,9
= 350 5
32,66
t = 0 menit
500 100
(19 2,5)0,002 0,9
= 350 5
32,66
= 0,0181

HIDROLISA PATI
0,0181
=
0,0236
= 0,7706
t = 5 menit
500 100
(19 2,4)0,002 0,9
= 350 5
32,66
= 0,0182
0,0182
=
0,0236
= 0,7711
t = 10 menit
500 100
(19 2,3)0,002 0,9
= 350 5
32,66
= 0,0184
0,0184
=
0,0236
= 0,7796
t = 15 menit
500 100
(19 2,1)0,002 0,9
= 350 5
32,66
= 0,0186
0,0186
=
0,0236
= 0,7881
t = 20 menit
500 100
(19 2)0,002 0,9
= 350 5
32,66

HIDROLISA PATI
= 0,0186
0,0187
=
0,0236
= 0,7923
Katalis H2SO4
0 = 0,0241
500 100
( ) 0,9
= 350 5
32,99
t = 0 menit
500 100
(19 7)0,002 0,9
= 350 5
32,66
= 0,0130
0,0130
=
0,0241
= 0,539
t = 5 menit
500 100
(19 3,5)0,002 0,9
= 350 5
32,66
= 0,0169
0,0169
=
0,0241
= 0,702

HIDROLISA PATI
t = 10 menit
500 100
(19 2)0,002 0,9
= 350 5
32,66
= 0,0185
0,0185
=
0,0241
= 0,7676
t = 15 menit
500 100
(19 2)0,002 0,9
= 350 5
32,66
= 0,0185
0,0185
=
0,0241
= 0,7676
t = 20 menit
500 100
(19 1,8)0,002 0,9
= 350 5
32,66
= 0,0187
0,0187
=
0,0241
= 0,7759

HIDROLISA PATI
d. Konstanta kecepatan reaksi

- HCl
t M (ml) XP XA -ln(1- X2 XY
(menit) XA)
(x) (y)
0 2,5 0,0181 0,7706 1,4722 0 0
5 2,4 0,0182 0,7711 1,4744 25 7,372
10 2,3 0,0184 0,7796 1,5123 100 15,123
15 2,1 0,0186 0,7881 1,5516 225 23,274
20 2 0,0187 0,7983 1,5716 400 31,432
= 50 =7,5821 750 77,201

=
2 ()2
(477,201) (507,5821)
=
(4750) (50)2
= 0,169
2
=
2 ()2
(7507,581) (5077,201)
=
(4750) (50)2
= 0,768
=
= 0,169

HIDROLISA PATI
- H2SO4
t M (ml) XP XA -ln(1- X2 XY
(menit) XA)
(x) (y)
0 7 0,0130 0,539 0,7743 0 0
5 3,5 0,0169 0,702 1,2079 25 6,0395
10 2 0,0185 0,7676 1,459 100 14,59
15 2 0,0185 0,7676 1,459 225 21,885
20 1,8 0,0187 0,7759 1,495 400 29,9
= 50 =6,3952 750 72,4145

=
2 ()2
(472,4145) (506,3952)
=
(4750) (50)2
= 0,027
2
=
2 ()2
(7506,3952) (5072,4145)
=
(4750) (50)2
= 0,602
=
= 0,027

Carbohydrate Polymers 98 (2013) 15051513
Contents lists available at ScienceDirect
Carbohydrate Polymers
journal homepage: www.elsevier.com/locate/carbpol
The effect of pH on hydrolysis, cross-linking and barrier properties of

starch barriers containing citric acid
Erik Olsson a,,1 , Carolin Menzel b,1 , Caisa Johansson a , Roger Andersson b , Kristine Koch b ,
Lars Jrnstrm a
a
Department of Engineering and Chemical Sciences, Karlstad University, SE-65188 Karlstad, Sweden
b
Department of Food Science, Uppsala BioCenter, Swedish University of Agricultural Sciences, Box 7051, SE-75007 Uppsala, Sweden
a r t i c l e i n f o a b s t r a c t
Article history: Citric acid cross-linking of starch for e.g. food packaging applications has been intensely studied during
Received 4 April 2013 the last decade as a method of producing water-insensitive renewable barrier coatings. We managed to
Received in revised form 30 June 2013 improve a starch formulation containing citric acid as cross-linking agent for industrial paper coating
Accepted 18 July 2013
applications by adjusting the pH of the starch solution. The described starch formulations exhibited both
Available online 26 July 2013
cross-linking of starch by citric acid as well as satisfactory barrier properties, e.g. fairly low OTR values
at 50% RH that are comparable with EVOH. Furthermore, it has been shown that barrier properties of
Keywords:
coated papers with different solution pH were correlated to molecular changes in starch showing both
Starch
Cross-linking
hydrolysis and cross-linking of starch molecules in the presence of citric acid. Hydrolysis was shown
Hydrolysis to be almost completely hindered at solution pH 4 at curing temperatures 105 C and at pH 5 at
Molecular structure curing temperatures 150 C, whereas cross-linking still occurred to some extent at pH 6.5 and drying
Barrier properties temperatures as low as 70 C. Coated papers showed a minimum in water vapor transmission rate at pH
Citric acid 4 of the starch coating solution, corresponding to the point where hydrolysis was effectively hindered
but where a signicant degree of cross-linking still occurred.
2013 Elsevier Ltd. All rights reserved.
1. Introduction sensitivity. Strategies being utilized involve blending with natural

(Gaspar, Benko, Dogossy, Reczey, & Czigany, 2005) or synthetic (Shi
The search for renewable materials to replace synthetic oil- et al., 2008) polymers, the incorporation of inorganic ller such as
based materials as barriers for consumer packages is an intense clay (Magalhes & Andrade, 2009), incorporation of organic llers
research eld. Thermoplastic starch, TPS, is one of the most promis- based on cellulose (Lopez-Rubio et al., 2007; Svagan, Hedenqvist,
ing biopolymer materials to be used in packaging applications since & Berglund, 2009), or starch (Habibi & Dufresne, 2008), modifying
it is renewable, is abundant at low cost, and is fully biodegradable. the starch (Jonhed, Andersson, & Jarnstrom, 2008; Yu, Chang, & Ma,
However, the inherent water sensitivity of TPS is a major problem 2010), studying different plasticizers and plasticizer concentrations
limiting its use as a barrier material. Plasticization lowers the glass (Huang, Yu, & Ma, 2006; Lourdin, Coignard, Bizot, & Colonna, 1997;
transition temperature, Tg , and increases the molecular mobility Olsson, Hedenqvist, Johansson, & Jrnstrm, 2013), using combi-
in the lms. Water is one of the most efcient plasticizers for TPS nations of plasticizers (Shi et al., 2007) and by cross-linking the
(Mathew & Dufresne, 2002). Since starch and many of its plasticiz- starch (Yoon, Chough, & Park, 2006) for instance with citric acid,
ers are hygroscopic, which means that the moisture content in the CA (Menzel et al., 2013). CA is generally recognized as safe accord-
lms increases with increasing relative humidity, RH, the barrier ing to FDA and EFSA and is thus suitable for use in food packaging
properties are signicantly reduced with increasing RH (Forssell, applications. The incorporation of CA and other natural polycar-
Lahtinen, Lahelin, & Myllarinen, 2002; Gaudin, Lourdin, Forssell, & boxylic acids into lms based on renewable materials has been the
Colonna, 2000). subject of recent research since CA has been shown to improve
Recent research regarding TPS has focused on how to bet- the mechanical and barrier properties of such materials, which has
ter utilize its potential as a barrier by decreasing its water mainly been attributed to cross-linking (Dastidar & Netravali, 2012;
Ning, Xingxiang, Na, & Jianming, 2010; Olivato, Grossmann, Bilck,
& Yamashita, 2012; Wang, Yu, Chang, & Ma, 2007). Cross-linking
Corresponding author. Tel.: +46 547002172. of polysaccharide materials such as thermoplastic starch (Shi et al.,
E-mail address: erik.olsson@kau.se (E. Olsson). 2007), paper (Yang, Xu, & Wang, 1996), cellulose derivatives (Coma,
1
These authors are contributed equally. Sebti, Pardon, Pichavant, & Deschamps, 2003), starch granules (Xie
0144-8617/$ see front matter 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.carbpol.2013.07.040
1506 E. Olsson et al. / Carbohydrate Polymers 98 (2013) 15051513
Fig. 1. Schematic illustration of the acid-catalyzed esterication of citric acid and starch to starch citrate (mono-esteried) and a possible structure of citric acid cross-linked
starch.
& Liu, 2004), and cotton bers (Andrews & Reinhardt, 1996) has reducing the pH or by adding Lewis acids. The schematic reaction
been shown to occur with different polycarboxylic acids such as mechanism is shown in Fig. 1. In most previous studies, very high
butanetetracarboxylic (Yang et al., 1996), malic, tartaric, malonic, reaction temperature in most cases well above 100 C was used to
glutaric, succinic, adipic and citric acid (Seidel et al., 2001; Shi et al., initiate cross-linking (Dastidar & Netravali, 2012; Ning et al., 2010;
2007) with or without a catalyst such as sodium hypophosphite Olivato et al., 2012; Shi et al., 2007; Wang et al., 2007; Wing, 1996).
(Reddy & Yang, 2010) or dihydrogen sodium phosphate (Wing, However, in a recent study, it was possible to achieve signicant
1996) at elevated temperatures. The addition of CA to starch has cross-linking at low temperature (70 C) if the amount of CA was
previously shown to give solution cast lms a reduced water- high. The reaction efciency of the CA was low in that case and
sensitivity and improved barrier properties, both by a reduction only a very small fraction of the added CA took part in the cross-
of the diffusion coefcient of water and by a reduction of the linking reaction (Menzel et al., 2013). Nevertheless, unreacted CA
moisture content of the lms (Olsson et al., 2013; Ghanbarzadeh, and monoesteried CA can work as a plasticizer for the starch as
Almasi, & Entezami, 2011). The reaction mechanism for the cross- seen i.e. by a reduction in the Tg (Menzel et al., 2013). It has even
linking, i.e. inter molecular di-ester formation, is the well-known been shown that CA can be used as the sole plasticizer for starch
Fischer-esterication between the carboxylic acid groups of CA and (Olsson et al., 2013) and increase the elongation at break for starch-
the hydroxyl groups in starch occurring twice within the same poly (vinyl alcohol) blends. The increase in elongation is stronger
CA molecule. The formation of an ester-bond can be catalyzed by than for glycerol as plasticizer (Yoon et al., 2006). It has also been
E. Olsson et al. / Carbohydrate Polymers 98 (2013) 15051513 1507
shown to increase the elongation at break in combination with speed of 6 m/min. The coated sheets were subsequently dried in an
other plasticizers such as glycerol for starch (Jiugao, Ning, & Xiaofei, oven for 90 s at 70, 105 or 150 C.
2005; Ning et al., 2010; Wang et al., 2007) or starch-poly (vinyl alco-
hol) blends (Ning et al., 2010; Shi et al., 2008; Wang et al., 2007). It 2.2. Multi-angle laser-light scattering for the determination of
has also been shown that CA reduces the Tg (Menzel et al., 2013). weight-average molecular weight (MW )
In excess water, CA can promote starch fragmentation, the onset
of gelatinization and acid hydrolysis. Acid hydrolysis during gela- 2.2.1. MW after de-esterication
tinization is promoted by both an increase of CA content and a Starch lm samples were dissolved in 0.1 M NaOH to ensure that
reduction in pH (Campbell & Briant, 1957; Hansuld & Briant, 1954; starch citrate esters were completely de-esteried and that only
Hirashima, Takahashi, & Nishinari, 2005; Yamada, Morimoto, & molecular changes in starch molecules were measured. The deter-
Hisamatsu, 1986). Cleavage of glycosidic linkages in starch by acid mination of MW is described elsewhere (Menzel et al., 2013) and
hydrolysis involves both the protonation of the glycosidic oxygen was carried out using high-performance size-exclusion chromatog-
and the addition of a water molecule to yield the reducing sugar raphy (HPSEC) coupled with multi-angle laser-light scattering
end group. However, the addition of acid to a starch formulation (MALLS) and refractive index (RI) detectors. The measurements
does not automatically lead to hydrolysis. For acid hydrolysis to were performed in duplicate.
take place with carboxylic acids, both low pH and high tempera-
ture are needed, as demonstrated in a study by Hirashima et al. 2.2.2. MW of the water-soluble fraction before and after
(2005) where the addition of acids before and after gelatinization de-esterication and solubility in water before de-esterication
was studied, and it was found that no hydrolysis occurred without The water solubility of the starch lms before the addition
sufcient temperature (Hirashima et al., 2005). It has been sug- of NaOH was determined according to Menzel et al. (2013). To
gested that it is possible to produce CA cross-linked starch lms detect changes in molecular weight in the starch lms, i.e. cross-
without excessive hydrolysis by pre-drying at a low temperature, linking due to CA esterication, the MW of the water-soluble
thereby removing most of the water before raising the tempera- sample fraction of the starch lms, before and after subsequent de-
ture (Wing, 1996). However, a recent study showed that exposure esterication was determined, as described in Menzel et al. (2013),
to high temperature, (150 C) of starch lms with a high CA con- using the HPSEC-MALLS-RI system. The measurements were per-
tent (30 pph) resulted in severe hydrolysis even in pre-dried lms formed in duplicate.
(Menzel et al., 2013).
In the present work, the effect of pH and reaction temperature 2.3. Molecular weight distribution of amylose and amylopectin
(curing) on material properties of CA containing starch lms was
examined. A high amount of CA was used both for the improved The amylopectin and amylose distribution was determined
barrier properties and the higher reaction kinetics seen in previ- according to Menzel et al. (2013). In brief, starch lm samples
ous studies (Menzel et al., 2013; Olsson et al., 2013) Factors such were dissolved to a concentration of 5 mg/mL in 0.1 M NaOH to
as molecular weight and degree of cross-linking were linked to break CA ester-linkages and an aliquot was subsequently frac-
the barrier properties, i.e. water vapor and oxygen transmission tionated by size-exclusion chromatography on a Sepharose CL-2B
rates. column (GE-Healthcare, Uppsala, Sweden). The prole of the elu-
ting fractions was determined using the phenolsulfuric acid
method (DuBois, Gilles, Hamilton, Rebers, & Smith, 1956) and
2. Materials and methods iodine staining (Morrison & Laignelet, 1983). The measurements
were performed in duplicate.
2.1. Materials
2.4. Degree of di-esterication according to complexometric
The starch lms and coatings consisted of a hydroxypropylated titration of CA with copper (II)-sulfate
and oxidized potato starch (Solcoat 155) (with an amylose content
of 21%, a degree of substitution of hydroxypropyl groups 0.11 and The concentration of CA di-ester was determined using the
an average 0.01 carboxylic acid groups per anhydroglucose unit complexometric titration method (Graffman, Domels, & Strater,
(Jonhed et al., 2008)) generously supplied by Solam (Kristianstad, 1974; Klaushofer, Berghofer, & Steyrer, 1978) with the modica-
Sweden), and CA (anhydrous citric acid, Puriss, SigmaAldrich Inc., tions described by Menzel et al. (2013). Based on the stable complex
St. Louis, USA), added at a concentration of 30 parts per hundred, formation reaction of free and asymmetrically mono-esteried CA
pph based on the dry weight of starch. The initial pH of the starch molecules with copper (II)-ions, two titrations were performed
solution with 30 pph CA was approximately 2 and it was adjusted on the starch lms. First, the lm samples were hydrolyzed with
to 3, 4, 5, and 6.5 with 40 wt% NaOH. 0.1 M KOH (pH > 12), adjusted to a pH of 8.5 with a borax/boric acid
The starch gelatinization, lm casting and paper coating were buffer and titrated with a 0.02 M copper (II)-sulfate solution. In a
performed according to Olsson et al. (2013). In brief, the starch second titration, lm samples were not hydrolyzed butonly pre-
was gelatinized in a boiling water bath under vigorous stirring for swollen in water, borax/boric acid buffer solution was added and
45 min. CA was added to the gelatinized starch after it had cooled to the sample solution was titrated with the 0.02 M copper (II)-sulfate
room temperature and the pH was adjusted with NaOH. The lms solution. The difference between the titrations of the hydrolyzed
used for molecular characterization were cast in Petri dishes and and non-hydrolyzed lms enabled the degree of di-esterication
dried at 70 C for 5 h with or without subsequent curing at 105 to be calculated (Menzel et al., 2013). The measurements were
or 150 C for 10 min. A paper substrate, Super Perga WS Parch- performed in triplicate.
ment, 50 g/m2 (Nordic Paper Greker, Norway), was coated and
used for the subsequent evaluation of barrier properties. This paper 2.5. Gel content of cross-linked lms
is a greaseproof paper with very low porosity. It has a specied
air permeance according to SCAN P26 of 0.01 nm Pa s according to The gel content in formic acid was determined according to
the supplier. The coatings were applied in two layers with a wire- Reddy & Yang (2010) with small modications described in Menzel
wound bar, 0.31 mm in wire diameter, using a bench coater (K202 et al. (2013). In brief, 0.1 g lm sample was placed in 5 mL formic
Control Coater, RK Coat Instruments Ltd., Royston, UK) at a coating acid under mild stirring for both 24 h at 23 C and 5 h at 50 C. The
gel content was determined as the dry weight of the material that desiccant. The test conditions were 23 C and 50% RH and the results
did not pass through a stainless steel mesh (Tyler mesh 30) and are presented in g/(m2 24 h). The measurements were performed
was expressed as percentage of dry lm weight. The measurements in triplicate.
were performed in triplicate.
2.11. Oxygen transmission rate
2.6. Moisture content
The oxygen transmission rate, OTR, was measured according
The determination of the equilibrium moisture content for lms to ASTM D3985-05 using a Mocon OxTran oxygen transmis-
stored at 23 C and 50% RH is described in detail elsewhere (Olsson sion rate tester, Model 2/21 (Mocon Inc., Minneapolis, MA, USA).
et al., 2013). The samples were equilibrated at 23 C and 50% RH and The OTR measurements were performed at a constant tempera-
the moisture content was determined gravimetrically after drying ture of 23 C but using two different approaches to control the RH.
at 105 C for 1 h. The measurements were performed in at least In one approach the RH was kept constant at 50% RH during the
triplicate. entire experiment, while in the other approach equilibration and
measurements were performed at 70% RH, directly followed by an
2.7. Moisture sorption increase to 80% RH (measurements were performed on the same
lms without removing from the machine). The measurements
The method to determine the moisture sorption is described in were performed in duplicate.
more detail elsewhere (Olsson et al., 2013), however this time it was
performed on lm fragments instead of whole lms. The measure- 3. Results and discussion
ments and equilibration was performed at 23 C. Film fragments
were rst equilibrated at a low RH, below 25% RH, before being put 3.1. Multi-angle laser-light scattering for the determination of
into the controlled moisture generator where the measurements weight-average molecular weight (MW )
started at 50% RH. The weight was recorded as a function of time
and RH. For the lms which had not reached equilibrium at the end The results of the MW determination are shown in Table 1.
of each % RH stage, the equilibrium value was calculated assum- As reported in a previous study (Menzel et al., 2013), the MW
ing that the moisture uptake reached equilibrium asymptotically of the reference starch lms without CA was about 9.0 106 to
according to Olsson et al. (2013). After the trials, the dry weight of 9.9 106 g/mol for non-cured and cured lms, respectively. Addi-
the lms was determined gravimetrically by drying overnight at tion of 30 pph CA and curing at 150 C severely degraded the starch,
105 C and used to calculate the moisture content at the different resulting in MW of 0.2 106 g/mol. As expected, an increase in pH
RH. The measurements were performed in at least duplicate. value of the starch solution resulted in higher MW up to the MW of
the reference starch lms without CA. As indicated by the results in
2.8. Thermal analysis Table 1, no signicant starch degradation was observed at pH 4
and curing temperatures of 105 C or lower. At the higher curing
The effects of solution pH and curing temperature on the Tg of temperature (150 C), however, pH 5 was required to prevent
the starch lms were determined by modulated differential scan- starch degradation. In a previous study (Menzel et al., 2013), it
ning calorimetry, MDSC, using the method described in Menzel was shown that increasing CA content and curing the temperature
et al. (2013). In brief, a TA Systems DSC Q2000 (TA Instruments, resulted in reduction in MW . This was explained in terms of more
New Castle, USA) was used and the scans were performed by heat- pronounced starch degradation at high CA levels and high temper-
ing the samples from 0 to 170 C with an underlying heating rate of atures due to acid hydrolysis of the glycosidic bonds in the starch
2 K/min and a modulated heat amplitude of 0.318 K with a period of molecules.
60 s. The samples were equilibrated and sealed in hermetical cups
at 23 C and 50% RH. The inection point in the reversible heat ow 3.2. Molecular characterization of amylose and amylopectin
curve was taken as the Tg . All measurements were performed in at
least triplicate. Changes in the amylose and amylopectin distribution can be
detected by iodine staining and changes in the elution prole
2.9. Coat weight by size-exclusion chromatography, as reported by (Menzel et al.,
2013), and are shown in the chromatograms in Fig. 2. There
The coat weight of the coated paper was measured gravimet- were two main peaks in the chromatogram, the rst peak elu-
rically as the difference in weight between coated and uncoated ting between 55 and 70 mL corresponding mainly to amylopectin
papers of the same area after conditioning at 23 C and 50% RH for molecules with typically maximum absorbance, max around
at least 24 h. The measurements were performed in triplicate. 540 nm, and a broad second peak eluting between 80 and 150 mL
corresponding mainly to amylose molecules and small amylopectin
2.10. Water vapor transmission rate molecules with max up to 610 nm (Morrison & Laignelet, 1983).
At a lower solution pH and at higher curing temperature, the
The water vapor transmission rate, WVTR, was measured rst eluting peak decreased due to acid hydrolysis of glycosidic
with the gravimetric dish method, ISO 2528, using silica gel as bonds within the large amylopectin molecules. Amylose molecules,
Table 1
Weight-average molecular weight (MW ) of de-estered starch molecules in cured and non-cured starch lms produced at different pH values of the starch-containing solution.
Error limits indicate standard deviation based on duplicates.
Curing MW [106 g/mol] in NaOH
pH 2 pH 3 pH 4 pH 5 pH 6.5
Non-cured 6.1 0.08 8.7 1.18 9.2 0.53 8.5 1.12 9.7 0.42
105 C 5.8 0.05 8.1 0.76 9.1 0.16 8.8 1.15 9.4 0.27
150 C 0.2 0.01 3.9 0.14 4.8 0.75 8.3 0.93 9.5 0.36
0.07
0.06
Degree of di-esterification
0.05
0.04
0.03
0.02
0.01
Fig. 2. Chromatograms (lines) and max (dots) for starch lms cured at 150 C with- 0.00
out CA and with 30 pph CA at pH 2, pH 4 and pH 6.5.
pH2 pH3 pH4 pH5 pH6.5
Fig. 3. Degree of di-esterication of starch lms prepared at different pH values and

eluting later in the chromatogram, were probably also degraded, curing temperatures, non-cured (white) and cured (105 C light gray and 150 C
as seen by the decrease in absorbance in the elution interval 71 to gray). Mean value of triplicates, error bars indicate standard deviation.
120 ml as well by the decline in max between 108 and 118 ml
for the non pH adjusted starch lm. However, the adjustment of evaporates during the curing at 150 C, thereby increasing the yield
the starch lm formulation to a higher pH value reduced the starch of ester-bonds.
degradation during lm preparation, drying and curing. From the
MW measurements reported in Table 1 at pH 5, it was possible
3.4. MW of the water soluble-starch fraction before and after
to cure lms at 150 C without any considerable degradation of
de-esterication and solubility in water
starch. However, as seen in the chromatograms a low degree of
degradation was detectable even at pH 6.5 (Fig. 2).
The MW of the water soluble starch fraction before and after
de-esterication was measured in order to detect cross-linkages
3.3. Degree of di-esterication according to complexometric between CA and starch molecules, as described in Menzel et al.
titration of CA with copper (II)-sulfate (2013). The results of the MW measurements are shown in Table 2
and water-solubility determinations are shown in Fig. 4. The MW
The amount of di-esteried CA ranged from 0.3 to 21% of total of the water-soluble fraction of the starch lms showed that the
added CA (data not shown) and the corresponding degree of di- starch lms prepared at pH 2 and the lms prepared at pH 3 and 4
esterication (DDE) is given in Fig. 3. Films prepared at pH 2 showed and cured at 150 C contained fractions with low MW , which implies
the highest DDE values ranging from 0.008 to 0.054 indicating that the starch was severely hydrolyzed. This was not seen in the
that in the latter case every 19th anhydroglucose monomer is di- lms prepared and cured at other solution pH levels and temper-
esteried (data from Menzel et al., 2013). An increase in pH resulted atures. A signicant reduction in MW after de-esterication of the
in a lower di-ester content. For lms with a pH between 2 and 5, an water-soluble starch fraction was seen for the lms prepared at
increase in curing temperature resulted in higher di-ester content solution pH 4 and in the lms cured at 150 C implying that cross-
(Fig. 3). At pH 6.5, no such trend was seen which can be a result of linking of starch by CA occurred at pH values between 2 and 6.5. One
the low precision of the method. drawback of this analysis is that the cross-linking information is
The lower di-ester content at higher pH can be explained by limited to the water-soluble part of the starch. The water-insoluble
the reaction mechanism for ester-bond formation (Fig. 1) since the part of the starch lm is expected to be even more cross-linked.
reaction is catalyzed at low pH. The increase in di-ester content However, the decrease in MW after de-esterication indicated by
with higher temperature could be a result of the changed equi- this method showed that at least some of the di-esters shown in
librium due to residual moisture in the pre-dried starch lms that Fig. 3 are formed between two different starch molecules.
Table 2
Weight-average molecular weight (MW ) of water soluble starch lms before (water) and after de-esterication (+ NaOH) and decrease of MW after de-esterication of cured
and non-cured starch lms produced at different pH values of starch-containing solution. Error limits indicate standard deviation based on duplicates.
Curing MW [106 g/mol] and decrease in MW after de-esterication
pH 2 pH 3 pH 4 pH 5 pH 6.5
Non-cured
Water 0.41 0.11 8.1 1.81 9.5 0.07 8.8 0.54 10.3 0.09
+ NaOH 0.33 0.01 5.1 1.26 8.7 0.22 8.5 0.63 10.1 0.66
Decrease 19% 37% 8% 3% 2%
105 C
Water 0.27 0.01 8.3 1.15 9.6 0.38 8.5 1.10 10.8 0.04
+ NaOH 0.22 0.02 4.9 0.7 8.5 0.01 8.1 0.81 10.1 0.06
Decrease 18% 41% 11% 4% 6%
150 C
Water 0.34 0.01 0.16 0.01 0.19 0.01 7.8 0.95 10.2 0.60
+ NaOH 0.051 0.03 0.13 0.01 0.15 0.06 6.5 0.55 9.5 0.51
Decrease 85% 19% 21% 18% 7%
100 Table 4
Moisture content of lms at 23 C and 50% RH of starch lms produced at different
pH and different curing temperatures. Error limits indicate standard deviation based
[g Anhydroglu/ 100 g starch]
on at least triplicates.
80
water soluble starch
Curing Moisture content [%]
pH 2 pH 3 pH 4 pH 5 pH 6.5
60
Non-cured 8.0 0.34 8.0 0.59 7.3 0.06 7.8 0.04 11.2 0.49
105 C 7.6 0.39 6.7 0.71 6.9 0.29 7.0 0.48 11.1 0.29
150 C 8.6 0.29 6.0 0.90 4.9 0.77 5.6 0.43 11.2 1.62
40
sufcient to create a gel (Table 3). This is consistent with the data for
20
the degree of di-esterication (Fig. 3) and water solubility (Fig. 4). At
the highest curing temperature 150 C, the gel content was higher
0 and water solubility was lower at pH 3 and pH 4 than at pH 2 and 5.
pH 2 pH 3 pH 4 pH 5 pH 6.5 The difference in gel content might be due to less hydrolysis with
increasing pH (Table 1), at the same time as cross-linking occured
Fig. 4. Water-soluble starch content of starch lms at different pH, non-cured (Fig. 3).
(white) and cured (105 C light gray and 150 C gray). Mean value of duplicates, For the lms subjected to formic acid at 50 C for 5 h (Table 3),
error bars indicate standard deviation.
i.e. more harsh conditions, it is evident that pH 3 and pH 4 resulted
in the highest gel content. This measurement indicated that the
The starch lms prepared at pH 2 showed an increasing solu- predominant part of these lms were strongly cross-linked, but
bility in water with increasing curing temperature (Fig. 4), which without excessive hydrolysis of starch which would promote dis-
was attributed to more extensive hydrolysis and hence more small solution.
starch molecules being dissolved (data from Menzel et al., 2013). In
contrast, starch lms with pH 3 and pH 4 showed a higher solubil- 3.6. Moisture content
ity in water when non-cured or cured at 150 C than lms prepared
at pH 2, and the solubility decreased with increasing curing tem- The equilibrium moisture content at 23 C and 50% RH for
perature probably due to higher cross-linking of starch. At pH 5 the CA-containing starch lms at different pH values is shown in
and 6.5, however, there were no signicant changes in the water- Table 4, where it is obvious that a reduction in moisture content in
solubility with temperature. Nevertheless, all CA-containing starch the cured samples occurred when the pH of the starchCAwater
lms showed a lower solubility in water than the reference starch solutions was raised to pH 3, 4 and 5. For this, there are several
lms without CA (data not shown), which could be a result of the possible explanations. It may be due to the reduction of hydroly-
cross-linking of starch by CA. The complex interaction of the two sis (Table 1) with increasing pH although cross-linking still occurs
concurrent reactions hydrolysis and cross-linking at different (Fig. 3). In the case of the lms adjusted to a pH between 3
pH values resulted in the complicated water-solubility scatter. and 5, there is a substantial reduction in the moisture content
with increasing curing temperature. This can be explained by the
3.5. Gel content in formic acid increasing degree of di-esterication with increasing curing tem-
perature (Fig. 3). At the same time, excessive hydrolysis is avoided,
The gel content corresponds to the fraction of the lm that has as seen in Table 1. Films with pH 6.5 showed much higher mois-
a degree of cross-linking greater than the gel point, i.e. the amount ture content compared to lms prepared at lower pH values. At
of cross-linking needed for a partially cross-linked system not to be pH 6.5, there were no changes in moisture content with increasing
completely dissolved in a good solvent (Larson, 1999). The degree curing temperature, which may be due to the small differences in
of cross-linking that is needed to reach the gel point is affected by the degree of di-esterication (Fig. 4) and hydrolysis (Table 1) with
the molecular weight. Hence, the measured gel content is affected increasing curing temperature.
by both hydrolysis and cross-linking.
The results of the gel content measurements are shown in 3.7. Moisture sorption
Table 3. The non-pH-adjusted lms (i.e. pH 2 in starch solution)
have been described in detail elsewhere (Menzel et al., 2013). No The equilibrium moisture content at different RH levels for non-
starch gel was detectable neither in lms prepared at pH 6.5 nor cured lms with pH 2, 4 and 6.5 is shown in Fig. 5. Up to 60% RH,
lms cured below 150 C prepared from pH-adjusted solutions. the lms prepared at pH 4 had the lowest moisture content and the
This demonstrated that in these starch lms the degree of cross- lms prepared at pH 6.5 had the highest moisture content. At higher
linking was not sufcient to create a gel, whereas the conditions for RH, however, the increase in moisture content was signicantly
preparation of starch lms cured at 150 C and adjusted to pH 5 larger for the lms prepared at pH 4 compared to those at pH 2
and starch lms at all curing temperatures prepared at pH 2 were or 6.5. The lms adjusted to pH 6.5 showed an exceptionally small
Table 3
Gel content for CA-containing starch lms produced at different curing temperatures and different pH values. Error limits indicate standard deviation based on triplicates.
Curing Dissolving conditions % Gel content in formic acid
pH 2 pH 3 pH 4 pH 5 pH 6.5
Non-cured 23 C 24 h 41 0 0 0 0
50 C 5 h 11 0 0 0 0
105 C 23 C 24 h 75 1 0 0 0 0
50 C 5 h 0 0 0 0 0
150 C 23 C 24 h 53 6 80 1 76 5 46 0
50 C 5 h 86 64 7 64 12 0 0
Table 5
Tg from MDSC for lms produced at different pH and different curing temperatures.
Error limits indicate standard deviation based on at least triplicates.
Curing Tg [ C]
pH 2 pH 3 pH 4 pH 5 pH 6.5
Non-cured 58 0.9 54.6 0.7 56.6 1.9 52.9 3.1 63.9 0.5
105 C 59.6 0.7 54.2 1.2 55.3 1.6 54.7 3.0 65.5 2.9
150 C 57.5 0.2 52.6 1.0 53.1 1.0 52.0 2.0 64.8 4.4
3.8. Thermal analysis
In a previous study, the addition of CA to starch was shown to

reduce the Tg of starch lms, especially at high CA content, but there
were only rather small changes in the Tg between 20 and 30 pph
CA (Menzel et al., 2013). It was also shown that at 20 pph CA or
above, curing had no signicant effect on Tg . That was attributed
to the competitive reactions of hydrolysis and cross-linking and
that mono-esteried CA can work as an internal plasticizer in the
Fig. 5. Moisture content of non-cured lms at different relative humidity. Error bars
indicate standard deviation based on at least duplicates. starch lms (Menzel et al., 2013). The Tg of starch lms containing
30 pph CA at different pH values and curing temperatures is shown
in Table 5. There were clearly only minor differences in Tg between
the different curing temperatures. Furthermore, there were only
increase in moisture content between 70 and 80% RH, and the raw
minor differences in the Tg at pH levels between 2 and 5, whereas
data, as exemplied in Fig. 6, showed that the lms prepared from
the Tg increased signicantly when the pH was raised to 6.5.
solutions adjusted to pH 6.5 exhibited a moisture uptake behavior
At rst glance, these results seemed rather strange due to the dif-
different from what was expected, represented by the pH 4 sample.
ferences in MW (Table 1), cross-linking (Fig. 3) and moisture content
When the conditions changed from 70 to 80% RH, the lms adjusted
(Table 4). For polymeric materials in general, Tg has been shown
to pH 6.5 showed a rapid initial increase in moisture content and
to increase with increasing molecular weight and crystallinity and
the moisture content then slowly dropped. This also occurred when
to decrease with increasing degree of branching and exibility of
the RH was changed from 80 to 90% even though the behavior was
the chains (Bizot et al., 1997). However, cross-linking has been
less pronounced.
shown to only slightly affect the Tg of starch (Chang, Cheah, &
This behavior, which at rst glance seems quite unlikely, has
Seow, 2000). Similar results were reported regarding the MW (Bizot
been shown previously for both starch and gluten (Oliver &
et al., 1997). In some reports, it has been shown that the Tg changes
Meinders, 2011). However, the authors attributed this to an insta-
only slightly with moisture content when the amount of plasti-
bility in the equipment at high RH levels and therefore did not
cizer is high, e.g. for glycerol (Chaudhary, Adhikari, & Kasapis, 2011;
further discuss their observations. However, it is clear from the data
Forssell, Mikkil, Moates, & Parker, 1997; Lourdin, Bizot, & Colonna,
of Oliver and Meinders (2011), that this behavior started when the
1997), glycerol-diacetate (Lourdin, Coignard, et al., 1997), xylitol
Tg of the starch and gluten lms was the same as the experimental
(Chaudhary et al., 2011) and for starch plasticized only by water
temperature for both starch and gluten. With increasing moisture
(Chang et al., 2000; Perdomo et al., 2009).
content, the Tg of the lms in the study by Oliver and Meinders
The moisture content increased signicantly when the pH was
(2011) dropped below the experiment temperature. This suggests
raised from 5 to 6.5, but despite this a signicant increase in Tg was
that the behavior was not an artifact but was rather due to a change
observed (Table 5). One reason could be different Tg values for the
in the material properties, such as crystallization, occurring at or
sodium salts of CA, which increase with increasing sodium content.
above Tg .
Tg of CA is 11 C and Tg of the sodium salts of CA are 69 C, 115 C
and 158 C for the mono-, di-, and tri-sodium citrate, respectively
(Shalaev & Gatlin, 2010). However, the addition of salts to starch is
very complex since it has also been reported to signicantly reduce
the Tg (Farahnaky, Farhat, Mitchell, & Hill, 2009).
3.9. Barrier properties of coated paper
The coat weight of double coated paper is given in Table 6, where

it is clear that the coat weight depends on the drying temperature;
the higher the drying temperature the lower the coat weight. One
possible explanation of this is that a higher drying temperature
did result in a smoother coating surface which was clearly seen by
visual observation as the rod pattern was weaker with increasing
drying temperature. A more even surface would lead to a slightly
lower volumetric addition in the second coating. However, at least
some of the decrease in coat weight seen with increasing drying
temperature is probably also due to differences in moisture content
of the coatings as seen in Table 4.
The WVTR at 23 C and 50% RH according to the dry cup method
Fig. 6. Moisture uptake with time at different relative humidity for non-cured lms is presented in Table 6. For comparison, the WVTR of the non-
with pH 4 and 6.5. coated paper was roughly 200 g/(m2 24 h). An increase in drying
Table 6 pH 6.5 the OTR is reduced when the RH is increased from 70 to

