UV−Visible and 1H or 13C NMR Spectroscopic Studies on the Specific

J. Phys. Chem. B 2007, 111, 1759-1768

1759

UV-Visible and 1H or 13C NMR Spectroscopic Studies on the Specific Interaction between Lithium Ions and the Anion from Tropolone or 4-Isopropyltropolone (Hinokitiol) and on the Formation of Protonated Tropolones in Acetonitrile or Other Solvents Masashi Hojo,* Tadaharu Ueda, Tomonori Inoue, and Michitaka Ike Department of Chemistry, Faculty of Science, Kochi UniVersity, Akebono-cho, Kochi 780-8520, Japan

Masato Kobayashi and Hiromi Nakai Department of Chemistry, School of Science and Engineering, Waseda UniVersity, Tokyo 169-8555, Japan ReceiVed: October 14, 2006; In Final Form: December 7, 2006

The specific interaction between lithium ions and the tropolonate ion (C7H5O2-: L-) was examined by means of UV-visible and 1H or 13C NMR spectroscopy in acetonitrile and other solvents. On the basis of the electronic spectra, we can propose the formation of not only coordination-type species (Li+(L-)2) and the ion pair (Li+L-) but also a “triple cation” ((Li+)2L-) in acetonitrile and acetone; however, no “triple cation” was found in N,N-dimethylformamide (DMF) and in dimethylsulfoxide (DMSO), solvents of higher donicities and only ion pair formation between Li+ and L- in methanol of much higher donicity and acceptivity. The 1H NMR chemical shifts of the tropolonate ion with increasing Li+ concentration verified the formation of (Li+)2Lspecies in CD3CN and acetone-d6, but not in DMF-d6 or CD3OD. With increasing concentration of LiClO4 in CD3CN, the 1H NMR signals of 4-isopropyltropolone (HL′) in coexistence with an equivalent amount of Et3N shifted first toward higher and then toward lower magnetic-fields, which were explained by the formation of (Li+)(Et3NH+)L′- and by successive replacement of Et3NH+ with a second Li+ to give (Li+)2L′-. In CD3CN, the 1,2-C signal in the 13C NMR spectrum of tetrabutylammnium tropolonate (n-Bu4NC7H5O) appeared at an unexpectedly lower magnetic-field (184.4 ppm vs TMS) than that of tropolone (172.7 ppm), while other signals of the tropolonate showed normal shifts toward higher magnetic-fields upon deprotonation from tropolone. Nevertheless, with addition of LiClO4 at higher concentrations, the higher and lower shifts of magnetic-fields for 1,2-C and other signals, respectively, supported the formation of the (Li+)2L- species, which can cause redissolution of LiL precipitates. All of the data with UV-visible and 1H and 13C NMR spectroscopy demonstrated that the protonated tropolone (or the dihydroxytropylium ion), H2L+, was produced by addition of trifluoromethanesulfonic or methanesulfonic acid to tropolone in acetonitrile. The order of the 5-C and 3,7-C signals in 13C NMR spectra of the tropolonate ions was altered by addition of less than an equivalent amount of H+ to the tropolonate ion in CD3CN. Theoretical calculations satisfied the experimental 13 C NMR chemical shift values of L-, HL, and H2L+ in acetonitrile and were in accordance with the proposed reaction schemes.

Introduction Tropolone is formally an R-diketone with two oxygen atoms being exocyclic to the seven-membered ring.1 In many respects, its coordination chemistry resembles that of a β-diketone, acetylacetone. Nonbenzenoid aromaticity and apparent nonrigid character have made tropolone and its derivatives the subject of considerable interest, both experimentally2 and theoretically.3 In solution, the 1H and 13C NMR spectra of tropolone always exhibit averaged signals due to the two tautomeric forms for which a lower limit of 108 s-1 can be set for the rate of the hydrogen shift at room temperature.4,5 It goes without saying that R-diketonates as well as β-diketonates are ready to form complexes (MLn) with transition metals (Mn+). The 1:1 complexes between transition metals (Mn+-L-) or alkaline earth metal ions (M2+-L-) in aqueous solution have been reported.6 Raban and Yamamoto7 reported experimental evidence for a chelated lithium bisacetylacetonate * Corresponding author. E-mail: [email protected].

(Li(acac)2-) complex in methanol solution. The formation of such a normal coordination-type species can be easily accepted by many chemists because of the possible stability on chelating in solution. The “dilithium complex” of a ligand anion (L-),8 on the other hand, has still been regarded as an uncertain matter. Murray and Hiller9 first suggested involvement of two lithium ions in a ligand loss during one-electron reduction of Fe(acac)3 in acetonitrile containing LiClO4 as the supporting electrolyte:

Fe(acac)3 + 2Li+ + e- / Fe(acac)2 + Li2(acac)+ (1) Similarly, the formation of [CH3COOLi2]+ species was suggested for the effect of LiClO4 on the polarographic reduction of dimeric copper(II) acetate in acetonitrile.10 The idea of both normal coordination and “reverse coordination” between a carboxylate ion and monovalent metal ions is illustrated in Scheme 1. The existence of lithium triple anion ([(CF3COO)2Li]-) has been observed in solution;11 the formation of (M+)2L--type species for Ag+ has also been reported in pyridine solution.12 Fuoss and Kraus13 proposed the concept of “symmetrical” triple

10.1021/jp066756n CCC: $37.00 © 2007 American Chemical Society Published on Web 01/27/2007

1760 J. Phys. Chem. B, Vol. 111, No. 7, 2007 SCHEME 1

ions, that is, anionic and cationic triple ions (to the same extent), based on mere electrostatic interaction between (spherical) ions in an electric field of low permittivity. By means of voltammetry,14 conductometry,15,16 and UVvisible and NMR spectroscopy,17-22 we have certainly demonstrated that alkali metal (M+) and alkaline earth metal ions (M2+) have “weak” but unexpectedly distinct chemical (i.e., covalent bonding or coordinating) as well as Coulombic interactions with many simple anions, not only in solvents of low permittivity (r < 10)16 but also in higher permittivity media (20 < r < 65) with poor solvating ability.14,15 The coordination chemistry of the lithium ion with simple anions as well as cyclic ligands has been reviewed by Olsher et al.23 Stable carbocations can be produced from trityl halides by addition of perchlorates of Li+, Na+, Mg2+, and Ba2+ in acetonitrile;17-19 the carbocations have been detected by UV-visible,17 1H, and 13C NMR spectroscopy.18,19 Salt effects at higher concentrations on solvolysis reaction rates of typical SN1 substrates, such as t-butyl chloride and 1-adamantyl halides (RX), have been elucidated on the basis of the (chemical) interaction between M+ or M2+ and the halide ions (X-) from the substrates in organic solvents and water mixtures.24 We have proposed that the properties of bulk water are altered into those of “dihydrogen ether” R[H]O-[H]R when the huge network of bulk water is distorted by added organic solvents and concentrated salts.24,25 We have also explained the large salt effects on the indicator acidity (or Hammett acidity function)20 and on the proton transfer from tropolone21 or nitrophenols22 to amine bases in acetonitrile solution in terms of the chemical interaction between the anions and M+ or M2+. We have reported the direct effects of M+ or M2+ on the equilibria of acid-base and metal indicators.26 Binder and Kreevoy27 examined the interaction of Li+ with a betaine dye and p-nitophenolate in acetonitrile. Specific interactions between many anions and ammonium ions have been observed in protophobic aprotic solvents, such as acetonitrile, benzonitrile, acetone, nitromethane, and nitrobenzene.28 In the previous paper,21 we have proposed that a cationic “triple ion” or a “reverse-coordinated” species, C7H5O2-(Li+)2, can be formed through an intermediate species, where both Li+ and Et3NH+ interact with a tropolonate ion when a large excess of LiClO4 is added to a tropolone-MeCN solution containing an equivalent amount of Et3N:

C7H5O2H---NEt3 + Li+ / C7H5O2-(Li+)(H+NEt3) (2) C7H5O2-(Li+)(H+NEt3) + Li+ / C7H5O2-(Li+)2 + Et3NH+ (3) Pocker and Ciula29 have deduced an n-value to be from 1.5 to 4 for the equilibrium of the proton-transfer reaction of tropolone (HA) with amine bases (B) in diethyl ether, as shown in eq 4.

