Communication pubs.acs.org/JACS
Formation of (HCOO−)(H2SO4) Anion Clusters: Violation of Gas-Phase Acidity Predictions Gao-Lei Hou,† Xue-Bin Wang,*,† and Marat Valiev*,‡ †
Physical Sciences Division, Pacific Northwest National Laboratory, 902 Battelle Boulevard, P.O. Box 999, MS K8-88, Richland, Washington 99352, United States ‡ Environmental Molecular Sciences Laboratory, Pacific Northwest National Laboratory, P.O. Box 999, Richland, Washington 99352, United States S Supporting Information *
significant deviations from the behavior that would be expected based on aqueous ionization constants. A similar situation can be also observed in biological systems, where proton transfer reaction processes in active sites are not always in line with the standard pKa values.17 Given these uncertainties, it seems natural to consider using the gas-phase proton affinity (PA) of participating species as a reference point for the analysis of proton transfer processes in nonaqueous solutions and developing appropriate acidity/ basicity measures tailored to a particular environment. Should one take such an approach, there is an implicit presumption that PA values portray an accurate picture under gas-phase conditions. Initial evidence, contrary to this fact, has surfaced in the analysis of the molecular cationic clusters,18−22 where it has been observed that individual PA values do not convey the right chemistry. Recently more examples have been reported, but this time affecting the anionic complexes. One of these cases involves thiourea/acetic acid complexes,23 which exhibit deprotonation of acetic acid despite the unfavorable differences in PA values. It was suggested that the resulting complex gains additional stability due to creation of additional intermolecular hydrogen bonds. In another study of bisulfate/succinic acid (HOOC(CH2)2COOH) clusters, it was found that a proton migrates from the higher PA of the singly deprotonated succinic base anion to the lower PA of HSO4−.24 In this case, it was suggested that formation of the intramolecular hydrogen bond in succinic acid may have increased its gas-phase acidity. The natural question is whether these examples are exceptions to the rule or is there a more fundamental reason behind it? This is the issue investigated in this work, where we perform detailed temperature-dependent negative ion photoelectron spectroscopy (NIPES)25 and high level computational analysis26 of a 1:1 gas-phase cluster involving two common molecular species: sulfuric acid and formate. The large difference between PA values of formate (345.4 ± 2.2 kcal/ mol)27 and bisulfate (309.5 ± 2.6 kcal/mol)27 would seem to imply exclusive formation of HSO4−(HCOOH); however, our results indicate this may not be entirely the case. Figure 1 presents the 157 nm photoelectron spectra of the (HSO4−)·H+·(HCOO−) cluster measured at both 20 and 300 K (see Supporting Information (SI) for experimental methods). It can be seen that in the lower electron binding energy (EBE)
ABSTRACT: Sulfuric acid is commonly known to be a strong acid and, by all counts, should readily donate its proton to formate, which has much higher proton affinity. This conventional wisdom is challenged in this work, where temperature-dependent negative ion photoelectron spectroscopy and theoretical studies demonstrate the existence of the (HCOO−)(H2SO4) pair at an energy slightly below the conventional (HCOOH)(HSO4−) structure. Analysis of quantum-mechanical calculations indicates that a large proton affinity difference (∼36 kcal/ mol), favoring proton transfer to formate, is offset by the gain in intermolecular interaction energy between HCOO− and H2SO4 through the electron delocalization and formation of two strong hydrogen bonds. However, this stabilization comes with a severe entropic penalty, requiring the two species in the precise alignment. As a result, the population of (HCOO−)(H2SO4) drops significantly at higher temperatures, rendering (HCOOH)(HSO4−) to be the dominant species. This phenomenon is consistent with the photoelectron data, which shows depletion in the spectra assigned to (HCOO−)(H2SO4), and has also been verified by ab initio molecular dynamics (AIMD) simulations.
