Anis Silfiyatus Sa'adah - 01111940000016 - Review Jurnal
Anis Silfiyatus Sa'adah - 01111940000016 - Review Jurnal
Oleh :
Anis Silfiyatus Sa’adah
01111940000016
DEPARTEMEN FISIKA
TAHUN 2022
1
Pengamatan in-situ grafena menggunakan Mikroskop Optik
Mikihiro Kato, Sujun Guan, Xinwei Zhao (2021)
Tujuan : Dengan mempertahankan sampel pada 9000C dalam atm,
1. Mengamati kelayakan pengamatan in-situ grafena hene tergores secara bertahap.
dekomposisi grafena dengan menggunakan OM
(Optical Microscope) yang dilengkapi dengan tahap 2. Pengamatan graphene dari Cu foil
vakum pemanas
2. Mengamati perilaku multi-layers graphene (MLG)
pada substrat logam
3. Mengamati pengaruh film graphene yang
ditumbuhkan CVD (Chemical Vapor Deposition)
pada foil logam selama pemanasan dalam vakum dan
atmosfer gas Ar + H2.
Hasil :
1. Pengamatan graphene dari Ni foil
Kesimpulan :
Perubahan MLG pada Ni dan grafena pada Cu dapat
dikenali melalui citra mikroskop optik. Pengamatan in-situ
dekomposisi grafena telah dilakukan dengan menggunakan
mikroskop optik yang dilengkapi dengan ruang vakum
pemanas. Karakteristik perubahan permukaan tersebut
sesuai dengan kondisi pertumbuhan grafena sebagai
katalisator Ni dan Cu. Puncak G-band bergeser ke sisi
energi yang lebih rendah, dengan meningkatnya suhu
panas dalam ruang hampa dan atmosfer gas Ar + H2.
Metode ini sederhana dan cukup efektif untuk memperoleh
informasi tentang kondisi pertumbuhan area grafena yang
luas pada substrat seperti Ni, yang memiliki tingkat
kelarutan karbon yang tinggi. Mikroskop optik dapat
mengamati grafena pada logam dengan mudah. Hasil
Gambar 2. Menunjukkan hasil OM in-situ dari grafena
pengamatan menunjukkan secara nyata dengan gambar
yang ditumbuhkan CVD pada permukaan sampel Ni berwarna dan akan memberikan cara sederhana untuk
selama kenaikan suhu di atmosfer Ar + H2. Ketika sampel memahami keadaan grafena dalam berbagai kondisi
dipanaskan dalam Ar + H2, area berwarna biru tua dan
atmosfer.
oranye muncul pada permukaan sampel pada Gambar 2.
Proses pereaksian ini berbeda dengan proses pereaksian
pada kondisi vakum. Daerah ini adalah karbon amorf (aC).
Setengah dari area grafena tergores oleh gas Ar + H2.
2
Dehidroksilasi dan kristalisasi gelas: Studi DTA
D.C. Lago, M.O. Prado (2013)
Differential Thermal Analysis (DTA) adalah teknik yang
sering digunakan untuk mempelajari kinetika kristalisasi
serbuk kaca selama sintering. Penelitian ini bertujuan
untuk menganalisis bagaimana puncak endotermik
dehidroksilasi yang muncul pada pengukuran DTA
bergantung pada area spesifik dan pada laju pemanasan.
Penelitian ini dilakukan pada gelas yttrium aluminosilicate
(YAS), samarium aluminosilicate dan soda-lime silica,
slab dan powder.
A R T I C L E I N F O A B S T R A C T
Keywords: In-situ observation of graphene by using visible light was very few. Herein, we investigate the feasibility of in-situ
In-situ observation observation of graphene decomposition by using the optical microscope (OM) equipped with a heating vacuum
Graphene stage. In this work, in addition to in-situ observation, we investigated the behavior of CVD-grown graphene on
Optical microscope
metal (Ni and Cu) surfaces using in-situ Raman spectroscopy and X-ray photoelectron spectroscopy to understand
CVD
the surface changes observed in the OM images. Observations were also made in vacuum and argon hydrogen gas
flow conditions. As a result of the observation, it was possible to observe the decomposition of graphene using
visible light at high temperatures. We expect it to be an easy way to observe graphene under various atmospheric
conditions.
* Corresponding authors.
E-mail addresses: 1214702@alumni.tus.ac.jp (M. Kato), xwzhao@rs.kagu.tus.ac.jp (X. Zhao).
https://doi.org/10.1016/j.apsadv.2021.100138
Received 27 March 2021; Received in revised form 2 July 2021; Accepted 5 July 2021
Available online 14 July 2021
2666-5239/© 2021 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
M. Kato et al. Applied Surface Science Advances 6 (2021) 100138
2. Experimental condition nor the vacuum (3.0 × 10− 5 Torr) changed noticeably, until
700 ºC. As shown in Fig. 1(c), above 750 ◦ C, the changes began to occur
2.1. CVD-grown graphene films in the region I and II, circled in black. The vacuum level (3.0 × 10− 5
Torr) did not change, suggesting that dissolution into Ni started to occur.
The starting sample was the polycrystalline Ni foil (>99%: The Reported that the precipitation and segregation reactions have been
Nilaco Corporation, Japan) surface coated with a thin carbon film (in the observed shown from 725 ◦ C to 900 ◦ C [24]. In addition, In-situ SEM
form of disordered few layer graphene) grown by thermal CVD. The observations also show dissolution into Ni at a similar temperature
carbon deposition was carried out with ethanol vapor supplied by range. [16]
bubbling liquid ethanol with an Ar gas at 870 ◦ C for 30 min (total Therefore, it is revealed that the change is due to the carbon dis
pressure of 5 Torr). solved in Ni started to re-deposit or the surface MLG started to dissolve
In the same way, graphene films were formed on the surface of the into Ni, during the warming-up process. Afterwards, the observation
polycrystalline Cu foil (>99.99%: The Nilaco Corporation, Japan) at 950 with heating as shown in Fig. 1(d) - (f), the phenomenon that occurred in
ºC for 10 min by thermal CVD using bubbled ethanol as the carbon the image of Fig. 1(c) triggered a gradual change in the appearance of
source. Carbon deposition was conducted with ethanol vapor supplied the entire sample as the temperature increased. From the image in Fig. 1
by bubbling liquid ethanol with an Ar gas (total pressure of 5 Torr). (e) - (f), the surface change in region III is more pronounced than in the
lower temperature image, and the difference in the change in the three
2.2. Characterization regions is likely due to the crystal orientation and surface shape of Ni.
[16] Compared to regions I and II, region III is assumed to be a place
The change of the CVD-grown graphene films had been observed by with fewer irregularities that are an obstacle to solid solution and pre
an in-situ OM instrument (KH-7700, Hirox). The magnification range of cipitation. There are black lumps on the surface of regions I and II. The
the objective lens is from 140 to 1400 times. The light source is a metal carbon that could not be solidly solubilized in the uneven areas has
halide lamp. The heating stage for the in-situ observation is set in a become amorphous carbon. It was observed at 900 ◦ C for about 5 min
vacuum chamber (S.T. Japan Co., Ltd) with an observation window. The (See the movie in the video). In addition, the field of view is narrower
observation window is made of quartz. The temperature was from RT to than in Fig. 1 due to the specifications of the device. Further observation
900 ºC. In the case of vacuum heating, the pressure of the vacuum with heating changed the state of the surface significantly. From the
chamber was kept better than 10− 5 Torr. While heated in Ar + H2 gas movie, it is observed that appears to be "green" on the whole is
atmosphere, the Ar + H2 gas has flowed at a rate of 30 ml/min. A decreasing. In the case of graphene with multiple layers, the RGB count
portable Raman spectroscopy system (i-Raman EX, B&W TEC) was used of the CCD can theoretically be shown to have a higher G (green) count
for in-situ observation of Raman. The 532-nm of an Nd-YAG laser was than R (red) and B (blue) [25]..
used for exciting the Raman scattering. The spectra of C 1 s was analyzed On the other hand, when the sample was heated in Ar + H2, the dark
by X-ray photoelectron spectroscopy (XPS, JPS-9010MC, JEOL). blue and orange areas appeared on the sample surface in Fig. 2. This
reacting process was different from that at vacuum condition. These
3. Results and discussion areas are amorphous carbon (a-C). Half of the graphene area was etched
by the Ar + H2 gas. While keeping the sample at 900 ◦ C in the atmo
3.1. Observation of graphene of Ni foil sphere, the graphene was etched away gradually (Fig. 2e~i).
Fig. 1(a) shows the OM images of the MLG/Ni sample at RT. The 3.2. Observation of graphene of Cu foil
sample surface covered with MLG shows. This sample was heated from
RT to 900 ºC, with a rate of 20 ºC /min. It focused on the three regions I, Fig. 3 shows that the change of CVD-grown graphene films on Cu foil
II and III as shown in Fig. 1(a). From Fig. 1(b), neither the surface during heated in vacuum and Ar + H2 gas atmosphere by an in-situ OM
Fig. 1. (a) OM image of RT. From (b) to (f) In-situ OM images of graphene temperature rising.