HA + B + nLiClO4 / Li+A-(n - m - 1)LiClO4 + BH+ClO4-mLiClO4 (4) In a paper from 2002, however, Pocker and Spyridis30 have

Hojo et al. described that two lithium ions are interacting with a single molecule of acetylacetone in its keto-form in diethyl ether. In the present paper, interactions between lithium ions and the anion of tropolone (Scheme 2) or 4-isopropyltroplone (hinokitiol) in acetonitrile and other solvents are examined by means of UV-visible absorption and by 1H and 13C NMR spectroscopy. We would like to report decisive evidence for the formation of “reverse-coordinated” species, (Li+)2L-, for the anion (L-) from tropolone or 4-isopropyltropolone in acetonitrile and acetone but not in DMF. We took care of the “effective” donicity or bulk donicity of a protic solvent, such as methanol and water. The “effective” donicity differs much from the original donor number by Gutmann,31 depending on the solution conditions, for example, DNbulk ) 3332 or 42.3 (cf., Table 1, ref 33) for bulk water, as compared to DN ) 18.031 for isolated H2O molecules in an organic solvent. In the course of the present study, we have encountered the formation of protonated tropolones (H2L+) upon the addition of rather strong acids to tropolone and 4-isopropyltropolone in acetonitrile. Doering and Knox35 postulated the conjugate acid of tropolone (the protonated tropolone) as a reaction intermediate in esterification and hydrolysis reactions. Olah and Calin36 have reported the 1H NMR spectrum of the protonated tropolone in SbF5-FSO3H-SO2 or FSO3H-SO2 at low temperatures. Protons gave quite interesting effects on the 1H and 13C NMR chemical shifts of the tropolonate ion, tropolone, or 4-isopropyltropolone at 25 °C in acetonitrile; the experimental results are reported in this connection. Quantum chemical calculations were performed to verify the experimental 13C NMR chemical shift values. Experimental Section Tropolone (2-hydroxycyclohepta-2,4,6-triene-1-one) and 4-isopropyltropolone of GR grade were purchased from Aldrich and TCI, respectively. Tetrabutylammonium tropolonate (n-Bu4NC7H5O2) was prepared from tertrabutylammonium hydroxide (10% aqueous solution, Wako, GR grade) and an equivalent amount of tropolone in methanol. The fine crystals were obtained by the complete evaporation of the solvent with repeated addition of acetone (a state of ionic liquids), followed by brief soaking of the flask into liquid nitrogen (formation of an amorphous solid), and by final warming of the flask to ambient temperature. The obtained crystals were dried in vacuo (Anal. Calcd for C23H41NO2‚2H2O: C, 69.13; H, 11.35; N, 3.51. Found: C, 69.56; H, 11.51; N, 3.53). Crystals of tetrabutylammonium 4-isopropyltropolonate could not be obtained probably because of their lower melting point. Perchlorates with no hydrated water, LiClO4 (GR grade from Wako), NaClO4, Mg(ClO4)2, and Ba(ClO4)2 (Aldrich), were used as received. Tetraethylammonium perchlorate was prepared by adding an equivalent amount of HClO4 to Et4NBr in water. The precipitate was filtered from the solution, washed with cold water, and recrystallized several times from water, followed by drying at 70 °C in vacuo. Triethylamine, and trifluoromethanesulfonic and methanesulfonic acids, from Wako, and 1,1,3,3-tetramethylguanidine (1,1,3,3-TMG) and tetramethylammonium hydroxide (25 wt % in methanol) from Aldrich were used as received. Commercially obtained solvents of GR grade (Wako), acetonitrile, acetone, N,N-dimethylformamide (DMF), dimethylsulfoxide (DMSO), methanol, and ethanol, were also used as received. Water was purified by a MilliQ system (Millipore Corp.). UV-visible absorption spectra were measured at room temperature using a Shimadzu double-beam spectrophotometer

Formation of Protonated Tropolones in Acetonitrile

J. Phys. Chem. B, Vol. 111, No. 7, 2007 1761

SCHEME 2

(model UV-2550) in 0.01, 0.1, and 1.0 cm path-length quartz cuvettes. When precipitation occurred, the solution was sonicated for a few minutes, and the supernatant solution was measured after centrifugation. Solutions of tetrabutylammonium tropolonate were prepared in brown-colored containers to avoid effects of light. All 1H and 13C NMR measurements were carried out in a single NMR tube at 25 ( 0.1 °C with a JEOL FT NMR spectrometer (model JNM-LA400). NMR solvents containing TMS were obtained from Aldrich: acetonitrile-d3, 99.8 atom% D (containing 0.03% TMS); acetone-d6, 99.9 atom% D (0.1% TMS); N,N-dimethylformamide-d7, 99.5 at. % D (1% TMS); and methanol-d4, 99.8 at. % D (0.1% TMS). For the component analyses of the LiL and NaL precipitates, the concentrations of Li+ and Na+ in 0.1 mol dm-3 HCl aqueous solutions of the precipitates were determined by a Hitachi polarized Zeeman atomic absorption spectrometer (model Z-6100), and the amount of tropolone (HL) in the solutions was determined using a UVvisible spectrophotometer. The assignment of 1H and 13C NMR signals of 4-isopropyltropolone (hinokitiol) was made by using the database in SDBSWeb: http://www.aist.go.jp/RIODB/SDBS/(National Institute of Advanced Industrial Science and Technology, October 27, 2005). Results and Discussion UV-Visible Absorption Spectra of the Tropolonate Ion in Various Solvents. In acetonitrile, the tropolonate ion (nBu4N+C7H5O2-) exhibits three main absorption bands at 414, 348, and 250 nm, while tropolone (C7H5O2H) gives UV absorption bands at 368, 352, 320, and 226 nm (cf., ref 21). Figure 1 shows the spectral changes of the tropolonate ion (1.0 × 10-3 mol dm-3 n-Bu4NC7H5O2) by addition of LiClO4 over a wide concentration range, from 5.0 × 10-4 to 0.5 mol dm-3, in acetonitrile. In the presence of a half-equivalent amount of Li+ (5.0 × 10-4 mol dm-3), the peaks of 414 and 348 nm due

Figure 1. UV-visible spectra of 1.0 × 10-3 mol dm-3 n-Bu4NC7H5O2 (0.1 cm path-length) in acetonitrile as a function of LiClO4 concentration: (1 -) 0; (- - -) 5.0 × 10-4; (- - -) 1.0 × 10-3; (- - -) 0.10; (5 -) 0.50 mol dm-3 of LiClO4.

