Article pubs.acs.org/JPCA
Beryllium Fluoride Exchange Rate Accelerated by Mg2+ as Discovered by 19F NMR Yixiang Liu,†,§ Xi-an Mao,‡ Maili Liu,† and Ling Jiang*,† †
Key Laboratory of Magnetic Resonance in Biological Systems, National Center for Magnetic Resonance in Wuhan, State Key Laboratory of Magnetic Resonance and Atomic and Molecular Physics, Wuhan Institute of Physics and Mathematics, Chinese Academy of Sciences, Wuhan 430071, China ‡ Department of Pharmacology, Case Western Reserve University School of Medicine, Cleveland, Ohio 44106, United States § Graduate University of Chinese Academy of Sciences, 19A Yuquanlu, Beijing 100049, China S Supporting Information *
ABSTRACT: Beryllium fluoride is widely used as a phosphoryl analogue in macromolecular studies, which are not only fluoride-sensitive but also magnesium-dependent. The beryllium fluorides are a mixture of different species including BeF3− and BeF42− exchanging under thermodynamic equilibrium in neutral aqueous solutions. In the cases of mimicking phosphate group transfer, both beryllium fluoride and the magnesium ion are generally needed. However, the impact of magnesium on the bioactivity of beryllium fluoride is not clear. We have found by 19F NMR spectroscopy that Mg2+ can severely affect the chemical exchange kinetics between BeF3− and BeF42−. When the F− concentration is relatively low, the presence of 10.0 mM Mg2+ can accelerate the exchange rate 3−4 fold. However, when the F− concentration is relatively high, the Mg2+ effect on the chemical exchange vanishes. On the basis of these findings, we proposed a possible mechanism that BeF42− and Mg2+ form an ion pair that affects the distribution of beryllium fluoride species and thus the activity in the solution.
■
that the magnesium ion has significant effects on the 19F NMR spectral intensities and chemical shifts of different beryllium fluoride species. In the presence of 10.0 mM magnesium(II), the optimal concentration ratio of F− to Be2+ to form BeF3− was changed from 6:1 to 10:1.11 In particular, the line widths of beryllium fluoride signals were significantly broadened after the addition of a magnesium ion. The kinetics of the Al3+/F− system has been extensively studied;12 however, no kinetic NMR studies of the Be2+/F− system have been performed, especially in the presence of magnesium. In this paper, we presented our investigations on the chemical exchange kinetics between BeF42− and BeF3− by 19F NMR spectroscopy. Titration spectra of beryllium fluorides were recorded as a function of the concentration of magnesium chloride. Chemical exchange rates between the two species were measured and compared at different ionic strengths. Our results suggested that in the solution containing 5.0 mM BeCl2, 50.0 mM NaF, and 10.0 mM MgCl2 at pH 7.0, an ion pair may form between the BeF42− anion and the Mg2+ cation.
INTRODUCTION The low atomic weight element beryllium has been found to be useful in biochemistry only since the late 1980s.1,2 Its fluoride complexes have been widely used in studying phosphorylation processes of a variety of NTPases. When G proteins are studied, for example, it is necessary to inhibit the activity of GTP from quickly converting to GDP. BeF3− is regarded as the best analogue to the phosphoryl group and forms a very stable GDP−BeF3− complex, which can be interpreted as an ideal model for mimicking GTP.1 The bovine F1-ATPase can be irreversibly inhibited by BeF3−.2 BeF2 was found to lock the UMP Kinase from Dictyostelium discoideum in its active conformation.3 The phosphoryl analogue model generated by beryllium fluoride is well accepted not only for NTPases but also for other proteins. In recent years, beryllium fluoride has facilitated numerous structural studies of proteins in their phosphorylated forms.4−6 Notably, the phosphorylation processes of most kinases or proteins are magnesium-dependent.7 After bound to the activated protein, beryllium fluoride coordinated with one or more magnesium ions is suspected to stabilize the transition state of the protein. Beryllium fluoride can form a variety of complexes, BeF42−, BeF3−, BeF2, and BeF+. These complexes, together with free fluoride, undergo chemical exchange in solution, as revealed by 19 F NMR.8−10 The contribution of individual beryllium fluoride species depends on the concentration ratio of the fluoride ion to the beryllium ion in the solution. Recently, we have found © 2014 American Chemical Society
