CHEMICAL ENGINEERING RESEARCH AND DESIGN

99 ( 2 0 I 5 ) 228-235

Contents lists available at ScienceDirect

Chemical Engineering Research and Design

IChemE

journal homepage: w w w.elsevie r.com /locate /cherd

Influences of the experimental setup configuration

on mass transfer measurements in absorption
systems

<D

CrossMark

Verena Wolfa*, Markus Lehnera, Karin Hoffmannb

3 Montanuniuersitaet Leoben, Leoben, Austria
b RVT Process Equipment GmbH, Steinuiiesen, Germany

ARTICLE

INFO

Article history:
Received 20 October 2014
Received in revised form 12 March
2015
Accepted 3 April 2015
Available online 13 April 2015
_______ _______________________
Keywords:
Absorption
Mass transfer m easurement
Standardisation
Packing
Experimental setup

1.

ABSTRACT
The standardisation of mass transfer m easurements for absorption systems is a key factor
for the deduction of accurate mass transfer models for random and structured packing.
Several papers (Hoffmann et al. (2007). Trans IChemE, A, 85 (Al),40;Kunzeetal. (2012). Chem.
Ingen. Techn., 84, 1931; Rejl et al. (2009). Chem. Eng. Res. Des., 87, 695) already deal with
this problem and recommendations are given for appropriate test systems, the execution
of the experiments and particularly also for the experimental setup to be used. However,
systematic investigations of the influences of the experimental setup configuration on the
results of mass transfer measurements in absorption systems are published rarely.
Mass transfer measurements with the system ammonia-air/water have been performed
at a pilot plant consisting of a DN 600 saturation column and a DN 450 measuring column, both made from polypropylene, and equipped with different random and structured
packing. Although the used experimental setup of the pilot plant follows strictly the recommendations published in Hoffmann et al., experimental results may differ significantly
depending on, for example, the locations of the gas and liquid sampling in the column, the
raw gas concentrations of ammonia or the pre-treatment of packing.
The paper presents the results of the test series considering different effects on the derived
mass transfer performance of the packing and addresses also problems as well as solutions
concerning the sampling in a two-phase regime. The aim is to provide a valuable contri
bution to the efforts for the standardisation of mass transfer measurements in absorption
systems.
2015 The Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.

Introduction

Mass transfer m easurem ents at pilot plants are the basis for
the dim ensioning of industrial colum ns as well as for the
m ass transfer m odels developed during the last years. For
this reason the quality of m ass transfer m easurem ents is very
im portant w hich have to be conducted as conscientiously as

possible. A lot of influences on m ass transfer m easurem ents

exist w hich need to be considered. Many of th em are already
m entioned in Hoffman et al. (2007), Rejl et al. (2009) as well as
in Kunze et al. (2012). Nevertheless, there are further im por
ta n t influences to be reflected and in this paper th e effects of
th e chosen raw gas concentration, sam pling position, and the
p re-treatm en t of plastic packing are considered.

* Corresponding author. Tel.: +43 38424025010.

E-mail address: verena.wolf@unileoben.ac.at (V. Wolf).
http://dx.doi.Org/10.1016/j.cherd.2015.04.005
0263-8762/ 2015 The Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.

CHEMICAL ENGINEERING RESEARCH AND DESIGN 9 9

( 2 0 I 5 ) 2 2 8 -2 3 5

229

Symbols
A
c
DN
F
FTIR
H
H20
He(TG)

cross sectional area of the column [m2]

concentration [mol/m3]
diameter nominal
F-factor [Pa0 5]
Fourier transformed infrared
height of packed bed [m]
water
Henry constant dependent on gas temperature
[bar]
HTUov height of a transfer unit (overall) [m]
feNH3 (t g; Henry constant dependent on gas tem pera
ture [mol/(l3 atm)]
MM
molar mass [kg/mol]
NH3
ammonia
NTU
number of transfer units [-]
pressure [Pa]
V
T
temperature [C], [K]
liquid load [m3/(m2 h)]
UL
V
volume flow [m3/h]
X
liquid mole fraction [mol/molges]
X
liquid mole load [mol/moljnert]
gas mole fraction [mol/molges]
y
Y
gas mole load [mol/mol;nert;]
density [kg/m3]
p
Indices
G,g
i
in
U

out
OV
S

gas phase
component i
inlet stream
liquid phase
outlet stream
overall
saturation

2.