2
M. Kato et al. Applied Surface Science Advances 6 (2021) 100138
Fig. 2. In-situ OM images of CVD-grown graphene on the Ni sample surface during temperature rising in Ar + H2 atmosphere.
Fig. 3. OM images of CVD-grown graphene films on Cu foil during heated in vacuum (a) and Ar + H2 gas atmosphere (b).
3
M. Kato et al. Applied Surface Science Advances 6 (2021) 100138
observation. Fig. 3a hardly shows the color change until the temperature situ observation. The information from Raman spectroscopy reflects
up to 400 ◦ C during heated in vacuum. With increasing the temperature information from a much smaller area. It hints that the graphene films
to 500 ◦ C, the color change is obvious. While with further heating over were etching away with H2 during heated in Ar + H2 gas atmosphere at
800 ◦ C, the change is slight. The color changes in the OM image might be high temperatures. In general, Raman peaks are temperature dependent,
caused by the graphene decomposition, which is related on the escape of with the peak position shifting to the lower frequency side and the in
intercalated gas, such as residual oxygen between the metal and gra tensity decreasing as the temperature of the sample increases. The same
phene, and/or the removal of the oxide film [26-28]. While the change trend was obtained in this experiment. The temperature dependence of
of CVD-grown graphene films on Cu foil during heated in Ar + H2 gas the Raman shift in the G-band is independent of the sample. Therefore,
atmosphere is shown as Fig. 3b. Graphene is stable with lower than in the range where the G-band can be observed, we can say that the
400 ◦ C. With increasing the temperature to 700 ◦ C, the color of graphene temperature of graphene on the substrate can be converted from the
becomes part of the green, which might be caused by the etching away Raman shift of the G-band. The slope of the G-band versus temperature
by the Ar + H2 atmosphere at high temperatures. With further heating was about − 0.024 cm− 1/ ◦ C for Ni and Cu foils. However, in the case of
over 900 ◦ C, the change is obvious, which hints that the graphene films heated in Ar + H2 gas atmosphere, the G-band disappeared at 600 ºC,
have completely been etched. due to the etching effect of H2 at high temperatures. The graphene films
It is known that high concentrations of hydrogen and oxygen on the were completely etched away at 900 ºC, which has been confirmed by
sample surface can from highly volatile molecules such as CH4 and CO2 the in-situ OM observation, as shown in Fig. 2 and Fig. 3.
[28,29]. From the experiment in 3.1 and 3.2, we could assume that the Fig. 5 shows that C 1 s XPS spectroscopy of graphene films on Ni and
chemical reaction between oxygen and hydrogen on the sample surface Cu foil, before and after heating in vacuum at 900 ºC for 10 min,
or intercalated in the sample layer formed highly volatile molecules and respectively. From the C 1 s XPS spectroscopy in Fig. 5, the peaks at
etched the graphene. The color of the metal surface also changed when 287.5, 286.1 and 284.6 eV can be attributed to the bonds of C=O, C–O,
comparing the observation in vacuum and the observation with the and C–C, respectively. The presence of the peaks at 287.5 and 286.1 eV
introduction of argon hydrogen. This result indicates that the underlying indicates the replacement of C atom by O atom, which might be from
metal was also affected in some way [30]. Therefore, in addition to the residual oxygen between the metal and graphene during CVD fabrica
basic parameter such as temperature and pressure, impurities formed by tion [27,31]. It could be seen that the C 1 s intensity peak shifts to higher
chemical reaction on the sample surface also affect the observation. energy after heating, which could be caused by metal distortion, and the
escape of intercalated gas such as residual oxygen between the metal
3.3. In-situ Raman and XPS analyzed and graphene [26,27,31]. From the XPS results, the peaks 286.1 eV and
287.5 eV, which indicate bonding with oxygen, are less intense when
To further investigate the influence of the atmosphere on heated graphene on nickel is heated compared to graphene on Cu, as shown in
graphene on Ni and Cu foil, the change of G-band peak position during (a) and (b) of Fig. 5. Therefore, the difference in the in-situ optical im
heated in vacuum and Ar + H2 gas atmosphere is summarized in Fig. 4. ages when heated in a vacuum may be because the change on Ni is due to
Fig. 4 shows the in-situ Raman spectroscopy measured from the CVD- the "dissolution or precipitation of graphene on the metal surface", while
grown graphene films on Ni and Cu foil during heated in vacuum and the change on Cu is due to the "reaction between graphene on the metal
Ar + H2 gas atmosphere. From Fig. 4, the signal of the G-band of gra surface and oxygen adsorbed on the surface or intercalated between the
phene films could be found till 900 ºC on Ni foil and till 600 ºC on Cu foil surface and the substrate". The optical images obtained from the ex
in vacuum condition. This might be caused by the strong background, periments are shown in Fig. 1 and Fig. 3(a), the optical images obtained
and/or the damage of graphene networks, which is matched with that of from the experiments reflect the characteristics of metal catalysts with
in-situ OM. While heated in Ar + H2 gas atmosphere, Fig. 4 shows that different degrees of carbon solidification.
the signal of G-band disappears at the temperature of 800 ºC by Ni foil
and 600 ºC by Cu foil. Excepting Ni in Vacuum experiment condition, 4. Conclusion
there is a difference between the in-situ OM image and the disappear
ance of the signal in the Raman spectra. We thought this was because the In summary, we have investigated the influence of the CVD-grown
scale of Raman laser irradiation was very different from the scale of in- graphene films on metal foil during heated in vacuum and Ar + H2
Fig. 4. G-band peak position (cm-1) of Raman spectroscopy for graphene films on Ni and Cu foil during heated in vacuum and Ar + H2 gas atmosphere.
4
M. Kato et al. Applied Surface Science Advances 6 (2021) 100138
Fig. 5. C 1 s XPS spectroscopy of CVD-grown graphene films on Ni (a) and Cu (b) foil.
gas atmosphere, by combining a portable Raman spectroscopy system observation by real time. Our results are real time observation with color
with the in-situ OM observation. The change of MLG on Ni and graphene image and will provide a simple way of understanding the graphene
on Cu could be recognized by optical microscope images. We have state in various atmosphere condition.
demonstrated that the in-situ observation of graphene decomposition
has been carried out by using an optical microscope equipped with a
heating vacuum chamber. The characteristics of surface change corre Declaration of Competing Interests
sponded to the condition of graphene growth as the using Ni and Cu
catalyst. The G-band peak shifts to lower energy side, with increasing The authors declare that they have no known competing financial
heated temperature in vacuum and Ar + H2 gas atmosphere, respec interests or personal relationships that could have appeared to influence
tively. This method is simple and effective enough to obtain information the work reported in this paper.
on the growth conditions of large areas of graphene on a substrate such
as Ni, which has a high degree of carbon solubility. It could be able to Supplementary materials
observe of graphene on the metal easily by optical microscopy and
Raman spectrum could be capable of atmosphere control as well as an Supplementary material associated with this article can be found, in
the online version, at doi:10.1016/j.apsadv.2021.100138.
