Influence of Water Content on the Acidities in Acetonitrile. Quantifying

Oct 4, 2010 - Brian D. McCarthy , Daniel J. Martin , Eric S. Rountree , Alexander C. ... Lauri Lipping , Agnes Kütt , Karl Kaupmees , Ivar Koppel , P...
5 downloads 0 Views 144KB Size
Download PDF
11788

J. Phys. Chem. A 2010, 114, 11788–11793

Influence of Water Content on the Acidities in Acetonitrile. Quantifying Charge Delocalization in Anions Karl Kaupmees, Ivari Kaljurand, and Ivo Leito* Institute of Chemistry, UniVersity of Tartu, RaVila 14a Street, 50411 Tartu, Estonia ReceiVed: June 19, 2010; ReVised Manuscript ReceiVed: September 9, 2010

The effect of traces of water on the relative strengths of acids (∆pKa values) in acetonitrile was quantitatively evaluated experimentally and computationally (COSMO-RS). Water affects first of all the anions by selective solvation. Expectedly, the more localized is the charge in acid anions the higher is the effect of water. The energetic effect of increasing water content from 0 to ca. 10 000 ppm on solvation enthalpies of anions ranged from 0.2-0.4 kcal mol-1 (anions with delocalized charges) to 15 kcal mol-1 in the case of the highly chargelocalized acetate ion. In the case of ∆pKa values the change ranges from 0.01 to ca. 1.7 pKa units (acid pair involving acetic acid). The COSMO-RS method was found to satisfactorily describe the trends in ∆pKa values. To quantify the extent of charge localization/delocalization in anions a parameter, weighted average positive σ (WAPS), was introduced, which can be conveniently computed using the COSMO approach. WAPS characterizes the distribution of charge density across the molecular surface and was found to correlate well with the extent of water influence on the dissociation of the respective acid. Introduction Experimentally measured acid and base strength of molecules (usually expressed as pKa values) in different media are among the most important chemical characteristics of molecules. Acetonitrile (AN) is one of the most popular nonaqueous solvents for acid-base chemistry. To date, a large amount of experimental acidity data in AN has been published starting with the groups of Coetzee1,2 and Kolthoff.3,4 Systematic collections of measured data over several decades have been compiled,5,6 and the data are still constantly accumulating.7-9 Choosing the experimental method, concentrations of compounds and conditions for pKa determination is of great importance for the reliability of the results. Among others, potentiometry,10-12 UV-vis spectrophotometry,8 UV-vis in combination with potentiometry,11 NMR spectrometry,13,14 and conductometry11 have been used depending on the availability, compounds’ properties, etc. Ideally, the results are independent of the method used. Actually, the observed acidities are influenced by side processes (most commonly homoconjugation and ion-pairing),1 which depend on the compound’s nature, its concentration, and also the nature of the solvent. Another wellknown fact is that impurities influence physicochemical properties of compounds and chemical equilibria in solutions. The best known and most often reported impurity in solvents is water. It is well-known that amphiprotic water as a cosolvent in an aprotic solvent influences physicochemical properties of compounds and chemical equilibria substantially.1,15-18 Nevertheless, only very few quantitative data can be found on the impact of water residues (at impurity level) on the acidities19 or basicities in AN. The vast majority of reports on acidities and basicities use either pure (at a reasonable level) solvent7,8 or binary water-AN mixtures with water content of at least 40%.20-22 Recently, a comprehensive acidity scale of neutral Brønsted acids in acetonitrile spanning over 24 pKa units and including * To whom correspondence should be addressed. Phone: +372 7 375 259. Fax: +372 7 375 264. E-mail: [email protected].

more than 100 compounds8,9 was built. Dry acetonitrile (less than 40 ppm of water) and inert-gas glovebox conditions were used. CH acids with hindered acidity centers were used as the backbone of the scale to minimize possible influence of water in the solvent. However, no information was given on possible effect of traces of water. Some of the acids included in the scale of ref 8 are given in Table 1. They range from CH acids with hindered acidity centers and essentially without HB donor properties (6, 7) to highly polar molecules with naked acidity centers and strong HB donor properties (11, 12). On dissociation the former produce anions with delocalized charge and low affinity for water while the latter produce anions with localized charge and strong affinity for water. Obviously dissociation of the latter acids is more influenced by water. However, the magnitude of the influence of water on either group on acids has never been studied. Also, connected to this, the very concept of charge delocalization in an ion is in our opinion insufficiently quantified; it is used at qualitative level. There are different well-known approaches available for expressing partial charges on atoms in molecules and ions,23,24 but at present, to the best of our knowledge, there is no quantitative parameter available that would characterize the charge distribution in an ion as a whole. At the same time in numerous situations (especially in interactions, such as ion-aggregation, hydrogen bonding between neutrals and ions, etc.) it is the extent of charge delocalization of the ion as the whole that plays role. Brønsted acidity of an acid HA is expressed by the equilibrium

