Thermoanalytical studies on acid-treated rice husk and production of some silicon based ceramics from carbonised rice husk

Thermoanalytical studies on acid-treated rice husk and production of some silicon based ceramics from carbonised rice husk

G Model ARTICLE IN PRESS JASCER 170 1–6 Journal of Asian Ceramic Societies xxx (2015) xxx–xxx H O STED BY Contents lists available at ScienceDire...

1MB Sizes 2 Downloads 19 Views

G Model

ARTICLE IN PRESS

JASCER 170 1–6

Journal of Asian Ceramic Societies xxx (2015) xxx–xxx

H O STED BY

Contents lists available at ScienceDirect

Journal of Asian Ceramic Societies journal homepage: www.elsevier.com/locate/jascer

Thermoanalytical studies on acid-treated rice husk and production of some silicon based ceramics from carbonised rice husk

1

2

3 4 5

Q1

Muhammad Ali a,∗ , M. Ahmad Tindyala b a b

Institute of Advanced Materials, Bahauddin Zakariya University, Multan 60800, Pakistan Department of Metallurgical & Materials Engineering, University of Engineering & Technology, Lahore 54890, Pakistan

6

7 21

a r t i c l e

i n f o

a b s t r a c t

8 9 10 11 12 13

Article history: Received 3 April 2015 Received in revised form 21 May 2015 Accepted 13 June 2015 Available online xxx

14

20

Keywords: Acid-treated rice husk Non-isothermal TG Thermal degradation Energy of activation Flynn and Wall expression

22

1. Introduction

15 16 17 18 19

Acid pre-treatment of rice husk activated amide groups, thereby rendering the material to earlier decomposition as compared to raw rice husk when subjected to thermal analysis under non-isothermal conditions. The role of heating rate on thermal characteristics of acid-treated rice husk and energy of activation of thermal degradation were discussed in this study. The latter was dealt using Flynn and Wall equation and studied over a range of thermal destruction. Comparison of TG curves, at 5 and 10 ◦ C/min revealed that degradation rate was faster in case of faster heating rate up to about 50% weight loss and then retarded resulting in lower final degradation temperature. It can be concluded from energy calculations that, as degradation proceeded, overall energy of activation increased up to 50% weight reduction. TG analysis was followed by high temperature pyrolysis of carbonised rice husk under argon and nitrogen atmospheres to obtain carbide, nitride, and oxynitride of silicon. Formation of different products was confirmed by FT-IR and XRD, and yield was calculated through HF dissolution method. Argon pyrolysis was proved efficient, as compared to nitrogen and closed tube pyrolysis, in yielding 30% by weight of silicon carbide from carbonised rice husk. © 2015 The Ceramic Society of Japan and the Korean Ceramic Society. Production and hosting by Elsevier B.V. All rights reserved.

Silicon carbide and silicon nitride have several high temperature applications, but their traditional production methods are 24 energy intensive. An inexpensive method that uses low cost agro25 industrial by-product is the pyrolysis of rice husk (RH), first carried 26 out by Lee and Cutler [1]. RH, as separated from rice grain, contains 27 28Q2 75–85% of organic mass and remaining 15–25% ash [2]. Organic mass, in turn, comprises lignin, cellulose, hemicellulose, etc. RH-ash 29 is obtained by complete combustion and can contain up to 95% of 30 silica [3]. Morphological studies on RH reveal that micron size silica 31 particles are distributed in its cellulosic part [4,5]. When these par32 ticles are made to react with carbon in the biomass part of RH under 33 specific experimental conditions, silicon carbide can result. More34 over, besides silicon carbide, modifications in process mechanism 35 lead to formation of some other industrially useful products, viz. sil36 icon nitride, silicon oxynitride, ultra-fine silica, and solar-cell grade 37 23

∗ Corresponding author. Tel.: +92 332 410 9686. E-mail address: [email protected] (M. Ali). Peer review under responsibility of The Ceramic Society of Japan and the Korean Ceramic Society.

