Differentiating Grotthuss Proton Conduction Mechanisms by Nuclear

Sep 4, 2014 - Samuel Simon Araya , Fan Zhou , Vincenzo Liso , Simon Lennart Sahlin , Jakob Rabjerg Vang , Sobi Thomas , Xin Gao , Christian Jeppesen ...
6 downloads 0 Views 2MB Size
Download PDF
Letter pubs.acs.org/ac

Differentiating Grotthuss Proton Conduction Mechanisms by Nuclear Magnetic Resonance Spectroscopic Analysis of Frozen Samples Takaya Ogawa,† Kazuhiro Kamiguchi,§ Takanori Tamaki,†,‡ Hideto Imai,§ and Takeo Yamaguchi*,†,‡ †

Chemical Resources Laboratory, Tokyo Institute of Technology, R1-17, 4259 Nagatsuta, Midori-ku, Yokohama, Kanagawa 226-8503, Japan ‡ Kanagawa Academy of Science and Technology, R1-17, 4259 Nagatsuta, Midori-ku, Yokohama, Kanagawa 226-8503, Japan § Device-Functional Analysis Department, NISSAN ARC Ltd., 1 Natsushima, Yokosuka, Kanagawa 273-0061, Japan S Supporting Information *

ABSTRACT: Available methods to analyze proton conduction mechanisms cannot distinguish between two protonconduction processes derived from the Grotthuss mechanism. The two mechanistic variations involve structural diffusion, for which water movement is indispensable, and the recently proposed “packed-acid mechanism,” which involves the conduction of protons without the movement of water and is typically observed in materials consisting of highly concentrated (packed) acids. The latter mechanism could improve proton conductivity under low humidity conditions, which is desirable for polymer electrolyte fuel cells. We proposed a method with which to confirm quantitatively the packed-acid mechanism by combining 2H and 17O solid-state magicangle-spinning nuclear magnetic resonance (MAS-NMR) measurement and 1H pulsed-field gradient (PFG)-NMR analysis. In particular, the measurements were performed below the water-freezing temperature to prevent water movement, as confirmed by the 17O-MAS-NMR spectra. Even without water movement, the high mobility of protons through short- and long-range proton conduction was observed by using 2H-MAS-NMR and 1H-PFG-NMR techniques, respectively, in the composite of zirconium sulfophenylphosphonate and sulfonated poly(arylene ether sulfone) (ZrSPP−SPES), which is a material composed of highly concentrated acids. Such behavior contrasts with that of a material conducting protons through structural diffusion or vehicle mechanisms (SPES), in which the peaks in both 2H and 17O NMR spectra diminished below water-freezing temperature. The activation energies of short-range proton movement are calculated to be 2.1 and 5.1 kJ/mol for ZrSPP−SPES and SPES, respectively, which indicate that proton conduction in ZrSPP−SPES is facilitated by the packed-acid mechanism. Furthermore, on the basis of the mean-square displacement using the diffusivity coefficient below water-freezing temperature, it was demonstrated that long-range proton movement, of the order of 1.3 μm, can take place in the packed-acid mechanism in ZrSPP− SPES.

P

These conduction mechanisms have been examined in many ways, including analyzing the difference of activation energy (Ea),15−18 determining the electroosmotic drag coefficient (Kdrag),10−14 and by conducting nuclear magnetic resonance (NMR)-based investigations.19−21 The proton conduction mechanism of a material with an Ea of 0.1−0.4 eV for proton conductivity has been empirically estimated to follow the Grotthuss mechanism.15−18 However, the E a of proton conduction varies with material properties such as ion exchange capacity.18 Thus, discussions based on only Ea values may not be precise, and further supporting data is required. Alternatively, Grotthuss and vehicle mechanisms can be distinguished by determining Kdrag; this value represents the number of water molecules moving with a single proton through a membrane,