Ka

HA + S {\} A- + HS+

(1)

and is quantified by dissociation constant Ka or (more often) its negative logarithm pKa (also known as absolute pKa):

10.1021/jp105670t  2010 American Chemical Society Published on Web 10/04/2010

Acidities in Acetonitrile

J. Phys. Chem. A, Vol. 114, No. 43, 2010 11789

TABLE 1: Change of Relative Acidities and Solvation Enthalpies upon Change of Water Content acid pairsa

Hint (kcal mol-1)g CH2O (ppm)b 103CH2O (M) ∆pKa_expb s(∆pKa)c ∆∆pKad ∆pKa_calce ∆pKa_c_corrf Aneutral Aanion Bneutral Banion

picric acid, 1 (11.00) and 4-CF3-C6F4CH(CN)2, 2 (10.19) a

8 77 1020 11937 4 2,4-(NO2)2-phenol, 3 (16.66) and (4-CF3-C6F4)CH(CN) 71 COOEt, 4 (16.08) b 997 9076 6 2-NO2-phenol, 5 (22.85) and (4-Me-C6F4) 87 (C6F5)CHCN, 6 (21.94) c 1156 9516 9-COOMe-fluorene, 7 (23.53) 6 and (4-Me-C6F4)(C6F5)CHCN, 82 6 (21.94) d 906 9676 2,4,6-(SO2OCH2CF3)3-aniline, 19 8 (22.54)i and (4-Me-C6F4) 54 (C6F5)CHCN, 6 (21.94) e 1060 11394 (4-NC5F4)(C6H5)NH, 9 (26.34) 8 and (C6H5)(C6F5)CHCN, 61 10 (26.14) f 1031 12167 6 (4-Me-C6F4)(C6F5)CHCN, 6 (21.94) and benzoic acid, 57 11 (21.51) g 1224 10791 7 HCl, 12 (10.30)j and 2,3,4,5,6-(CF3)5C6CH(CN)2, 49 13 (8.86)k h 992 ∼10000 9-COOMe-fluorene, 7 (23.53) 10 and acetic acid, 14, (23.51) i 39 1379 ∼10000 10584 (4-Me-C6F4)(C6F5)CHCN, 6 (21.94) and acetic acid, 14, (23.51) j

0.36 3.33 44.4 519 0.17 3.08 43.4 395 0.27 3.79 50.3 414 0.24 3.58 39.4 421 0.80 2.35 46.1 496 0.36 2.65 44.9 529 0.27 2.49 53.3 469 0.30 2.13 43.2

-0.83 -0.83 -0.83 -0.82 -0.61 -0.62 -0.60 -0.52 -0.89 -0.90 -0.79 -0.33 -1.55 -1.57 -1.55 -1.48 -0.56 -0.56 -0.55 -0.61 -0.20 -0.20 -0.19 0.00 -0.37 -0.33 -0.56 -1.4 -1.10 -1.16 -0.80

0.01 0.02 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.03 0.02 0.02 0.05 0.02 0.02 0.03 0.03 0.01 0.01 0.01 0.02 0.05 0.1 0.13 0.1 0.02 0.05 0.11

0.44 1.70 56.0 435 460

-0.21 -0.17 -0.63 -1.74l -0.26

0.08 0.1 0.03

0 0 0.01 -0.01 0.01 0.09 -0.01 0.10 0.56 -0.02 0 0.07 0 0.01 -0.05 0 0.01 0.20 0.04 -0.19 -1.03 -0.06 0.30 0.00 0.04 -0.41 -1.53

0.02

-3.95 -3.95 -3.95 -3.95 -3.91 -3.91 -3.90 -3.87 -4.85 -4.85 -4.80 -4.53 -1.39 -1.39 -1.39 -1.39 -5.56 -5.56 -5.56 -5.58 -3.69 -3.69 -3.68 -3.63 3.61 3.58 3.18 2.03 -3.96 -3.96 -3.92