Coat wt, WVTR (dry cup) and OTR at different relative humidity for coatings with
80% RH. This behavior was extraordinary, but it coincided with the
different pH and different drying temperatures. Error limits indicate standard devi-
ation based on triplicates. For OTR, measurements performed in duplicate, both uncharacteristic moisture uptake seen in Fig. 6, and even though the
measurements are indicated. moisture content increased slightly in the free lms, the OTR of the
coated paper decreased. This suggests that a structural change such
Drying/property Coat wt [g/m2 ], WVTR [g/m2 24 h] OTR [ml/m2 24 h]
as crystallization occured in the material. This is a phenomenon
pH 2 pH 3 pH 4 pH 5 pH 6.5 which could be of great interest for producing an oxygen barrier
70 C from starch that can be used at high relative humidity.
Coat wt 18.1 0.7 17.3 0.1 16.1 0.6 16.8 1.2 18.5 1.4
WVTR 28.7 1.7 26.2 1.2 25.5 2.9 27.1 0.5 41.4 3.2
OTR 50 7/8 7/49 40/200
4. Conclusions
105 C
One major drawback of using CA as cross-linking agent and
Coat wt 15.9 0.7 15.5 0.5 16.6 1.1 16.4 0.5 16.8 0.8
WVTR 25.8 2.0 22.5 0.5 20.3 2.1 20.3 2.9 33.3 1.8 plasticizer in starch coatings for renewable barrier coatings in the
OTR 50 3/6 5/8 5/35 6/140 3/100 packaging sector is the degradation of the starch material due to
OTR 70 12/13 7/9 560/880 acid hydrolysis during industrial processing. This study shows that
OTR 80 56/60 48/51 500/690 it is possible to produce coatings with improved barrier properties
150 C from starch cross-linked and plasticized with CA by adjusting the
Coat wt 14.80.28 14.7 0.56 15.4 0.8 15.0 0.4 15.1 0.1 pH of the starch coating formulation. In order to prevent hydrolysis
WVTR 20.0 0.9 17.9 0.7 15.6 1.7 17.6 3.0 30.1 0.6
an adjustment to pH of 4 was shown to be sufcient. In addition,
OTR 50 3/7 3/6 2/6
a minimum in WVTR was seen at pH 4, which was a reduction of
more than 20% compared to non-adjusted starch coatings formu-
lations (pH 2). At higher pH values, the WVTR increased and at pH
temperature led to a lower WVTR for all the lms at all the pH 6.5, the WVTR was strongly elevated, compare to pH 2. Besides,
levels examined. One explanation, at least for pH levels between fairly low OTR values at 50% RH (between 3 to 8 ml/(m2 24 h)) for
3 and 5, is the reduction in moisture content in the free lms pH 4 coatings were achieved. This shows that it is possible to use
(Table 4). Another explanation may be an increasing cross-linking starch with CA as a barrier material which has barrier properties
of the material with increasing curing temperature, which is evi- comparable with todays petroleum based plastics.
dent at all pH values except at pH 6.5 (Fig. 4) and which could As mentioned, two reactions were shown to affect the bar-
reduce the starch chain mobility. At all curing temperatures, the rier properties of starch based coated paper during processing,
WVTR reached a minimum at about pH 4. At pH 4, the cross-linking hydrolysis and cross-linking. In this regard, several molecular tech-
reaction, which would inhibit molecular movement, is still fairly niques were used to show that the two concurrent reactions which
efcient (Fig. 4), whereas the starch hydrolysis that would enable affect the starch molecular weight in opposite directions, can be
movement in the lm was almost stopped (Table 1). The moisture controlled by a suitable choice of pH and reaction temperature
content (Table 4) was also at a minimum at pH 4 for all curing (curing). It was shown that the hydrolysis reaction was affected
temperatures. more strongly by an increase in pH than the cross-linking, being
Results of the OTR measurement at 50, 70 and 80% RH are given reduced more than the cross-linking reaction. Hydrolysis stopped
in Table 6. It was possible to produce coatings with fairly low OTR almost completely at pH 4 at curing temperatures 105 C and at
values at 50% RH (between 3 to 8 ml/(m2 24 h)) at all pH values, even pH 5 at curing temperatures 150 C, whereas cross-linking still
though there was a clear indication that the OTR values were higher occurred to some extent even at pH values as high as 6.5 and at
for coatings produced at pH 6.5. For some experimental points, the drying temperatures as low as 70 C.
difference between the duplicates was substantial, indicating that This study describes a possibility to produce renewable barrier
it was difcult to produce defect-free coatings. There was a weak coatings based on solutions of starch with a non-toxic cross-linker,
trend toward lower OTR values at higher curing temperatures. It can citric acid, with minimized hydrolysis. The coatings are indus-
nevertheless be concluded that an increase in curing temperature trial applicable in a cost-efcient manner and are an alternative
and low pH values did not increase the OTR. This is important since to petroleum based products.
even though cross-linking in general reduces the mobility of the
material, sometimes it can lead to increased permeability due to
Acknowledgements
reduced density and increased free volume (Bresser, Boolchand, &
Suranyi, 1986; Byun et al., 2007) which has been shown to strongly
The authors thank StoraEnso AB for performing the moisture
affect OTR (Byun et al., 2007).
uptake measurements and BillerudKorsns AB for performing the
When the OTR measurements were performed rst at 70% RH,
oxygen transmission measurements. This study was performed
directly followed by measurements at 80% RH on the same sample
within the project Renewable Functional Barriers which is a part
(Table 6), there was a weak tendency toward lower OTR values at
of the BFP research program, jointly launched by the Swedish Gov-
pH 4 compared to pH 2. One of the major aims of cross-linking is to
ernmental Agency for Innovation Systems (VINNOVA), The Swedish
render the lms less water-sensitive at high RH. Fig. 4 shows that
Forest Industries Federation and Tr-och Mbelfretagen (TMF). The
the cross-linking reaction takes place with less hydrolysis at pH 4
authors acknowledge VINNOVA and the industrial companies tak-
compared to pH 2. This combination of a relatively high degree of
ing part in the project for their nancial support.
crosslinking and low degree of hydrolysis may be one explanation
of the lower OTR values of coatings prepared at pH 4 compared to
those prepared at pH 2. The OTR at 70 and 80% RH were signicantly References
lower at pH 2 and pH 4 compared to pH 6.5. This is probably due
Andrews, B. A. K., & Reinhardt, R. M. (1996). Enhanced cotton textiles from utilization
to a low degree of cross-linking at pH 6.5 (Fig. 4). However, defects research. In Glenn Fuller, Thomas A. Mckeon, & Donald D. Bills (Eds.), Agricul-
in these lms could also result in high OTR values of the coatings tural materials as renewable resources (pp. 3245). Washington, DC: American
produced at pH 6.5. For most water-sensitive barrier materials such Chemical Society.
Bizot, H., Le Bail, P., Leroux, B., Davy, J., Roger, P., & Buleon, A. (1997). Calorimetric
as starch, it is normal that gas transport increases with increasing evaluation of the glass transition in hydrated, linear and branched polyanhy-
RH, as seen for starch coatings with pH 2 and pH 4. However, at droglucose compounds. Carbohydrate Polymers, 32, 3350.
Bresser, W., Boolchand, P., & Suranyi, P. (1986). Rigidity percolation and molecular Lourdin, D., Coignard, L., Bizot, H., & Colonna, P. (1997). Inuence of equilibrium
clustering in network glasses. Physical Review Letters, 56, 24932496. relative humidity and plasticizer concentration on the water content and glass
Byun, Y. J., Hong, S. I., Kim, K. B., Jeon, D. H., Kim, J. M., Whiteside, W. S., et al. (2007). transition of starch materials. Polymer, 38, 54015406.
Physical and chemical properties of gamma-irradiated EVOH lm. Radiation Magalhes, N. F., & Andrade, C. T. (2009). Thermoplastic corn starch/clay hybrids:
Physics and Chemistry, 76, 974981. Effect of clay type and content on physical properties. Carbohydrate Polymers,
Campbell, A. M., & Briant, A. M. (1957). Wheat starch pastes and gels containing 75, 712718.
citric acid and sucrose. Journal of Food Science, 22, 358366. Mathew, A. P., & Dufresne, A. (2002). Plasticized waxy maize starch: Effect of polyols
Chang, Y. P., Cheah, P. B., & Seow, C. C. (2000). Plasticizingantiplasticizing effects of and relative humidity on material properties. Biomacromolecules, 3, 11011108.
water on physical properties of tapioca starch lms in the glassy state. Journal Menzel, C., Olsson, E., Plivelic, T. S., Andersson, R., Johansson, C., Kuktaite, R., et al.
of Food Science, 65, 445451. (2013). Molecular structure of citric acid cross-linked starch lms. Carbohydrate
Chaudhary, D. S., Adhikari, B. P., & Kasapis, S. (2011). Glass-transition behaviour of Polymers, 96, 270276.
plasticized starch biopolymer system a modied GordonTaylor approach. Morrison, W. R., & Laignelet, B. (1983). An improved colorimetric procedure for
Food Hydrocolloids, 25, 114121. determining apparent and total amylose in cereal and other starches. Journal
Coma, V., Sebti, I., Pardon, P., Pichavant, F. H., & Deschamps, A. (2003). Film properties of Cereal Science, 1, 920.
from crosslinking of cellulosic derivatives with a polyfunctional carboxylic acid. Ning, W., Xingxiang, Z., Na, H., & Jianming, F. (2010). Effects of water on the properties
Carbohydrate Polymers, 51, 265271. of thermoplastic starch poly(lactic acid) blend containing citric acid. Journal of
Dastidar, T. G., & Netravali, A. N. (2012). Green crosslinking of native starches with Thermoplastic Composite Materials, 23, 1934.
malonic acid and their properties. Carbohydrate Polymers, 90, 16201628. Olivato, J. B., Grossmann, M. V. E., Bilck, A. P., & Yamashita, F. (2012). Effect of organic
DuBois, M., Gilles, K. A., Hamilton, J. K., Rebers, P. A., & Smith, F. (1956). Colorimetric acids as additives on the performance of thermoplastic starch/polyester blown
method for determination of sugars and related substances. Analytical Chemistry, lms. Carbohydrate Polymers, 90, 159164.
28, 350356. Oliver, L., & Meinders, M. B. J. (2011). Dynamic water vapour sorption in gluten and
Farahnaky, A., Farhat, I. A., Mitchell, J. R., & Hill, S. E. (2009). The effect of sodium starch lms. Journal of Cereal Science, 54, 409416.
chloride on the glass transition of potato and cassava starches at low moisture Olsson, E., Hedenqvist, M. S., Johansson, C., & Jrnstrm, L. (2013). Inuence of citric
contents. Food Hydrocolloids, 23, 14831487. acid and curing on moisture sorption, diffusion and permeability of starch lms.
Forssell, P., Lahtinen, R., Lahelin, M., & Myllarinen, P. (2002). Oxygen permeability Carbohydrate Polymers, 94, 765772.
of amylose and amylopectin lms. Carbohydrate Polymers, 47, 125129. Perdomo, J., Cova, A., Sandoval, A. J., Garca, L., Laredo, E., & Mller, A. J. (2009).
Forssell, P. M., Mikkil, J. M., Moates, G. K., & Parker, R. (1997). Phase and glass Glass transition temperatures and water sorption isotherms of cassava starch.
transition behaviour of concentrated barley starchglycerolwater mixtures, a Carbohydrate Polymers, 76, 305313.
model for thermoplastic starch. Carbohydrate Polymers, 34, 275282. Reddy, N., & Yang, Y. (2010). Citric acid cross-linking of starch lms. Food Chemistry,
Gaspar, M., Benko, Z., Dogossy, G., Reczey, K., & Czigany, T. (2005). Reducing water 118, 702711.
absorption in compostable starch-based plastics. Polymer Degradation and Sta- Seidel, C., Kulicke, W., Hess, C., Hartmann, B., Lechner, M. D., & Lazik, W. (2001).
bility, 90, 563569. Inuence of the cross-linking agent on the gel structure of starch derivatives.
Gaudin, S., Lourdin, D., Forssell, P. M., & Colonna, P. (2000). Antiplasticization and Starch/Strke, 53, 305310.
oxygen permeability of starch-sorbitol lms. Carbohydrate Polymers, 43, 3337. Shalaev, E. Y., & Gatlin, L. A. (2010). The impact of buffer on solid-state properties
Ghanbarzadeh, B., Almasi, H., & Entezami, A. A. (2011). Improving the barrier and and stability of freeze-dried dosage forms. In F. Jameel, & S. Hershenson (Eds.),
mechanical properties of corn starch-based edible lms: Effect of citric acid and Formulation and process development strategies for manufacturing biopharmaceu-
carboxymethyl cellulose. Industrial Crops and Products, 33, 229235. ticals. Hoboken, NJ, USA: John Wiley & Sons, Inc.
Graffman, G., Domels, H., & Strater, M. (1974). Rapid determination of citrate Shi, R., Bi, J., Zhang, Z., Zhu, A., Chen, D., Zhou, X., et al. (2008). The effect of
in washing and cleansing agents by titration with Cu-solution. Fette, Seifen, citric acid on the structural properties and cytotoxicity of the polyvinyl alco-
Anstrichmittel, 76, 218220. hol/starch lms when molding at high temperature. Carbohydrate Polymers, 74,
Habibi, Y., & Dufresne, A. (2008). Highly lled bionanocomposites from functional- 763770.
ized polysaccharide nanocrystals. Biomacromolecules, 9, 19741980. Shi, R., Zhang, Z., Liu, Q., Han, Y., Zhang, L., Chen, D., et al. (2007). Characteriza-
Hansuld, M. K., & Briant, A. M. (1954). The effect of citric acid of selected edible tion of citric acid/glycerol co-plasticized thermoplastic starch prepared by melt
starches and ours. Journal of Food Science, 19, 581589. blending. Carbohydrate Polymers, 69, 748755.
Hirashima, M., Takahashi, R., & Nishinari, K. (2005). Effects of adding acids before Svagan, A. J., Hedenqvist, M. S., & Berglund, L. (2009). Reduced water vapour sorp-
and after gelatinization on the viscoelasticity of cornstarch pastes. Food Hydro- tion in cellulose nanocomposites with starch matrix. Composites Science and
colloids, 19, 909914. Technology, 69, 500506.
Huang, M., Yu, J., & Ma, X. (2006). High mechanical performance MMT-urea and Wang, N., Yu, J., Chang, P. R., & Ma, X. (2007). Inuence of citric acid on the proper-
formamide-plasticized thermoplastic cornstarch biodegradable nanocompos- ties of glycerol-plasticized dry starch (DTPS) and DTPS/poly(lactic acid) blends.
ites. Carbohydrate Polymers, 63, 393399. Starch/Strke, 59, 409417.
Jiugao, Y., Ning, W., & Xiaofei, M. (2005). The effects of citric acid on the properties Wing, R. E. (1996). Starch citrate: Preparation and ion exchange properties.
of thermoplastic starch plasticized by glycerol. Starch/Strke, 57, 494504. Starch/Strke, 48, 275279.
Jonhed, A., Andersson, C., & Jarnstrom, L. (2008). Effects of lm forming and Xie, X., & Liu, Q. (2004). Development and physicochemical characterization of new
hydrophobic properties of starches on surface sized packaging paper. Packaging resistant citrate starch from different corn starches. Starch/Strke, 56, 364370.
Technology and Science, 21, 123135. Yamada, T., Morimoto, Y., & Hisamatsu, M. (1986). Effect of citric acid on potato
Klaushofer, H., Berghofer, E., & Steyrer, W. (1978). Starch citrates production and starch gelatinization. Starch/Strke, 38, 264268.
technical application properties. Strke, 30, 4751. Yang, C. Q., Xu, Y., & Wang, D. (1996). FT-IR spectroscopy study of the polycarboxylic
Larson, R. G. (1999). The structure and rheology of complex uids. New York: Oxford acids used for paper wet strength improvement. Industrial and Engineering
University Press. Chemistry Research, 35, 40374042.
Lopez-Rubio, A., Lagaron, J. M., Ankerfors, M., Lindstrom, T., Nordqvist, D., Mattozzi, Yoon, S., Chough, S., & Park, H. (2006). Properties of starch-based blend lms using
A., et al. (2007). Enhanced lm forming and lm properties of amylopectin using citric acid as additive. II. Journal of Applied Polymer Science, 100, 25542560.
micro-brillated cellulose. Carbohydrate Polymers, 68, 718727. Yu, J., Chang, P. R., & Ma, X. (2010). The preparation and properties of dialde-
Lourdin, D., Bizot, H., & Colonna, P. (1997). Antiplasticization in starch glycerol hyde starch and thermoplastic dialdehyde starch. Carbohydrate Polymers, 79,
lms? Journal of Applied Polymer Science, 63, 10471053. 296300.
Food Hydrocolloids 36 (2014) 369e373
Food Hydrocolloids
journal homepage: www.elsevier.com/locate/foodhyd
Effect of pHs on dispersity of maize starch nanocrystals in aqueous

medium
Benxi Wei, Xiuting Hu, Hongyan Li, Chunsen Wu, Xueming Xu, Zhengyu Jin*, Yaoqi Tian**
The State Key Laboratory of Food Science and Technology, School of Food Science and Technology, Jiangnan University, Wuxi 214122, China
Article history: The effect of different dispersion pHs on zeta potentials, size distribution, and aggregation behavior of
Received 14 May 2013 starch nanocrystals (SNC) prepared by sulfuric acid hydrolysis was examined in this study. The results
Accepted 13 August 2013 showed that zeta potential of starch nanocrystals decreased from 6.7 mV to 34.5 mV as the dispersion
pH increased from 2.07 to 11.96. Smaller starch nanocrystals and wider distribution peaks were observed
Keywords: with the increase of dispersion pH. The data obtained from eld emission scanning electron microscopy
Waxy maize starch
(FE-SEM) also indicated that the aggregated parallelepiped nanoplatelets (1.5 mm) changed to mono-
Nanocrystals
dispersed spherical-like nanoparticles (50 nm) with increasing of dispersion pH from 2.07 to 11.96.
Dispersion pH
Zeta potentials
Considering these ndings, the stable SNC suspension could be obtained by adjusting the dispersion pH
to the range of 7.44e9.45.
Crown Copyright 2013 Published by Elsevier Ltd. All rights reserved.
1. Introduction SNC are crystalline platelets originating from breakdown of the

semicrystalline structure of starch granules by acid hydrolysis of
Over the last decade, starch nanocrystals (SNC) have attracted amorphous parts. For instance, sulfuric acid (H2SO4) was
growing interests due to not only their nanoscaled size but also commonly used (LeCorre, Vahanian, Dufresne, & Bras, 2012). The
their renewable and biodegradable nature (Kim, Lee, Kim, Lim, & stability of SNC suspension is, to a large extent, dependent on the
Lim, 2012; Le Corre, Bras, & Dufresne, 2011). Because of their surface charge of SNC and can be interpreted by DLVO-theory.
unique properties, SNC have been widely used as particle Generally, the stability of a disperse system depends on the bal-
emulsiers to prepare Pickering emulsions (Li, Sun, & Yang, 2012) ance of electrostatic repulsive energy and the van der Waals
and as reinforcement to prepare nanocomposites, such as attraction energy (Adamczyk & Weron ski, 1999). The electrostatic
biodegradable lms and natural polymers (Angellier, Molina- repulsive forces among particles can be illustrated by Electrical
Boisseau, Dole, & Dufresne, 2006; Lin, Huang, & Dufresne, Double Layer theory. Briey, the electrical double layer may consist
2012; Zheng, Ai, Chang, Huang, & Dufresne, 2009). However, due of a layer of electrons (if the non-electrolytic phase is a metal or
to their natural hydrophilicity, SNC are easily self-aggregated and electronic conductor), a layer of adsorbed ions or ionizable groups
settled in water forming agglomerates in micrometer scale on the solid particles, and a diffuse double layer consisting of an
(Chang, Ai, Chen, Dufresne, & Huang, 2009). The aggregation ionic atmosphere, in which ions of one sign are in excess of their
behavior of SNC greatly limits their applications, since a proce- normal concentrations whereas those of the other sign are in
dure of mixing SNC aqueous suspensions with matrix solution is defect (Grahame, 1947).
needed in the casting fabrication of the nanocomposites (Ren, During the formation of SNC, carboxyl groups and sulfate esters
Jiang, Zhou, & Tong, 2011). Consequently, a homogeneous could be introduced to the surface of SNC or microcrystalline cel-
dispersion of SNC is required for high mechanical performance of lulose after H2SO4 hydrolysis (Angellier, Choisnard, Molina-
nanocomposites. Boisseau, Ozil, & Dufresne, 2004; Araki, Wada, Kuga, & Okano,
1998). The dispersion pH might generate signicant inuence on
the ionization of carboxyl groups and sulfate esters according to
DLVO- and Electrical Double Layer theories. This inuence might
decide the dispersity of SNC suspension. Therefore, this study tried
* Corresponding author. Tel./Fax: 86 510 85913299.
to explore the effect of dispersion pHs on zeta potentials, size dis-
** Corresponding author. Tel./Fax: 86 510 85328571.
E-mail addresses: weibenxi@hotmail.com (B. Wei), jinlab2008@yahoo.com
tribution, aggregation behaviors of waxy maize starch nanocrystals
(Z. Jin), yqtian@jiangnan.edu.cn (Y. Tian). prepared by H2SO4 hydrolysis in aqueous medium.
0268-005X/$ e see front matter Crown Copyright 2013 Published by Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.foodhyd.2013.08.015
370 B. Wei et al. / Food Hydrocolloids 36 (2014) 369e373
2. Materials and methods
2.1. Materials
Waxy maize starch (WMS) was kindly donated by Tianjin

Tingfung Starch Development Co., Ltd. (Tianjin, CHN). All other
chemicals and reagents were purchased from Sinopharm Chemical
Reagent Co., Ltd. (Suzhou, CHN) and of analytical grade, unless
otherwise stated. Milli-q water was used in all experiments.
2.2. Preparation of SNC
SNC were prepared according to the procedure described by

Angellier et al. (2004) with minor modication. Starch powders (50 g)
were mixed with 500 mL of 3.16 M H2SO4 solution at 40 C for 7 d with Fig. 1. Zeta potentials of SNC suspensions at different dispersion pH placed for 2 h and
constant stirring at the speed of 200 rpm. The suspensions were 72 h.
washed by successive centrifugations in distilled water until the pH of
supernatant was constant. The resultant samples were dispersed 3. Results and discussion
again using Ultra Turrax T20 (IKA) for 3 min at 13,500 rpm to avoid
aggregates and stored at 4 C with several drops of chloroform. 3.1. Zeta potentials and stability of SNC suspensions
SNC concentrations were determined by weighting lyophilized
powders of the SNC suspensions (10 mL), and expressed as weight Zeta potentials of H2SO4-hydrolyzed SNC suspensions (0.01%, w/
percentage relative to the volume of the water phase. The con- v) at different dispersion pHs placed for 2 and 72 h were shown in
centration of SNC suspension was 3.5% (w/v) with the tests per- Fig. 1. Zeta potentials of all the samples were negative, indicating
formed in triplicate and averaged. Different concentrations of SNC that SNC had negative surface charges. This was due to the presence
suspensions were derived by diluting the stock homogeneous of carboxyl and sulfate esters on the surface of SNC after H2SO4
dispersion of SNC. hydrolysis (Angellier, Putaux, Molina-Boisseau, Dupeyre, &
Dufresne, 2005). Zeta potentials of SNC suspensions (placed for 2 h)
2.3. Size distribution and zeta potentials of SNC decreased from 6.7 mV to 34.5 mV as the dispersion pH
increased from 2.07 to 11.96. However, great changes were
Size distribution and zeta potential measurements were per- observed after storage for 72 h. As the dispersion pH below 5.6,
formed at 25 C with a commercial Malvern Zetasizer Nano ZS 90. little changes could be observed. Zeta potentials for SNC suspen-
Zeta potentials were determined in the presence of 1 mM KCl. The sions signicantly increased when the dispersion pH was in the
pH of the suspension (0.01%, w/v) was adjusted by HCl (0.1 M) or range of 5.6e9.6 (especially pH at 8.8). An ionization equilibrium
NaOH (0.1 M). The refractive index of uid and particle used were state of carboxyl and sulfate esters might contribute to the increase
1.33 and 1.53, respectively. Measurements were carried out in of zeta potentials. On the contrary, zeta potentials for SNC sus-
quintuplicate for error analysis. pensions decreased as the dispersion pH above 9.6. This may result
from the degradation of sulfate esters under alkaline condition.
2.4. Anion-exchange chromatography
SNC suspensions (0.1%, w/v) were heated in a boiling water bath

for 15 min to dissolve the SNC. The dextrin composition was
analyzed by high-performance anion-exchange chromatography
with pulsed amperometric detection system (HPAEC-PAD). The
system was a Dionex DX 500 instrument (Sunnyvale, CA, USA)
consisting of a GP40 gradient pump and an electrochemical detector
for pulsed amperometric detection (PAD). Filtered samples (25 mL)
were eluted at 1 mL min1 with 150 mM NaOH and a gradient of
NaOAc as described earlier by Koch, Andersson, and man (1998).
2.5. Field emission scanning electron microscopy (FE-SEM)
FE-SEM was performed using an S-4800 (Hitachi, Japan) at an

acceleration voltage of 1 kV. SNC suspensions (0.01%, w/v) at
different dispersion pHs were sonicated at 4 C for 15 min. A drop of
suspension was spread onto the copper grids coated with carbon
support lm and then coated with gold for observation.
2.6. Statistical analysis
Statistical analysis was performed by using ORIGIN 7.5 (Origin-

Lab Inc., USA). Data were expressed as means standard deviations
of at least triplicate and analyzed by a one-way analysis of variance Fig. 2. Illustration of SNC suspensions (0.1%, w/v) prepared after 2 h and 72 h at
(ANOVA). A probability P < 0.05 was considered signicant different dispersion pHs (The dispersion pH ware 2.07, 3.92, 5.38, 7.44, 9.45, 10.25,
throughout the study. 10.92, and 11.96 from left to right, respectively).
B. Wei et al. / Food Hydrocolloids 36 (2014) 369e373 371
Fig. 3. Particle size distribution of SNC suspensions at different dispersion pHs.
150
120
19
90
mV
35
60
30
0
0 5 10 15 20 25 30 35 40
Time/min
Fig. 4. HPAEC-PAD chromatograms of waxy maize starch nanocrystals.
Fig. 2 illustrates SNC suspensions (0.1%, w/v) prepared after 2 h repulsion forces. As the dispersion pH increased to the range of
and 72 h at different dispersion pHs. Obviously, the aggregation 3.92e10.25, the tested samples exhibited two prominent peaks at
behaviors of SNC suspensions (2 h) were weakened as dispersion around 50 and 250 nm. The smaller particles resulted from the
pH increased from 2.07 to 5.38. A homogeneous dispersion was isolated SNC and the larger particles were originated from their
obtained when the dispersion pH above 5.38. This result was minor aggregates. The changes in distribution indicated that,
consistent with zeta potentials of the SNC suspension, since comparing with van der Waals attractive forces and hydrogen
carboxyl and sulfate esters were in protonated state under acidic bonding forces among the SNC, the repulsion forces became the
conditions. The repulsive forces caused by the surface charges were dominant forces, although some SNC aggregates might be still
much weaker than van der Waals and hydrogen bonding forces. formed. A wider area of peaks appeared at the range of 4e250 nm,
Therefore, SNC aggregated and caused fast sedimentation of sus- when the dispersion pH was higher than 10.25. The smaller peak
pensions. However, carboxyl and sulfate esters were deprotonated might be derived from either partial dissolution of the smaller SNC
as pH increased. The surface charge, measured by zeta potential, to dextrin or the dissolution of larger particles to smaller ones
was strong enough to repulse SNC away from each other and pro- under alkaline condition. Homogeneous transparent solution was
duced stable suspensions. This could be demonstrated by the fact obtained due to the complete dissolution of SNC as the dispersion
that an absolute zeta potential of at least 30 mV for electrostatic pH reached 11.96. The size distribution spectra showed three peaks
stabilized systems was desired to obtain a physically stable sus- at about 6, 30, and 150 nm, respectively. The former two peaks were
pension (Mller & Jacobs, 2002). Consequently, the decrease of attributed to the dissolution of SNC into dextrin, as two normal
absolute zeta potential resulted in the aggregation behavior of the distribution peaks with peak degree of polymerization (DP) 19 and
SNC suspensions as the pH was above 9.6 after storage for 72 h. 35 were detected in HPAEC-PAD chromatograms of SNC (Fig. 4). The
dextrin distribution of SNC was in agreement with previous nd-
ings (Angellier et al., 2009; Yoo et al., 2009). The third peak at
3.2. Effect of dispersion pHs on particle size distribution
150 nm might result from the incompletely dissolved SNC.
The effect of dispersion pHs on size distribution of SNC sus-
pensions (0.01%, w/v) was shown in Fig. 3. The suspensions showed 3.3. Appearance and morphology transition of SNC
one peak at about 1000e2000 nm when the dispersion pH was
below 3.92, which was ascribed to the serious SNC aggregates. This Fig. 5 illustrates the size, morphology, and aggregation behavior
was due to the fact that van der Waals attractive forces and of SNC suspensions (placed for 2 h) at different dispersion pHs. SNC
hydrogen bonding forces among SNC displayed stronger than the had the shape of parallelepiped nanoplatelets and were generally
372 B. Wei et al. / Food Hydrocolloids 36 (2014) 369e373
Fig. 5. FE-SEM micrographs of SNC suspensions at different dispersion pHs after 2 h (a, b, c, d, e, and f represented samples at pH of 2.07, 3.92, 5.38, 9.45, 10.25, and 10.92,
respectively).
observed in aggregates at about 1.5 mm, while the dispersion pH monodispersed spherical-like nanoparticles as the dispersion pH
was in the range of 2.07e3.92 (Fig. 5a and b). As the dispersion pH increased. This was owing to the increase of repulsion forces among
increased to 5.38e9.45, monodispersed SNC with size about 100e SNC and partial dissolution of larger SNC to smaller ones. Therefore,
200 nm were observed (Fig. 5c and d). This was attributed to the the dispersity of aqueous SNC suspensions could be controlled by
fact that strong aggregation tendency has been greatly reduced as a adjusting the pH of suspensions.
result of the increase of repulsion forces among SNC. This result was
in good agreement with the size distribution peaks in Fig. 3. The Acknowledgments
platelet-like SNC became spherical and with the size range at about
50e100 nm, when the dispersion pH above 10.25. It could be This study was nancially supported by the Project of State Key
interpreted that alkaline condition promoted the dissolution of SNC Laboratory of Food Science and Technology, Jiangnan University
and resulted in the spherical-like and even smaller particles (Fig. 5e (No. SKLF-ZZB-201206), Natural Science Foundation of China (No.
and f). 31201288), and Natural Science Foundation of Jiangsu Province (No.
BK2012115).
4. Conclusions
References
Zeta potential of SNC suspensions was negative and decreased
Adamczyk, Z., & Weron ski, P. (1999). Application of the DLVO theory for particle
with increase of dispersion pHs after storage for 2 h. After storage
deposition problems. Advances in Colloid and Interface Science, 83(1), 137e226.
for 72 h, zeta potentials increased under alkaline condition. Angellier, H., Choisnard, L., Molina-Boisseau, S., Ozil, P., & Dufresne, A. (2004).
Aggregated parallelepiped nanoplatelets were changed to Optimization of the preparation of aqueous suspensions of waxy maize starch
B. Wei et al. / Food Hydrocolloids 36 (2014) 369e373 373
nanocrystals using a response surface methodology. Biomacromolecules, 5(4), with pulsed amperometric detection. Journal of Chromatography A, 800(2),
1545e1551. 199e206.
Angellier, H., Molina-Boisseau, S., Dole, P., & Dufresne, A. (2006). Thermoplastic Le Corre, D., Bras, J., & Dufresne, A. (2011). Evidence of micro and nano-scaled
starch-waxy maize starch nanocrystals nanocomposites. Biomacromolecules, particles during starch nanocrystals preparation and their isolation. Bio-
7(2), 531e539. macromolecules, 12(8), 3039e3046.
Angellier, H., Putaux, J. L., Molina-Boisseau, S., Dufresne, A., Bertoft, E., & Perez, S. LeCorre, D., Vahanian, E., Dufresne, A., & Bras, J. (2012). Enzymatic pretreatment for
(2009). The molecular structure of waxy maize starch nanocrystals. Carbohy- preparing starch nanocrystals. Biomacromolecules, 13(1), 132e137.
drate Research, 344(12), 1558e1566. Lin, N., Huang, J., & Dufresne, A. (2012). Preparation, properties and applications of
Angellier, H., Putaux, J. L., Molina-Boisseau, S., Dupeyre, D., & Dufresne, A. (2005). polysaccharide nanocrystals in advanced functional nanomaterials: a review.
Starch nanocrystal llers in an acrylic polymer matrix. Macromolecular Sym- Nanoscale, 4(11), 3274e3294.
posia, 221, 95e104. Li, C., Sun, P., & Yang, C. (2012). Emulsion stabilized by starch nanocrystals. Starch-
Araki, J., Wada, M., Kuga, S., & Okano, T. (1998). Flow properties of microcrystalline Strke, 64, 497e502.
cellulose suspension prepared by acid treatment of native cellulose. Colloids and Mller, R., & Jacobs, C. (2002). Buparvaquone mucoadhesive nanosuspension:
Surfaces A: Physicochemical and Engineering Aspects, 142(1), 75e82. preparation, optimisation and long-term stability. International Journal of
Chang, P. R., Ai, F., Chen, Y., Dufresne, A., & Huang, J. (2009). Effects of starch Pharmaceutics, 237(1), 151e161.
nanocrystal-graft-polycaprolactone on mechanical properties of waterborne Ren, L., Jiang, M., Zhou, J., & Tong, J. (2012). A method for improving dispersion of
polyurethane-based nanocomposites. Journal of Applied Polymer Science, 111(2), starch nanocrystals in water through crosslinking modication with sodium
619e627. hexametaphosphate. Carbohydrate polymers, 87(2), 1874e1876.
Grahame, D. C. (1947). The electrical double layer and the theory of electro- Yoo, S.-H., Perera, C., Shen, J., Ye, L., Suh, D.-S., & Jane, J.-L. (2009). Molecular
capillarity. Chemical Reviews, 41(3), 441e501. structure of selected tuber and root starches and effect of amylopectin structure
Kim, H. Y., Lee, J. H., Kim, J. Y., Lim, W. J., & Lim, S. T. (2012). Characterization of on their physical properties. Journal of Agricultural and Food Chemistry, 57(4),
nanoparticles prepared by acid hydrolysis of various starches. Starch-Strke, 1556e1564.
64(5), 367e373. Zheng, H., Ai, F., Chang, P. R., Huang, J., & Dufresne, A. (2009). Structure and prop-
Koch, K., Andersson, R., & man, P. (1998). Quantitative analysis of amylopectin erties of starch nanocrystal-reinforced soy protein plastics. Polymer Composites,
unit chains by means of high-performance anion-exchange chromatography 30(4), 474e480.
International Journal of Advanced Research in
IT and Engineering ISSN: 2278-6244
OPTIMUM HYDROLYSIS CONDITIONS OF CASSAVA STARCH FOR GLUCOSE

PRODUCTION
A. Ayodeji Ayoola*
A. Opeyemi Adeeyo**
C. Vincent Efeovbokhan***
D. Adeola Olasimbo****
Abstract: Acid and enzymatic hydrolysis of cassava starch to glucose (fermentable sugar)
were investigated and compared. And the effects of acid concentration, pH, temperature
and time on the yield of glucose were studied. Experiments were carried out at a
temperature range of (60 100)0C between 30 minutes and 4 hours. (0.2 1.0)M strength of
H2SO4 acid was used and pH values range of 4 7 was considered during enzymatic
hydrolysis. The study revealed that maximum concentration of glucose was obtained at
1000C using 1.0M H2SO4 acid for 4 hours during acid hydrolysis. At pH of 4, temperature of
600C and 4 hours of operation, highest concentration of glucose was obtained during
enzymatic hydrolysis. Enzymatic hydrolysis produced higher yield of glucose when compared
to that obtained from acid hydrolysis.
Keywords: Cassava starch, hydrolysis, - amylase, glucoamylase, glucose.
*Lecturer II in Chemical Engineering Department, Covenant University, Nigeria.

**Lecturer II in Chemical Engineering Department, Covenant University, Nigeria.
***Lecturer II in Chemical Engineering Department, Covenant University, Nigeria.
****Undergraduate student in Chemical Engineering Department, Covenant University,
Nigeria.
Vol. 2 | No. 1 | January 2013 www.garph.co.uk IJARIE | 93

1. INTRODUCTION
Drop in the World reserves of petroleum has sensitized the world of the danger embedded
in total dependency on fossil fuel, as the main source of energy in the world of geometric
increase in energy demand. In many countries, Nigeria inclusive, energy consumption is
based on imported refined fossil fuel. But there is need for alternative sources of energy
which can successively compete with fossil fuel, in terms of cost and quality. Energy
obtained from feedstock (such as sorghum, maize, cassava etc.) is such a good alternative, if
efficiently harnessed [9].
Cassava (Manihot esculenta) is a very promising feedstock for glucose production (an energy
source) and a good glucose production technology results in energy generation [4], [2], [11].
The energy in cassava is preserved in a form of carbohydrate, mainly as starch, the greatest
component of dry matter in fresh roots [8]. Starches are generally insoluble in water at
room temperature. There are two types of linkage in starchy structures: -1,4 and -1,6,
linkages. Amylose, a kind of starch, is an unbranched, single chain polymer containing 500 to
2000 glucose subunits with only -1,4 glycosidic links [6]. The presence of -1,6 glycosidic
linkages in some starchy materials results in a branched glucose polymer called amylopectin
[1]. The breaking down of the -1,4 and -1,6 linkages to small units of glucose
(monosaccharide) is made possible by the actions of - amylase and glucoamylase
(enzymes) respectively [13].
The two commonly used technologies in the conversion of starch to glucose are acid and
enzymatic hydrolyses. Acid concentration, operating temperature and duration of hydrolysis
play significant roles in determining both the quantity and quality of glucose, during acid
hydrolysis [12]. During enzymatic hydrolysis, enzymes break both the -1,4 and -1,6
molecular bonds of starch to more simplified smaller units of monomers [6]. -amylase
splits -1,4 bonds in amylose and amylopectin. It is an endo-acting enzyme and its action is
often considered to be random. The -1,6 glycosidic bonds are not hydrolyzed. The
properties as well as the action of -amylase depend on the microorganisms or plants from
which it is derived. However, -amylases rapidly decrease the viscosity of starch solutions
[8]. Glucoamylase is an exo-acting enzyme, hydrolyzing -1,4 and -1,6, glycosidic linkages
in amylose and amylopectin. The rates of hydrolysis depend on the molecular size and
structure of the substrates [8].

The aim of this study is to compare the two methods of acid hydrolysis and enzymatic
hydrolysis for the conversion of cassava starch to glucose (fermentable sugar). Also to
observe the effects of pH and temperature on the action of the enzymes involved during
enzymatic hydrolysis. Also, to establish the effects of variation in acid concentration,
temperature and time of operation on glucose obtained during acid hydrolysis.
2. MATERIALS AND METHODS
Fresh Manihot esculenta was obtained from a local market at Oshodi, Lagos State, Nigeria.
amylase and amyloglucosidase were obtained from the culture collection unit of the
Department of Biotechnology, Federal Institute of Industrial Research, Oshodi (FIIRO) Lagos,
Nigeria. The reagents used during the course of this study include Sulphuric acid (H2SO4)
solution, Sodium hydroxide (NaOH) solution, Anhydrous DGlucose, Distilled water, 3,5-
Dinitrosalicyclic acid (DNSA) and Potassium Sodium Tartrate (Rochelle Salt).
Preparation of Cassava Starch: The tubers were peeled and washed. The tubers were
grated and then soaked in water for three days. After which it was pressed using a muslin
cloth to extract starch content. The starch was allowed to settle down and water was
decanted and dry white powder of starch was obtained using a dryer.
Acid Hydrolysis: 50g of cassava starch was hydrolyzed by dispersing in 150ml of H2SO4 with
solution strength ranging from 0.2M 1.0M. The slurries obtained were kept in water bath
set at different temperatures between 60C and 100C within the time range of 30mins and
4hrs. After the specified time, 50ml of 0.1M NaOH was used to neutralize the activity of any
trace of H2SO4 that may be present. The sugar brix and concentration of clear glucose syrup
obtained were determined using DNSA reagent (Ranken method 1984), with
spectrophotometer operated at 540nm wavelength.
Enzymatic Hydrolysis: Enzymatic hydrolysis of cassava starch was done by dispersing 50g
starch sample each in 150ml of distilled water. The slurry obtained was gelatinized at 80C
for 30mins in a water bath. The slurry was then transferred to water bath at 90C, 2.0ml of
alpha amylase (Termanyl) was added to the slurry and allowed to liquefy for 1 hr. The
liquefied starch was cooled at 60C, pH range of 4.0 7.0 was observed. 2.0ml of
glucoamylase enzyme was added to each of the liquefied slurries and then maintained at
different temperature between 60C and 100 C, between 30mins and 4hrs for

saccharification to take place. Sugar brix and glucose content of the glucose syrup obtained
were determined.
3. RESULTS AND DISCUSSION
The following are the results obtained during acid and enzymatic hydrolysis.
0.07
0.06
Concentration of Glucose, mg/mL
0.05
0.04
60 deg. Celsius
0.03 80 deg. Celsius
100 deg. Celsius
0.02
0.01
0
0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Concentration of Acid, M
Figure 1: Graphs of concentration of glucose obtained during 30 minutes of acid hydrolysis

at different temperatures.
0.07
0.06
0.05
0.04
60 deg. Celsius
80 deg. Celsius
0.02
0.01
0
0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Figure 2: Graphs of concentration of glucose obtained during 1 hour acid hydrolysis at

different temperatures.

0.08
0.07

0.06
0.05
60 deg. Celsius
100 deg. Celsius
0.03
0.02
0.01
0
0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Figure 3: Graphs of concentration of glucose obtained during 2 hours acid hydrolysis at

0.09
0.08
0.07
0.06
0.05
60 deg. Celsius
100 deg. Celsius
0.03
0.02
0.01
0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Figure 4: Graphs of concentration of glucose obtained during 4 hours acid hydrolysis at


0.045
60 deg. Celsius
100 deg.Celsius
Concentration of Glucose, mg/mL 0.035
0.03
0.025
0.02
0.015
0.01
0.005
4 4.5 5 5.5 6 6.5 7
pH Value
Figure 5: Graphs of glucose concentration produced and pH during 30 minutes enzymatic

hydrolysis at various temperatures.
0.07
60 deg. Celsius
80 deg. Celsius
0.06 100 deg.Celsius
0.05
0.04
0.03
0.02
0.01
4 4.5 5 5.5 6 6.5 7
pH Value
Figure 6: Graphs of glucose concentration produced and pH during 1 hour enzymatic


0.09
100 deg. Celsius
60 deg.Celsius

0.07
0.06
0.05
0.04
0.03
0.02
0.01
4 4.5 5 5.5 6 6.5 7
pH Value
Figure 7: Graphs of glucose concentration produced and pH during 2 hours enzymatic

0.11
60 deg. Celsius
0.1 80 deg. Celsius
100 deg.Celsius
0.09
0.08
0.07
0.06
0.05
0.04
0.03
0.02
0.01
4 4.5 5 5.5 6 6.5 7
pH Value
Figure 8: Graphs of glucose concentration produced and pH during 4 hours

enzymatic hydrolysis at various temperatures.
Acid Hydrolysis
Highest glucose concentration was observed at 1000C while lowest concentration was
obtained at 600C, during the 30 minutes of operation (Figure 1). The results of the three
graphs showed a progressive trend indicating that the 30 minutes of hydrolysis was not

sufficient to generate maximum glucose concentration. Similar result is observed at 1 hour

of acid hydrolysis (Figure 2). Figure 3 showed that at 2 hours of acid hydrolysis, maximum
glucose concentration was obtained at 0.8M acid concentration and 1000C. That is, during
acid hydrolysis, increase in the acid concentration, time and temperature favour increase in
concentration of glucose produced (Figure 4). That is, the complete breaking of 1,4
and 1,6 glycosidic bonds to glucose (fermentable sugar) required 4 hours of operation
at acid concentration of 1.0M at 1000C.
Enzymatic Hydrolysis
In Figure 5 8 above, during enzymatic hydrolysis, decrease in pH and temperature favour
increase in concentration of glucose produced, while increase in time favour glucose
production. Highest concentration of glucose was obtained at pH of 4.0, temperature of
600C during 4 hours of operation. These results agreed with the findings of Teerapatr & et. al.
[12].
4. CONCLUSION
During hydrolysis, highest concentration of glucose obtained was during 4 hours of
operation using 1.0M H2SO4 acid concentration during acid hydrolysis. Enzymatic hydrolysis
produced highest yield of glucose at pH of 4.0, temperature of 600C during 4 hours of
operation. And it was found that enzymatic hydrolysis produced higher concentration of
glucose when compared to that obtained from acid hydrolysis. Further work can still be
done on this study. These may be consideration of the effect of: (i) using HCl acid instead of
H2SO4, (ii) increasing time of operation above 4 hours and (iii) decreasing the pH below 4.
BIBLIOGRAPHY
[1] Amutha, R. and Gunasekaran, P. Production of ethanol from liquefied cassava starch
using co-immobilised cells of Zymomonas mobilis and Saccharomyces diastaticus. J. Biosci.
Bioeng., Vol. 92, p 560-564, 2001.
[2] Bede N. O. Nutritional implications of projects giving high priority to the production of
staples of low nutritive quality: The case for Cassava in the Humid Tropics of West Africa,
International Institute of Tropical Agriculture, Ibadan, Nigeria. 2006.
[3] Carl, A.B., Edward, R.A., and David, E.B. Fundamentals of Clinical Chemistry. Saunders,
Imprint of Elsevier. Inc. 2008.

[4] Demirbas, A. Biofuels sources, biofuel policy, biofuel economy and global biofuel
projections. Energy Convers. Manage., 49: 21 06-21 16. 2008.
[5] Erhabor, P.O., Azaiki, S.S. and Ingawa, S.A. Cassava: The White Gold. Initiative
Publication Company, Benin City, Edo State. 2008.
[6] Kanlaya Y. and Jirasak K. A Study of Optimal Conditions for Reducing Sugars Production
from Cassava Peels by Diluted Acid and Enzyme, (Nat. Sci.) Vol.38, p.29 35, 2004.
[7] Klanarong, S. and Kuakoon P. Cassava Ethanol Technology Growth in Thailand. 2008.
[8] Kosaric, N. and Vardar-Sukan, F. ). Potential Source of Energy and Chemical Products. In:
The Biotechnology of Ethanol: Classical and Future Applications, Roehr, M. (Ed.). Wiley-VCH,
Weinheim. 2001.
[9] Mousdale, D.M. Biofuels: Biotechnology, Chemistry and Sustainable Development. CRC
Press, Taylor & Francis Group, U.S.A. 2008.
[10] Ranken, M.D. Food Industries Manual, 21st Edition. Kaptain Szabo Publiser Washinton
DC p138-139. 1984
[11] Sriroth, K.K., Piyachomkwan, K., Amornitikul, N. T. and Buranares, I. A Study of
Directions for the Developments of Ethanol Industry in Thailand, Ministry of Industry.
Cassava and Starch Technology Research Unit. 2003.
[12] Teerapatr, S., Cholada, S., Bongotrat, P., Wichien, K. and Sirintip, C. Utilization of Waste
from Cassava Starch Plant for Ethanol Production. 2004.
[13] Wong, Y.C., Wong, C.T., Onyiruka, S.O. and Akpanisi, L.E. University Organic Chemistry,
The Fundamentals. Africana-FEP Publishers Limited. 2001.

