# Fileset

[Enormous CO2 Sorption Capacity of PDMS Rubbers near Condensation Pressure and at Low Temperatures.pdf](https://mdr.nims.go.jp/filesets/09514606-62dd-4de0-851a-28f8c1a57fa8/download)

## Creator

[Mizuki Inoue](https://orcid.org/0000-0003-3098-5448), Edhuan Ismail, [Sadaki Samitsu](https://orcid.org/0000-0002-4139-1656), Hirofumi Kanoh, [Izumi Ichinose](https://orcid.org/0000-0002-2236-0942)

## Rights

[In Copyright](http://rightsstatements.org/vocab/InC/1.0/)

## Other metadata

[Enormous CO2 Sorption Capacity of PDMS Rubbers near Condensation Pressure and at Low Temperatures](https://mdr.nims.go.jp/datasets/15661849-f7db-411c-9c81-beeb385f40d5)

## Fulltext

Microsoft Word - Chem Lett1  Mizuki Inoue,1 Edhuan Ismail,1 Sadaki Samitsu,2 Hirofumi Kanoh,3 Izumi Ichinose*1 1Research Center for Functional Materials, NIMS, 1-1 Namiki, Tsukuba, 305-0044 Japan 2 Research and Services Division of Materials Data and Integrated System, NIMS, 1-2-1 Sengen, Tsukuba 305-0047, Japan 3 Graduate School of Science, Chiba University, 1-33 Yayoi-cho, Inage, Chiba 263-8522, Japan E-mail: ICHINOSE.Izumi@nims.go.jp We report for the first time that PDMS rubber sorbs an 1 extremely large amount of CO2 (792 mg g−1) at near the 2 condensation pressure at a certain low temperature (−3 ℃), 3 giving the sorption enthalpy of 9.3–10.8 kJ mol−1. The 4 sorption isotherms were evaluated by means of the Flory–5 Huggins equation, and the resultant  parameters (1.3–0.2) 6 were used to estimate the solubility parameter (SP). The 7 CO2/CH4 and CO2/N2 selectivities were 9.6–44 at 100–8 3000 kPa and were far larger than those of activated carbon. 9  10 Keywords: PDMS, Flory-Huggins theory, CO2 sorption 11  The solubility of CO2 in polymeric materials has been 12 extensively studied in relation to membrane separation.1–8 13 The miscibility of polymers with supercritical CO2 has 14 become another topic of interest in recent decades.9–13 High-15 pressure CO2 often plasticizes polymers, decreasing their 16 glass-transition temperature. Rubbers such as 17 polydimethylsiloxane (PDMS) sorb large amounts of CO2; 18 these amounts increase with increasing partial pressure in 19 accordance with Henry’s law. The dual-sorption model 20 proposed by Barrer et al. has been used to estimate the 21 Henry’s and Langmuir’s constants of CO2 in membranes.14 22 Although the latter constant is a measure of the adsorption 23 capacity at a certain pressure, the dual-sorption model cannot 24 predict the maximum amount of CO2 that is sorbed near the 25 condensation pressure.   26 A large sorption capacity of CO2 is urgently required for 27 systems for recovering greenhouse gases.15,16 The sorption 28 capacities of PDMS rubbers are generally far lower than 29 those of porous adsorbents such as activated carbons, zeolites, 30 or metal–organic frameworks. Merkel et al. reported that 31 PDMS in membrane form sorbs 75 mg g−1 of CO2 at 32 2500 kPa and 35 ℃.7 The solubility of CO2 has in PDMS 33 been reported to be in the range 70–120 mg g−1  at 2000–34 3000 kPa and 25 ℃, as calculated by a first-principles 35 calculation.17 A study of the sorption of CO2 in bulk PDMS 36 rubber by Shah et al. gave a CO2 solubility of 94 mg g−1 at 37 2000 kPa and 10 ℃.18 However, to the best of our knowledge 38 there have been no detailed investigations of the maximum 39 amount of CO2 sorbed by PDMS rubbers. In this study, we 40 evaluated CO2 sorption behaviors of PDMS rubbers at low 41 temperatures and we observed an extremely large sorption 42 near the condensation pressure.  