P
roton transfer processes play an important role in a broad range of chemical and biological phenomena.1−8 Fundamental understanding of the preferred protonation state as a function of molecular composition has a long history of extensive research9 and presents an important direction in rationalizing proton driven processes.10−14 A time-honored approach to this problem consists of using the concept of acidity/basicity, which aims to quantify the tendency for proton loss/addition for a given species.15 It has been recognized that designation of a given substance as an acid (proton donating) or a base (proton accepting) is a relative concept, which depends not only on the species involved but also on the external environment in which process occurs.9,13 This requires a careful definition in terms of basic reaction processes in a given medium, and a vast amount of such work has been done for an aqueous solution case. There is sometimes an expectation that aqueous phase results can be transferred to other cases, but it is not always true. This is particularly well illustrated in the case of ionic liquids,16 which often show © 2017 American Chemical Society
Received: June 8, 2017 Published: August 7, 2017 11321
DOI: 10.1021/jacs.7b05964 J. Am. Chem. Soc. 2017, 139, 11321−11324
Communication
Journal of the American Chemical Society
the components I and III are similar in both temperature cases (1/15.7 for 20 K and 1/17.5 for 300 K). This indicates that the signals corresponding to components I and III may come from the same isomer, with component I corresponding to ground state and component III to excited state transitions (see Figure S1 in SI for the density of states and time-dependent density functional theory (TD-DFT) calculation analyses). Note that the small deviation of the ratios is most likely due to the different thermal broadening effect on the spectral features at different temperatures. This implies that the component II may be identified as a contribution coming from a second isomer whose population is reduced at higher temperatures. In agreement with experimental results, ab initio electronic structure calculations indeed show the presence of two stable isomers (Figure 3, and Tables 1 and S1; see SI for theoretical
Figure 1. 157 nm photoelectron spectra of (HSO4−)·H+·(HCOO−) cluster recorded at 20 (blue) and 300 (red) K, respectively.
range of ∼5.8−6.5 eV, the two spectra have nearly perfect overlap with each other. The 300 K spectrum does have small additional features at EBE around 5.8 eV, which are most likely due to hot bands of the vibrationally excited clusters at high temperatures, as has been observed in previous temperature controlled NIPES experiments.28 In the higher EBE range of 6.5−7.2 eV, the two spectra start to deviate from each other. In particular, we observe that the high temperature (300 K) signal exhibits a depletion of intensity compared to that at the low temperature of 20 K. Such behavior cannot be accounted for by the temperature broadening effect and suggests the possibility of existence of two different isomers that contribute differently based on the temperature. To help understand the origin of these differences, the experimental spectra were deconvoluted into three Gaussian line bands, I (blue), II (orange), and III (green) (see SI for the deconvolution details), covering 5.8−6.5, 6.5−7.2, and >7.2 eV energy regions, respectively.29−31 As shown in Figure 2, the resulting Gaussian shapes provide a good fit to the spectra in both temperature regimes. We found that the relative ratios of
Figure 3. Optimized structures of the two isomers (iso A, (HCOOH)(HSO4−); iso B, (H2SO4)(HCOO−))and spin density distributions (inserts) of the neutral radicals at the corresponding anionic geometries.
details). The first isomer (isomer A) was identified as (HCOOH)(HSO4−), a complex of formic acid and bisulfate anion. Such a structure would be naturally expected due to a much higher PA of formate versus bisulfate (345.4 vs 309.5 kcal/mol). It was therefore unusual to see the reverse protonation pattern in the second isomer (isomer B), a complex of sulfuric acid and formate, (H2SO4)(HCOO−). Surprisingly, the two isomers were found to be essentially degenerate in energy with a small preference for isomer B by 0.43 kcal/mol (Table 1) at the CCSD(T) level including the B3LYP zero-point energy (ZPE) corrections. The nature of the negatively charged species is different for isomers A and B. For isomer A, the excess electron density is associated with HSO4−, while for B it is located on HCOO− (Figure S2). As a result, there expected to be a significant difference in their vertical detachment energies (VDEs). Indeed calculations at the coupled cluster level of theory predict the VDE value for (HSO4−)(HCOOH) isomer A to be 5.78 eV and (H2SO4)(HCOO−) isomer B to be 6.55 eV (Table 1), consistent with the experimental data fitting procedure, which assigns features I and III to isomer A and feature II to isomer B (Figure 2). Our density of states and TD-DFT calculations given in Figure S1 support such assignments as well. The stability of isomer B, (H2SO4)(HCOO−), seems puzzling at a first glance, and warrants further analysis and discussion. It should be noted that while PA values seem to provide a good metric for quantifying protonation states, these numbers refer to isolated species. If the molecules in the acid− base pair are in close contact with each other and experience strong mutual interactions, the PA values alone cannot predict the final energy balance. To illustrate this point we can consider a process of building cluster complexes by first protonating the
Figure 2. Fitted photoelectron spectra of (HSO4−)H+(HCOO−) cluster (black) overlaid with the experimental spectra at 20 (left) and 300 (right) K, respectively (middle panel). The three Gaussian bands used for the fitting, and the resulting errors in the fitting are shown in the bottom and upper panels, respectively. 11322
DOI: 10.1021/jacs.7b05964 J. Am. Chem. Soc. 2017, 139, 11321−11324
Communication
Journal of the American Chemical Society Table 1. Energetics of the Two Isomers of (HSO4−)·H+·(HCOO−)a B3LYP
CCSD(T)
Exptl.