to the tropolonate ion shifted to 403 and 343.5 nm, respectively. Precipitation occurred after addition of an equivalent amount of Li+ (1.0 × 10-3 mol dm-3); the supernatant solution after centrifugation gave a lower intensity of the peaks, accompanied by a slight hypsochromic shift to 401 nm. Further addition of LiClO4 (up to 2.0 × 10-3 mol dm-3) caused a further decrease in intensity of the peaks. However, in the presence of a 10-fold excess of Li+ (1.0 × 10-2 mol dm-3), the stunted peak around 401 nm began to increase again, accompanied by a further hypsochromic shift (392 nm). In 0.5 mol dm-3 LiClO4, the tropolonate ion gave two bands at 386 and 326.5 nm above 300 nm. The chemical analyses of the pale yellow precipitates (by the addition of 1.0 or 2.0 × 10-3 mol dm-3) confirmed their composition to be [Li+]:[C7H5O2-] ) 1:1. We would like to stress that the addition from 5.0 × 10-4 to 0.5 mol dm-3 Et4NClO4 to a 1.0 × 10-3 mol dm-3 solution of the tropolonate ion caused no effects on the absorption spectrum of the tropolonate ion in acetonitrile. On the basis of the above experimental results, we can propose the products due to the interaction between Li+ and the free ligand (L-: the tropolonate ion) in acetonitrile solution as follows: (1) a half-equivalent amount of Li+ to a ligand ion, Li+(L-)2, the coordination-type species; (2) an equivalent amount of Li+ to a ligand ion, Li+L-, the non-charged species of low solubility; and (3) with a large excess of Li+ to a ligand ion, (Li+)2L-, the “triple cation” or “reverse-coordinated” species. We have definitely demonstrated the formation of both Li+(A-)2 and (Li+)2A- for 4-acyl-5-pyrazolonates and related ligands (A-: β-diketonates) in acetonitrile.37 For a higher concentration of the tropolonate ion (1.0 × 10-2 mol dm-3, using a 0.01 cm path-length cuvette), more distinct features of the interaction between the tropolonate ion and Li+ in acetonitrile were observed (cf., Figure 2). The addition of an equivalent amount of Li+ (1.0 × 10-2 mol dm-3) to the free ligand, L-, caused an undersized intensity from the tropolonate ion. The absorbance of the band peak was further decreased by addition of more LiClO4 (2.0 × 10-2 mol dm-3), presumably due to “common ion effect” for the Li+L- salt formation. However, the band peak increased abruptly in the presence of more than 5.0 × 10-2 mol dm-3 LiClO4. In contrast, with a 1.0 × 10-4 mol dm-3 concentration of tropolonate ion, no apparent precipitation was observed for the whole range of LiClO4 concentrations; therefore, the absorptivity of the bands due to Li+L- (the ion pair) could be estimated as /cm-1 mol-1 dm3 ) 1.3 × 104. Sodium perchlorate caused a copious precipitation with the tropolonate ion in acetonitrile. Even in the presence of a 500fold excess of Na+ to the tropolonate ion (1.0 × 10-3 mol dm-3), only a limited part of Na+L- crystals dissolved in the solution. For a 1.0 × 10-4 mol dm-3 solution of the tropolonate ion, no

1762 J. Phys. Chem. B, Vol. 111, No. 7, 2007

Hojo et al.

Figure 2. Changes in absorbance (the first absorption band) of the tropolonate ion in the presence of various concentrations of LiClO4 or NaClO4 in acetonitrile: (0) LiClO4, 1.0 × 10-2 mol dm-3 C7H5O2-; (O) LiClO4, 1.0 × 10-3 mol dm-3 C7H5O2-; (b) LiClO4, 1.0 × 10-4 mol dm-3 C7H5O2-; (4) NaClO4, 1.0 × 10-3 mol dm-3 C7H5O2-; (2) NaClO4, 1.0 × 10-4 mol dm-3 C7H5O2-. For the supernatant solution after centrifugation when precipitates formed, cuvettes of 0.01, 0.1, and 1.0 cm path-lengths were used for 1.0 × 10-2, 1.0 × 10-3, and 1.0 × 10-4 mol dm-3 C7H5O2-, respectively.

Figure 3. UV-visible spectra of 1.0 × 10-3 mol dm-3 n-Bu4NC7H5O2 (0.1 cm path-length) in DMF as a function of LiClO4 concentration: (1 -) 0; (2 - - -) 5.0 × 10-4; (- - -) 1.0 × 10-3; (- - -) 1.0 × 10-2; (5 -) 0.10; (6 - - -) 0.50 mol dm-3 of LiClO4.

precipitation of Na+L- was observed by addition of 0.5 mol dm-3 NaClO4, probably because of the formation of (Na+)2L-. The composition of the precipitate was found to be [Na+]:[L-] ) 1:1 by the chemical analyses. In acetone (DN ) 17.0),31 the interaction between Li+ and L- was found to be as in MeCN, except that the peak wavelengths of the tropolonate species in acetone were slightly longer than in MeCN (cf., Table 1). In DMF, however, UVvisible spectral changes were observed for the tropolonate ion (1.0 × 10-3 mol dm-3, Figure 3) for various concentrations of LiClO4. Although two-step successive shifts in peak wavelength were observed with increasing concentration of LiClO4, that is, at 416, 405, and 404.5 nm with 0, 0.5 × 10-3, and 1.0 × 10-3 mol dm-3 LiClO4, respectively, no further shift was observed even by addition of a large excess of the salt: the formation of Li+(L-)2 and Li+L- occurred, but no (Li+)2L-

Figure 4. λmax/nm values of the tropolonate ion (1.0 × 10-3 mol dm-3) in mixed solvents between MeOH or H2O and acetonitrile: (O) MeOH; (b) H2O.

species appeared to be formed in DMF. The higher donicity of DMF (DN ) 26.6)31 than that of MeCN (DN ) 14.1)31 ought to lower the activity of Li+ in DMF more than in MeCN. In a solvent of even higher donicity, DMSO (DN ) 29.8),31 the interaction between Li+ and L- should be weaker than in DMF; actually more than 1.0 × 10-2 mol dm-3 of Li+ was needed for the observation of the ion pair (Li+L-). Araki et al.38 have reported that lithium β-diketonate (LiL) exists mainly as Li(uni-L) in DMSO. In a protic solvent, methanol, the interaction between Li+ and L- appeared to be much different from that in the aprotic solvents (as shown above). The two absorption bands at 400 and 335 nm for >300 nm were altered to 393 and 330 nm by the addition of 0.5 mol dm-3 LiClO4. Neither Li+(L-)2 nor (Li+)2L- seemed to be present in methanol: only the equilibrium between the ligand (L-) and the ion pair (Li+L-) was demonstrated by the appearance of isosbestic points at 334, 361, and 396.5 nm. Using mixed solvents between methanol and MeCN, we further explored the cause of the hypsochromic effect (blue shift) in the absorption bands of the tropolonate ion in methanol, as compared to those observed in an aprotic solvent as MeCN. Figure 4 shows the changes in wavelength of the first absorption band peak of 1.0 × 10-3 mol dm-3 solution of n-Bu4NC7H5O2 with increasing amounts of methanol to MeCN. The peak at 414 nm in MeCN shifted suddenly to 409.5 nm upon the addition of only 2.0% (v/v) methanol. Further hypsochromic shifts were caused by more methanol, at last, up to 400 nm in pure methanol. We also observed that the intensity of the peak decreased (hypochromic effect) with increasing content of methanol. The hypsochromic and hypochromic effects by the addition of methanol to MeCN were attributed to an increasing

TABLE 1: λmax (of the First Absorption Band) Valuesa of the Species Produced from Li+ and the Tropolonate Ion (L-) with Increasing Concentration of Li+ to a Constant Concentration of L- in Various Solvents acetonitrile acetone DMF DMSO methanol water

b

DNc

ANc

L-

Li+ (L-)2

Li+L-

(Li+)2L-

35.95 20.70 36.71 46.7 32.70 78.30

14.1 17.0 26.6 29.8 33.8d (19)e 42.3d (18.0)e

19.3 12.5 16.0 19.3 41.3 54.8

414 416 416 417 400 392

403 405 405 406

401 403 404.5 405 393 (386)f

386 389

a Only band peaks of the longest wavelength are listed. b The permittivities of the solvents at 25 °C (cf., ref 34). c Gutmann’s donor and accepter numbers. d The donicity of a bulk solvent (DNbulk)33 and not of isolated molecules. e The donor number for isolated molecules of MeOH or H2O. f The formation of the ion pair occurs probably at higher concentrations of LiClO4.

