Article pubs.acs.org/JPCA
Combinations of Aromatic and Aliphatic Radiolysis Jay A. LaVerne*,†,‡ and Jennifer Dowling-Medley† †
Radiation Laboratory and ‡Department of Physics University of Notre Dame, Notre Dame, Indiana 46556, United States ABSTRACT: The production of H2 in the radiolysis of benzene, methylbenzene (toluene), ethylbenzene, butylbenzene, and hexylbenzene with γ-rays, 2−10 MeV protons, 5−20 MeV helium ions, and 10−30 MeV carbon ions is used as a probe of the overall radiation sensitivity and to determine the relative contributions of aromatic and aliphatic entities in mixed hydrocarbons. The addition of an aliphatic side chain with progressively from one to six carbon lengths to benzene increases the H2 yield with γ-rays, but the yield seems to reach a plateau far below that found from a simple aliphatic such as cyclohexane. There is a large increase in H2 with LET (linear energy transfer) for all of the substituted benzenes, which indicates that the main process for H2 formation is a second-order process and dominated by the aromatic entity. The addition of a small amount of benzene to cyclohexane can lower the H2 yield from the value expected from a simple mixture law. A 50:50% volume mixture of benzene−cyclohexane has essentially the same H2 yield as cyclohexylbenzene at a wide variation in LET, suggesting that intermolecular energy transfer is as efficient as intramolecular energy transfer.
■
INTRODUCTION Experiments have long shown that the sensitivity of aliphatic compounds to ionizing radiation is very different from that of aromatic compounds,1−5 but the interactions occurring in mixtures of these two classes of organic compounds is not at all understood. Similar uncertainty exists in our knowledge of the radiolytic decomposition of hydrocarbons containing both an aromatic and an aliphatic entity. The dominant radiationinduced process in the initial decomposition of a hydrocarbon in the condensed phase is breakage of the C−C or C−H bonds. Carbon-centered radicals can then go on to make a variety of stable products by combination, disproportionation, and other types of reactions. H· atoms tend to produce H2 by abstraction or addition reactions. Measurement of H2 is much more straightforward than the determination of the many carbon containing products, so the yield of H2 can be taken as a good indicator of the relative overall decomposition of the hydrocarbon by radiolysis. For instance, H2 yields with simple liquid hydrocarbons such as hexane and cyclohexane irradiated with γrays are about five to six molecules/100 eV of total energy deposition.4 Yields of this magnitude represent the maximum values for any product that are not due to a chain process. Aromatic compounds such as benzene and its derivatives have yields of H2 at 0.04 molecules/100 eV, showing their relative inertness to radiation.5 No one really understands why the yields of H2 are so low with aromatic compounds, but more information on the early energy-transfer processes is of fundamental importance in elucidation of the overall mechanism.6 Aromatic and aliphatic compounds have such a different response to radiation that an obvious point of interest is what © 2015 American Chemical Society
happens in mixtures of these compounds or when a molecule has both aromatic and aliphatic components. Even the earliest studies on the radiolysis of mixtures of aliphatic and aromatic compounds found that the yields of H2 were much lower than expected from the law of mixtures, which assumes that the response of each component of a mixture is independent of the others. Energy deposition is proportional to the electron density fraction of each component of the medium, and so the production of products should be based on this same fraction.3,7−9 The addition of an aliphatic side chain to benzene increases the H2 yield only slightly, which again shows the overwhelming influence of the aromatic entity on H 2 yields.10−12 Such a phenomenon has been called many things including the “protective effect” and can be used to lower the overall sensitivity of a medium to radiolysis. More knowledge on the mechanisms involved will allow the construction of molecules designed for a specific radiation response. Another interesting facet of the radiolysis of aromatic and aliphatic compounds is the relative response to track structure. With increasing linear energy transfer (LET, equal to the stopping power, −dE/dx) the density of energy deposition becomes greater about the particle path. This increase in local energy density leads to higher concentrations of reactive species than would be obtained using conventional fast electron or γradiolysis. Yields of products that are formed due to secondorder reactions will be enhanced with increasing LET, while those due to first order processes will be unaffected. Aromatic Received: July 31, 2015 Revised: September 16, 2015 Published: September 21, 2015 10125
DOI: 10.1021/acs.jpca.5b07414 J. Phys. Chem. A 2015, 119, 10125−10129
Article
The Journal of Physical Chemistry A
determined using an inline technique where the sample cell is connected by a four-way valve to the gas chromatograph (GC), which was an SRI 8610 equipped with a thermal conductivity detector. The column was a 6.4 mm diameter 13X molecular sieve 3 m long, maintained at 40 °C. The sample was purged with argon for 10 min at room temperature, and the cell was sealed with a four-way valve during the radiolysis. Following the irradiation, the sample cell was opened to the column using the four-way valve. Calibration was performed by injection of pure H2 under normal conditions. Errors in gas analysis are estimated at 10%.