isomer A (HCOOH)(HSO4−)
isomer B (HCOO−)(H2SO4)
0.52 5.55 4.94 21.71 0.43 5.78 5.09 23.42 6.10
0.00 5.95 4.96 53.63 0.00 6.55 5.08 56.99 6.85
ΔE (kcal/mol) VDE (eV) ADEb (eV) ΔEintc (kcal/mol) ΔE(kcal/mol) VDE (eV) ADEb (eV) ΔEintc (kcal/mol) VDE (eV)
The experimental uncertainty is ±0.1 eV. bAdiabatic detachment energy. cΔEint = E(HSO4−) + E(HCOOH) − E[A] for isomer A and ΔEint = E(H2SO4) + E(HCOO−) − E[B] for isomer B. The energies include ZPE corrections. a
isomer B essentially disappears (Figure S4). Analysis of the trajectories indicates that the primary reason for reduced population of isomer B at higher temperatures indeed involves perturbation of the relative orientation of the two molecules. These results are qualitatively consistent with experimentally observed depletion of the isomer B population at higher temperatures, suggesting that isomer B is an energetically favored, but entropically penalized conformer. Overall, our results indicate that relative PA values may not always provide a reliable prediction of possible protonation states in situations where the participating molecules are in close contact with each other and experience strong mutual interactions. The fact that we observe an unusual protonation pattern even in the seemingly straightforward case of the sulfuric acid formate complex is unexpected and indicates that such behavior may be quite common.33 Our analysis suggests that electron delocalization accompanied by formation of multiple hydrogen bonds17,36,37 may be the main stabilization mechanisms. However, such stabilization will result in a significant entropic penalty, which indicates that such structures would probably be observed only in very special cases, such as perhaps protein active sites.
separated species and then bringing them together. The energy balance in the first step is driven by PA values, however the second step proves to be quite important. Indeed our calculations indicate that isomer A is characterized by mutual interaction energy of ∼23 kcal/mol, while in the isomer B that number goes up to ∼57 kcal/mol (Scheme S1 and Table S2). The end result is that despite significant energetic penalty associated with the protonation of bisulfate group in isomer B (∼36 kcal/mol), the apparent gain in mutual interaction energy (∼34 kcal/mol) has a potential to compensate for that. The same argument of the importance of the mutual interaction energy has been made in the previous studies of proton location in the cationic protonated clusters.18,19 The origin for enhanced stability of isomer B, (H2SO4)(HCOO−), is likely to arise from the increased electron delocalization. This can be anticipated by the comparison of the electron binding energies of bisulfate (4.75 eV)24 and formate (3.50 eV).32 The less bound excess electron in formate will be much easier to delocalize compared to bisulfate. This conclusion is supported by charge analysis showing that the anion group contains 90% of the excess negative charge in isomer A and 75% in isomer B as well as the electron cloud distributions (see Figure S2 in SI for the natural populated charge distributions and the highest occupied molecular orbitals). Similarly, the analysis of the spin densities of the neutral complexes after electron removal shows much more pronounced delocalization in isomer B (Figures 3 and S3). It is possible that the same electron