Methods

2.1.

Experimental setup

The experimental plant at the Montanuniversitaet Leoben

consists of a DN 600 saturation column and a DN 450 m ea
surement column made of polypropylene which can be seen
in Figs. 1 and 2.
The air is sucked from the environment and flows through
the saturation column where it is saturated with water. The
relative humidity of the sucked air and of the air leaving the
saturation column are measured. After passing a tube type gas
distributor, positioned 1 m below the packing, the gas streams
through the m easurem ent column. Afterwards, it enters a
radial blower (frequency converted gear, max. 5000 m 3/h, max.
5000 Pa pressure increase) and is then finally released to the
environment. The gas flow is measured by a flow grid (airflow)
downstream of the blower. Furthermore, the temperature of
the gas is measured at every in- and outlet of both columns
(Endress & Hauser, PT 100 type Omnigrad M TR 13; WIKA,
type 55) which have their own liquid circulation. The pack
ing is irrigated by a trough type liquid distributor which is
situated 0.15 m above the packing (128 dripping points per
m2, free cross sectional area of 41%, three different oper
ating ranges are available: 50-30 m 3/(m2 h), 10-60 m 3/(m2h),
30-120 m 3/(m2h)). While the saturation column possesses a

Fig. 1 - Experimental plant packed column at

Montanuniversitaet Leoben.
closed liquid circulation, the liquid in the measurement col
umn is only used for one single passage. It is collected below
the gas entrance of the column in a pan type collector and
conducted to a bin outside of the column. The used radial
pumps are able to handle a volume flow of up to 50 m 3/h. The
liquid flow is measured by a magnetic inductive flow meter
(Krohne, type optiflux 4000) and the temperature of the inlet
stream at the m easurement column is gauged by a resistance
thermometer (Voltcraft, type K).
The mass transfer characteristic has been determined by
using the absorption model system ammonia (NH3 )-air/water.
The mass transfer resistance for this system is located in
both phases although it is predominant in the gas phase
(Hoffmann et al., 2007). This mass transfer system has been
chosen because it is established at many experimental plants
and therefore m easurement data for validation exists.
The gaseous ammonia is injected in the gas duct between
the saturation and the measuring column and has to pass a
static mixer to ensure an optimal mixing. A constant ammonia
dosage is ensured by the use of a pressure regulator, a valve
as well as a float-type flow meter.
The concentrations of ammonia in front of and behind
the packing are measured for each operating point (constant
F-factor and liquid load) by an FTIR spectrometer (Gasmet
Technologies Oy, DX4000) in real time at a frequency of 1Hz.
Furthermore, the aqueous vapour concentration of the gas
stream is also measured every second by the FTIR. This m ea
surement is a key factor in monitoring that no water droplets
are sucked in at the gaseous sampling location so that the test
results do not get falsified as ammonia is readily soluble in
water.
The concentration of ammonia in the liquid phase is ana
lysed by photometric determination after it was preserved by
the use of hydrochloric acid as receiver. This method has been
selected as it is not sensitive to the C02-CaCC>3-equilibrium
present in the piped water used due to a lack of huge amounts
of deionised water for the necessary measurements (Gucher,
2013).
At the beginning of a test run, the experimental plant is
only operated with water and air until steady-state conditions
for the chosen operation point are reached. The mass transfer