5
M. Kato et al. Applied Surface Science Advances 6 (2021) 100138
References [16] K. Takahashi, K. Yamada, H. Kato, H. Hibino, Y. Homma, Surf. Sci. 606 (7) (2012)
728–732.
[17] Z.J. Wang, G. Weinberg, Q. Zhang, T. Lunkenbein, A. Klein-Hoffmann,
[1] A.K. Geim, K.S. Novoselov, Nature Mater. 6 (2007) 183–191.
M. Kurnatowska, M. Plodinec, Q. Li, L. Chi, R. Schloegl, M.G. Willinger, ACS Nano
[2] F. Bonaccorso, Z. Sun, T.A. Hasan, A.C. Ferrari, Nat. Photonics 4 (9) (2010)
9 (2) (2015) 1506–1519.
611–622.
[18] P. Yan, Y.J. Jeong, M.F. Islam, P.C. Pistorius, RSC Adv. 6 (103) (2016)
[3] A. Reina, X. Jia, J. Ho, D. Nezich, H. Son, V. Bulovic, M.S. Dresselhaus, J. Kong,
101235–101241.
Nano Lett. 9 (1) (2008) 30–35.
[19] C.T. Pan, J.A. Hinks, Q.M. Ramasse, G. Greaves, U. Bangert, S.E. Donnelly, S.
[4] X. Li, W. Cai, J. An, S. Kim, J. Nah, D. Yang, R. Piner, A. Velamakanni, I. Jung,
J. Haigh, Sci. Rep. 4 (1) (2014) 1–7.
E. Tutuc, K.S. Banerjee, L. Colombo, R.S. Ruoff, Science 324 (5932) (2009)
[20] Z. Liu, Y.C. Lin, C.C. Lu, C.H. Yeh, P.W. Chiu, S. Iijima, K. Suenaga, Nat. Commun.
1312–1314.
(2014) 5.
[5] X. Li, C.W. Magnuson, A. Venugopal, R.M. Tromp, J.B. Hannon, E.M. Vogel,
[21] K. Kim, J.C. Yoon, J. Kim, J.H. Kim, S.W. Lee, A. Yoon, Z. Lee, Applied Microscopy
L. Colombo, R.S. Ruoff, J. Am. Chem. Soc. 133 (9) (2011) 2816–2819.
49 (1) (2019) 1–8.
[6] D. Wei, Y. Liu, Y. Wang, H. Zhang, L. Huang, G. Yu, Nano Lett. 9 (5) (2009)
[22] T.O. Terasawa, K. Saiki, Nat. Commun. 6 (2015) 6834.
1752–1758.
[23] T. Taira, S. Obata, K. Saiki, Appl. Phys. Express 10 (5) (2017), 055502.
[7] D.R. Lenski, M.S. Fuhrer, J. Appl. Phys. 110 (1) (2011), 013720.
[24] A. Dahal, M. Batzill, Graphene–nickel interfaces: a review, Nanoscale 6 (5) (2014)
[8] S. Bae, H. Kim, Y. Lee, X. Xu, J.S. Park, Y. Zheng, J. Balakrishnam, T. Lei, R.H. Kim,
2548–2562.
I.Y. Song, J.Y. Kim, S.K. Kim, B. Özyilmaz, J.H. Ahn, H.B. Hong, S. Iijima, Nat.
[25] Y.Y. Wang, R.X. Gao, Z.H. Ni, H. He, S.P. Guo, H.P. Yang, C.X. Cong, T. Yu,
Nanotechnol. 5 (8) (2010) 574–578.
Nanotechnology 23 (49) (2012), 495713.
[9] L. Baraton, Z.B. He, C.S. Lee, C.S. Cojocaru, M. Châtelet, J.L. Maurice, H.Y. Lee,
[26] P. Sutter, J.T. Sadowski, E.A. Sutter, J. Am. Chem. Soc. 132 (2010) 8175–8179.
D. Pribat, EPL (Europhysics Letters) 96 (4) (2011) 46003.
[27] P.R. Kidambi, B.C. Bayer, R. Blume, Z. Wang, C. Baehtz, R.S. Weatherup, M.
[10] F. Abild-Pedersen, J.K. Nørskov, J.R. Rostrup-Nielsen, J. Sehested, S. Helveg, Phys.
G. Willinger, R. Schloegl, S. Hofmann, Nano Lett. 13 (2013) 4769–4778.
Rev. B 73 (11) (2006), 115419.
[28] J. Cho, L. Gao, J. Tian, H. Cao, W. Wu, Q. Yu, E.N. Yitamben, B. Fisher, J.R. Guest,
[11] Y. Wang, A.J. Page, Y. Nishimoto, H.J. Qian, K. Morokuma, S. Irle, J. Am. Chem.
Y. Chen, N.P. Guisinger, ACS Nano 5 (2011) 3607–3613.
Soc. 133 (46) (2011) 18837–18842.
[29] E. Starodub, N.C. Bartelt, K.F. McCarty, J. Phys. Chem. C 114 (11) (2010)
[12] C. Mattevi, H. Kim, M. Chhowalla, J. Mater. Chem. 21 (10) (2011) 3324–3334.
5134–5140.
[13] J.D. Wood, S.W. Schmucker, A.S. Lyons, E. Pop, J.W. Lyding, Nano Lett. 11 (11)
[30] S. Choubak, M. Biron, P.L. Levesque, R. Martel, P. Desjardins, J. Phys. Chem. Lett. 4
(2011) 4547–4554.
(7) (2013) 1100–1103.
[14] X. Li, C.W. Magnuson, A. Venugopal, J. An, J.W. Suk, B. Han, M. Borysiak, W. Cai,
[31] R. Larciprete, S. Ulstrup, P. Lacovig, M. Dalmiglio, M. Bianchi, F. Mazzola,
A. Velamalanni, Y. Zhu, L. Fu, M.E. Vogel, E. Voelkl, L. Colombo, R.S. Ruoff, Nano
L. Hornekær, F. Orlando, A. Baraldi, P. Hofmann, S. Lizzit, ACS Nano 6 (2012)
Lett. 10 (11) (2010) 4328–4334.
9551–9558.
[15] G. Odahara, H. Hibino, N. Nakayama, T. Shimbata, C. Oshima, S. Otani, M. Suzuki,
T. Yasue, T. Koshikawa, Appl. Phys. Express 5 (3) (2012), 035501.
6
Journal of Non-Crystalline Solids 381 (2013) 12–16
a r t i c l e i n f o a b s t r a c t
Article history: Differential thermal analysis (DTA) is a technique frequently used for the study of the crystallization kinetics of
Received 5 July 2013 glass powders during sintering. The effect of specific surface area on crystallization kinetics has been reported
Received in revised form 31 August 2013 earlier. It has been shown that there is a shift in the DTA crystallization peaks towards lower temperatures for
Available online 2 October 2013
smaller particle size. This paper analyzes how the dehydroxilation endothermic peaks that appear in DTA
measurements depend on specific area and on the heating rate. Experiments were done on yttrium aluminosil-
Keywords:
Dehydroxilation;
icate, samarium aluminosilicate and soda-lime silica glasses, slabs and powders. We found that some of them
Crystallization; exhibit a dehydroxilation endothermic peak that overlaps at different extent with the crystallization peak. The
Yttrium aluminosilicate glass; dehydroxilation peak is not detected when yttrium aluminosilicate glass powder is preheated at melting temper-
Differential Thermal Analysis atures, or its surface is previously hydrated. For the yttrium aluminosilicate glass a dehydroxilation apparent
activation energy of 255 ± 50 kcal mol−1 was obtained using the Kissinger plot.
© 2013 Elsevier B.V. All rights reserved.
1. Introduction SiO2, and also showed an endothermic peak before the crystallization
peak.