43 CO2 separation at high pressure conditions is strongly 44 demanded in oil & gas industry. For example, natural gas 45 produced from deep underground has a pressure exceeding 46 120 MPa and often contains several tens of percent of CO2. 47 Most of the CO2 can be separated by reducing pressure and 48 subsequent cooling. However, partial pressure of CO2 is still 49 above 2.5 MPa even at −10 ℃. Therefore, the sorbents that 50 can selectively remove CO2 from CH4/CO2 mixture at low 51 temperature are required. Furthermore, if considered the 52 efficient transportation, CO2 separation process needs to be 53 operated at high pressure. In this paper, we report PDMS 54 powder has a high potential as the solution for this problem. 55 A sample of PDMS rubber was prepared from 56 SYLGARD™ 184 (Dow Corning Co.). PDMS macromer 57 (1.0 g) was mixed with a crosslinker (0.1 g) and the mixture 58 was stirred for five minutes and then cured at room 59 temperature for several days. To achieve a short equilibrium 60 time in sorption measurements, submillimeter-sized PDMS 61 particles were prepared by cryogenic grinding (Pulverisette 62 14, Fritsch) (Figure S1). CO2 sorption isotherms were 63 measured by using a BELSORP-HP (MicrotracBEL). The 64 sorption temperature range was −30 to 25 ℃, and the 65 equilibrium criterion was set as a pressure deviation of less 66 than 0.1% for 500 s. 67 Figure 1. CO2 sorption isotherms of PDMS rubber at −30 to 25 ℃. The 68 upper limit of the sorption pressure in the system was 3200 kPa.  69 The condensation pressure of CO2 at −30 ℃ is 1420 kPa, 70 and it reaches a supercritical state at 31 ℃ and 7377 kPa. The 71 sorption measurement equipment used in this study had an 72 upper pressure limit of 3200 kPa. Figure 1 shows the CO2 73 sorption isotherms for PDMS rubber at −30 to 25 ℃. The 74 dotted lines represent the condensation pressures at each 75 temperature. Interestingly, the maximum sorption increased 76 uniformly from 210 mg g−1 at −30 ℃ to  332 mg g−1 at −20 ℃, 77 489 mg g−1 at −10 ℃, and 792 mg g−1 at −3 ℃. The final 78 value was 4.4 times larger than that at −30 ℃. At 3100 kPa, 79 the amounts of CO2 sorbed at 4 ℃ and at 25 ℃ were 80 260 mg g−1 and 180 mg g−1, respectively.  81 Enormous CO2 Sorption Capacity of PDMS Rubbers near Condensation Pressure and at Low Temperatures 2   Because of the high flexibility of the polymer chains, 1 the sorption of CO2 by PDMS rubber can be addressed in 2 terms of solubility in a liquid polymer, and it can be 3 quantitatively discussed by means of the Flory–Huggins 4 equation (1):19–22 5 ln 𝑃/𝑃 = ln(1 − 𝜑 ) + 1 −1𝑚𝜑 + 𝜒𝜑      (1) 6 where, 𝑃  is the vapor pressure (Pa) of the dissolving gas 7 (CO2), 𝑃  is its saturated vapor pressure (Pa), 𝜑  is the 8 volume fraction of the polymer in the liquid phase, 𝑚 is the 9 degree of polymerization, and 𝜒 is the interaction parameter. 10 Note that this equation does not express the effect of elastic 11 modulus. For the 𝑚  value, we employed the degree of 12 polymerization of the macromer (𝑚 = 765) from which the 13 PDMS rubber was prepared. 