230

CHEMICAL ENGINEERING RESEARCH AND DESIGN 9 9

measurements are operated at these adjusted liquid and gas

loads. Then, the metering of ammonia is started where the
amount of ammonia is adjusted in a way that the measured
outlet concentration above of the packing reaches approxi
mately 100 ppm. Steady-state conditions are assumed when
a constant NH3 -concentration in the gas outlet stream for a
time period of 1-2 min is achieved. Now, the gas measuring
position inside the column is switched to the gas sampling
location below the packing and again the measurement is
operated until a constant value for 1-2 min is realised. After
wards, the liquid samples above and below the packing are
taken simultaneously as described in Section 2.1.1.
The concentration of approximately 100 ppm in the gas
outlet stream has been chosen to ensure a good measuring
result of the FTIR. Furthermore, a bed height of 0.75 m has been
chosen to ensure that at least 5% of the inlet gas concentration
can be measured in the outlet gas stream as recommended by
Rejl et al. (2009).

( 2 0 I 5 ) 2 2 8 -2 3 5

of water, ammonia, and carbon dioxide are analysed. Based

on the fact that ammonia is readily soluble in water and as
there is interference between water and ammonia concerning
their infrared spectra, it is essential that only gas is sucked
in and no water droplets. As the sampling position is located
straight above and below the packing, in particular in the twophase regime, a special sampling device is necessary, which
was developed especially for this purpose at the Chair of Pro
cess Engineering. The m easurem ent pipe which leads directly
to the FTIR spectrometer is situated in a double-tube, where
a small, but long (over the whole cross section of the column)
slot is milled in both pipes, the inner one at the top and the
outer one at the bottom side as shown in Fig. 4.
2.1.1.2. Liquid sampling. The liquid should also be taken
straight above and below the packing. For this purpose a sam
pling device has been developed which is illustrated in Fig. 5.

2.1.1. Sampling
Concerning the sampling of the liquid and the gaseous phase,
following has to be considered:
The sample needs to be representative, which means that
there should not be any disturbing components present and
the sample volume should be high enough.
On the other side, it is very important to avoid a signifi
cant disturbance of the gas and the liquid phase and thus
maintaining the fluid dynamic conditions unchanged.
Furthermore, the sampling positions should be located in
such a way that only the mass transfer capacity of the pack
ing itself is measured, which means the sampling should
take place straight above (Fig. 3) and below the packing.
Therefore, various sampling devices have been developed
which are described below.
2.1.1.1. Gas sampling. For the analysis of the gas phase an
amount of 1.41/min is sucked from the column (optional from
above or below the packing) through a heated pipe (180 C) to
the pre-heated FTIR spectrometer, where the concentrations

Fig. 3 - Gas and liquid sampling device directly above the

packing.

CHEMICAL ENGINEERING RESEARCH AND DESIGN 9 9

231

( 2 O I 5 ) 2 2 8 -2 3 5

Fig. 4 - View of the gas measuring probe (Photograph and

schematic view).

It is designed in such a way that the liquid is collected in the

bulk phase of the column. This prevents the sampling of liq
uid close to the column walls which interaction with the gas is
limited and which is therefore not representative (Griinewald
et al., 2011).
During sampling the total volume of the liquid sample col
lector is emptied three times before the sample is taken to
ensure a fresh and representative sample. Furthermore, the
horizontal position of the three possible liquid collectors along
the height of the column is chosen in a way that they do not
interact with each other. As shown in Fig. 6, they collect the
liquid at the same diameter inside the column but they are
shifted by angles of 45 and 90 degrees. The exact positions of
the liquid sampling devices along the column height can be
seen in Fig. 7.

3.

Calculations

3.1.

Gas load factor F

Fig. 6 - Arrangement of the liquid sampling devices.

ps is calculated with the Antoine equation (Eq. (3)) with the

constants given in Gmehling and Kolbe (1992).
Pair, humid

X \ / ; x [kg/m3]

[Pa]_______

287.06 x (273.15 + Tajr [-C])

^1 - 0.378 X

Ps [Pa] \
Pair [Pa

Ps

[Pa] =

])

1 0 (1019621-(1730.63/233.426+Tair[C]))

(2)

(3)

Irrigation density

The liquid flow is also applied to the cross-sectional area of

the empty column as shown in the following equation:
Vl
u l

3.3.