■
EXPERIMENTAL SECTION Two series of beryllium fluoride samples were prepared from stock solutions as described in the previous paper.11 The first Received: September 23, 2014 Revised: December 12, 2014 Published: December 22, 2014 24
dx.doi.org/10.1021/jp509640d | J. Phys. Chem. A 2015, 119, 24−28
The Journal of Physical Chemistry A
Article
series consisted of nine samples with [NaF]/[BeCl2] fixed at 50.0:5.0 (mM/mM) and a variable concentration of MgCl2 of 0.0, 5.0, 10.0, 15.0, 20.0, 30.0, 60.0, 100.0, and 150.0 mM. The second series had a fixed [NaF]/[BeCl2] of 50.0:5.0 (mM/ mM) and varied [NaCl] of 0.0, 15.0, 30.0, 45.0, 60.0, 90.0, 180.0, 300.0, and 450.0 mM. The pH of all NMR samples was carefully adjusted to pH 7.0 to eliminate the dependence of 19F resonances on pH.13 19F NMR experiments were conducted at 298 K on a Bruker 600 MHz spectrometer equipped with a 5.0 mm CPQCI 1H/19F−13C/15N/D Z-GRD cryoprobe. Chemical shifts were referenced to the free fluoride signal in a solution only containing 5.0 mM BeCl2 and 50.0 mM NaF at pH 7.0 (F− at δ 0.0 ppm). Exchange rates between BeF3− and BeF42− were evaluated by two-dimensional exchange spectroscopy (2D EXSY) conducted on samples with three different [NaF]/[BeCl2] concentration ratios of 50.0 mM/5.0 mM; 100.0 mM/5.0 mM, and 300.0 mM/5.0 mM, respectively. 2D 19F-EXSY spectra were recorded with a NOESY pulse sequence, a data matrix of 2048 × 256, and four scans. The relaxation delay was 2.0 s. The spectral width was 5647 Hz in both dimensions. Mixing times were optimized based on 19F relaxation rate. To calculate the exchange rate, a first-order reaction was assumed for the exchanging process between two species, BeF3− and BeF42− k 34
BeF3− ⇌ BeF4 2 −
(1)
k43
Figure 1. 19F NMR spectra (564 MHz, 25 °C) of beryllium fluorides as a function of the concentration of MgCl2. In the solution, the [NaF]/[BeCl2] ratio was kept at 50.0:5.0 (mM/mM).
Equation 1 is the commonly accepted formula in NMR for chemical exchange. Because BeF3− is converted to BeF42− by accepting a fluoride ion, eq 1 should be written as eq 2 for discussing the mechanism of fluoride exchange ka
BeF3− + F− ⇌ BeF4 2 − kd
MgCl2 and became significant between 15.0 and 20.0 mM MgCl2. Although line broadening of BeF2 resonance became observable at about 30.0 mM MgCl2, the signal intensity increment was started between 5.0 and 10.0 mM MgCl2. At the MgCl2 concentration of 30.0 mM, a new and very broadened signal appeared at −37.0 ppm and can be assigned to magnesium fluoride complexes MgFx(H2O)6−x. The chemical shift value agrees with the reports in the literature.14−17 This MgFx signal kept growing as the concentration of MgCl2 was increased, and the BeFx signals became even more broadened, as can be seen from Figure 1. These results indicated that the presence of MgCl2 not only affects the equilibrium between different species of beryllium fluoride but also affects the line shapes of the resonances. The latter case implied that there is interaction between MgCl2 and BeFx and that the interaction affinity is in the order of BeF42− > BeF3− > BeF2. The line broadening effect of magnesium(II) observed in Figure 1, however, was not observed when magnesium chloride was replaced by sodium chloride, as shown in Figure 2, where the concentrations of 50.0 mM NaF and 5.0 mM BeCl2 were the same as those used in Figure 1 and the concentration range of NaCl was more than doubled as to match the ion strength and the chloride difference between NaCl and MgCl2. When MgCl2 was replaced by NaCl, the 19F NMR fine structures of BeFx remained unchanged. In contrast to MgCl2, the presence of a high concentration of NaCl was in favor of formatting the high fluoridated BeF42− but not the low fluoridated BeF2, as indicated by the relative intensity and chemical shift changes of the resonances in Figure 2. The results provided further evidence of the interaction between BeFx and MgCl2.