roton conduction supported by water is an essential chemical process in a wide variety of fundamental research fields. For example, in biological systems, proton channels filled with water molecules, such as the D channel in cytochrome c oxidase, drive the synthesis of adenosine triphosphate (ATP) by proton movement.1,2 Such water-assisted proton conduction is also extremely important in materials science3−9 because the proton conductivity of electrolytes used for polymer electrolyte fuel cells (PEFCs) increases under high relative humidity (RH) conditions.4−8 Two main mechanisms of proton conduction via water have been proposed: the vehicle mechanism and the Grotthuss mechanism. In the vehicle mechanism, protonated water forms clusters such as H3O+ and the proton moves through the medium as a water cluster by molecular diffusion (Figure 1a). In this case, some water molecules accompany the proton.8−14 In the Grotthuss mechanism, protons move from oxygen to oxygen by simultaneously breaking and forming hydrogen bonds; i.e., water does not accompany proton motion8−14 (Figure 1b,c). © XXXX American Chemical Society

Received: June 11, 2014 Accepted: August 31, 2014

A

dx.doi.org/10.1021/ac5021485 | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Letter

Figure 1. Mechanisms of proton movement: (a) vehicle mechanism, (b) structural diffusion, and (c) packed-acid mechanism. (d) 17O and (e) 2H MASNMR analysis of ZrSPP−SPES. The larger and sharper peak of 17O at −15.3 °C compared with that at 47.3 °C is because of either 17O quadrupole relaxation23 or paramagnetic relaxation with zirconium in ZrSPP−SPES.24,25 Panels (d and e) were adapted from ref 22 with permission from the Royal Society of Chemistry (Copyright 2014) and are repeated here for clarity.

Here, we propose a method that can be used to confirm the occurrence of the packed-acid mechanism. The approach involves application of 2H- and 17O-solid-state magic-angle spinning (MAS)-NMR and 1H-pulsed-field gradient (PFG)NMR techniques, especially at temperatures below the waterfreezing point. The temperature at which the 17O peak of a material incubated in H217O diminishes in the 17O-MAS-NMR spectrum was defined as “water-freezing temperature” because the diminished 17O peak indicates that water in the material does not move. Such sudden cessation of H217O movement can be interpreted as freezing water. Below the water-freezing temperature, water movement must be eliminated; under these conditions, whereas both structural diffusion and vehicle mechanisms will be disrupted, the packed-acid mechanism can still operate. We utilized a composite of zirconium sulfophenylphosphonate (ZrSPP) and SPES (ZrSPP−SPES) as a material composed of highly concentrated acids that would facilitate migration of protons through the packed-acid mechanism. Indeed, the packed-acid mechanism was previously indicated by 2 H- and 17O-MAS-NMR analysis to take place in such composites. Figure 1d,e was published in the previous report22 and shows the 17O- and 2H-MAS-NMR spectra of ZrSPP−SPES, respectively, at 47.3, −15.3, and −38.8 °C. As a general property of NMR spectra, broadened peaks demonstrate that the movement of a measured atom is slow, whereas sharp peaks mean the opposite. At low temperature, a peak profile usually becomes broadened because of the reduced movement of atoms. The broadened and decreased peak height of 17O, observed upon cooling the sample to below the water-freezing point (−38.8 °C), indicates that mobility of 17O diminished, as expected (Figure 1d). In contrast, the 2H-MAS-NMR results obtained with ZrSPP−SPES (Figure 1e) show that protons in the ZrSPP− SPES sample remain as active at −38.8 °C as at 47.3 °C. In the previous report, these MAS-NMR results were considered as