Formation of Protonated Tropolones in Acetonitrile

J. Phys. Chem. B, Vol. 111, No. 7, 2007 1763 TABLE 2: 1H NMR Chemical Shift Valuesa of the Tropolonateb Ion in the Presence of LiClO4 in Various Solvents at 25 °C [LiClO4]/mol dm-3 0

Figure 5. 1H NMR spectra of 1.0 × 10-3 mol dm-3 n-Bu4NC7H5O2 in the presence of LiClO4: (1) 0; (2) 1.0 × 10-3; (3) 0.50 mol dm-3 LiClO4.

acceptivity of the mixed solvents; the acceptor numbers (AN) of methanol and MeCN are defined to be 19.3 and 41.3, respectively.31 The effect of water (AN ) 54.8)31 upon the wavelength of the tropolonate ion in MeCN was similar but more distinct than that of methanol. We have demonstrated that only a very small amount of water or “residual” water in acetonitrile has no longer the properties of bulk water but behaves more like an ether.25 On the other hand, the bathochromic (red shift) and hyperchromic effects for the tropolonate ion, observed in acetone (AN ) 12.5)31 and in DMF (AN ) 16.0),31 can be accounted for by the smaller acceptor numbers of these two solvents. 1H or 13C NMR Evidence of “Dilithium Tropolonate”. On the basis of the electronic spectra, we can propose the formation of not only coordination-type species (Li+(L-)2) and the ion pair (Li+L-) but also the “triple cation” ((Li+)2L-) in acetonitrile and acetone; however, no “triple cation” seems to be found in DMF and DMSO, solvents of higher donicities, and only the ion pair formation between Li+ and L- in methanol, of higher donicity and acceptivity. 1H or 13C NMR data can supply information of the electron density on a specific proton or carbon atom in molecules. Hiller et al.8 have reported that the 1H NMR spectrum of the Li2(acac)+ species exhibits signals at δ ) 1.85 and 5.28 ppm versus TMS, signals which were assigned to the methyl and methine protons in acetylacetone, respectively. A sample of tetraethylammonium acetylacetonate in CH3CN exhibits methine resonance at 4.68 ppm, the upfield shift being consistent with the removal of the lithium cation field. Relative 1H chemical shifts of tropolone in acetonitrile have been reported by Bertelli et al.5 Figure 5 shows the changes of the 1H NMR spectra of 1.0 × 10-3 mol dm-3 n-Bu4NC7H5O2 with increasing concentration of LiClO4 in CD3CN. The addition of an equivalent amount of LiClO4 (1.0 × 10-3 mol dm-3) leads to remarkably lower magnetic-field shifts, as compared to the δ values without salt. By further addition of LiClO4, the chemical shifts were observed at lower magnetic-fields (as listed in Table 2). Definitely, 5-H was affected to the most extent by the addition of Li+. Figure 6 shows the changes in δ values of the 3,7-H with increasing LiClO4 concentration in the various solvents. We can notice enhanced increases in δ values (after the 1:1 interaction between Li+ and the tropolonate ion) in the presence of a larger excess of LiClO4 in acetone ((CD3)2CO) as well as in acetonitrile. It is worth pointing out that a nonmetallic salt, Et4NClO4, showed only minor effects on the δ value over the wide concentration range of the salt in CD3CN. In DMF-d7, the δ(3,7-Η) value, 6.410, of the tropolonate ion without LiClO4 being present, increased suddenly to 6.724 on addition of an

1.0 × 10-3

0.01

0.1

0.5

4,6-H 3,7-H 5-H

6.724 6.386 5.973

CD3CN 7.132 7.228 6.837 6.947 6.532 6.678

7.289 7.009 6.766

7.390 7.127 6.891

4,6-H 3,7-H 5-H

6.593 6.342 5.794

(CD3)2CO 7.036 7.108 6.769 6.837 6.383 6.475

7.202 6.970 6.628

7.287 7.097 6.740

4,6-H 3,7-H 5-H

6.687 6.410 5.926

DMF-d7 7.055 7.066 6.724 6.735 6.382 6.399

7.070 6.744 6.405

7.105 6.805 6.457

4,6-H 3,7-H 5-H

6.660 7.018 7.178

CD3OD 6.709 6.735 7.041 7.042 7.225 7.263

6.740 7.035 7.270

6.760 7.041 7.291

Et4NClO4 in CD3CN [Et4NClO4]/mol dm-3 4,6-H 3,7-H 5-H

0c

1.0 × 10-3

0.01

0.1

0.5

(6.730) (6.390) (5.978)

6.733 6.395 5.983

6.739 6.401 5.991

6.738 6.397 5.992

6.791 6.443 6.053

ppm versus TMS as internal reference. b 1.0 × 10-3 mol dm-3 n-Bu4NC7H5O2. c A different run for no salt, cf., the LiClO4 part. a

Figure 6. Changes in 1H NMR chemical shift values, δ(3,7-H), of 1.0 × 10-3 mol dm-3 n-Bu4NC7H5O2 with increasing concentration of LiClO4 in various solvents: (]) CD3OD; (O) CD3CN; (4) (CD3)2CO; (0) DMF-d7; (b) Et4NClO4 in CD3CN.

equivalent amount of Li+; however, no further shifts were observed with the addition of a large excess of LiClO4. In CD3OD, a very large δ value (7.018) was given for 3,7-H of the tropolonate ion itself. On the addition of an equivalent amount of Li+ to the tropolonate ion, the δ value increased slightly, the value remaining constant with further addition of LiClO4. Methanol as the bulk solvent has not only a strong donicity (DNbulk ) 33.8)33 but also a large acceptivity (AN ) 42.3).31 The carbonyl oxygen atoms of the tropolonate ion ought to be well solvated strongly in methanol and lead to the lower magnetic-field shift. The Li+ ion will have limited possibility to interact, except by electrostatic attraction, with the solvated tropolonate ion, because the Li+ ion also is strongly solvated in this solvent. Figure 7 shows the changes in 13C NMR chemical shifts of 1.0 × 10-2 mol dm-3 n-Bu4NC7H5O2 with increasing concen-

1764 J. Phys. Chem. B, Vol. 111, No. 7, 2007

Figure 7. Changes in 13C NMR chemical shift values (δ) of 1.0 × 10-2 mol dm-3 n-Bu4NC7H5O2 with increasing concentration of LiClO4 (0.20 mol dm-3 and more) in CD3CN: (b) 1,2-C; (0) 4,6-C; (4) 3,7C; (O) 5-C.

SCHEME 3: Formal Resemblance between Dilithium Tropolonate, (Li+)2L-, and the Protonated Tropolone, H2L+

tration of LiClO4 in CD3CN. The tropolonate ion gave four 13C NMR signals at 184.4, 134.6, 121.8, and 114.2 ppm for 1,2-, 4,6-, 7,3-, and 5-C, respectively, an order being partially different from that for tropolone (1,2-, 4,6-, 5-, and 7,3-C). We may note that the 1,2-C signal of the tropolonate ion appeared at an unexpectedly lower magnetic-field (184.4 ppm) than that of tropolone (172.7 ppm), while other signals of the tropolonate showed normal shifts toward higher magnetic-fields upon the deprotonation from tropolone (vide infra). The extensive precipitation by addition of less than 0.2 mol dm-3 LiClO4 kept us from observing 13C NMR signals; the signals, however, could be detected with 0.2 mol dm-3 or more of LiClO4 (cf., Figure 2): all of the signals, except for 1,2-C, shifted toward lower magnetic-fields. The 1,2-C signal shifted slightly toward higher magnetic-fields, despite that positive charged species (Li+) are interacting with the tropolonate ion. Protons (H+) caused similar but more dramatic shifts for the 1,2-C signal. We would like to point out that the signal order for 7,3- and 5-C was not altered by addition of LiClO4, even for concentrations up to 1.0 mol dm-3. The “normal” order of tropolone was realized by addition of proton donors, such as CF3SO3H or CH3SO3H, to the tropolonate ion (vide infra). The formal resemblance between dilithium tropolonate, (Li+)2L-, and the protonated tropolone is illustrated in Scheme 3; a comparison of lithium and hydrogen bonds has been discussed.39 Effects of Li+ on the Deprotonation of 4-Isopropyltropolone (Hinokitiol) in the Presence of Various Bases. In acetonitrile, 4-isopropyltropolone (hinokitiol) gave three absorption peaks at 349 nm (/cm-1 mol-1 dm3 ) 6.1 × 103), 321 nm ( ) 6.6 × 103), and 238 nm ( ) 3.1 × 104) in the UVvisible absorption spectrum. Hinokitiol is a weak acid (pKa ) ca. 7.0,40 cf., 6.6841 for tropolone) in aqueous solution; therefore, it cannot dissociate in the aprotic solvent. Figure 8 shows the

Hojo et al.