𝜑  was calculated directly from 14 the amount of CO2 sorbed. The solid lines in Figure 1 show 15 the predicted isotherms calculated by using Equation (1), and 16 the best-fit values of the  parameter. The Flory–Huggins 17 equation accurately traced the CO2 sorption isotherms at all 18 temperatures. The  value was 1.3 at −30 ℃, 0.81 at −3 ℃, 19 and 0.20 at 25 ℃, decreasing with increasing temperature. In 20 our experimental temperature range, the amount of CO2 21 sorbed was predicted to be highest at 25 ℃, near the saturated 22 vapor pressure, because of the minimum  value. However, 23 the sorption pressure also becomes extremely high, to the 24 point where it is unrealistic for practical separation processes. 25 At a constant CO2 pressure, the amount of CO2 sorbed 26 increased with decreasing temperature. For example, at 27 a partial pressure of 1000 kPa, the amount of CO2 sorbed was 28 124 mg g−1 at −30 ℃ and 33 mg g−1 at 25 ℃. The former 29 amount is 3.8 times larger than the latter. 30 Figure 2. (a) Changes in the solubility parameters of CO2 and PDMS 31 rubber at 243–298 K. The values of CO2 were obtained from literature.23 32 (b)  Sorption enthalpies of AC and PDMS rubber for various sorption 33 amounts. The dashed lines are the CO2 condensation heats at 243 K and 34 270 K. 35 The maximum amounts of CO2 absorbed in PDMS are 36 210–792 mg g−1, and we believe that CO2 exists as a 37 continuous body within the polymer chains at such high 38 concentrations. Therefore, the CO2 is in a condensed form. 39 The miscibility of the polymer and solvent can be predicted 40 from the values of the Hansen solubility parameter (SP), and 41 its relation to the  parameter can be described by the 42 following equation:24 43 𝜒 =𝑀𝜌𝑚𝑅𝑇𝛿 − 𝛿       (2) 44 where, Mpoly is the molecular weight of the polymer (PDMS 45 macromer),  is the polymer density (kg m−3), m is the degree 46 of polymerization, R is the universal gas constant 47 (8.314 J mol−1 K−1), T is the temperature (K), and poly and sol 48 are the solubility parameters (Pa0.5) of PDMS and CO2, 49 respectively. The PDMS rubber used here has Mpoly = 57, 50 = 1030, and m = 765. The SP (poly) for PDMS can be 51 calculated from its  value and the known SP of CO2 (sol), 52 and the results are shown in Figure 2(a). The SP for CO2 53 decreases with increasing temperature; it is 19 MPa0.5 at 54 −30 ℃, 15 MPa0.5 at −3 ℃, and 9.4 MPa0.5 at 25 ℃.23 On the 55 other hand, the SP for PDMS was calculated to be 24 MPa0.5 56 at −30 ℃, 19 MPa0.5 at −3 ℃, and 10 MPa0.5 at 25 ℃.25 The 57 SP at −30 ℃ is therefore 80% larger than that at room 58 temperature, and this is interesting new knowledge regarding 59 PDMS rubber. 60 The sorption enthalpy change (𝛥𝐻 ) for CO2 can be 61 calculated by using the Clausius–Clapeyron equation: 62 ln𝑃𝑃= −𝛥𝐻𝑅1𝑇−1𝑇      (3) 63 where, 𝑃  and 𝑃  are the gas partial pressures (Pa) at certain 64 sorption amounts, 𝛥𝐻  is the heat of sorption (J mol－1), 𝑅 is 65 the universal gas constant (J mol−1 K−1) and  𝑇  and 𝑇  are the 66 temperatures (K) used for the sorption measurements at 𝑃  67 and 𝑃 , respectively. Figure 2(b) shows the change in the 68 values of 𝛥𝐻  for PDMS rubber and for activated carbon 69 (SORBONORIT 4) for sorptions of 30–180 mg g−1. The 𝛥𝐻  70 values were calculated at 𝑇 = 243 K and 𝑇 = 270 K. The 71 CO2 condensation heats are 13 kJ mol−1 at 243 K and 11 kJ 72 mol−1 at 270 K.26 The 𝛥𝐻  values for activated carbon 73 decreased with the sorption amount from 24 kJ mol−1 at 74 30 mg