DOI: 10.1002/chem.201404500
Full Paper
& Reaction Kinetics
Quantification of Ion-Pairing Effects on the Nucleophilic Reactivities of Benzoyl- and Phenyl-Substituted Carbanions in Dimethylsulfoxide Francisco Corral-Bautista, Roland Appel, Johanna S. Frickel, and Herbert Mayr*[a] Dedicated to Professor Mieczysław Ma˛kosza on the occasion of his 80th birthday
Abstract: Second-order rate constants for the reactions of acceptor-substituted phenacyl (PhCOCHAcc) and benzyl anions (PhCHAcc) with diarylcarbenium ions and quinone methides (reference electrophiles) have been determined in dimethylsulfoxide (DMSO) solution at 20 8C. By studying the kinetics in the presence of variable concentrations of potassium, sodium and lithium salts (up to 102 mol L1), the influence of ion-pairing on the reaction rates was examined. As the concentration of K + did not have any influence on the rate constants at carbanion concentrations in the range of 104–103 mol L1, the acquired rate constants could be assigned to the reactivities of the
free carbanions. The counter ion effects increase, however, in the series K + < Na + < Li + , and the sensitivity of the carbanion reactivities toward variation of the counter ion strongly depends on the structure of the carbanions. The reactivity parameters N and sN of the free carbanions were derived from the linear plots of log k2 against the electrophilicity parameters E of the reference electrophiles, according to the linear-free energy relationship log k2(20 8C) = sN(N + E). These reactivity parameters can be used to predict absolute rate constants for the reactions of these carbanions with other electrophiles of known E parameters.
Introduction Reactions of carbanions with alkyl halides, Michael acceptors, or carbonyl groups are among the most important methods for the formation of carbon–carbon bonds.[1] In previous work, we studied the nucleophilic reactivities of different types of carbanions[2] and found that pKaH values are not a reliable measure for their relative reactivities.[2b,d,f,g] Since these investigations were mostly performed in DMSO solution at low ion concentrations, ion pairing was neglected. We now examine the concentration range in which this assumption is justified and characterize the nucleophilic reactivities of the benzoyl-substituted carbanions 1 and the analogously substituted benzyl anions 2 (Scheme 1) by determining the rate constants of their reactions with the electrophiles shown in Table 1. The secondorder rate constants are then analyzed by the linear freeenergy relationship Equation (1), where E is an electrophilespecific parameter, and N and sN are solvent-dependent nucleophile-specific parameters.[3] log k2 ð20 CÞ ¼ sN ðN þ EÞ
Scheme 1. Benzoyl- (1-X) and benzyl-substituted carbanions (2-X) and their absorption maxima lmax (in nm) in DMSO. [a] pKaH in DMSO solution. From reference [4]; [b] the nucleophilic reactivities of the anions 2-CO2Et and 2NO2 were reported in references [2b] and [2g]; [c] lmax from reference [2g]; [d] 2-COPh = 1-Ph.
Relationships between structures and nucleophilic reactivities of the carbanions 1 and 2 are discussed.
ð1Þ
[a] Dr. F. Corral-Bautista, Dr. R. Appel, J. S. Frickel, Prof. Dr. H. Mayr Department Chemie, Ludwig-Maximilians-Universitt Mnchen Butenandtstraße 5-13 (Haus F), 81377 Mnchen (Germany) Fax: (+ 49) 89-2180-77717 E-mail: [email protected]
Results and Discussion Preparation of the alkali salts The alkali salts (1-X)M were generated by treatment of the corresponding CH acids with alkali tert-butoxide (LiOtBu,
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/chem.201404500. Chem. Eur. J. 2014, 20, 1 – 11
These are not the final page numbers! ÞÞ
1
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
Full Paper Table 1. Benzhydrylium ions 3 a–f and Michael acceptors 3 g–p employed as reference electrophiles in this study. Electrophile
E[a]
lmax[b] [nm]
R = NMe2 R = N(CH2)4
3a 3b
7.02 7.69
613 620
n=2 n=1
3c 3d
8.22 8.76
618 627
Scheme 3. Possible configurations of b-dicarbonyl enolates.