silicon [6–12]. Instead of carrying out synthesis process directly, RH may or may not be subjected to pre-treatments depending upon the desired products. Hence the starting material can be raw RH [4,13–15], burnt RH [13,16–18], or a mixture of RH silica and carbon black [19]. During the synthesis through pyrolysis, heating sequence may comprise single step, two steps, or more. Single step process (direct pyrolysis at 1400 ◦ C) resulted in the highest yield as compared to other heating routes [5,20]. In two step method, the husks are heated at about 900 ◦ C for appropriate soaking period and then pyrolysed under vacuum at about 1500 ◦ C. Moustafa et al. [21] discussed the chemistry of formation of silicon carbide, and Padhi and Patnaik [6] elucidated chemical reactions of silicon nitride and silicon oxynitride formation as well. RH is pyrolysed under inert or reducing media. As a general rule, argon or nitrogen atmosphere is employed for synthesis of SiC [4], NH3 atmosphere or 95% N2 –5% H2 atmosphere for Si3 N4 [4,22]. Influence of several catalysts on product yield was also focused in various reports. Martinez et al. [23] showed that use of FeCl2 ·4H2 O or CoCl2 ·6H2 O as catalyst results in formation of SiC in shorter time than without catalyst. Same effect of CoCl2 was also inferred by Krishnarao [18]. A study on influence of Na2 SiO3 and its concentration on yield of SiC production concluded that comparable yield can be obtained in one third times by using catalyst

http://dx.doi.org/10.1016/j.jascer.2015.06.003 2187-0764 © 2015 The Ceramic Society of Japan and the Korean Ceramic Society. Production and hosting by Elsevier B.V. All rights reserved.

Please cite this article in press as: M. Ali, M.A. Tindyala, Thermoanalytical studies on acid-treated rice husk and production of some silicon based ceramics from carbonised rice husk, J. Asian Ceram. Soc. (2015), http://dx.doi.org/10.1016/j.jascer.2015.06.003

38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

G Model JASCER 170 1–6

M. Ali, M.A. Tindyala / Journal of Asian Ceramic Societies xxx (2015) xxx–xxx

2 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78

79

ARTICLE IN PRESS

[5]. Some catalysts have also been used to control morphology of the products, such as V2 O5 fosters formation of ␤-Si3 N4 [24], and Fe2 O3 solution treatment increases proportion of both ␤-SiC and its whiskers [13]. Treating RH, prior to thermal exposure, with a mineral acid (H2 SO4 , HNO3 or HCl) solution or with boron compound (as H3 BO3 ) proved beneficial to increase yield as well as purity of SiC as compared to that produced from raw RH [25,26]. Using RH for synthesising various products requires sound knowledge of its thermal characteristics and pyrolysis kinetics. Comprehension of the behaviour of RH against thermal exposure helps in designing proper equipment features and proposing suitable process parameters as well. A common practice to study thermal characteristics of RH is through thermogravimetric analysis (TGA) in which rate of mass change is measured either as a function of time (as in isothermal TG) or as a function of temperature (as in non-isothermal TG). Energy of activation, using TG data, can be computed using Flynn and Wall expression [3]: d(log ˇ) −R Ea = · 0.457 d(1/T )

106

where Ea represents the energy of activation, R universal gas constant, ˇ heating rate and T absolute temperature. Product of the constant and slope of the linear relationship between log ˇ and T−1 gives the value of the activation energy. This method eliminates the calculations of pre-exponential or frequency factor (k) as required in Arrhenius equation but, as it involves logarithm of heating rate, it is no more applicable to isothermal TGA. To probe kinetics of pyrolysis and evaluate energy of activation, RH can be subjected to TG either under isothermal or non-isothermal conditions, the latter being more common as it makes possible to explore kinetics over a continuous range of temperature. It is known that thermal decomposition of RH proceeds in three distinct stages [8,27] viz. loss of hygroscopic water (40–150 ◦ C), volatilisation of organic matter (215–350 ◦ C) and, at the end, combustion of carbon (350–690 ◦ C). To study the role of heating rate [26,28] and that of different atmospheres [29–33] on kinetic characteristics are common interests while analysing RH thermally. Williams and Besler [28] conducted thermal analysis of components of RH (lignin, cellulose, etc.) and compared thermogrammes of components with those of RH. Thermal analysis of RH (and other materials) is usually carried out under inert or reducing atmospheres, and sometimes under oxidative atmosphere [6,7,34]. Purging the inert gas ensures that the sample, while undergoing decomposition, responds to temperature only. However, reactive gas, as CO2 [35], can be employed for some purpose; e.g. O2 is purged to get C% in material (by monitoring the quantity of CO2 evolved), and CO2 atmosphere for TG of RH can gasify char above 700 ◦ C.