(2)
The kinetic rate constants in eq 2 are given in eq 3 d[BeF4 2 −] = ka[F−free][BeF3−] − kd[BeF4 2 −] dt
(3)
and there is a relationship between the chemical exchange rate constants in eqs 2 and 3 and the magnetization exchange rate constants in eq 1 ka[F−free] = k 34
kd = k43
(4)
The integrations of cross peaks collected in Topspin 2.1 were used to extract kinetic data. Exchange rates were derived by data fitting with MATLAB 7.0 and EXSYCALC.
■
RESULTS AND DISCUSSION Line Broadening Effect of Magnesium(II) on Beryllium Fluorides. The top spectrum in Figure 1 was recorded from the solution containing 5.0 mM BeCl2 and 50.0 mM NaF without MgCl2. Three well-resolved 19F signals of the beryllium fluoride complexes were observed, each showing a quartet due to the coupling to 9Be (spin I = 3/2). These signals were assigned to BeF42−, BeF3−, and BeF2, respectively, according to the reported chemical shifts.11 Introducing Mg2+ to the solution immediately changed the chemical shifts and the line width of 19 F resonances. BeF42− was the most sensitive species to the presence of Mg2+ because its fine coupling structure was hardly observable at 5.0 mM MgCl2, and the peaks nearly vanished at 20.0 mM MgCl2. The effect on BeF3− started at about 10.0 mM 25
dx.doi.org/10.1021/jp509640d | J. Phys. Chem. A 2015, 119, 24−28
The Journal of Physical Chemistry A
Article
Figure 3. Comparison of the line width of the 19F quartet of BeF42− and BeF3− in the absence (blue) and presence (red) of 10.0 mM Mg2+. In experiments, the concentration of BeCl2 was kept at 5.0 mM, and the concentration of NaF was changed from 50.0 to 300.0 mM. While introducing the Mg2+ cation into the solution seriously broadened the 19 F signal when [F−total] was 50.0 mM, as shown in (a) and (b), it did not cause appreciable change in the 19F line width when [F−total] was 300.0 mM, as shown in (c) and (d).
Figure 2. 19F NMR spectra (564 MHz, 25 °C) of beryllium fluorides as a function of the concentration of NaCl. In the solution, the concentration ratio [NaF]/[BeCl2] was kept at 50.0:5.0 (mM/mM).
The change in the line widths of the 19F NMR signals shown in Figure 1, however, was observed only when the fluoride to beryllium concentration ratio was relatively low (10:1). When the ratio reached 20:1 or 60:1, no appreciable change in line width was found (Figure 3). In addition, relative contents of different BeFx species were alerted, and BeF42− became the dominant one in the system at the higher [NaF]/[BeCl2] ratio. The results revealed that an excess concentration of fluorine enhances the distribution of BeF42− but suppresses the interaction between BeFx and MgCl2. Fluoride Exchange Rates between BeF42− and BeF3−. The line broadening effect observed in Figure 1 could be explained by the change of the fluoride exchange rate. Fluoride exchange between different beryllium fluoride complexes was observed as early as in the 1960s, and exchange networks were described in the 1970s.8,10,18 In the cited studies, exchange rate constants were measured through temperature-dependent line shape simulation. Here, we measured the exchange rates between BeF3− and BeF42− by using 2D EXSY spectroscopy. As shown in Figure 4, under the given conditions, four sets of peaks appeared in the EXSY spectrum, two diagonal peaks and two cross peaks. The BeF2 signal was very weak so that its exchange with other beryllium fluorides could be ignored. The results of exchange rate measurements are collected in Table 1, together with the equilibrium constants. In order to optimize the parameters of 2D EXSY experiments, 19F relaxation times of BeF3− and BeF42− were measured for the
Figure 4. 2D exchange spectrum of beryllium fluorides at a [NaF]/ [BeCl2] concentration ratio of 100.0 mM/5.0 mM with a mixing time of 140 ms in the absence of MgCl2.