measured within the cell by applying electrodes across an electrolyte membrane. For example, a typical electrolyte polymer, sulfonated poly(arylene ether sulfone) (SPES), has a value of 0.4 for Kdrag at −25 and −10 °C,11 suggesting that 0.4 water molecules accompany each proton and more than 60% of protons permeate SPES through the Grotthuss mechanism.10−14 Hence, the Grotthuss mechanism is more dominant than the vehicle mechanism in SPES. NMR spectroscopic analysis has been used to determine proton conduction mechanisms; thus, pulsed-field gradient (PFG)-NMR was applied to measure differences in the diffusivity coefficient.19−21 However, these methods cannot be used to distinguish between mechanisms that take place without significant water movement. Recent research has revealed that the Grotthuss mechanism can be subdivided into two mechanisms. One is structural diffusion, known as the common conduction mechanism, for which water fluctuation is indispensable (Figure 1b) because it can overcome the ratedetermining reorientation step.8,9 The second is a “packed-acid mechanism,” which occurs in materials containing highly concentrated (packed) acids.22 In contrast to structural diffusion, the packed-acid mechanism can lead to the conduction of protons without water movement or fluctuation (Figure 1c). Reorientation is overcome by acid−acid interactions.22 Neither of the Grotthuss-derived mechanisms requires water molecules to accompany the proton, although structural diffusion involves water fluctuation. Hence, the difference between these two mechanisms does not affect Kdrag or other physical constants that can be obtained by conventional NMR measurement; i.e., structural diffusion and packed-acid mechanisms are difficult to distinguish from each other based on the previously obtained measurements. A method for distinguishing the packed-acid mechanism from other conduction mechanisms is desired for a complete understanding of many biological systems and the development of proton-conducting materials. B

dx.doi.org/10.1021/ac5021485 | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Letter

Figure 2. SPES measured by (a) 17O-MAS-NMR and (b) 2H-MAS-NMR.

Figure 3. 1H-PFG-NMR of (a) SPES at −15.3 °C, (b) SPES at −38.8 °C, and (c) proton diffusivity of SPES plotted against temperature. 1H-PFG-NMR of (d) ZrSPP−SPES at −15.3 °C, (e) ZrSPP−SPES at −38.8 °C, and (f) proton diffusivity of ZrSPP−SPES plotted against temperature.

evidence for the packed-acid mechanism, which can conduct protons without water movement.22 Although these results indicate that protons move through the packed-acid mechanism under these conditions, MAS-NMR

measurements can only detect local proton movement. Therefore, there remains the possibility that the observed behavior of the protons in this study results from movement on an atomic scale. Furthermore, a quantitative study on the packed-acid C

dx.doi.org/10.1021/ac5021485 | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Letter

Considering that Kdrag = 0.4 in SPES at low temperature,11 i.e., protons in SPES mainly migrate through the Grotthuss mechanism, the data presented in Figures 2 and 3 suggest that the Grotthuss mechanism in SPES can be more accurately described as structural diffusion rather than a packed-acid mechanism. In addition, the peak maximum of the 2H NMR signal in ZrSPP−SPES does not shift as temperature decreases (Figure 1e), whereas the 2H peak maximum in the SPES shifted to lower field. Usually, low water contents decreased the contribution to proton conduction from higher-field signals derived from 2H2O compared with those from SO32H.30,31 Then, 2H peak generally shifts to lower field with decreasing water content in proton movement by freezing of water as shown in the case of SPES (Figure 2b). However, the 2H peak does not shift to lower field in ZrSPP−SPES (Figure 1e), although water freezes. This phenomenon can be interpreted that mobile proton is still influenced by oxygen atoms in the water to the same extent, irrespective of whether water is frozen or not (see Figure S2 in the Supporting Information). Thus, protons in the ZrSPP−SPES composite move via frozen H2O. This phenomenon matches the concept of the packed-acid mechanism in that water movements are not required. Analyzing the 2H NMR peak shift with decreasing temperature may therefore also be considered as a method by which to determine whether a packed-acid mechanism is operating. We then measured the 1H-PFG-NMR spectra of ZrSPP− SPES with decreasing temperature. In contrast to the 1H NMR peak profile in SPES, Figure 3d,e shows that the 1H NMR peak of ZrSPP−SPES was still detected at −38.8 °C even though the mobility of water was restricted by freezing, as suggested by the 17 O-MAS-NMR spectra (Figure 1d). The proton diffusivity in ZrSPP−SPES was plotted against temperature, and proton mobility was retained at temperatures below the water-freezing temperature (Figure 3f). Therefore, it can be concluded that ZrSPP−SPES conducts protons through a packed-acid mechanism, which does not require the movement of water molecules even though the Ea of diffusivity calculated from the Arrhenius plot changed from 17 to 41 kJ/mol at temperatures above and below the water-freezing point, respectively. In addition, the Ea for local proton movement was calculated to be 2.1 kJ/mol based on the line width at half height of the 2H-MAS-NMR signal measured below water-freezing temperature in Figure 1e (see Figure S1 in the Supporting Information). The Ea values for atomic-scale proton movement in ZrSPP−SPES and SPES suggest that, in contrast to proton diffusion in SPES, local proton movement in ZrSPP−SPES was certainly facilitated by the packed-acid mechanism. However, above water-freezing temperature, the diffusivity and Ea values for long-range proton movement were similar for both ZrSPP−SPES and SPES. Therefore, in systems in which water can move, the main proton conduction mechanism can be attributed to structural diffusion even in ZrSPP−SPES. The change in the Ea of diffusivity in ZrSPP−SPES at water-freezing temperature can be attributed to a change in the main conduction mechanism from structural diffusion to a packed-acid mechanism below the water-freezing temperature, while retaining proton mobility. Furthermore, the distance of proton movement in the PFG-NMR sample is 1.3 μm, calculated from the mean-square displacement using the diffusivity at −38.8 °C. Hence, this measurement confirmed that proton movement in the packed-acid mechanism can take place on a micrometer scale and not just on an atomic scale. This finding demonstrates that the packed-acid mechanism can