Figure 8. Effects of LiClO4 on the UV-visible absorption spectra (1.0 cm path-length) of 2.0 × 10-4 mol dm-3 hinokitiol containing 2.0 × 10-4 mol dm-3 Et3N in acetonitrile: (1 -) no base; (2 - - -) 0; (- - -) 2.0 × 10-5; (- - -) 1.0 × 10-4; (- - - -) 2.0 × 10-4; (-) 1.0 × 10-3; (- - -) 0.010; (7 - - -) 0.10 mol dm-3 LiClO4.

effects of LiClO4 on the absorption spectra of hinokitiol in the coexistence of an equivalent amount of Et3N (2.0 × 10-4 mol dm-3). The spectrum indicates no proton transfer from hinokitiol to Et3N without LiClO4 being added. However, upon the addition of only 2.0 × 10-5 mol dm-3 LiClO4, an ionized hinokitiol species was observed at around 400 nm. The band developed to give two peaks of 397 and 389 nm with 2.0 × 10-4 mol dm-3 LiClO4. The intensity of the band reached its maximum at 1.0 × 10-3 mol dm-3 LiClO4, that is, a 5-fold amount as compared to the hinokitiol concentration. The peak at 389 nm increased at the expense of the peak at 397 nm with increasing concentration of LiClO4 up to 0.01 mol dm-3; the peak shifted further to 384 nm with 0.1 mol dm-3 LiClO4. The effect of LiClO4 on the absorption spectra of hinokitiol-Et3N was consistent with that for tropolone-Et3N in acetonitrile (cf., ref 21). Effects of a stronger base, 1,1,3,3-TMG, were also examined; the presence of an equivalent amount of 1,1,3,3-TMG in 2.0 × 10-4 mol dm-3 hinokitiol-MeCN solution caused a partial proton-transfer from hinokitiol to the base. In the presence of an equivalent amount of the base, the band peak at 410 nm was shifted to ca. 400 nm by the addition of 2.0 × 10-5 mol dm-3 LiClO4. A maximum formation of (Li+)(BH+)L′- species (λmax ) 397 nm) was achieved for concentrations as low as 2.0 × 10-4 mol dm-3 (cf., at 1.0 × 10-3 mol dm-3 for Et3N as the base). The band peak shift to 384 nm with 0.1 mol dm-3 LiClO4 indicated the formation of (Li+)2L′-. The addition of an equivalent amount of a strong base, Me4NOH, to a hinokitiolMeCN (2.0 × 10-4 mol dm-3) solution caused an increase in the concentration of the 4-isopropyltropolonate ion; with 0.1 mol dm-3 LiClO4, a peak at 384 nm appeared. Figure 9 shows the changes in UV-visible absorption spectra by the addition of LiClO4 to 2.0 × 10-4 mol dm-3 hinokitiol in the presence of 2.0 × 10-4 mol dm-3 Et3N in DMF. We can notice that only a small fraction of hinokitiol was deprotonated even without the base being present in DMF. The addition of an equivalent amount of Et3N, however, did not alter the spectrum (no increase in deprotonation). The band with 401 nm peak increased monotonously with increasing concentration of LiClO4 up to 0.1 mol dm-3 or higher, showing a clear isosbestic point at 367 nm (for >360 nm). Thus, only (Li+)(BH+)L′- species and no formation of (Li+)2L′- took place in DMF; cf., the tropolonate ion in DMF gave no (Li+)2L-type species with increasing concentration of LiClO4 (vide supra). In ethanol (DNbulk ) 30.4,33 AN ) 37.1),31 the addition of LiClO4 to a hinokitiol-Et3N (2.0 × 10-3 mol dm-3) solution caused a monotonous increase of a peak at 391 nm with

Formation of Protonated Tropolones in Acetonitrile

J. Phys. Chem. B, Vol. 111, No. 7, 2007 1765

Figure 9. Effects of LiClO4 on the UV-visible absorption spectra (1.0 cm path-length) of 2.0 × 10-4 mol dm-3 hinokitiol containing 2.0 × 10-4 mol dm-3 Et3N in DMF: (1 -) no base; (2 - - -) 0; (- - -) 1.0 × 10-4; (- - -) 2.0 × 10-4; (5 -) 1.0 × 10-3; (- - -) 0.010; (7 - - -) 0.10 mol dm-3 LiClO4. Figure 11. Changes in 1H NMR chemical shift values (δ) of 1.0 × 10-3 mol dm-3 hinokitiol containing 1.0 × 10-3 mol dm-3 with increasing concentration of LiClO4 in CD3CN: (O) no base; (2) without LiClO4; (b) LiClO4; (]) Et4NClO4. In the figure, (a), (b), (c), and (d) show the δ values for 6-, 3-, 7-, and 5-H of hinokitiol, respectively.

Figure 10. 1H NMR spectra of 1.0 × 10-3 mol dm-3 hinokitiol containing 1.0 × 10-3 mol dm-3 Et3N in CD3CN: (a) no base; (b) 0; (c) 0.10 mol dm-3 LiClO4.

increasing LiClO4 concentration from 1.0 × 10-4 to 0.1 mol dm-3. The final product must be (Li+)(BH+)L′- or the ion pair (Li+L′-) in the protic solvent. In benzonitrile (DN ) 11.9, AN ) 15.5),31 on the other hand, the formation of (Li+)2L′- species was strongly suggested by the increase of a band at 385 nm with 0.01-0.1 mol dm-3 LiClO4 under similar conditions. Tropolone-triethylamine equilibria in various solvents, observed by IR and 1H NMR spectroscopic methods, have been reported.42 Figure 10 shows the 1H NMR spectra of hinokitiol (1.0 × 10-3 mol dm-3) in CD3CN under different conditions: (a) in the absence of Et3N, (b) in the presence of an equivalent amount of Et3N, and (c) 0.1 mol dm-3 LiClO4 in coexistence with the base. Although δ values were just slightly shifted to higher magnetic-fields in the coexistence with Et3N, they were definitely shifted to higher magnetic-fields by addition of 0.1 mol dm-3 LiClO4. Figure 11 shows astonishing changes in the δ values of hinokitiol-Et3N (1.0 × 10-3 mol dm-3) in the presence of a wide concentration range of LiClO4 in CD3CN. Upon the addition of small amounts of LiClO4 (0.5, 1.0, and 5.0 × 10-3 mol dm-3), the signals were extensively shifted to higher magnetic-fields; however, they shifted in the opposite direction with more than 0.01 mol dm-3 LiClO4 being added and approached the original δ values of hinokitiol. The shifts in the δ values toward higher magnetic-fields indicate a proton transfer from tropolone to Et3N by the addition of LiClO4 to give (Li+)(Et3NH+)L′-, that is, less positive charge on the tropolone-H atoms. In contrast, the successive shifts toward lower magnetic-fields should correspond to the replacement of Et3NH+ on the tropolonate ion with a second Li+ to give (Li+)2L′-, the “triple cation” from (or among) Li+ and a ligand anion. A