Differential scanning calorimetry (DSC) and differential thermal In Ref. [6], Reben M., in a thermal study of oxyfluoride glass with
analysis (DTA) have been extensively used in the study of the crystalli- strontium fluoride, also found an endothermic peak before the crystalli-
zation kinetics of glasses. Barandiaran and Colmenero [1] used DTA runs zation exothermic event.
to estimate the critical cooling rates of glasses. Cabral et al. [2] used DTA This endothermic peak before crystallization is generally not
to study the stability on heating of glasses that nucleate homogeneous- discussed in the glass literature although it is thoroughly studied in
ly. Both last studies were run with glass slabs instead of powders in mineral structure works during heating [7,8]. Moreover, very often the
order to avoid surface crystallization. temperature for crystallization start, TS, is chosen at the minimum of
There are other papers of DTA measurements, run on glass powders. the endothermic peak which we will show that is not correct.
One of them is that of R. Müller [3]. Müller presented a model for surface
crystallization kinetics of glasses and compared the model results with
experimental data on fresnoite glass. In Ref. [3], the model satisfactorily 2. Experimental
explains the shift towards lower temperatures of the crystallization DTA
peaks for smaller powder particle size. Although not discussed in Ref. 2.1. Glass preparation
[3], DTA runs on some of the fresnoite glass powders exhibited an endo-
thermic event before the crystallization peak, especially for particles Yttrium aluminosilicate glass samples were prepared with composi-
with sizes of 112 and 65 micrometers. For particles of 1.25 micrometers tion: 1Y2O3 · 1Al2O3 · 3SiO2 which corresponds approximately to the
no endothermic peak is seen. core of the ternary glassy domain [9]. Samarium aluminosilicate glass
In Ref. [4], C. Lara et al. studied the crystallization of RO–BaO–SiO2 samples were prepared with composition: 1Sm2O3 · 1Al2O3 · 3SiO2.
(R = Mg, Zn), and it is shown in their DTA runs that for R = Zn there In each case, a mixture of high purity oxide precursors was homog-
is an endothermic peak before the crystallization one, but no when enized in a rotating mill and melted in a platinum crucible for 2 h at
R = Mg. 1600 °C. The melt was poured on a stainless steel plate and immediately
In Ref. [5], Sung-Bum Sohn et al. studied the influence of CeO2 addi- splat-cooled using a second steel plate. A fraction of the glass was sepa-
tion on the crystallization behavior in the glass system MgO–Al2O3– rated and identified as slab. The remainder glass was milled and sieved
to obtain three different powder size samples.
⁎ Corresponding author at: Centro Atómico Bariloche, Avenida Bustillo 9500, R8402AGP,
Soda-lime silica glass samples were prepared with composition
S.C. de Bariloche, Río Negro, Argentina. Tel./fax: +54 294 4445241. 1Na2O · 1CaO · 2SiO2 (NCS). The melting procedure was the same as
E-mail address: dlago@cab.cnea.gov.ar (D.C. Lago). detailed above, at a lower temperature (1500 °C).
0022-3093/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jnoncrysol.2013.09.008
D.C. Lago, M.O. Prado / Journal of Non-Crystalline Solids 381 (2013) 12–16 13
Glass pieces were milled in an RS 200 vibratory disk mill and sieved. 3.2.1. The effect of particle size distribution
In the case of the yttrium aluminosilicate (YAS), the glass fractions Fig. 2 shows the DTA results for a monolith and the different frac-
obtained were those remaining between the following pairs of sieves: tions of YAS glass powder in alumina crucibles. The trace corresponding
50 and 100 μm (fraction 1), between 25 and 37 μm (fraction 2), and to the slab exhibits the Tg inflection and the crystallization peak as
less than 25 μm (fraction 3). The last particle size was obtained by wet expected (being TS the temperature corresponding to the start of the
sieving. crystallization peak and TX, the temperature of the peak maximum).
Samarium aluminosilicate glass and soda-lime silica glass particles The traces corresponding to glass powders present the same structure
between 25 and 37 μm sieves were collected. as for the slab, but moreover they have an endothermic peak at TE.
The particle size distribution of the different glass fractions was mea- The results of the DTA analysis are resumed in Table 1.
sured by laser diffraction with a Mastersizer Micro Malvern Instrument, Fig. 3 shows the overlapping of the DTA trace for YAS fraction 3 and
with simultaneously applied ultrasound to disaggregate particle clus- the cross section area of a powder cylinder during sintering in a hot
ters. The dispersive medium was distilled water. stage microscope (HSM). Shrinking is previous to the endothermic
peak.
Fig. 2. DTA traces for slab and different YAS glass powder fractions. Tg, TE, TS and TX are
Fig. 1. Particle size distribution of different YAS glass powder fractions. indicated.
14 D.C. Lago, M.O. Prado / Journal of Non-Crystalline Solids 381 (2013) 12–16
Table 1
Glass transition and crystallization temperatures for different YAS fractions.
3.2.5.2. Soda-lime silica glass. A soda-lime metasilicate glass (NCS) was YAS powder (b) and YAS powder 30 h in water (c) show at about
also tested. Fig. 7 shows the DTA trace for a slab and powder of a NCS 300 °C a loss of adsorbed water, chemically adsorbed water at 600 °C
glass. No endothermic peak is observed. and gas expulsion due to crystallization above 1122 °C.
From the results shown in this section, we do not detect a change
3.2.6. The effect of the heating rate on the endothermic reaction of weight during the endothermic event. A simple calculation using
For the YAS glass powder, fraction 2, DTA runs were done at 5, 10, 15, Zhuravlev (Ref. [13]) data of maximum silanols (≡Si―OH) surface den-
20 and 40 °C min−1 (see Fig. 8). A Kissinger plot of the temperatures of sity equal to 5 ≡SiOH per nm−2 in silica, and considering that our com-
maximum endothermic reaction rate is shown in Fig. 9. An apparent position is 1Y2O3 · 1Al2O3 · 3SiO2 and that 10% of surface Si has a
activation energy of 255 ± 50 kcal mol−1 is determined for the endo- silanol functional group, 3000 cm2∙gr−1 sample specific surface area,
thermic reaction. and sample mass approximately 50 mg, shows that the mass loss dur-
ing dehydration should be less than 0.3 micrograms, i.e. near the sensi-
bility of our TG equipment.
3.2.7. Thermogravimetry
Fig. 10 shows thermogravimetric behavior of a YAS slab and of YAS 4. Discussion
powder with three different surfaces.
Let us remark that only YAS powder (b) exhibits endothermic peak Fig. 2 shows a shift of the crystallization peaks of YAS powder sam-
in Fig. 5. ples towards lower temperatures as expected from Ref. [3]. However,
We assign the initial mass increase of the slab (a) to measurement the endothermic peaks exhibited by all the YAS powder samples except
error. Above Tg there is a mass decrease of about 1o/oo that we assign by the monolithic one are not explained in the literature.
to a decrease in dissolved gases. For temperatures higher than TS there The same occurs with SmAS glasses and other observed in this paper
is a slight mass decrease due to the decrease of gases solubility in the and the referenced literature.
crystalline phases. The glass transition temperature is the same for all YAS samples,
Glass microspheres (d) show at about 300 °C a slight mass decrease being in slab or powder form, suggesting that there has not been intro-
due to physical adsorbed water and again above Tg due to a decrease in duced glass structure changes during grinding. If grinding changed the
dissolved gases. Both processes show very slight mass changes since glass structure by plastic deformation, this plastically deformed glass
during flame spheroidization most water and dissolved gases were should have a different Tg value, and this is not observed. The glass tran-
expulsed. sition temperature is maintained in all YAS samples, indicating that the
supercooled liquid structure inherited is the same for all samples. We
Fig. 3. Comparison of DTA and HSM curves on the same temperature scale for YAS glass Fig. 5. DTA of YAS glass: (a) slab, (b) powder (fraction 2), (c) powder (fraction 2) + 30 h
powder (heating rate 10 °C min−1). Solid line: DTA trace; dot line: HSM trace. in contact with water at room temperature, (d) glass powder microspheres.