mechanism was not performed. PFG-NMR spectroscopic measurements can be used to extract information on proton movements on the micrometer scale; thus, PFG-NMR analysis is an appropriate method with which to check for proton movement on a substantial scale. By combining MAS- and PFG-NMR techniques, differences in the movement of protons on the atomic and substantial scale can be identified. SPES was employed as a material for comparison because this polymer is a well-known proton-conducting material, for which Kdrag is 0.4; its dominant proton-conduction mechanism is through the Grotthuss mechanism at low temperature.11 In this report, the use of both MAS- and PFG-NMR measurements, especially below the water-freezing temperature, is proposed as a method with which to distinguish the packed-acid mechanism from other proton-conducting mechanisms.

■

EXPERIMENTAL SECTION

■

RESULTS AND DISCUSSION

We synthesized SPES and ZrSPP−SPES as previously reported.22 The dynamic behavior of protons and water molecules in the composites was investigated by 1H-PFGNMR analysis and by 2H- and 17O-MAS-NMR analysis. SPES and the ZrSPP−SPES composite were incubated in 1H2O, 2H2O, and H217O at 100% RH for 3 days prior to NMR measurements. The incubation in 2H2O or H217O resulted in exchange of the protons of sulfonic acid and adsorbed water with 2H or water in the materials by H217O, respectively. Thus, the 2H-MAS-NMR spectrum included only signals from protons that participate in proton conduction. Similarly, 17O-MAS-NMR analysis targets only water in the materials. 2 H- and 17O-MAS-NMR measurements were carried out with an Agilent 400WB NMR system at a magnetic field of 9.4 T. An airtight 4 mm diameter rotor was used. The spinning speed was 12 kHz. 2H- and 17O-1D-NMR spectra were obtained by using a single pulse sequence. Sample temperature was calibrated by measuring the 79Br signal of KBr powder under the same measurement conditions described elsewhere.26 1 H-PFG-NMR measurements were carried out with a JNM LA-400 spectrometer, which was equipped with a maximum gradient strength of 12 T/m. The diffusion coefficient (D) was measured by using a PFG-stimulated-echo pulse sequence27,28 at a magnetic field of 9.4 T.