Food Chemistry 156 (2014) 713
Food Chemistry
journal homepage: www.elsevier.com/locate/foodchem
Microwave-assisted acid hydrolysis to produce xylooligosaccharides

from sugarcane bagasse hemicelluloses
Jing Bian a, Pai Peng b, Feng Peng a, Xiao Xiao a, Feng Xu a, Run-Cang Sun a,c,
a
Beijing Key Laboratory of Lignocellulosic Chemistry, College of Materials Science and Technology, Beijing Forestry University, Beijing 100083, China
b
College of Forestry, Northwest A&F University, Yangling 712100, China
c
State Key Laboratory of Pulp and Paper Engineering, South China University of Technology, Guangzhou 510640, China
Article history: Hemicelluloses from sugarcane bagasse were subjected to microwave-assisted acid hydrolysis at mild
Received 6 November 2013 temperature to produce xylooligosaccharides (XOS). The hydrolysis was performed with dilute H2SO4
Received in revised form 8 January 2014 at 90 C and the inuence of acid concentration (0.10.3 M) and reaction time (2040 min) on the XOS
Accepted 28 January 2014
production was ascertained with response surface methodology based on central composite design.
Available online 6 February 2014
The tted models of XOS and xylose yields were in good agreement with the experimental results. Com-
pared to hydrolysis time, acid concentration was a more signicant coefcient in the production of XOS. A
Keywords:
well-dened degree of polymerisation of XOS and the monomer in the hydrolysates were quantied. No
Microwave-assisted hydrolysis
Sulfuric acid
sugar-degraded byproduct was detected. The maximum XOS yield of 290.2 mg g1 was achieved by
Sugarcane bagasse hydrolysis with 0.24 M H2SO4 for 31 min. The results indicated that the yields of xylose and the byprod-
Response surface methodology ucts can be controlled by the acid concentration and reaction time in microwave-assisted acid hydrolysis.
2014 Elsevier Ltd. All rights reserved.
1. Introduction XOS are usually produced from xylan rich lignocellulosic mate-
rials by breaking down some ether bonds in the xylan backbone to
Currently, there has been increasing interest in the production of give compounds with a lower degree of polymerisation (Vzquez
high-value products from ligocellulosic biomass. Particular empha- et al., 2000). In recent years, several methods are being explored
sis has been placed on xylooligosaccharides (XOS), which are sugar to generate XOS. The technologies used are classied into physical,
oligomers made up of xylose residues through b-(1 ? 4)-linkages chemical, and biochemical methods, among which autohydrolysis,
where the number of xylose residues involved in their formation enzymatic and acid hydrolysis have been studied extensively.
varies from 2 to 7 (Caparrs, Garrote, Ariza, Daz, & Lpez, 2007). Autohydrolysis involves the deacetylation of xylans to produce
This is mainly due to their remarkable potential for utilisation in acetic acid, which hydrolyses the hemicelluloses from lignocellu-
many elds including pharmaceuticals, feed formulations and agri- losic materials. This process eliminates the use of corrosive
cultural purposes (Vzquez, Alonso, Domnguez, & Paraj, 2000). chemicals for the extraction and hydrolysis of xylan. However, spe-
Additionally, XOS possess a variety of excellent physiological prop- cial equipment operated at high temperatures and pressures are
erties such as lowering cholesterol, improving bowel function, employed (Wang, Sun, Cao, Tian, & Wang, 2009). In addition,
calcium absorption and lipid metabolism, and reducing the risk of extensive purication processes are needed because of the produc-
colon cancer, since they are not metabolised by human digestive tion of a variety of undesirable products such as degraded lignin,
system but promote the growth of benecial intestinal bacteria monosaccharides, as well as their degraded products. Enzymatic
such as Bidobacterium and Lactobacillus (Chapla, Pandit, & Shah, hydrolysis does not produce undesirable byproducts, but the time
2012; Moure, Gulln, Domnguez, & Paraj, 2006; Vzquez et al., is relatively longer as compared to autohydrolysis. It also requires
2000). Thus, XOS have also been found numerous applications in special conditions for storage and handling of the enzymes
food-related elds. (Samanta, Jayapal, Kolte, Senani, Sridhar, Suresh, et al., 2012). In
addition, the complete enzymatic hydrolysis of xylan depends on
different enzymes acting in synergy since some of them prefer to
randomly cleave unsubstituted xylan, but some prefer more highly
Corresponding author at: Beijing Key Laboratory of Lignocellulosic Chemistry,
College of Materials Science and Technology, Beijing Forestry University, Beijing
branched xylan (de Menezes, Silva, Pavarina, Gumaro Dias,
100083, China. Tel./fax: +86 1062336903. Gumaro Dias, Grossman, et al., 2009). Acid hydrolysis randomly
E-mail address: rcsun3@bjfu.edu.cn (R.-C. Sun). cleaves the glycosidic bonds between adjacent xylose units and
http://dx.doi.org/10.1016/j.foodchem.2014.01.112
0308-8146/ 2014 Elsevier Ltd. All rights reserved.
8 J. Bian et al. / Food Chemistry 156 (2014) 713
usually large amounts of undesirable degraded by-products are NaOH to wash the column and then a 15 min elution with 18
generated. With respect to the heating process, acid hydrolysis is mM NaOH was carried out to re-equilibrate the column.
usually conducted through convection or conduction, which has Calibration was performed with a standard solution of L-arabinose,
low process efciency. D-galactose, D-glucose, D-xylose, D-mannose, galacturonic acid, and
Microwave irradiation is an attractive technology with advanta- glucuronic acid (SigmaAldrich, China). The molecular weights of
ges of lower energy requirements, uniform and selective heating, the precipitated xylan-rich fraction were determined by gel
the ability to start and stop instantaneously, as well as low equip- permeation chromatograph (GPC) on a PL aquagel-OH 50 column
ment size and waste (Delazar, Nahar, Hamedeyazdan, & Sarker, calibrated with PL pullulan polysaccharide standards. The eluent
2012; Gabriel, Gabriel, Grant, Grant, Halstead, & Mingos, 1998; was 0.02 M NaCl in 0.005 M sodium phosphate buffer (pH 7.5). A
Warrand & Janssen, 2007). These benets lead to its wide applica- ow rate of 0.5 ml min1 was maintained. FT-IR spectrum was
tions such as food processing, wood drying, natural products measured using a Nicolet iN 10 FT-IR Microscope (Thermo Nicolet
extract, and lignocellulosic biomass treatment. When biomass is Corporation, Madison, WI) equipped with a liquid nitrogen cooled
subjected to microwave irradiation, it is assumed that the induced MCT detector. The spectrum was measured in the range of
hot spots in the biomass disrupt the structure and thus improve 4000750 cm1 at a resolution of 8 cm1.
the disintegration of the material (Janker-Obermeier, Sieber,
Faulstich, & Schieder, 2012). Recently, reports on microwave-as-
2.4. Microwave-assisted hydrolysis of hemicellulosic fraction
sisted acid hydrolysis for the production of xylose or furfural have
been already available (Kumar & Wyman, 2008; Rose & Inglett,
Microwave irradiation was performed by using a microwave
2010; Yemis & Mazza, 2011), but less attention has been paid on
oven (Beijing Xiang-Hu Science and Technology Development Re-
the treatment of hemicelluloses to produce oligosaccharides.
agent Co., Ltd.) equipped with a magnetic stirrer and a temperature
The aim of the present study was to apply microwave-assisted
probe. Microwave intensity and irradiation time were set by a con-
acid hydrolysis to produce XOS from sugarcane bagasse, an indus-
trol panel. According to the experimental design, the xylan-rich
trial crop with an annual production of up to 100 million tons
hemicelluloses were suspended in dilute sulfuric acid solution
(Zhao, Wang, Lin, & Guo, 2012). The effects of sulfuric acid concen-
with a solid to liquor ratio of 1:25 (g ml1) in each run. A micro-
tration and hydrolysis time on the hydrolysis were assessed in
wave irradiation power of 700 W was applied to heat the suspen-
terms of the composition and degree of polymerisation (DP) of
sion to 90 C and held for the desired time with reuxing. After
hydrolysates. In order to maximise efciency in the production of
irradiation, the reactant was immediately cooled in an ice bath to
XOS, the optimum microwave-assisted hydrolysis conditions were
room temperature and centrifuged to separate solubilised fraction
identied by response surface methodology (RSM).
for analysis. Acid concentration and hydrolysis time for the exper-
imental design are given in Table 1. A total of 11 runs were carried
2. Experimental out and the experimental runs were randomised to minimise the
effect of unexpected variability in the observed responses.
2.1. Raw material
Sugarcane bagasse used in this work was collected from a sugar 2.5. Experimental design and data analysis
factory in Guangzhou, China. It was dried in an oven at 55 C for
16 h, ground using a laboratory mill, and screened to obtain a frac- Preliminary experiments with microwave-assisted hydrolysis of
tion with average mesh size of 2080. The composition of the sug- sugarcane bagasse hemicelluloses indicated that sulphuric acid
arcane bagasse was cellulose 43.6%, hemicelluloses 33.5%, lignin concentration and hydrolysis time had signicant effects on the
18.1%, ash 2.3%, and wax 0.8% on a dry weight basis. yield of xylooligosaccharides. The hydrolysis parameters were
optimised using response surface methodology. Central composite
design (CCD) for 2 factors and three replicates at the center point
2.2. Preparation of hemicellulosic fraction from sugarcane bagasse
was employed. Sulfuric acid concentration (U1) and hydrolysis
time (U2) were chosen for independent variables. The range of
Wax and other extracts of sugarcane bagasse were rstly re-
the two independent variables and the center point values are pre-
moved by reuxing in a Soxhlet apparatus with ethanol/toluene
sented in Table 1 based on the preliminary experiments. The yields
(1:2, v/v) for 8 h. Then the powder obtained was delignied with
of XOS and xylose were determined as the response variable (Y).
sodium chlorite in acidic solution (pH 3.84.0, adjusted by acetic
The variables were coded according to the equation:
acid) at 75 C for 2 h. Subsequently, the delignied material was
extracted with 10% KOH, with a solid to liquor ratio of 1:20 X i U i U i;0 =DU i ; 1
(g ml1) at 30 C for 10 h. The liquid fraction was collected, neutra-
lised to pH 5.56.0 with acetic acid, concentrated, and precipitated where xi is the coded value of the variable Ui, Ui,0 is the value of Ui at
by the addition of three equivalent volumes of 95% ethanol. The the center point, and DUi is the step change.
resulting precipitate was dissolved in distilled water after separat- The relationship of the independent variables and response was
ing by ltration, dialysed (molecular weight cut-off 3500 g mol1) calculated by the quadratic polynomial equation:
against water, and then freeze-dried.
X X XX
Y A0 Ai U i Aii U 2i Aij U i U j ; 2
2.3. Chemical characterisation of isolated hemicellulosic fraction
where Ui and Uj are independent variables, which inuence the re-
The composition of xylan-rich hemicelluloses was determined sponse variable Y, A0 is the intercept, Ai is the linear coefcient, on
by high-performance anion-exchange chromatography (HPAEC) the response Y; Aii is the quadratic coefcient and Aij estimates the
ICS3000 gradient using a Carbopac PA-20 column (4 250 mm, interaction effect between variables i and j on the response Y. The
Dionex). The monosaccharides and uronic acids were separated statistical software Design-Expert (Version 8.0.6) was used for the
in 18 mM NaOH (carbonate free and purged with nitrogen), with regression analysis of the experimental data and the response sur-
post-column addition of 0.3 M NaOH at a rate of 0.5 ml min1. face plots. Analysis of variance (ANOVA) was used to estimate the
Run time was 45 min, followed by a 10 min elution with 0.2 M statistical parameters.
J. Bian et al. / Food Chemistry 156 (2014) 713 9
Table 1
Independent variables of the central composite design and the experimental results.
Run Variables Coded levels Responses

U1 (M) U2 (min) x1 x2 XOS yield (mg g1) Xylose yield (mg g1)
1 0.1 20 1 1 72.3 8.7
2 0.1 30 1 0 138.3 19.9
3 0.1 40 1 1 160.8 28.0
4 0.2 20 0 1 204.7 42.6
5 0.2 30 0 0 290.1 85.5
6 0.2 30 0 0 277.9 99.9
7 0.2 30 0 0 280.5 97.6
8 0.2 40 0 1 262.2 100.2
9 0.3 20 1 1 248.9 101.2
10 0.3 30 1 0 255.1 116.9
11 0.3 40 1 1 232.5 172.8
2.6. Hydrolysate analysis by HPAEC were also determined. The weight-average molecular weight was
76,880 g mol1 and the polydispersity was 3.12, which indicated
Released xylooligosaccharides were quantied by HPAEC Dionex the relatively homogeneous structure of the isolated hemicellulose
ICS 3000 using a Carbopac PA-100 column (4 250 mm, Dionex) preparation. The structure of the hemicelluloses was determined
in combination with a PA-100 guard column (4 50 mm, Dionex). by FT-IR (Fig. 1) since it has been proven to be a easy and quick tool
The column temperature was 30 C and the ow rate of the gradient for studying the structure and property of polysaccharides
elution was 0.4 ml min1. XOS was separated with 080 mM NaAc (Kacurakova & Mathlouthi, 1996). The absorbances at 3386,
gradient in a 100 mM NaOH isocratic (carbonate free and purged 2908, 1643, 1458, 1331, 1241, 1160, 1030, 975 and 899 cm1 in
with nitrogen) for 15 min, followed by a 80300 mM NaAc gradient the spectrum are associated with typical hemicelluloses. The pres-
in 100 mM NaOH for 10 min, then a 10 min elution with 100 mM ence of arabinosyl side chains is documented by the two low-
NaOH was used to re-equilibrate the column before the next injec- intensity shoulders at 1160 and 975 cm1. The intensive band at
tion. The concentration of the oligosaccharides was quantied using 1030 cm1 is due to CAO, CAC stretching and the glycosidic
peak area and compared with those for xylobiose (X2), xylotriose (CAOAC) contributions, indicating a dominant xylan in the iso-
(X3), xylotetraose (X4), xylopentaose (X5), and xylohexose (X6), lated hemicelluloses, in good agreement with the compositional
purchased from Megazyme (Ireland). XOS yield (w/w) was calcu- analysis aforementioned. The sharp band at 899 cm1 is indicative
lated as (X2 + X3 + X4 + X5 + X6)/xylan weight. The free monosac- of the b-conguration for the glycosidic linkages between xylopy-
charides in the hydrolysates were determined by HPAEC as ranose units in the main xylan chains. Data from analytical inves-
described above. The quantitative analysis of the degradation prod- tigations indicated that the isolated hemicelluloses were
ucts in hydrolysate was performed on an high performance liquid potentially applicable for XOS production. Although the branched
chromatography (HPLC, Agilent 1200 series, Agilent Technologies, structure causes the XOS to have an enhanced variety of structures,
USA) equipped with a refractive index detector (Yang, Wang, Xu, some substitute unit, especially arabinose linked to the xylan back-
Sun, & Lu, 2012). The separation was performed on an aminex col- bone, is susceptible to hydrolysis and gives furfural and other
umn (HPX-87H ion exclusion column with a length of 300 mm undesired reaction byproducts during the production (Paraj
and an inner diameter of 7.8 mm, Bio-Rad Laboratories, USA) at et al., 2004). Therefore, the efcient and proper production method
50 C with 5 mM sulfuric acid at a ow rate of 0.6 ml min1. Before and operation conditions should be chosen ideally.
measurements, all the samples were ltered through 0.22 lm syr-
inge lter. 3.2. Effects of microwave-assisted acid hydrolysis of hemicelluloses
Acid hydrolysis leads to partial degradation of hemicelluloses

3. Results and discussion into soluble lower molecular weight polymers, oligosaccharides
and monosaccharides. The proportion of the soluble compounds
3.1. Chemical composition of hemicellulosic preparation depends on the operation conditions. Temperature, acid concentra-
tion, and reaction time are the most crucial parameters in the
Sugarcane bagasse is made up of structural and non-structural hydrolysis of hemicelluloses, since they affect hydrolysis rate and
components. The non-structural components of sugarcane bagasse selectivity. In the present study, a mild temperature of 90 C was
were removed and the dewaxed feedstock was then subjected to utilised as this was the maximum temperature measured after
delignication and alkaline extraction to obtain hemicellulosic the longest microwave irradiation in atmospheric pressure
fraction. The fraction obtained was chemically characterised, since (Warrand & Janssen, 2007). In fact, both the mild temperature
an accurate determination of the starting material is of vital impor- and the way of applying the heat were important in the study.
tance in the XOS production. The structure and purity of the XOS According to the preliminary experiment, the dilute acid concen-
were strongly affected by the chemical composition of the hemi- trations from 0.1 to 0.3 M and reaction time 2040 min were
celluloses utilised (Paraj, Garrote, Cruz, & Dominguez, 2004). selected for hydrolysis treatment.
Compositional analysis indicated that xylose was the dominant After the microwave-assisted acid hydrolysis of sugarcane
component of the isolated hemicellulosic fraction, comprising bagasse hemicelluloses was completed, the hydrolysis products
83.8% of the total neutral sugars. Arabinose (11.3%) appeared to in the solution of each sample were analysed. Although the DP lar-
be the secondary major sugar and galactose (0.5%), glucose (3.1%) ger than X6 could be seen, the amount of each sample was re-
and uronic acid (1.4%) were detected in minor amounts. The results ported up to DP 6, due to the lack of X7 of the standard (Gullon,
showed that the fraction obtained was mainly composed of xylan, Gonzalez-Munoz, & Parajo, 2011). The relative contents of X2, X3,
which was benecial for XOS production. The molecular weights X4, X5, and X6 in XOS are shown in Fig. 2. When the acid concen-
and the polydispersity of the obtained hemicellulose preparation tration increased from 0.1 to 0.3 M with 20 min reaction time, the
Fig. 1. FT-IR spectrum of hemicellulosic fraction from sugarcane bagasse.
35
X2 X3 X4 X5 X6
30
25
Content (%)
20
15
10
0
0.1 M 0.1 M 0.1 M 0.2 M 0.2 M 0.2 M 0.3 M 0.3 M 0.3 M
20 min 30 min 40 min 20 min 30 min 40 min 20 min 30 min 40 min
Acid concentration - Hydrolysis time
Fig. 2. The relative contents of xylobiose (X2), xylotriose (X3), xylotetraose (X4), xylopentaose (X5), and xylohexose (X6) in XOS produced from sugarcane bagasse
hemicelluloses with different acid concentrations and times.
content of X6 decreased drastically from 24.5% to 12.1%; the con- to form byproducts such as furfural and hydroxymethylfurfural
tent of X5 decreased from 25.4% to 18.5%, but the contents of X2 (HMF), reducing XOS purity (Otieno & Ahring, 2012). In the present
and X3 increased from 14.9% and 15.1% to 24.9% and 23.8%, respec- study, no side products were detected by HPLC under the condi-
tively. This indicated that xylopentose and xylohexaose continued tions used. This is probably a consequence of the mild hydrolysis
to be hydrolysed into smaller oligosaccharides after release with conditions at a low temperature of 90 C. This can be supported
the increased acid concentration. As similar observation was also by Neureiter, Danner, Thomasser, Saidi, and Braun (2002) who con-
obtained with the hydrolysis times of 30 and 40 min. The effect cluded that temperature has an important impact on the formation
of different reaction times on XOS production can also be observed of sugar degradation products.
in the gure. When the reaction time increased from 20 to 40 min
at an acid concentration of 0.1 M, the contents of X2 and X3 3.3. Fitting models
slightly increased from 14.9% and 15.1% to 17.3% and 17.9%,
accompanied by a decrease in the contents X5 and X6 from From the above results, it is clear that acid concentration and
25.4% and 24.5% to 23.9% and 20.1%, respectively. The results were hydrolysis time are the major factors that inuence the resulting
paralleled with the acid concentrations of 0.2 and 0.3 M. In addi- products. When the acid concentration increases to a certain value,
tion, higher level of X2 and X3 concentrations and lower amounts the degradation of sugars takes place. It is accepted that the
of X5 and X6 were observed with the higher acid concentrations. oligosaccharide formation is more prominent in dilute acids
However, the content of X4 in all of the hydrolysates remained at (Maki-Arvela, Salmi, Holmbom, Willfor, & Murzin, 2011). However,
a constant level. It was concluded that the distribution of DP for if the acid concentration is too low, the process becomes econom-
the products was dependent on both acid concentration and ically unfavourable. Thus, acid concentration and hydrolysis time
hydrolysis time under the conditions used. should be optimised to achieve high XOS yield. Response surface
Since the DP decreased with the increasing acid concentration methodology is an effective statistical procedure using a minimum
and hydrolysis time, the formation of monomers in the hydrolysate set of experiments to determine the coefcients of a mathematical
was unavoidable. Usually, the treatment with concentrated acid model and the optimum conditions (Yemis & Mazza, 2012). Values
and/or high temperature causes the degradation of carbohydrates of the independent process variables were studied and the
responses obtained from 11 different combinations of microwave-

assisted reaction conditions are shown in Table 1. The yields of XOS
and xylose were in the range of 72.3290.1 mg g1 and 8.7
172.8 mg g1, respectively. The t summary report produced by
Design-Expert recommends the quadratic and linear models for
XOS and xylose yields, respectively. The following regression
equations represent the description of XOS and xylose yields from
the experimental responses:
Y XOS 642:37632 4181:4605U 1 27:14816U 2 26:225U 1 U 2

6965:52632U 2I 0:32905U 22
Y xylose 106:2924242 557:1666667U 1 2:475U 2

The correlation measure for testing good-of-t of the regression
equation is the determination coefcient R2, which is dened as
the ratio of the explained variation to the total variation and dem-
onstrates the agreement between the observed and predicted re-
sults (Nath & Chattopadhyay, 2007). It is suggested that at least
0.80 is a good t for a model (Yemis & Mazza, 2012). In the present
study, R2 for XOS yield was found to be 0.982 and 0.915 for the xy-
lose, respectively. The high values of R2 of the models indicated a
close agreement between the experimental results and the theo-
retical values predicted by the model. This similarity can also be
veried by the high correlations between the observed and pre-
dicted values in Fig. 3.
In order to obtain the best-t model, analysis of variance (ANO-
VA) was considered to determinate the adequacy of the models.
Regression coefcients and analysis of variance for XOS and xylose
yields are presented in Table 2. The probability value (P-value) was
used to check the signicance of the coefcient, which indicates
the interaction between each independent variable. P-values less
than 0.05 for the regression model are statistically signicant and
reversed to the not signicant model term (Oomah & Mazza, 2008).
ANOVA analysis showed that the P-values were 0.1385 and 0.1644
for the two models respectively, indicating the lack of t was not
signicant and further demonstrated the two models were of good
t. The analysis of XOS yield showed that the P-values of U1, U2,
U1U2, U 21 , and U 22 were <0.0001, 0.0103, 0.0106, 0.0004, and
0.01078, respectively, indicating the independent variables U1,
U2, and the quadratic terms of U1U2, U 21 , and U 22 had signicant ef-
fects on XOS yield. This also demonstrated that acid concentration
and reaction time had signicant effects on the production of XOS.
P-value is also an indication of the order of signicance for process Fig. 3. Actual vs. predicated yields of XOS and xylose.
variables affecting the yields, where smaller P-values indicate
higher signicance of the corresponding coefcient (Muralidhar,
Chirumamila, Marchant, & Nigam, 2001). Compared to hydrolysis
The linear effect of acid concentration on XOS yield was ob-
time, the effect of acid concentration was higher on XOS yield,
served only in the range of 0.10.24 M, and further increases in
since a smaller P-value of acid concentration indicated a more sig-
acid concentration resulted in a decrease in XOS yield. Similarly,
nicant coefcient. As for xylose, both U1 and U2 had signicant ef-
an increase in XOS yield was observed with the increasing time
fects on xylose yield as indicated by the low P-values of <0.0001
from 20 to 30 min and then XOS yield decreased as the time in-
and 0.0054, respectively. Obviously, the effect of acid concentra-
creased. The yield of XOS increased as acid concentration increased
tion was more signicant than that of hydrolysis time on xylose
from 0.1 to 0.24 M, then went down as acid concentration further
yield.
increased, which indicated that a sulfuric acid concentration of
0.24 M was required to achieve maximum. Likewise, hydrolysis
3.4. Optimisation of process variables time increased from 20 to 30 min led to an increase in XOS yield
and then a decrease with the further increasing time. The results
Two-dimensional contour plot and three-dimensional response showed that these two variables substantially affected the produc-
surface are the graphical representations of the regression equation of XOS. The experiments performed at both higher acid con-
tion, which provide a method to evaluate the interactions between centration and longer hydrolysis time led to a decrease in the
variables and predict the optimum values for each variable at max- yield of XOS. This was in accordance with the increase of xylose
imum value (Qiao, Hu, Gan, Sun, Ye, & Zeng, 2009). In the present yield discussed below. The red zone in Fig. 4 is the optimal condi-
study, two-dimensional contour plots and three-dimensional re- tion for the production of XOS. The hydrolytic depolymerisation of
sponse surface plots were generated using Design-Expert as pre- the xylan-rich hemicelluloses proceeded most efciently at 0.24 M
sented in Fig. 4. sulfuric acid and a hydrolysis time of 31 min, with a predicted
Table 2
Analysis of variance of the models for XOS and xylose yields.
Source Sum of squares Degrees of freedom Mean square F-value P-value

1
XOS (mg g )
Model 47285.86 5 9457.171 54.16213 0.0002
U1 22216.34 1 22216.34 127.2351 <0.0001
U2 2799.36 1 2799.36 16.0322 0.0103
U1U2 2751.003 1 2751.003 15.75526 0.0106
U12 12291.37 1 12291.37 70.39385 0.0004
U22 2742.983 1 2742.983 15.70933 0.0107
Residue 873.0428 5 174.6086
Lack of t 790.4561 3 263.4854 6.380821 0.1385
Pure error 82.58667 2 41.29333
Corrected total 48158.9 10
Xylose (mg g1)
Model 22301.46 2 11150.73 43.30648 <0.0001
U1 18626.08 1 18626.08 72.33878 <0.0001
U2 3675.375 1 3675.375 14.27418 0.0054
Residual 2059.872 8 257.4841
Lack of t 1940.186 6 323.3643 5.403514 0.1644
Pure error 119.6867 2 59.84333
Corrected total 24361.33 10
Model 22301.46 2 11150.73 43.30648 <0.0001
Fig. 4. Response surface and contour plot of acid concentration vs. hydrolysis time on the XOS (a) and xylose (b) yields.
optimum value of 290.2 mg g1. The XOS yield in the present study followed by 0.2 M, and then 0.3 M. The content of xylose also in-
was higher than the results reported in the hydrolysis of tobacco creased with the increasing reaction time. The phenomenon was
stalk xylan in 0.25 M sulfuric acid for 30 min heated by boiling also consistent with the distribution of the DP. The XOS with high-
water bath (140 mg g1), as well as the hydrolysis by xylanase er DP were hydrolysed into both XOS with lower DP and xylose
(114 mg g1) for 24 h (Akpinar, Erdogan, Bakir, & Yilmaz, 2010). with the increase in acid concentration and hydrolysis time. Under
As shown in Fig. 3, xylose yield in the hydrolysis process the optimum conditions for the production of XOS aforementioned,
increased with the increasing of acid concentration and hydrolysis a xylose yield of 104.8 mg g1 was achieved. This indicated that a
time. The surface analysis showed no interaction between the two certain amount of xylose was obtained when maximum XOS was
variables, and xylose yield linearly increased as a function of the produced during the microwave-assisted acid hydrolysis process.
variables under the conditions used. In general, the concentration Above all, the phenomena indicated that the monosaccharide and
of 0.1 M sulfuric acid resulted in the lowest level of xylose, byproduct can be controlled by acid concentration and reaction
time in the microwave-assisted hydrolysis process at mild Kacurakova, M., & Mathlouthi, M. (1996). FTIR and laser-Raman spectra of
oligosaccharides in water: Characterization of the glycosidic bond.
temperature.
Carbohydrate Research, 284, 145157.
Kumar, R., & Wyman, C. E. (2008). The impact of dilute sulfuric acid on the
selectivity of xylooligomer depolymerization to monomers. Carbohydrate
4. Conclusions Research, 343, 290300.
Maki-Arvela, P., Salmi, T., Holmbom, B., Willfor, S., & Murzin, D. Y. (2011). Synthesis
In this study, the effect of microwave-assisted acid hydrolysis of of sugars by hydrolysis of hemicellulosesa review. Chemical Reviews, 111,
56385666.
sugarcane bagasse on the production of XOS was investigated. The Moure, A., Gulln, P., Domnguez, H., & Paraj, J. C. (2006). Advances in the
statistical analysis showed that acid concentration was more manufacture, purication and applications of xylo-oligosaccharides as food
signicant as compared to the hydrolysis time in the production additives and nutraceuticals. Process Biochemistry, 41, 19131923.
Muralidhar, R., Chirumamila, R., Marchant, R., & Nigam, P. (2001). A response
of XOS. Response surface analysis indicated that the microwave- surface approach for the comparison of lipase production by Candida cylindracea
assisted acid hydrolysis performed with 0.24 M H2SO4 for 31 min, using two different carbon sources. Biochemical Engineering Journal, 9, 1723.
produced a maximum XOS yield of 290.2 mg g1. Xylose yield Nath, A., & Chattopadhyay, P. (2007). Optimization of oven toasting for improving
crispness and other quality attributes of ready to eat potato-soy snack using
increased with an increasing acid concentration and residence time
response surface methodology. Journal of Food Engineering, 80, 12821292.
and no sugar-degraded products were formed during the process. Neureiter, M., Danner, H., Thomasser, C., Saidi, B., & Braun, R. (2002). Dilute-acid
Thus, microwave-assisted acid hydrolysis at mild temperature pro- hydrolysis of sugarcane bagasse at varying conditions. Applied Biochemistry and
vides an efcient technique to produce XOS with great application Biotechnology, 98, 4958.
Oomah, B. D., & Mazza, G. (2008). Optimization of a spray drying process for
potential. axseed gum. International Journal of Food Science and Technology, 36, 135143.
Otieno, D. O., & Ahring, B. K. (2012). A thermochemical pretreatment process to
produce xylooligosaccharides (XOS), arabinooligosaccharides (AOS) and
Acknowledgements mannooligosaccharides (MOS) from lignocellulosic biomasses. Bioresource
Technology, 112, 285292.
The authors are grateful to Grants from the Fundamental Paraj, J. C., Garrote, G., Cruz, J. M., & Dominguez, H. (2004). Production of
xylooligosaccharides by autohydrolysis of lignocellulosic materials. Trends in
Research Funds for the Central Universities (No. BLX2013005, Food Science and Technology, 15, 115120.
JC2013-3), the Major State Basic Research Projects of China Qiao, D., Hu, B., Gan, D., Sun, Y., Ye, H., & Zeng, X. (2009). Extraction optimized by
(973-2010CB732203), and Ministry of Science and Technology using response surface methodology, purication and preliminary
characterization of polysaccharides from Hyriopsis cumingii. Carbohydrate
(863-2012AA023204, 10-25 Project-2012BAD32B06). Polymers, 76, 422429.
Rose, D. J., & Inglett, G. E. (2010). Production of feruloylated arabinoxylo-
oligosaccharides from maize (Zea mays) bran by microwave-assisted
References
autohydrolysis. Food Chemistry, 119, 16131618.
Samanta, A. K., Jayapal, N., Kolte, A. P., Senani, S., Sridhar, M., Suresh, K. P., et al.
Akpinar, O., Erdogan, K., Bakir, U., & Yilmaz, L. (2010). Comparison of acid and (2012). Enzymatic production of xylooligosaccharides from alkali solubilized
enzymatic hydrolysis of tobacco stalk xylan for preparation of xylan of natural grass (Sehima nervosum). Bioresource Technology, 112, 199205.
xylooligosaccharides. LWT-Food Science and Technology, 43, 119125. Vzquez, M. J., Alonso, J. L., Domnguez, H., & Paraj, J. C. (2000).
Caparrs, S., Garrote, G., Ariza, J., Daz, M. J., & Lpez, F. (2007). Xylooligosaccharides Xylooligosaccharides: Manufacture and applications. Trends in Food Science
production from Arundo donax. Journal of Agricultural and Food Chemistry, 55, and Technology, 11, 387393.
55365543. Wang, J., Sun, B., Cao, Y., Tian, Y., & Wang, C. (2009). Enzymatic preparation of wheat
Chapla, D., Pandit, P., & Shah, A. (2012). Production of xylooligosaccharides from bran xylooligosaccharides and their stability during pasteurization and
corncob xylan by fungal xylanase and their utilization by probiotics. Bioresource autoclave sterilization at low pH. Carbohydrate Polymers, 77, 816821.
Technology, 115, 215221. Warrand, J., & Janssen, H. G. (2007). Controlled production of oligosaccharides from
de Menezes, C. R., Silva, . S., Pavarina, . C., Gumaro Dias, E. F., Gumaro Dias, F., amylose by acid-hydrolysis under microwave treatment: Comparison with
Grossman, M. J., et al. (2009). Production of xylooligosaccharides from conventional heating. Carbohydrate Polymers, 69, 353362.
enzymatic hydrolysis of xylan by the white-rot fungi Pleurotus. International Yang, H. Y., Wang, K., Xu, F., Sun, R. C., & Lu, Y. B. (2012). H2SO4-catalyzed
Biodeterioration and Biodegradation, 63, 673678. hydrothermal pretreatment of Triploid poplar to enhance enzymatic hydrolysis.
Delazar, A., Nahar, L., Hamedeyazdan, S., & Sarker, S. (2012). Microwave-assisted Industrial and Engineering Chemistry Research, 51, 1159811604.
extraction in natural products isolation. In S. D. Sarker & L. Nahar (Eds.), Natural Yemis, O., & Mazza, G. (2011). Acid-catalyzed conversion of xylose, xylan and straw
products isolation (pp. 89115). Humana Press. into furfural by microwave-assisted reaction. Bioresource Technology, 102,
Gabriel, C., Gabriel, S., Grant, E. H., Grant, E. H., Halstead, B. S. J., & Mingos, D. M. P. 73717378.
(1998). Dielectric parameters relevant to microwave dielectric heating. Yemis, O., & Mazza, G. (2012). Optimization of furfural and 5-
Chemical Society Reviews, 27, 213224. hydroxymethylfurfural production from wheat straw by a microwave-assisted
Gullon, P., Gonzalez-Munoz, M. J., & Parajo, J. C. (2011). Manufacture and prebiotic process. Bioresource Technology, 109, 215223.
potential of oligosaccharides derived from industrial solid wastes. Bioresource Zhao, L. C., Wang, Y., Lin, J. F., & Guo, L. Q. (2012). Adsorption and kinetic behavior of
Technology, 102, 61126119. recombinant multifunctional xylanase in hydrolysis of pineapple stem and
Janker-Obermeier, I., Sieber, V., Faulstich, M., & Schieder, D. (2012). Solubilization of bagasse and their hemicellulose for xylo-oligosaccharide production.
hemicellulose and lignin from wheat straw through microwave-assisted alkali Bioresource Technology, 110, 343348.
treatment. Industrial Crops and Products, 39, 198203.
Acid hydrolysis of native corn starch: Morphology, crystallinity,

rheological and thermal properties
R.G. Utrilla-Coello a,b , C. Hernndez-Jaimes a , H. Carrillo-Navas a , F. Gonzlez a ,
E. Rodrguez a , L.A. Bello-Prez b , E.J. Vernon-Carter a , J. Alvarez-Ramirez a,
a
Universidad Autnoma Metropolitana-Iztapalapa, Departamento de Ingeniera de Procesos e Hidrulica, Apartado Postal 55-534, Mxico, D.F. 09340,
Mexico
b
Centro de Desarrollo de Productos Biticos (CEPROBI) del Instituto Politcnico Nacional, Yautepec, Morelos, Mexico
Article history: The acid hydrolysis of native corn starch at 35 C was monitored during 15 days. After this time, the
Received 11 December 2013 residual solids were about 37.0 3.0%. First-order kinetics described the hydrolysis data, giving a con-
Received in revised form 9 January 2014 stant rate of kH = 0.18 0.012 days1 . Amylose content presented a sharp decrement of about 85% and
X-ray diffraction results indicated a gradual increase in crystallinity during the rst 3 days. SEM micro-
Available online 21 January 2014
graphs showed that hydrolysis disrupted granule morphology from an initial regular shape to increasingly
irregular shapes. Fractal analysis of SEM images revealed an increase in surface roughness. Fast changes
Keywords:
in the thermal effects were caused by molecular rearrangements after fast hydrolysis of amylose in the
Corn starch
Acid hydrolysis
amorphous regions in the rst day. Steady shear rate and oscillatory tests showed a sharp decrease of
Kinetics the apparent viscosity and an increase of the damping factor (tan()) caused by amylose degradation.
Morphology 2014 Elsevier Ltd. All rights reserved.
Crystallinity
1. Introduction (Wang et al., 2003; Wang, Gao, Yu, & Xiao, 2006). For relatively short
hydrolysis times (up to 72 h) acid thinning did not cause disruption
Starch is widely used in the food industry as the main ingredient of the granular crystalline structure (Singh, Singh Sodhi, & Singh,
for the preparation and processing of different products (sausages, 2009; Singh Sandhu, Singh, & Lim, 2007). Recent studies indicated
bread, pasta, noodles, etc.) or/and as gelling agent, thickener, emul- that the hydrogen ion primarily attacked rstly the interior and
sion stabilizer and fat replacer (Nehir El & Simsek, 2012). Starch then the exterior of C-type starch granules. B-type starch granules
is an important determinant of texture, consistency and senso- started to crack after about 4 days of hydrolysis, while C-type starch
rial acceptance (Bjrck & Asp, 1994). The structural arrangement granules cracked until the hydrolysis progressed up to 16 days
of starch components determines the micro-structure of swollen (Pang et al., 2007). The degradation of amylose and amylopectin
starch granules and amylose/amylopectin ratio, which in turn affect by a high acid concentration resulted in a decrease in storage mod-
the characteristics of food products (Genovese & Rao, 2003; Lu, Duh, ulus (G ), loss modulus (G ), gelling temperature, and gel strength
Lin, & Chang, 2008; Ortega-Ojeda, Larsson, & Eliasson, 2004). Acid of acid-thinned starches (Wang et al., 2003). The microstructure of
hydrolysis is widely used in industry for chemical treatment of starch granules was strongly disrupted by the action of hydrogen
starch particles (BeMiller & Whistler, 2009). The underlying idea ion, depending on acid concentration and temperature. From an
regarding acid hydrolysis is to exploit the susceptibility differ- industrial applications standpoint, a close monitoring of the effects
ences of semi-crystalline and amorphous starch lamellae to acid. of acid hydrolysis on the functionality and micro-structure of starch
Crystalline lamellae were being more resistant to hydrolysis than granules should provide important guidelines for hydrolysis pro-
amorphous lamellae, tending to display negligibly slow hydrolysis cess design oriented to the production of starches with desired
while starch amorphous zones were prone to fast acid hydroly- functional properties (e.g., crystallinity, shear thinning, etc.).
sis (Hoover, 2000; Wang, Truong, & Wang, 2003). The degree of The aims of this work were: (i) to study the acid hydrolysis of
crystallinity increased gradually with the time of acid hydrolysis corn starch at 35 C and to determine the kinetics dynamics of the
process; (ii) to monitor the changes in morphology, crystallinity,
rheology and particle size with hydrolysis time; (iii) to establish
Corresponding author. Tel.: +52 55 5804 4650; fax: +52 55 5804 4650. an interrelationship between the progression of these parameters
E-mail address: jjar@xanum.uam.mx (J. Alvarez-Ramirez). with the hydrolysis kinetics.
http://dx.doi.org/10.1016/j.carbpol.2014.01.046
R.G. Utrilla-Coello et al. / Carbohydrate Polymers 103 (2014) 596602 597
2. Materials and methods measured between 10 and 30 , with a 2 step of 0.020371 , for
38 s per point.
2.1. Materials The crystallinity index was used for quantifying crystallinity
changes as function of the hydrolysis time. The crystallinity index,
Native corn starch was obtained from Gluten y Almidones Indus- CIHW , was computed with the classical method by Hermans and
triales, Mexico City, Mexico. H2 So4 (98%), I2 (98%) and NaOH (99%) Weidinger (1948). Briey, the diffractograms were deconvoluted
were obtained from J.T. Baker (Mexico City, Mexico). Deionized using Origin 8.0 software and the starch crystallinity index, CIHW ,
water was used in all experiments. was calculated by the following equation:
A A
T a
2.2. Acid hydrolysis CIHW = 100 (1)
AT
where Aa is the value of the area under the curve corresponding to
Acid hydrolysis was carried out according to method reported by
the amorphous portion of the diffractograms and AT is the total area
Kim, Lee, Kim, Lim, and Lim (2012) with slight modications. Starch
of the diffractograms, i.e., the sum of the areas of all the resulting
(15 g, dry basis) was dispersed in an aqueous sulphuric acid solu-
deconvoluted peaks.
tion (100 mL, 3.16 M), by stirring and keeping at 35 C for different
time periods (015 days). For monitoring the hydrolysis advance,
the solution was cooled down to 5 C for recovering non-hydrolyzed 2.6. Amylose content
material, including dissolved and suspended starch granules. After-
wards, the suspension was centrifuged (6000 g for 15 min) and The apparent amylose content in the aggregate was determined
the precipitates were washed in distilled water until neutral pH by the Hoover and Ratnayake (2002) test. Starch (20 mg, dry basis)
was reached. The precipitated solids were air-dried at 35 C for 24 h, was dissolved in 90% DMSO (8 mL) by using 10-mL screw cap reac-
put into a sealed glass container and stored at 4 C. In this way, the tion vials. The contents were vigorously mixed for 20 min and then
hydrolysis was monitored as the percent of both suspended solids heated in a water bath (with intermittent shaking) at 85 C for
and dissolved non-hydrolyzed starch relative to the initial starch 15 min. The vials were cooled to room temperature (25 C), and
solids. the contents diluted with water to 25 mL in a volumetric ask.
The diluted solution (1.0 mL) was mixed with water (40 mL) and
mixed with 5 mL I2 /KI solution (2.5 mM I2 and 6.5 mM KI), adjus-
2.3. Scanning electron microscopy (SEM)
ting the nal volume to 50 mL. The contents were let stand for
15 min at room temperature before measuring the absorbance
Native and acid hydrolyzed corn starch particles subjected to
at 600 nm.
different hydrolysis times were mounted on carbon sample hold-
ers using double-side sticky tape and were observed using a JEOL
2.7. Differential scanning calorimetry (DSC)
JMS 7600F scanning electron microscope (Akishima, Japan) with
the GB-H mode at 1 kV accelerating voltage. Micrographs at 3000
Gelatinization properties of native and acid-modied corn
magnication are presented.
starch particles subjected to different hydrolysis times were ana-
lyzed by differential scanning calorimetry (DSC) (TA Instruments,
2.4. Fractal dimension (Df ) Q1000, New Castle, DE, USA) previously calibrated with indium fol-
lowing the procedure described by Palma-Rodriguez et al. (2012)
The SEM images were gray leveled as coded images (1000) with slight modications. A 3.0 mg starch sample (dry basis) was
on 256 values, 0 (black) to 255 (white) with a resolution of weighed in an aluminum pan 6.0 L of deionized water was added.
1280 1024 pixels, each pixel being of size of about 3 m. In order Afterwards, the pan was sealed and allowed to equilibrate for
to minimize the anisotropy effects presented by the images, three 16 h at room temperature before being loaded. The mixture was
non-overlapped sample sections of 400 400 pixels (1200 m by heated in the DSC cell from 20 to 140 C applying a heating rate of
side) were taken for each image. For each sample, it was estimated 5 C min1 . An empty pan was used as the reference. All measure-
the fractal dimension using the regularization dimension method ments were done by triplicate.
(Roueff & Vhel, 1998) available at FracLab 2.1 version. The regular-
ization method was chosen for robust computation in the presence 2.8. Rheological properties of the starch dispersions
of noise contaminated signals. The method consists basically in
the estimation of fractal dimension via the behavior of the signal Rheological properties of hydrolyzed starch were determined
length for increasingly less regularized version of the image. A frac- on starch dispersions, prepared by suspending the starch (15%,
tal image exhibits a power law for the lengths as a function of the w/w) in deionized water. The dispersions were gently stirred and
regularization degree. heated at 90 C for 20 min to allow complete gelatinization of the
starch granules. The swollen dispersions were then cooled down
2.5. X-ray diffraction (XRD) and kept in sealed containers. Dynamic oscillatory measurements
of the starch dispersions were carried out using a Physica MCR
Native and acid hydrolyzed corn starch particles subjected to 300 rheometer (Physica Metechnik GmbH, Stuttgart, Germany),
different hydrolysis times were stored in a sealed container at with a coneplate geometry, in which the rotating cone was 50 mm
a relative humidity of 85% for achieving constant moisture con- in diameter, and cone angle of 2 with a gap of 0.05 mm. About
tent. The X-ray powder diffractograms were measured in air at 1.25 mL of sample was carefully placed in the measuring system,
room temperature following the procedure of Hernndez-Nava, and left to rest for 10 min at 25 C for structure recovery. Ampli-
Bello-Perez, San Martin-Martinez, Hernandez-Sanchez, and Mora- tude sweeps were carried out in the range of 0.11000% at 25 C.
Escobedo (2011) with slight modications using a Bruker D-8 Temperature maintenance was achieved with Physica TEK 150P
Advance diffractometer with the BraggBrentano geometry, Cu temperature control and measuring system. The storage modulus
K radiation, a Ni 0.5% Cu-K lter in the secondary beam, and a (G ) and the loss modulus (G ) were obtained from the equipment
one-dimensional position-sensitive silicon strip detector (Bruker, software (US200/32 V2.50) in all cases. Flow curves of the starch
Lynxeye). The diffraction intensity as a function of 2 angle was dispersions were obtained by varying the shear rate from 0.001 to
598 R.G. Utrilla-Coello et al. / Carbohydrate Polymers 103 (2014) 596602
1000 s1 and the corresponding shear stress and apparent viscosity 100
values measured. Analysis was performed by triplicate.
90 Experimental Data
Residual Solids, R (%)