compounds that are normally thought to be radiation inert are found to have a relatively high yield of H2 with radiation of high LET.12 Phase does not seem to affect the outcome and even polymers, and resins are found to have an increasing yield of H2 with increasing LET.12−14 This work examines the production of H2 in various combinations of aromatic and aliphatic compounds to gain more information on the early processes involved. Benzenes with different lengths of aliphatic side chains are examined to determine the effects of intramolecular processes. Mixtures of benzene and cyclohexane are compared with the results of cyclohexylbenzene to examine the relative competition between intra- and intermolecular processes. Variation of H2 yields with LET is used to give more information on the overall reaction order of these processes and to highlight common similarities. Comparison of the high LET results with that from γ-radiolysis is used to probe overall mechanisms for the formation of H2.
■
RESULTS AND DISCUSSION Radiolytic decomposition of organic compounds can be difficult to unravel because of the many products possible; however, for every molecule of the medium decomposed to make H2 there is a corresponding formation of a carboncontaining compound. The prototypical aromatic benzene has a yield of H2 of 0.04 molecules/100 eV, while the yield of H2 with a typical aliphatic compound like cyclohexane is several orders of magnitude higher at 5.9 molecules/100 eV.5 One can then question what the yield of H2 will be when both aromatic and aliphatic components make up a particular molecule. One simple approach to this enquiry is to add various length aliphatic side chains to a benzene ring. Yields of H2 are shown in Figure 1 as functions of the track average LET for
■
EXPERIMENTAL SECTION Benzene (99.5%), cyclohexane (99%), toluene (99.5%), ethylbenzene (99%), butylbenzene (99%), hexylbenzene (98%), and cyclohexylbenzene (97%) were obtained from Alfa Aesar in the highest purity available. All compounds were used as received. Irradiations with γ-rays were carried out using a selfcontained Shepherd 109-68R 60Co γ-source in the Radiation Laboratory at the University of Notre Dame. The dose rate was nominally about 180 Gy/min, as estimated by the Fricke dosimeter.15 Heavy ion irradiations were performed using the Tandem FN Van de Graaff facility of the Notre Dame Science Laboratory in the University of Notre Dame Physics Department. The ions used in these experiments were: 2, 5, 10, and 15 MeV 1H+; 5, 10, 15, and 20 MeV 4He2+; and 10, 20, and 30 MeV 12C6+ ions. These energies were incident to the sample with energy loss to windows determined using standard stopping power compilations.16 Irradiation methods and ion characteristics have been previously described.17 Absolute dosimetry techniques were used by combining ion energy with integrated beam currents. Ion beam currents were typically 1.5 nA charge current. Excitation with 185 nm UV was obtained by using a lowpressure mercury lamp with a high-purity quartz window from Atlantic Ultraviolet. The sample was purged with ultra-highpurity argon before irradiation and continuously stirred throughout the irradiation. Actinometry was performed in the same configuration as used with the hydrocarbons and consisted of an aqueous solution of ethanol.18 Photolytic quantum yields were determined by dividing the amount of products formed by the number of photons absorbed. Because all photon energy is absorbed by the media, the absolute photolytic yields were determined by dividing the quantum yields by the energy per photon and converted to the standard radiation chemical unit. Radiolysis of all compounds was performed under an inert atmosphere of ultra-high-purity (99.999%) argon. Sample cells for γ-radiolysis and UV radiolysis were made from quartz cuvettes containing an entrance and exit port for inline connection to the gas chromatograph. The volume of these cells was typically about 4 mL. Heavy ion radiolysis cells were made of Pyrex with a beam entrance window of nominally 4−6 mg/cm2 mica epoxied to the front and a magnetic stir bar in the rear; the volume was about 2 mL. Molecular hydrogen was
Figure 1. Yields of H2 from benzene, methylbenzene (toluene), ethylbenzene, butylbenzene, hexylbenzene, and cyclohexane14 as a function of track average LET for 1H (▲), 4He (●), and 12C (■) ions. The dashed lines show the low LET limiting yields obtained with γrays.