The local behavior of protons and water molecules in SPES was measured by 2H- and 17O-MAS-NMR with changing temperature (Figure 2). With decreasing temperature, the peaks of both 2 H and 17O diminished. Ea values of local proton movement were calculated from the Arrhenius plot of line width at half height in 2 H-MAS-NMR results to be 5.1 kJ/mol29 (see Figure S1 in the Supporting Information). Figure 3a,b shows a series of 1H-PFGNMR spectra of SPES samples measured at −15.3 °C (Figure 3a) and −38.8 °C (Figure 3b) as a function of time. These data show that the 1H peak of SPES diminished severely at −38.8 °C, i.e., below the water-freezing temperature, as also indicated in Figure 2a. The proton diffusivity in SPES is plotted against temperature in Figure 3c. The Ea for proton movement on the substantial scale was calculated from the Arrhenius plot to be 17 kJ/mol. The diffusivity below −38.8 °C is not shown because no peak could be detected below this temperature. The drastic change of proton diffusivity confirms that protons in SPES cannot move without water fluctuation, which corresponds to the expected proton behavior for the structural diffusion or vehicle mechanisms. D

dx.doi.org/10.1021/ac5021485 | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Letter

(6) Ogawa, T.; Ushiyama, H.; Lee, J. M.; Yamaguchi, T.; Yamashita, K. J. Phys. Chem. C 2011, 115, 5599−5606. (7) Ogawa, T.; Ohashi, H.; Tamaki, T.; Yamaguchi, T. Phys. Chem. Chem. Phys. 2013, 15, 13814−13817. (8) Kreuer, K. D.; Paddison, S. J.; Spohr, E.; Schuster, M. Chem. Rev. 2004, 104, 4637−4678. (9) Eigen, M.; et al. Angew. Chem., Int. Ed. Engl. 1964, 3, 1−72. (10) Zawodzinski, T. A.; Davey, J.; Valerio, J.; Gottesfeld, S. Electrochim. Acta 1995, 40, 297−302. (11) Gallagher, K. G.; Pivovar, B. S.; Fuller, T. F. J. Electrochem. Soc. 2009, 156, B330−B338. (12) Peng, Z.; Morin, A.; Huguet, P.; Schott, P.; Pauchet, J. J. Phys. Chem. B 2011, 115, 12835−12844. (13) Ge, S. H.; Yi, B. L.; Ming, P. W. J. Electrochem. Soc. 2006, 153, A1443−A1450. (14) Luo, Z. P.; Chang, Z. Y.; Zhang, Y. X.; Liu, Z.; Li, J. Int. J. Hydrogen Energy 2010, 35, 3120−3124. (15) Colomban, P.; Phamthi, M.; Novak, A. J. Mol. Struct. 1987, 161, 1−14. (16) Gosalawit, R.; Chirachanchai, S.; Shishatskiy, S.; Nunes, S. P. Solid State Ionics 2007, 178, 1627−1635. (17) Dai, C. A.; Liu, C. P.; Lee, Y. H.; Chang, C. J.; Chao, C. Y.; Cheng, Y. Y. J. Power Sources 2008, 177, 262−272. (18) Hara, N.; Ohashi, H.; Ito, T.; Yamaguchi, T. J. Phys. Chem. B 2009, 113, 4656−4663. (19) Telfah, A.; Majer, G.; Kreuer, K. D.; Schuster, M.; Maier, J. Solid State Ionics 2010, 181, 461−465. (20) Saito, M.; Arimura, N.; Hayamizu, K.; Okada, T. J. Phys. Chem. B 2004, 108, 16064−16070. (21) Saito, M.; Hayamizu, K.; Okada, T. J. Phys. Chem. B 2005, 109, 3112−3119. (22) Ogawa, T.; Aonuma, T.; Tamaki, T.; Ohashi, H.; Ushiyama, H.; Yamashita, K.; Yamaguchi, T. Chem. Sci. 2014, DOI: 10.1039/ C4SC00952E. (23) Zhu, J. F.; Ye, E.; Terskikh, V.; Wu, G. J. Phys. Chem. Lett. 2011, 2, 1020−1023. (24) Dees, A.; Zahl, A.; Puchta, R.; Hommes, N.; Heinemann, F. W.; Ivanovic-Burmazovic, I. Inorg. Chem. 2007, 46, 2459−2470. (25) Gale, E. M.; Zhu, J.; Caravan, P. J. Am. Chem. Soc. 2013, 135, 18600−18608. (26) Thurber, K. R.; Tycko, R. J. Magn. Reson. 2009, 196, 84−87. (27) Stejskal, E. O.; Tanner, J. E. J. Chem. Phys. 1965, 42, 288−292. (28) Tanner, J. E. J. Chem. Phys. 1970, 52, 2523−2526. (29) Lee, Y. J.; Bingol, B.; Murakhtina, T.; Sebastiani, D.; Meyer, W. H.; Wegner, G.; Spiess, H. W. J. Phys. Chem. B 2007, 111, 9711−9721. (30) Bunce, N. J.; Sondheimer, S. J.; Fyfe, C. A. Macromolecules 1986, 19 (3), 33−339. (31) Ma, Z.; Jiang, R.; Myers, M. E.; Thompson, E. L.; Gittleman, C. S. J. Mater. Chem. 2011, 21, 9302−9311.