107

2. Materials and methods

80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105

108 109 110 111 112 113 114 115 116 117 118 119 120 121

Ample quantity of RH was procured from a local rice industry. This raw RH was thoroughly rinsed with distilled water to remove any residual rice grains or soil particles. A part of water-treated RH was soaked in 5–6 N sulphuric acid solution for one and a half hour and then again washed with distilled water. Acid-treated and untreated RH were dried, pulverised to a size −100 mesh through ASTM standard sieving and stored in a drying oven at 100 ◦ C till further usage. Acid-treatment effectively removes oily substance (if any) and also oxidises carbohydrates in the husk, resulting in blackened surface. Thermogravimetric analysis of acid-treated RH was carried out using LINSIES PT1600 thermal analyser. High purity nitrogen gas was purged at a rate of 0.2 L/min to conduct TGA from ambient to 800 ◦ C at two different heating rates: 5 and 10 ◦ C/min.

Energy of activation of acid-treated RH thermal destruction was calculated, from 10% weight loss to 60% to study its trend with decomposition, by Flynn and Wall expression. Before high temperature pyrolysis, untreated RH was carbonised at 700 ◦ C for 2 h under nitrogen atmosphere. This carbonised material was subjected to high temperature pyrolysis in three different ways: under argon atmosphere, under nitrogen atmosphere and in the closed tube. A porcelain tube with 5 mm I.D., closed from one end with ceramic wool, was filled with sample and heated from ambient to 1350 ◦ C at a rate of 5 ◦ C/min by means of SentroTech tube furnace STT-1650. Argon and nitrogen gases were introduced directly into the tube at a flow rate of 0.2 L/min in respective runs. In case of pyrolysis in the closed tube, both ends were closed with high temperature resistant ceramic wool to stop any air ingress and heated to 1350 ◦ C through the same furnace. Soaking time provided at pyrolysis temperature was 2 h in all the three cases. Pyrolysed material was oxidised at 700 ◦ C for 2 h to eliminate unreacted carbon, and then analysed through FT-IR and XRD techniques. For infrared spectroscopy, 200–250 mg of KBr powder was mixed with 10–15 mg of sample and palletised under 40 MPa compaction load. The pallets were loaded in Jasco FT/IR 4100 spectrometer to generate absorption spectra for each sample. PANanalytical X’pert powder Diffractometer was employed to produce X-ray diffraction patterns over a scan range of 2 = 5–120◦ . Unreacted silica was removed through 40% HF solution treatment to quantify the products.

3. Results and discussion

122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147

148

Thermogrammes of acid-treated RH at two different heating rates are shown in Fig. 1. Initial downward trend in TG curves around 100 ◦ C depicts loss of hygroscopic water. Decomposition and volatilisation of organic mass started around 190 ◦ C and continuously proceeded to final degradation temperature. Similar results have been presented in other reports [7,26,35]. It is apparent from two thermogrammes that decomposition started slightly earlier under faster heating rate [3]. Furthermore, rate of thermal degradation was slightly faster under faster heating rate up to 50% weight reduction and then slowed down resulting in lower final degradation temperature. This confirms the fact that an increase in heating rate decreases Ea , thereby essentially increasing the rate of thermal degradation or of carbonisation [3,26,28]. Table 1 lists characteristic temperature values along with total weight reduced in the two cases. Liou [26] conducted TG on acid (HCl) leached RH at different heating rates. The initial and final degradation temperatures recorded therein, 152 ◦ C and 528 ◦ C (for ˇ = 5 ◦ C/min) and 163 ◦ C and 546 ◦ C (for ˇ = 10 ◦ C/min) respectively, are closer to those given in the table. Raw RH started decomposing at about 250 ◦ C [28], whereas acid leaching makes the husk to undergo degradation earlier at about 185 ◦ C, because acid-treatment activated amide groups in the material, such as NH2 and CN as represented by the absorption peaks at 1630–1600 cm−1 and 1265–1200 cm−1 respectively in Fig. 2. It can be deduced that this earlier decomposition will start chemical reaction at lower temperature, thereby contributing to the economy of the process.