selected samples (see the Supporting Information, Table S1). From Table 1, it can be seen that at the lower [NaF]/[BeCl2] concentration ratio (50.0 mM/5.0 mM), the presence of 10.0 mM MgCl2 increased the exchange rates by a factor about 4, while at the higher ratios (100.0 mM/5.0 mM and 300.0 mM/ 5.0 mM), the presence of MgCl2 had little effect on the exchange rates. The data in Table 1 explain well that the line broadening effect of magnesium(II) in Figure 3 was due to the exchange acceleration effect. At the lower [NaF]/[BeCl2] ratio (50.0 mM/5.0 mM), when kd gained a 4.7-fold increase by the presence of 10.0 mM Mg2+ (see data in the first row of Table 1), the associate kinetic rate constant ka gained only a 3.3-fold increase. The equilibrium 26
dx.doi.org/10.1021/jp509640d | J. Phys. Chem. A 2015, 119, 24−28
The Journal of Physical Chemistry A
Article
Table 1. Exchange Rates between BeF42− and BeF3− and the Equilibrium Constants Measured through 2D EXSY at 298 K without MgCl2 [NaF]/[BeCl2]
ka/M−1 s−1
50.0 mM/5.0 mM 50.0 mM/5.0 mMa 100.0 mM/5.0 mM 100.0 mM/5.0 mMa 300.0 mM/5.0 mM
× × × × ×
0.54 0.84 0.55 0.74 0.73
2
10 102 102 102 102
with MgCl2(10.0 mM)
kd/s−1 2.74 3.47 2.90 3.34 3.39
log K
ka/M−1 s−1
kd/s−1
log K
1.29 1.38 1.27 1.34 1.33
× × × × ×
12.54 12.70 2.75 3.16 3.15
1.15 1.32 1.13 1.27 1.24
1.78 2.66 0.37 0.59 0.55
2
10 102 102 102 102
a Experiments were repeated with ionic strength correction by adding NaCl so that the ionic strength was equal to the solution with [NaF]/[BeCl2] = 300.0 mM/5.0 mM.
Figure 5. Diagrams illustrating an ion pair of BeF42− and Mg2+ in (a) and the fluoride exchange between BeF42− and BeF3− in (b) and (c). In the diagrams, only three of six water molecules in the first coordination sphere of Mg2+ have been drawn. Dashed lines represent hydrogen bonds.
constant seemed to have been affected by Mg2+, with log K being changed from 1.29 to 1.15. The 11% decrease of the equilibrium constant was also found in the [NaF]/[BeCl2] sample at a concentration of 100.0 mM/5.0 mM. In the repeated experiment with ionic strength correction, it was found that both kd and ka increased in parallel and the equilibrium constant only dropped 4% (see data in the second row of Table 1; log K = 1.38 without the presence of Mg2+, and log K = 1.32 with the presence of Mg2+). Experiments at higher [NaF]/[BeCl2] ratio (100.0 mM/5.0 mM) showed the same results that the change of equilibrium constant dropped 5% after ionic strength correction. This is true because magnesium fluoride complexes were formed in the solution that decreased the fluoride concentration in the chemical equilibrium. Interpretation of the Accelerated Exchange Rates by the Ion Pair Model. Now, we need to answer the question why the Mg2+ ion catalyzes the fluoride exchange in some cases while it does not have any kinetic effect in some other cases. At this point, the mechanism of fluoride exchange should be discussed. The equilibrium in eq 2 is a simplified form of eq 5 BeF3(H 2O)− + F− ⇌ BeF4 2 − + H 2O
ion in BeF42− becomes activated (marked with a star) and leaves the position for the invading water molecule (Figure 5c). Then, the fluoride exchange between BeF42− and BeF3− is accomplished. The mechanism should apply to fluoride exchange without the presence of Mg2+. However, the formation of the BeF42−·Mg2+ ion pair accelerates the leaving of the negatively charged F− by exerting a positive electric force and thus increases the dissociation kinetic rate kd. The ion pair model proposed in Figure 5 has also been supported by the spectra shown in Figure 3. When the concentration of NaF was rather high (300.0 mM) while the concentration of BeCl2 was still kept at 5.0 mM, the F− ion started to enter into the first coordination sphere of Mg2+ to form a complex MgF(H2O)5+, which is demonstrated in Figure 1. As a result, the BeF42−·Mg2+ ion pair started to break. The higher the concentration of F − , the lower that the concentration of the ion pair should be. Finally, when all ion pairs were broken, magnesium(II) did not help the F− ion leave at all. This explains well the experimental observations shown in Figure 3c,d and Table 1. In solutions, ion pairs have been catalogued as doublesolvent-separated (2SIP), solvent-shared (SIP), and contact (CIP) ion pairs.24 The mechanism illustrated in Figure 5 suggests that the ion pair of BeF42− and Mg2+ is either 2SIP or SIP but not CIP because the leaving F− ion needs a pathway to escape into the solution and CIP does not provide such a pathway.