contribute to proton conductivity in materials that are utilized for applications such as PEFC, in which the operation may need to start at or below water-freezing temperature (−40 °C). The key point of this method is to freeze water and completely prevent water movement. Structural diffusion cannot occur without water movement, as observed for protons in SPES. Therefore, on the basis of such measurements, it is possible to distinguish between packed-acid mechanism and structural diffusion. In addition, static 2H-MAS-NMR experiments on ZrSPP− SPES were undertaken. The signals were broad, and no clear quadrupole splitting was observed (see Figure S3 in the Supporting Information); thus, the materials studied in this research did not show anisotropy.

■

CONCLUSION We have proposed a method for distinguishing between two proton-conduction mechanisms derived from the Grotthuss mechanism: structural diffusion and the packed-acid mechanism. These two mechanisms are different in that structural diffusion requires water fluctuation whereas the packed-acid mechanism does not. In SPES, in which proton conduction occurs through structural diffusion, the mobility of protons diminished, as shown by NMR measurements on samples below the water-freezing temperature. Proton movement in ZrSPP−SPES remains active even below the water-freezing temperature, indicating that a packed-acid mechanism occurs. In addition, 1H-PFG-NMR analysis can be used to extract information on the scale of proton movement, confirming that proton movement in the packed-acid mechanism occurs on a micrometer scale and not on an atomic scale. Furthermore, comparison of the results obtained by MASand PFG-NMR analysis show that proton movement can take place on either the atomic or substantial scales. On the basis of the results of these measurements, we hereby propose a method with which to distinguish between packed-acid and structural diffusion mechanisms.

■

ASSOCIATED CONTENT

S Supporting Information *

Additional information as noted in text. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Author Contributions

T.O. synthesized the materials and analyzed the results. K.K. performed NMR measurements. T.O. and T.T mainly wrote this paper. H.I and T.Y. conceived this research and managed the project. Notes

The authors declare no competing financial interest.

■

REFERENCES

(1) Voet, D.; Voet, J. G.; Pratt, C. W. Fundamentals of Biochemistry: Life at the Molecular Level, 4th ed.; Wiley: Hoboken, 2012. (2) Decoursey, T. E. Physiol. Rev. 2003, 83, 475−579. (3) Merte, L. R.; Peng, G. W.; Bechstein, R.; Rieboldt, F.; Farberow, C. A.; Grabow, L. C.; Kudernatsch, W.; Wendt, S.; Lægsgaard, E.; Mavrikakis, M.; Besenbacher, F. Science 2012, 336, 889−893. (4) Ogawa, T.; Ushiyama, H.; Yamashita, K.; Lee, J. M.; Yamaguchi, T. Chem. Lett. 2010, 39, 736−737. (5) Ogawa, T.; Kajikawa, Y. Technol. Forecasting Soc. Change 2014, DOI: 10.1016/j.techfore.2014.04.002. E

dx.doi.org/10.1021/ac5021485 | Anal. Chem. XXXX, XXX, XXX−XXX