Table 1 Important temperatures and total weight reduced (nomenclature: ˇ = heating rate, ◦ C/min; T1 = degradation start temperature, ◦ C; T2 = degradation end temperature, ◦ C; Tmax = temperature of maximum rate of degradation, ◦ C). ˇ

T1

T2

Tmax

wt.% reduced

5 10

190 185

560 530

369 344

72.8 70.5

Please cite this article in press as: M. Ali, M.A. Tindyala, Thermoanalytical studies on acid-treated rice husk and production of some silicon based ceramics from carbonised rice husk, J. Asian Ceram. Soc. (2015), http://dx.doi.org/10.1016/j.jascer.2015.06.003

149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175

G Model

ARTICLE IN PRESS

JASCER 170 1–6

M. Ali, M.A. Tindyala / Journal of Asian Ceramic Societies xxx (2015) xxx–xxx

3

Fig. 1. TG of acid-treated RH at 5 and 10 ◦ C/min.

Q3

Fig. 2. FT-IR of acid-treated RH.

176 177 178 179 180 181 182 183

Activation energy was calculated for each degradation interval of 10% from ˛ = 0.1 to ˛ = 0.6 (where ˛ = 0.1 means 10% degradation region, ˛ = 0.2 means 20% and so forth). This range was selected because after the combustion of more than 50% of the sample mass, proportion of residues becomes accountable. Values of heating rate and absolute temperature were converted as per requirement of the energy expression, shown in Table 2, to draw a graph between the two parameters [3]. Table 2 Relationship between log ˇ and 1/T. Log ˇ

0.699 1.000

1/T × 103 (K−1 ) ˛ = 0.1

˛ = 0.2

˛ = 0.3

˛ = 0.4

˛ = 0.5

˛ = 0.6

1.968 1.911

1.760 1.729

1.644 1.620

1.554 1.508

1.453 1.412

1.369 1.345

Graph between 1/T and log ˇ, shown in Fig. 3, presents six straight lines each corresponding to a certain degradation region (˛). Slope of each line was substituted in Flynn and Wall expression to obtain energy of activation for each weight fraction, as listed in Table 3.

Table 3 Linear expression and energy of activation for each degradation region. ˛

Equation of straight line

Ea (kJ/mol)

Ea (kJ/mol) [3]

0.1 0.2 0.3 0.4 0.5 0.6

Y = −0.189X + 2.10 Y = −0.071X + 1.80 Y = −0.080X + 1.70 Y = −0.122X + 1.63 Y = −0.128X + 1.54 Y = −0.075X + 1.42

3.436 1.291 1.454 2.218 2.327 1.363

1.299 1.299 1.294 1.414 1.450 –

Please cite this article in press as: M. Ali, M.A. Tindyala, Thermoanalytical studies on acid-treated rice husk and production of some silicon based ceramics from carbonised rice husk, J. Asian Ceram. Soc. (2015), http://dx.doi.org/10.1016/j.jascer.2015.06.003

184 185 186 187 188

G Model JASCER 170 1–6 4

ARTICLE IN PRESS M. Ali, M.A. Tindyala / Journal of Asian Ceramic Societies xxx (2015) xxx–xxx

Fig. 3. Graph between 1/T and log ˇ by Flynn and Wall expression.

189 190 191 192

It is evident from the table that, as the degradation proceeded, the value of activation energy increased up to ˛ = 0.5 [3], except its maiden value, the higher value of which was responsible for initiating thermal destruction. As the degradation exceeded 50%