Figure 12. Changes in 1H NMR chemical shift values (δ) of 1.0 × 10-3 mol dm-3 hinokitiol containing 1.0 × 10-3 mol dm-3 with increasing concentration of LiClO4 in DMF-d7: (O) no base; (2) without LiClO4; (b) LiClO4. In the figure, (a), (b), (c), and (d) show the δ values for 6-, 3-, 7-, and 5-H of hinokitiol, respectively.

nonmetallic salt, Et4NClO4, at least up to 0.5 mol dm-3, gave almost no effect on the δ values (cf., Figure 11). When a stronger base, 1,1,3,3-TMG, was used, the δ values shifted substantially toward higher magnetic-fields. Nevertheless, with increasing concentration of LiClO4 (more than 0.01 mol dm-3), the δ values shifted toward lower magnetic-fields after δ minimum values at 5 × 10-3 mol dm-3 LiClO4, in accordance with what was observed when using Et3N as the base. In DMF-d7 (cf., Figure 12), the changing behavior of the δ values (of hinokitiol in the coexistence of Et3N) with increasing concentration of LiClO4 was much different from that in CD3CN. After decreasing sharply with up to 0.01 mol dm-3 LiClO4, the δ values remained constant or decreased only slightly; that is, no increase in the δ values was observed. A tropolone-Et3N system (both 1.0 × 10-3 mol dm-3) was also examined by 1H NMR in CD3CN. The δ values of 4,6-, 3,7-, and 5-H of tropolone shifted as for 6-, 3-, 7-, and 5-H of hinokitiol in CD3CN under similar conditions. However, the δ(4,6-H) value at 0.5 mol dm-3 LiClO4 exceeded even the original δ value in the absence of both Et3N and LiClO4: δ(4,6-H) ) 7.471 ppm at 0.5 mol dm-3 LiClO4, cf., δ(4,6-H) )

1766 J. Phys. Chem. B, Vol. 111, No. 7, 2007

Hojo et al.

Figure 13. Changes in 13C NMR chemical shift values (δ) of 5.0 × 10-3 mol dm-3 tropolone with increasing concentration of CF3SO3H or CH3SO3H in CD3CN: (O) 1,2-C, (0) 4,6-C; (]) 5-C; (4) 3,7-C. Open and solid symbols show CF3SO3H and CH3SO3H, respectively.

7.449 in the absence of both Et3N and LiClO4. The δ values for 3,7- and 5-H with higher LiClO4 concentrations approached the original values, but did not exceed these. The reversal in the δ(4,6-H) values is not a wonder, but it suggests that the negative charge density of 4,6-H in (Li+)2L- must be less than that of HL (L-: the tropolonate ion). Formation of Protonated Tropolones in Acetonitrile. With increasing concentration of CF3SO3H in a 1.0 × 10-3 mol dm-3 tropolone-MeCN solution, the absorption peaks at 368, 352, and 320 nm decreased while the peaks at 337 nm ( ) 6.4 × 103), 318 nm ( ) 6.1 × 103), and 302 nm ( ) 6.3 × 103) increased, giving several isosbestic points, for example, at 345 and 290 nm. The spectral changes demonstrated that a 2-fold amount of CF3SO3H to tropolone was needed to achieve almost complete protonation of tropolone in MeCN. In 2 mol dm-3 HClO4 aqueous solution, the formation of the protonated tropolone (or the dihydroxytropylium ion), H2L+, has been indicated by a UV spectral change.41 While trifluoromethanesulfonic acid is a strong acid in aqueous solution, it is only a fairly stronger acid in MeCN (pKa ) 2.60).43 Hosoya and Nagakura44 have determined the basicity of tropolone (HL) (or the acidity of H2L+) to be pKa ) -0.86 in sulfuric acid solutions. The addition of CF3SO3H to a 1.0 × 10-3 mol dm-3 tropolone-CD3CN solution caused the δ values in 1H NMR to shift toward lower magnetic-fields. The δ value changes with various concentration of CF3SO3H also verified that a 2-fold amount of the acid was nearly sufficient to give the protonated tropolone in CD3CN. The effects of CF3SO3H on UV-visible and 1H NMR spectra of 4-isopropyltropolone were most similar to those on tropolone. In the UV-visible spectrum of 4-isopropyltropolone, a new peak at 336 nm ( ) 7.7 × 103) developed at the expense of the peaks at 348 and 321 nm. The 1H NMR chemical shift values for tropolone and 4-isopropyltropolone in the presence of CF3SO3H are listed in Table S1. It is natural that the chemical shift values in 1H NMR of 1.0 × 10-3 mol dm-3 n-Bu4NC7H5O2 also will be enlarged with increasing CF3SO3H concentration in CD3CN. The addition of 1.0 × 10-3 mol dm-3 CF3SO3H to 1.0 × 10-3 mol dm-3 tropolone ion may produce the tropolone molecule in CD3CN. The changes in δ values clearly indicated the formation of the protonated tropolone, H2L+, from the tropolonate ion (L-) and the rather strong acid in acetonitrile. The chemical shifts in 13C NMR of 5.0 × 10-3 mol dm-3 tropolone in CD3CN with increasing concentrations of CF3SO3H are shown in Figure 13 (and Table S2). Imashiro et al.2j have

Figure 14. Changes in 13C NMR chemical shift values (δ) of 5.0 × 10-3 mol dm-3 n-Bu4NC7H5O2 with increasing concentration of CH3SO3H in CD3CN: (b) 1,2-C; (0) 4,6-C; (O) 5-C; (2) 3,7-C.

reported the 13C NMR chemical shift values of tropolone in DMSO-d6. Upon the addition of CF3SO3H, the signals of 4,6-, 5-, and 3,7-C all shifted toward lower magnetic-fields; the signal of 1,2-C, however, shifted the opposite direction. The δ(5-C) value shifted most with the complete formation of H2L+. The analyses of the tangent lines in Figure 13 verified the 1:1 reaction between tropolone (HL) and the acid (HA) to give the H2L+ species. Much higher acid concentrations were needed to produce the H2L+ species when a weaker acid, CH3SO3H, in acetonitrile was used as the proton donor (>0.2 mol dm-3, cf., Table S2). It must be a natural behavior that the δ value shifts toward lower magnetic-fields with increasing concentration of proton donor. However, it is not an easy task to understand the higher magnetic-field shifts of the 1,2-C with the addition of H+. In ab initio calculations with the MP2/6311+G(d,p)//MP2/6-31+G(d,p) level,3e the negative charge densities of 1-C in tropolone and in the protonated tropolone are reported to be 5.660 and 5.661, respectively, though. At any rate, more precise calculations with suitable configurations are highly desired; we performed ab initio calculations with a Gaussian 03 quantum chemistry package (vide infra). As for carboxylic acids in aqueous solution, it is known that ionization of carboxylic acids results in downfield shifts for the carboxyl carbons in the 13C NMR chemical shifts.45 Further exciting features were observed in the 13C NMR chemical shift of the tropolonate ion with increasing concentration of a proton donor. Figure 14 shows the 13C NMR chemical shift values of 5.0 × 10-3 mol dm-3 n-Bu4NC7H5O2 in the presence of various concentrations of methanesulfonic acid. In the presence of 2.5 × 10-3 mol dm-3 CH3SO3H, although the 1,2-signal shifted toward higher magnetic-fields (∆δ ) - 4.4 ppm), the 5-C signal shifted remarkably (∆δ ) +7.8 ppm) toward lower magnetic-fields, while other signals shifted to small extents (∆δ < 2.0). On the addition of 5.0 × 10-3 mol dm-3 CH3SO3H, the 5-C signal (127.5 ppm) finally exceeded the 3,7-C signal (124.2 ppm). The relatively weaker intensity of the 5-C signal allowed us to distinguish rather easily this signal from the 3,7- or 4,6-C signal. At the same time, the 1,2-C signal shifted toward higher magnetic-fields to a great extent (∆δ ) -10.2 ppm, as compared to the value without the proton donor). With 0.01 mol dm-3 CH3SO3H, the conversion to tropolone from the tropolonate ion appeared to be complete as judged by the chemical shift values of the product and those of

Formation of Protonated Tropolones in Acetonitrile

J. Phys. Chem. B, Vol. 111, No. 7, 2007 1767 HL. Therefore, it is concluded that the theoretical results support the suggested reaction schemes for tropolone and n-Bu4NC7H5O2. Acknowledgment. Quantum chemical calculations were performed in part at the Research Center for Computational Science (RCCS) of the Okazaki National Research Institutes. M.K. is indebted to the Research Fellowship for Young Scientists from Japan Society for the Promotion of Science (JSPS).