F ^Pa'5] =

__________ Pair

3.2.

For the description of the gas flow in a mass transfer columns

the F-factor is used, where the gas flow is based on the cross
sectional area of the column and multiplied with the square
root of the gas density to get comparable results.

[k&'m3]

[ m 3/ h ]

A [m2]

(4 )

Height of a transfer unit-HTU

(1)

The therefore required gas density is calculated based on

the ideal gas equation by Eq. (2) with the assumption that
the air is completely saturated with water after passing the
saturation column. The required saturated vapour pressure

The experimental results of the mass transfer measurements

are used for the calculation of the required packing height of
industrial columns. For this purpose the kinetic mass trans
fer model of Chilton and Colburn as shown in Eq. (5) is used
where the number of transfer units (NTU) is multiplied with
the height of one transfer unit (HTU).
H [m] = HTU x NTU = HTUVx NTU9V= HTU|,Vx NTU^

(5)

For dimensioning a new column the number of transfer

units can be calculated from the specified concentrations
in the in- and outgoing streams. Conceptually, this value
describes how often the potential of the driving force has to
be exhausted to reach the desired outgoing concentrations
based on the ingoing concentrations. In Eqs. (6a) and (6b) as
well as in the further described equations the mass transfer

232

CHEMICAL ENGINEERING RESEARCH AND DESIGN 9 9

( 2 0 I 5 ) 2 2 8 -2 3 5

With the assumption of a linear equilibrium as well

as balance line, which is fulfilled concerning the used
ammonia/air-water test system under the used operating con
ditions, as well as if the calculation is done with molar loads
instead of molar fractions, the calculation of NTU can be done
with a logarithmic average (Eqs. (7a) and (7b)) (Lehner et al.,
2011)

Yin -Yout x In [' AYin \

AYin AYout
^AYout /

(7a)

NTU[,V=

Xout Xin
/ AX0ut \
x In (
AX0ut ~ AXin
^ AXin /

(7b)

n ti 4

AY describes the difference between the existing molar load in

the gas phase and the theoretically to this value corresponding
equilibrium load in the liquid phase, besides AX defines the
difference between the equilibrium gas load and the existing
molar load in the liquid phase. The equilibrium liquid, respec
tively, gas load can be calculated by the use of Henrys law for
molar loads, as shown in the following equations:
(He (Tg) /p) x (Xj/1 + Xj)
1 _ (He (Tg) /p) x (X;/l + Xj)
AXj = X* - X;

(Yj/1 + Yj) x (p/He (Tg))

1 _ (Yj/1 + Yj) x (p/He (Tg))

(8a)

(8b)

The herein used temperature dependent Henry-constant

is calculated by the help of Eq. (9) based on the Vant Hoff
equation. The therefore required values, in case ammonia is
absorbed by aqueous solutions, are published by Sander (1999)
which are established on the relationship given by Edwards
et al. (1978).
feNH3 (Tg) [mol/13 atm] = 61 x exp ^4200 x ^

Tg [K]

298.15

))
(9)

He (Tg) [bar]
^Solution

[kg/m3]

MMsolution [kg/mol] X feNH3 [mol/13atm]

1000 [l3/m3]
(10)

Eq. (10) is used to translate the calculated Henry-constant

in the pressure form. The herein used density of the
ammonia-water solution which is dependent on the concen
tration of ammonia is shown in Eq. (11) and has been found
by fitting the values given in Kiister et al. (1972).
/solution [ k g to ta l^ to ta l] =

4-3 x 10 8 x c,H3

- 0.0074 x cNh3 + 997.99

calculation for the mass transition for the gas side g as well
as for the liquid side 1are outlined.
xin

NTUr

dY
Y-Y*

-/<

(11)

The molar mass of the ammonia-water solution can after

wards be calculated with the density of the solution by using
Eq. (12) and the water concentration by Eq. (13).