methylbenzene (toluene), ethylbenzene, butylbenzene, and hexylbenzene along with previous results for benzene and cyclohexane.14 The yields are expressed in the traditional radiation chemical unit of molecules/100 eV (1 molecule/100 eV = 0.104 μmol J−1). Also shown are the results with γ-rays for comparison. Examination of the results with γ-rays shows an increase in the yield of H2 with increasing length of the aliphatic side chain as one would expect; however, the results are not linear with the size of the aliphatic side chain. The H2 yields appear to approach a saturation of H2 production with a molecule only slightly larger than hexylbenzene. The yield of H2 for the latter is 0.43 molecules, which is still considerably lower than the value of 5.9 molecules/100 eV found with cyclohexane. Clearly, the aliphatic to aromatic fraction has an effect on the yield of H2, but the presence of an aromatic component overwhelms the overall response of the molecule. 10126
DOI: 10.1021/acs.jpca.5b07414 J. Phys. Chem. A 2015, 119, 10125−10129
Article
The Journal of Physical Chemistry A
LET for all aromatics suggests that a similar mechanism is responsible for the significant H2 production at high LET. One can see that the yields of H2 with the butylbenzene and hexylbenzene are nearly that found with cyclohexane in carbon ion radiolysis. Clearly, aromatic compounds can readily decompose with certain types of radiation. All of the various length side-chain benzene compounds are dominated by the response of the aromatic part of the molecule at high LET. Energy deposited by the ionization radiation must quickly localize on the aromatic component of these benzene-based molecules. Variation of the aliphatic to aromatic fraction of the medium can also be attained by a simple mixture of cyclohexane and benzene. The results for various mixtures of benzene and cyclohexane with γ-rays, 5 MeV He ions, and 185 nm UV photons are shown in Figure 2, and agreement with previously
Variation in track structure by changing the LET of the incident radiation is a convenient method for observing the overall order of processes leading to product formation. Second-order processes will lead to an increase in H2 yields with increasing LET, whereas first-order processes will not be affected by LET. The LET dependence of cyclohexane as shown in Figure 1 is nearly independent of LET. The overall decomposition and reaction of most aliphatic hydrocarbons to give H2 can be summarized by the following reactions. RH 2* → RH· + H· ·
H + RH 2 → H 2 + RH ·
·
H + H → H2
(1) ·
(2) (3)
Reaction 3 is a second-order process and should lead to an observed increase in H2 yield with increasing LET; however, reaction 2 is a pseudo-first-order abstraction process that dominates the production of H2 in this system and so the overall response to LET is minimal. The relative response of these two reactions in γ-radiolysis is not surprising. An H· atom upon formation will be exposed to about 10 M of parent molecules in a typical aliphatic system, and it will quickly abstract an H· atom from a neighbor molecule. An H· atom will have to diffuse through the medium to find another H· atom, and with low LET radiation it is highly likely to encounter a medium molecule before another H· atom. Reaction 3 takes two H· atoms to produce H2 so it is actually less efficient than reaction 2, which only needs one. The competition between reactions 2 and 3 can be seen as a slight decrease in the H2 yield at moderate LET. All of the aromatic compounds of Figure 1 show a large increase in H2 yields with increasing LET. Such a dependence on LET indicates that second-order processes are responsible for H2 formation in all cases. The near-similarity in the LET response strongly suggests that the mechanism is nearly the same for all of these simple aromatic liquids. The exact mechanism is not known, but it is generally expected to be described by the following scheme. RH* → R· + H·
(4)
H· + RH → RH 2·
(5)
H· + H· → H 2
(6)
R* + R* → H 2
(7)
Figure 2. Yields of H2 as a function of the percentage of benzene in cyclohexane for 185 nm UV, γ-ray, and 5 MeV 4He radiolysis along with previous γ-ray results.8 The dashed line shows the expected result using the mixture law.