Figure 15. Calculated 13C NMR chemical shift values (δ) of L-, HL, and H2L+: (O) 1,2-C; (0) 4,6-C; (]) 5-C; (4) 3,7-C.

TABLE 3: Calculated Magnetic Shielding Constants and Chemical Shiftsa of Tropolonate (L-), Tropolone (HL), and Protonated Tropolone (H2L+) in ppm chemical shifts δ

shielding constants 1,2-C 4,6-C 5-C 3,7-C

L-

HL

H2L+

L-

HL

H2L+

1.10 54.22 83.19 67.81

12.52 46.12 60.34 64.53

21.76 36.31 41.28 52.13

188.02 134.90 105.93 121.31

176.60 143.00 128.79 124.59

167.36 152.81 147.84 136.99

a Magnetic shielding constant of carbons of TMS is calculated to be 189.12 ppm.

tropolone in CD3CN. The 1:1 neutralization between L- and HA may partly be affected by the homoconjugation reaction,

L- + 2HA / HL + HA2-

(5)

because methanesulfonic acid is only a weak acid in acetonitrile, pKa ) 10.046 or 8.36,43 and the equilibrium constant for the homoconjugation reaction, K(HA2-), is quite large, log K(HA2-) being 3.8546 or 2.9.43 Theoretical Calculation of 13C NMR Chemical Shifts for L , HL, and H2L+. Geometry optimizations for L-, HL, and H2L+ have been performed with the density functional theory (DFT) employing the PBE1PBE hybrid functional.47 Dunning’s correlation-consistent triple-ζ (cc-pVTZ) basis set48 was adopted. Magnetic shielding tensors of these molecules at the optimized geometries were evaluated by using the gauge-including atomic orbital (GIAO) method49 at the same level of theory. To evaluate 13C chemical shifts, the magnetic shielding calculation of the tetramethylsilane (TMS) standard was also performed. All calculations were carried out by use of the Gaussian 03 quantum chemistry package.50 Figure 15 shows the calculated 13C NMR chemical shifts of L-, HL, and H2L+. Because the experimental signals due to (1-C and 2-C), (3-C and 7-C), or (4-C and 6-C) in HL and H2L+ as well as in L- cannot be distinguished, only the averaged chemical shifts of these pairs are shown in the figure. The calculated magnetic shielding constants and chemical shifts for L-, HL, and H2L+ are listed in Table 3. As the number of protons bonded to oxygen atoms increases, the calculated chemical shift of (1-C, 2-C) decreases. The opposite is true for (3-C, 7-C) and (4-C, 6-C). This behavior agrees with the experimental results. The changes in the experimental chemical shifts shown in Figure 13 correspond to the calculated changes from HL to H2L+. Furthermore, the changes shown in Figure 14 accord with the calculated ones from L- to H2L+ through

Supporting Information Available: Tables S1 and S2 for the experimental 1H and 13C chemical shift values of the tropolonate ion, tropolone, or 4-isopropyltropolone in the presence of proton donors. This material is available free of charge via the Internet at http://pubs.acs.org. References and Notes (1) Siedle, A. R. In ComprehensiVe Coordination Chemistry; Wilkinson, G., Gillard, R. D., McCleverty, J. A., Eds.; Pergamon: Oxford, 1987; Vol. 2, p 397. (2) (a) Martinez-Richa, A.; Joseph-Nathan, P. Solid State Nucl. Magn. Reson. 2003, 23, 119. (b) Redington, R. L.; Sams, R. L. Chem. Phys. 2002, 283, 135. (c) Tekely, P.; Reichert, D.; Zimmermann, H.; Lus, Z. J. Magn. Reson. 2000, 145, 173. (d) Olender, Z.; Reichert, D.; Muller, A.; Zimmermann, H.; Poupko, R.; Luz, Z. J. Magn. Reson., Ser. A 1996, 120, 31. (e) Wang, J.; Gilson, D. F. R. Spectrochim. Acta, Part A 1995, 51, 1683. (f) Takegoshi, K.; Hikichi, K. J. Am. Chem. Soc. 1993, 115, 9747. (g) Titman, J. J.; Luz, Z.; Spiess, H. W. J. Am. Chem. Soc. 1992, 114, 3756. (h) Mori, A.; Sugiyama, S.; Mametsuka, H.; Tomiyasu, K.; Takeshita, H. Bull. Chem. Soc. Jpn. 1987, 60, 4339. (i) Szeverenyi, N. M.; Bax, A.; Maciel, G. E. J. Am. Chem. Soc. 1983, 105, 2579. (j) Imashiro, F.; Maeda, S.; Takegoshi, K.; Terao, T.; Saika, A. Chem. Phys. Lett. 1982, 92, 642. (k) Jackman, L. M.; Trewella, J. C.; Haddon, R. C. J. Am. Chem. Soc. 1980, 102, 2519. (3) (a) Redington, R. L. J. Phys. Chem. A 2006, 110, 1600. (b) Casadesus, R.; Vendrell, O.; Moreno, M.; Lluch, J. M. Chem. Phys. Lett. 2005, 405, 187. (c) Burns, L. A.; Murdock, D.; Vaccaro, P. H. J. Chem. Phys. 2006, 124, 204307. (d) Nakai, H.; Sodeyama, K. J. Mol. Struct. (THEOCHEM) 2003, 637, 27. Nakai, H.; Sodeyama, K. Chem. Phys. Lett. 2002, 365, 203. (e) Mo, O.; Yanez, M.; Essenfar, M.; Herreros, M.; Natario, R.; Abboud, J. L.-M. J. Org. Chem. 1997, 62, 3200. (f) Redington, R. L.; Bock, C. W. J. Phys. Chem. 1991, 95, 10284. (g) Sekiya, H.; Nagashima, Y.; Tsuji, T.; Nishimura, Y.; Mori, A. J. Phys. Chem. 1991, 95, 10311. (h) Rios, M. A.; Rodriguez, J. Can. J. Chem. 1991, 69, 201. (4) Weiler, L. Can. J. Chem. 1972, 50, 1975. (5) Bertelli, D. J.; Andrews, T. G., Jr.; Crews, P. O. J. Am. Chem. Soc. 1969, 91, 5286. (6) Stability Constants of Metal-Ion Complexes, Part B, Organic Ligands; IUPAC Chemical Data Series, No. 22; Perrin, D. D., compiled; Pergamon: Oxford, 1979; p 479. (7) Raban, M.; Yamamoto, G. Inorg. Nucl. Chem. Lett. 1976, 12, 946. Raban, M.; Noe, E. A.; Yamatoto, G. J. Am. Chem. Soc. 1977, 99, 6527. (8) In the SciFinder Scholar system, a paper has been obtained by three technical terms, dilithium, complex, and diketonate: Hiller, L. K., Jr.; Cockrell, J. R., Jr.; Murray, R. W. J. Inorg. Nucl. Chem. 1969, 31, 765. (9) Murray, R. W.; Hiller, L. K., Jr. Anal. Chem. 1967, 39, 1221. (10) Itabashi, E. J. Electroanal. Chem. 1972, 36, 179. (11) Regis, A.; Corset, J. Chem. Phys. Lett. 1975, 32, 462. (12) (a) Mukherjee, L. M.; Kelly, J. J.; Richards, M.; Lukacs, J. M., Jr. J. Phys. Chem. 1969, 73, 580. Mukherjee, L. M.; Lukacs, J. M., Jr. J. Phys. Chem. 1969, 73, 3115. (b) Ahrland, S.; Ishiguro, S.; Persson, I. Acta Chem. Scand. A 1986, 40, 418. (c) Salomon, M. Electrochim. Acta 1990, 35, 509. (13) Fuoss, R. M.; Kraus, C. A. J. Am. Chem. Soc. 1933, 55, 2387. Fuoss, R. M. Chem. ReV. 1935, 17, 27. Kraus, C. A. Science 1939, 90, 281. (14) Hojo, M.; Imai, Y. Bull. Chem. Soc. Jpn. 1983, 56, 1963. (15) Hojo, M.; Takiguchi, T.; Hagiwara, M.; Nagai, H.; Imai, Y. J. Phys. Chem. 1989, 93, 955. Hojo, M.; Hasegawa, H.; Miyauchi, Y.; Moriyama, H.; Yoneda, H.; Arisawa, S. Electrochim. Acta 1994, 39, 629. Hojo, M.; Hasegawa, H.; Hiura, N. J. Phys. Chem. 1996, 100, 891. Hojo, M.; Hasegawa, H.; Morimoto, Y. Anal. Sci. 1996, 12, 521. (16) Chen, Z.; Hojo, M. J. Phys. Chem. B 1997, 101, 10896. Hojo, M.; Ueda, T.; Chen, Z.; Nishimura, M. J. Electroanal. Chem. 1999, 468, 110. (17) Hojo, M.; Hasegawa, H.; Tsurui, H.; Kawamura, K.; Minami, S.; Mizobe, A. Bull. Chem. Soc. Jpn. 1998, 71, 1619. (18) Hojo, M.; Ueda, T.; Yamasaki, M. J. Org. Chem. 1999, 64, 4939. (19) Hojo, M.; Ueda, T.; Yamasaki, M.; Inoue, A.; Tokita, S.; Yanagita, M. Bull. Chem. Soc. Jpn. 2002, 75, 1569.