(6a)
^Solution [k g t0ta i/m t30taI]

MMSo]uti0n [kgtotal/moltotal]

Yout

^Solution [ m o lto ta l/m [otal]

Xout

n tu |,v =

(6b)

___________ /Solution [kgtotal/rc^otal]___________

ch2o

[molH2o/mt3otal] + cNh3 [molNH3/m3otal]

(12)

CHEMICAL ENGINEERING RESEARCH AND DESIGN 9 9

ch 2o

^Solution [kgtotal/m t0tal]

[molHjo/m,total
3

233

( 2 0 I 5 ) 2 2 8 -2 3 5

/partial,NH3 [^ S N H s ^ to ta l]

MMh2o [kgH20/m olH2o]

/^Solution fkgtotal^m !;otal] ^ ^ 3

(13)

j^molNH3/m 30^ j x MMnh3 [kgNH3/m olNH3

MMh2o [kgH20/molH2o]
It is im portant to use the molar loads instead of the molar
fractions for the calculations as otherwise the mass flow of
the inlet and outlet stream would change too much. The cal
culation of the loads out of the fractions is shown in Eq. (14)
exemplary for the gas side (Draxler and Siebenhofer, 2014).

Furthermore, the presented results concerning the location

of the sampling as well as the raw gas concentration were
repeated to show the reproducibility of the measurements.
4.1.

y, [molj/moltotaI]
Yj [molj/moljnert] =
1 - y ; [mol;/moltotal]

(14)

The height of transfer unit is necessary for the calculation

of a column height. It can be calculated as shown in Eqs. (15a)
and (15b) as the used packing height is known and the number
of transfer units can be calculated with the measured in- and
outgoing concentrations during the experiment as shown in
Eqs. (7a) and (7b).
HTU3V[m] =

H
HTUoV

(15a)

HTUoV[m] =

H
HTUL

(15b)

For the calculation of the NTU value the measured concen

trations have to be transformed into molar loads which are
calculated with Eq. (14) for the gas phase and Eqs. (15)(17) for
the liquid phase.
Yj [molj/molinert = molNH3/m olair] =
L

^ [PPm l * 10
1 - y; [ppm] x 10-b
(16)

X, [molj/moijne,, - molBH!/mol!0] = ai*

Location of sampling

Hoffmann et al. (2007) recommend that the sampling position

for the gas and the liquid phase should be situated directly
below the packing to eliminate the mass transfer which takes
place below the packing but they also suggest locating the
sampling in the inlet and outlet tubes to get a representative
sample. Rejl et al. (2009) suggest to situate the sampling loca
tion inside the packing to eliminate end effects. Therefore,
the influence of the sampling location has been examined.
All examined sampling positions are marked with numbers
(gas sampling) or letters (liquid sampling) in Fig. 7. In fact
only the sampling positions below the packing were varied.
Therefore, the liquid inlet (a) and the gas outlet concentration
(1) were measured straight above the packing. The gas inlet
concentration was measured straight below the packing (2)
and alternatively in the inlet tube (4). The liquid outlet con
centration was measured directly below (b) as well as 60 cm
below (c) the packing and in the outlet tube (d). The results
of these measurements are depicted in Fig. 8 together with
the mass balance calculations, where the HTUov values are
plotted against the F-factor. The mass balance is calculated by
comparing the total NH3 amount of the inlet gas and liquid
stream with the one of the outgoing streams (Eq. (18)) and the
determination of the deviation of thus received values.