g−1 to 21 kJ mol−1 at 180 mg g−1. These values are 9–75 12 kJ mol−1 larger than the average heat of condensation of 76 CO2. In contrast, the 𝛥𝐻  values for PDMS increased with 77 increasing sorption, giving a value of 9.3 kJ mol−1 at 78 30 mg g−1 and 11 kJ mol−1 at 180 mg g−1. These values were 79 39–52% of the corresponding values for activated carbon, 80 and were slightly lower than the average values of the heat of 81 condensation of CO2. Interestingly, the 𝛥𝐻  values increase 82 with increasing sorption amount. That is, CO2 readily 83 dissolved in the swollen PDMS rubber. 84 Figure 3. (a) Comparison of CO2 sorption isotherms of PDMS rubber 85 and activated carbon. (b) CO2, CH4, and N2 sorption isotherms of PDMS 86 rubber at 270 K. 87 3   Figure 3 (a) shows a comparison of the CO2 isotherms 1 for PDMS rubber and activated carbon. The amount sorbed 2 by activated carbon was 644 mg g−1 at 3130 kPa, which is 3 19% less than that of PDMS rubber. In the CO2 partial-4 pressure range 0–2000 kPa, activated carbon adsorbed a large 5 amount of CO2. However, complete desorption required the 6 pressure to be reduced to below atmospheric pressure. In 7 contrast, PDMS rubber desorbed 80% of CO2 when the 8 pressure was reduced from the saturation pressure to 9 2000 kPa. 10 A comparison of the CO2, CH4, and N2 sorption 11 isotherms at 270 K is shown in Figure 3(b). The inset contains 12 magnified plots for CH4 and N2. The amounts of CH4 and N2 13 sorbed at 3100 kPa were 19 and 15 mg g−1, respectively. At 14 this pressure, the amount of CO2 sorbed was about 37–53 15 times larger than these values. The ideal gas selectivities of 16 CO2/CH4 and CO2/N2 were therefore 9.6–44 at 100–3000 kPa, 17 and higher selectivities were observed at a higher pressure 18 (Figure S2, Table S1). In contrast, these selectivities were 19 1.9–8.5 for activated carbon, and the values decreased at 20 higher pressures. 21 Figure 4. Comparison of desorption rates under (a) high-pressure and (b) 22 low-pressure conditions. 23 In Figure 1, the amounts of CO2 sorbed were measured 24 by a volumetric method from the equilibrated pressure 25 changes in a gas container of known volume (manifold) and 26 a sample container. Figure 4(a) shows one of the desorption 27 cycles, in which the pressure changes of the manifold at 28 2828 kPa were plotted after the sample container at 3134 kPa 29 was opened. In this case, the initial manifold pressure was 30 306 kPa lower than that of the sample container. After 420 s, 31 95% of CO2 was desorbed and the manifold reached an 32 equilibrium pressure of 2988 kPa. Figure 4(b) shows the 33 pressure changes of the manifold initially at 2429 kPa when 34 the sample container at 2637 kPa was opened; in this case 35 95% desorption was reached after 130 s.  36 As mentioned before, a large amount of CO2 37 (792 mg g−1) was sorbed at 3134 kPa, and the desorption is 38 an endothermic process. Presumably, when liquid CO2 39 changes to the gas state, a longer desorption time is required. 40 At a pressure of 2637 kPa, the amount sorbed was 290 mg g−1 41 and the CO2 in PDMS rubber started to show a slightly gas-42 like behavior. The diffusion coefficient and the desorption 43 rate of CO2 then increased. 44 From the viewpoint of separation engineering, CO2 45 recovery at a low temperature, that is, the amount sorbed at 46 Figure 5. CO2 sorption isotherms of PDMS rubber and other polymer 47 rubbers at 243 K. NBR: PN20HA (JSR), PVDF-HFP: G701BP 48 (DAIKIN). 