(5)
It is reasonably well established that hydrogen bonding is found between all of the components in the highly polar aqueous solutions. In this case, ion pair formation may be taken into consideration for interpreting the results discovered in this study. It is known that Mg2+ is a strong “structure-making ion”.19−23 Mg2+ usually strengthens the hydrogen bond of water, making the solution have higher viscosity. The Mg2+ cation and the BeF42− anion are very likely to have formed an ion pair, as depicted in Figure 5a. The ion pair complex might be a weak association, and the BeF42− might still be in the outside of the second coordination sphere of Mg2+. However, a water molecule can attack the beryllium atom at any time (Figure 5b). The exchange reaction follows a standard associative bimolecular nucleophilic substitution mechanism (SN2) at the tetrahedral coordinated central atom (Be2+). The water molecule serves as the nucleophile, while the bonded F−
■
CONCLUSION In summary, in the 19F NMR study of the beryllium fluoride in aqueous solutions, we have found that the presence of Mg2+ accelerates the fluoride exchange between BeF3− and BeF42− at low fluoride concentration while it has little effect when the total fluoride concentration is high. We have discussed the mechanism of fluoride exchange and indicated that an ion pairing interaction is the decisive factor for the exchange rate acceleration. Our data have demonstrated that in the solutions of the mixture of BeCl2, NaF, and MgCl2, an ion pair was 27
dx.doi.org/10.1021/jp509640d | J. Phys. Chem. A 2015, 119, 24−28
The Journal of Physical Chemistry A
Article
formed between BeF42− and Mg(H2O)62+ when [BeCl2] = 5.0 mM and [NaF] = 50.0 mM. The ion pair was broken when more fluoride ions were added to the solution because more fluoride ions would bind to Mg2+ and thus destroy the ion pair structure.
■
(13) Schmidt, M.; Schmidbaur, H. Ligand Redistribution Equilibria in Aqueous Fluoroberyllate Solutions. Z. Naturforsch., B: Chem. Sci. 1998, 53b, 1294−1300. (14) Baxter, N. J.; Olguin, L. F.; Goličnik, M.; Feng, G.; Hounslow, A. M.; Bermel, W.; Blackburn, G. M.; Hollfelder, F.; Waltho, J. P.; Williams, N. H. A Trojan Horse Transition State Analogue Generated by MgF3− Formation in an Enzyme Active Site. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 14732−14737. (15) Henry, G. D.; Maruta, S.; Ikebe, M.; Sykes, B. D. Observation of Multiple Myosin Subfragment 1-ADP-fluoroberyllate Complexes by Fluorine-19 NMR Spectroscopy. Biochemistry 1993, 32, 10451−10456. (16) Fovet, Y.; Gal, J.-Y. Formation Constants β2 of Calcium and Magnesium Fluorides at 25°C. Talanta 2000, 53, 617−626. (17) Shibata, N.; Sato, H.; Sakaki, S.; Sugita, Y. Theoretical Study of Magnesium Fluoride in Aqueous Solution. J. Phys. Chem. B 2011, 115, 10553−10559. (18) Lincoln, S.; Sandercock, A.; Stranks, D. A 19F Nuclear Magnetic Resonance Study of the Exchange of the Fluoro Ligands of the Tetrafluoroberyllium(II) Ion. Aust. J. Chem. 1977, 30, 271−279. (19) Holm, N. G. The Significance of Mg in Prebiotic Geochemistry. Geobiology 2012, 10, 269−279. (20) Bourret, R. B. Receiver Domain Structure and Function in Response Regulator Proteins. Curr. Opin. Microbiol. 2010, 13, 142− 149. (21) Marcus, Y. Effect of Ions on the Structure of Water: Structure Making and Breaking. Chem. Rev. 2009, 109, 1346−1370. (22) Gessner, R. V.; Quigley, G. J.; Wang, A. H. J.; Van der Marel, G. A.; Van Boom, J. H.; Rich, A. Structural Basis for Stabilization of ZDNA by Cobalt Hexaammine and Magnesium Cations. Biochemistry 1985, 24, 237−240. (23) Stirnemann, G.; Wernersson, E.; Jungwirth, P.; Laage, D. Mechanisms of Acceleration and Retardation of Water Dynamics by Ions. J. Am. Chem. Soc. 2013, 135, 11824−11831. (24) Buchner, R.; Chen, T.; Hefter, G. Complexity in “Simple” Electrolyte Solutions: Ion Pairing in MgSO4(aq). J. Phys. Chem. B 2004, 108, 2365−2375.