delocalization argument may also apply to the recent results by Yacovitch et al.,33 showing the existence of (NO3−)(H2SO4) complex, contrary to PA values. Similar to our case, the electron binding energy of NO3− (3.94 eV34) is much lower than that of HSO4−. The enhanced stability of isomer B is also consistent with the fact that HCOO− and H2SO4 are being bound together by two hydrogen bonds of equal strength. The isomer A is more asymmetrical in nature and essentially contains one intermolecular hydrogen bond involving the OH group of formic acid. To enable formation of the double hydrogen bonded structure shown in isomer B, the two molecules have to be locked in a very specific configuration, which may be easily destroyed given enough thermal fluctuations. Ab initio molecular dynamics (AIMD) simulations confirm this picture. In qualitative agreement with experimental data, we find that at the low temperature T = 70 K35 there is a sizable population of both isomers: 30% for isomer A and 70% for isomer B. An increase in temperature results in a sharp decline of the isomer B population. At T = 140 K the population of isomer B is reduced to 30%, at T = 210 K it is 10%, and at T = 300 K the
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/jacs.7b05964. Experimental methods; Theoretical details; Spectra deconvolution details; Optimized structures of the two isomers; Calculated proton affinities and mutual interaction energies; Density of states and TD-DFT calculation analyses; NPA charge distributions and HOMOs; Spin densities and spin populations; AIMD trajectory analyses; Transition state calculations; Absolute energies and coordinates of optimized structures (PDF)
■
AUTHOR INFORMATION
Corresponding Authors
*xuebin.wang@pnnl.gov *marat.valiev@pnnl.gov ORCID
Gao-Lei Hou: 0000-0003-1196-2777 Xue-Bin Wang: 0000-0001-8326-1780 Marat Valiev: 0000-0001-6127-1988 Notes
The authors declare no competing financial interest. 11323
DOI: 10.1021/jacs.7b05964 J. Am. Chem. Soc. 2017, 139, 11321−11324
Communication
Journal of the American Chemical Society
■
(31) Cui, X.; Cai, W.; Shao, X. RSC Adv. 2016, 6, 105729. (32) Kim, E. H.; Bradforth, S. E.; Arnold, D. W.; Metz, R. B.; Neumark, D. M. J. Chem. Phys. 1995, 103, 7801. (33) Yacovitch, T. I.; Heine, N.; Brieger, C.; Wende, T.; Hock, C.; Neumark, D. M.; Asmis, K. R. J. Chem. Phys. 2012, 136, 241102. (34) Weaver, A.; Arnold, D. W.; Bradforth, S. E.; Neumark, D. M. J. Chem. Phys. 1991, 94, 1740. (35) The low temperature NIPES experiments were conducted with the cryogenic trap temperature set at 20 K. The actual temperature of the molecular clusters should be higher than the setting temperature by ∼30−50 K depending on the clusters. (36) Shokri, A.; Wang, Y.; O’Doherty, G. A.; Wang, X.-B.; Kass, S. R. J. Am. Chem. Soc. 2013, 135, 17919. (37) Shokri, A.; Schmidt, J.; Wang, X.-B.; Kass, S. R. J. Am. Chem. Soc. 2012, 134, 2094.
ACKNOWLEDGMENTS This work was supported by the U.S. Department of Energy (DOE), Office of Science, Office of Basic Energy Sciences, Division of Chemical Sciences, Geosciences, and Biosciences, and performed using EMSL, a national scientific user facility sponsored by DOE’s Office of Biological and Environmental Research and located at Pacific Northwest National Laboratory, which is operated by Battelle Memorial Institute for the DOE. Discussions with S. M. Kathmann are greatly appreciated.