49 low partial pressures, is important. In Figure 5, the CO2 50 sorption isotherms for PDMS rubber, acrylonitrile–1,3–51 butadiene rubber (NBR), and poly(vinylidene difluoride)–52 hexafluoro-propylene rubber (PVDF-HFP) are compared in 53 the pressure range 0–900 kPa. At −30 ℃ and 900 kPa, the 54 amounts of CO2 sorbed by PDMS rubber, NBR, and PVDF-55 HFP were 100, 59, and 15 mg g−1, respectively. It is clear that 56 PDMS rubber is far superior sorbent to PVDF-HFP at low 57 temperatures. The glass-transition temperatures (Tg) of 58 PDMS rubber, NBR, and PVDF-HFP are −120 ℃, −43 ℃, 59 and −35 ℃, respectively.27–29 At −30 ℃, the elastic modulus 60 of NBR is about 600 times larger than that of PDMS rubber 61 and the elastic modulus of PVDF-HFP reaches 2 GPa.30,31 62 The swelling of these two polymers by CO2 must be strongly 63 limited by their high elastic moduli. 64 Figure 6. Schematic illustration of PDMS rubbers swollen with 792 mg 65 g−1 of CO2.  66 Figure 6 is a schematic illustration of PDMS rubber 67 containing 792 mg g−1 of sorbed CO2. The crosslinked rubber 68 structure has 76 [Si(CH3)2O] units and 106 CO2 molecules. 69 By using a CO2 density of 0.946 g cm−3 at −3 ℃, the volume 70 ratio of PDMS to CO2 was calculated to be 54:46. The high 71 CO2 sorption capacity of nonporous PDMS rubber is 72 surprisingly high, even in comparison to the corresponding 73 values for porous sorbents. 74 The relationship between the CO2 partial pressure and 75 the amount of the gas sorbed was precisely analyzed by using 76 the Flory–Huggins equation. From a mathematical viewpoint, 77 this equation predicts that there is no particular upper limit to 78 the amount sorbed when the χ parameter is less than ~0.7. 79 4   However, swelling of the PDMS rubber will stop when 𝛥𝐻  1 is balanced by the entropic elasticity of the polymer chains. 2 This is quite different from the adsorption behavior of porous 3 adsorbents such as activated carbon. Our results show that a 4 lower elastic modulus at low temperatures favors the sorption 5 of a large amount of CO2. When choosing a polymer sorbent, 6 the SP should be close to that of CO2 at a low temperature. 7 We also need to consider that the SP of CO2 increases with 8 decreasing temperature. The amounts of CO2 sorbed by our 9 PDMS rubber are an order of magnitude larger than those 10 reported in the literature (Figure S3, Table S2). This is 11 because previous studies used a dual-sorption model and 12 examined only the range in which Henry’s law is valid. In 13 this study, by using a volumetric method, we have 14 demonstrated that rubber powders can sorb large amounts of 15 CO2 at low temperatures and near the saturation pressure. 16 PDMS rubber is therefore a good sorbent for CO2, even 17 at low partial pressures, provided the temperature is low. This 18 result opens up new applications for PDMS in CO2 separation 19 and recovery. In particular, because of the large selectivity 20 between CO2 and CH4 sorption, we believe that PDMS rubber 21 powders might be useful in the purification of CO2-rich 22 natural gas and biogas. 