-0.83 -0.83 -0.83 -0.83 -0.61 -0.61 -0.61 -0.57 -0.89 -0.89 -0.84 -0.57 -1.55 -1.55 -1.55 -1.55 -0.56 -0.56 -0.56 -0.58 -0.20 -0.20 -0.19 -0.13 -0.37 -0.40 -0.79 -1.95 -1.10 -1.10 -1.06

3.13 3.10 2.41 0.88 2.21

-0.21 -0.24 -0.94 -2.46

-6.97 -6.97 -6.98 -7.02 -5.83 -5.83 -5.84 -5.86 -4.67 -4.67 -4.67 -4.68 -7.19 -7.19 -7.19 -7.21 -12.25 -12.25 -12.26 -12.29 -6.99 -6.99 -6.99 -7.02 -7.38 -7.38 -7.38 -7.38 -4.58 -4.58 -4.61 -4.61 -7.19 -7.19 -7.19 -7.21 -7.38

-6.62 -6.62 -6.63 -6.76 -5.91 -5.92 -5.99 -6.54 -5.07 -5.12 -5.64 -8.02 -8.77 -8.77 -8.80 -9.05 -12.20 -12.20 -12.23 -12.50 -7.62 -7.63 -7.78 -8.73 -8.43 -8.43 -8.47 -8.74 -1.13 -1.15 -1.51 -1.51 -8.77 -8.77 -8.81 -9.06 -8.74

-6.69 -6.73 -6.69 -6.73 -6.69 -6.75 -6.72 -6.87 -7.13 -7.88 -7.13 -7.89 -7.13 -7.92 -7.14 -8.18 -7.38 -8.43 -7.38 -8.43 -7.38 -8.47 -7.38 -8.71 -7.38 -8.43 -7.38 -8.43 -7.38 -8.46 -7.38 -8.72 -7.38 -8.43 -7.38 -8.43 -7.38 -8.46 -7.38 -8.76 -7.03 -7.97 -7.03 -7.98 -7.03 -8.02 -7.04 -8.40 -7.14 -4.49 -7.14 -4.91 -7.20 -8.99 -7.56 -16.03 -7.14 -7.37 -7.14 -7.37 -7.15 -7.39 -7.15 -7.39 -5.37 -2.17 -5.38 -2.61 -5.48 -9.30 -5.94 -18.42 -5.97 -18.73

Hint _corr (kcal mol-1)h A B 0 0.00 -0.01 -0.09 0 -0.01 -0.08 -0.59 0 -0.04 -0.56 -2.94 0 0.00 -0.03 -0.27 0 0.00 -0.03 -0.26 0 -0.01 -0.16 -1.08 0 0.00 -0.04 -0.31 0 -0.02 -0.35 -0.35 0 0.00 -0.04 -0.28 -4.56

0 0.00 -0.01 -0.11 0 0.00 -0.03 -0.28 0 0.00 -0.04 -0.28 0 0.00 -0.03 -0.28 0 0.00 -0.03 -0.32 0 0.00 -0.04 -0.42 0 -0.41 -4.43 -11.12 0 0.00 -0.02 -0.02 0 -0.44 -7.02 -15.68 -11.18

a Members of the acid pairs, literature pKa values (from ref 8 unless specified otherwise) in parentheses following compound name and number. b The approximate uncertainty of water content determination is ca. (5 ppm in the middle range and ca. (2 ppm in the low range; ∆pKa_exp denotes experimentally determined relative acidity ∆pKa_exp ) pKa(HA2) - pKa(HA1). c Experimental standard deviation of ∆pKa allows evaluation of the within-series agreement of ∆pKa values; for details see the Supporting Information. d Change in pKa units compared to the driest solvent for that acid pair. e Calculated ∆pKa according to eq 8. f Corrected ∆pKa values (∆pKa_corr ) ∆pKa_calci ∆pKa_calc0 + ∆pKa_exp0, 0 indicates the driest solvent used). g Calculated interaction enthalpies of neutral and anionic forms of both acids in pair; acid with higher pKa value is marked as A. h Corrected values of Hint, Hint_corr ) ∆Hint - ∆Hint0, ∆Hint ) Hint_anion - Hint_neutral; 0 indicates the driest solvent used. i Reference 9. j Reference 29. k Reference 32. l Estimated from measurement against compound 6; see text for details.

pKa ) -log

a(HS+) · a(A-) a(HA)

(2)

To eliminate the necessity to routinely measure the activity of hydrogen ion a(HS+), which can be problematic in nonaqueous media, the experimental method used in ref 8 was such that relative acidities of two acids are measured directly:

When the hydrogen-bond donor is the conjugate acid of the anion, the process is called homoconjugation.