n=2 n=1
R1 = Ph R1 = Ph R1 = tBu R1 = tBu R1 = tBu R1 = OMe R1 = Me R1 = tBu
R2 = OMe R2 = NMe2 R2 = NO2 R2 = Me R2 = OMe R2 = OMe R2 = NMe2 R2 = NMe2
3e 3f
9.45 635 10.04 630
3g 3h 3i 3j 3k 3l 3m 3n
12.18 13.39 14.36 15.83 16.11 16.38 16.36 17.29
3o
17.90 521
3p
14.68 520
Previous NMR investigations of the sodium salt of pentane2,4-dione (R = R’ = CH3 in Scheme 3) in methanol showed the presence of a mixture of the S- and U-configurations.[6] The absence of the W-configuration can be explained by the steric interaction between the coplanar R and R’ groups. The U/S ratio increased when sodium iodide was added, due to the formation of the sodium complex of the bidentate U-configuration. Addition of more than one equivalent of 18-crown-6 led to the disappearance of the U-configuration, showing that the free enolate ion exists in the S-configuration exclusively; obviously, in the free carbanion the U-configuration is destabilized by the repulsive Coulomb interaction between the negatively charged oxygen atoms. Based on these investigations,[6, 7] we assign the two sets of signals for 1-COMe to the E,Z and Z,E configurations, the two nonidentical S-configurations of unsymmetrical b-dicarbonyl compounds. The increasing NMR shielding of the carbanionic carbon and the attached proton from Li + to Na + and K + reflects the increasing electron density with increasing freedom of the carbanion 1-CO2Et (Scheme 2). Due to their low stability, the potassium salts of the anion of deoxybenzoin (1-Ph) and of the benzyl anions 2-CN, 2-COMe, and 2-SO2Ph, were not isolated, but were generated by deprotonation of the corresponding CH acids with KOtBu or dimsyl potassium (typically 1.05 equiv) in DMSO prior to combining them with the electrophiles. As the UV/Vis absorbances of the carbanions 1-Ph, 2-CN, 2-COMe and 2-SO2Ph reached a limiting value when the corresponding CH acids were treated with 1.05 equivalents of base and did not increase further when additional 2–5 equivalents of base were added, one can conclude that 1.05 equivalents of base are sufficient for their complete deprotonation.
422 533 374 371 393 407 490 486
[a] Electrophilicity-parameters E were taken from references [3b,c,f, 5]; [b] in DMSO solution.
Product studies The benzhydrylium salt 3 eBF4 was added as a solid to equimolar solutions of the potassium salts (1-X)K in DMSO. After decoloration of the blue solutions (typically within one to two minutes), water was added to precipitate the products 4-X. After filtration, the crude products were recrystallized from ethanol or ethanol/CH2Cl2 mixtures and characterized by NMR spectroscopy and mass spectrometry (Scheme 4). Combination of solutions of the potassium salts (1-X)K with solutions of the quinone methides 3 j,k,n in DMSO (with 5–10 % of dichloromethane as cosolvent) and workup with aqueous acetic acid yielded the crude reaction products 5-X which were purified by chromatography and characterized by
Scheme 2. 1H and 13C NMR chemical shifts of the anionic centres of the alkali salts 1-X (0.16 < c < 0.24 mmol L1) in [D6]DMSO. [a] Generated by treatment of the conjugate CH acid with KOtBu in DMSO solution; [b] two isomers in solution (ratio = 1:1.3).
NaOtBu, KOtBu) in ethanol and isolated after washing the precipitated salts with dry diethyl ether (Scheme 2). The anions 1-X gave rise to a single set of NMR signals in [D6]DMSO solution, except for (1-COMe)K, which shows two sets of broad signals in the ratio 1:1.3. For this anion, four possible configurations have to be considered (Scheme 3). &
&
Chem. Eur. J. 2014, 20, 1 – 11
www.chemeurj.org
2
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!
Full Paper
Scheme 4. Reaction of the potassium salts (1-X)K with the benzhydrylium salt 3 eBF4 in DMSO.
Scheme 6. Reactions of the carbanions (2-X)K with the quinone methide 3 n in DMSO at ambient temperature. [a] Determined by 1H NMR spectroscopy after purification by chromatography.
Scheme 5. Reactions of the potassium salts (1-X)K with the quinone methides 3 in DMSO. [a] Determined by 1H NMR spectroscopy after purification by chromatography; [b] could not be determined. Figure 1. Decrease of the absorbance A (at 630 nm) during the reaction of (1-CN)K (1.35 104 mol L1) with 3 f (1.71 105 mol L1) in DMSO at 20 8C. Inset: Correlation of the first-order rate constants kobs with the concentrations of (1-CN)K. Filled circles in absence, empty circles in presence of 2.4 equivalents of 18-crown-6.