D.C. Lago, M.O. Prado / Journal of Non-Crystalline Solids 381 (2013) 12–16 15
Fig. 6. DTA of (a) YAS glass slab, (b) SmAS glass slab and (c) SmAS glass powder showing Fig. 8. Heating rate effect on the DTA results for YAS glass powder (fraction 2). The endo-
the endothermic peak before crystallization. thermic and the crystallization peak shifts towards higher temperatures with the increase
of the heating rate.
note that the energy involved in this endothermic peak is comparable to nonadjacent hydroxyls are involved, proton diffusion is needed, as
the crystallization enthalpy. was proved by the electrical conductivity increase in measurements
The endothermic peak is due to a surface effect during heating of done up to the dehydroxylation temperature.
the crushed sample, since the monolithic sample does not present the We expect that the particles that were flame polished contain a
endothermic behavior. lower surface density of silanols (≡Si―OH) which conducts to the
However, neither a soda-lime silicate glass powder nor a Pyrex glass almost absence of the endothermic peak (Fig. 5d). In the case of glass
powder tested in this work (DTA not shown in this work) exhibits endo- powder previously in contact with water (Fig. 5c), the increase of sur-
thermic peak (Fig. 7). face silanol density allows the condensation reaction [12,13]:
Moreover in the literature, some glass powder compositions exhibit
a DTA endothermic peak [3-5] and some do not [10,11]. In Ref. [4], mea- ≡Si―OH þ HO―Si≡→≡Si―O―Si≡ þ H2 O↑
surements on two sintering glass powders are shown; one of them ex- to take place at lower temperatures, showing an important mass loss
hibits endothermic peak and the other does not. below 600 °C (Fig. 10c).
The partial overlapping of this peak with the crystallization peak Thus, this surface endothermic reaction, with an apparent reaction
complicates the interpretation of this evidence, which is more clear in enthalpy of 255 kcal mol−1 for YAS glass, as shown in Fig. 9, can be
the case of the YAS glass powder fraction 3 where the endothermic assigned to a dehydroxilation reaction. This value is in agreement with
peak is well defined before the crystallization peak. dehydroxilation measurements for clays [7].
With YAS glass powders, the endo-peak is observable either using
alumina crucibles or platinum crucibles, which eliminates the possibili-
ty that it is due to a sample-alumina crucible chemical reaction. 5. Conclusions
Moreover, the X-ray diffraction analysis confirms that within the
limit of detection of XRD there is no crystallization at 1040 °C endother- The correct interpretation of the DTA/DSC experiments spectra is
mic peak. possible as long as we are aware of the different physical phenomena
Fig. 5 shows that the endo-peak becomes sluggish when the YAS that occur during glass powder heating. Particularly, on studying glass
glass powder is passed through a propane–oxygen flame or when the powders crystallization, there is an effect on the crystallization peaks
powder is put for about 30 h in contact with distilled water. of moving toward lower temperatures, when compared with the results
It was shown for clays [7] that a possible reaction mechanism for obtained with glass slabs.
dehydroxilation can be proposed. The proposed mechanism needs a Moreover when run on some glass powders, YAS glasses for exam-
proton delocalization at a specific hydroxyl site. For H2O to form, the ple, DTA traces exhibit an endothermic peak before (or overlapped
delocalized proton must migrate to a second hydroxyl site. If however,
Fig. 7. DTA of (a) NCS glass slab, and (b) NCS glass powder showing the absence of an en- Fig. 9. Kissinger plot of YAS temperatures of maximum crystallization rates at different
dothermic peak before crystallization. heating rates (data from Fig. 7).
16 D.C. Lago, M.O. Prado / Journal of Non-Crystalline Solids 381 (2013) 12–16
Acknowledgements
This work has been financed by the project PICT 2008-39 ANPCyT
Argentina.
Fig. 10. Thermogravimetry of YAS samples with different surface quality.
References
with) the crystallization peak. This endothermic peak appears when we
[1] J.M. Barandiaran, J. Colmenero, J. Non-Cryst. Solids 46 (1981) 277–287.
work with glass powders indicating that it is due to an endothermic sur- [2] A. Cabral, A. Cardoso, E.D. Zanotto, J. Non-Cryst. Solids 320 (2003) 1–8.
face process. [3] R. Müller, J. Therm. Anal. 35 (1989) 823–835.
This endothermic peak has height, width and temperature span that [4] C. Lara, M.J. Pascual, A. Durán, J. Non-Cryst. Solids 348 (2004) 149–155.
[5] Sung-Bum Sohn, Se-Young Choi, J. Non-Cryst. Solids 282 (2001) 221–227.
depends upon the glass surface chemistry. [6] Manuela Reben, J. Non-Cryst. Solids 357 (2011) 2653–2657.
As the process is endothermic, it is not due to strain relaxation, crys- [7] R.L. Frost, A.M. Vasallo, Clay Clay Miner. 44 (5) (1996) 635–651.
tallization (we checked experimentally this point with XRD just at the [8] R.E. Grim, W.F. Bradley, Am. Mineral. 33 (1–2) (1948) 50–59.
[9] J.E. White, D.E. Day, Key Eng. Mater. 181 (1994) 94–95.
maximum of the endothermic peak) or available energy due to the [10] Xiaojie J. Xu, Chandra S. Ray, Delbert E. Day, J. Am. Ceram. Soc. 74 (5) (1991)
sintering decreasing surface. Sintering can show a change in the DTA 909–914.
baseline but not an endothermic peak. [11] Najim Sadiki, Jean Pierre Coutures, Catherine Fillet, Jean Luc Dussossoy, J. Nucl.
Mater. 348 (2006) 70–78.
Since YAS glass components have a unique oxidation state, we do
[12] Quan Wan, Christopher Ramsey, George Baran, J. Therm. Anal. Calorim. 99–1 (2010)
not expect an endochemical (or even exothermic) reaction to take 237–243.
place during the heating process. [13] L.T. Zhuravlev, Colloids Surf. A Physicochem. Eng. Asp. 173 (2000) 1–38.
Radiation Physics and Chemistry 181 (2021) 109334
A R T I C L E I N F O A B S T R A C T
Keywords: We measured the metal-protein compounds X-ray fluorescence by the laser-induced X-ray source with an
Laser plasma aluminum tape target. The laser-induced X-ray generated by the high repetition rate (1 kHz) femtosecond (40 fs)
High repetition rate Ti: Sapphire laser system. To characterize a laser-induced Kα and bremsstrahlung X-ray source, we changed the
Kα
laser intensity and pulse focused position. We measured the X-ray fluorescence signal with the copper plate,
Laser-induced X-ray
X-ray fluorescence
selenium powder, lyophilized hemoglobin, and lyophilized seleno-methionine, which was detected by X-ray
Metal-protein spectrometer. With a few micrograms of samples, the X-ray fluorescence can be measured with high accuracy and
high stability.
* Corresponding author.
E-mail address: dkko@gist.ac.kr (D.-K. Ko).
https://doi.org/10.1016/j.radphyschem.2020.109334
Received 20 July 2020; Received in revised form 14 December 2020; Accepted 25 December 2020
Available online 28 December 2020
0969-806X/© 2020 Elsevier Ltd. All rights reserved.