1768 J. Phys. Chem. B, Vol. 111, No. 7, 2007 (20) Hojo, M.; Hasegawa, H.; Yoneda, H. Bull. Chem. Soc. Jpn. 1996, 69, 971. (21) Hojo, M.; Hasegawa, H.; Yoneda, H. J. Chem. Soc., Perkin Trans. 2 1994, 1855. (22) Hojo, M.; Hasegawa, H.; Mizobe, A.; Ohkawa, Y.; Miimi, Y. J. Phys. Chem. 1995, 99, 16609. (23) Olsher, U.; Izatt, R. M.; Bradshaw, J. S.; Dalley, N. K. Chem. ReV. 1991, 91, 137. (24) Manege, L. C.; Ueda, T.; Hojo, M. Bull. Chem. Soc. Jpn. 1998, 71, 589. Manege, L. C.; Ueda, T.; Hojo, M.; Fujio, M. J. Chem. Soc., Perkin Trans. 2 1998, 1961. Hojo, M.; Ueda, T.; Inoue, S.; Kawahara, Y. J. Chem. Soc., Perkin Trans. 2 2000, 1735. Hojo, M.; Ueda, T.; Ueno, E.; Hamasaki, T.; Fujimura, D. Bull. Chem. Soc. Jpn. 2006, 79, 751. (25) Hojo, M. Bunseki Kagaku 2004, 53, 1279. Hojo, M. KharkoV UniVersity Bulletin (Ukraine), No. 626, Chemical Series 2004, 11, 47. (26) Hojo, M.; Ueda, T.; Kawamura, K.; Yamasaki, M. Bull. Chem. Soc. Jpn. 2000, 73, 347. Hojo, M.; Ueda, T.; Inoue, A. Bull. Chem. Soc. Jpn. 2002, 75, 2629. (27) Binder, D. A.; Kreevoy, M. M. J. Phys. Chem. 1994, 98, 10008. (28) Hojo, M.; Miyauchi, Y.; Nakatani, I.; Mizobuchi, T.; Imai, Y. Chem. Lett. 1990, 1035. Hojo, M.; Watanabe, A.; Mizobuchi, T.; Imai, Y. J. Phys. Chem. 1990, 94, 6073. Hojo, M.; Miyauchi, Y.; Imai, Y. Bull. Chem. Soc. Jpn. 1990, 63, 3288. Hojo, M.; Miyauchi, Y.; Tanio, A.; Imai, Y. J. Chem. Soc., Faraday Trans. 1991, 87, 3847. Miyauchi, Y.; Hojo, M.; Ide, N.; Imai, Y. J. Chem. Soc., Faraday Trans. 1992, 88, 1425. Miyauchi, Y.; Hojo, M.; Moriyama, H.; Imai, Y. J. Chem. Soc., Faraday Trans. 1992, 88, 3175. Hojo, M.; Hasegawa, H.; Morimoto, Y. J. Phys. Chem. 1995, 99, 6715. (29) Pocker, Y.; Ciula, J. C. J. Am. Chem. Soc. 1988, 110, 2904. (30) Pocker, Y.; Spyridis, G. T. J. Am. Chem. Soc. 2002, 124, 10373. (31) Gutmann, V. The Donor-Acceptor Approach to Molecular Interactions; Plenum: New York, 1978; Chapter 2. (32) Erlich, R. H.; Roach, E.; Popov, A. I. J. Am. Chem. Soc. 1970, 92, 4989. (33) Marcus, Y. J. Solution Chem. 1984, 13, 599. (34) Covington, A. K.; Dickinson, T. Physical Chemistry of Organic SolVent Systems; Plenum: London, 1973; Chapter 1. (35) Doering, W. v. E.; Knox, L. H. J. Am. Chem. Soc. 1951, 73, 828. (36) Olah, G. A.; Calin, M. J. Am. Chem. Soc. 1968, 90, 938.

Hojo et al. (37) Hojo, M.; Ueda, T.; Nishimura, M.; Hamada, H.; Matsui, M.; Umetani, S. J. Phys. Chem. B 1999, 103, 8965. (38) Araki, Y.; Iwase, A.; Kudo, S.; Suzuki, T. M.; Yokoyama, T. Bull. Chem. Soc. Jpn. 1991, 64, 2931. (39) Scheiner, S. In Lithium Chemistry; Sapse, A.-M., Schleyer, P. v. R., Eds.; John Wiley: New York, 1995; Chapter 3. (40) Aulin-Erdtman, G. Acta Chem. Scand. 1950, 4, 1031. (41) Hirai, M.; Oka, Y. Bull. Chem. Soc. Jpn. 1970, 43, 778. (42) Poh, B.-L.; Siow, H.-L. Aust. J. Chem. 1980, 33, 491. (43) Fijinaga, T.; Sakamoto, I. J. Electroanal. Chem. 1977, 85, 185. (44) Hosoya, H.; Nagakura, S. Bull. Chem. Soc. Jpn. 1966, 39, 1414. (45) (a) Hagen, R.; Roberts, J. D. J. Am. Chem. Soc. 1969, 91, 4504. (b) Nagata, C.; Hasegawa, T.; Tanaka, S. J. Chem. Soc. Jpn. 1977, 1087. (46) Kolthoff, I. M.; Chantooni, M. K., Jr. J. Am. Chem. Soc. 1965, 87, 4428. (47) (a) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. ReV. Lett. 1996, 77, 3865. (b) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. ReV. Lett. 1997, 78, 1396. (48) Dunning, T. H., Jr. J. Chem. Phys. 1989, 90, 1007. (49) (a) Ditchfield, R. Mol. Phys. 1974, 27, 789. (b) Wolinski, K.; Hinton J. F.; Pulay, P. J. Am. Chem. Soc. 1990, 112, 8251. (50) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03, revision B.05; Gaussian, Inc.: Wallingford, CT, 2003.