["W C - l * MM- [ W 1 h, o]

(17)

Ppartial,H20 [kgH2CI/ m 30tal]

Ppartial,H20 [^Stotal^m total] = ^solution [^total^m total] ^partial,NH3 [k g NH3/ m tttal]

Psolution [kgtotal/m?otal]

4.

cnh3

(18)

[molNH3/m t3otal] X MMNh3 [kgNH3/molNH3]

Results

Although the used experimental setup of the pilot plant fol

lows strictly the recommendations published by Hoffmann
et al. (2007), further influences on the experimental results
have been observed, which are described below. All experi
ments were carried out with random plastic (polypropylene)
packing with a nominal size of 50 mm, a void fraction of
94%, and a specific surface area of 90m2/m 3. Due to the fact
that the used air is sucked in from the environment it is
important to mention that the ambient conditions (relative
humidity, temperature, and ambient pressure) were recorded
and considered for interpretation of the results. All presented
experimental results are normalised based on the highest
value shown in each diagram due to a non-disclosure agree
ment. The uncertainty analysis has been performed based on
the propagation of uncertainty of GaufS (DIN 1319,1983). The
measured and the calculated variables as well as their uncer
tainties can be obtained from Table 1.

cNH3,L,in

X Vl +

CNH3,G,in x

Vg = CNH3,L,out X Vl + CNH3,G,out X Vq

(19)
Comparing the measurements [g:l2, l:a-b] and [g:l2, 1:
a-c] it can be recognised that the HTUov values differ only by
2% but the mass balance has an deviation of about 60%. Fur
ther it can be seen that the HTUov values of the measurements
[g:l- 2 , l:a-b] and [g:l- 2 , l:a-d] show a deviation of 10% and the
mass balance already of about 150-300%.
Of course the huge difference concerning the mass bal
ance can be explained due to the not closed mass balance
area. Therefore, the m easurement [g:l
4, l:a-d] took this into
consideration which is also visible in the best matching mass
balance (-3-10%) in Fig. 8 . However, these sampling positions
lead to 65% lower HTUov values than [g:l- 2 , l:a-b] due to the

234

CHEMICAL ENGINEERING RESEARCH AND DESIGN 9 9

( 2 0 I 5 ) 2 2 8 -2 3 5

1 Table 1 - Results from the uncertainty analysis.

Calculated variable

M easured variable

Max. uncertainty of calculated variable

F-factor

Gas flow (3 Pa + 0.5% of m easuring range)

Column diam eter (1 mm)
Atm ospheric pressure (100 Pa)
Gas tem perature (1C)

0.2 Pa0-5 or 7%

Irrigation density

Liquid flow (0.2% of m easured value)

Column diam eter (1 mm)
Packing height (1 cm)
Gas concentration (4% of m easured value)
Liquid concentration (5% of m easured value)

0.4m 3/(m 2 h) or 0.3%

HTUov

0.025 m or 17%

H T U 0V = f(F )

Fig. 8 - Influence of sampling location and mass balance therefore.

interaction of gas and the liquid droplets in the column section

below the packing. Therefore, it is obvious, that the sampling
position has a huge influence on the determined HTUov val
ues, whereas the liquid concentrations do not have such a
decisive effect on the HTUov values. Furthermore, a closed
mass balance is not an indicator for the right sampling posi
tions.
4.2.

calculated and are shown in Fig. 9. The gas side HTUov values
are lower compared to the liquid side ones. As shown in Fig. 9
an almost doubled raw gas concentration leads to 10% lower
HTUov values for the gas phase as well as for the liquid phase.
This indicates that the influence of the chosen gas concen
tration cannot be neglected. In order to achieve comparable
measurements, additional examinations would be necessary
to give an indication what gas concentration should be used.

Raw gas concentrations

4.3.