J. Kang and D.-K. Ko Radiation Physics and Chemistry 181 (2021) 109334
its unique characteristics. Various research groups are working on 2. Experimental setup and methods
improving the conversion efficiency and the photon flux of the
laser-induced Kα X-ray by changing laser pulse parameters such as the Fig. 1 shows the experimental setup for laser-induced X-ray gener
incident angle and contrast ratio. (Goryaev et al., 2014; Azamoum et al., ation and fluorescence measurement. This experiment was performed
2018). with a double-amplified CPA femtosecond Ti:sapphire laser source (36 fs
We have optimized the characteristic emission of laser-induced X-ray pulses, 805 nm wavelength, 1 kHz repetition rate, 10 W power). All
and also optimized the bremsstrahlung emission required for x-ray experiments were performed in a high vacuum environment below 10− 4
fluorescence measurements. Various types of metal targets such as wire, Torr with a vacuum chamber. Various types of solid targets such as wire,
plate, tape, rotating wheel, and liquid metal are used to generate laser- tape, wheel, and plate were used, depending on the characteristics of the
induced X-rays. Tape target is useful for experiments using high repe laser used for X-ray generation. We used a 1 kHz high repetition rate
tition rate laser pulses because it allows continuous experiments, but its laser, so we used a tape target for continuous X-ray generation. The
stability is relatively low due to laser pulse pressure (Zhang et al., 2014). focused high energy pulse creates a micrometer size crater on the target.
To maintain the high stability and accuracy of the laser-induced X-ray Since generation efficiency is lowered due to the crater, the laser must be
fluorescence measurement, we adjusted the thickness of the tape target irradiated with pulse-by-pulse in the non-damaged area (Goryaev et al.,
and manufactured a new tape rotating holder. In this study, we devel 2014). We rotated the tape target at a constant rate of 4 cm/s consid
oped the laser-induced X-ray source for XRF by the combination and ering the crater size and repetition rate. The thickness of the tape target
control of characteristic emission and bremsstrahlung emission and was 40 μm, which was designed considering the metal strength and the
measured the fluorescence of metallic substances, metal compounds, fluctuation due to the focusing of the laser pulse. The disadvantages of
and metal-protein materials (seleno-peptides) with high stability and using a tape target are frequent fluctuations and low stability (Zhang
high accuracy. et al., 2014). Therefore, we designed and manufactured the tape rotation
mount to improve the stability of X-ray generation. The Al target has low
threshold energy and can be generated with high efficiency, even with
Fig. 1. Top view of the experimental setup for laser-induced X-ray generation with a femtosecond laser. The experimental setup consists of two parts, (a) opti
mization of laser-induced X-ray source by controlling the laser intensity, focal position, and pulse width and (b) measurement of X-ray fluorescence by the laser-
induced source. All experiments were performed in a vacuum chamber of 10-4 Torr or less
2
J. Kang and D.-K. Ko Radiation Physics and Chemistry 181 (2021) 109334
relatively low laser intensity. Since this property is suitable for compact The laser-induced X-ray fluorescence of copper plate (Cu), human he
table-top X-ray sources, we used an Al target. In addition, Al has the moglobin (Fe), selenium powder (Se), and seleno-peptide (Se) was
advantages of low cost, easy fabrication, and disposal. measured to research the X-ray fluorescence in various energy ranges.
The femtosecond laser pulses focused at an intensity of 1016 W/cm2 Metal compound samples were commercially available products (Sigma
or more by the off-axis parabolic mirror. The laser-induced X-ray gen Aldrich).
eration efficiency varies with the angle at which the focused laser irra
diates the aluminum tape target. To maximize efficiency, the X-ray was 3. Results and discussion
generated by the laser incident at an angle of 30◦ (Goryaev et al., 2014).
The femtosecond laser-induced X-ray sources have a flux of 107 photo 3.1. Al tape target laser-induced X-ray generation and characterization
n/s⋅srad in the Al Kα energy region and 107 photon/s⋅srad in the
Bremsstrahlung region. Characterizing a laser-induced Kα and brems To improve and characterize the flux of the laser-induced X-ray
strahlung X-ray source by changed the laser intensity, pulse focused source, we controlled a few physical parameters. Fig. 2(a) shows the
position, and pulse width (Fig. 1 a). The metal compound X-ray fluo spectrum of femtosecond laser-induced X-ray sources with an Al tape
rescence measurement is performed by the laser-induced X-ray source, target. This spectrum is the raw spectrum data without concerned the
which has a wide energy range, from 0.1 keV to 20 keV. The metal detector efficiency and any affected with the filter. It corresponds to the
compound sample was placed behind the laser-induced X-ray source, laser power of 1.5 W, the focused beam diameter was 22 μm, and the
and the metal fluorescence was measured by using an Amptek X-ray peak power of 1.97ⅹ1016 W/cm2. The spectrum shows the Al Kα lines at
detector (XR-100CR). To reduce noise, the Pb plate was wrapped around 1.486 keV and the pile-up peak at 2.972 keV. In the energy region of 3.5
the X-ray detector to block the laser-induced X-ray source (main beam) keV or more, the bremsstrahlung continuum emission can be confirmed.
and optimized the fluorescence signal by adjusting the position and The maximum photon flux of the Kα X-ray source is 4.84ⅹ107 photon/
velocity of the Al tape target (Fig. 1 b). In order to prevent the noise s∙srad and the photon flux of bremsstrahlung emission is 1.34ⅹ107
caused by the reflected laser pulse and the electrons from the photon/s∙srad which is 3.6 times less than the Kα emission. The pile-up
laser-produced plasma, the X-ray signal was optimized by adjusting the peak at 2.972 keV and 4.5 keV were not included in the line emission
threshold and the detecting range on the detector program. Using the flux calculation.
magnets to remove electron noise is also a good option, but we chose the The metal inner ionization energy threshold depends on the Kα value
threshold condition control option due to the limitation of the experi of the metal target and laser irradiance value. The ionization threshold
mental space. To prevent debris, a debris-proof box and acrylic tape condition is given by (Gibbon and Förster, 1996).
products were manufactured and installed near the pulse focusing site.
Fig. 2. (a) Spectrum of laser-induced X-ray with aluminum. It corresponds to the laser intensity of 1016 W/cm2 and an angle of incidence of 30◦ . (b) Dependence of
the aluminum Kα flux on the femtosecond laser power. (c) Dependence of the aluminum Kα (1.486 keV) and bremsstrahlung emission (>3.5 keV) photon numbers on
the femtosecond laser power. (d) The conversion efficiency of laser-induced aluminum Kα change by the laser power.
3
J. Kang and D.-K. Ko Radiation Physics and Chemistry 181 (2021) 109334
v
√̅̅̅̅̅̅̅̅̅̅̅
2U 1
(
I λ2
)1/2 is 1.18ⅹ1016 W/cm2, and the conversion efficiency is 7.86ⅹ10− 6. Two
= = 0.84 / (1) fitting lines correspond to a lognormal fitting of the experimental points,
c me c 1018 W cm2 μm2
which is a guide for the eye and follows the theoretical model. The
where U, I and λ are the Kα characteristic emission energy of the target corresponding conversion efficiency value is about 80% compared to the
material, the laser peak power and the wavelength of the laser pulse. As theoretical calculation maximum value of 10− 5.
the atomic number Z of the target material increased, Kα emission en Fig. 3 shows the dependence of aluminum Kα flux on the laser pulse
ergy increased, and a higher ionization threshold peak power is width, which changed from 40 fs to 120 fs. Since the laser pulse width
required. According to the calculation, the ionization threshold intensity affects the Kα flux, we used the shortest pulse width (36 fs) for all ex
of aluminum, copper and molybdenum are 1.28ⅹ1016 W/cm2, periments. Fig. 4 shows the Al Kα (1.478 keV) and bremsstrahlung
6.92ⅹ1016 W/cm2 and 1.5ⅹ1017 W/cm2, respectively. The generated emission (>3.5 keV) on the change of focal position in the micrometer
electrons are accelerated by the strong laser field and it collides to the scale. We set the focal position zero at the maximum K alpha emission.