A1- + HA1 a [A1- · · · HA1]

(5)

This needs to be taken into account through homoconjugation constant if it can be assumed that homoconjugation process is present.

K

HA2 + A1- {\} A2- + HA1

(3) KAHA )

a(A1 · · · HA1) a(A1-) · a(HA1)

(6)

The difference of acidities can be calculated as follows

-log K ) ∆pKa_exp ) pKa(HA2) - pKa(HA1) ) log

a(A1 ) · a(HA2) a(HA1) · a(A2)

(4)

In AN anions are weakly solvated.1 This leads to different association processes forming hydrogen-bonded complexes.

When the hydrogen bond donor is another substance (another acid, water, or some other impurity), then the association process is called heteroconjugation. In the present work heteroconjugation was not taken into account as such. Instead, it was observed by the impact of water on the ∆pKa_exp values. Experimental determination of the effect of water on the absolute pKa values in acetonitrile is not easy. Most of the methods used for pKa measurement are relative; i.e., the obtained values are either directly or indirectly dependent on the pKa

11790

J. Phys. Chem. A, Vol. 114, No. 43, 2010

Kaupmees et al.

value(s) of some reference acid(s). Water present in acetonitrile affects both the measured acid and the reference acid(s). Also, the largest influence of water is exerted on the solvated proton, whose activity can change markedly with changing water content and is difficult to quantify. Therefore, in this work we have chosen to investigate the effect of water in AN on the ∆pKa values. This way the problem with the changing pKa value of the reference acid is avoided. And, as we show below, by choosing suitable acid pairs, inference can be made also on the behavior of single acids, not only acid pairs. Due to the above difficulties, an efficient computational method for evaluation of the effect of traces of water would be desirable from two points of view: (1) as a generally usable tool, more convenient than experiments, and (2) as an independent confirmation method offering at least some level of support to the deductions made from the experimental data. For computational prediction of the influence of water traces on acidities in AN we chose COSMO-RS (conductor-like screening model for real solvents).25 This approach combines the COSMO26 method with statistical thermodynamics of pairwise interactions of segments of molecular surfaces. In somewhat simplified terms, the COSMO method takes into account (in an averaged way) the dispersion forces between the molecules and the statistical thermodynamics part takes into account the electrostatic interactions and hydrogen bond interactions. A particular strength of COSMO-RS is its ability to handle liquid mixtures with different component ratios.27 COSMO-RS has been demonstrated to predict the pKa values in AN with reasonable accuracy.28 This is especially true for acids that on dissociation form anions with delocalized charge. The pKa_calc values of such acids can be calculated from the dissociation reaction ∆G values using the theoretical expression (eq 7) without any empirical adjustments with a root-mean-square error of 1.1 pKa units.

pKa_calc )

∆G G(HS+) - G(S) + - log [S] RT ln(10) RT ln(10)

(7)

For calculating ∆pKa_calc the equation above transforms to

∆pKa_calc )

∆GA G(HS+) - G(S) + - log [S] RT ln(10) RT ln(10) ∆GB G(HS+) - G(S) + - log [S] RT ln(10) RT ln(10) ∆GA ∆GB 1 ) (∆GA - ∆GB) RT ln(10) RT ln(10) RT ln(10)

(

)

)

(8)

where indexes A and B mark two different acids. The COSMO calculation of the COSMO-RS method gives the so-called σ profile: distribution of charge density (σ) on a molecular surface.25 This distribution can be used for evaluating the charge delocalization in the species. For this we define a parameter: weighted average positive σ (WAPS). WAPS is defined as the weighted mean of positive σ values divided by the ion surface area. The surface segments with negative charge are (due to definition of σ) characterized by positive σ values. All together the WAPS can be calculated as follows:

∞

∫ σ · p(σ) dσ

WAPS )

σ)0

∞

A

(9)