The amount of ammonia dispensed into the air should be high

enough to limit any measurement mistakes on the one hand,
but on the other hand ammonia gas should also not be wasted.
Therefore, it was investigated, if the chosen raw gas concen
tration has an influence on the measured mass transfer per
formance. Due to the fact that the mass transfer resistance in
the used system ammonia/air-water is located in both phases,
the gas as well as the liquid side HTUqv values have been

In general, it is known that plastic packing improve their mass

transfer performance during operation compared to their vir
gin state at the start-up. However, there are no publications
about how long or in which way plastic packing should be
pre-treated before they are examined in mass transfer mea
surements. Results concerning the fluid dynamic behaviour
of this issue are published by Wolf and Lehner (2013). Woicke

H T U 30V = f(F )
D e p e n d a n c e o f G a s C o n c e n tra tio n
-lo w concentration
low concentration repeated

H TU 'ov = f(F )
D e p e n d a n c e o f G a s C o n c e n tra tio n
.

System: NH3-alr/w ater

H=0.75m
uL=30m 3/(m 2h)

-h ig h concentration
high concentration repeated

o 0,100
>

Pre-treatment of plastic packing

l o w concentration

1,3

1,2

-O low concentration repeated

L l,1

S y s te m : N H 3-a lr/w a te t
H = 0 .7 5 m
uL= 3 0 m 3/(m 2h)

- - h i g h concentration
high concentration repeated
a s *

i P 09
i 1 0,8

I
1

3,0
F-Factor F [Pa0 5]

low cone.

2.3
3.4
3.9

115
112
121

2.3
3.4
3.9

111
109
122

2.3
3.4
3.9

215
230
206

512
410
424
rep.
478
405
430
C.
1090
1004
833

high cone. rep.

2.3
3.4
3.9

0,6
0 ,5

2,5

y.

IPa05] [ppml [ppm]

low cone.
-

! ^ 1>
n

1,4

2 5

3 0

F-Factor F [Pa05]

Fig. 9 - Influence of gas concentration.

3 5

4, 0

185
202
198

944
902
814

CHEMICAL ENGINEERING RESEARCH AND DESIGN

H TU % f(F)

Fig. 10 - Influence of the pre-treatment of plastic packing.

et al. (2012) have already reported the occurring effect concern

ing the mass transfer performance. Nevertheless, the results
in Fig. 10 should point out how long it takes to establish
stable conditions. The HTUov values at the 4th and 5th irri
gation are about 40% lower at an F-factor of 1.2Pa0-5, 20% at
F-factors 2.3 Pa0-5 and 3.2 Pa0-5 and 11% lower at an F-factor of
3.9 Pa05 than the ones found at the 1st irrigation. The therefore
deduced recommendation is to perform first mass transfer
m easurem ents under the same conditions with a brand-new
plastic packing as long as it takes to achieve stable conditions.
Only then valid measurements are obtained.
Summing up the alternating irrigation times for
the presented results it required all together 2h of
ammonia-air/water and 2h of simple air/water irrigation
to achieve the presented stable conditions after the fourth
irrigation. The packing was left wet between the m easure
ments of the single irrigation runs, which also can have
an influence on the surface of the packing. It needs to be
mentioned that the presented results were measured over a
complete duration of five days.

5.

Conclusion and outlook

It has been shown that the influences of the chosen raw

gas concentration, the sampling location as well as the pre
treatm ent of the plastic packing have a significant influence
on mass transfer measurements and should therefore not be
neglected. This work is intended to contribute to the efforts
to standardise mass transfer measurements although for an
accurate understanding of the shown influences further sys
tematic investigations would be necessary, particularly with
other test systems as recommended by Hoffmann et al. (2007)
and Rejl et al. (2009). Nevertheless, the following recommen
dations can be pointed out:
Location of sampling: The sampling positions where the
concentrations are measured have to be chosen carefully
and it is necessary to define whether they should be situ
ated in the pipes to/from the m easurem ent column, directly
above or below of the packing or/and inside of the packing.
Due to the presented results and to avoid effects of the plant
configuration the positions directly above, below, and inside
of the packing are recommended. Furthermore, if the sam
pling is done inside of the column, it is essential to get an
integral sample over the whole cross section of the column
diameter as the mass transfer in the bulk and the wall zone
differs.