metal. Characteristic emission occurs due to the collision of the accel Since the scale of the laser-produced plasma is tens of μm, Al Kα and
erated electrons and metal inner shell. The laser-induced X-ray source is bremsstrahlung emission were measured 30 times at 10 μm intervals
affected by the hot electron temperature, and the hot electron temper before and after the focal point. As shown in Fig. 4, the aluminum Kα and
ature is proportional to (Iλ2)α, where α is between 1/3 and 1 depends on bremsstrahlung emission decreased as the position moved away from
the numerical calculation assumptions of the laser absorption. (Four the focal point. As the position of the tape target moved away from the
maux and Kieffer, 2016). focal point, the radius of the focused laser increased and the peak power
In this study, since the aluminum tape target with relatively low Kα decreased. Since the peak power change affected the hot electrons
energy was used, the laser-induced metal inner ionization was suffi accelerated by the laser field, the Al Kα emission decreased as the Al tape
ciently performed, even at a peak power of 1016 W/cm2. Fig. 2(b) is a target moved away from the focal point. The position of the maximum
peak signal of aluminum Kα according to the change of the laser power. bremsstrahlung emission is shifted by 10–20 μm relative to the Al Kα,
The Al Kα signal increased with increasing laser power and then satu and the bremsstrahlung emission also decreased as the tape target
rated from 0.9 W or higher. At the maximum value, the FWHM of the Al moved away from the focal position. The increase of the bulk electron
Kα peak is 0.117 keV, and the flux is 5.42ⅹ107 photon/s∙srad. The temperature affects the increase of the bremsstrahlung emission, but it
dependence of the aluminum Kα (1.486 keV) and the bremsstrahlung has the opposite effect on the Kα characteristic emission. According to
emission (>3.5 keV) photon numbers per second on the femtosecond previous research, the Ti Kα decreased with the increasing of the bulk
laser power is shown in Fig. 2(c). As the laser power increased, the Kα electron temperature (Cristoforetti et al., 2013). When the bulk electron
photon increased linearly, and then it saturated from 0.8 W or higher. temperature increased, the Al Kα emission deteriorated, which causes
The metal inner ionization threshold power value is 0.7 W, and the the Al Kα maximum position to shift by 10–20 μm. At the X-ray fluo
corresponding experimental result satisfies the ionization threshold rescence, the fluorescence signal was optimized at this shifted position.
condition. Two fitting lines correspond to a Boltzmann fit and the bar is a We could obtain the high-efficiency, compact laser-induced X-ray source
standard error with 10 measurement trials. The intensity of the brems with 80% efficiency of the theoretical calculations and studied the
strahlung emission increased dramatically in the Al Kα saturation re characteristics on the laser intensity, pulse width and focused position.
gion, after which it also become saturated. The intensity of the
bremsstrahlung emission generated in the laser-induced plasma varies 3.2. Femtosecond laser-induced X-ray fluorescence measurement of
with the bulk electron temperature. It is described as follows (Alkhi metal-protein compounds
mova et al., 2018).
/( ) To precisely calibrate the X-ray fluorescence signal, we measured the
Ibrs ̃Aexp− 12.4/λTe λ2 (2) laser-induced X-ray fluorescence with selenium powder and the copper
plate. As shown in Fig. 4, we set the focal position zero at the maximum
where Ibrs, Te, λ and A are the intensity of the bremsstrahlung emission, K alpha emission and the maximum Bremsstrahlung was obtained at the
the bulk electron temperature (keV), the wavelength (Å) and a constant. position shifted by 10–20 μm from zero position. X-ray fluorescence
As the bulk electron temperature increased, the intensity of the Brems measurement was performed at the shifted position. In this experiment,
strahlung emission increased rapidly and become saturated and the X-ray fluorescence produced directly by laser-produced plasma
converged. The bulk electron temperature can be adjusted by the laser electrons was not considered. Application experiments are possible
intensity and focused pulse alignment. (Alkhimova et al., 2018) We using the penetration depth of the electrons generated in laser produced
optimized Te and Ibrs, by controlling the corresponding parameters. plasma (ValleBrozas et al., 1007). Further research can be performed
Fig. 2(d) shows the conversion efficiency of Al Kα versus the laser using the electrons generated by the laser produced plasma.
power. The correlation between laser power and conversion efficiency Fig. 5(a) shows the X-ray fluorescence of the selenium powder and
has already been reported in previous research when using the femto copper plate. We confirmed the selenium Kα and Kβ signal at 11.22 keV
second Ti sapphire laser pulse and aluminum target (Nakano et al., and 12.49 keV, respectively. The ratio of the Kβ/Kα is 0.159. It shows the
2004). The optimum laser intensity for conversion efficiency is driven as copper Kα and Kβ signal at 8.04 keV and 8.90 keV, respectively. The
(Nakano et al., 2004). ratio of Kβ/Kα is 0.097. The stability of laser-induced X-ray fluorescence
2 with selenium powder is shown in Fig. 5. Fig. 5(b) shows the photon
(Z/13)
(opt)
I16 ≈ 2/3
(3) number per second at the selenium Kα peak. The average is 37.75, the
(L/λ) sin2 θ + 0.1ηa standard deviation is 1.29, and the standard error is 0.41. Fig. 5(b) is the
energy value graph of the X-ray fluorescence of the selenium Kα peak.
where Z, L, θ and ηa are the atomic number of target, the density scale
The average energy, which was measured 10 times, was 11.218 keV,
length, the incident angle of the laser and the absorption coefficient,
with a standard deviation of 0.002 and a standard error of 0.0007. This
respectively. For example, the atomic number of aluminum target Z is
opt data is within the error range of 0.037% compared to the theoretical
13, the optimal intensity is I16 ≈ 1.5 and, for a copper target (Z = 29),
value of Selenium Kα 1 energy of 11.2224 keV.
opt
I16 ≈ 8. (Nakano et al., 2004) According to the theoretical calculation, Factors that affect the stability and accuracy of laser-induced X-ray
when the laser intensity is 1.5ⅹ1016 W/cm2, it has a maximum conver fluorescence include external pressure, due to the laser pulse focusing. It
sion efficiency value. When the laser power is 1.15 W, the laser peak causes the change of the flux and peak signal of x-ray fluorescence,
power is 1.5ⅹ1016 W/cm2, and it has maximum conversion efficiency. As resulting in lower stability and lower accuracy. Therefore, we optimized
shown in Fig. 2(d), when the laser power is 0.9 W, the laser peak power the thickness of the Al tape as 40 μm for the experiment and
4
J. Kang and D.-K. Ko Radiation Physics and Chemistry 181 (2021) 109334
5
J. Kang and D.-K. Ko Radiation Physics and Chemistry 181 (2021) 109334
Fig. 5. Laser-induced X-ray fluorescence spectrum of the selenium powder and the copper plate(a). Laser-induced X-ray fluorescence stability with selenium
powder(b)
Fig. 6. Laser-induced X-ray fluorescence spectrum of the lyophilized powder of human hemoglobin (a) and the laser-induced X-ray fluorescence spectrum of the
lyophilized powder of seleno-methionine(b)
measured with a laser-induced X-ray source. By using the wide brems Declaration of competing interest
strahlung energy region of the laser-induced X-ray source, it is possible
to analyze the X-ray fluorescence of various metal compounds that have The authors declare that they have no known competing financial
specific characteristic X-ray energy. To optimize the laser-induced X-ray interests or personal relationships that could have appeared to influence
source intensity, we changed the pulse width, laser power, and pulse the work reported in this paper.
focusing position. Al Kα and bremsstrahlung emission trends were
measured according to the change in laser intensity to find the optimum Acknowledgments
conversion efficiency condition, and the difference of 10–20 μm spacing
between Al Kα maximum position and bremsstrahlung maximum posi This research was supported by the National Research Foundation of
tion was confirmed. We measured the X-ray fluorescence with copper Korea (NRF) (Grant Nos. NRF-2015R1A5A1009962 and NRF-
plate, selenium powder, lyophilized Hemoglobin, and lyophilized 2016K1A3A1A61005950).
seleno-methionine. As the content of metal atoms in the metal com
pound decreases, the photon number per second of X-ray fluorescence References
decreases, and the signal-to-noise ratio and accuracy of the Kα/Kβ ratio
also decrease. It was confirmed that the X-ray fluorescence measurement Alkhimova, M.A., Faenov, A. Ya, Pikuz, T.A., Skobelev, I. Yu, Pikuz, S.A., Nishiuchi, M.,
Sakaki, H., Pirozhkov, A.S., Sagisaka, S., Dover, N.P., Kondo, Ko, Ogura, K.,
was also possible with a few milligrams of metal compounds. The ac Fukuda, Y., Kiriyama, H., Esirkepov, T., Bulanov, S.V., Andreev, A., Kando, M.,
curate characteristic emission X-ray energy signal was measured with a Zhidkov, A., Nishitani, K., Miyahara, T., Watanabe, Y., Kodama, R., Kondo, K., 2018.
small amount of metal atoms. We think this technique has a potential to X-ray emission from stainless steel foils irradiated by femtosecond petawatt laser
pulses. IOP Conf. Series : J. Phys. Conf. 946 https://doi.org/10.1088/1742-6596/
be used as a step for high-throughput drug screening. 946/1/012018, 012-018.