∫ p(σ) dσ

σ)0

where σ is the polarization charge density, p(σ) is the probability function of σ, and A is the surface area of the anion. This work was undertaken for three purposes: (1) to quantitatively evaluate the influence of water on the relative acidities of some neutral OH, NH, and CH acids from ref 8 and HCl from ref 29 with varied strength and structure (for this, the relative acidities of the acids of Table 1 were studied in AN solutions of different water content), (2) to assess the ability of the COSMO-RS computational approach to predict the influence of traces of water on relative acidities in AN, and (3) to propose a parameter for evaluating the delocalization of charge in an ion (WAPS) and to evaluate its usefulness for predicting the influence of traces of water on pKa values in AN. Experimental Section Chemicals, equipment, conditions, titration method, and ∆pKa calculation methods were the same as in ref 8. The only difference in experiments was the water content. For each acid pair, one titration experiment was made with commercial AN without any treatment, for one experiment AN was dried by keeping it on molecular sieves and for higher concentrations water was added gravimetrically to AN before making solutions from it. After each experiment the water content of used solutions was determined also by coulometric Karl Fischer titration. The water content from gravimetry and determined by coulometry agreed well. Compound Pairs. The compound pairs were chosen in such a way that one of the acids gives upon dissociation delocalized anion and thus that anion is presumably little influenced by water (2, 4, 6, 9, 13). The other acid was in some cases (7, 8, 10) also with delocalized charge but mostly it was some acid that has moderately (1, 3) or highly (5, 11, 12, 14) localized charge in the dissociated anionic form. See Table 1 for experimental ∆pKa_exp values. Computations. For COSMO-RS computations the COSMOtherm software package30 (version C2.1, revision 01.10) was used. Geometry optimization and generation of COSMO input files was done using the Turbomole software package31 at the DFT BP TZVP level with full geometry optimization. The sequence of computations was the same as in ref 28. σ-profiles for calculating WAPS values were provided by COSMOtherm software. See Table 1 for calculated ∆pKa_calc and ∆pKa_c_corr. For more detailed information about experimental setup, structures of used substances, their geometries and σ-profiles see Supporting Information. Results The experimental data are presented in Table 1 and Figure 1. Nine acid pairs were studied. The ∆pKa_exp of each pair was measured in four solvent mixtures having water contents of approximately 10, 100, 1000, and 10 000 ppm. Acid pairs a, b, c, and f at roughly the 100 ppm level were measured both in the present work and in the work by Ku¨tt et al.8 The experimental ∆pKa_exp values in solvents of similar water content did not have statistically significant differences. This is evidence

Acidities in Acetonitrile

Figure 1. Influence of water content (expressed as logarithm of molar concentration) on the relative acidities (∆pKa) of acid pairs. Each pair is designated with a letter from Table 1. The crosshatched area on the graph represents the water content during experiments in ref 8. Vertical error bars represent the double value of standard deviation.

Figure 2. Comparison of corrected calculated (∆pKa) values of acid pairs most influenced by changing water content to experiment at different water contents (expressed as logarithm of molar concentration). The solid line indicates series of experiments, fragmental line calculated values.

that the experimental conditions in this work and ref 8 can be considered the same. Discussion The calculated ∆pKa values are listed in the seventh and eighth columns of Table 1. The directly calculated ∆pKa values (∆pKa_calc) differ significantly from experimental values but the trends in ∆pKa on increasing the water content expressed by ∆pKa_c_corr reproduce the experimental trends well. In calculations no correction factors for the slopes 1/RT ln(10) were used. However, they are needed when pKa values of acids are calculated with localized charges in their anions.28 Use of correction factors would also possibly enable us to get ∆pKa_calc values closer to the experiment. The correction factors were not used in this work because of their unavailability for the mixtures of AN and water. Computations generally underestimate the impact of water, except for pair g and i, where overestimation occurs (Figure 2). It can be concluded that underestimation occurs when the negative charge is delocalized in the anion. In the case of localized anionic charge the effect of water is overestimated. As mentioned before, the influence of water on the acidity of an acid is related to the delocalization of negative charge in the anionic form. The proposed WAPS value for quantifying the charge delocalization presented in Table 2 seems to give