published results for γ-rays is good.8 According to the mixture law the response of each component of a mixture should be independent of the others and the production of H2 in simple mixtures of cyclohexane and benzene should follow the dashed line of Figure 2. All of the radiations show the expected trend of the H2 yield decreasing with increasing benzene fraction; however, the variation of H2 yield with benzene fraction is not linear. There is a sharp decrease in H2 production with only a very small addition of benzene to cyclohexane. Radiation response varies slightly with all three types of radiation, but the same overall trend is noted. UV radiolysis is only exciting to the lower level energy states of the molecules, while the ionizing radiations are populating all electronic energy levels. Benzene has lower energy levels than cyclohexane, and much of the deposited energy can be expected to transfer to the benzene.23 Previous results on the formation of H2 with variation in the level of excitation by UV photolysis suggest that H2 mainly is produced by higher energy excited states, and the low yields of H2 observed here at 185 nm are not too surprising; however, the little H2 that is produced by UV is strongly affected by the presences of the aromatic benzene. Photolysis studies at 160 nm have shown that benzene quenches the emission of cyclohexane and the present results show a decreased H2 yield so possibly the same energy states are involved.24 The results for all three types of radiation show that in a mixture of
Reaction 6 is a second-order process that could lead to the formation of H2; however, previous studies have shown that the simple addition of I2, an efficient H· atom scavenger, does not lead to a decrease in H2 yields in benzene.19 Such a result suggests that H· atoms are not a source of H2, and another second-order process is responsible for H2 formation in aromatic systems. Again, the results with γ-rays are not too surprising as aromatics are known to be good radical scavengers and H· atoms have a high probability of adding to a neighboring parent molecule, reaction 5, following their formation. Photochemical and fluorescence studies seem to suggest that H2 formation in aromatics is due to second-order combination reactions of highly excited states, reaction 7.6,20,21 The exact nature of the excited states involved in reaction 7 and the associated products are not yet known. Ring opening could be occurring, but the most likely scenario is complete ring destruction to give acetylene.22 The common dependence on 10127
DOI: 10.1021/acs.jpca.5b07414 J. Phys. Chem. A 2015, 119, 10125−10129
Article
The Journal of Physical Chemistry A
single molecule is that intermolecular energy transfer is essentially as efficient as intramolecular energy transfer. With increasing LET the yields of H2 increase with benzene, cyclohexylbenzene, and the 50:50% volume benzene−cyclohexane mixture. All of these systems seem to be approaching a common H2 yield that is observed with neat cyclohexane. Although the mixture law fails miserably with γ-rays, it becomes more predictive at high LET for both the benzene−cyclohexane mixture and cyclohexylbenzene. At high LET the benzene is essentially behaving radiolytically like an aliphatic compound with respect to H2 production and its overall radiation sensitivity. There is no longer an “aromatic sink”, so the presence of the aromatic entity has little effect. Any redistribution of energy within a mixture or a single molecule does not seem to matter, and one cannot tell from these experiments if energy transfer occurs under these conditions. The actual mechanism for H2 production in aromatic and aliphatic mixtures or in mixed component molecules is not yet resolved in this work, but some speculation on the possible pathways can be made. Ionization yields to give the molecular cation and free electron in each component of the benzene− cyclohexane mixture are probably similar because the initial energy loss is more or less equally distributed to each constituent. Ionization is an extremely fast process, so there is little redistribution of energy before this process occurs. Transfer of the positive charge from the cyclohexane radical cations to the benzene is possible. This process would be followed by neutralization reactions to give the highly excited states of benzene responsible for H2 production. Neutralization reactions to give highly excited states of each constituent can also occur without transfer of positive charge to be followed by energy transfer from the highly excited states of cyclohexane to benzene. Either of these processes will essentially transfer energy from cyclohexane to benzene, leading to products that are derived primarily from the benzene decomposition. Transfer of the charge or excited state energy must be incredibly efficient in benzene−cyclohexane mixtures to give similar results to that of the cyclohexylbenzene.
aromatic and aliphatic hydrocarbons the aromatic entity has a strong effect on the overall radiation response. The mixtures of benzene and cyclohexane were further explored at an equal volume fraction of each as a function of LET to obtain additional information on the processes occurring in these mixtures. Figure 3 shows the radiation
Figure 3. Yields of H2 from benzene,14 cyclohexane,14 cyclohexylbenzene, and an equal volume mixture of cyclohexane and benzene (56.3 mol percent benzene) as a function of track average LET for 1H (▲), 4He (●), and 12C (■) ions. The dashed lines show the low LET-limiting yields obtained with γ-rays. The solid line is the predicted H2 yield from the mixture law; see the text.