Azamoum, Y., Tcheremiskine, V., Clady, R., Ferré, A., Charmasson, L., Utéza, O.,
Author statements Sentis, M., 2018. Impact of the pulse contrast ratio on molybdenum Kα generation by
ultrahigh intensity femtosecond laser solid interaction. Sci. Rep. 8, 4119. https://
doi.org/10.1038/s41598-018-22487-3.
Jungu Kang: Methodology, Investigation, Data Curation, Formal M. Barberio, S. Veltri, M. Sciscio, P. Antici, Laser-Accelerated proton beams as
analysis, Writing-Original Draft. diagnostics for cultural heritage, Sci. Rep. 7. 40415. https://doi.org/10.1038/srep
40415 .
Do-Kyeong Ko: Supervision, Conceptualization, Methodology, Proj
ect administration, Funding acquisition, Writing-Review & Editing.
6
J. Kang and D.-K. Ko Radiation Physics and Chemistry 181 (2021) 109334
Cristoforetti, G., Anania, M.P., Faenov, A. Ya, Giulietti, A., Giulietti, D., Hansen, S.B., Li, Runze, Ashour, Omar A., Chen, Jie, Elsayed-Ali, H.E., Rentzepis, Peter M., 2017b.
Koester, P., Labate, L., Levato, T., Pikuz, T.A., Gizzi, L.A., 2013. Spatially resolved Femtosecond laser induced structural dynamics and melting of Cu (111) single
analysis of Kα x-ray emission from plasmas induced by a femtosecond weakly crystal. An ultrafast time-resolved x-ray diffraction study. J. Appl. Phys. 121 https://
relativistic laser pulse at various polarizations. Phys. Rev. 87 https://doi.org/ doi.org/10.1063/1.4975198, 055-102.
10.1103/physreve.87.023103, 023-103. Morak, A., Kämpfer, T., Uschmann, I., Lübcke, A., Förster, E., Sauerbrey, R., 2006.
Feurer, T., Morak, A., Uschmann, I., Ziener, Ch, Schwoerer, H., Förster, E., Sauerbrey, R., Acoustic phonons in InSb probed by time-resolved X-ray diffraction. Phys. Status
2001. An incoherent sub-picosecond X-ray source for time-resolved X-ray-diffraction Solidi B 243 (No. 12), 2728–2744. https://doi.org/10.1002/pssb.200542387.
experiments. Appl. Phys. B 72, 15–20. https://doi.org/10.1007/s003400000456. Nakano, H., Andreev, A.A., Limpouch, J., 2004. Femtosecond X-ray line emission from
Fourmaux, S., Kieffer, J.C., 2016. Laser-based Kα X-ray emission characterization using a multilayer targets irradiated by short laser pulses. Appl. Phys. B 79, 469–476.
high contrast ratio and high-power laser system. Appl. Phys. B 122, 162. https://doi. https://doi.org/10.1007/s00340-004-1582-7.
org/10.1007/s00340-016-6442-8. Otendal, M., Tuohimaa, T., Hertz, H.M., 2007. Stability and debris in high-brightness
Gibbon, P., Förster, E., 1996. Short-pulse laser–plasma interactions. Plasma Phys. Contr. liquid metal-jet-anode microfocus x-ray sources. J. Appl. Phys. 101 https://doi.org/
Fusion 38, 769–793. https://doi.org/10.1088/0741-3335/38/6/001. 10.1063/1.2423229, 026102.
Goryaev, A.A., Andreev, A.A., Platonov, K. Yu, Sedov, M.V., Zhavoronkov, N.I., 2014. X- Sokolowski-Tinten, Klaus, von der Linde, Dietrich, 2004. Ultrafast phase transitions and
ray generation under oblique incidence of femtosecond laser pulses with kHz pulse lattice dynamics probed using laser-produced x-ray pulses. J. Phys. Condens. Matter
repetition rate on a metal foil with irregular relief. LASERS AND THEIR 16. https://doi.org/10.1088/0953-8984/16/49/r04.
APPLICATION 117 (6), 990–994. https://doi.org/10.1134/s0030400x14120091. Tillman, C., Persson, A., Wahlstrim, C.-G., Svanberg, S., Herrlin, K., 1995. Imaging using
H€ofer, S., K€ampfer, T., F€orster, E., St€ohlker, T., Uschmann, I., 2016. Communication: hard X-rays from a laser-produced plasma. Appl. Phys. B 61, 333–338. https://doi.
the formation of rarefaction waves in semiconductors after ultrashort excitation org/10.1007/bf01081532.
probed by grazing incidence ultrafast time-resolved x-ray diffraction. Structural F.Valle Brozas, A. Crego, L. Roso, A. Peralta Conde, Laser-based X-ray and electron
dynamics 3. https://doi.org/10.1063/1.4963011, 051-101. source for X-ray fluorescence studies, Appl. Phys. B 122, 220. https://doi.org/10.100
Iqbal, M., Urrehman, Z., Im, H., Son, J.G., Seo, O., Stiel, H., Nickles, P.V., Noh, D.Y., 7/s00340-016-6494-9 .
Janulewiez, K.A., 2014. Performance improvement of a Kerformance improvement Yu, J., Jiang, Z., Kieffer, J.C., Krol, A., 1999. Hard x-ray emission in high intensity
of a Kα source by high-resolution thin-layer-graphite spectrometer and a femtosecond laser–target interaction. Phys. Plasmas 6, 13–18. https://doi.org/
polycapillary lens. Appl. Phys. B 116, 305–311. https://doi.org/10.1007/s00340- 10.1063/1.873372.
013-5691-z. Zamponi, F., Ansari, Z., Korff Schmising, C.v., Rothhardt, P., Zhavoronkov, N.,
Larsson, J., Allen, A., Bucksbaum, P.H., Falcone, R.W., Lindenberg, A., Naylor, G., Woerner, M., Elsaesser, T., Bargheer, M., Trobitzsch-Ryll, T., Haschke, M., 2009.
Missalla, T., reis, D.A., Scheidt, K., Sjögren, A., Sondhauss, P., Wulff, M., Wark, J.S., Femtosecond hard X-ray plasma sources with a kilohertz repetition rate. Appl. Phys.
2002. Picosecond X-ray diffraction studies of laser-excited acoustic phonons in InSb. A 96, 51–58. https://doi.org/10.1007/s00339-009-5171-9.
Appl. Phys. A 75, 467–478. https://doi.org/10.1007/s003390201421. Zhang, B.B., Sun, S.S., Sun, D.R., Tao, Y., 2014. Note: a novel normalization scheme for
Li, Minghua, Huang, Kai, Chen, Liming, Yan, Wenchao, Tao, Mengze, Zhao, Jiarui, laser-based plasma x-ray sources. Rev. Sci. Instrum. 85 https://doi.org/10.1063/
Maa, Yong, Li, Yifei, Zhang, Jie, 2017a. Laser-driven powerful kHz hard x-ray 1.4896252, 096-110.
source. Radiat. Phys. Chem. 137, 78–82. https://doi.org/10.1016/j.
radphyschem.2016.01.042.