23  24 Acknowledgments 25 The authors are grateful for the financial support of the 26 Cabinet Office, Government of Japan, Cross-Ministerial 27 Moonshot Agriculture, Forestry and Fisheries Research and 28 Development Program, ‘Technologies for Smart Bioindustry 29 and Agriculture’ (No. JPJ009237). The authors also 30 appreciate Mr. R. Ema of National Institute of Technology, 31 Tokyo College for his supports in literature arrangement. 32  33 References and Notes 34 1 A. Bos, I. G. M. Pünt, M. Wessling, H. Strathmann, J. Memb. 35 Sci. 1999, 155, 67. 36 2 V. I. Bondar, B. D. Freeman, I. Pinnau, J. Polym. Sci. Part B 37 Polym. Phys. 1999, 37, 2463. 38 3 K. Tanaka, M. Okano, H. Toshino, H. Kita, K. I. Okamoto, J. 39 Polym. Sci. Part B Polym. Phys. 1992, 30, 907. 40 4 T. C. Merkel, V. Bondar, K. Nagai, B. D. Freeman, Y. P. 41 Yampolskii, Macromolecules 1999, 32, 8427. 42 5 H. D. Kamaruddin, W. J. Koros, J. Memb. Sci. 1997, 135, 147. 43 6 Y. Hirayama, Y. Kase, N. Tanihara, Y. Sumiyama, Y. Kusuki, 44 K. Haraya, J. Memb. Sci. 1999, 160, 87. 45 7 T. C. Merkel, V. I. Bondar, K. Nagai, B. D. Freeman, I. Pinnau, 46 J. Polym. Sci. Part B Polym. Phys. 2000, 38, 415. 47 8 H. Lin, B. D. Freeman, J. Mol. Struct. 2005, 739, 57. 48 9 M. Houben, R. van Geijn, M. van Essen, Z. Borneman, K. 49 Nijmeijer, J. Memb. Sci. 2021, 620, 118922. 50 10 J. D. Wind, S. M. Sirard, D. R. Paul, P. F. Green, K. P. 51 Johnston, W. J. Koros, Macromolecules 2003, 36, 6433. 52 11 J. Von Schnitzler, R. Eggers, J. Supercrit. Fluids 1999, 16, 81. 53 12 O. Akin, F. Temelli, J. Memb. Sci. 2012, 399–400, 1. 54 13 M. Houben, J. Kloos, M. van Essen, K. Nijmeijer, Z. 55 Borneman, J. Memb. Sci. 2022, 647, 120292. 56 14 R. M. Barrer, J. A. Barrie, J. Slater, J. Polym. Sci. 1958, 27, 57 177. 58 15 D. M. D’Alessandro, B. Smit, J. R. Long, Angew. Chemie Int. 59 Ed. 2010, 49, 6058. 60 16 M. Bui, C. S. Adjiman, A. Bardow, E. J. Anthony, A. Boston, 61 S. Brown, P. S. Fennell, S. Fuss, A. Galindo, L. A. Hackett, J. 62 P. Hallett, H. J. Herzog, G. Jackson, J. Kemper, S. Krevor, G. 63 C. Maitland, M. Matuszewski, I. S. Metcalfe, C. Petit, G. Puxty, 64 J. Reimer, D. M. Reiner, E. S. Rubin, S. A. Scott, N. Shah, B. 65 Smit, J. P. M. Trusler, P. Webley, J. Wilcox, N. Mac Dowell, 66 Energy Environ. Sci. 2018, 11, 1062. 67 17 W. Shi, N. S. Siefert, B. D. Morreale, J. Phys. Chem. C 2015, 68 119, 19253. 69 18 V. M. Shah, B. J. Hardy, S. A. Stern, J. Polym. Sci. Part B 70 Polym. Phys. 1986, 24, 2033. 71 19 Y. Kamiya, Y. Naito, D. Bourbon, J. Polym. Sci. Part B Polym. 72 Phys. 1994, 32, 281. 73 20 C. P. Ribeiro, B. D. Freeman, Macromolecules 2008, 41, 9458. 74 21 E. Favre, P. Schaetzel, Q. T. Nguygen, R. Clément, J. Néel, J. 75 Memb. Sci. 1994, 92, 169. 76 22 C. A. Scholes, G. W. Stevens, S. E. Kentish, J. Memb. Sci. 77 2010, 350, 189. 78 23 Y. Marcus, ACS Omega 2018, 3, 524. 79 24 T. Lindvig, M. L. Michelsen, G. M. Kontogeorgis, Fluid Phase 80 Equilib. 2002, 203, 247. 81 25 H. Watanabe T. Miyauchi, J. Chem. Eng. Jpn. 1973, 6, 109. 82 26 NIST Chemistry WebBook, NIST Standard Reference 83 Database Number 69. 84 27 K. Zalewski, Z. Chyłek, W. A. Trzciński, Polymers. 2021, 13. 85 28 P. Rybiński, A. Kucharska-Jastrza̧bek, G. Janowska, Polym. 86 Sci. Ser. B 2014, 56, 477. 87 29 L. F. Malmonge, J. A. Malmonge, W. K. Sakamoto, Mater. 88 Res. 2003, 6, 469. 89 30 C. Rajesh, P. Divia, S. Dinooplal, G. Unnikrishnan, E. 90 Purushothaman, Polym. Polym. Compos. 2021, 29, S1328. 91 31 A. J. de J. Silva, C. R. Nascimento, M. F. da Costa, J. Mater. 92 Sci. 2016, 51, 9074. 93  94  95 5