weight loss, a decrease in Ea value came forth, thereby increasing the degradation rate. Formation of different products was confirmed by characterisation through FT-IR and XRD. The peaks at 1092.48 cm−1 , in Figs. 4–6, show Si O bonding, i.e. silica was present after each thermal treatment. The absorption peaks at 801.278 cm−1 , in Figs. 4 and 5, depict Si C bonding, i.e. silicon carbide was formed during pyrolysis in argon and nitrogen atmospheres [4,5]. No absorption peak matched clearly to the Si N O or Si N bonding type that may be due to the limited IR data available on inorganic compounds. X-ray diffraction patterns (with 2 on abscissa and counts on ordinate) are shown in Fig. 7(a)–(c) for products of Ar pyrolysis, N2 pyrolysis and pyrolysis in the closed tube respectively. Successful pyrolysis under argon atmosphere produced considerable quantity of silicon carbide as evident from the peaks at 2 = 35.7, 60.1, and 71.8, whereas the peak at 26.4 shows the presence of tridymite. Pyrolysis under nitrogen atmosphere produced silicon oxynitride rather than pure silicon carbide, as can be seen in Fig. 7(b), wherein peaks at 2 = 20.2, 33.2, and 37.2 are characteristic for

Fig. 4. FT-IR of Ar pyrolysis products.

Fig. 5. FT-IR of N2 pyrolysis products.

Please cite this article in press as: M. Ali, M.A. Tindyala, Thermoanalytical studies on acid-treated rice husk and production of some silicon based ceramics from carbonised rice husk, J. Asian Ceram. Soc. (2015), http://dx.doi.org/10.1016/j.jascer.2015.06.003

193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211

G Model JASCER 170 1–6

ARTICLE IN PRESS M. Ali, M.A. Tindyala / Journal of Asian Ceramic Societies xxx (2015) xxx–xxx

5

Fig. 6. FT-IR of closed tube pyrolysis products.

Fig. 7. (a) XRD of Ar pyrolysis products, (b) XRD of N2 pyrolysis products, and (c) XRD of products of pyrolysis in the closed tube ( Si2 N2 O, Si3 N4 ).

221

silicon oxynitride. XRD pattern of closed tube pyrolysis products depicts presence of silicon nitride at 2 = 64.8. For quantification, all the pyrolysed products were treated with HF solution to detach silica contents. In case of pyrolysis under argon atmosphere considerable yield of silicon carbide (27.3%) was achieved which is very close to that obtained by Janghorban and Tazesh [5] under same conditions of atmosphere and pyrolysis temperature. A mixture of carbide and oxynitride was obtained after pyrolysis under nitrogen atmosphere and pyrolysis in closed tube produced about 15% of silicon nitride.

222

4. Conclusions

212 213 214 215 216 217 218 219 220

223 224

Rice husk has silica contents which can be utilised to produce various useful silicon based materials, including silicon carbide,

tridymite, 䊐 cristobalite, ♦ SiC, 

oxynitride and nitride. Acid-treatment of RH converted carbohydrates into complex amino acids containing CN and NH2 bond groups, due to which, when exposed to controlled heating, the material undergoes decomposition earlier than raw husk does. It is more useful practice to calculate energy of activation over a range of temperature because its value does not remain constant at all stages of degradation. It was found that the first value of Ea was greater, but then abruptly decreased once the degradation started, and then gradually rose up to 50% weight loss. Ea value again decreased beyond 50% weight reduction, thereby increasing the rate of decomposition. The feasibility of producing some ceramic materials from RH was studied and pyrolysis in argon atmosphere was found capable of yielding substantial quantity of silicon carbide. Pyrolysis in the closed tube, although by no means equivalent to vacuum, produced

Please cite this article in press as: M. Ali, M.A. Tindyala, Thermoanalytical studies on acid-treated rice husk and production of some silicon based ceramics from carbonised rice husk, J. Asian Ceram. Soc. (2015), http://dx.doi.org/10.1016/j.jascer.2015.06.003