J. Phys. Chem. A, Vol. 114, No. 43, 2010 11791 acceptable results. The larger the value, the more localized the negative charge. The last six columns of Table 1 contain interaction enthalpies calculated for both neutral and anionic form of each acid by COSMOtherm.25 The listed Hint values characterize both energetic effects of specific and nonspecific solvation which are separately evaluated by COSMO-RS calculation. Hint is the sum of misfit HMF, hydrogen-bond HHB, and van der Waals HvdW interaction energies in the mixture. The changes in solvation enthalpies of acids present a possibility to understand whether one of the acids in the pair is mainly responsible for the change in the ∆pKa or both of them contribute. In the case of the low effect of water on the ∆pKa value it can be determined if this is due to the negligible influence of water on the acids or because the influences mutually cancel. It should be noted that for acids 6 and 7 that were used in several pairs, the difference in the values of Hint is due to the slightly different water content in different mixtures. Picric acid (1) and compound 2 are strong acids and both have delocalized charge in their anions with WAPS values of 4.3 and 2.4, respectively. Data presented in Table 1 show that water has no influence on the relative acidity at concentrations up to 11 000 ppm. From the changes in solvation enthalpies it can be seen that water has little influence on both anions. The relative acidity in an acid pair b (2,4-dinitrophenol (3) and compound 4), weaker by 5 orders of magnitude, behaves in a slightly different way. At water concentrations over 9000 ppm contraction of ∆pKa by 0.1 units is observed. This can be interpreted as the strength of the OH acid increases more due to the specific solvation of its anions by the water molecules when water content increases. The strengthening of 2,4dinitrophenol can be seen from the solvation enthalpies as well. In the wettest solvent the energetic effect of specific solvation is calculated to be -0.59 kcal mol-1 on the 2,4-dinitrophenol whereas the effect on compound 4 is only -0.28 kcal mol-1. For the pair of acids 5 and 6, which are yet by another 5 orders of magnitude weaker, this effect is even more expressed. The relative strength of the OH acid 2-nitrophenol increases by 0.1 units at a water content of ca. 1000 ppm, and at 9500 ppm the relative strengthening is already 0.56 units. The 2-nitrophenolate ion is solvated better at a water content of 1% by 2.94 kcal mol-1. Notable changes in solvation enthalpies can be seen at a water content of 0.1%, which is consistent with the experiment. Based on the results of pairs b and c, the influence of traces of water on the ∆pKa is small if the WAPS value is 5 or lower. The comparison of acid pairs b and c is informative from the aspect that the steric hindrances of the anionic center are identical for anions of 3 and 5, at the same time the influence of water is quite different. This demonstrates that the accessibility of the anionic center is not the only factor and the degree of charge localization on the anionic center, which is higher in 2-nitrophenolate, is important as well. For acid pairs d and e slight changes of ∆pKa are observed at a water content of around 10 000 ppm, by 0.07 and 0.05 pKa units, respectively, but the impact of water is not nearly as large as for 2-nitrophenol, an OH acid. The pKa values of CH and NH acids with large molecules can be considered to have similar sensitivities to water. Both the experimental and calculated data suggest that in pair e compound 6 was influenced more by addition of water relative to compound 8. This can be explained by the slightly smaller WAPS value and the bulky oxygen rich -SO2OCH2CF3 substituent groups in 8, which shield the acidity center on the nitrogen atom and are good hydrogen-bond acceptors. According to the solvation enthalpies all four acids

11792

J. Phys. Chem. A, Vol. 114, No. 43, 2010

Kaupmees et al.

TABLE 2: Charge Delocalization in Anions Expressed by the WAPS Parameter acid WAPS (e/Å4)a a

1 4.3

2 2.4

3 4.9

4 2.1

5 6.2

6 1.7

7 3.1

8 1.3

9 2.8

10 2.4

11 7.1

12 35.9

13 1.5

14 16.2

e is electron charge; for better readability here and in subsequent discussion WAPS values are multiplied by 105.

are little influenced by water. The changes are so small and similar that only slight changes in ∆pKa can be observed experimentally. To further investigate the effect of water on NH acids, another pair, f, consisting of 4 pKa units weaker (relative to pair e) NH and CH acids (9 and 10, respectively) was studied. The anion of the NH acid in this case becomes more stabilized at a water content over 12 000 ppm, and this results in a 0.2 pKa unit contraction of the ∆pKa value. The WAPS values are 2.8 for 9 and 2.4 for 10. This suggests that the charge delocalizations in both anions are similarly extensive. The observed change in ∆pKa could be explained by the pyridyl group in 9. The negative charge concentration is high on the exposed nitrogen atom of the pyridyl group, which leads to higher sensitivity toward water. The changes are also observable in solvation enthalpies. Although the WAPS values of compounds 2 and 10 are similar, the differences in Hint_corr of compound 10 are quite noticeable among the CH acids used in this study. This confirms that the influence of water is related to the acid strength. For benzoic acid (11), which was measured against compound 6, the impact of water was severe. Its relative strength increased by 0.19 units at a water content of 1000 ppm and by 1.03 units at 10 000 ppm. This is easily interpreted: negative charge in the benzoate anion is only weakly delocalized into the aromatic ring, because only the field-inductive effect is operational. This leads to a quite high WAPS value of 7.1. The charge is concentrated on the two exposed oxygen atoms, which are good hydrogen bond acceptors. In the anion of 2-nitrophenol (5) the charge is delocalized more to the aromatic ring and to the nitro group. The hydrogen bonds formed are therefore weaker. This is supported by the difference in energetic effects of solvation of benzoic acid and 2-nitrophenol. The Hint_corr values are -11.12 and -2.94, respectively. For the pair h the ∆pKa value at a water content around 1000 ppm in Table 1 is not fully accurate (the ∆pKa value should actually be larger). This is because the neutral form of the acid 13 could not be achieved in the medium of high water content. The spectral data suggest that a large fraction of the acidic titrant was consumed for protonating water. It was impossible to take this effect into account due to the fact that HCl is an acid transparent in the UV region. The experiment at water content around 10 000 ppm was also carried out, but the protonation of water was so extensive that getting even near the protonated forms of the acids was impossible and the attempts to calculate the ∆pKa were unsuccessful. Spectra of neutral form from drier solvents can be used to calculate ∆pKa if both of the studied compounds have spectra in the UV-vis region. The gathered data for pair h suggest that the behavior of HCl is similar to that of weaker OH acids: the low water content ( 20), in particular, on the acidities of OH acids (unless with a strongly sterically shielded acidity center) compared to NH and CH acids. From the experimental point of view, the water content for acidity determination experiments in AN should be as low as possible, especially for acids with WAPS values higher than 4.5. The results of this work confirm that the acidity scale in ref 8 is essentially unaffected by the water traces in solvent used