■
REFERENCES
(1) Blake, R. S.; Monks, P. S.; Ellis, A. M. Chem. Rev. 2009, 109, 861. (2) Weinberg, D. R.; Gagliardi, C. J.; Hull, J. F.; Murphy, C. F.; Kent, C. A.; Westlake, B. C.; Paul, A.; Ess, D. H.; McCafferty, D. G.; Meyer, T. J. Chem. Rev. 2012, 112, 4016. (3) Silverman, D. N.; McKenna, R. Acc. Chem. Res. 2007, 40, 669. (4) Ferguson, S. J. Curr. Biol. 2000, 10, R627. (5) Wolke, C. T.; Fournier, J. A.; Dzugan, L. C.; Fagiani, M. R.; Odbadrakh, T. T.; Knorke, H.; Jordan, K. D.; McCoy, A. B.; Asmis, K. R.; Johnson, M. A. Science 2016, 354, 1131. (6) Whiting, A. K.; Peticolas, W. L. Biochemistry 1994, 33, 552. (7) Childs, W.; Boxer, S. G. Biochemistry 2010, 49, 2725. (8) Hwang, S.; Shao, Q.; Williams, H.; Hilty, C.; Gao, Y. Q. J. Phys. Chem. B 2011, 115, 6653. (9) Bell, R. P. The Proton in Chemistry, 2nd ed.; Springer US: Fletcher & Son Ltd.: Norwich, 1973. (10) Ishikita, H.; Saito, K. J. R. Soc. Interface 2013, 11, 1. (11) Garczarek, F.; Gerwert, K. Nature 2006, 439, 109. (12) Freier, E.; Wolf, S.; Gerwert, K. Proc. Natl. Acad. Sci. U. S. A. 2011, 108, 11435. (13) Eustis, S. N.; Radisic, D.; Bowen, K. H.; Bachorz, R. A.; Haranczyk, M.; Schenter, G. K.; Gutowski, M. Science 2008, 319, 936. (14) Graham, J. D.; Buytendyk, A. M.; Wang, D.; Bowen, K. H.; Collins, K. D. Biochemistry 2014, 53, 344. (15) Brönsted, J. N. Rec. Trav. Chim. 1923, 42, 718. (16) Johnson, K. E.; Pagni, R. M.; Bartmess, J. Monatsh. Chem. 2007, 138, 1077. (17) Elsasser, B.; Valiev, M.; Weare, J. H. J. Am. Chem. Soc. 2009, 131, 3869. (18) Fridgen, T. D. J. Phys. Chem. A 2006, 110, 6122. (19) Gardenier, G. H.; Roscioli, J. R.; Johnson, M. A. J. Phys. Chem. A 2008, 112, 12022. (20) Alata, I.; Broquier, M.; Dedonder-Lardeux, C.; Jouvet, C.; Kim, M.; Sohn, W. Y.; Kim, S.-s.; Kang, H.; Schütz, M.; Patzer, A.; Dopfer, O. J. Chem. Phys. 2011, 134, 074307. (21) Bing, D.; Hamashima, T.; Tsai, C.-W.; Fujii, A.; Kuo, J.-L. Chem. Phys. 2013, 421, 1. (22) Cheng, T. C.; Bandyopadhyay, B.; Mosley, J. D.; Duncan, M. A. J. Am. Chem. Soc. 2012, 134, 13046. (23) Beletskiy, E. V.; Wang, X.-B.; Kass, S. R. J. Phys. Chem. A 2016, 120, 8309. (24) Hou, G.-L.; Lin, W.; Deng, S. H. M.; Zhang, J.; Zheng, W.-J.; Paesani, F.; Wang, X.-B. J. Phys. Chem. Lett. 2013, 4, 779. (25) Wang, X.-B.; Wang, L.-S. Rev. Sci. Instrum. 2008, 79, 073108. (26) Valiev, M.; Bylaska, E. J.; Govind, N.; Kowalski, K.; Straatsma, T. P.; Van Dam, H. J. J.; Wang, D.; Nieplocha, J.; Apra, E.; Windus, T. L.; de Jong, W. A. Comput. Phys. Commun. 2010, 181, 1477. (27) Linstrom, P. J., Mallard, W. J., Eds. NIST Chemistry WebBook, NIST Standard Reference Database Number 69, National Institute of Standards and Technology, Gaithersburg MD, 20899. http:// webbook.nist.gov/chemistry/ (retrieved March 10, 2017). (28) Wang, X.-B.; Woo, H.-K.; Wang, L.-S.; Minofar, B.; Jungwirth, P. J. Phys. Chem. A 2006, 110, 5047. (29) Kim, J. B.; Wenthold, P. G.; Lineberger, W. C. J. Chem. Phys. 1998, 108, 830. (30) Chen, B.; Hrovat, D. A.; West, R.; Deng, S. H.; Wang, X. B.; Borden, W. T. J. Am. Chem. Soc. 2014, 136, 12345. 11324
DOI: 10.1021/jacs.7b05964 J. Am. Chem. Soc. 2017, 139, 11321−11324