response for the formation of H2 from both neat benzene and cyclohexane, and the large differences in radiation response are readily observed. Energy deposition by a high-energy charged particle is due to Coulombic interactions of the incident particle with electrons of the medium molecules. The fraction of energy loss to each component in a mixture is equal to their electron density fraction. Equal volumes of benzene and cyclohexane have electron densities of 51.1 and 48.9%, respectively, or roughly equal electron fractions. Approximately half of the initial energy deposited will go into each component of this mixture, and one would expect yields equivalent to the solid line given in Figure 3. This response was obtained by a simple mixture law that predicts the total yield of H2 is equal to the sum of the contribution of each component, which is determined by the electron fraction times the H2 yield for that component. Mixtures of benzene and cyclohexane at 50:50 volume percent were irradiated with radiation of different of LET, and the results are shown in Figure 3. The experimental results with γ-rays are almost an order of magnitude lower than that expected from the simple mixture law. Energy initially deposited in the cyclohexane is being transferred to the benzene, and the H2 produced in the mixture more closely matches that expected from neat benzene rather than the value expected from an ideal mixture. A further check of the transfer of energy from an aliphatic component to an aromatic one can be probed by examination of the cyclohexylbenzene molecule. Within this molecule the distribution of electrons is 48.1 to 51.9% for the cyclohexyl and phenyl components, respectively, or essentially equally divided between the two entities. The results given in Figure 3 show that the H2 yields are very similar to that observed with 50:50% volume mixtures. These data also show that energy is being transferred from the aliphatic to the aromatic entity. The startling observation between radiolysis of the mixture and the
■
CONCLUSIONS The addition of an aliphatic side chain with progressively from one to six carbon lengths to benzene increases the H2 yield. Although the H2 yield increases with side chain length, the yield seems to plateau at a value far below that found from a simple aliphatic such as cyclohexane alone. There is a large increase in H2 with LET for all of the substituted benzenes, which indicates that the main process for H2 formation follows second-order kinetics. A similar trend for H2 production for all of the substituted benzenes to that of benzene itself suggests a common mechanism, which probably involves highly excitedstate combination processes. Aromatic compounds are very sensitive to radiation at high LET as opposed to their response with low LET γ-rays. The production of H2 in the radiolysis of mixtures of simple aromatic and aliphatic hydrocarbons is highly influenced by the radiation response of the aromatic component. Hydrogen production does not follow the simple mixture law where the total yield is dependent on the electron fraction of each constituent and its corresponding H2 yield. The addition of a small amount of benzene can dramatically lower the H2 yield with γ-rays in a mixture of cyclohexane-benzene from the expected value. Energy loss in a 50:50% volume mixture of benzene−cyclohexane is essentially equally divided between the 10128
DOI: 10.1021/acs.jpca.5b07414 J. Phys. Chem. A 2015, 119, 10125−10129
Article
The Journal of Physical Chemistry A
(17) Enomoto, K.; LaVerne, J. A.; Araos, M. S. Heavy Ion Radiolysis of Liquid Pyridine. J. Phys. Chem. A 2007, 111, 9−15. (18) Sokolov, U.; Stein, G. Photolysis of Liquid Water at 1849 A. J. Chem. Phys. 1966, 44, 3329. (19) Enomoto, K.; LaVerne, J. A.; Pimblott, S. A. Products of the Triplet Excited State Produced in the Radiolysis of Liquid Benzene. J. Phys. Chem. A 2006, 110, 4124−4130. (20) Burns, W. G. Decomposition of Aromatic Substances by Different Kinds of Radiation 1. Proton Irradiation of Benzene. Trans. Faraday Soc. 1962, 58, 961. (21) LaVerne, J. A. Fluorescence in the Heavy Ion Radiolysis of Benzene. J. Phys. Chem. 1996, 100, 18757−18763. (22) Burns, W. G. Decomposition of Aromatic Substances by Different Kinds of Radiation. Trans. Faraday Soc. 1962, 58, 961−970. (23) Doering, J. P. Electronic Energy Levels of Benzene Below 7 eV. J. Chem. Phys. 1977, 67, 4065−4070. (24) Johnston, D. B.; Lipsky, S. Energy Transfer from Cyclohexane to Benzene in Their Irradiated Mixtures. Radiat. Phys. Chem. 1991, 38, 51−60.