2.9. Statistical analyses 32.25 + 67.75exp(-t/5.5)
80
Data were analyzed using a one way analysis of variance 70

(ANOVA) and a Tukeys test for a statistical signicance P 0.05,
using the SPSS Statistics 19.0. All experiments were done in tripli- 60
cate.
50
3. Results and discussion 40
3.1. Hydrolysis kinetics 30

0 5 10 15
The hydrolysis kinetics during 15 days is presented in Fig. 1, Hydrolysis Time (days)
expressed in terms of residual unhydrolyzed solids, t. It is apparent
Fig. 1. Starch hydrolysis kinetics during 15 days. The experimental data can be
that the residual solids exhibited an exponential decaying pattern.
described by a rst-order kinetics model.
In this way, a simple rst-order kinetics model of the following
form was used for tting the experimental data: where R represents the residual solids, kH is the hydrolysis kinetics
dR constant and R is the residual solids for very long times. The results
= kH (R R) (2) are also presented in Fig. 1. The estimated model parameters are
dt
kH = 0.18 0.012 days1 and R = 32.25 2.50%.
Fig. 2. SEM images of native and hydrolyzed starch particles at different hydrolysis times: (a) native, (b) 12 h, (c) 1 day, (d) 5 days, (e) 7 days, (f) 9 days, (g) 15 days.
3.0 (a)
15 days
Intensity (a.u.)
7 days
Fractal Dimension, Df
2.9 5 days
3 days
2 days
2.8 1 days
0 days
10 15 20 25 30
2.7
Diffraction Angle, 2
44
(b)
Crystallinity, CIHW (%)

2.6 42
0.01 0.1 1 10
40
Hydrolysis Time (days)
38
Fig. 3. Regularization fractal dimension as a function of hydrolysis time. Increase of
the fractal dimension indicates increase of the surface roughness. 36
34
The extent of the corn hydrolysis achieved at day 15 (ca. of 32
63.0 3.0%) fell short relative to reported conversions of about 85% 0 3 6 9 12 15
at 40 C (Kim et al., 2012). Susceptibility of starches to hydrolysis Hydrolysis Time (days)
is strongly affected by temperature as the mobility and activ-
ity of the hydrogen ion is largely increased. However, hydrolysis Fig. 4. (a) XRD patterns of native corn starch and the hydrolyzed fractions at dif-
ferent times. Prominent intensity peaks at about 15.0 , 17.0 , 18.0 and 23.0 in 2,
may be also affected by other factors as well, such as the branch-
which are indicative of A-type crystallinity. (b) Evolution of crystallinity index CIHW
ing location of internal chains, i.e., the clustered versus scattered with the hydrolysis time.
branching structure (Espinosa-Solis, Sanchez-Ambriz, Hamaker, &
Bello-Prez, 2011; Hoover, 2000).
the relative fraction of the crystalline regions increased due to the
hydrolysis of the amorphous regions of the starch granules. How-
3.2. SEM study ever, a decrease in the peaks intensity occurred after fth day.
Although crystalline lamellae are more resistant to hydrolysis by
Native and acid treated starches obtained at different hydrolysis chemicals than the amorphous fractions, sulphuric acid disrupted
times were observed by SEM (Fig. 2). Native starch granules exhib- the crystalline structure after the depletion of the rings in the amor-
ited a regular, polygonal-like shape with smooth surface. After the phous regions. In principle, this feature should be reected in the
rst day of hydrolysis, only some starch granule presented slight XRD pattern shown in Fig. 4a. Fig. 4b presents the evolution of the
exo-erosion on the surface, gradually increasing by the third day, crystallinity index with the hydrolysis time. As already observed
but with most of the starch granules maintaining their polygonal in Fig. 4b, the crystallinity index CIHW achieved a maximum value
shapes. At longer hydrolysis times (5 days and thereafter), defor- (about a 42% increment) after three hydrolysis days. Afterwards,
mations on the surface of acid-treated granules were observed. the crystallinity index decreased to achieve values of about 38%.
Adhesion between some of the granules (days 5, 7 and 9) and The decreased crystallinity for long hydrolysis times suggests that
progressive surface erosion and fracture favored by stirring (Le the acid degraded mainly amylose and long amylopectin chains,
Corrre, Bras, & Dufresne, 2011) affected strongly the granule mor- leading to an increment in the proportion of less ordered short
phology. This pattern became more noticeable at the nal stage chains. This corroborates previous ndings establishing that, in a
of the hydrolysis time (15 days). The effect of acid hydrolysis on rst stage, acid hydrolysis affected predominantly the amorphous
the granule surface was quantied by computing the regularization regions, so that residual starch had more rened crystallinity. The
fractal dimension Df (Roueff & Vhel, 1998) based on SEM images results in Fig. 4b suggest that the crystallinity index can be used
of starch granules. The results shown in Fig. 3 indicate that the as suitable quantities for monitoring the evolution of starch crys-
fractal dimension was nearly constant (Df 2.65) during the rst tallinity during acidic hydrolysis for obtaining, e.g., starch granules
24 h. Afterwards, the fractal dimension exhibited a gradual increase with maximum crystallinity content.
with hydrolysis time, which is related to an increase of the surface
roughness. 3.4. Amylose content
3.3. XRD study The variation of the amylose content is described in Fig. 5a. A fast
decrease occurred during the rst 3 days, subsequently achieving
XRD patterns of native corn starch and the hydrolyzed fractions a constant value of about 2.94%. The amylose degradation behavior
at different times are shown in Fig. 4a. Results in Fig. 4a were dis- was described by rst-order kinetics with mean surviving time-
played for this range for better visualization of the intensity peaks constant of about A = 0.85 0.05 days. Starch granule swelling
characterizing the starch crystallinity. In this way, the dotted verti- allowed amylose leaking from the amorphous regions. Similar to
cal lines were used for highlighting four prominent intensity peaks gelatinized starch, once the amylose was in the aqueous phase,
at about 15.0 , 17.0 , 18.0 and 23.0 in 2, which were indicative of it was easily hydrolyzed by hydrogen ions. The residual amylose
A-type crystallinity. The peak at 18.0 was much more prominent (about 2.94%) is apparently related to the residual starch resistant
than the other peaks, and the peak at 23.0 was broader. These to acid hydrolysis. Note the difference between the mean survival
1
intensity peaks appeared in the native and in all of the hydrolyzed time constants between whole starch (H = kH = 5.5 0.35 days)
fractions. In the rst days of hydrolysis, the intensity of the peaks and amylose fraction (A = 0.85 0.05 days). This indicates that
gradually increased, while their width decreased, indicating that amylose was hydrolyzed during the rst 3 days, while amylopectin
35 100000
Amylose Content (%)
native
30 (a) 30 min
(a) 1000
1 day
25 3 days
5 days
7 days
20
app (Pa s)
12 days
Experimental Data 10 15 days
15 Exponential Decay
10
0.1
5
0
1E-3
0 2 4 6 8 10 12 14 1E-3 0.01 0.1 1 10 100 1000
Shear Rate (1/s)
Hydrolysis Time (Days)
10000
14 (b)
Enthalpy, H (J/g)
(b) 1000
app,max (Pa s)
12
100
Exponential Growth
10 10
8 1
0.1
6 0 3 6 9 12 15
28 30 32 34 36 38 40 42 44
Hydrolysis Time (days)
Crystallinity, CIHW (%)
Fig. 6. (a) Constant shear rate viscosity for different values of the hydrolysis time. A
Fig. 5. (a) Variation of the amylose content. A fast decrease is exhibited in the rst 3 shear thinning behavior is observed, with sharp decrease of viscosity after the rst
days. (b) Gelatinization enthalpy H as a function of the crystallinity content CIHW . hour of hydrolysis.
was more resistant to acid hydrolysis as the latter is located mainly shear thinning behavior due to two main factors: (1) the ocs
in the semi-crystalline regions of the starch granules. In fact, it is are deformed and become aligned with the shear eld, which
well-known that amylose disrupts the structural order within the decreases their resistance to ow, and (2) the ocs are disrupted
amylopectin crystallites (Jenkins & Donald, 1995). by shear forces, which decreases their effective volume fraction
(McClements, 2005). A dispersion containing occulated solids
3.5. DSC analysis tends to show a higher apparent viscosity than one than contains
the same concentration of unocculated solids. This is because the
The onset, peak, conclusion temperatures (To , Tp and Tc ), and effective volume fraction of a oc is greater than the sum of volume
enthalpy (H) of the phase transition for the native and acid fractions of the individual solids due to the presence of the con-
hydrolyzed starches for selected hydrolysis times, are shown in tinuous phase trapped within it (Dickinson & Stainsby, 1982). The
Table 1. The temperatures To and Tp decreased and the temperature rate at which the ocs in dispersions are deformed and disrupted
Tc increased during the rst day. The increase in Tc with the acid decreases as the number and the strength of the attractive inter-
hydrolysis was related with the removal of the amorphous regions actions between solids increases. Thus it may be inferred from the
of starch granules, due to the melting of the remaining crystalline curves in Fig. 6a that the native corn starch dispersion had a higher
regions at higher temperature, but without altering the content and effective volume fraction and attained stronger interaction forces
order of the double helices of amylopectin. The gelatinization tem- than the hydrolyzed starch dispersions. As hydrolysis time pro-
perature range Tc To increase in the rst hydrolysis day, indicating ceeded, and residual solids decayed, the effective volume fraction
that the crystallinity heterogeneity was broader for the hydrolyzed of the hydrolyzed starch dispersions decreased, and the apparent
granules. The changes in the gelatinization temperatures during the viscosity decreased. Fig. 6b depicts the drop in maximum apparent
rst hydrolysis day coincides with the fast decrement of the amy- viscosity as a function of hydrolysis time. The decrease in apparent
lose content (Fig. 5a), indicating that heterogeneity is linked to the viscosity was of about 99.9% in the rst 12 h, and is in line with the
amylose distribution within starch granule structure. In contrast, decrease of the amylose content (Fig. 5a). In fact, gelatinized amy-
the gelatinization enthalpy increased to achieve a maximum at the lose is the main contributor to the viscosity in starch dispersions. In
third hydrolysis day to decrease gradually afterwards. It is appar- particular, the reduction of short-chain amylose affects largely the
ent that the gelatinization enthalpy reected the average starch viscosity of starch dispersions (Zhang, Zhu, Shao, Gu, & Liu, 2012).
crystallinity given by the index CIHW (Fig. 4b). Fig. 5b presents In this way, it may be argued that the residual amylose content
the enthalpy H as a function of the crystallinity content CIHW . (Fig. 5a) is not leaked within the aqueous phase, but is contained
A positive correlation between these two parameters was appar- within the intrinsic crystalline structure. In turn, this makes the
ent, which can be described as an exponential behavior. In this way, residual amylose resistant to acid hydrolysis.
increased starch crystallinity was reected as increased gelatiniza-
tion enthalpy, so that high crystallinity values were also related to 3.7. Viscoelasticity of starch dispersions
improvements in the order of amylopectin molecules within the
double helices arrangement. Fig. 7a and b presents the storage G and the loss G modulus as
a function of the applied strain. For all hydrolysis times, the storage
3.6. Starch dispersions apparent viscosity modulus G exhibited a monotonous decreasing behavior with the
applied strain, indicating that the elasticity of the starch dispersions
The apparent viscosity-shear rate behavior of the starch dis- was reduced under shear action. For very short hydrolysis times (up
persions subjected to different hydrolysis times is presented in to about 1 h), the loss modulus G exhibited a Newtonian region for
Fig. 6a. All the starch dispersions showed typical behavior of oc- small strains and a small overshoot in the region of about 2060%
culated dispersions. Flocculated dispersions exhibit pronounced strain. Following the classication of complex uids (Hyun, Kim,
Table 1
Thermal parameters of native and hydrolyzed corn starch at different hydrolysis times.
Time To ( C) Tp ( C) Tc ( C) H (J/g) Tc To ( C)
Native 57.73 0.72b,c

72.16 0.26 e
81.13 0.92 a
7.65 0.31 a
22.91 0.25a
0.5 d 58.03 0.82c 71.59 0.38d,e 82.99 1.10a 7.54 0.31a 24.96 0.28a
1d 49.89 0.63a 65.24 0.59a 94.69 0.53b,c 9.69 0.66b 44.80 1.17c,d
3d 51.88 0.02a 70.37 0.10c,d 98.58 0.08d 11.35 0.08c 46.69 0.05d
5d 50.48 2.28a 66.91 0.76b 95.58 0.00c 13.68 0.36d 45.09 2.28c,d
7d 51.26 1.05a 67.11 0.16b 93.25 0.56b 9.47 0.00b 41.99 1.61b,c
12 d 55.52 0.49b,c 70.15 0.60c 95.21 0.73b,c 7.35 0.45a 39.69 0.24b
15 d 55.00 0.91b 74.63 0.25f 102.03 1.00e 9.93 0.40b 47.03 1.91d
Values are means standard error, of three replicates.

Superscripts with different letters in same column indicate signicant differences (P 0.05).
d = days; To , Tp , and Tc = onset, peak, and conclusion temperatures; H = enthalpy change, Tc To = gelatinization temperature range.
100
100
(a) (b)
10
10
1
G'' (Pa)
G' (Pa)
0.1
Native Native
15 min 15 min
30 min
0.01 1 day 1 30 min
1 day
3 days 3 days
5 days 5 days
1E-3 7 days 7 days
12 days 12 days
15 days 15 days
1E-4
0.1
0.1 1 10 100 1000 0.1 1 10 100 1000
Strain (%) Strain (%)
(c) (d)
100
10
Modulus (Pa)
G'
10
G''
tan()
1
1
0.1
0.01
0.1
0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14
Hydrolysis Time (days) Hydrolysis Time (days)
Fig. 7. (a) Storage G and (b) loss G moduli as a function of the applied strain. An overshoot in the loss modulus is observed for native weakly hydrolyzed corn starch
dispersions. (c) Decrease of the complex modulus as a function of the hydrolysis time. (d) Damping factor tan(), showing the loss of viscoelasticity with hydrolysis time.
Ahn, & Lee, 2002), starch dispersions were characterized by Type dispersions had a liquid-like behavior (Wang et al., 2003). This
III microstructure. The overshoot of the loss modulus was caused by shows that important viscoelasticity reductions for starch disper-
molecule shear-induced aggregation, forming a weakly structured sions were obtained with relatively short hydrolysis times of the
material. This complex structure was destroyed by the applica- order of 1 h.
tion of large deformations over the critical strain, after which the
biomolecules chains aligned with the ow eld and the loss mod- 4. Conclusions
ulus decreased. However, the loss modulus overshoot disappeared
after 1 h of hydrolysis time. The dispersions aggregates responsi- In this work a detailed characterization of acid hydrolysis of
ble of the loss modulus overshoot were induced by gelatinized native corn starch for short and long time horizons is given. Differ-
amylose molecules. Starch granules with low amylose content pro- ent techniques were used to this end, including XRD, DSC, amylose
duced weak gels (shear thinning) for all strain range after the rst content determination, fractal analysis of SEM images and rheology
hour of hydrolysis. of starch dispersions. Results showed that amylose content played
For 1% of strain, Fig. 7c presents the complex modulus G and an important role in the hydrolyzed starch characteristics for short
G as a function of the hydrolysis time. Both moduli decreased time scales. In contrast, disruption of the starch granule integrity
sharply in the rst hydrolysis hours, reecting the depletion of amy- was determinant in the long term, affecting mainly crystallinity and
lose content. The corresponding damping factor tan() = G /G , thermal properties. The results showed that hydrolysis is a exi-
an index of the gel uidization, is presented in Fig. 7d. After the ble process for starch treatment under desired characteristics. In
rst hour, the damping factor tan() > 1, indicating that the starch this way, the production of shear thinning starch with relatively
low hydrolysis advance required only few hours. In contrast, the Hyun, K., Kim, S. H., Ahn, K. H., & Lee, S. J. (2002). Large amplitude oscillatory shear as
production of starch granules with high crystallinity required mod- a way to classify the complex uids. Journal of Non-Newtonian Fluid Mechanics,
107, 5165.
erate hydrolysis advance and longer times of the order of 23 days. Jenkins, P. J., & Donald, A. M. (1995). The inuence of amylose on starch granule
Overall, the results in this work showed that acid hydrolysis is a structure. International Journal of Biological Macromolecules, 17, 315321.
complex process underlying starch microstructure changes in time Kim, H. Y., Lee, J. H., Kim, J. Y., Lim, W. J., & Lim, S. T. (2012). Characterization of
nanoparticles prepared by acid hydrolysis of various starches. Starch/Strke, 64,
scales from hours to days. 367373.
Le Corrre, D., Bras, J., & Dufresne, A. (2011). Evidence of micro- and nanoscaled
particles during starch nanocrystals preparation and their isolation. Biomacro-
Acknowledgements
molecules, 12, 30393046.
Lu, T. J., Duh, C. S., Lin, J. H., & Chang, Y. H. (2008). Effect of granular characteristics
The authors want to thank CONACyT-Mxico for providing on the viscoelastic properties of composites of amylose and waxy starches. Food
Hydrocolloids, 22, 164173.
nancial support under project INFR-2011-1-163250. Also, special
McClements, D. J. (2005). Food emulsions: Principles, practices, and techniques (2nd
thanks to LDRX (T-128) UAM-I for XRD measurements. ed.). Florida: CRC Press.
Nehir El, S., & Simsek, S. (2012). Food technological applications for optimal nutri-
tion: An overview of opportunities for the food industry. Comprehensive Reviews
References in Food Science and Food Safety, 11, 212.
Ortega-Ojeda, F. E., Larsson, H., & Eliasson, A. C. (2004). Gel formation in mixtures
BeMiller, J. N., & Whistler, R. L. (2009). Starch: Chemistry and technology (3rd ed.). of high amylopectin potato starch and potato starch. Carbohydrate Polymers, 56,
New York: Elsevier. 505514.
Bjrck, I., & Asp, N. G. (1994). Controlling the nutritional properties of starch in foods Palma-Rodriguez, H. M., Agama-Acevedo, E., Mendez-Montealvo, G., Gonzalez-Soto,
A challenge to the food industry. Trends in Food Science and Technology, 5(7), R. A., Vernon-Carter, E. J., & Bello-Prez, L. A. (2012). Effect of acid treatment on
213218. the physicochemical and structural characteristics of starches from different
Dickinson, E., & Stainsby, G. (1982). Colloids in foods. London: Applied Science Pub- botanical sources. Starch/Strke, 64, 115125.
lishers. Pang, J., Wang, S., Yu, J., Liu, H., Yu, J., & Gao, W. (2007). Comparative studies on
Espinosa-Solis, V., Sanchez-Ambriz, S. L., Hamaker, B. R., & Bello-Perez, L. A. (2011). morphological and crystalline properties of B-type and C-type starches by acid
Fine structural characteristics related to digestion properties of acid-treated hydrolysis. Food Chemistry, 105, 989995.
fruit starches. Starch/Strke, 63, 717727. Roueff, F., & Vhel, J. L. (1998). A regularization approach to fractional dimension
Genovese, D. B., & Rao, M. A. (2003). Role of starch granule characteristics (volume estimation. Fractals, 98, 113.
fraction, rigidity, and fractal dimension) on rheology of starch dispersions with Singh, H., Singh Sodhi, N., & Singh, N. (2009). Structure and functional proper-
and without amylose. Cereal Chemistry, 80, 350355. ties of acid thinned sorghum starch. International Journal of Food Properties, 12,
Hermans, P. H., & Weidinger, A. (1948). Quantitative X-ray investigations on the 713725.
crystallinity of cellulose bers. A background analysis. Journal of Applied Physics, Singh Sandhu, K., Singh, N., & Lim, S. T. (2007). A comparison of native and acid
19, 491506. thinned normal and waxy corn starches: Physicochemical, thermal, morpholog-
Hernndez-Nava, R. G., Bello-Prez, L. A., San Martn-Martnez, E., Hernndez- ical and pasting properties. LWT Food Science and Technology, 40, 15271536.
Snchez, H., & Mora-Escobedo, R. (2011). Effect of extrusion cooking on Wang, Y. J., Truong, V. D., & Wang, L. (2003). Structures and rheological properties of
the functional properties and starch components of lentil/banana blends: corn starch as affected by acid hydrolysis. Carbohydrate Polymers, 52, 327333.
Response surface analysis. Revista Mexicana de Ingeniera Qumica, 10, Wang, S. J., Gao, W. Y., Yu, J. L., & Xiao, P. G. (2006). The crystalline changes of
409419. starch from Rhizoma Dioscorea by acid hydrolysis. Chinese Chemical Letters, 17,
Hoover, R. (2000). Acid-treated starches. Food Reviews International, 16, 369392. 12551258.
Hoover, R., & Ratnayake, W. S. (2002). Starch characteristics of black bean, chick pea, Zhang, C., Zhu, L., Shao, K., Gu, M., & Liu, Q. (2012). Toward underlying reasons for rice
lentil, navy bean and pinto bean cultivars grown in Canada. Food Chemistry, 78, starches having low viscosity and high amylose: Physiochemical and structural
489498. characteristics. Journal of the Science of Food and Agriculture, 93, 15431551.
Hydrolysis of concentrated raw starch: A new very efcient -amylase from

Anoxybacillus avothermus
Georges Tawil a , Anders Viks-Nielsen b , Agns Rolland-Sabat a , Paul Colonna a , Alain Bulon a,
a
Unit Biopolymres, Interactions, Assemblages, Institut National de la Recherche Agronomique, Rue de la Graudire, BP 71627, F-44316 Nantes Cedex 3, France
b
Novozymes A/S, Krogshoejvej 36, 2880 Bagsvaerd, Denmark
Article history: A new -amylase from Anoxybacillus avothermus (AFA) was found to be effective in hydrolyzing raw
Received 13 May 2011 starch in production of glucose syrup at temperatures below the starch gelatinization temperature. AFA
Received in revised form 4 July 2011 is very efcient, leading to 77% hydrolysis of a 31% raw starch suspension. The nal hydrolysis degree
Accepted 6 July 2011
is reached in 23 h at starch concentrations lower than 15% and 824 h at higher concentrations. AFA
Available online 22 July 2011
is also very efcient in hydrolyzing the crystalline domains in the starch granule. The major A-type
crystalline structure is more rapidly degraded than amorphous domains in agreement with the observed
Keywords:
preferential hydrolysis of amylopectin. Amyloselipid complexes are degraded in a second step, yielding
-Amylase
Maize starch
amylose fragments which then re-associate into B-type crystalline structures forming the nal -amylase
Crystallinity resistant fraction. The mode of action of AFA and the factors limiting complete hydrolysis are discussed
Amyloselipid complex in details.
Molar mass distribution 2011 Elsevier Ltd. All rights reserved.
Resistant fraction
Oligosaccharides
1. Introduction ule, the amylose content, the crystalline structure or the presence
of amylose-lipid complexes were shown to be limiting factors to
Starch and its two main constituents, amylose and amylopectin, hydrolysis of the starch granule.
are degraded by -amylases which are the major enzymes involved While acid hydrolysis degrades preferentially the amorphous
in the hydrolysis of (14) glycosidic bonds. -Amylases are very parts of the starch granule, -amylases can solubilize both amor-
important in biological reactions such as fermentation, germination phous and crystalline domains (Colonna, Buleon, & Lemarie, 1988;
or digestion, but also widely used in e.g. the industry for production Gerard et al., 2001). The mechanisms involved in the hydrolysis of
of glucose syrups, control of anti-staling in bread products or in the crystalline domains and especially the disruption of the dou-
detergents to remove starch based stains. ble helices from the crystallite and their disentanglement are not
As native starch is water insoluble at room temperature, many well known. As double helices are too wide to enter the catalytic
applications of amylases are carried out at high temperature and site of -amylases (Andr, Buleon, Haser, & Tran, 1999), their dis-
pressure where the starch is gelatinized. During gelatinization, the entanglement is speculated to occur during the adsorption stage.
granular architecture and the molecular order (double helices) of Adsorption of amylase onto starch granule was found to be a pre-
the starch granule are disrupted. This change in the physical state is requisite for hydrolysis, but it can be inhibited by the presence of
known to increase the susceptibility of starch to enzymatic hydrol- oligosaccharides such as maltose or maltotriose (Leloup, Colonna, &
ysis (Lauro, Suortti, Autio, Linko, & Poutanen, 1993). Ring, 1991). We recently studied a highly efcient fungal -amylase
By contrast, the hydrolysis of solid starchy substrates strongly from Rhizomucor sp. (RA) optimized for biofuel production, and able
depends on starch structure and amylase source (Buleon, Colonna, to degrade preferentially the crystalline fraction of maize starch
Planchot, & Ball, 1998; Colonna, Leloup, & Buleon, 1992; Gallant, (Tawil, Viks-Nielsen, Rolland-Sabat, Colonna, & Buleon, 2011).
Bouchet, Buleon, & Perez, 1992; Gerard, Colonna, Buleon, & For this enzyme the slower degradation of amylose was attributed
Planchot, 2001; Gernat, Radosta, Anger, & Damaschun, 1993; Oates, to the presence of lipid complexed amylose in the rst stages of
1997; Williamson, Belshaw, Self, Noel, Ring, Cairns, Morris, Clark hydrolysis and then in the nal stage to recrystallization of amylose
& Parker, 1992). The morphology and the surface of the gran- fragments released by action of the amylase. Amylose-lipid com-
plexes have proved to be present in cereal starches which naturally
contain lipids (Morrison, Tester, Gidley, & Karkalas, 1993; Morrison,
Corresponding author. Tel.: +33 2 40 67 50 47; fax: +33 2 40 67 50 43. Tester, Snape, Law, & Gidley, 1993) but can also form when heat-
E-mail address: buleon@nantes.inra.fr (A. Bulon). ing lipid containing starch in presence of water (Biliaderis, 1992;
doi:10.1016/j.carbpol.2011.07.005
G. Tawil et al. / Carbohydrate Polymers 87 (2012) 4652 47
Buleon & Colonna, 2007; Le Bail, Bizot, Ollivon, Keller, Bourgaux, & 2.2.3. Sample preparation for high-performance size-exclusion
Buleon, 1999). In the Vh-type, the most common crystalline form chromatography coupled with multiangle laser light scattering
obtained by complexation of amylose with lipids, the single helices and differential refractometric detection (HPSEC-MALLS-DRI)
are packed in an orthorhombic unit cell with 16 water molecules Samples (10 mg) were pretreated in Me2 SO/H2 O (90/10 v/v),
within the unit cell (Rappenecker and Zugenmaier, 1981). The precipitated in 80% ethanol and dried. They were then solubilized
amyloselipid complexes have also been shown to be more resis- in water by microwave heating under pressure (Rolland-Sabat,
tant to -amylolysis than the A-type native structure present in Amani, Dufour, Guilois, & Colonna, 2003). The resulting solutions
cereal starches (Gernat et al., 1993; Tawil et al., 2011). However, were ltered through 5 m DuraporeTM membranes. Carbohy-
hydrolysis is completed when an excess amount of enzyme or drate concentration was determined by the orcinol sulfuric method
a longer hydrolysis time is applied (Biliaderis & Galloway, 1989; (Planchot, Colonna, & Buleon, 1997). Sample recovery rates were
Holm et al., 1983). calculated from the ratio of the initial concentration to the nal
In this study, we focus on a new bacterial -amylase from Anoxy- concentration in solution.
bacillus avothermus (AFA) found to be very efcient in hydrolysis
of concentrated raw starch. This bacterial amylase contains a Starch 2.3. Kinetics of -amylolysis in heterogenous phase
Binding Domain (SBD) belonging to the CBM20 family and has
been evaluated for use in low temperature glucose syrup produc- -Amylases from Porcine Pancreas (PPA) and A. avothermus
tion (Viks-Nielsen, Andersen, Hoff, & Pedersen, 2006). The use of (AFA) were used at 37 C, phosphate buffer pH 7 for PPA and 61 C,
this enzyme compared to traditional high temperature -amylases acetate buffer pH 4.5 for AFA (resembling relevant application con-
would require less energy and water than in the traditional liq- ditions). The following three starch concentrations were used: 5,
uefaction process. Its mode of action was studied in comparison to 15 and 31% dry basis (d.b.), respectively. The amount of AFA and
porcine pancreatic -amylase (PPA) which has been widely studied PPA was dosed at the same activity, i.e. 15.5 U per mg dry starch.
for digestion of starch products and as model enzyme for hydrolysis Thus, 235, 710, and 1466 L of AFA and 1.8, 5.5 and 11.4 mL of PPA
of raw starch. AFA specicity was assessed by monitoring the evolu- were used for 5, 15 and 31% starch respectively, the nal volume
tion of starch structure especially morphology, crystalline structure being adjusted to 20 mL with buffer. The suspension was shaken
and molar mass distribution during hydrolysis of maize starch at continuously in a water bath at 200 rpm and 2 mL aliquots were
various concentrations. The limiting factors for a complete hydrol- withdrawn at different time intervals (1, 2, 8, 24, 48, 72 and 96 h).
ysis were also investigated. For each aliquot, the reaction was stopped by adding 80 L of 1 M
KOH and then centrifuged at 4167 g at 4 C for 10 min. The pre-
2. Experimental cipitate was used for structural analysis. Total solubilized sugars
were measured in the supernatant by the orcinol sulfuric method
2.1. Materials (Planchot et al., 1997), and the extent of hydrolysis was expressed
as the ratio of soluble sugars from starch hydrolysis to the initial
2.1.1. Substrates mass of starch (d.b.).
Normal maize starch was from Cerestar (Cargill Vilvoorde,
Belgium). 2.4. Analysis of the residual starch
2.1.2. Enzymes 2.4.1. Crystalline structure

Puried preparation of AFA (Viks-Nielsen et al., 2006) was X-ray diffraction (XRD) analysis was performed on native starch
provided by Novozymes, Denmark. Crystallized and lyophilized and residual starch withdrawn at different time intervals during
Porcine Pancreatic -amylase and Proteinase K (from Tritirachium hydrolysis. The water content of samples was adjusted by water
album) were purchased from Sigma Chemical Company (St. Louis, phase sorption for 10 days in desiccators under partial vacuum at a
MO). All other reagents were of analytical grade. relative humidity of 90% (using a saturated salt solution of baryum
chloride). Hydrated samples (20 mg) were then sealed between
2.2. Sample preparation two tape foils to prevent any signicant change in water con-
tent during the measurement. XRD diagrams were recorded on
2.2.1. Enzymes a BRUKERTM (Wissembourg. France) D8 Discover diffractometer.
PPA was solubilized in 20 mM phosphate buffer, pH 6.57.0, Cu K1 radiation ( = 0.15405 nm), produced in a sealed tube at
containing 2 mM NaCl, 0.25 mM CaCl2 and 0.2 g L1 NaN3 . The solu- 40 kV and 40 mA, was selected using a Gobl mirror parallel optics
tion was centrifuged at 4167 g and the supernatant was used for system and collimated to produce a beam of 500 m diameter.
hydrolysis after protein content determination. AFA was provided The diffracted beam was collected with a two-dimensional GADDS
as 4 mg mL1 solution in sodium acetate buffer. The concentra- detector and recording time was 600 s. The distance from the sam-
tion was determined by A280 and calculated using a molecular ple to the detector was 100 mm. After normalization of all recorded
absorbance of 2.728 of the puried AFA, i.e. absorbance at 280 nm diagrams at the same integrated scattering between 3 and 30
of a 1 mg mL1 AFA solution. (2), relative crystallinity was determined as described previously
(Maache-Rezzoug, Zarguili, Loisel, Queveau, & Bulon, 2008).
2.2.2. Measurement of the enzyme activity
Enzyme activity was determined according to the Ceralpha 2.4.2. Molar mass distribution
procedure (McCleary et al., 2002). It consists to hydrolyze non The molar mass distribution of residues at different hydroly-
reducing-end blocked p-nitrophenyl maltoheptaoside (BPNPG7) by sis times was determined using high-performance size-exclusion
-amylase, in the presence of excess -glucosidase. The amount of chromatography (HPSEC) with multiple angle laser light scatter-
released para-nitrophenol PNP was determined spectrophotomet- ing (MALLS) detection (Rolland-Sabat, Guilois, Jaillais, & Colonna,
rically at a wavelength of 400 nm. One unit of activity (U) is dened 2011). Sample recovery was calculated from the ratio of the mass
as the amount of amylase required to release 1 mole of PNP from eluted from the column (integration of the refractometric signal)
BPNPG7 in one min at 40 C, pH 7. Specic activities of AFA and PPA to the injected mass which was determined using the orcinol sul-
were 17 U and 3.5 U per mg of protein, respectively. phuric method (Planchot et al., 1997).
48 G. Tawil et al. / Carbohydrate Polymers 87 (2012) 4652
2.5. Analysis of the soluble products 100
The composition of soluble products released by enzymatic 90

hydrolysis was determined as previously described (Tawil et al., 80
2011) by using high-performance anion-exchange chromatogra-
% of Hydrolysis
phy (HPAEC) with a pulsed amperometric detector (PAD) system 70
(Pohu, Putaux, Planchot, Colonna, & Buleon, 2004). A detector
60
response correction was performed for quantitative analysis using
gluco-oligosaccharides from DP1 to DP7 (Sigma, Chemical Com- 35
pany). 30 PPA
25
2.6. Determination of the limiting factors for a complete 20
hydrolysis 15
10
Potential inhibition of amylases by the end products was 5
checked using washing of residual starch and a new 96 h hydrol- 0
ysis after adding a new dosage of amylase solution as described 0 20 40 60 80 100
previously (Tawil et al., 2011). Time (h)
The concentration of amylase in the supernatant at the end of
this new hydrolysis was determined according to the Bradfords Fig. 1. Effect of raw starch concentration on AFA hydrolysis kinetics: 31% d.b. ()
procedure (Bradford, 1976) and the amount of adsorbed protein 15% d.b. (), 5% d.b. (). Data obtained for PPA (+) on 31% starch are shown for
comparison.
determined as C = ((C0 C)/C0 ) * 100 with C0 being the initial con-
centration. To check if the amylase adsorbed onto residual starch
and/or aggregated at this stage was responsible for the hydrolysis 3.2. Hydrolysis of the crystalline structure
rate decline, any starch bound enzyme was hydrolyzed by pro-
teinase K. Thus, residual starch was washed one time in water and The crystalline structure was essentially assessed by X-ray
four times in TrisHCL. Aliquot of proteinase K (0.015 mg per mg diffraction (XRD). Degree of crystallinity and crystalline type of
of starch) was then added onto each residual starch sample for residual starch are summarized, with the corresponding hydrol-
30 min. Proteinase K was removed by washing the residue 5 times ysis degree, in Table 2 for 5 and 31% starch suspensions at different
with acetate buffer for AFA and with a phosphate buffer for PPA times of hydrolysis. Native maize starch has A-type crystallinity
followed by centrifugation at 4167 g. Subsequent hydrolysis was with characteristic peaks at Bragg angle (2) = 15, 17, 18 and 23
conducted as described above. (Fig. 2). The crystallinity decreases from 31% to 20% during the rst
2 h of hydrolysis of the 31% starch suspensions by AFA (Table 2).
3. Results and discussion The hydrolysis of crystalline domains is less pronounced in the 5%
starch suspensions since a slightly lower decrease of crystallinity
3.1. Effect of raw starch concentration on the hydrolysis kinetics is observed while the hydrolysis extent is signicantly higher (82%
versus 60%). This ability to hydrolyse the crystalline domains at
Fig. 1 shows the evolution of hydrolysis kinetics for AFA with high starch concentrations is a unique feature of AFA, as for RA
increasing starch concentration from 5 to 31% d.b. and Table 1 (Tawil et al., 2011), compared to more classical -amylases as PPA
summarizes the corresponding values determined for AFA, and for which the degree of crystallinity is kept constant all along the
PPA after 96 h hydrolysis. No matter at what starch concentration hydrolysis of 31% starch suspensions.
tested, AFA is very effective on raw starch and surprisingly active on The intensity of the peak at 2 around 20 , which corresponds
the 31% starch suspension since the nal hydrolysis degree (FHD) to amyloselipid complexes formed between amylose and endoge-
reaches 77% for AFA, versus 32% for PPA after 96 h (Fig. 1 and neous fatty acids present in maize starch (Vh-type), is remarkably
Table 1). The hydrolysis degree decreases slightly when increas- stable at 31% starch. It develops more at 5% starch when the hydrol-
ing the starch concentration with the FHD decreasing from 88 to ysis extent reaches more than 80% and is also present in the 96 h
77% for 5 and 31% starch, respectively. All hydrolysis curves have a PPA residue obtained from 5% starch suspensions. Thus it appears
classical two phase shape but with a very rapid and short rst stage that the Vh-type structure is much more resistant to -amylases
which does not exceed 23 h for 5% and 8 h for 15% and 31% starch than the A-type, as already described by Gernat et al. (1993).
suspensions respectively (Table 1). Hydrolysis kinetics is much less
dependent on starch concentration than for PPA. Even it is a little 3.3. Changes in the macromolecular structure of starch during
smaller than the values observed for RA (Tawil et al., 2011), the hydrolysis
very high rate of hydrolysis and its extent at 28 h make AFA very
interesting for rapid production of short chain maltodextrins from The HPSEC-DRI traces of raw starch as well as AFA and PPA resid-
raw maize starch. ual starch at 96 h (initial starch concentration 31%) are reported
Table 1
Kinetic data as a function of the starch concentration for AFA (and PPA in brackets).
Time (h) 5% db 15% db 31% db
1 78. 1 (25.0 1.8) 67.4 2.5 (22.0 2.6) 51.8 1.4 (19.0 8.0)
2 82.7 0.8 (35.0 15.0) 70.1 3.6 (29.0 4.4) 60.0 3.0 (23.0 2.8)
8 82.2 4.0 (50.0 3.6) 71.0 5.7 (34.0 3.0) 70.6 4.0 (27.0 3.4)
24 80.7 1.4 (66.0 3.8) 75.7 4.6 (45.0 8.0) 73.1 4.0 (18.4 1.6)
48 82.8 1.7 (78.0 9.6) 79.0 0.4 (48.0 1.4) 74.7 2.0 (32.0 1.2)
72 82.0 3. (84.0 6.4) 79.0 0.4 (54.0 1.6) 74.6 2.0 (31.5 0.8)
96 (FHD) 87.9 1.8 (89.0 10.6) 79.0 0.4 (67.0 8.0) 76.6 1.0 (32.0 2.8)
Diffracted
Intensity Vh-type
1
96 h (77%)
48 h (75%) 2
Relative scale
3
8 h (71%)
2 h (60%)
A-type
Native starch 4
5
0 5 10 15 20 25 30 35
4 4,5 5 5,5 6 6,5 7 7,5 8
Bragg angle (2)
volume (mL)
Fig. 2. Evolution of the crystalline structure as a function of time and hydrolysis
degree (in brackets) during hydrolysis of a 31% starch suspension by AFA. Fig. 3. HPSEC-DRI traces of residual starch after hydrolysis of 31% maize starch
suspensions by AFA and PPA. 1: native starch (control); 23: residual starch obtained
after hydrolysis with PPA at 48 h and 96 h respectively; 45: residual starch obtained
in Fig. 3. Raw starch presents two peaks, which are attributed to after hydrolysis with AFA at 48 h and 96 h respectively. Relative scale was obtained
amylopectin (elution volume around 5.7 mL) and amylose (elution by dividing each initial normalized prole (normalized refractometric response) by
the amount of residual starch.
volume around 6.6 mL) according to previous work (Rolland-Sabat
et al., 2003). Amylopectin weight average molar mass decreases
from 3.08 108 to 2.35 108 g mol1 after 48 h. Structural infor- of the corresponding DRI peak being 2 times less after 96 h, decreas-
mation could be obtained by plotting the radius of gyration versus ing from 5.65 106 to 2.58 106 g mol1 (Table 3). Nevertheless,
molar mass from the exponent G according to the empirical these latter values do not represent the absolute molar mass of
power law: RG MG . The radius of gyration decreases with hydrol- amylose population since amylose and amylopectin peaks are not
ysis and G , which was calculated on the whole amylopectin fully separated, and then low molar mass amylopectin fractions
fraction, increases from 0.38 to 0.41 after 96 h of hydrolysis. G elute at the same elution volume as amylose ones. Moreover, a part
value depends on the polymer shape, the temperature, and the of the amylose peak could involve partially hydrolyzed amylopectin
polymersolvent interactions: G = 0.33 for a sphere; G = 0.50.6 eluting at the same volume. These results are in agreement with
for a linear random coil, and G = 1 for a rod. The values determined the observed large decrease of crystallinity during hydrolysis, since
here show that the amylopectin remaining fraction is slightly amylopectin is usually assumed to support the framework of the
less dense in solution than native amylopectin and then probably crystalline domains in the starch granule. In any case the behavior
slightly less branched. When looking at the chromatograms (Fig. 3) is different from that of PPA for which residual starch present the
the amount of amylopectin has been more intensively reduced same bimodal shape as for native starch (Fig. 3), with no decrease of
than that of amylose which is consistent with the decrease of crys- amylopectin weight average molar mass which is consistent with
tallinity observed during hydrolysis. At the same time amylose is a granule per granule action mode as already proposed (Colonna
highly hydrolyzed, the weight average molar mass taken at the apex et al., 1988).
Table 2
Evolution of the crystalline structure CS (degree of crystallinity in %, crystalline type A, B or Vh) and degree of hydrolysis HD (%) of maize starch during hydrolysis by AFA and
PPA.
Native starch Hydrolyzed 2 h Hydrolyzed 8 h Hydrolyzed 48 h Hydrolyzed 96 h
CS CS HD CS HD CS HD CS HD
AFA 5% starch 31, A 22, A + Vh 82 15, A + Vh 82 14, A + Vh 83 14, A + Vh 87