9 9

(2 0 I 5)

228-235

235

Raw gas concentration: It is not useful to leave the deci

sion open about which gas concentration shall be used. The
choice of the used raw gas concentration is dependent on
the measuring equipment as well as on the packing height
as the clean gas concentration has to be in the measuring
range. The authors suggest to define special gas concentra
tions for the different available test systems to ensure better
comparability. The definition of these concentrations would
need further investigations.
Pre-treatment of packing: A detailed procedure for the pre
treatm ent of plastic as well as metal packing has to be
defined to guarantee that mass transfer measurements in
different experimental plants lead to the same results. To
develop such procedures further investigations are neces
sary in particular addressing the surface properties and how
they are changeable.

Acknowledgement
Financial support from RVT Process Equipment GmbH is grate
fully acknowledged.

References
DIN 1319,1983. Grundlagen der Messtechnik. Beuth, Berlin.
Draxler, J., Siebenhofer, M., 2014. Verfahrenstechnik in Beispielen.
Problemstellungen, Losungsansatze. In: Rechenwege.
Springer Vieweg, Wiesbaden.
Edwards, T.J., et al., 1978. Vapor-liquid equilibria in
multicomponent aqueous solutions of volatile weak
electrolytes. AIChE J. 24, 966-976.
Gmehling, J., Kolbe, B., 1992. Thermodynamik. 2. iiberarbeitete
Auflage. VCH-Verl.-Ges, Weinheim.
Griinewald, M., et al., 2011. Auslegung von
AbsorptionskolonnenNeue Problemstellungen fur eine
altbekannte Aufgabe. Chem. Ingen. Techn. 83 (7), 1026-1035.
Gucher, A., 2013. Entwicklung eines Verfahrens zur quantitativen
Bestimmung von Ammoniak in der Fliissigphase einer
Absorptionskolonne. Montanuniversitat Leoben, Lehrstuhl fur
Verfahrenstechnik des industriellen Umweltschutzes, Leoben
(Masterarbeit).
Hoffmann, A., et al., 2007. Standardization of mass transfer
measurementsa basis for the description of absorption
processes. Trans. IChemE., A 85 (Al), 40-49.
Kunze, A.K., et al., 2012. Die Notwendigkeit einer
Standardisierung von Stofftransportmessungen in der Abund Desorption. Chem. Ingen. Techn. 84,1931-1938.
Kuster, F.W., et al., 1972. Logarithmische Rechentafeln. fur
Chemiker, Pharmazeuten. In: Mediziner und Physiker. 101.
Walter de Gruyter, Auflage, Berlin.
Lehner, M., et al., 2011. Hydrodynamic and mass transfer
characteristic of a novel grid structured plastic packing. Heat
Mass Transf. 47 (8), 1035-1041.
Rejl, J.F., et al., 2009. Methods standardization in the
measurement of mass-transfer characteristics in packed
absorption columns. Chem. Eng. Res. Des. 87, 695-704.
Sander, R., 1999. Compilation of Henrys Law constants for
inorganic and organic species of potential importance in
environmental chemistry. Max Planck Institute of Chemistry,
Air Chemistry Department, Mainz,
(http://www.henrys-law.org/henry.pdf).
Woicke, N., et al., 2012. Grofiere
StoffaustauschleistungVerbesserung der Benetzbarkeit von
Kunststoffpackungen. CIT Plus 15 (11-12), 56-58.
Wolf, V., Lehner, M., 2013. Einfliisse bei der experimentellen
Bestimmung der hydraulischen Parameter von Fiillkorpern
und Packungen. 9. Minisymposium der Verfahrenstechnik. In:
Book of Abstract., pp. 73-77.

Copyright of Chemical Engineering Research & Design: Transactions of the Institution of

Chemical Engineers Part A is the property of Elsevier Science and its content may not be
copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for
individual use.