two constituents, and yet energy is rapidly transferred from the cyclohexane to the benzene. Similar H2 yields are found for 50:50 volume mixtures of benzene and cyclohexane as with cyclohexylbenzene. Intermolecular energy transfer seems to be as efficient as intramolecular energy transfer.
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS
■
REFERENCES
We thank Prof. Michael Wiescher for making available the facilities of the Notre Dame Nuclear Structure Laboratory, which is supported by the U.S. National Science Foundation. The work reported here was supported by the Division of Chemical Sciences, Geosciences and Biosciences, Basic Energy Sciences, Office of Science, United States Department of Energy through grant number DE-FC02-04ER15533. This contribution is NDRL-5079 from the Notre Dame Radiation Laboratory.
(1) Schoepfle, C. S.; Fellows, C. H. Gaseous Products from Action of Cathode Rays on Hydrocarbons. Ind. Eng. Chem. 1931, 23, 1396− 1398. (2) Burton, M. Radiation Chemistry. J. Phys. Colloid Chem. 1947, 51, 611−625. (3) Gordon, S.; Burton, M. Radiation Chemistry of Pure Organic Compounds - Benzene and Benzene-D6. Discuss. Faraday Soc. 1952, 12, 88−98. (4) Gyorgy, I. Aliphatic Alkanes. In Radiation Chemistry of Hydrocarbons; Foldiak, G., Ed.; Elsevier: Amsterdam, 1981; pp 61− 176. (5) Roder, M. Aromatic Hydrocarbons. In Radiation Chemistry of Hydrocarbons; Foldiak, G., Ed.; Elsevier: Amsterdam, 1981; pp 393− 468. (6) Baidak, A.; Badali, M.; LaVerne, J. A. Role of the Low-energy Excited States in the Radiolysis of Aromatic Liquids. J. Phys. Chem. A 2011, 115, 7418−7427. (7) Manion, J. P.; Burton, M. Radiolysis of Hydrocarbon Mixtures. J. Phys. Chem. 1952, 56, 560−569. (8) Burton, M.; Patrick, W. N. Radiation Chemistry of Mixtures Cyclohexane and Benzene-D6. J. Phys. Chem. 1954, 58, 421−423. (9) Foldiak, G.; Roder, M.; Wojnarovits, L. Influence of Ionization Density on Radiolysis of Cyclohexane-Benzene Mixtures. Radiat. Phys. Chem. 1992, 39, 77−80. (10) Burton, M.; Gordon, S.; Hentz, R. R. Effect of Ring on Radaition Chemistry of Alkyl-Substituted Benzenes. J. Chim. Phys. 1951, 48, 190−194. (11) Hentz, R. R.; Burton, M. Studies in Photochemistry and Radiation Chemistry of Toluene, Mesitylene and Ethylbenzene. J. Am. Chem. Soc. 1951, 73, 532−536. (12) La Verne, J. A.; Baidak, A. Track Effects in the Radiolysis of Aromatic Liquids. Radiat. Phys. Chem. 2012, 81, 1287−1290. (13) Baidak, A.; LaVerne, J. A. Radiation-Induced Decomposition of Anion Exchange Resins. J. Nucl. Mater. 2010, 407, 211−219. (14) LaVerne, J. A.; Chang, Z.; Araos, M. S. Heavy Ion Radiolysis of Organic Materials. Radiat. Phys. Chem. 2001, 60, 253−257. (15) LaVerne, J. A.; Schuler, R. H. Radiation Chemical Studies with Heavy-Ions - Oxidation of Ferrous Ion in the Fricke Dosimeter. J. Phys. Chem. 1987, 91, 5770−5776. (16) Ziegler, J. F.; Biersack, J. P.; Littmark, U. The Stopping and Range of Ions in Solids; Pergamon: New York, 1985. 10129
DOI: 10.1021/acs.jpca.5b07414 J. Phys. Chem. A 2015, 119, 10125−10129