AFA 31% starch 31, A 20, A 60 20, A 71 15, A 75 15, A + Vh 77
PPA 5% starch 31, A 30, A 34 25, A 49 20, A 78 20, A + Vh 89
PPA 31% starch 31, A 30, A 23 30, A 27 30, A 32 30, A 32
Table 3
Weight average molar mass (Mw ), z-average radius of gyration (RGZ ) and the slope of the log/log plot of molar mass versus radius of gyration (G ) of amylose and amylopectin
after hydrolysis of maize starch (31% d.b.) by AFA; nd: not determined; (*) values obtained over the whole amylopectin peak; (+) values at the maximum of amylose peak;
the experimental uncertainty was 5%.
Amylopectin Amylose Amylose/amylopectin ratio

1 * 1 +
M w (g mol ) RGZ (nm) *
G M w (g mol )
Native starch 3.08 108 270.0 0.38 5.65 106 0.5

AFA 48 h 2.38 108 236.0 0.44 1.92 106 nd
AFA 96 h 2.35 108 240.0 0.41 2.58 106 nd
Table 4
DP1 to DP7 composition in the soluble fraction during hydrolysis of 31% starch suspensions by AFA.
Time (h) DP1 (%) DP2 (%) DP3 (%) DP4 (%) DP5 (%) DP6 (%) DP7 (%)
2 14.1 34.2 16.5 2.9 17.6 9.1 5.6

AFA 8 16.7 34.5 14.1 4.1 17.9 8.8 3.8
48 19.0 35.3 9.5 5.6 18.7 8.2 3.7
3.4. Nature of the soluble oligosaccharides produced. et al., 1993; Kwasniewska-Karolak, Nebesny, & Rosicka-Kaczmarek,
2008; Lauro, Forssell, Suortti, Hulleman, & Poutanen, 1999).
The distribution of oligosaccharides in the soluble fraction after The B-type structure can originate from rearrangements of
2, 8 and 48 h hydrolysis using AFA on 31% starch suspensions linear amylose-like fragments released by the enzyme. Such a rear-
is shown in Table 4. Only the values obtained for DP (degree rangement into B-type is favoured at higher starch concentrations.
of polymerization) <8 are shown, since the HPAEC response is B-type structure is known to result from retrogradation of starch
not quantitative above DP 7. The distribution is similar to the (Buleon & Colonna, 2007) or crystallization of short chain amy-
usual distribution obtained from hydrolysis by classical -amylase lose in water (Buleon, Veronese, & Putaux, 2007). B-type structures
as PPA (Pohu et al., 2004; Sharma, Yadav, & Ritika, 2008). The have proved to be very resistant to hydrolysis (Colonna et al.,
major oligosaccharide present in the soluble fraction is maltose 1992; Planchot et al., 1997). Such rearrangement during hydroly-
(DP2), which is known to induce a potential inhibition of amy- sis has already been described by Lopez-Rubio, Flanagan, Shrestha,
lases (Colonna et al., 1992; Faisant et al., 1993). The composition Gidley, and Gilbert (2008) when looking at hydrolysis of high amy-
changes slightly between 2 and 48 h (Table 4). The ratio DP1/DP3 lose starch. The B-type structure formed prevents any progress of
and DP4/DP7 increases, evidencing a partial hydrolysis of DP3 into hydrolysis, once the Vh-type is degraded.
DP1 and DP7 into DP4 and DP3.
4. General discussion
3.5. Limiting factors for complete hydrolysis
AFA is highly effective in hydrolysis of raw starch suspensions,
The relative activity of AFA as a function of time under the buffer being able to degrade 83% of a 5% and 60% of a 31% starch suspen-
and temperature conditions used for hydrolysis was determined. sion respectively, within 2 h at 61 C. The hydrolysis degree of a 31%
Under these conditions AFA loses its activity very rapidly and within starch suspension reaches 7577% within 24 h. This very high ef-
24 h. It could explain why an incomplete hydrolysis of starch ciency at high starch concentration is remarkable. The hydrolysis
was observed using this enzyme. Nevertheless, as the amount of degree observed on raw starch is very comparable to that deter-
hydrolyzed product slightly increases from 2 h to 8 h of hydrolysis mined for RA (Tawil et al., 2011) but is reached more rapidly, 70%
at 5% starch and from 2 h to 24 h at higher starch concentrations, it for AFA versus 30% for RA after 8 h on a 31% starch suspension.
was assumed that the substrate has a protective effect against AFA This makes AFA more suitable for low temperature glucose syrup
denaturation. A similar observation has been already reported in production. RA reaches similar extents of hydrolysis in 48 h but
the literature (De Cordt, Hendrickx, Maesmans, & Tobback, 1994;
Gorinstein, 1993). Starch is known to stabilize -amylase activity
at higher dry weight concentrations.
V type
The degree of hydrolysis increases from 87 to 90% and from 77
to 82% for the 5% and 31% starch suspensions, respectively, after B type
washing the residue and adding a fresh dosage of AFA. It is difcult
to attribute this change to a slight inhibition by reaction products
since AFA also lost its activity during the rst hydrolysis.
No further hydrolysis was observed after removal of the
Diffracted
adsorbed or aggregated AFA by proteinase K and adding a new
Intensity c
amylase solution, i.e. indicating that no unproductive binding or
aggregation of AFA prevents a complete hydrolysis, although less
than 10% of enzyme was still present in the supernatant at the end
of hydrolysis.
Thus, it was concluded that hydrolysis ends due to the pres-
ence of a residue resistant to AFA. The crystalline structure of this b
residue was analyzed by X-ray diffraction on the samples that
were most hydrolyzed: 90% and 82% originating from 5 and 31%
starch suspensions, respectively (after washing the residue and
adding and fresh AFA as described above). As shown in Fig. 4,
a mixture of A- + Vh-type is found in the 5% starch residue and
a
almost pure B-type, with characteristic peaks at 2 around 5.6, 17
and 24 , in the 31% starch residue hydrolyzed by AFA. It demon-
A type
strates that the mode of action of AFA is not the same on a 5
versus a 31% starch suspension. The resistance of the Vh-type at
5% starch could be due to improvement of the Vh crystals per- 5 10 15 20 25 30
fection by annealing at 61 C. While at 31% starch, less water is
2 thetas
available and annealing is less important. This explains the higher
susceptibility of the Vh-type at 31% starch in comparison to 5%
Fig. 4. Crystalline structure of (a) native maize starch (A type), (b) residual starch
starch. The higher resistance to amylolysis of Vh-type when com- at 192 h hydrolysis of 31% starch by AFA (FHD 83%, B type) and (c) residual starch at
pared to A-type in cereal starches has been observed earlier (Gernat 192 h hydrolysis of 5% starch by AFA (FHD 94%, A + Vh type).
was more suitable for bioethanol production since it works very amylose into B-type is also a major way to prepare highly resistant
efciently at 32 C and releases essentially glucose (Tawil et al., starch (Buleon & Colonna, 2007; Leloup, Colonna, & Buleon, 1991).
2011). The very high efciency of AFA for hydrolysis of raw starch,
The mode of action of AFA is very specic. Its efciency is almost compared to that of PPA, is evidently due to the presence of a par-
independent on starch concentration when compared to other - ticularly important starch binding domain (CBM20) in this enzyme
amylases. Its behavior differs with changes in starch concentration (Viks-Nielsen et al., 2006). Such domain has been reported to facil-
since the residue from 5% starch consists of a mixture of resid- itate adsorption of amylases on raw starch, the disentanglement of
ual A-type structure and a large amount of crystalline Vh-type the double helices present in the crystalline lamellae and therefore
amyloselipid complexes while at 31% it consists of pure B-type the disruption of the starch structure results in a faster breakdown
structure. Therefore it seems that amylose-lipid complexes are of the starch granules (Juge et al., 2006; Christiansen et al., 2009).
more resistant to enzymatic attack at low starch concentration and
that crystalline structures are more rapidly degraded at high starch
concentration (Table 2). It differs from cellulases which usually 5. Conclusion
degrade preferentially amorphous cellulose leading to an increase
in the degree of crystallinity during hydrolysis (Zhang & Lynd, AFA is shown to be very efcient on concentrated raw starch
2004). Moreover in this work the temperature used for hydrolysis suspensions with hydrolysis degrees reaching 60% after 2 h and
(61 C) may provoke some starch reorganization during hydroly- 77% after 48 h for 31% starch. The resistance to hydrolysis is due
sis as annealing or amyloselipid complexing, especially when the to the presence of lipid-complexed amylose at low starch concen-
-glucan chains mobility increases after partial hydrolysis. tration, recrystallization of complexes being favoured at 61 C and
At 31% starch, amount of amylopectin is more intensively in excess water. Finally, at 31% starch rearrangement of amylose
reduced than that of amylose at 4896 h, which is consistent with fragments into B-type structures is probably the nal limiting fac-
both rapid hydrolysis of the crystalline domains, i.e. originating tor for complete hydrolysis. One of the most surprising results is
from the amylopectin, and the resistance to hydrolysis of lipid com- that AFA is able to hydrolyze the crystalline domains easier than
plexed amylose present in maize starch. Nevertheless, both total the amorphous ones. The ability of AFA to hydrolyze preferen-
amount and average molar mass of amylose decrease also sub- tially crystalline structures at high starch concentration, as already
stantially which evidences hydrolysis of non complexed amylose. observed for RA (Tawil et al., 2011), opens a new way in terms of
Indeed Morrison, Law, and Snape (1993c) showed that the rate fundamental approach of enzymatic hydrolysis in condensed sys-
of complexed amylose in native starch is about 15% of the total tems but also for the design of new industrial processing of native
amylose content. In this work, the presence of Vh-type structure starch.
after AFA hydrolysis for 48 h and longer hydrolysis times could
result from 2 mechanisms: (i) Vh-type crystalline domains could
Acknowledgements
be present initially in the native granules but in a relative amount
too small to be detected by X-ray diffraction before hydrolysis of the
Authors thank S. Guilois, M. de Carvalho and B. Pontoire for
A-type structure. Moreover their crystallinity could be improved by
excellent technical assistance. B. Henrissat (AFMB, CNRS Marseille)
annealing at 61 C. (ii) Vh-type could be formed during hydrolysis
and G. Veronese (LISBP, INSA Toulouse) are also acknowledged for
between amylose fragments released by enzyme and lipid present
helpful discussion.
in maize starch. By contrast, amylose and amylopectin amounts
decrease concomitantly during PPA hydrolysis, which is consis-
tent with a mode of attack granule by granule already observed for References
hydrolysis of wheat starch and other starches in solid state (Colonna
et al., 1988; Oates, 1997). Andr, G., Buleon, A., Haser, R., & Tran, V. (1999). Amylose chain in an interacting
context, II. Molecular modeling of a maltopentaose fragment in the barley -
The higher resistance of Vh-type structure, characteristic of
amylase catalytic site. Biopolymers, 50, 751762.
lipid-complexed amylose, compared to A-type is controversial. The Biliaderis, C. G., & Galloway, G. (1989). Crystallization behavior of amylose V com-
resistance has been observed during enzymatic hydrolysis of cereal plexes Structure properties relationships. Carbohydrate Research, 189, 3148.
Biliaderis, C. G. (1992). Structures and phase-transitions of starch in food systems.
starch at prolonged hydrolysis times by Gernat et al. (1993) and
Food Technology, 46, 98105.
Lauro et al. (1999). In contrast Gerard et al. (2001) have shown Bradford, M. M. (1976). A rapid and sensitive method for the quantication of
on maize starch mutants that A- and Vh-types have the same microgram quantities of protein utilizing the principle of protein-dye binding.
susceptibility to PPA hydrolysis. Moreover, lipid-complexed amy- Analytical Biochemistry, 72, 248254.
Buleon, A., Colonna, P., Planchot, V., & Ball, S. (1998). Starch granules: Structure and
lose formed in vitro is completely degraded by -amylases when biosynthesis. International Journal of Biological Macromolecules, 23, 85112.
an excess amount of enzyme or long hydrolysis time is applied Buleon, A., Veronese, G., & Putaux, J. L. (2007). Self-association and crystallization of
(Biliaderis & Galloway, 1989; Faisant et al., 1993; Holm et al., 1983). amylose. Australian Journal of Chemistry, 60, 706718.
Buleon, A., & Colonna, P. (2007). Physicochemical behaviour of starch in food applica-
Here, the temperature used for AFA assays was sufciently high to tions. The chemical physics of food. Oxford, U.K.: Blackwell Publishing., p. 2059.
favour either the annealing of existing complexes in native maize Christiansen, C., Hachem, M. A., Janecek, S., Viks-Nielsen, A., Blennow, A., & Svens-
starch into a more crystalline structure or the formation of new son, B. (2009). The carbohydrate-binding module family 20 diversity, structure,
and function. FEBS Journal, 276, 50065029.
complexes between amylose fragments released and lipid present Colonna, P., Buleon, A., & Lemarie, F. (1988). Action of Bacillus subtilis alpha-amylase
in maize starch. on native wheat starch. Biotechnology and Bioengineering, 31, 895904.
B-type structure appears at very high degrees of hydrolysis and Colonna, P., Leloup, V., & Buleon, A. (1992). Limiting factors of starch hydrolysis.
European Journal of Clinical Nutrition, 46, S17S32.
at high starch concentration. It probably results from recrystal-
De Cordt, S., Hendrickx, M., Maesmans, G., & Tobback, P. (1994). The inuence
lization of linear fragments of amylose released by the amylase of polyalcohols and carbohydrates on the thermostability of alpha-amylase.
when remaining amylose is widely hydrolyzed. The high initial Biotechnology and Bioengineering, 43(2), 107114.
Faisant, N., Champ, M., Colonna, P., Buleon, A., Molis, C., Langkilde, A. M., et al. (1993).
concentration of starch favours crystallization and precipitation,
Structural features of resistant starch at the end of the human small intestine.
and this type of structure probably constitutes the real resistant European Journal of Clinical Nutrition, 47, 285296.
part to enzymatic hydrolysis. Such reorganization during enzy- Gallant, D. J., Bouchet, B., Buleon, A., & Perez, S. (1992). Physical characteristics and
matic hydrolysis has already been observed by Lopez-Rubio et al. susceptibility of starch granules. European Journal of Clinical Nutrition, 46(Suppl.
2), S3S16.
(2008) on resistant starch and high amylose maize starch and more Gerard, C., Colonna, P., Buleon, A., & Planchot, V. (2001). Amylolysis of maize mutant
recently on maize starch (Tawil et al., 2011). Recrystallization of starches. Journal of the Science of Food and Agriculture, 81, 12811287.
Gernat, C., Radosta, S., Anger, H., & Damaschun, G. (1993). Crystalline parts of 3 Morrison, W. R., Tester, R. F., Snape, C. E., Law, R., & Gidley, M. J. (1993). Swelling and
different conformations detected in native and enzymatically degraded starches. gelatinization of cereal starches. 4. Some effects of lipid-complexed amylose
Starch-Staerke, 45, 309314. and free amylose in waxy and normal barley starches. Cereal Chemistry, 70(4),
Gorinstein, S. (1993). Kinetic studies during enzyme hydrolysis of potato and cassava 385391.
starches. Starch-Staerke, 45(3), 9195. Morrison, W. R., Tester, R. F., Gidley, M. J., & Karkalas, J. (1993). Resistance to acid-
Holm, J., Bjorck, I., Ostrowska, S., Eliasson, A. C., Asp, N. G., Larsson, K., et al. (1983). hydrolysis of lipid-complexed amylose and lipid-free amylose in lintnerised
Digestibility of amyloselipid complexes in vitro and in vivo. Starch-Staerke, 35, waxy and non-waxy barley starches. Carbohydrate Research, 245(2), 289302.
294297. Morrison, W. R., Law, R. V., & Snape, C. E. (1993). Evidence for inclusion complexes
Juge, N., Nohr, J., Le Gal-Coeffet, M. F., Kramhoft, B., Furniss, C. S. M., Planchot, V., of lipids with v-amylose in maize, rice and oat starches. Journal of Cereal Science,
Archer, D. B., Williamson, G., & Svensson, B. (2006). The activity of barley alpha- 18(2), 107109.
amylase on starch granules is enhanced by fusion of a starch binding domain Oates, C. G. (1997). Towards an understanding of starch granule structure and
from Aspergillus niger glucoamylase. Biochimica et Biophysica Acta Proteins & hydrolysis. Trends in Food Science and Technology, 8, 375382.
Proteomics, 1764, 275284. Planchot, V., Colonna, P., & Buleon, A. (1997). Enzymatic hydrolysis of -glucan
Kwasniewska-Karolak, I., Nebesny, E., & Rosicka-Kaczmarek. (2008). Characteriza- crystallites. Carbohydrate Research, 298, 319326.
tion of amyloselipid complexes derived from different wheat varieties and their Pohu, A., Putaux, J. L., Planchot, V., Colonna, P., & Buleon, A. (2004). Origin of the lim-
susceptibility to enzymatic hydrolysis. Food Science and Technology International, ited alpha-amylolysis of debranched maltodextrins crystallized in the A form:
14, 2937. A TEM study on model substrates. Biomacromolecules, 5, 119125.
Lauro, M., Suortti, T., Autio, K., Linko, P., & Poutanen, K. S. (1993). Accessibility of Rappenecker, G., & Zugenmaier, P. (1981). Conformation and packing analysis of
barley starch granules to -amylase during different phases of gelatinization. polysaccharides and derivatives, 8. detailed renement of the crystal structure
Journal of Cereal Science, 17(2), 125136. of Vh amylose. Carbohydrate Research, 89, 1119.
Lauro, M., Forssell, P. M., Suortti, M. T., Hulleman, S. H. D., & Poutanen, K. S. (1999). Rolland-Sabat, A., Amani, N. G., Dufour, D., Guilois, S., & Colonna, P. (2003). Macro-
Alpha-amylolysis of large barley starch granules. Cereal Chemistry, 76, 925930. molecular characteristics of ten yam (Dioscorea spp) starches. Journal of the
Le Bail, P., Bizot, H., Ollivon, M., Keller, G., Bourgaux, C., & Buleon, A. (1999). Monitor- Science of Food and Agriculture, 83, 927936.
ing the crystallization of amyloselipid complexes during maize starch melting Rolland-Sabat, A., Guilois, S., Jaillais, B., & Colonna, P. (2011). Molecular size and
using synchrotron X-ray diffraction. Biopolymers, 50, 99110. mass distributions of native starches using complementary separation meth-
Leloup, V. M., Colonna, P., & Ring, S. G. (1991). Diffusion of a globular protein ods: Asymmetrical Flow Field Flow Fractionation (A4F) and Hydrodynamic and
in amylose and amylopectin gels. Biotechnology and Bioengineering, 38, 127 High Performance Size Exclusion Chromatography (HDC-HPSEC). Analytical and
134. Bioanalytical Chemistry, 399(4), 14931505.
Leloup, V. M., Colonna, P., & Buleon, A. (1991). Inuence of amylose-amylopectin Sharma, A., Yadav, B. S., & Ritika. (2008). Resistant starch: Physiological roles and
ratio on gel properties. Journal of Cereal Science, 13, 113. food applications. Food Reviews International, 24, 193234.
Lopez-Rubio, A., Flanagan, B. M., Shrestha, A. K., Gidley, M. J., & Gilbert, E. P. (2008). Tawil, G., Viks-Nielsen, A., Rolland-Sabat, A., Colonna, P., & Buleon, A. (2011). In
A novel approach for calculating starch crystallinity and its correlation with depth study of a new highly efcient hydrolyzing -amylase from Rhizomucor
double helix content: A combined XRD and NMR study. Biomacromolecules, 9, sp. Biomacromolecules, 12, 3442.
19511958. Viks-Nielsen, A., Andersen, C., Hoff, T., & Pedersen, S. (2006). Development of new
McCleary, B. V., McNally, M., Monaghan, D., Mugford, D. C., Black, C., Broad- alpha-amylases for raw starch hydrolysis. Biocatalysis and Biotransformation,
bent, R., et al. (2002). Measurement of alpha-amylase activity in white 24(12), 121127.
wheat our, milled malt, and microbial enzyme preparations, using the Williamson, G., Belshaw, N. J., Self, D. J., Noel, T. R., Ring, S. G., Cairns, P., Morris, V.
ceralpha assay: Collaborative study. Journal of AOAC International, 85, J., Clark, S. A., & Parker, M. L. (1992). Hydrolysis of A-type and B-type crystalline
10961102. polymorphs of starch by alpha-amylase, beta-amylase and glucoamylase-1. Car-
Maache-Rezzoug, Z., Zarguili, I., Loisel, C., Queveau, D., & Bulon, A. (2008). bohydrate Polymers, 18, 179187.
Structural modications and thermal transitions of standard maize Zhang, Y. H. P., & Lynd, L. R. (2004). Toward an aggregated understanding of enzy-
starch after DIC hydrothermal treatment. Carbohydrate Polymers, 74(4), matic hydrolysis of cellulose: Noncomplexed cellulase systems. Biotechnology
802812. and Bioengineering, 88, 797824.
Review
Digestion of starch: In vivo and in vitro kinetic models used to characterise

oligosaccharide or glucose release
Anthony C. Dona a,b, Guilhem Pages a, Robert G. Gilbert b, Philip W. Kuchel a,*
a
School of Molecular & Microbial Biosciences, University of Sydney, Sydney, NSW 2006, Australia
b
The University of Queensland, Centre for Nutrition and Food Sciences, Hartley Teakle Building 83, Brisbane QLD 4072, Australia
Article history: We give an overview of the kinetics of starch digestion, emphasizing in vitro studies and the various
Received 8 May 2009 mathematical models used to analyse the data. The emphasis in this review is on ungelatinised starch
Received in revised form 14 December 2009 (wherein granules are still intact and unswollen), applicable to domestic animal feed and to some human
food. The mammalian digestive system uses a complex but well ordered series of processes to degrade
Available online 7 January 2010
and absorb nutrients from the diet of an individual. Mechanical action like mastication and churning
of food throughout the various subsections of the gastrointestinal tract work together with biochemical
Keywords:
components in secretions containing acids, buffers, and hydrolytic enzymes. Many attempts have been
Amyloglucosidase
a-Amylase made to mimic digestion in vitro in an effort to model its complexities. Characterisation of enzyme-cat-
Starch digestion alysed hydrolysis in vitro using puried enzymes, separately for simplication, has proven difcult, as the
Hydrolysis kinetic models starch granules complex structure causes enzyme action to follow unconventional kinetics. The suscep-
Product inhibition tibility of starch granules to digestion by glucohydrolases depends on a set of factors that include the
Rapidly digested starch granular structure, the method of preparation and the nature of the starch, and those molecules bound
Starch to it. Within a starch granule, the branching structure, molecular size and molecular weight distributions,
Time-resolved 1H nuclear magnetic and crystallinity, may all affect its physical properties, thus controlling digestibility. Characteristics such
resonance
as solubility and the presence of bre, fat and protein all contribute to the rate of digestion.
Crown Copyright 2010 Published by Elsevier Ltd. All rights reserved.
1. Introduction conditions ? structure of starch and of starch-containing sub-

stances ? digestion properties.
The digestion of starch is not a simple, single chemical process. In other words, biosynthetic events control the formation and
The process can be partially quantied by several measures which hence structure of the starch, which is dependent on growth con-
differ from one another depending on the enzyme(s) and reaction ditions (Buleon, Colonna, Planchot, & Ball, 1998; Copeland, Blazek,
conditions used to catalyse the hydrolysis. These measures include Salman, & Tang, 2009; Smith, 2007) (e.g., diurnal temperature,
the rate of starch loss, rate of glucose appearance, and the rate of water availability, etc.) and on any processing, such as cooking. It
appearance of various oligosaccharides. is important to be aware that causal relations follow the arrows gi-
Understanding the factors inuencing starch digestion is best ven above. While there might be a correlation between, say, plant
done through a causal, mechanistically-based approach through variety and rapid digestibility, this is not causal; observing that a
the following paradigm: biosynthesis ? growth and processing particular plant variety gives rise to a structure which is rapidly di-
gested is causal (Copeland et al., 2009).
Abbreviations: ADP-glucose, adenosine50 -diphosphoglucose; AF4, asymmetric
As discussed in detail below, there are a number of different
ow-eld ow fractionation; AUC, area under curve; CEST, chemical exchange structural scales to be considered in starch-containing foods, rang-
saturation transfer; DMSO, dimethylsulfoxide; DP, degree of polymerisation; DSC, ing through the distribution of individual starch branches, through
differential scanning calorimetry; FACE, Fluorescence-assisted capillary electro- the overall branching structure of the starch molecules in a gran-
phoresis; GI, Glycemic Index; GL, Glycemic Load; GPC, gel-permeation chromatog-
ule, through to the macroscopic structure of a grain. This structure
raphy; HPAEC, high-performance anion-exchange chromatography; HR MAS, high-
resolution magic-angle spinning; NMR, nuclear magnetic resonance; NSP, non- includes not only the starch but also proteins, lipids and non-starch
starch polysaccharide; RDS, rapidly digested starch; RS, resistant starch; SAXS, polysaccharides that may be present. Properties such as the kinet-
small-angle X-ray scattering; SDC, slowly digested carbohydrate; SEC, size-exclu- ics of digestion are controlled by this structure at one or more of
sion chromatography; SDS, slowly digested starch; SEM, scanning electron micros- the various structural levels (Smith, 2007).
copy; XRD, X-ray diffraction.
* Corresponding author. Tel.: +61 293513709.
The starch consumed by domestic animals is largely uncooked
E-mail address: p.kuchel@mmb.usyd.edu.au (P.W. Kuchel). (ungelatinised), whereas most human starch-containing food is
0144-8617/$ - see front matter Crown Copyright 2010 Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.carbpol.2010.01.002
600 A.C. Dona et al. / Carbohydrate Polymers 80 (2010) 599617
cooked (gelatinised), with some exceptions such as muesli. Cook- fraction of protein located in the starch granules themselves is rel-
ing greatly changes, and may destroy, granule structure. Unlike atively small (0.10.3%) (Csoti, Bako, Tamas, & Gardonyi, 2005).
most enzymic reactions, the digestion of starch in its ungelatinised Similarly, in sorghum, the starch is tightly bound to a protein (ka-
form is controlled, for the most part, by the granular architecture of rin), and this plays an equivalent role to the hull in wheat (Belton,
the substrate. The focus of this review is on the characterisation of Delgadillo, Halford, & Shewry, 2006).
the functional property of digestibility in starches. Although struc-
tural and physical properties of starch are presented, the introduc-
3. Molecular architecture of starch
tion is kept brief by using references to many literary sources
covering this broad area.
Regardless of their botanical origins, starch varieties contain
primarily two different types of anhydroglucose polymers, amylo-
2. Starch sources pectin and amylose; both are linked by a (1,4) bonds in linear seg-
ments, with a (1,6) bonds at branching points. Amylopectin makes
Starch is abundant in the grains of all cereal crops including up 7080% of most starch varieties and so is the major compo-
rice, wheat, maize, barley, and sorghum, as well as in pulses and nent, strongly inuencing the physiochemical and culinary charac-
tubers. The structure of starch and its packaging into granules var- teristics of starch. Amylopectin contains many short branches with
ies widely between plant species, but the most important as foods a non-random distribution of branching points (45%). On the
are rice and wheat so these are emphasized here. The generalisa- other hand, amylose is primarily a linear macromolecule with less
tion of concepts for studying digestion, to other sources of starch, than 1% long-chain branches forming predominately single chain
can be readily made. helices. Amylose is also known to exhibit a sixfold left-handed
Native starches exist in many varieties. For example, rice can be double helical conformation.
waxy, non-waxy, and sticky; while maize can contain low or high The structure of starch in a grain can be categorised (Ball et al.,
amylose. In other words these types differ in their amylose, amylo- 1996) into six levels (Fig. 1), ranging in scale from nm to mm: i.e., 6
pectin, protein, and lipid contents together with other structural orders of magnitude.
differences. About 1530% of the rice grain is protein, lipid, and Level 1: Individual branches (see Fig. 1) This is the distribution of
non-starch polysaccharides (Buleon et al., 1998; Chiou, Fellows, the chain lengths (degrees of polymerisation, DP) of the branches
Gilbert, & Fitzgerald, 2005). Protein fractions in most other grains, in a sample, often termed the chain-length distribution. The length
such as wheat, are signicantly higher. scale of chains is of the order of 1 nm.
The protein is also an important structural component in grain. Level 2: Whole starch molecules (see Fig. 1) This is the structure
We here consider only grains which have been hulled (or milled), of the branched molecules, which is exceptionally hard to fully
i.e., from which the hard outer layer(s) has/have been removed. characterise both theoretically (Gray-Weale & Gilbert, 2009) and
In wheat, much of the protein, predominantly friabilin, links the experimentally, but which can usefully be characterised by aver-
surface of starch granules to a complex matrix of proteins and lip- ages such as the number- and weight-average molecular weight,
ids. Matrix proteins in wheat seeds create a need for special extrac- or the weight-average size distribution (Gidley et al., 2010). The
tion methods to release the starch granules. Once released, the branching properties of each amylopectin anhydroglucose chain
Fig. 1. Six supramolecular levels of the rice grain, highlighting the microscopic structural contribution of starch.
A.C. Dona et al. / Carbohydrate Polymers 80 (2010) 599617 601
can be separated into three distinct groups (Peat, Whelan, & Tho-
mas, 1952) (Fig. 2): (1) A chains have no consequential branches
and are joined to a B or C chain through an a (1,6) bond at their
reducing end. (2) B chains differ from A chains by possessing sub-
sequent branches that are connected to them via an a (1,6) bond at
the reducing end of branches. (3) C chains are the backbone of the
amylopectin molecule having a free reducing end. Numerous
branches are attached to a C chain, thus laying the foundation for
the complex structure of amylopectin.
Level 3: Lamellar structure (see Fig. 1) In native starch, the tight
packing of glucan chains enables them to intertwine to form dou-
ble helices, which in turn create regions of crystalline order, with
the crystalline component largely comprising clusters of shorter
portions (typically 17 monomer units) of amylopectin branches;
the amylose is thought largely to reside in the amorphous layers.
This structure can be characterised by small-angle X-ray and neu-
tron scattering, and by scanning electron microscopy (SEM). Iso-
lated amylose molecules can form into a sixfold left-handed
double helical structure (Fig. 3), which also can exist in whole
starch after gelatinisation (Section 4).
Level 4: Granules (see Fig. 1) The lamellar structures occupy the in-
ner architecture of native starch granules in concentric growth ring
shells of thickness 100400 nm; these are separated by regions of
amorphous structure. The crystalline and amorphous lamellae alter-
nate radially, creating concentric shells. The amorphous regions of
the granule contain the branch points of amylopectin molecules, as
they tend not to align, and amylose, as it is also does not form tightly
packed crystalline regions (Donald, 2004). These semi-crystalline
and amorphous growth rings that make up the granules are 15
30 lm in size. Characterisation is by SEM. This level of structure is
greatly changed by the process of gelatinisation.
A conceptual model of starch granule structure called the hairy
billiard ball (Jane, Wong, & McPherson, 1997; Lineback, 1984,
1986; Robertson et al., 2006) illustrates both the single helical
organisation of amylose, and the double helical organisation and Fig. 3. Double helical conformation of amylose is dissimilar to other carbohydrates
clustering of amylopectin. The model describes the outside of the of glucose like cellulose and b (1,3) glucan that form single and triple helices,
respectively. The helix is tightened upon binding with iodine and iodide ions, or
granule with protruding branches of amylopectin (hair) rather
short-chain alcohols.
than a smooth exterior. The outside of the billiard ball, inside the
hairy layer, denes a boundary between initially accessible, or sur-
face-substrate sites, and initially inaccessible or interior substrate synthetic polymers (Jenkins et al., 1994). Instead, the double heli-
sites (Robertson et al., 2006). The hairy billiard ball model was ces appear to arrange side-by-side in a smectic or nematic-type
developed to explain short range surface-order in X-ray diffraction structure (Waigh et al., 1997). These properties led to another con-
(XRD) patterns. These patterns are unrelated to the X-ray patterns ceptual model of the granular architecture of starch, based on side
observed due to the semi-crystalline lamellae of granules (Seve- chain liquidcrystalline polymers. The rigid double helices of amy-
nou, Hill, Farhat, & Mitchell, 2002). The digestion time courses of lopectin branches provide the fundamental units of structural or-
glucohydrolase action on starch granules are also best explained der in the liquid crystals, although they need to be adequately
by variations in accessibility of the enzyme, seen in the hairy bil- decoupled from the backbone for liquid crystalline packing to oc-
liard ball model. Unlike synthetic polymers, such as polyethylene, cur (Shibaev, 1994).
starch amylopectin crystallites do not assemble into semi-crystal- The alternating semi-crystalline shells, elliptical in shape, regu-
line rings with chain-folded lamellae. The repeat distance of starch larly alternate with a repeat distance of 9 nm (Waigh, Donald,
crystallites upon annealing also does not appear to change as with Heidelbach, Riekel, & Gidley, 1999; Waigh et al., 2000). Regardless
of their botanical origins, amylopectin molecules exhibit the same
repeat distance for crystalline and amorphous regions, as mea-
sured by XRD (Waigh et al., 1997, 1999). The molecular order of
starch granules is also conrmed by their birefringent nature when
viewed in polarised light, displaying a Maltese cross (BeMiller,
2007).
Level 5: Endosperm (see Fig. 1) This comprises the starch gran-
ules, together with protein and lipids. This level of structure is
not usually important for digestion, with the exception of grains
such as sorghum that lack a hull and instead have a dense pro-
teinstarch matrix which is not readily broken down.
Level 6: Whole grain (see Fig. 1) This nal level, 1 mm in size,
includes the highest-level structures such as the hull, etc.
Fig. 2. Cluster model of amylopectin structure. The organisation of the chains is
made evident. An amylopectin molecule contains only one reducing end (denoted
The role of the granular structure in nature is as an energy-stor-
by the black dot). (Inferred from Peat et al., 1952.) age medium for the germinating plant, to release glucose slowly
given the right external stimuli. The compact crystalline structure Castignolles, Gilbert, & Gaborieau, 2009); the latter can be carried
of amylopectin is the most important structural component for this out at 80 C with minimal shear and mild mechanical disruption
purpose. Although amylose generally makes up a signicant frac- of the granule structure, and it is a recommended technique for
tion (2030%) of starch varieties, its role in the starch granule as characterisation. Organic solvents such as DMSO are not useful as
a source of glucose for energising an embryo is yet to be com- media (rather than for subsequent characterisation) for studies of
pletely understood. While amylopectin alone is sufcient for the digestion, as these environments may inhibit hydrolytic enzyme
formation of starch granules, amylose also plays a central role in action.
the initial stages of granule crystallisation (Ziegler, Creek, & Runt, Once starch has been gelatinised, the granules have swollen to
2005). Interestingly, details such as the amylopectinamylose ra- many times their original volume. In the absence of shear, the
tio, size, morphology, and size distribution of the starch granule granules maintain their molecular integrity (Parker & Ring,
all determine the physicochemical characteristics of the granule 2001). The swelling of starch granules is quantitatively related to
(Ellis et al., 1998). the pasting temperature: this is the rst temperature at which
the viscosity of a starch suspension rapidly increases on heating.
Swelling of granules is accompanied by leaching of amylose mole-
4. Functional properties cules and lipids, normally bound in the granule, into the continu-
ous phase (Bhattacharya, Sowbhagya, & Swamy, 1972, 1978). The
The addition of water and application of heat to raw grains is presence of amylose and lipids in the granule is known to retard
essential to transform them into a food with pleasing textural attri- the swelling because amylopectin contributes mainly to the uptake
butes. There are several stages that occur during cooking, including of water during swelling (Sasaki & Matsuki, 1998; Tester & Morri-
glass transition, gelatinisation (the non-equilibrium swelling of the son, 1990). The proportion of long chains in amylopectin correlates
crystalline regions (Slade & Levine, 1988)), swelling, pasting, and with the rate of starch swelling. Starches with higher swelling po-
retrogradation (see below). The stages evolve in this order during tential tend to contain higher proportions of long chains (DP P 35)
the heating and subsequent cooling of suspended granules, often in amylopectin (Bogracheva, Wang, & Hedley, 2001a; Sasaki & Mat-
depending on one another (Slade & Levine, 1988). In its native suki, 1998). Amylose and lipid content, and amylopectin structure,
form, a starch granule is insoluble in cold water, thus creating a are therefore all related to the swelling and gelatinisation charac-
suspension when mechanically dispersed in water. Heat treatment teristics of starch granules.
of these suspensions in excess water is generally used to make Gelatinised starch samples are far more susceptible to degrada-
homogeneous starchwater mixtures. The hydration and rheolog- tion by a-amylase than are native starch granules. Furthermore,
ical properties of carbohydrate polymers is well described (Matser characteristics such as the amylopectin/amylose ratio and the
& Steeneken, 1997; Steeneken, 1989). amylose complexes with lipids affect the rate of hydrolysis. The ex-
The glass transition temperature is that at which the amor- tent of digestibility of starches generally decreases as the amylose
phous (glassy) regions of crystalline polymers become soft or rub- content increases (Vesterinen, Myllrinen, Forssell, Sderling, &
bery. The semi-crystalline nature of starch granules mean that Autio, 2002), although amylose content alone is not a sole predic-
starch must lose its glassy, brittle properties before gelatinisation tor of digestibility (Htoon et al., 2009; Lopez-Rubio, Flanagan Bern-
can occur (Biliaderis, Page, Maurice, & Juliano, 1986). The glass adine, Shrestha Ashok, Gidley Michael, & Gilbert Elliot, 2008).
transition temperature is depressed by water, so at higher mois- Similarly, amylose complexed with lipid is more resistant to attack
ture contents the glass transition occurs at lower temperature by hydrolytic enzymes than is free carbohydrate (Nebesny, Rosi-
(Biliaderis et al., 1986; Levine & Slade, 1989; Slade & Levine, cka, & Tkaczyk, 2004).
1988). Unlike most synthetic polymers, which return to their The molecules in a gelatinised starch sample, when stored un-
glassy state upon cooling below the glass transition temperature, der certain conditions, can undergo inter- and intra-molecular
cooling starch granules to below the glass transition temperature association into an ordered structure; this is the process known
does not necessarily reverse the formation of soft, rubbery amor- as retrogradation. The association of amylose and amylopectin
phous regions (Biliaderis et al., 1986; Levine & Slade, 1989). crystals as gelatinised samples are cooled can lead to the formation
The gelatinisation of starch occurs in the neighbourhood of of a gel, or gelation, during the retrogradation process (Ring et al.,
60 C (see above) depending on its source (Shiotsubo, 1983). 1987). To retrograde, amylose must rst be heated to greater than
Above this temperature, the starch granule absorbs considerable 100 C, destroying the granular architecture (Miles, Morris, Orford,
amounts of water and loses its semi-crystalline nature while & Ring, 1985). Amylopectin recrystallises at a rate which is much
swelling to many times its original volume. Complete homogene- slower than that at which amylose forms its single or double heli-
ity of starch may only occur when swollen granules are heated ces, and the amylose-derived helices can thus aggregate (Baik, Kim,
well beyond the gelatinisation temperature, causing the disrup- Cheon, Ha, & Kim, 1997). Consequently, the duration of the rst
tion of the starch granules into smaller aggregates. At even higher stage of retrogradation depends on the amylose content of starch.
temperatures and pressures, spontaneous hydrolysis of starch High molecular weight amylose promotes retrogradation more
molecules increases the textural homogeneity of the mixture by than lower molecular weight polymers suggesting that the molec-
producing smaller oligosaccharides (Miyazawa, Ohtsu, Nakagawa, ular weight distribution of amylose long-chain branches (and the
& Funazukuri, 2006). main chain of the amylose) in the sample are one of the determi-
To characterise the size distribution of whole starch in mixtures nants of the temperature of the initial stage of retrogradation (Tsai
requires dissolution of the starch without degradation of the mol- & Lii, 2000). The later stages of retrogradation correlates with the
ecules. Some starch varieties become completely solubilised after chain-length distribution of the amylopectin, in particular the pro-
heating in water to 100 C. Autoclaving starch suspensions is also portion of short A chains (Fredriksson, Silverio, Andersson, Elias-
used to create a homogenous solution, although the effectiveness son, & Aman, 1998; Silverio, Fredriksson, Andersson, Eliasson, &
of autoclaving varies with the starch variety; this process may well Aman, 2000; Tsai & Lii, 2000). It is noted that the correlations cited
degrade the starch (Gidley et al., 2010; Ratnayake & Jackson, 2009). here are not necessarily causal. The retrogradation of starch has
Starch may be completely dissolved in anhydrous dimethylsulfox- been extensively studied by using several different techniques
ide (DMSO), either in the presence of a small amount of water (Bel- including differential scanning calorimetry (DSC) (Jane et al.,
lo-Prez, Roger, Baud, & Colonna, 1998) or with the addition of LiBr 1999; Lai, Lu, & Lii, 2000; Tako & Hizukuri, 2000; Tsai & Lii,
as a hydrogen-bond disruptor (Dona et al., 2007; Schmitz, Dona, 2000), XRD (Jagannath, Jayaraman, Arya, & Somashekar, 1998;
Jouppila, Kansikas, & Roos, 1998), enzyme kinetics (Kim, Kim, & such as temperature and pH, although they occur generally at
Shin, 1997; Tsuge, Tatsumi, Ohtani, & Nakazima, 1992), and dy- the optimal pH of 5 and at temperatures around 37 C. Two basic
namic viscoelastic measurements (Morikawa & Nishinari, 2000; types of enzyme catalyse polysaccharide degradation. Although
Otobe, Yoshii, Sugiyama, & Kikuchi, 1995; Tako & Hizukuri, 2000). phosphorylases play a key role in the metabolism of starch in the
Starch granules can display different diffraction patterns (via plant kingdom, we concentrate here on the action of specic
XRD) depending on their botanical origin; an A-type for normal hydrolases in the digestion of carbohydrates in mammals (Bae
starches and B-type for tuber starches (Katz, 1930) (Fig. 2). Differ- et al., 2005; Schinzel & Nidetzky, 1999).
ences between the two allomorphs relate to the relative amounts Endohydrolases (Werner & Keilich, 1965) are generally excreted
of water and the organisation of the double helices in the unit cell by cells and so operate outside them. Carbohydrate endohydrolas-
of the crystal (Bogracheva, Wang, & Hedley, 2001b; Bogracheva, es cleave large carbohydrates that are unable to diffuse into cells,
Wang, Wang, & Hedley, 2002). Amylopectin molecules in A-type to give smaller products that can traverse cell membranes. Endo-
starches have a larger number of short-chain fractions (Hizukuri, hydrolases randomly cleave a hydrated starch molecule into two
1985), but the crystallite is more susceptible to enzymic attack smaller molecules; in the specic case of a-amylase this is done
over longer times (Planchot, Colonna, & Buleon, 1997). by cleaving any accessible a (1,4) bond. Exohydrolases (Akerberg,
In native starch, the fraction of double helical order (Fig. 3) is Zacchi, Torto, & Gorton, 2000) release a monomer or dimer from
signicantly greater than crystalline order, suggesting not all heli- the non-reducing end of the substrate molecule. Enzymes such as
cal formations are involved in starch crystallites (Cooke & Gidley, amyloglucosidase, and b-amylase, release glucose or maltose units,
1992). Even so, helical starch molecules resist amylase hydrolysis, respectively, from the non-reducing end of starch and
explaining, at least in part, why high amylose starches resist diges- oligosaccharides.
tion more than waxy varieties, even though they are often less The surfaces of starch granules are affected in different ways by
crystalline (Tester, Qi, & Karkalas, 2006). enzyme action, as seen by light microscopy, confocal laser scan-
Crystallinity of the starch granule is not the only structural ning microscopy (Apinan et al., 2007), and SEM (Tester, Morrison,
property affecting enzymic degradation. Amyloselipid complexes Gidley, Kirkland, & Karkalas, 1994; Zhang, Ao, & Hamaker, 2006)
also affect the rate of digestion. These complexes accumulate in the (Fig. 4). Many different modes of enzyme attack have been identi-
granule as starch granules are degraded (Gernat, Radosta, Anger, & ed, including pin-hole, sponge-like erosion, medium-sized holes,
Damaschun, 1993). Whether amyloselipid complexes observed single holes in individual granules, and surface erosion (Sujka &
are always present or formed during digestion is still unknown, Jamroz, 2007). Depending on the enzyme and variety of starch, en-
although they are resistant to enzymic attack. zymes can erode the entire granule surface (exo-corrosion) or di-
gest channels at specic locations on the surface towards the
centre (endo-corrosion). Starches from wheat, barley and rye have
5. Digestive enzymes specic susceptible zones that become pitted as a result of endo-
corrosion (Fig. 4). Pits become enlarged and form numerous chan-
There are many hydrolytic enzymes within the digestive tract of nels within the granule, thus weakening its structure. The granules
animals that catalyse the breakdown of polymeric macromole- eventually fragment, leaving behind residual starch (Fig. 4d).
cules. Rates of enzymic action are very dependent on conditions Although the physiochemical properties of the crystalline and
Fig. 4. SEM images (Zhang et al., 2006) of normal maize starch hydrolysed for: (A) 0 min (control); (B) 40 min (note the pit-hole action of a-amylase and amyloglucosidase);
and (C) 120 min. The nal image, D, shows residual pyramid shaped fragments that remain after digestion.
amorphous regions of the granule are different, the residual starch zymes that are integral to the plasma membrane of enterocytes
after digestion has a virtually unchanged debranched distribution, in the small intestine; these include mucosal maltaseglucoamy-
as measured by size-exclusion chromatography with multiple-an- lase, and sucraseisomaltase. These are exo-glucosidases that act
gle laser light scattering and refractive index detection (Zhang on the non-reducing end of glucose oligomers and catalyse not
et al., 2006). A similar prole after debranching implies a progres- only the hydrolysis of a (1,4) bonds but also to a lesser extent that
sion of digestion through the crystalline and amorphous lamellae; of a (1,6) branch bonds, ensuring further degradation of nonlinear
this is also seen with SEM (Zhang et al., 2006) (Fig. 4). Residues left oligosaccharides. The resulting monosaccharides, such as glucose
after digestion also show comparable properties, such as gelatini- and galactose, are absorbed by secondary active transport across
sation temperature and enthalpy, with native granules as analysed the apical membrane of enterocytes and subsequently exit the gas-
by differential scanning calorimetry (Zhang et al., 2006). trointestinal tract across the basolateral membrane, into the blood-
High amylose starches characterised at different stages of stream (Boron & Boulpaep, 2009).
in vitro digestion display properties that suggest that amorphous The large intestine (colon) contains a very large population of
areas of the granule are hydrolysed by a-amylase, leaving the heli- micro-organisms (showing considerable variation between indi-
cal crystalline region intact. Analysis with SEM, small-angle X-ray viduals) that are benecial to the digestion of polysaccharides in
scattering (SAXS), infrared spectroscopy, solid state 13C nuclear humans who might be enzyme-decient, or in normal subjects
magnetic resonance (NMR) spectroscopy, and XRD show that an with carbohydrates that have not been broken down in the jeju-
increase in the molecular order is favoured by the hydrolytic action num and ileum. The bacteria inhabiting this segment of the diges-
of the enzyme (Lopez-Rubio et al., 2008). The changes suggest that tive tract metabolise, through fermentation, undigested
resistant starch (RS) does not have a specic structure in pre-di- polysaccharides like RS and soluble ber. Fermentation creates
gested samples, but may be formed during digestion by a rear- short-chain fatty acids, which stabilise blood glucose levels and
rangement of amylose chains into resistant structures of higher suppress cholesterol synthesis in the liver (MacFarlane & MacFar-
crystallinity (Leloup, Colonna, & Ring, 1991; Oates, 1997). Conse- lane, 2006; Wong, de Souza, Kendall, Emam, & Jenkins, 2006).
quently, the resistance to enzyme digestion is a result of competi-
tion between the kinetics of enzyme hydrolysis, and the kinetics of
amylose retrogradation. Furthermore, when hydrolysis is stopped 6. Digestibility indices
after different times, the yield of high amylose RS fragments dimin-
ishes, but the size of the fragments stays constant (Jane & Robyt, In the last 25 years the in vivo digestibility of foods has been
1984). This is attributed to the amorphous regions being digested classied by a number of metrics, the most popular of which is
more readily, leaving the chains in the helical crystalline regions the Glycemic Index (GI). The GI was rst dened by Jenkins et al.
(Fig. 3). (1981) as the total glycemic response in the blood [area under
the curve of glucose concentration versus time (Table 1)] in the
5.1. Human digestive system 2 h period immediately subsequent to the consumption of a xed
amount of carbohydrate; it is expressed as a value relative to that
Starch is met by salivary amylase in the mouth, which is the of a standard food, normally white bread or glucose (Roberts,
rst enzyme to act on carbohydrates during digestion (Hoebler, 2000). Many intrinsic and extrinsic factors affect the nature of
Devaux, Karinthi, Belleville, & Barry, 2000). In a relatively short starch and so affect the GI of a food, including starch structure at
time, the bolus of food is carried by oesophageal peristalsis into all six levels (Fig. 1); these factors are gastrointestinal motility,
the stomach. One of the key cells in the stomach for starch diges- the method of cooking, and the presence of bers, fats and proteins
tion is the parietal cell, which secretes HCl. The pH of the gastric (Krezowski, Nuttall, Gannon, & Bartosh, 1986; Thorne, Thompson,
juice is 2.6, retarding the action of a-amylase but increasing & Jenkins, 1983). The extent of starch retrogradation, the granular
the acid hydrolysis of starch. Also in the upper gastrointestinal morphology and the dietary ber content are factors that have
tract, lipids bound to starch are hydrolysed by lipases secreted been examined in vitro (Urooj & Puttraj, 1999).
by various exocrine glands. A key step in lipid digestion is the cre- Through the process of retrogradation, gelatinized or solubilised
ation of an emulsion that increases the area of the oilwater inter- starch can be transformed from an unstructured into a more or-
face enabling more efcient action of the enzymes. The emulsion is dered or crystalline state. This large physical change causes heat-
produced rst through the mechanical process of mastication and processed starchy foods to harden or become stale as they sponta-
then peristaltic movements of the digestive tract. The emulsion is neously approach a metastable state of lower free energy. This has
stabilised by its droplets being coated with membrane lipids, dena- been reported to decrease the GI value, due to an increased resis-
tured proteins, and fatty acids, thus preventing them from coalesc- tance to amylase (Chung, Lim, & Lim, 2006).
ing (Boron & Boulpaep, 2009). Other systems used to rank the in vivo digestibility of carbohy-
From the stomach, the ingested food proceeds to the duodenum drates are Glycemic Load (GL), Glycemic Response, and RS (Sec-
where it encounters the pancreatic secretion whose rate of release tion 7.1). The GL is a metric used as a basis for weight loss, or for
is controlled by satiety hormones. Pancreatic uid contains two diabetes control. Also controversial (Das et al., 2007), the GL is cal-
important components for starch digestion. Sodium hydrogencar- culated by multiplying the GI by the percentage of carbohydrate
bonate(bicarbonate) neutralises the acidity of the uid arriving present within the food consumed. Like the GI, the GL has a scale
from the stomach to a pH of 8. Pancreatic uid also contains a- used to collect food varieties into groups suitable for certain diets.
amylase that continues the hydrolysis of starch into glucose and The Glycemic Response is a measure of the increase in the glucose
oligosaccharides. The latter include linear and branched structures concentration in the blood at any given time after the ingestion of
that are not absorbed into the bloodstream without further hydro- carbohydrate and RS, along with the other starch fractions
lysis to glucose. Glucose is only a very minor product of endohy- (Section 7.1).
drolases, such as a-amylase which operates during the initial Conducting experiments on commercial foods to estimate their
stages of digestion, so further enzymic processes are denitely GI, or other in vivo measures, is an expensive process, so recent re-
required. search has been aimed at developing in vitro methods to predict GI
Substrates not digested by a-amylase, such as a-limit dextrins, or an equivalent metric (Garsetti et al., 2005; Goni, Garcia-Alonso,
and small linear oligomers, along with larger a-glucans, are later & Saura-Calixto, 1997; Okuda, Aramaki, Koseki, Satoh, & Hashiz-
degraded into single glucose units. This conversion occurs via en- ume, 2005) (Section 7).
Table 1
Methods of kinetic analysis of the enzymic digestion of starch sorted by type of model.
Source and treatment Techniques used Model used Results Commentary