ASSOCIATED CONTENT
S Supporting Information *
T1 relaxation time of beryllium fluorides (Table S1). This material is available free of charge via the Internet at http:// pubs.acs.org.
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work was supported by grants from the National Basic Research Program of China (#2013CB910200) and the Natural Science Foundation of China (#21173257, #21120102038, and #21221064).
■
REFERENCES
(1) Bigay, J.; Deterre, P.; Pfister, C.; Chabre, M. Fluoride Complexes of Aluminium or Beryllium Act on G-Proteins as Reversibly Bound Analogues of the Gamma Phosphate of GTP. EMBO J. 1987, 6, 2907− 2913. (2) Kagawa, R.; Montgomery, M. G.; Braig, K.; Leslie, A. G.; Walker, J. E. The Structure of Bovine F1-ATPase Inhibited by ADP and Beryllium Fluoride. EMBO J. 2004, 23, 2734−2744. (3) Schlichting, I.; Reinstein, J. Structures of Active Conformations of UMP Kinase from Dictyostelium Discoideum Suggest Phosphoryl Transfer Is Associative. Biochemistry 1997, 36, 9290−9296. (4) Cho, H.; Wang, W.; Kim, R.; Yokota, H.; Damo, S.; Kim, S.-H.; Wemmer, D.; Kustu, S.; Yan, D. BeF3− Acts as a Phosphate Analog in Proteins Phosphorylated on Aspartate: Structure of a BeF3− Complex with Phosphoserine Phosphatase. Proc. Natl. Acad. Sci. U.S.A. 2001, 98, 8525−8530. (5) Hastings, C. A.; Lee, S.-Y.; Cho, H. S.; Yan, D.; Kustu, S.; Wemmer, D. E. High-Resolution Solution Structure of the Beryllofluoride-Activated NtrC Receiver Domain. Biochemistry 2003, 42, 9081−9090. (6) Wemmer, D. E.; Kern, D. Beryllofluoride Binding Mimics Phosphorylation of Aspartate in Response Regulators. J. Bacteriol. 2005, 187, 8229−8230. (7) Sternweis, P. C.; Gilman, A. G. Aluminum: A Requirement for Activation of the Regulatory Component of Adenylate Cyclase by Fluoride. Proc. Natl. Acad. Sci. U.S.A. 1982, 79, 4888−4891. (8) Feeney, J.; Haque, R.; Reeves, L. W.; Yue, C. P. Nuclear Spin− Spin Coupling Constants; Equilibrium and Kinetic Studies for Fluoroberyllate Complexes in Solution. Can. J. Chem. 1968, 46, 1389−1398. (9) Mesmer, R. E.; Baes, C. F. Fluoride Complexes of Beryllium(II) in Aqueous Media. Inorg. Chem. 1969, 8, 618−626. (10) Kotz, J. C.; Schaeffer, R.; Clouse, A. O. High-Resolution Nuclear Magnetic Resonance Spectroscopy of Some Beryllium-Containing Compounds. Beryllium-9 and Fluorine-19 Spectra. Inorg. Chem. 1967, 6, 620−622. (11) Liu, Y.; Mao, X.; Liu, M.; Jiang, L. Impact of Magnesium(II) on Beryllium Fluorides in Solutions Studied by 19F NMR Spectroscopy. Chin. J. Chem. 2014, 32, 878−882. (12) Bodor, A.; Tóth, I.; Bányai, I.; Szabó, Z.; Hefter, G. T. 19F NMR Study of the Equilibria and Dynamics of the Al3+/F− System. Inorg. Chem. 2000, 39, 2530−2537. 28
dx.doi.org/10.1021/jp509640d | J. Phys. Chem. A 2015, 119, 24−28