225 226 227 228 229 230 231 232 233 234 235 236 237 238 239

G Model JASCER 170 1–6

M. Ali, M.A. Tindyala / Journal of Asian Ceramic Societies xxx (2015) xxx–xxx

6 240 241 242 243

244

245 246 247 248 249 250 251 252 253 254 255 256 257 258 259

ARTICLE IN PRESS

about 15% of silicon nitride. RH should be processed on commercial scale to utilise its potential of producing various high-tech materials including those mentioned above and ultra-fine silica, poly-crystalline silicon, etc. References [1] G.J. Lee and I.B. Cutler, Am. Ceram. Soc. Bull., 54, 195–198 (1975). [2] A. Muthadi, R. Anitha and S. Kothandaraman, IE(I)Journal-CV, 88, 50–56 (2007). [3] H.J. Kim and Y.G. Eom, Mokchae Konghak (J. Korean Wood Sci. Technol.), 29, 59–67 (2001). [4] R.V. Krishnarao, Y.R. Mahajan and T.J. Kumar, J. Eur. Ceram. Soc., 18, 147–152 (1998). [5] K. Janghorban and H.R. Tazesh, Ceram. Int., 25, 7–12 (1999). [6] B.K. Padhi and C. Patnaik, Ceram. Int., 21, 213–220 (1995). [7] L.T. Vlaev, I.G. Markovska and L.A. Lyubchev, Therm. Acta, 406, 1–7 (2003). [8] A. Chakraverty, P. Mishra and H.D. Banerjee, Therm. Acta, 94, 267–275 (1985). [9] K.K. Larbi, Synthesis of High-Purity Silicon from Rice Husks (M.S. thesis), University of Toronto (2010). [10] E. Swatistang and M. Krochai, J. Met. Mater. Min., 19, 91–94 (2009). [11] H.D. Banerjee, S. Sen and H.N. Acharya, Mater. Sci. Eng., 52, 173–179 (1982). [12] D.N. Bose, P.A. Govinda and H.D. Banerjee, Sol. Energy Mater., 7, 319–321 (1982).

[13] K. Sujirote and P. Leangsuwan, J. Mater. Sci., 38, 4739–4744 (2003). [14] R.V. Krishnarao and M.M. Godkhindi, Ceram. Int., 18, 243–249 (1992). [15] R.V. Krishnarao, M.M. Godkhindi, M. Chakraborty and P.G. Mukunda, J. Am. Ceram. Soc., 74, 2869–2875 (1991). [16] R.V. Krishnarao and M.M. Godkhindi, J. Mater. Sci., 27, 1227–1230 (1992). [17] R.V. Krishnarao and M.M. Godkhindi, Ceram. Int., 18, 185–191 (1992). [18] R.V. Krishnarao, J. Eur. Ceram. Soc., 12, 395–401 (1993). [19] R.V. Krishnarao, J. Mater. Sci. Lett., 12, 1268–1271 (1993). [20] K. Janghorban and H.R. Tazesh, Proceedings of 2nd Pacific Rim Int. Conference on Advanced Materials and Processing, (1995). [21] S.F. Moustafa, M.B. Morsi and A.A. Din, Can. Metall. Q., 36, 355–358 (1997). [22] I.A. Rahman and F.R. Riley, J. Eur. Ceram. Soc., 5, 11–22 (1989). [23] V. Martinez and M.F. Valencia, Ceram. Int., 32, 891–897 (2006). [24] I.A. Rahman, Ceram. Int., 24, 293–297 (1998). [25] T. Minaru, K. Tadashi, K. Masafumi, US Patent 4591492. [26] T.-H. Liou, Carbon, 42, 785–794 (2004). [27] M.A. Hamad, J. Chem. Technol. Biotechnol., 31, 624–626 (1981). [28] P.T. Williams and S. Besler, Fuel, 75, 151–159 (1993). [29] K.G. Mansary and A.G. Ghaly, Energy Sources, 20, 653–663 (1998). [30] K.G. Mansary and A.G. Ghaly, Energy Sources, 21, 453–466 (1999). [31] K.G. Mansary and A.G. Ghaly, Energy Sources, 21, 773–774 (1999). [32] K.G. Mansary and A.G. Ghaly, Biomass Bioenergy, 17, 19–31 (1999). [33] K.G. Mansary and A.G. Ghaly, Bioresour. Technol., 65, 13–20 (1998). [34] I.G. Markovska and L.A. Lyubchev, J. Therm. Anal. Calorim., 89, 809–814 (2007). [35] A. Sharma and T.J. Rao, Bioresour. Technol., 67, 53–59 (1999).

Please cite this article in press as: M. Ali, M.A. Tindyala, Thermoanalytical studies on acid-treated rice husk and production of some silicon based ceramics from carbonised rice husk, J. Asian Ceram. Soc. (2015), http://dx.doi.org/10.1016/j.jascer.2015.06.003

260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284