In vivo monitoring of starch digestion
Jenkins et al. Range of Blood glucose concentrations GI calculated by comparing Commercial products GI The GI measure of starch
(1981), Ross, carbohydrate-based of subjects measured in area under the curve of blood ranged from 30130 AUC. hydrolysis to glucose is
Brand, Thorburn, meals tested prepared 30 min intervals after glucose concentration for 2 h Products with a GI below 55 dependent on many factors
and Truswell using a range of consumption of commercial, after ingestion relative to a AUC are generally including the age, sex, race,
(1987) cooking methods starch-based products white bread standard considered healthier protein ingestion (Krezowski
et al., 1986), food processing
(Brand, Nicholson, Thorburn,
& Truswell, 1985), and health
(Brand-Miller et al., 2002;
Jenkins et al., 2002; Thorne
et al., 1983) of the subject.
Also GI only considers
glucose absorption into the
blood stream leaving other
sugars unconsidered. The
meaningfulness of this
measure remains
controversial (DeVries, 2007)
Exponential kinetic models
Frei et al. (2003), Range of Digestions performed both A simple exponential model The in vitro data show a No mechanistic basis for
Goni et al. (1997) carbohydrate-based in vivo and in vitro using a- is used to calculate the area correlation with the in vivo the exponential model used
meals tested prepared amylase and under in vitro digestion GI data to analyse in vitro data
using a range of amyloglucosidase. The areas curves Time points only taken
cooking methods under the curves are every 30 min both in vivo and
compared for various foods in vitro leaving interesting
parts of the digestion
unmonitored during in vitro
analysis
Apar & Ozbek (2007) No prior treatment, Digestion time courses of a- Simple exponential model Both glucose and maltose No mechanistic basis for
rice starch granules amylase monitored over the tted to a range of initial cause uncompetitive the exponential model used
digested at 60 C and rst 10 min of digestion with concentrations. Initial product inhibition to analyse data
pH 6.5 residual starch concentration velocities analysed using Only the rst 10 min of
measured by iodine binding LineweaverBurk plots digestion was monitored
Remaining starch
determined by iodine
binding rather than sugar
production measured by
reducing power of solution,
decreasing reliability
Wang et al. (2006) No prior treatment, Digestion of starch using Simple exponential model Glucose found to inhibit Mathematical model was
corn starch granules amyloglucosidase monitored used to describe the product amyloglucosidase after a created although no
digested in an by glucose assay and inhibition of glucose grace period at low simulations of the model or
incubator at 50 C spectrophotometry glucose concentrations t to the data were shown
Optimal activity at 40 C No mechanistic basis for
the exponential model used
to analyse in vitro data
Al-Rabadi et al. Barley and sorghum Digestion performed Simple exponential model First order kinetics Although model closely
(2009) starch digested at mimicking conditions in vivo. tted to fraction of digested describes grain fragment describes starch digestion,
37 C in three stages Mechanical shaking of starch using linear ts to digestion only the SDS stage was
with pH ranging from samples along with a- logarithmic plots Small starch particle sizes monitored as during the rst
2.0 to 7.0 amylase and increase the rate of enzymic 30 min of digestion no
amyloglucosidase solutions action aliquots were taken
subsequently followed by
glucose assay and
spectrophotometry
Starch fractions determined, no tted model
Benmoussa, Englyst Starch Englyst test (Englyst et al., Fractions quantied for Amylopectin ne structure Englyst and related
Moldenhauer, and based meals prepared 1992) used to quantify commercial foods and correlated to digestibility methods only uses two time
Hamaker (2007), using various cooking hydrolysis various starches of differing values (RDS and SDS) points during a digestion,
Englyst and techniques but ne structure leaving the mechanism of
Hudson (1996), digested at 37 C and starch hydrolysis
Pizzoferrato, pH 5.2 unconsidered
Rotilio, and Paci Pizzoferrato
(1999) Chestnuts roasted
and ground and
incubated at 37 C
and pH 5.0
Benmoussa Rice
our samples were
boiled for 20 min and
digested at 37 C
(continued on next page)

Table 1 (continued)

Chung et al. (2006) Waxy rice starch was Englyst digestion values and Fractions quantied Gelatinized and retrograded Glucose concentration
gelatinised at various time courses of a-amylase although no kinetic model samples digestibility values measured while other
temperatures for digestion plotted tted to time courses compared oligosaccharides ignored
5 min and digested at
37 C
Ao et al. (2007), Zhang Maize, potato Englyst digestion values and Fractions quantied Modied starches with No attempt is made at
Zhang et al. and wheat starch time courses of a-amylase although no kinetic model larger branch density and characterising digestion by
(2006) with no prior and amyloglucosidase tted to time courses smaller chain length have using a kinetic model
treatment were digestions of native and slower digestion properties
incubated at 37 C modied starches from
and pH 5.2 different varieties were
Ao Maize starch characterised
heated to 80 C for
10 min, autoclaved
for 20 min then
incubated at 50 C
and pH 5.0 during
digestion
Manelius et al. Manelius maize and Total reducing sugar after Time points taken every Resistant starch No kinetic model used to
(1997), Manelius wheat starch digestion with a-amylase >30 min although no model demonstrated slow and analyse in vitro data
et al. (2000), Shu digestion performed determined for native and used to t data incomplete hydrolysis. Time points only taken
et al. (2006) at 25 C and pH 6.5 resistant starch varieties. Furthermore, high degrees every >30 min leaving
with no prior Also different degrees of of substitution and large- interesting parts of the
hydrothermal substitution and different granule sizes decrease the digestion unmonitored
treatment sized granules were analysed rate of enzymic attack
Shu rice starch using reducing sugar assay or
samples cooked at gel-permeation
50 C and digested at chromatography
37 C
Bertoft and Manelius Waxy maize starch a-Amylase digestion No kinetic model used to t The method described Recovery of sample from
(1992) boiled for 60 min monitored by gel- to the data from chain- allows for the hydrolysis of chromatography is not
then digested at 23 C permeation chromatography length distributions granules to be monitored complete (Ward et al., 2006)
and pH 5.5 at various time points and the products collected No attempt made to
at any stage during a describe the digestion of
digestion starch using a kinetic model
MichaelisMenten multiple stage model
Dona et al. (2009) Granular wheat and Digestion of wheat our and RDS and SDS stages of starch Both stages of starch LineweaverBurk plots
rice starch rice starch by both a- digestion analysed digestion could be used have large systematic
suspensions were amylase and separately, initially using characterised separately errors
digested at pH 5.2 and amyloglucosidase monitored LineweaverBurk plots using classical Michaelis
25 C without prior by 1H NMR spectroscopy Menten kinetics
hydrothermal
treatment
MichaelisMenten without inhibition or inactivation
Nitta et al. (1979) Soybean starch and The kinetics of b-amylase Data tted to HanesWoolf The optimal pH of b- An elaborate kinetic model
potato amylose action of soy bean starch was plot to nd Michaelis amylase was found to be was designed to consider
suspensions were analysed over a range of pH parameters at various pH 5.85 enzymic action and pH
digested at a range of values by testing the values and a model dependence although no
pHs at 25 C without reducing end concentration developed to predict examples of the predictions
prior hydrothermal over time degradation at a given pH of the model were compared
treatment to a digestive time course
Heitmann, Wenzig, Potato starch a-Amylase digestion Reducing sugar liberated Kinetics of three native Very limited time points,
and Mersmann digestion of samples monitored by GPC estimated by Michaelis starches tted using whole time course not
(1997) incubated at 60 C Menten analysis LineweaverBurk analysis considered
Recovery of sample from
chromatography is not
complete (Ward et al., 2006)
Park and Rollings Puried amylose and a-Amylase digestion Reducing sugar liberated Kinetics of amylopectin and Same criticisms as above
(1995) amylopectin from monitored by SEC with low estimated by Michaelis amylose compared so Digestion halted by NaOH
Sigma incubated for angle laser light scattering Menten analysis branching density affect on although starch is further
30 min at 50 C in digestion characterised degradated in alkaline
H2O/DMSO mixture environments
Very limited time points,
whole time course not
considered
Akerberg et al. (2000) Wheat starch heated Digestion with both a- A model based on Michaelis Various starch Although digestions were
to 90 C for 15 min amylase and Menten kinetics considering concentrations had their run over 80 h normally only
and then incubated at amyloglucosidase was hydrolysis by both enzymes products during digestion 10 measurements were
30 C and pH 5.0 characterised by HPAEC and various sizes of simulated accurately taken of products
during digestion substrate along with pH The number of variable
dependence was simulated parameters in the kinetic
model created allows the
simulated model to t
experimental data accurately
with many different
parameter values
Table 1 (continued)

Amato et al. (2004) Wheat our Digestion of wheat our by MichaelisMenten model No inhibition occurs HR MAS spectroscopy takes
suspensions with no endogenous enzymes without inhibition or during hydrolysis 30 min to prepare the
prior hydrothermal monitored by HR MAS NMR inactivation Spectroscopy can be used sample for analysis after
treatment monitored spectroscopy to monitor hydrolysis enzyme addition, missing the
at 27 C and pH 6.8 initial stages of digestion
Considering the adsorption of enzyme onto substrate before enzymic hydrolysis
Slaughter, Ellis, and Wheat, potato and Monitoring a-amylase Freundlich plot considering A linear relationship found No t to the integrated
Butterworth rice starch digestion by means of adsorption used to analyse with Freundlich plot MichaelisMenten equation
(2001) suspensions were reducing-end assay suspended samples and suggesting enzyme was displayed for either
incubated at 37 C regular MichaelisMenten adsorption onto the surface gelled samples of suspended
and pH 7.4 during used for gelled starch of starch limits the rate of samples after considering
digestion suspended starch digestion adsorption step
Product or substrate inhibited MichaelisMenten
Fujii and Kawamura Starch of unspecied Digestion with both a- Product-inhibited Michaelis The action of a-amylase can Digestion halted by strong
(1985) source incubated at amylase and Menten model used to be neglected after substrate alkaline solution although
30 C and pH 5.25 amyloglucosidase were estimate kinetic parameters molecular weight decreases starch is further degraded in
characterised by measuring by used of LineweaverBurk below 5000 alkaline environments
the reducing power of plots Product inhibition equation
aliquots used to t data although
other papers report no
inhibition of glucosidases
Pastrana et al. (1998) Starch of unspecied Digestion by glucoamylase Kinetic model used to t Digestion is inhibited by the The rheological properties
source incubated at was monitored by glucose digestion data was classical substrate only due to of starch solutions depend
23 C and pH 5.37 assay and diffusional MichaelisMenten changing rheological greatly upon the preparation
restriction in starch solutions multiplied by an exponential properties of the reacting technique (autoclaved or
of varying concentration describing restriction of the system, not by classical heated or suspended only).
enzyme due to an increase in substrate inhibition effects This paper has not explained
solution viscosity how starch solutions were
prepared although
rheological properties were
used to explain inhibition
Acronyms: AUC, area under curve; HR MAS, high-resolution magic-angle spinning; RDS, rapidly digested starch; SDS, slowly digested starch.
6.1. Glycemic Index the expense of fat oxidation, encouraging body fat gain (Brand-
Miller, Holt, Pawlak, & McMillan, 2002). The consumption of
The GI of food is a measure of the rate at which the contained high-GI foods can therefore be related to obesity for physiological
carbohydrate is hydrolysed in the digestive system and absorbed reasons, but obesity is also thought to develop from an altered
via active and facilitated diffusion across enterocyte membranes appetite (Thornley, McRobbie, Eyles, Walker, & Simmons, 2008).
into the bloodstream. The immediate effects of carbohydrates on Foods that are hydrolysed relatively quickly in the digestive system
an individuals blood glucose concentration are of interest not only leave subjects hungry in a relatively short time, increasing a ten-
for nutritional guidance but the glucose concentration has various dency to eat between meals. This increased appetite created by
health implications (Jenkins et al., 2002). The human subjects used high-GI foods is also thought to contribute to obesity (Roberts,
to test varieties of food must be carefully chosen, as the rate of 2000).
digestion of carbohydrates varies with health status, race and gen-
der. Although widely used, both for research and commercially, the 7. In vitro starch digestion
validity of the GI as a guide for dietary design is controversial.
Skeptics question fundamental properties of this functional mea- The digestibility of starch measured in vivo is a time-consum-
sure, including its reproducibility and therefore its meaning (Bar- ing, expensive process that requires many human subjects with
clay et al., 2008; DeVries, 2007; Livesey, Taylor, Hulshof, & specic attributes. Since there is no human subject dependence
Howlett, 2008), as well as its relevance to diseases with a nutri- on the measurement of in vitro starch digestion, investigation of
tional basis. in vitro digestibility as a replacement for the GI is an increasingly
researched topic (Table 1) (Frei, Siddhuraju, & Becker, 2003; Goni
6.2. Health inferences of GI et al., 1997).
Although controversial, the GI has proven useful as a numerical 7.1. Starch fractions
classication in the treatment of diabetes (Wolever et al., 2008).
This health inference derives from blood glucose concentration For nutritional purposes, starch can be classied into three cat-
and the insulin response of diabetics, relating directly to the rate egories by the Englyst test (Englyst, Kingman, & Cummings, 1992),
of digestion of carbohydrates (Jenkins et al., 2002). Obesity, which depending on their rate and extent of digestion; these include rap-
eventually leads to a variety of circulatory and respiratory prob- idly digested starch (RDS), slowly digested starch (SDS), and resis-
lems, is another major aspect of health that can be related to GI. tant starch (RS). The kinetic time course of the digestion of
Obesity was once thought to be best controlled by low-fat, high- suspended starch samples reveals very different rates between
carbohydrate diets, although commonly these are counterproduc- stages, suggesting that a change in the physical nature of the sub-
tive to weight loss due to promotion of carbohydrate oxidation at strate determines the kinetics of digestion (Table 1).
RDS is the fraction of starch granules that cause a rapid increase

in blood glucose concentration after ingestion of carbohydrates.
The fraction of starch that is said to be RDS in vitro is dened as
the amount of starch digested in the rst 20 min of a standard
digestion reaction mixture (Englyst et al., 1992). In practice, the
end of the RDS stage is considered to involve a change in the prop-
erties of the substrate, causing a transition in the rate of oligosac-
charide or glucose production. The extent of reaction at which this
occurs can vary, depending on the conditions of digestion and of
course will not always occur in the rst 20 min after enzyme addi-
tion (Akerberg et al., 2000; Ao et al., 2007; Apar & Ozbek, 2007;
Wang, Zeng, Liu, & Yuan, 2006); this time depends on how much
enzyme is added. Although RDS is dened by experimental analy-
sis of digestion in vitro, the rate of starch conversion to sugar fol-
lows similar kinetics in the human digestive system. Both free
sugar and RDS in a meal must travel to the small intestine before
the majority of the sugar is absorbed into the bloodstream. Coinci-
Fig. 5. Time courses of starch hydrolysis simulated with various enzyme kinetics
dently, RDS is not to be confused with the total sugar content of a
models (real data in blue). The tting of parameters used least squares regression
meal, which also contributes to the rate of increase in blood glu- analysis (also see Fig. 6). For all models: C0 = 130 mmol L1. Red Michaelis
cose concentration. Menten where Km = 8.9 102 mmol L1 and Vmax = 1.3 mmol L1 min1 (Eq. (2));
SDS is the fraction of starch that is digested slowly but com- Orange MichaelisMenten with product inhibition where Km = 1.5 mmol L1,
Vmax = 86 mmol L1 min1 and Ki = 6.2 104 mmol L1 (Eq. (8)); Green Michae-
pletely in the human small intestine. From studies of in vitro diges-
lisMenten with high-substrate inhibition where Km = 5.3 102 mmol L1,
tion it is clear that there is a transition in the smoothness of the Vmax = 9.1 101 mmol L1 min1 and Ks = 8.4 103 mmol L1 (Eq. (9)); and Pur-
progress curves of reducing sugar production from RDS to SDS ple MichaelisMenten with diffusional restriction where
(Figs. 5 and 6). SDS is dened as the starch that is digested after Km = 1.3 105 mmol L1, Vmax = 5.7 102 mmol L1 min1, Rm = 2.1 and
the RDS but in no longer than 120 min under standard conditions l = 3.8 103 mmol1 L (Eq. (10)). (For interpretation of the references to colour
in this gure legend, the reader is referred to the web version of this article.)
of substrate and enzyme concentration (Englyst et al., 1992). The
potential health benets of SDS in vivo include stable glucose
metabolism, diabetes management, mental performance, and sati-
ety (Lehmann & Robin, 2007). Commercially, there are no SDS
products available, although new slowly digestible carbohydrates
(SDCs) such as isomaltulose (Palatinose, Palatinit) have been com-
mercialised. These substances yield a slow and sustained blood
glucose concentration after intake (Lina, Jonker, & Kozianowski,
2002).
The fraction of starch that escapes digestion in the small intes-
tine, and may be subject to bacterial fermentation in the large
intestine, is termed RS, derived from in vitro studies where starch
undergoes limited enzymic hydrolysis. RS can have both benecial
and detrimental effects on a persons health; e.g., polysaccharides
undigested until reaching the large intestine become metabolised
to short-chain fatty acids by bacteria. Saturated fatty acids are
commonly linked to cardiovascular disease including heart attack,
heart failure, and stroke. On the other hand, the benecial action of
RS is similar to that of brous polysaccharides that resist endoge-
nous human enzymes. Dietary bre is composed mostly of non-
starch polysaccharides (NSPs) such as arabinoxylan and b-glucan. Fig. 6. Digestion of 6% (w/w) starch suspended in H2O with 12 U mL1 a-amylase at
25 C. The RDS (<120 min) tted with the integrated MichaelisMenten equation
By decreasing the plasma insulin and glucose response, NSP and
(Eq. (2)) used parameters Km = 1.4 102 mmol L1 and Vmax = 5.3 mmol L1 min1
RS are acknowledged as important contributors to the prevention and the SDS (>120 min) had parameters Km = 3.6 102 mmol L1 and
and management of diabetes, obesity, and colon and rectal cancers Vmax = 1.9 101 mmol L1 min1.
(Topping et al., 2008). A range of physiological inuences affect
starch digestibility: e.g., people who chew their food sparingly
swallow larger particle sizes. Larger particles allow less access for to correlate digestibility and the physical properties of starch
the digestive enzymes and have a faster oro-ileocaecal transit time, granules.
making the presence of RS more likely, therefore increasing the
contribution to the various health benets associated with RS. 8.1. Monitoring digestion
8.1.1. Reducing capacity

8. Kinetic models of in vitro digestion Digestion of starch is most commonly monitored by measuring
the reducing capacity of the ltrate of aliquots taken at intervals
As discussed above, the trajectory of digestion time courses is during the reaction (Goni et al., 1997; Sayago-Ayerdi, Tovar, Oso-
changed dramatically by variations in the methods of starch prep- rio-Diaz, Paredes-Lopez, & Bello-Perez, 2005; Shu et al., 2006).
aration; starch digestibility depends on the many physical proper- For exohydrolases, such as amyloglucosidase, an exact concentra-
ties of the starch granule (Fig. 1). This section describes a range of tion of product can be determined by this method, due to the spec-
kinetic models that are summarised in Table 1; these have been icity of the enzyme. When this method is used for a-amylase, the
used to characterise the kinetics of starch digestion, in an attempt concentration of oligosaccharide can be estimated, although the
amount of starch left undigested must be determined by weighing 8.1.3. 1H NMR spectroscopy
the residual substrate, after ltration and drying. Measuring both Techniques that can directly reect the concentration of sugars
oligosaccharide production and remaining substrate allows for a in a solution, including NMR spectroscopy (Berthon & Kuchel,
number-average weight of the products to be calculated, as a- 1995; Dona, Pages, Gilbert, Gaborieau, & Kuchel, 2009; Kuchel,
amylase can produce a range of products from glucose to large- 1981, 1989), and high-resolution magic-angle spinning (HR MAS)
1
branched limit dextrins (Manelius, Qin, Avall, Andtfolk, & Bertoft, H NMR, have been implemented (Amato et al., 2004). The initial
1997). The oligosaccharide products can be converted to glucose phase of the reaction is often overlooked when using HR MAS
via an exohydrolase and the glucose concentration measured using NMR, as there is an inability to acquire data at a sufciently rapid
a glucometer, to estimate the amount of starch digested (Aarathi, rate, immediately after mixing the sample and enzyme.
Urooj, & Puttaraj, 2003). Solution-state 1H NMR spectroscopy is often used to monitor di-
Strong alkali which is generally used to terminate enzymic reac- rectly the kinetics of chemical processes in homogeneous, and het-
tions is known to hydrolyse polysaccharides. This leads to a change erogeneous, solutions such as starch mixtures (Dona et al., 2009;
in the reducing capacity of the solution after the digestion is pre- Minerath, Casale, & Elrod, 2008; Walker et al., 2007). Recently,
sumed to have ceased. The estimated reducing capacity will there- NMR methodology has been developed to monitor progress of
fore always be biased when hydroxides are used. physical and chemical processes in heterogeneous starch suspen-
sions (Dona et al., 2007, 2009) provided that: (1) the nucleus (nu-
clei) monitored is in solution during the entirety of the reaction;
8.1.2. Chromatography
and (2) the heterogenous solution can be kept from settling (pro-
Digestion can be monitored by two different chromatography
ducing a vertical concentration gradient) during the time course.
techniques (Gidley et al., 2010; Morell, Samuel, & OShea, 1998).
NMR spectroscopy only measures from a small portion of the sam-
Size-exclusion chromatography (SEC, often termed gel-permeation
ple (within the receiving coil), creating a bias in the measured con-
chromatography, GPC) can be used to measure the evolution in the
centration if a heterogeneous sample settles.
molecular size distribution of the carbohydrate substrate during
Introduction of 1H NMR spectroscopy for the direct analysis of
the course of digestion. It is essential to realise that SEC separates
both a- and b-reducing ends of oligosaccharides produced during
by size (as indicated by its name), not by molecular weight, and for
hydrolysis enables the accurate estimation of product formation
a branched polymer such as starch there is normally no unique
during digestion of carbohydrate (Fig. 7). The method can be ap-
relationship between hydrodynamic radius and molecular weight
plied to many forms of starch including extracted, milled our,
(although there is a unique size/molecular weight relationship
and cooked, uncooked, macerated, and whole grain starch. Record-
for linear polymers). Obtaining the full, quantitative size distribu-
ing the progress of a reaction using NMR spectroscopy is less time-
tion of starch in a sample, without degradation, is not a simple
consuming than current assay techniques, and yields large data
matter (Cave, Seabrook, Gidley, & Gilbert, 2009; Gidley et al.,
sets which include the initial RDS stage. Investigation of RDS is sig-
2010; Gray-Weale & Gilbert, 2009; Hoang et al., 2008; Schmitz
nicant as the initial rate of absorption of sugar into the blood is
et al., 2009). When it becomes possible to achieve routinely such
determined by the RDS stage, and the total sugar content of a meal,
undegraded distributions, it is anticipated that such data, as a func-
thus directly affecting the in vivo measurement of starch digestion.
tion of starch source and of time and site of digestion, will be of
NMR spectroscopy, while able to monitor rigorously the kinet-
considerable help in understanding digestive mechanisms.
ics of slow (minute time scale) reactions in solution, has some
Fluorescence-assisted capillary electrophoresis (FACE) (Morell,
drawbacks. NMR equipment is expensive making it unsuitable
Samuel, & OShea, 1998; OShea et al., 1998), and high-performance
for routine use as a large-scale industrial technique. Also, most
anion-exchange chromatography (HPAEC) are used to obtain the
applications of NMR require the solvents to be deuterated, adding
distributions of oligosaccharide products, including the chain-
to its expense.
length distribution of the amylopectin after treatment with a
debranching enzyme (Manelius, Nurmi, & Bertoft, 2000). These
techniques are restricted to relatively low degrees of polymerisa- 8.2. MichaelisMenten kinetics
tion (less than 80 anhydroglucose units) and so cannot character-
ise chain-lengths distributions of amylose. The latter can be MichaelisMenten kinetics describes the action of many reac-
determined by using SEC. tions where the enzyme concentration is small relative to the sub-
Fig. 7. 1H NMR (400.13 MHz) spectra displaying the time dependence of digestion of starch (4% w/w) with: (A) a-amylase (3 U mL1); and (B) amyloglucosidase (0.3 U mL1).
Adapted from Dona et al. (2009). The signal from both a and b-anomer reducing ends allow for measurement of the concentration of mono- (amyloglucosidase) or
oligosaccharide (a-amylase). The a (1,4) bond signal at 5.4 ppm increases due to oligosaccharide production by a-amylase (A), and decreases as amyloglucosidase (B)
hydrolyses all accessible a (1,4) bonds.
strate concentration. The derivation of the MichaelisMenten tion of starch hydration in solution during digestion affects the
equation assumes that a slow, product forming reaction follows accessibility of the enzyme to the starch.
the rapid, reversible formation of an enzymesubstrate complex Substrate branching characteristics have been shown to inu-
(Eq. (1)), ence depolymerisation reaction rates, as enzyme susceptibility de-
creases with increased branching density (Ao et al., 2007; French &
k1 k2
E S
ES ! E P; 1 Knapp, 1950; Kerr, Cleveland, & Katzbeck, 1951) (Section 9). In
k1 some cases, kinetic models that account for the multiple phases
of starch degradation are used to give an explanation for the vari-
where E is the enzyme, S is the substrate (starch) and P is the prod- ations in starch hydrolysis rates with the changes in branch density
uct. The MichaelisMenten equation (Eq. (2)) is then derived by of the starch sub-fractions (Dona et al., 2009; Park & Rollings,
using the steady-state approximation for the ES complex: speci- 1995) (Section 8.7). In an attempt to elucidate features of starch
cally the concentration of the enzymesubstrate complex is as- digestion, many other single-phase models have been proposed
sumed to change much more slowly than the concentration of the to describe the time courses, although they are generally found
substrate, so the rate equation takes the form, to be inadequate (Section 8.48.6).
dP V max S
; 2
dt K m S 8.3. Solutions of the MichaelisMenten equation
where Km is the Michaelis constant (Eq. (3)), and Vmax is the maxi- As mentioned previously, the kinetic constants of the Michae-
mum velocity of the reaction achieved when the enzyme active lisMenten equation can be estimated by using LineweaverBurk
sites in the sample are all complexed with substrate all the time, plots. Although this analysis avoids solving the MichaelisMenten
and [P] is the concentration of product at any given time during equation, the double reciprocal plot has a large systematic error.
the time course. The relationship between the Km and the unitary Fitting the transformation of an inherently nonlinear equation to
rate constants in the reaction scheme is: experimental data distorts the errors in the measured variables,
subsequently impacting on the veracity of estimates of the kinetic
k1 k2
Km : 3 parameters. A more accurate graphical analysis of enzyme kinetics,
k1 the HanesWoolf plot, graphs the ratio of the initial substrate con-
Taking the reciprocal of both sides of the MichaelisMenten centration to the reaction velocity plotted against substrate con-
equation (Eq. (2)) gives the LineweaverBurk equation (Eq. (4)) centration (Nitta, Kunikata, & Watanabe, 1979). Even more
that is often used to graphically analyse enzyme kinetic data. The statistically robust is the direct linear plot method of Eisenthal
equation is: and Cornish-Bowden (1974). Direct linear plots are created by plot-
ting experimental observations in parameter space and extrapolat-
1 Km 1 ing lines through the observed points to a point of common
: 4
V V max S V max intersection. Errors associated with experimental observations
normally mean there is no unique intersection point; however,
This relationship was used to estimate Vmax and Km values be-
the most accurate estimates of the Michaelis parameters from di-
fore the days of nonlinear least squares regression of Eq. (2) di-
rect linear plots are generally easily located (Eisenthal & Cornish-
rectly onto the data. The latter approach is now routinely used
Bowden, 1974).
with readily available software packages (e.g., Castillo, Hadi, &
Integration of the MichaelisMenten equation (Eq. (2)) yields an
Minguez, 2009).
expression that relates time (t) to the concentration of substrate [S]
Several time courses measured under identical conditions, ex-
at any given time. The solution therefore describes a reaction time
cept for the amount of enzyme used, should be superimposable
course:
when the time axis is scaled by the initial enzyme concentration.
This is known as Selwyns test (Selwyn, 1965); failure to pass this V max t S0 St K m lnS0 =St ; 5
test means that the shape of the time course curve is not entirely
governed by product formation or substrate consumption during where [S]0 is the initial substrate concentration. The solution is non-
the enzyme-catalysed reaction. Commonly, instability of the en- linear and clearly implicit with respect to the substrate concentra-
zyme, or its aggregation, is responsible for this failure, although tion. An explicit form of the integrated MichaelisMenten equation
instability of substrate or product could also be factors. These (Eq. (5)) is often approximated by introducing reasonable con-
problems are often rectied by changing the reaction conditions. straints, causing only one of the terms on the right hand side to dic-
Expressions have been derived to describe the initial reaction rate tate its value. If [S]0 << Km, then the rst term dominates, restricting
of a simple enzyme with a single substrate and unknown mecha- the value of the solution, indicating that the depletion of the sub-
nism of inactivation (Schnell & Hanson, 2007). Unlike mecha- strate is approximated by a single exponential function. If the oppo-
nism-based enzyme inactivation, product formation and enzyme site occurs such that Km << [S]0, then the solution becomes a simple
inactivation are considered to occur independently of one another. zero-order decay where the apparent rate constant that describes
Hydrolysis of carbohydrates of similar properties to amylopec- the reaction is independent of the extent of reaction.
tin, such as glycogen and many small soluble oligosaccharides, The product-inhibition model for enzymic action (Eq. (9)) can
by glucohydrolases can be described by MichaelisMenten kinetics also be integrated to produce, like the integrated MichaelisMen-
(Table 1) (Inokuchi, Takahashi, & Irie, 1981; Miranda, Murado, San- ten equation, a nonlinear expression which is implicit in the sub-
roman, & Lema, 1991). On the other hand, digestion of starch, and strate concentration:
amylopectin samples digested with either a-amylase or amyloglu-