Acidities in Acetonitrile because the core of the scale is composed using acids with delocalized charges in their anions. Acknowledgment. This work was supported by grants nos. 6699 and 7374 from the Estonian Science Foundation. We are thankful to Dr. Sulev Sild, Prof. Ilmar A. Koppel, and Dr. Aleksander Trummal for useful discussions. Supporting Information Available: Descriptions of experiment, structures of the molecules, notes on obtained ∆pKa values, computational geometries, and σ-profiles of the acid anions. This material is available free of charge via the Internet at http://pubs.acs.org. References and Notes (1) Coetzee, J. F. Prog. Phys. Org. Chem. 1967, 4, 45–92. (2) Coetzee, J. F.; Padmanabhan, G. R. J. Phys. Chem. 1965, 69, 3193– 3196. (3) Kolthoff, I. M.; Bruckenstein, S.; Chantooni, M. K., Jr. J. Am. Chem. Soc. 1961, 83, 3927–3935. (4) Kolthoff, I. M.; Chantooni, M. K., Jr.; Bhowmik, S. J. Am. Chem. Soc. 1966, 88, 5430–5439. (5) Tables of Rate and Equilibrium Constants of Heterolytic Organic Reactions; Palm, V., Ed.; VINITI: Moscow-Tartu, 1975-1985. (6) Izutsu, K. Acid-Base Dissociation Constants in Dipolar Aprotic SolVents; IUPAC Chemical Data Series No. 35; Blackwell Scientific: Oxford, U.K., 1990. (7) Kaljurand, I.; Ku¨tt, A.; Soova¨li, L.; Rodima, T.; Ma¨emets, V.; Leito, I.; Koppel, I. A. J. Org. Chem. 2005, 70, 1019–1028. (8) Ku¨tt, A.; Leito, I.; Kaljurand, I.; Soova¨li, L.; Vlasov, V. M.; Yagupolskii, L. M.; Koppel, I. A. J. Org. Chem. 2006, 71, 2829–2838. (9) Leito, I.; Raamat, E.; Ku¨tt, A.; Saame, J.; Kipper, K.; Koppel, I. A.; Koppel, I.; Zhang, M.; Mishima, M.; Yagupolskii, L. M.; Garlyauskayte, R. Yu.; Filatov, A. A. J. Phys. Chem. A 2009, 113, 8421–8424. (10) Koppel, I. A.; Koppel, J.; Leito, I.; Koppel, I.; Mishima, M.; Yagupolskii, L. M. J. Chem. Soc., Perkin Trans. II 2001, 229–232. (11) Garrido, G.; Koort, E.; Rafols, C.; Bosch, E.; Rodima, T.; Leito, I.; Roses, M. J. Org. Chem. 2006, 71, 9062–9067. (12) Wrobel, R.; Chmurzynski, L. Anal. Chim. Acta 2000, 405, 303– 308. (13) Abdur-Rashid, K.; Fong, T.; Greaves, B.; Gusev, D.; Hinman, J.; Landau, S.; Lough, A.; Morris, R. J. Am. Chem. Soc. 2000, 122, 9155– 9171. (14) Kolomeitsev, A. A.; Koppel, I. A.; Rodima, T.; Barten, J.; Lork, E.; Ro¨schenthaler, G.-V.; Kaljurand, I.; Ku¨tt, A.; Koppel, I.; Ma¨emets, V.; Leito, I. J. Am. Chem. Soc. 2005, 127, 17656–17666.