cosidase, show clear deviation of time courses from those pre- Km K m S0
V max t S0 St 1 Km lnS0 =St ; 6
dicted by the MichaelisMenten equation. This deviation cannot Ki Ki
be a property of the enzyme, so it is concluded that the deviation
is a consequence of either a chemical property of the substrate, where Ki is the product inhibition constant. In the special case
such as an indigestible starch core, or a chemical property of the where Ki is equal to Km, the integrated solution becomes a pure sim-
starch, such as its time-dependent interaction with water. Varia- ple exponential decay (Kuchel, 1985; Kuchel & Ralston, 1997):
V max t
S S0 eK m S0 ; 7 veal a grace zone where inhibition does not occur below a critical
concentration of product (Lim et al., 1999; Wang et al., 2006).
The implicit nature of both Eqs. (5) and (6) explains the use of However, the critical glucose concentration is much greater than
classical, graphical approaches to analyse experimental data, or that of glucose generated during typical starch digestion. Further-
more recently numerical approaches. Alternative techniques to more, product inhibition, if occurring mechanistically, will take
compute the substrate concentration as a function of time, such place regardless of the product concentration and does not corre-
as root-solving including the bisection method or NewtonRaph- spond well with the notion of a grace period. Studies on the inhi-
son root approximation, have been applied to analyse enzyme time bition of the main products of a-amylase, maltose and maltotriose,
courses (Scheid, 1988; Tjalling, 1995). Other approximation meth- are inconclusive as to whether the inhibition by these products is
ods such as the fourth order RungeKutta integration, and a low- competitive or non-competitive (Kazaz, Desseaux, Marchis-Mou-
order recursive method known as the decomposition method ren, Prodanov, & Santimone, 1998).
(Sonnad & Goudar, 2009) have provided a simple alternative to Experiments characterising inhibition of enzymic starch hydro-
numerical integration. However recently, using the Lambert W lysis use various amounts of glucose added to a starch solution be-
function solution (Goudar, Harris, McInerney, & Suita, 2004), the fore adding glucoamylase. At the concentrations of glucose
MichaelisMenten equation can be written in an explicit form in (<100 mmol L1) produced during standard starch digestions, no
substrate concentration: signicant change is observed in the initial velocity of the hydro-
lytic reaction (Wang et al., 2006). Similarly, the study of immobi-
S0 S0 V max t
S K m W e Km ; 8 lised a-amylase inactivation during the hydrolysis of starch
Km
particles shows the same relative enzymic activity for long periods
where W is the Lambert omega function. This solution allows a sys- after the beginning of digestion (Lim et al., 1999). Furthermore,
tematic analysis of the errors associated with the different approx- models of product-inhibited enzymic reactions still produce
imate methods of integrating the MichaelisMenten equation smooth time courses (of a different shape from that of Michae-
(Sonnad & Goudar, 2009). lisMenten reactions (Eq. (2))) not explaining the sharper transi-
Since numerical integration has become a routine procedure, tions that are seen during digestion of starch suspensions
multiple iterations of numerical integration can be applied to the (Akerberg et al., 2000). Evidence is therefore against product inhi-
analysis of experimental data from enzymic reactions. Thus non- bition as the explanation for deviations of MichaelisMenten
linear least squares regression analysis of the iterated solutions kinetics from real carbohydrate-polymer digestion curves.
can be used to determine the best-t estimates of the constants
in the MichaelisMenten equation. This data tting technique is 8.5. Substrate inhibition
constrained by the desirability of data being collected over the en-
tire reaction time course. By comparison, the LineweaverBurk Inhibition of enzymic processes is normally either competitive,
plot uses only the initial velocities of reactions over a range of sub- where the inhibitor binds to the enzyme competitively with the
strate concentrations. substrate, or non-competitive, in which case the inhibitor has iden-
tical afnities for the enzyme and the enzymesubstrate complex,
8.4. Product inhibition decreasing the maximum velocity of the reaction. Another special
case is uncompetitive inhibition where the inhibitor binds only
Deviations from MichaelisMenten kinetics in the early stages with the enzymesubstrate complex (Kuchel & Ralston, 1997).
of hydrolysis of starch by both endohydrolases and exohydrolases The binding of the inhibitor is at a subsidiary site on the enzyme,
have been reported (Akerberg et al., 2000; Ao et al., 2007; Apar & causing the catalytic action of the enzyme to be altered, and slow-
Ozbek, 2007; Wang et al., 2006; Zhang et al., 2006). After the initial ing the formation of product.
stage of starch digestion, time courses predicted theoretically by High-substrate inhibition is a form of uncompetitive inhibition
integrating the MichaelisMenten equation deviate from the that is characterised by a decrease in the reaction rate at high sub-
experimental data. Various hypotheses have been proposed to ex- strate concentrations. The kinetic behaviour of high-substrate
plain this outcome. Based on these hypotheses, existing variations inhibited enzymic processes follows the BriggsHaldane model
to the MichaelisMenten model, or new kinetic models, have been (Briggs & Haldane, 1925):
developed in order to explain starch digestion (Table 1).
Competitive-inhibition occurs when the binding of a ligand to dP V max S
; 10
the enzyme prevents binding of the substrate to the active site, dt K m S K S S2
and vice versa. Used previously to explain the kinetics of many en-
zymesubstrate systems, the competitive-inhibition model has where Ks is the high-substrate inhibition constant. The quadratic
been used to describe product inhibition of glucoamylase by glu- term is not interpreted mechanistically; it is only justied phenom-
cose (Fujii & Kawamura, 1985; Lim, Lee, Shin, & Lim, 1999; Wang enologically and many studies have used this model to describe
et al., 2006) (Eq. (9)): starch hydrolysis (Lopez, Torrado, Fucinos, Guerra, & Pastrana,
2006; Miranda et al., 1991; Pastrana, Gonzalez, Miron, & Murado,
dP V S 1998).
max : 9
dt K m 1 P S Models of substrate inhibiting a-amylase action on starch have
Ki
been based upon the uncompetitive mechanism in recent research
The parameter estimates obtained from tting experimental (Table 1) (Lopez et al., 2006; Pastrana et al., 1998). However, the
data of carbohydrate digestion with a product-inhibition model less commonly used competitive model of substrate inhibition is
are generally more precise. Whether the tting is improved be- phenomenologically a more likely explanation for the effect
cause product inhibition occurs, or because the number of oating (Fig. 8). Given the numerous sub-sites for glucose residues at the
variables in the theoretical model has increased, is an important catalytic centre of a-amylase (Fig. 8), ineffective binding or partial
consideration in the data analysis (Fig. 6). Studies that model occupancy of the sub-sites is more probably responsible for inhibi-
starch digestion using product inhibition assume an effect irre- tion than an uncompetitive process.
spective of the concentration of the product (Fujii & Kawamura, Physical properties of the solutions, such as viscosity and diffu-
1985). However, studies on both glucoamylase and a-amylase re- sional restrictions, have also been used to explain high-substrate
Fig. 8. Phenomenological interpretation of how the binding sites of a-amylase are bound during either product or substrate inhibition. The enzyme is rst shown unbound,
then regular binding without inhibition is presented. Ineffective binding shows how the enzyme might be bound in the instances of competitive-inhibition. Binding at the
subsidiary site, that opens up only on occupation of the catalytic site by the substrate, brings about uncompetitive inhibition. (Based on Davies, Wilson, & Henrissat, 1997;
Yoon & Robyt, 2003).
inhibition. In polysaccharide solutions, the concentration of solute (Hill, Macdonald, & Lang, 1997; Lopez et al., 2006). The Linewe-
strongly affects rheological properties. It is therefore reasonable to averBurk plots obtained from inhibition experiments have not
assume that particular kinetic proles are at least partially condi- been able to distinguish between the two mechanisms. Research
tioned by diffusional restrictions of the reaction mixture. The typ- characterising the binding of substrate molecules to the enzyme
ical MichaelisMenten differential equation is multiplied by a sub-sites, followed by subsequent reorganisation of the enzyme
decaying exponential term to take diffusional restriction into ac- substrate complex, has not been able to shed light on this question
count. Thus the reaction rate decreases asymptotically as the sub- either. This leads to multiple models still being used to describe the
strate concentration increases (Eq. (11)) (Pastrana et al., 1998), reaction mechanism of glucoamylase (Matsumura, Hirata, Ishii, &
Kobayashi, 1988; Natarajan & Sierks, 1997).
dP V max S Dv R
e ; 11
dt K m S
8.6. Empirical exponential models
where Dv is a rst order coefcient that characterises the relation-
ship between reaction velocity and diffusional restriction, and R ex- Given the above difculties with characterising starch digestion
presses the diffusional restriction (Pastrana et al., 1998). Thus, using product/high-substrate-inhibition models, other empirical
exponential models have been developed (Apar & Ozbek, 2007; Hill
R Rm R0 Rm elS ; 12 et al., 1997; Wang et al., 2006). Moreover, exponential models al-
low for easy manipulation, enabling different types of product or
where R0 is the diffusional restriction when [S] = 0, Rm is the largest
substrate inhibition to be added (Table 1). A grace zone (Sec-
value of diffusional restriction, and l is a specic coefcient for dif-
tion 8.4) where inhibition only occurs after a limiting value of
fusional restriction.
product has been formed can also be added to an empirical expo-
Combinations of product and substrate inhibition are used in
nential model; this is another reason for its use. Although easily
models to t digestion data from experiments where estimates
controlled, variations of exponential models used to t digestion
are not satisfactory with single mechanisms (Lopez et al., 2006).
time courses have no mechanistic basis.
Although both types of inhibition might well occur during starch
Hill et al. (1997) proposed an exponential model which ac-
digestion, as mentioned above, increasing the number of tting
counts for product inhibition as follows,
parameters in models of catalysis will always lead to a more pre-

cise t to the experimental data; this occurs at the expense of high dS dS
eK i P ; 13
coefcients of variation on parameter estimates. dt dt P0
Assignment of the reported inhibition by glucose of glucoamy-
lase as either competitive or non-competitive has proven difcult where Ki is the product inhibition constant.
This model was extended to incorporate a threshold concentra- supposes that enzyme attack is simultaneous for both linear and
tion of product, below which product inhibition does not occur: branch points, although it assumes different reaction rates of
hydrolysis for the different domains of amylopectin.
dS dS
eK i PPth ; 14 The nal model described here uses multiple applications of the
dt dt P0 MichaelisMenten equation to account for the various rates of
hydrolysis by a-amylase and amyloglucosidase (Akerberg et al.,
where [P]th is the threshold concentration of product below which
2000). The model simulates digestion by using various rates and
product inhibition does not occur. Introduction of this constant en-
binding afnities of oligosaccharides of different lengths, produced
abled a closer t of the equation to the multiple stages that are evi-
by a-amylase and amyloglucosidase. Although this analysis con-
dent during in vitro starch digestion. Often, to make the model more
siders both types of enzyme, and all lengths and complexities of
able to t the data a factor is included to account for the fact that 1 g
substrate, the system of starch digestion is, perhaps, better ana-
of starch is converted into 1.11 g of glucose (the extra mass being
lysed under simplied in vitro conditions, such as with single en-
from the added water) during digestion (Wang et al., 2006).
zymes, or with well dened oligosaccharides.
Another exponential model (Eq. (15)) introduced by Goni et al.
(1997) has been used to described closely the digestion of both
cooked (Frei et al., 2003) and raw (Al-Rabadi, Gilbert, & Gidley,
9. Starch properties affecting digestion
2009) grain starch samples. Specically,
C C 1 1 ekt ; 15 The kinetics of enzymic starch digestion depends largely on two

factors. The rst is the molecular architecture and physicochemical
where C corresponds to the concentration of glucose at any time, C1 characteristics of the starch granule, at all six levels of structure
is the concentration of glucose at the completion of the reaction, (Ao et al., 2007; French & Knapp, 1950; Kerr et al., 1951; Kruger
and k is a rst order kinetic constant. This equation is simply the & Marchylo, 1985) (Fig. 1; Section 3). Research correlating the
solution of the relevant rst order differential equation. botanical origin, largely responsible for the architecture of the
This same model has been used by Al-Rabadi et al. (2009) to de- starch granule, with both in vivo and in vitro digestibility is ad-
scribe the digestion by a-amylase of raw starch grain from both vanced; this is a correlation and does not follow the causal rela-
barley and sorghum. Results indicated that digestion of these raw tionship (biosynthesis causes structural properties that
grains was controlled by diffusion of enzymes within the porous determine digestibility) (Goni et al., 1997; Jenkins et al., 2002;
grain (Al-Rabadi et al., 2009). The simple exponential models used Thompson, 2000). The second is the hydration or physical confor-
to t both the degradation of starch and the production of oligosac- mations that starch granules assume in aqueous solution (Sec-
charide closely simulates the digestion process over the initial tion 4); these conformations control digestibility. Both heat and
stage (RDS stage) (Apar & Ozbek, 2007; Komolprasert & Ofoli, pressure during the preparation of starch suspensions alter the
1991) or over the entirety of the reaction (>24 h, or more correctly progression towards gelatinisation, which increases the availabil-
a large extent of reaction) where the RDS stage becomes compar- ity of polysaccharide starch chains to digestive enzymes, thus
atively insignicant (Al-Rabadi et al., 2009). Like the deviations affecting rates of hydrolysis (Chung, Yoo, & Lim, 2005; Chung
from the MichaelisMenten model, attempts to simulate digestion et al., 2006; Eerlingen, Jacobs, & Delcour, 1994; Holm, Lundquist,
time courses, where both stages of starch digestion are carefully Bjorck, Eliasson, & Asp, 1988).
monitored, have proven difcult (Dona et al., 2009; Hill et al., Digestibility is strongly inuenced by shearing and heating the
1997; Wang et al., 2006). sample (Farhat et al., 2001; Guraya, James, & Champagne, 2001;
Shiotsubo, 1983). The relative amount of starch in each digestibil-
8.7. Multiple stage models ity fraction can be altered by a range of various techniques for
starch preparation (Fassler et al., 2006). The hydrolysis patterns
The obvious transition in the velocity of a reaction from being of samples that are partially gelled after heating at different tem-
high to being much lower after the initial stage of a digestion time peratures, behave very differently during the early stages of hydro-
course has led to the implementation of two/multi-stage kinetic lysis (Chung et al., 2006; Granfeldt, Eliasson, & Bjorck, 2000). The
models (Table 1). One such model (Dona et al., 2009) uses Michae- time required for oligosaccharide formation to reach a plateau de-
lisMenten kinetics (Eq. (2)) to t two stages of starch digestion by creases as the temperature of heat treatment is increased, indicat-
using LineweaverBurk analysis to estimate Vmax and Km values ing that starch in its native state is very resistant to digestion. This
that are pertinent to each of two stages of the time course and later property is explained by the disruption of inter- and intra-molecu-
the whole time course is simulated by numerically integrating the lar hydrogen bonds between starch chains, allowing the chains to
MichaelisMenten equation (Eq. (2)). The model successfully de- separate and become more accessible to enzymes. Progressive gel-
scribes time courses from various starch samples prepared under atinisation then increases the availability of polysaccharide chains
different conditions, and digested with different hydrolytic en- to the digestive enzymes (Chung et al., 2006). The extent of starch
zymes. The time courses are measured with 1H NMR spectroscopy gelatinisation is positively correlated with its in vitro digestibility
(Dona et al., 2009). and the in vivo plasma glucose response in rats (Holm et al., 1988).
Current models of the conformation of starch in solution, espe- Starch retrogradation has also been correlated to rates of en-
cially the hairy billiard ball, support the use of such a kinetic model zyme-catalysed hydrolysis; it depends on the time and tempera-
to describe digestion time courses. Furthermore, the parameters ture of sample storage (Eerlingen et al., 1994) and also on the
estimated can be substituted back into the MichaelisMenten dif- amyloseamylopectin ratio (Fredriksson et al., 2000). Both storage
ferential equation to closely describe the whole time course of pro- conditions and the fraction of amylose which affect re-crystallisa-
duction of oligosaccharides during starch digestion (Fig. 6). tion are correlated with the digestion rates of retrograded starch
A similar approach was taken for tting starch digestion (Park & samples. Retrogradation over periods of up to a week of gelatinisa-
Rollings, 1995) with multiple stages, although the rationale for tion, and partially gelled starch, show greater rates of hydrolysis
using a multi-stage model was different from the previous case. than native starch (Chung et al., 2005, 2006).
The susceptibility of linear and branch points on a carbohydrate The granular architecture established during starch synthesis in
to enzymic hydrolysis is the mechanistic basis for tting the reac- the grain also plays a role in determining the digestibility of starch.
tion over multiple stages (Park & Rollings, 1995). The kinetic model Amyloglucosidase and a-amylase show a decrease in the rate of
hydrolysis, as the digestion approaches the branch points of the a proper mechanistic basis, whether this involves a model with
polysaccharide (French & Knapp, 1950; Kerr et al., 1951). It has multiple stages or is explained by a continuous mathematical func-
been proposed that partially shortening the length of exterior tion (Chung et al., 2006; Cooke & Gidley, 1992; Dona et al., 2009).
branched chains slows their rate of hydrolysis. Modication of nor- In an effort to elucidate the ner details of the features of the
mal maize starch, by treatment with b-amylase and maltogenic a- digestion of starch, new analytical methodologies are being devel-
amylase, removes maltose residues and reduces the chain length oped (Cave et al., 2009; Gidley et al., 2010; Hoang et al., 2008; Pod-
on the exterior of the starch molecule. Digestion of this modied zimek, Lebedat, & Johann, 2009; Roger, Baud, & Colonna, 2001;
maize starch is slower than unmodied maize starch (Ao et al., Rolland-Sabate, Colonna, Mendez-Montealvo, & Planchot, 2007).
2007). Further modication by increasing the branch density on One such technique being improved upon is non-aqueous SEC of
the exterior of the starch granule using transglucosidase slows starch substrate samples (Cave et al., 2009; Hoang et al., 2008;
the rate of hydrolysis (Ao et al., 2007). Aligned with this concept, Podzimek et al., 2009), as quantitative recovery of starch samples
altering the branch density by concentrating the number of a from SEC is contentious (Ward, Gao, De Bruyn, Gilbert, & Fitzgerald,
(1,6) bonds or changing the chain distributions in amylopectin 2006). Other chromatography techniques, specically asymmetric
molecules decreases the rate of hydrolysis of starch by a-amylase ow-eld ow fractionation (AF4) (Roger et al., 2001; Rolland-Sa-
(Ao et al., 2007). bate et al., 2007), are becoming more generally applied to analyse
Although it appears that an increased branch density decreases hydrodynamic volume distributions of starch before and during
the rate of glucosidases, in the other extreme where branch density digestion. It has been proved that SEC can cause signicant shear
is very low, the rate of digestion is also reduced. Comparison of the degradation of the amylopectin component of starch (Cave et al.,
rate of hydrolysis of amylopectin with an equal weight-concentra- 2009). Problems arising from chromatography of starch such as
tion of amylose shows that the highly linear amylose chains yield incomplete recovery and shear scission of samples are expected
sugars more slowly than highly branched amylopectin (Kerr et al., to be solved by using a more gentle technique such as AF4. Finally,
1951). This is postulated to occur because of the very low concen- NMR methods such as chemical exchange saturation transfer
tration of reducing ends on amylose that are able to be acted upon (CEST) (Ward, Aletras, & Balaban, 2000; Zhou & van Zijl, 2006), nor-
by glucosidases. Thus, it is concluded that both linear polymers of mally used in imaging sugars and carbohydrates, is currently being
anhydroglucose and highly branched polymers of anhydroglucose applied to the digestion of starch revealing before unseen features
inhibit the action of hydrolytic enzymes, slowing rates of digestion of oligosaccharides during digestion (Dona, Pages, & Kuchel,
(Kerr et al., 1951). unpublished).
Granules of starch that have been separated by sedimentation Overall the future is bright for the quantitative analysis of
analysis into different sizes, and inspected by light microscopy, starch digestion, using advanced analytical methods, mathematical
have also shown differences in rates of digestion by a-amylase modeling of enzyme kinetics with complex substrates, and there-
(Fig. 4). Small granules were much more efciently hydrolysed fore prediction of nutritional outcomes. An in vitro counterpart of
by digestive enzymes compared with their large-granule counter- the GI seems plausible and is possibly imminent.
parts (Kruger & Marchylo, 1985). The rate of digestion is not only
much greater in the case of small granules but also the total per-
centage of starch digested over the duration of the reaction is Acknowledgments
greater in the case of small granules (Manelius et al., 1997). The
initial rates of hydrolysis of granules of various sizes are explained We thank the Australian Research Council for Discovery Project
by the dependency on the amount of enzyme that is absorbed into grants to PWK (DP0877789) and RGG (DP0985694). We also thank
the granule and thus the increased surface area of starch (McLaren, Prof. Mike Gidley for valuable discussions.
1963).
References
10. Future directions Aarathi, A., Urooj, A., & Puttaraj, S. (2003). In vitro starch digestibility and
nutritionally important starch fractions in cereals and their mixtures. Starch/
Staerke, 55, 9499.
There is increasing evidence that the nutritional value of food is
Akerberg, C., Zacchi, G., Torto, N., & Gorton, L. (2000). A kinetic model for enzymatic
controlled by the microstructure of the contained carbohydrate. wheat starch saccharication. Journal of Chemical and Biotechnology, 75,
The structure of starch plays an important role in determining its 306314.
accessibility to digestion, thus inuencing the time course of the Al-Rabadi, G. J. S., Gilbert, R. G., & Gidley, M. J. (2009). Effect of particle size on
kinetics of starch digestion in milled barley and sorghum grains by porcine a-
postprandial blood glucose response. The many structural features amylase. Journal of Cereal Science, 50, 198204.
of starch granules have led to numerous models for estimating Amato, M. E., Ansanelli, G., Fisichella, S., Lamanna, R., Scarlata, G., Sobolev, A. P., et al.
in vitro digestion rates, even under the simplest of experimental (2004). Wheat our enzymatic amylolysis monitored by in situ 1H NMR
spectroscopy. Journal of Agricultural and Food Chemistry, 52, 823831.
conditions. Progress towards correlating structural and physical Ao, Z., Simsek, S., Zhang, G., Venkatachalam, M., Reuhs, B. L., & Hamaker, B. R. (2007).
properties of the starch granule to its digestibility, by use of more Starch with a slow digestion property produced by altering its chain length,
advanced kinetic models, is evident (Dona et al., 2009). However, branch density, and crystalline structure. Journal of Agricultural and Food
Chemistry, 55, 45404547.
quantitative mechanistic account for the rate limiting step of Apar, D. K., & Ozbek, B. (2007). Estimation of kinetic parameters for rice starch
starch digestion, obtained with different preparation techniques, hydrolysis inhibited by added materials. Chemical Engineering Communications,
and at various stages during digestion, is still being sought. 194, 334344.
Apinan, S., Yujiro, I., Hidefumi, Y., Takeshi, F., Myllarinen, P., Forssell, P., et al. (2007).
Analysis of the time dependence on the structural and physical Visual observation of hydrolyzed potato starch granules by a-amylase with
properties of starch granules during the course of digestion is of confocal laser scanning microscopy. Starch/Staerke, 59, 543548.
great importance (Gernat et al., 1993; Wang et al., 2006). Unveiling Bae, J., Lee, D., Kim, D., Cho, S.-J., Park, J. E., Koh, S., et al. (2005). Facile synthesis of
glucose-1-phosphate from starch by Thermus caldophilus GK24 a-glucan
what physiological property controls the rate of starch granule
phosphorylase. Process Biochemistry, 40, 37073713.
digestion at various stages during, and under various conditions Baik, M.-Y., Kim, K.-J., Cheon, K.-C., Ha, Y.-C., & Kim, W.-S. (1997). Recrystallization
of preparation, is the next step towards completely understanding kinetics and glass transition of rice starch gel system. Journal of Agricultural and
the processes of glucose release by hydrolytic enzymes. A better Food Chemistry, 45, 42424248.
Ball, S., Guan, H. P., James, M., Myers, A., Keeling, P., Mouille, G., et al. (1996). From
understanding of the rate limiting property of granular digestion glycogen to amylopectin: A model for the biogenesis of the plant starch granule.
by hydrolases will allow the development of kinetic models with Cell, 86, 349352.
Barclay, A. W., Petocz, P., McMillan-Price, J., Flood, V. M., Prvan, T., Mitchell, P., et al. Eisenthal, R., & Cornish-Bowden, A. (1974). Direct linear plot. New graphical
(2008). Glycemic index, glycemic load, and chronic disease risk-a meta-analysis procedure for estimating enzyme kinetic parameters. Biochemical Journal, 139,
of observational studies. American Journal of Clinical Nutrition, 87, 627637. 715720.
BeMiller, J. N. (2007). Starches, modied food starches, and other products from Ellis, R. P., Cochrane, M. P., Dale, M. F. B., Duffus, C. M., Lynn, A., Morrison, I. M., et al.
food starches. In R. L. Whistler & J. N. BeMiller (Eds.), Carbohydrate chemistry for (1998). Starch production and industrial use. Journal of the Science of Food and
food scientists (pp. 173223). St Paul, MN: AACC International. Agriculture, 77, 289311.
Bello-Prez, L. A., Roger, P., Baud, B., & Colonna, P. (1998). Macromolecular features Englyst, H. N., Kingman, S. M., & Cummings, J. H. (1992). Classication and
of starches determined by aqueous high-performance size exclusion measurement of nutritionally important starch fractions. European Journal of
chromatography. Journal of Cereal Science, 27, 267278. Clinical Nutrition, 46, 3350.
Belton, P. S., Delgadillo, I., Halford, N. G., & Shewry, P. R. (2006). Karin structure and Englyst, H. N., & Hudson, G. J. (1996). The classication and measurement of dietary
functionality. Journal of Cereal Science, 44, 272286. carbohydrates. Food Chemistry, 57, 1521.
Benmoussa, M., Moldenhauer, K. A. K., & Hamaker, B. R. (2007). Rice amylopectin Farhat, I. A., Protzmann, J., Becker, A., Valles-Pamies, B., Neale, R., & Hill, S. E. (2001).
ne structure variability affects starch digestion properties. Journal of Effect of the extent of conversion and retrogradation on the digestibility of
Agricultural and Food Chemistry, 55, 14751479. potato starch. Starch/Staerke, 53, 431436.
Berthon, H. A., & Kuchel, P. W. (1995). NMR studies of erythrocyte metabolism. In K. Fassler, C., Arrigoni, E., Venema, K., Hafner, V., Brouns, F., & Amado, R. (2006).
M. Brindle (Ed.), Advances in molecular and erythrocyte cell biology Digestibility of resistant starch containing preparations using two in vitro
(pp. 147205). Greenwich, CT: JAI Press Inc. models. European Journal of Nutrition, 45, 445453.
Bertoft, E., & Manelius, R. (1992). A method for the study of the enzymic hydrolysis Fredriksson, H., Silverio, J., Andersson, R., Eliasson, A. C., & Aman, P. (1998). The
of starch granules. Carbohydrate Research, 227, 269283. inuence of amylose and amylopectin characteristics on gelatinization and
Bhattacharya, K. R., Sowbhagya, C. M., & Swamy, Y. M. I. (1972). Interrelation retrogradation properties of different starches. Carbohydrate Polymers, 35,
between certain physicochemical properties of rice. Journal of Food Science, 37, 119134.
733735. Fredriksson, H., Bjorck, I., Andersson, R., Liljeberg, H., Silverio, J., Eliasson, A. C., et al.
Bhattacharya, K. R., Sowbhagya, C. M., & Swamy, Y. M. I. (1978). Importance of (2000). Studies on a-amylase degradation of retrograded starch gels from waxy
insoluble amylose as a determinant of rice quality. Journal of the Science of Food maize and high-amylopectin potato. Carbohydrate Polymers, 43, 8187.
and Agriculture, 29, 359364. Frei, M., Siddhuraju, P., & Becker, K. (2003). Studies on the in vitro starch
Biliaderis, C. G., Page, C. M., Maurice, T. J., & Juliano, B. O. (1986). Thermal digestibility and the glycemic index of six different indigenous rice cultivars
characterization of rice starches: A polymeric approach to phase transitions of from the Philippines. Food Chemistry, 83, 395402.
granular starch. Journal of Agricultural and Food Chemistry, 34, 614. French, D., & Knapp, D. W. (1950). The maltase of Clostridium acetobutylicum. Its
Bogracheva, T. Y., Wang, T. L., & Hedley, C. L. (2001a). Implications of genetic specicity range and mode of action. The Journal of Biological Chemistry, 187,
changes in starch granular structure to gelatinization behavior. Special 463471.
Publication Royal Society of Chemistry, 271, 7781. Fujii, M., & Kawamura, Y. (1985). Synergistic action of a-amylase and glucoamylase
Bogracheva, T. Y., Wang, Y. L., & Hedley, C. L. (2001b). The effect of water content on on hydrolysis of starch. Biotechnology and Bioengineering, 27, 260265.
the ordered/disordered structures in starches. Biopolymers, 58, 247259. Garsetti, M., Vinoy, S., Lang, V., Holt, S., Loyer, S., Brand-Miller, J., et al. (2005). The
Bogracheva, T. Y., Wang, Y. L., Wang, T. L., & Hedley, C. L. (2002). Structural studies of glycemic and insulinemic index of plain sweet biscuits: Relationships to in vitro
starches with different water contents. Biopolymers, 64, 268281. starch digestibility. Journal of the American College of Nutrition, 24, 441447.
Boron, W. F., & Boulpaep, E. L. (2009). Medical physiology. Saunders. Gernat, C., Radosta, S., Anger, H., & Damaschun, G. (1993). Crystalline parts of three
Brand-Miller, J. C., Holt, S. H. A., Pawlak, D. B., & McMillan, J. (2002). Glycemic index different conformations detected in native and enzymically degraded starches.
and obesity. American Journal of Clinical Nutrition, 76, 281S285S. Starch/Staerke, 45, 309314.
Brand, J. C., Nicholson, P. L., Thorburn, A. W., & Truswell, A. S. (1985). Food Gidley, M. J., Hanashiro, I., Hani, N. M., Hill, S. E., Huber, A., Jane, J.-L., et al. (2010).
processing and the glycemic index. American Journal of Clinical Nutrition, 42, Reliable measurements of the size distributions of starch molecules in solution:
11921196. Current dilemmas and recommendations. Carbohydrate Polymers, 79, 225261.
Briggs, G. E., & Haldane, J. B. (1925). A note on the kinetics of enzyme action. The Goni, I., Garcia-Alonso, A., & Saura-Calixto, F. (1997). A starch hydrolysis procedure
Biochemical Journal, 19, 338339. to estimate glycemic index. Nutrition Research (New York), 17, 427437.
Buleon, A., Colonna, P., Planchot, V., & Ball, S. (1998). Starch granules: Structure and Goudar, C. T., Harris, S. K., McInerney, M. J., & Suita, J. M. (2004). Progress curve
biosynthesis. International Journal of Biological Macromolecules, 23, 85112. analysis for enzyme and microbial kinetic reactions using explicit solutions
Castillo, E., Hadi, A. S., & Minguez, R. (2009). Diagnostics for non-linear regression. based on the Lambert W function. Journal of Microbiological Methods, 59,
Journal of Statistical Computation and Simulation, 79, 11091128. 317326.
Cave, R. A., Seabrook, S. A., Gidley, M. J., & Gilbert, R. G. (2009). Characterisation of Granfeldt, Y., Eliasson, A. C., & Bjorck, I. (2000). An examination of the possibility of
starch by size-exclusion chromatography: The limitations imposed by shear lowering the glycemic index of oat and barley akes by minimal processing. The
scission. Biomacromolecules, 10, 22452253. Journal of Nutrition, 130, 22072214.
Chiou, H., Fellows, C. M., Gilbert, R. G., & Fitzgerald, M. A. (2005). Study of rice-starch Gray-Weale, A., & Gilbert, R. G. (2009). General description of the structure of
structure by dynamic light scattering in aqueous solution. Carbohydrate branched polymers. Journal of Polymer Science Part A Polymer Chemistry, 47,
Polymers, 61, 6171. 39143930.
Chung, H.-J., Yoo, B., & Lim, S.-T. (2005). Effects of physical aging on thermal and Guraya, H. S., James, C., & Champagne, E. T. (2001). Effect of cooling, and freezing on
mechanical properties of glassy normal corn starch. Starch/Staerke, 57, 354362. the digestibility of debranched rice starch and physical properties of the
Chung, H.-J., Lim, H. S., & Lim, S.-T. (2006). Effect of partial gelatinization and resulting material. Starch/Staerke, 53, 6474.
retrogradation on the enzymatic digestion of waxy rice starch. Journal of Cereal Heitmann, T., Wenzig, E., & Mersmann, A. (1997). Characterization of three different
Science, 43, 353359. potato starches and kinetics of their enzymic hydrolysis by an a-amylase.
Cooke, D., & Gidley, M. J. (1992). Loss of crystalline and molecular order during Enzyme and Microbial Technology, 20, 259267.
starch gelatinisation: Origin of the enthalpic transition. Carbohydrate Research, Hill, G. A., Macdonald, D. G., & Lang, X. (1997). a-Amylase inhibition and
227, 103112. inactivation in barley malt during cold starch hydrolysis. Biotechnology Letters,
Copeland, L., Blazek, J., Salman, H., & Tang, M. C. (2009). Form and functionality of 19, 11391141.
starch. Food Hydrocolloids, 23, 15271534. Hizukuri, S. (1985). Relationship between the distribution of the chain length of
Csoti, I., Bako, A., Tamas, L., & Gardonyi, M. (2005). Investigation of the starch- amylopectin and the crystalline structure of starch granules. Carbohydrate
binding properties of wheat friabilin. Acta Biologica Szegediensis, 49, 2930. Research, 141, 295306.
Das, S. K., Gilhooly, C. H., Golden, J. K., Pittas, A. G., Fuss, P. J., Cheatham, R. A., et al. Hoang, N.-L., Landol, A., Kravchuk, A., Girard, E., Peate, J., Hernandez, J. M., et al.
(2007). Long-term effects of 2 energy-restricted diets differing in glycemic load (2008). Toward a full characterization of native starch: Separation and
on dietary adherence, body composition, and metabolism in CALERIE: A 1-y detection by size-exclusion chromatography. Journal of Chromatography A,
randomized controlled trial. American Journal of Clinical Nutrition, 85, 1205, 6070.
10231030. Hoebler, C., Devaux, M. F., Karinthi, A., Belleville, C., & Barry, J. L. (2000).
Davies, G. J., Wilson, K. S., & Henrissat, B. (1997). Nomenclature for sugar-binding Particle size of solid food after human mastication and in vitro simulation
subsites in glycosyl hydrolases. Biochemical Journal, 321, 557559. of oral breakdown. International Journal of Food Sciences and Nutrition, 51,
DeVries, J. W. (2007). Glycemic index: The analytical perspective. Cereal Foods 353366.
World, 52, 4549. Holm, J., Lundquist, I., Bjorck, I., Eliasson, A. C., & Asp, N. G. (1988). Degree of starch
Dona, A., Yuen, C.-W. W., Peate, J., Gilbert, R. G., Castignolles, P., & Gaborieau, M. gelatinization, digestion rate of starch in vitro, and metabolic response in rats.
(2007). A new NMR method for directly monitoring and quantifying the American Journal of Clinical Nutrition, 47, 10101016.
dissolution kinetics of starch in DMSO. Carbohydrate Research, 342, Htoon, A., Shrestha, A. K., Flanagan, B. M., Lopez-Rubio, A., Bird, A. R., Gilbert, E. P.,
26042610. et al. (2009). Effects of processing high amylose maize starches under controlled
Dona, A. C., Pages, G., Gilbert, R. G., Gaborieau, M., & Kuchel, P. W. (2009). Kinetics of conditions on structural organisation and amylase digestibility. Carbohydrate
in vitro digestion of starches monitored by time-resolved 1H nuclear magnetic Polymers, 75, 236245.
resonance. Biomacromolecules, 10, 638644. Inokuchi, N., Takahashi, T., & Irie, M. (1981). Purication and characterization of a
Donald, A. M. (2004). Understanding starch structure and functionality. Starch in minor glucoamylase from Aspergillus saitoi. Journal of Biochemistry (Tokyo,
Food, 156, 184. Japan), 90, 10551067.
Eerlingen, R. C., Jacobs, H., & Delcour, J. A. (1994). Enzyme-resistant starch. V. Effect Jagannath, J. H., Jayaraman, K. S., Arya, S. S., & Somashekar, R. (1998). Differential
of retrogradation of waxy maize starch on enzyme susceptibility. Cereal scanning calorimetry and wide-angle X-ray scattering studies of bread staling.
Chemistry, 71, 351355. Journal of Applied Polymer Science, 67, 15971603.
Jane, J.-l., Wong, K.-s., & McPherson, A. E. (1997). Branch-structure difference in MacFarlane, S., & MacFarlane, G. T. (2006). Composition and metabolic activities of
starches of A- and B-type X-ray patterns revealed by their Naegeli dextrins. bacterial biolms colonizing food residues in the human gut. Applied and
Carbohydrate Research, 300, 219227. Environmental Microbiology, 72, 62046211.
Jane, J., Chen, Y. Y., Lee, L. F., McPherson, A. E., Wong, K. S., Radosavljevic, M., Manelius, R., Qin, Z., Avall, A. K., Andtfolk, H., & Bertoft, E. (1997). The mode of
et al. (1999). Effects of amylopectin branch chain length and amylose content action of granular wheat starch by bacterial a-amylase. Starch/Staerke, 49,
on the gelatinization and pasting properties of starch. Cereal Chemistry, 76, 142147.
629637. Manelius, R., Nurmi, K., & Bertoft, E. (2000). Enzymic and acidic hydrolysis of
Jane, J. L., & Robyt, J. F. (1984). Structure studies of amylose-V complexes and cationized waxy maize starch granules. Cereal Chemistry, 77, 345353.
retrograded amylose by action of a-amylases, and a new method for preparing Matser, A. M., & Steeneken, P. A. M. (1997). Rheological properties of highly cross-
amylodextrins. Carbohydrate Research, 132, 105118. linked waxy maize starch in aqueous suspensions of skim milk components.
Jenkins, D. J., Wolever, T. M., Taylor, R. H., Barker, H., Fielden, H., Baldwin, J. M., et al. Effects of the concentration of starch and skim milk components. Carbohydrate
(1981). Glycemic index of foods: A physiological basis for carbohydrate Polymers, 32, 297305.
exchange. American Journal of Clinical Nutrition, 34, 362366. Matsumura, M., Hirata, J., Ishii, S., & Kobayashi, J. (1988). Kinetics of saccharication
Jenkins, D. J. A., Kendall, C. W. C., Augustin, L. S. A., Franceschi, S., Hamidi, M., of raw starch by glucoamylase. Journal of Chemical Technology and Biotechnology,
Marchie, A., et al. (2002). Glycemic index: Overview of implications in health 42, 5167.
and disease. American Journal of Clinical Nutrition, 76, 266S273S. McLaren, A. D. (1963). Enzyme reactions in structurally restricted systems. IV. The
Jenkins, P. J., Cameron, R. E., Donald, A. M., Bras, W., Derbyshire, G. E., Mant, G. R., digestion of insoluble substrates by hydrolytic enzymes. Enzymologia, 26,
et al. (1994). In situ simultaneous small and wide angle X-ray scattering: A new 237246.
technique to study starch gelation. Journal of Polymer Science Part B Polymer Miles, M. J., Morris, V. J., Orford, P. D., & Ring, S. G. (1985). The roles of amylose and
Physics, 32, 15791583. amylopectin in the gelation and retrogradation of starch. Carbohydrate Research,
Jouppila, K., Kansikas, J., & Roos, Y. H. (1998). Factors affecting crystallization and 135, 271281.
crystallization kinetics in amorphous corn starch. Carbohydrate Polymers, 36, Minerath, E. C., Casale, M. T., & Elrod, M. J. (2008). Kinetics feasibility study of
143149. alcohol sulfate esterication reactions in tropospheric aerosols. Environmental
Katz, J. R. (1930). Essays on the physical chemistry of starch and bread production I. Science & Technology, 42, 44104415.
Changes in the X-ray spectrum of starch during baking and when bread is Miranda, M., Murado, M. A., Sanroman, A., & Lema, J. M. (1991). Mass transfer
becoming frumpy. Zeitschrift Fur Physikalische Chemie-Abteilung a-Chemische control of enzymic hydrolysis of polysaccharides by glucoamylase. Enzyme and
Thermodynamik Kinetik Elektrochemie Eigenschaftslehre, 150, 3759. Microbial Technology, 13, 142147.
Kazaz, M. A., Desseaux, V., Marchis-Mouren, G., Prodanov, E., & Santimone, M. Miyazawa, T., Ohtsu, S., Nakagawa, Y., & Funazukuri, T. (2006). Solvothermal
(1998). The mechanism of porcine pancreatic a-amylase. Inhibition of treatment of starch for the production of glucose and maltooligosaccharides.
maltopentaose hydrolysis by acarbose, maltose and maltotriose. European Journal of Materials Science, 41, 14891494.
Journal of Biochemistry, 252, 100107. Morell, M. K., Samuel, M. S., & OShea, M. G. (1998). Analysis of starch structure
Kerr, R. W., Cleveland, F. C., & Katzbeck, W. J. (1951). The action of amyloglucosidase using uorophore-assisted carbohydrate electrophoresis. Electrophoresis, 19,
on amylose and amylopectin. Journal of the American Chemical Society, 73, 26032611.
39163921. Morikawa, K., & Nishinari, K. (2000). Effects of concentration dependence of
Kim, J. O., Kim, W. S., & Shin, M. S. (1997). A comparative study of retrogradation of retrogradation behaviour of dispersions for native and chemically modied
rice starch gels by DSC, X-ray, and a-amylase methods. Starch/Staerke, 49, potato starch. Food Hydrocolloids, 14, 395401.
7175. Natarajan, S. K., & Sierks, M. R. (1997). Minimizing nonproductive substrate
Komolprasert, V., & Ofoli, R. Y. (1991). Starch hydrolysis kinetics of Bacillus binding: A new look at glucoamylase subsite afnities. Biochemistry, 36,
licheniformis a-amylase. Journal of Chemical Technology and Biotechnology, 51, 1494614955.
209223. Nebesny, E., Rosicka, J., & Tkaczyk, M. (2004). Inuence of conditions of maize starch
Krezowski, P. A., Nuttall, F. Q., Gannon, M. C., & Bartosh, N. H. (1986). The effect of enzymatic hydrolysis on physicochemical properties of glucose syrups. Starch/
protein ingestion on the metabolic response to oral glucose in normal Staerke, 56, 132137.
individuals. American Journal of Clinical Nutrition, 44, 847856. Nitta, Y., Kunikata, T., & Watanabe, T. (1979). Kinetic study of soybean b-amylase.
Kruger, J. E., & Marchylo, B. A. (1985). A comparison of the catalysis of starch The effect of pH. Journal of Biochemistry, 85, 4145.
components by isoenzymes from the two major groups of germinated wheat a- OShea, M. G., Samuel, M. S., Konik, C. M., & Morell, M. K. (1998). Fluorophore-
amylase. Cereal Chemistry, 62, 1118. assisted carbohydrate electrophoresis (FACE) of oligosaccharides: Efciency of
Kuchel, P. W. (1981). Nuclear magnetic-resonance of biological samples. Critical labeling and high-resolution separation. Carbohydrate Research, 307, 112.
Reviews in Analytical Chemistry, 12, 155231. Oates, C. G. (1997). Towards an understanding of starch granule structure and
Kuchel, P. W. (1985). Kinetic analysis of multienzyme systems in homogeneous hydrolysis. Trends in Food Science & Technology, 8, 375382.
solution. In G. R. Welch (Ed.), Organized multienzyme system (pp. 303380). Okuda, M., Aramaki, I., Koseki, T., Satoh, H., & Hashizume, K. (2005). Structural
Orlando, FL: Academic Press. characteristics, properties, and in vitro digestibility of rice. Cereal Chemistry, 82,
Kuchel, P. W. (1989). Biological applications of NMR. In L. D. Field & S. Sternhell 361368.
(Eds.), Analytical NMR (pp. 157219). London: Wiley. Otobe, K., Yoshii, Y., Sugiyama, J., & Kikuchi, Y. (1995). Studies on dynamic
Kuchel, P. W., & Ralston, G. B. (1997). Enzyme kinetics. In Schaums outline of theory viscoelasticity of rice. II. Varietal properties related to dynamic
and problems in biochemistry (2nd ed., pp. 251274). New York, NY: McGraw- viscoelasticity on gelatinized rice starch. Nippon Shokuhin Kagaku Kogaku
Hill. Kaishi, 42, 8592.
Lai, V. M. F., Lu, S., & Lii, C.-Y. (2000). Molecular characteristics inuencing Park, J. T., & Rollings, J. E. (1995). Effects of substrate branching characteristics on
retrogradation kinetics of rice amylopectins. Cereal Chemistry, 77, 272278. kinetics of enzymatic depolymerization of mixed linear and branched
Lehmann, U., & Robin, F. (2007). Slowly digestible starch its structure and health polysaccharides: II. Amylose/glycogen a-amylolysis. Biotechnology and
implications: A review. Trends in Food Science & Technology, 18, 346355. Bioengineering, 46, 3642.
Leloup, V. M., Colonna, P., & Ring, S. G. (1991). a-Amylase adsorption on starch Parker, R., & Ring, S. G. (2001). Aspects of the physical chemistry of starch. Journal of
crystallites. Biotechnology and Bioengineering, 38, 127134. Cereal Science, 34, 117.
Levine, H., & Slade, L. (1989). A food polymer science approach to the practice of Pastrana, L., Gonzalez, M., Miron, J., & Murado, M. A. (1998). A new device for
cryostabilization technology. Comments on Agricultural and Food Chemistry, 1, measuring diffusional restrictions and modeling substrate inhibition in a
315396. starchglucoamylase system. Biotechnology Letters, 20, 127130.
Lim, S.-T., Lee, J.-H., Shin, D.-H., & Lim, H. S. (1999). Comparison of protein extraction Peat, S., Whelan, W. J., & Thomas, G. J. (1952). Evidence of multiple branching in
solutions for rice starch isolation and effects of residual protein content on waxy maize starch. Journal of the Chemical Society, 45464548.
starch pasting properties. Starch/Staerke, 51, 120125. Pizzoferrato, L., Rotilio, G., & Paci, M. (1999). Modication of structure and
Lina, B. A. R., Jonker, D., & Kozianowski, G. (2002). Isomaltulose (Palatinose): A digestibility of chestnut starch upon cooking: A solid state 13C CP MAS NMR
review of biological and toxicological studies. Food and Chemical Toxicology, 40, and enzymatic degradation study. Journal of Agricultural and Food Chemistry, 47,
13751381. 40604063.
Lineback, D. R. (1984). The starch granule organization and properties. Bakers Planchot, V., Colonna, P., & Buleon, A. (1997). Enzymatic hydrolysis of [a]-glucan
Digest, 58(16), 1821. crystallites. Carbohydrate Research, 298, 319326.
Lineback, D. R. (1986). Current concepts of starch structure and the impact on Podzimek, S., Lebedat, P., & Johann, C. (2009). Asymmetric ow eld ow
properties. Denpun Kagaku, 33, 8088. fractionation: A powerful method for polymer characterization. LCGC North
Livesey, G., Taylor, R., Hulshof, T., & Howlett, J. (2008). Glycemic response and America, 27, 6266.
health a systematic review and meta-analysis: Relations between dietary Ratnayake, W. S., & Jackson, D. S. (2009). Starch gelatinization. Advances in Food and
glycemic properties and health outcomes. The American Journal of Clinical Nutrition Research, 55, 221268.
Nutrition, 87, 258S268S. Ring, S. G., Colonna, P., IAnson, K. J., Kalichevsky, M. T., Miles, M. J., Morris, V. J., et al.
Lopez-Rubio, A., Flanagan Bernadine, M., Shrestha Ashok, K., Gidley Michael, J., & (1987). The gelation and crystallisation of amylopectin. Carbohydrate Research,
Gilbert Elliot, P. (2008). Molecular rearrangement of starch during in vitro 162, 277293.
digestion: Toward a better understanding of enzyme resistant starch formation Roberts, S. B. (2000). High-glycemic index foods, hunger, and obesity: Is there a
in processed starches. Biomacromolecules, 9, 19511958. connection? Nutrition Reviews, 58, 163169.
Lopez, C., Torrado, A., Fucinos, P., Guerra, N. P., & Pastrana, L. (2006). Enzymatic Robertson, G. H., Wong, D. W. S., Lee, C. C., Wagschal, K., Smith, M. R., & Orts, W. J.
inhibition and thermal inactivation in the hydrolysis of chestnut puree with an (2006). Native or raw starch digestion: A key step in energy efcient biorening
amylases mixture. Enzyme and Microbial Technology, 39, 252258. of grain. Journal of Agricultural and Food Chemistry, 54, 353365.
Roger, P., Baud, B., & Colonna, P. (2001). Characterization of starch polysaccharides Tester, R. F., Qi, X., & Karkalas, J. (2006). Hydrolysis of native starches with amylases.
by ow eld-ow fractionation-multi-angle laser light scattering-differential Animal Feed Science and Technology, 130, 3954.
refractometer index. Journal of Chromatography A, 917, 179185. Thompson, D. B. (2000). On the non-random nature of amylopectin branching.
Rolland-Sabate, A., Colonna, P., Mendez-Montealvo, M. G., & Planchot, V. (2007). Carbohydrate Polymers, 43, 223239.
Branching features of amylopectins and glycogen determined by asymmetrical Thorne, M. J., Thompson, L. U., & Jenkins, D. J. A. (1983). Factors affecting starch
ow eld ow fractionation coupled with multiangle laser light scattering. digestibility and the glycemic response with special reference to legumes.
Biomacromolecules, 8, 25202532. American Journal of Clinical Nutrition, 38, 481488.
Ross, S. W., Brand, J. C., Thorburn, A. W., & Truswell, A. S. (1987). Glycemic index of Thornley, S., McRobbie, H., Eyles, H., Walker, N., & Simmons, G. (2008). The obesity
processed wheat products. American Journal of Clinical Nutrition, 46, 631635. epidemic: Is glycemic index the key to unlocking a hidden addiction? Medical
Sasaki, T., & Matsuki, J. (1998). Effect of wheat starch structure on swelling power. Hypotheses, 71, 709714.
Cereal Chemistry, 75, 525529. Tjalling, J. Y. (1995). Historical development of the NewtonRaphson method.
Sayago-Ayerdi, S. G., Tovar, J., Osorio-Diaz, P., Paredes-Lopez, O., & Bello-Perez, L. A. Society for Industrial and Applied Mathematics, 37, 531551.
(2005). In vitro starch digestibility and predicted glycemic index of corn tortilla, Topping, D. L., Bajka, B. H., Bird, A. R., Clarke, J. M., Cobiac, L., Conlon, M. A., et al.
black beans, and tortilla-bean mixture: Effect of cold storage. Journal of (2008). Resistant starches as a vehicle for delivering health benets to the
Agricultural and Food Chemistry, 53, 12811285. human large bowel. Microbial Ecology in Health and Disease, 20, 103108.
Scheid, F. (1988). Least squares polynomial approximation. In Schaums outline of Tsai, M.-L., & Lii, C.-Y. (2000). Effect of hot-water-soluble components on the
theory and problems of numerical analysis (2nd ed., pp. 241274). New York, NY: rheological properties of rice starch. Starch/Staerke, 52, 4453.
McGraw-Hill. Tsuge, H., Tatsumi, E., Ohtani, N., & Nakazima, A. (1992). Screening of a-amylase
Schinzel, R., & Nidetzky, B. (1999). Bacterial a-glucan phosphorylases. FEMS suitable for evaluating the degree of starch retrogradation. Starch/Staerke, 44,
Microbiology Letters, 171, 7379. 2932.
Schmitz, S., Dona, A. C., Castignolles, P., Gilbert, R. G., & Gaborieau, M. (2009). Urooj, A., & Puttraj, S. (1999). Digestibility index and factors affecting rate of starch
Quantication of the extent of starch dissolution in dimethyl sulfoxide by 1H digestion in vitro in conventional food preparation. Starch/Staerke, 51, 430435.
NMR spectroscopy. Macromolecular Bioscience, 9, 506514. Vesterinen, E., Myllrinen, P., Forssell, P., Sderling, E., & Autio, K. (2002). Structural
Schnell, S., & Hanson, S. M. (2007). A test for measuring the effects of enzyme properties in relation to oral enzymatic digestibility of starch gels based on pure
inactivation. Biophysical Chemistry, 125, 269274. starch components and high amylose content. Food Hydrocolloids, 16, 161167.
Selwyn, M. J. (1965). Simple test for inactivation of an enzyme during assay. Waigh, T. A., Hopkinson, I., Donald, A. M., Butler, M. F., Heidelbach, F., & Riekel, C.
Biochimica et Biophysica Acta, 105, 193195. (1997). Analysis of the native structure of starch granules with X-ray
Sevenou, O., Hill, S. E., Farhat, I. A., & Mitchell, J. R. (2002). Organization of the microfocus diffraction. Macromolecules, 30, 38133820.
external region of the starch granule as determined by infrared spectroscopy. Waigh, T. A., Donald, A. M., Heidelbach, F., Riekel, C., & Gidley, M. J. (1999). Analysis
International Journal of Biological Macromolecules, 31, 7985. of the native structure of starch granules with small angle X-ray microfocus
Shibaev, V. (1994). Some new physicochemical aspects of side chain liquid crystal scattering. Biopolymers, 49, 91105.
polymers. Molecular Crystals and Liquid Crystals Science and Technology Section A Waigh, T. A., Kato, K. L., Donald, A. M., Gidley, M. J., Clarke, C. J., & Riekel, C. (2000).
Molecular Crystals and Liquid Crystals, 243, 201230. Side-chain liquid-crystalline model for starch. Starch/Staerke, 52, 450460.
Shiotsubo, T. (1983). Study of starch gelatinization. Part I. Starch gelatinization at Walker, G. S., Atherton, J., Bauman, J., Kohl, C., Lam, W., Reily, M., et al. (2007).
different temperatures as measured by an enzymic digestion method. Determination of degradation pathways and kinetics of acyl glucuronides by
Agricultural and Biological Chemistry, 47, 24212425. NMR spectroscopy. Chemical Research in Toxicology, 20, 876886.
Shu, X., Jiao, G., Fitzgerald, M. A., Yang, C., Shu, Q., & Wu, D. (2006). Starch structure Wang, J.-P., Zeng, A.-W., Liu, Z., & Yuan, X.-G. (2006). Kinetics of glucoamylase
and digestibility of rice high in resistant starch. Starch/Staerke, 58, 411417. hydrolysis of corn starch. Journal of Chemical Technology and Biotechnology, 81,
Silverio, J., Fredriksson, H., Andersson, R., Eliasson, A. C., & Aman, P. (2000). The 727729.
effect of temperature cycling on the amylopectin retrogradation of starches Ward, K. M., Aletras, A. H., & Balaban, R. S. (2000). A new class of contrast agents for
with different amylopectin unit-chain length distribution. Carbohydrate MRI based on proton chemical exchange dependent saturation transfer (CEST).
Polymers, 42, 175184. Journal of Magnetic Resonance, 143, 7987.
Slade, L., & Levine, H. (1988). Non-equilibrium melting of native granular starch: Ward, R. M., Gao, Q., De Bruyn, H., Gilbert, R. G., & Fitzgerald, M. A. (2006). Improved
Part I. Temperature location of the glass transition associated with methods for the structural analysis of the amylose-rich fraction from rice our.
gelatinization of A-type cereal starches. Carbohydrate Polymers, 8, 183208. Biomacromolecules, 7, 866876.
Slaughter, S. L., Ellis, P. R., & Butterworth, P. J. (2001). An investigation of the action Werner, R., & Keilich, G. (1965). An enzymatic method on kinetic studies of
of porcine pancreatic a-amylase on native and gelatinised starches. Biochimica endohydrolases: Activity determination of a-amylase. Biochemische Zeitschrift,
et Biophysica Acta General Subjects, 1525, 2936. 342, 226235.
Smith, A. M. (2007). Starch biosynthesis and degradation. Handbook of Plant Science, Wolever, T. M. S., Mehling, C., Chiasson, J. L., Josse, R. G., Leiter, L. A., Maheux, P.,
2, 11591165. et al. (2008). Low glycaemic index diet and disposition index in type 2 diabetes
Sonnad, J. R., & Goudar, C. T. (2009). Solution of the MichaelisMenten equation (the Canadian trial of Carbohydrates in Diabetes): A randomised controlled trial.
using the decomposition method. Mathematical Biosciences and Engineering, 6, Diabetologia, 51, 16071615.
173188. Wong, J. M. W., de Souza, R., Kendall, C. W. C., Emam, A., & Jenkins, D. J. A. (2006).
Steeneken, P. A. M. (1989). Rheological properties of aqueous suspensions of Colonic health: Fermentation and short chain fatty acids. Journal of Clinical
swollen starch granules. Carbohydrate Polymers, 11, 2342. Gastroenterology, 40, 235243.
Sujka, M., & Jamroz, J. (2007). Starch granule porosity and its changes by means of Yoon, S.-H., & Robyt, J. F. (2003). Study of the inhibition of four alpha amylases by
amylolysis. International Agrophysics, 21, 107113. acarbose and its 4IV-a-maltohexaosyl and 4IV-a-maltododecaosyl analogues.
Tako, M., & Hizukuri, S. (2000). Retrogradation mechanism of rice starch. Cereal Carbohydrate Research, 338, 19691980.
Chemistry, 77, 473477. Zhang, G., Ao, Z., & Hamaker, B. R. (2006). Slow digestion property of native cereal
Tester, R. F., & Morrison, W. R. (1990). Swelling and gelatinization of cereal starches. Biomacromolecules, 7, 32523258.
starches. I. Effects of amylopectin, amylose, and lipids. Cereal Chemistry, 67, Zhou, J., & van Zijl, P. C. M. (2006). Chemical exchange saturation transfer imaging
551557. and spectroscopy. Progress in Nuclear Magnetic Resonance Spectroscopy, 48,
Tester, R. F., Morrison, W. R., Gidley, M. J., Kirkland, M., & Karkalas, J. (1994). 109136.
Properties of damaged starch granules. III. Microscopy and particle size analysis Ziegler, G. R., Creek, J. A., & Runt, J. (2005). Spherulitic crystallization in starch as a
of undamaged granules and remnants. Journal of Cereal Science, 20, 5967. model for starch granule initiation. Biomacromolecules, 6, 15471554.
EKUILIBRIU M ISSN : 1412-9124
Vol. 11. No. 2. Halaman : 73 77 Juli 2012
PENGARUH JENIS DAN KONSENTRASI ASAM TERHADAP KINETIKA REAKSI

HIDROLISIS PELEPAH PISANG (Musa Paradisiaca L)
Enny Kriswiyanti Artati*, Feliciana Irvina W. H., Fatimah

Jurusan Teknik Kimia Fakultas Teknik Universitas Sebelas Maret
Jl. Ir. Sutami No. 36 A, Surakarta 57126 Telp/fax: 0271-632112
*Email: enny91@yahoo.com
Abstract: Banana stem is a biomass that contain high cellulose that can be converted into
glucose through hydrolysis reaction. This research purposes were to understand the influence of
any kind of acid and its concentration for banana leaves cellulose hydrolysis and to gain the
kinetic data or the reaction rate constant (k) of banana leaves hydrolysis reaction. The constant
variables in this research were mass of banana leaves (25 gram), mass ratio of solid to solvent
o
(1:10) and cooking temperature (100 C). The observated variables were kind of acid (H2SO4
and HCl) in each concentration of 1.0 N; 1.5 N; 2.0 N; 2.5 N; 3.0 N. The glucose samples were
then analyzed using refractometer. Datas showed that at higher acid concentration and longer
reaction time, the higher glucose concentration was formed. Datas also showed that HCl will
deliver higher glucose concentration than H2SO4. The maximum operation condition was gained
at 2 N HCl concentration. By assuming the order of hydrolysis reaction was one, the reaction
rate constant were in the range of 0.000182 0.0043/minute and 0.00209 0.0066/minute for
H2SO4 and HCl respectively.
Keywords: cellulose hydrolysis; banana stem; sulfite acid; chloride acid
PENDAHULUAN hidrolisis selulosa pelepah pisang serta

Indonesia merupakan penghasil biomassa memperoleh data kinetika atau tetapan
yang jumlahnya cukup besar. Biomassa dapat kecepatan reaksi (k) hidrolisis pelepah pisang.
berasal dari rumah tangga, pertanian, dan Tanaman pisang merupakan tanaman
industri. Biomassa merupakan sumber energi asli Indonesia yang banyak diusahakan di
yang menarik untuk dikembangkan karena daerah tropis. Pisang merupakan tanaman
jumlahnya yang melimpah dan sifatnya yang yang berbuah hanya sekali kemudian mati.
dapat diperbarui. Tingginya antara 2-9 m, berakar serabut dengan
Beberapa tahun belakangan ini telah batang di bawah tanah (bonggol) yang pendek.
dilakukan pengembangan biomassa selulosik Dari mata tunas yang ada pada bonggol tumbuh
(selulosa dan hemiselulosa) seperti limbah tanaman baru (Admin, 2006).
pertanian dan pengolahan hutan, kertas Pelepah pisang diperoleh dari batang
bekas, dan limbah industri, sebagai sumber pisang palsu (pseudo-stem) yang mengandung
glukosa untuk difermentasi menjadi etanol. selulosa sebesar 40,1% dan lignin 17,8%
Sumber selulosa yang dapat digunakan (Kamara et al., 2006).
antara lain, adalah sisa-sisa produk pertanian Selulosa merupakan penyusun utama
dan hasil hutan, kertas bekas, dan limbah kayu berupa polimer alami yang panjang dan
industri (White, 2000). linier terdiri dari residu -D-glukosa yang
Sisa produk pertanian yang banyak dihubungkan oleh ikatan glikosida pada posisi
dihasilkan salah satunya adalah pelepah pisang. C1 dan C4. Selulosa mempunyai sifat antara lain
Seperti halnya biomassa pada umumnya, berwarna putih, berserat, tidak larut dalam air
pelepah pisang mengandung selulosa sebesar dan pelarut organik serta mempunyai kuat tarik
40,1% dan lignin 17,8% (Kamara et. al., 2006) . yang tinggi.
Selulosa dapat digunakan untuk membuat Berdasarkan sifat kelarutan selulosa
bioetanol dengan cara hidrolisis dan fermentasi. dalam alkali, selulosa dibagi menjadi tiga macam
Hidrolisis pada selulosa ada beberapa cara yaitu -Selulosa (merupakan selulosa yang tidak
antara lain dillute acid hydrolisis, concentrated larut dalam larutan NaOH 17-18%), -Selulosa
acid hydrolisis, dan enzymatic hydrolisis. (merupakan jenis selulosa yang larut dalam
Penelitian ini bertujuan untuk mengetahui larutan NaOH 17-18%) dan -Selulosa.
pengaruh konsentrasi dan jenis asam untuk
73
Hidrolisis adalah suatu proses antara yang bereaksi dengan larutan sehingga
reaktan dengan air agar suatu senyawa pecah dihasilkan pula hasil yang semakin banyak
atau terurai. Reaksi ini merupakan reaksi orde (Supranto, 1998). Suhu berpengaruh terhadap
satu, karena air yang digunakan berlebih, konstanta kecepatan reaksi. Jika suhu tinggi,
sehingga perubahan reaktan dapat diabaikan. konstanta kecepatan reaksi akan semakin besar
Terdapat beberapa jenis proses hidrolisis antara sehingga reaksi dat semakin cepat (Kirk Othmer,
lain: hidrolisis murni (sebagai reaktan hanya air), 1983). Waktu reaksi yang semakin lama akan
hidrolisis dengan larutan asam (bisa berupa memperbanyak jumlah tumbukan zat-zat
asam encer atau pekat), hidrolisis dengan basa pereaksi sehingga molekul-molekul yang
(bisa berupa basa encer atau pekat) dan bereaksi semakin banyak dan memperbanyak
hidrolisis dengan menggunakan enzim. hasil yang terbentuk (Supranto, 1998).
Hidrolisis adalah suatu proses antara Pengadukan berkaitan dengan faktor frekuensi
reaktan dengan air agar suatu senyawa pecah tumbukan (A) pada persamaan Arhenius
terurai. Reaksi hidrolisis: sehingga dengan adanya pengadukan maka
kecepatan reaksi akan meningkat.
hidrolisis
(C6H10O5)n + n H2O nC6H12O6 Salah satu cara untuk menganalisis kadar
Selulosa Air Glukosa (1) glukosa yaitu menggunakan refraktometer.
Satuan skala pembacaan refraktometer yaitu
o o
Pada reaksi hidrolisis polisakarida dengan brix. Brix adalah satuan skala yang digunakan
air, air akan menyerang selulosa pada ikatan 1- untuk pengukuran kandungan padatan terlarut
4 glukosida menghasilkan dextrin, sirup atau (Purwono, 2002). Skala brix dari refraktometer
glukosa tergantung pada derajat pemecahan sama dengan berat gram glukosa dari 100 gram
rantai polisakarida dalam selulosa. Reaksinya larutan glukosa.
merupakan reaksi order satu jika digunakan air Reaksi hidrolisa yang terjadi pada pelepah
yang berlebih, sehingga perubahan reaktan pisang (Matz, 1970)adalah :
dapat diabaikan.
Tetapi reaksi antara air dan selulosa ini (C6H10O5)n + n H2O n(C6H12O6)
berlangsung sangat lambat sehingga diperlukan Selulosa Glukosa (2)
bantuan katalisator untuk memperbesar
kereaktifan air. Katalisator ini bisa berupa asam Dari persamaan reaksi hidrolisa di atas
maupun enzim. Katalisator asam yang biasa bila dianggap sebagai reaksi elementer dan
digunakan adalah asam klorida, asam nitrat dan reaksi samping diabaikan, maka persamaan
asam sulfat. kecepatan reaksi adalah :
n
Glukosa yang dihasilkan selama proses -rA= k.CA.CB (3)
hidrolisis difermentasi menjadi etanol. Hasil Karena konsentrasi B sangat besar, maka
hidrolisis enzim lebih dapat dikendalikan, konsentrasi B dapat dianggap bernilai konstan
sehingga dapat diatur kadar maltosa dan untuk setiap nilai n. Maka persamaan (3)
glukosanya. menjadi :
Proses hidrolisis dengan menggunakan - rA = k .CA
asam dipengaruhi oleh ukuran bahan, kecepatan
n
pengadukan, konsentrasi asam, rasio bahan, dengan k = k. CB
suhu, dan waktu. Semakin halus ukuran bahan
dC A
permukaan bidang kontak akan semakin luas - = k.CA (4)
sehingga kecepatan reksi akan bertambah cepat dt
dan akan memperbesar konversi reaksi karena CA = CAo (1-x) maka persamaan (4)
(Supranto, 1998). Laju proses hidrolisis akan menjadi:
bertambah oleh konsentrasi asam yang tinggi. d(1 x)
Meskipun konsentrasi asam yang tinggi dapat - =k.dt (5)
menambah laju hidrolisis, konsentrasi asam (1 - x)
yang tinggi juga akan mengakibatkan terikatnya Jika diintegralkan dengan batasan t=0, x=0 dan
material pengotor seperti SiO2, phospat, dan t=t, x=x, maka persamaan (5) menjadi:
garam Ca, Mg, K, Na, maka perlu perbandingan -ln (1-x) = k.t + C (6)
yang sesuai antara bahan yang akan dihidrolisis
dengan konsentrasi asam yang ditambahkan Persamaan (6) menunjukkan hubungan
(Kerr,1950). Rasio bahan yang semakin besar antara konversi reaksi dengan waktu. Dengan x
maka konsentrasi glukosa hasil hidrolisis adalah konversi reaksi yang menyatakan
semakin banyak pula. Karena dengan semakin perbandingan jumlah selulosa yang bereaksi
besar rasio bahan semakin besar pula bahan
74 E K U I L I B R I U M Vol. 11. No. 2. Januari 2012 : 73 77

dengan jumlah selulosa mula-mula, dan C glukosa yang terbentuk sedikit, karena larutan
adalah suatu konstanta. menjadi terlalu kental dan tumbukan antar
reaktan menjadi berkurang (Lamiya dan Mareta,
METODE PENELITIAN 2010).
Bahan yang digunakan dalam penelitian
ini adalah pelepah pisang dengan kadar air 5,8%
dan kadar selulosa 38,54%, larutan H2SO4,
larutan HCl, aquadest. Alat utama yang
digunakan adalah magnetik stirer, pendingin
bola, dan erlenmeyer.
Analisis Kadar Alpha Selulosa. Pulp
basah sebanyak 5 gram ditambah dengan 50
mL larutan NaOH 17,5% dan dibiarkan beberapa
menit. Kemudian melembutkan pulp dan
menambahkan 50 mL larutan NaOH dan
dibiarkan selama 30 menit. Menambahkan 300
mL air dan disaring dengan kain. Selanjutnya
dicuci dengan 1 L air dan 100 mL larutan asam
Gambar 1. Grafik Hubungan antara Waktu (menit)
asetat encer (1:4). Kemudian mencuci dengan 1 dengan Berat Glukosa Hasil Hidrolisa
L air panas, memeras air yang masih pada berbagai Konsentrasi H2SO4
terkandung dalam pulp dan mengeringkannya.
Proses Hidrolisis. Mula-mula pelepah
pisang yang sudah dikeringkan, dihancurkan
dengan menggunakan blender sampai menjadi
serbuk dan diayak menggunakan ayakan 50
mesh. Sebanyak 25 gram bahan dimasukkan ke
dalam erlenmeyer. Untuk memperoleh glukosa
optimum dilakukan penambahan larutan H2SO4
dengan konsentrasi 1 N sebanyak 250 mL.
Kemudian menghidupkan magnetik stirer dan
memanaskan larutan sampai suhu 100 oC.
Hidrolisis dilakukan sampai 120 menit.
Pengambilan sampel dimulai saat larutan
o
mencapai suhu 100 C dan dianggap sebagai
waktu ke-0. Sampel yang diambil langsung
didinginkan dengan air es untuk menghentikan
reaksi. Larutan sampel hasil hidrolisis dianalisis Gambar 2. Grafik Hubungan antara Waktu (menit)
menggunakan refraktometer. dengan Berat Glukosa Hasil Hidrolisa
pada berbagai Konsentrasi HCl
Langkah penelitian di atas dilakukan untuk
variasi konsentrasi asam 1 N, 1,5 N, 2 N, 2,5 N,
Dari Gambar 1, berat glukosa hasil
3 N serta jenis asam H2SO4 dan HCl.
hidrolisis terbesar diperoleh pada proses
hidrolisa dengan waktu 2 jam pada konsentrasi
HASIL DAN PEMBAHASAN
asam sulfat 2 N yaitu sebesar 8,2 gram. Dari
Data hasil pengambilan sampel pada
Gambar 2, berat glukosa hasil hidrolisis terbesar
berbagai variasi konsentrasi asam dan waktu
diperoleh pada proses hidrolisa dengan waktu 2
pengambilan sampel disajikan pada Gambar 1
jam pada konsentrasi asam klorida 2 N yaitu
dan 2. Berdasarkan Tabel Gambar 1 dan
sebesar 9 gram.
Gambar 2 dapat dilihat bahwa semakin lama
Berdasarkan gambar, pada nilai
waktu reaksi berat glukosa yang dihasilkan akan
konsentrasi asam yang sama, untuk katalis
semakin besar karena waktu kontak antara
asam sulfat menghasilkan berat glukosa yang
reaktan untuk bereaksi semakin besar.
lebih kecil. Dalam hal ini yang mempengaruhi
Pada reaksi hidrolisis, asam berfungsi +
kecepatan reaksi adalah jumlah H . Pada H2SO4
sebagai katalis yang bertujuan untuk + +
jumlah H lebih banyak daripada HCl. H dari
mempercepat jalannya reaksi. Asam akan -
asam akan berikatan dengan OH dari air
mempengaruhi penurunan energi aktivasi, -
membentuk gula reduksi. Untuk OH yang sama
sehingga reaksi berjalan dengan cepat dan berat +
pada H2SO4 terdapat sisa H yang tidak bereaksi
glukosa yang dihasilkan akan semakin besar.
sehingga menyebabkan glukosa yang dihasilkan
Akan tetapi jika konsentrasi asam terlalu besar,
Pengaruh Jenis dan Konsentrasi Asam Terhadap Kinetika Reaksi Hidrolisis 75

Pelepah Pisang (Musa Paradisiaca L) (Enny K. Artati, Feliciana Irvina W. H., Fatimah)
pada H2SO4 lebih sedikit (Hikmiyati dan Yanie, Tabel 5. Data Konversi (XA) dengan Waktu (menit)
2007). pada berbagai Konsentrasi Asam Klorida
Pengaruh jenis asam berupa H2SO4 dan Waktu Konversi (XA)
(menit) 1N 1,5 N 2N 2,5 N 3N
HCl dengan konsentrasi 1 N; 1,5 N; 2 N; 2,5 N; 3
N terhadap nilai k proses hidrolisis pelepah 0 0,25 0,56 0,72 0,57 0,56
pisang dipengaruhi oleh besarnya konversi tiap 30 0,33 0,57 0,82 0,65 0,61
60 0,36 0,61 0,85 0,72 0,65
variabel.
90 0,40 0,65 0,89 0,76 0,72
Tabel 3. Data Konversi (XA) dengan Waktu (menit) 120 0,42 0,69 0,93 0,80 0,78
pada berbagai Konsentrasi H2SO4
Konversi (XA) Tabel 6. Data ln ( 1-X A ) dengan Waktu (menit)
Waktu
pada berbagai Konsentrasi HCl
(menit) 1N 1,5 N 2N 2,5 N 3N
Waktu -ln(1-x)
0 0,23 0,41 0,60 0,54 0,52 (menit) 1N 1,5 N 2N 2,5 N 3N
30 0,29 0,47 0,70 0,60 0,57
0 0,29 0,62 1,01 0,86 0,82
60 0,33 0,50 0,73 0,65 0,63
30 0,39 0,66 1,30 1,06 0,95
90 0,36 0,52 0,75 0,69 0,67
60 0,45 0,74 1,45 1,30 1,06
120 0,38 0,58 0,77 0,71 0,69
90 0,51 0,82 1,62 1,45 1,30
120 0,54 0,91 1,83 1,62 1,53
Tabel 4. Data ln ( 1-XA ) dengan Waktu (menit)
pada berbagai Konsentrasi H2SO4
-ln(1-XA) Tabel 7. Data Konstanta Kecepatan Reaksi pada
Waktu
berbagai Konsentrasi dan Jenis Asam
(menit) 1N 1,5 N 2N 2,5 N 3N
Konsentrasi Konstanta Kecepatan Reaksi (k)
0 0,27 0,53 0,91 0,78 0,74 (N) H2SO4 HCl
30 0,34 0,63 1,17 0,91 0,84
1 0,00182 0,00209
60 0,40 0,70 1,30 1,06 0,98
1,5 0,00261 0,00240
90 0,45 0,74 1,37 1,17 1,12
2 0,0043 0,00660
120 0,48 0,86 1,45 1,24 1,17
2,5 0,00393 0,00638
3 0,00381 0,00591
Gambar 3. Grafik Hubungan antara Waktu ( menit) dengan ln(1-XA) pada berbagai Konsentrasi H2SO4
76 E K U I L I B R I U M Vol. 11. No. 2. Januari 2012 : 73 77

Gambar 4. Grafik Hubungan antara Waktu ( menit) dengan ln(1-XA) pada berbagai Konsentrasi HCl
Dari Tabel 7 terlihat jika semakin tinggi Kirk, R.E., and Othmer, D.F.,1983,
konsentrasi asam yang digunakan sebagai Enchyclopedia of Chemical Technology,
katalis, maka nilak k yang didapat semakin tinggi 3rd ed., John Wiley and Sons Inc., New
dan sampai kondisi tertentu nilai k akan York
-E/RT
menurun. Dari persamaan Arhenius k=A.e , Zahro, L. M. dan Istiorini, M., 2010, Penyiapan
energi aktivasi yang kecil mengakibatkan Bahan Baku dalam Proses Fermentasi
konstanta kecepatan reaksi menjadi besar, Fase Cair Asam Sitrat Melalui Proses
sehingga reaksi akan berjalan lebih cepat. Hidrolisa Ampas Singkong, Jurusan
Begitu pula sebaliknya, jika energi aktivasi Teknik Kimia, Fakultas Teknik, Universitas
meningkat, maka konstanta kecepatan reaksi Diponegoro
menurun dan reaksi akan berjalan semakin Levenspiel, O., 1972, Chemical Reaction
lambat. Pada kondisi tertentu konsentrasi asam Engineering, 2nd ed., John Wiley and
yang terlalu besar akan mengakibatkan energi Sons Inc., New York
aktivasi meningkat yang disebabkan karena Matz, S. A., 1970, Sereal Technology, The Avi
tumbukan antar reaktan berkurang. Publishing Co. Inc., West Port,
Connecticut
KESIMPULAN Hikmiyati, N dan Yanie, N. S., 2009, Pembuatan
Berdasarkan hasil penelitian dapat Bioetanol dari Limbah Kulit Singkong
disimpulkan bahwa semakin tinggi konsentrasi Melalui Proses Hidrolisa Asam dan
asam yang digunakan maka kadar glukosa akan Enzimatis, Jurusan Teknik Kimia,
semakin meningkat sampai konsentrasi asam Fakultas Teknik, Universitas Diponegoro
optimum. Konsentrasi asam yang optimum Nyimas Chairunisa, 2010, Tinjauan Hidrolisis
adalah 2 N. Semakin kuat jenis asam yang Pati Bonggol Pisang untuk Bahan Baku
digunakan maka kadar glukosa akan semakin Pembuatan Bioetanol, Jurusan Teknik
tinggi. Jenis asam yang menghasilkan kadar Kimia, Fakultas Teknik, Politeknik Negeri
glukosa tinggi adalah asam klorida. Sriwijaya
Tetapan kecepatan reaksi pada variabel Sukmawati, R. F. dan Milati, S., 2009,
konsentrasi asam sulfat sebesar 0,0043 /menit Pembuatan Bioetanol dari Kulit
dan tetapan kecepatan reaksi pada variabel Singkong, Program Studi D III Teknik
konsentrasi asam klorida sebesar 0,0066 /menit Kimia, Fakultas Teknik, Universitas
Sebelas Maret Surakarta
DAFTAR PUSTAKA Supranto, 1998, Proses Industri Kimia II,
Groggins, P.H., 1958, Unit Processes in Teknik Kimia FT UGM, Yogyakarta
Organic Synthesis, 5th ed., Mc. Graw Hill www.bps.go.id
Book Company Inc., New York www.iptek.net.id
Kerr, R.W., Chemistry and Industry of Starch,
2nd ed., Academic Press Inc., New York
Pengaruh Jenis dan Konsentrasi Asam Terhadap Kinetika Reaksi Hidrolisis 77

Pelepah Pisang (Musa Paradisiaca L) (Enny K. Artati, Feliciana Irvina W. H., Fatimah)
LEMBAR ASISTENSI
DIPERIKSA KETERANGAN TANDA

NO TANGGAL TANGAN
1 19 Mei 2016 - Atur Border Grafik
- Cek Abstrak
- Cek Format Penulisan
2 20 Mei 2016 ACC

Lapres 5 Senin

Diunggah oleh

Informasi Dokumen

Hak Cipta

Format Tersedia

Bagikan dokumen Ini

Bagikan atau Tanam Dokumen

Opsi Berbagi

Apakah menurut Anda dokumen ini bermanfaat?

Apakah konten ini tidak pantas?

Hak Cipta:

Format Tersedia

Lapres 5 Senin

Diunggah oleh

Hak Cipta:

Format Tersedia

LAPORAN RESMI

PRAKTIKUM PROSES KIMIA

Gema Adil Guspiani 21030114120040

Laboratorium Proses Kimia

Jurusan Teknik Kimia Fakultas Teknik

PRAKTIKUM PROSES KIMIA

Gema Adil Guspiani 21030114120040

Laboratorium Proses Kimia

Jurusan Teknik Kimia Fakultas Teknik

Semarang , 20 Mei 2016

Dosen Pembimbing Asisten Pembimbing

Dr. Siswo Sumardiono, ST, MT Ilham Dwiyanto Emzar

LABORATORIUM PROSES KIMIA 2016

LABORATORIUM PROSES KIMIA 2016

Starch is a carbohydrate with the major components amylose and amylopectin.

LABORATORIUM PROSES KIMIA 2016

Semarang, Mei 2016

LABORATORIUM PROSES KIMIA 2016

LABORATORIUM PROSES KIMIA 2016

3.3.1 Bahan ........................................................................................... 9

BAB V PENUTUP ................................................................................................ 16

LABORATORIUM PROSES KIMIA 2016

Gambar 3.1 Skema Rancangan Percobaan............................................................ 8

LABORATORIUM PROSES KIMIA 2016

LABORATORIUM PROSES KIMIA 2016

1.1 Latar Belakang

1.2 Rumusan Masalah

LABORATORIUM PROSES KIMIA 2016

1.3 Tujuan Percobaan

1.4 Manfaat Percobaan

LABORATORIUM PROSES KIMIA 2016

2.1 Pengertian Pati

2.2 Hidrolisa Pati

LABORATORIUM PROSES KIMIA 2016

Dimana xA = konversi reaksi setelah satu detik. Persamaan 7 dapat

2.3 Modifikasi Pati

LABORATORIUM PROSES KIMIA 2016

terbatas. Modifikasi adalah pati yang gugus hidroksinya telah mengalami

2.4 Variabel yang Berpengaruh

LABORATORIUM PROSES KIMIA 2016

keseluruhan, hidrolisa pati dengan katalis asam merupakan proses yang

LABORATORIUM PROSES KIMIA 2016

yang tinggi sehingga molekul-molekul zat pereaksi akan sulit bergerak.

LABORATORIUM PROSES KIMIA 2016

3.1 Rancangan Praktikum

Gambar 3.1 Skema Rancangan Praktikum

3.2. Gambar Alat Utama

LABORATORIUM PROSES KIMIA 2016

Gambar 3.2 Rangkaian alat hidrolisis

3.3 Bahan dan Alat yang digunakan

3.4 Prosedur percobaan

LABORATORIUM PROSES KIMIA 2016

LABORATORIUM PROSES KIMIA 2016

sampai 100 ml, diambil 5 ml. Kedalam Erlenmeyer dimasukkan, kemudian

Rumus penentuan kadar pati awal =

LABORATORIUM PROSES KIMIA 2016

LABORATORIUM PROSES KIMIA 2016

4.1 Pengaruh Jenis Katalis Terhadap Konversi (XA) Hidrolisa Pati

Tabel 4.1 Hidrolisa Pati katalis HCl 0,1 N

Tabel 4.2 Hidrolisa Pati katalis H2SO4 0,1 N

LABORATORIUM PROSES KIMIA 2016

Gambar 4.1 Pengaruh Jenis Katalis Terhadap Konversi Hidrolisa Pati

4.2 Pengaruh Jenis Katalis Terhadap Konstanta Kecepatan Reaksi (k)

LABORATORIUM PROSES KIMIA 2016