J. Phys. Chem. A, Vol. 114, No. 43, 2010 11793 (15) Reichardt, C. SolVents and SolVent Effects in Organic Chemistry, 3rd ed.; Wiley-VCH: Weinheim, Germany, 2003. (16) Anslyn, E. V.; Dougherty, D. A. Modern Physical Organic Chemistry; University Science Books: Sausalito, CA, U.S., 2006. (17) Porras, S. P.; Kuldvee, R.; Palonen, S.; Riekkola, M.-L. J. Chromatogr. A. 2003, 990, 35–44. (18) Zarzycki, P. K.; Zarzycka, M. B.; S´lczka, M. M.; Clifton, V. L. Anal. Bioanal. Chem. 2010, 397, 905–908. (19) Kolthoff, I. M.; Chantooni, M. K., Jr. J. Am. Chem. Soc. 1969, 81, 6907–6910. (20) Buckenmaier, S. M. C.; McCalley, D. V.; Euerby, M. R. J. Chromatogr. A 2003, 1004, 71–79. ¨ zkan, G.; (21) Beltra´n, J. L.; Sanli, N.; Fonrodona, G.; Barro´n, D.; O Barbosa, J. Anal. Chim. Acta 2003, 484, 253–264. (22) Espinosa, S.; Bosch, E.; Rose´s, M. Anal. Chim. Acta 2002, 454, 157–166. (23) Meister, J.; Schwarz, W. H. E. J. Phys. Chem. 1994, 98, 8245– 8252, 30. (24) Cramer, C. J. Essentials of Computational Chemistry: Theories and Models; John Wiley & Sons, Ltd: Chichester, England, 2003. (25) Klamt, A. COSMO-RS: From Quantum Chemistry to Fluid Phase Thermodynamics and Drug Design; Elsevier Science Ltd: Amsterdam, The Netherlands, 2005. (26) Klamt, A.; Schu¨u¨rmann, G. J. Chem. Soc., Perkins Trans. 2 1993, 5, 799–805. (27) MacDonald, S. M.; Opallo, M.; Klamt, A.; Eckert, F.; Marken, F. Phys. Chem. Chem. Phys. 2008, 10, 3925–3933. (28) Eckert, F.; Leito, I.; Kaljurand, I.; Ku¨tt, A.; Klamt, A.; Diedenhofen, M. J. Comput. Chem. 2009, 30, 799–810. (29) Ku¨tt, A.; Rodima, T.; Saame, J.; Raamat, E.; Ma¨emets, V.; Kaljurand, I.; Koppel, I. A.; Garlyauskayte, R. Y.; Yagupolskii, Y. L.; Yagupolskii, L. M.; Bernhardt, E.; Willner, H.; Leito, I. J. Org. Chem., submitted for publication. (30) Eckert, F.; Klamt, A. COSMOtherm, Version C2.1, ReVision 01.10; COSMOlogic GmbH&CoKG: Leverkusen, Germany, 2006; http://www. cosmologic.de. (31) Ahlrichs, R.; Ba¨r, M.; Baron, H.-P.; Bauernschmitt, R.; Bo¨cker, S.; Ehrig, M.; Eichkorn, K.; Elliott, S.; Furche, F.; Haase, F.; Ha¨ser, M.; Horn, H.; Hattig, C.; Huber, C.; Huniar, U.; Kattannek, M.; Ko¨hn, M.; ¨ hm, H.; Scha¨fer, A.; Ko¨lmel, C.; Kollwitz, M.; May, K.; Ochsenfeld, C.; O Schneider, U.; Treutler, O.; von Arnim, M.; Weigend, F.; Weis, P.; Weiss, H. Turbomole Version 5.8, 2005. (32) Ku¨tt, A.; Movchun, V.; Rodima, T.; Dansauer, T.; Rusanov, E. B.; Leito, I.; Kaljurand, I.; Koppel, J.; Pihl, V.; Koppel, I.; Ovsjannikov, G.; Toom, L.; Mishima, M.; Medebielle, M.; Lork, E.; Ro¨schenthaler, G.-V.; Koppel, I. A.; Kolomeitsev, A. A. J. Org. Chem. 2008, 73, 2607–2620. (33) Eckert, F. Personal communication.

JP105670T