NMR spectroscopy and mass spectrometry. The most reactive carbanion 1-Ph was not employed as a stable salt but generated by deprotonation of deoxybenzoin with KOtBu (ca. 1.05 equiv) in DMSO prior to adding the solution of the quinone methide 3 n. In all cases, the products 5-X were obtained in good yields with low diastereoselectivities (Scheme 5). Solutions of the benzyl anions (2-X)K in DMSO were obtained by treatment of the corresponding CH acids with 1.05 equivalents of dimsylK or KOtBu. Addition of the quinone methide 3 n, followed by workup with aqueous acetic acid, gave the products 6-X in good yields as mixtures of two diastereoisomers (Scheme 6).
order rate constants (see below, Tables 4 and 5) were obtained as the slopes of linear correlations of kobs with the concentrations of the nucleophiles (Figure 1). Counter ion effects All carbanions 1-X were obtained as alkali salts by treatment of ketones, b-diketones, b-ketoesters, and related compounds with alkali tert-butoxide. To examine how ion pairing affects the kinetics of their reactions with electrophiles, we determined the first-order rate constants at constant concentrations of the carbanion salts (ca. 104 mol L1) and the electrophiles (ca. 105 mol L1) in the presence of variable concentrations (0– 102 mol L1) of alkali salts (LiCl, LiBF4, NaBPh4, KBPh4, KOTf; representative examples are shown in Table 2; detailed experimental data for further nucleophiles are in the Supporting Information). The measured rate constants were related to the second-order rate constants of the free carbanions (krel = 1), which were taken from Tables 4 and 5 (see below) or from the quoted references. For comparison, we also studied the carbanions derived from the ester 2-CO2Et, the b-diketones 7, the diester 8-CO2Et and the b-ketoester 8-COMe (structures are
Kinetic investigations The kinetics of the reactions of the carbanions 1-X and 2-X with the reference electrophiles 3 were studied in DMSO solution at 20 8C by monitoring the absorptions of the electrophiles with conventional or stopped-flow UV/Vis spectrometers. To simplify the evaluation of the kinetic experiments, the carbanions were used in large excess (> 10 equiv). Thus, their concentrations remained almost constant throughout the reactions, and pseudo-first-order kinetics were obtained in all runs. The first-order rate constants kobs were derived by least-squares fitting of the exponential function At = A0·exp(-kobs t) + C to the time-dependent absorbances At of the electrophiles. SecondChem. Eur. J. 2014, 20, 1 – 11
www.chemeurj.org
These are not the final page numbers! ÞÞ
3
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
Full Paper Table 2. First-order rate constants kobs for the reactions of various carbanions with reference electrophiles in the presence of variable concentrations of alkali salts in DMSO at 20 8C. 100 krel was calculated from the second-order rate constants of the free anions given in Table 4. Reaction
c = 2.12 104 mol L1
c = 3.69 104 mol L1
c = 4.39 104 mol L1
c = 9.53 104 mol L1
c = 6.58 104 mol L1
M
Additive
[Alkali + ]total [mol L1]
K K K K K K K K K
KBPh4 KBPh4 KBPh4 KBPh4 KBPh4 KBPh4 KBPh4 KBPh4
3.51 104 4.80 104 7.36 104 1.48 103 2.79 103 5.27 103 7.79 103 1.16 102
0.761[a] 0.748 0.737 0.729 0.701 0.658 0.631 0.604 0.599
100 98 97 96 92 86 83 79 79 100 71 72 69 62 57 54 49 49
kobs [s1]
NaBPh4 NaBPh4 NaBPh4 NaBPh4 NaBPh4 NaBPh4 NaBPh4 NaBPh4
5.69 104 1.11 103 1.71 103 3.09 103 4.80 103 7.74 103 1.21 102 1.82 102
1.32[a] 0.950 0.935 0.909 0.823 0.757 0.707 0.652 0.649
K Li Li Li Li Li Li Li Li Li Li
none LiBF4 LiBF4 LiBF4 LiBF4 LiBF4 LiBF4 LiBF4 LiBF4 LiBF4
4.65 104 1.37 103 2.27 103 4.08 103 7.69 103 1.13 102 1.74 102 2.66 102 4.30 102 5.93 102
1.58[a] 1.18 1.10 0.976 0.871 0.688 0.616 0.480 0.333 0.178 0.145
100 75 69 62 55 44 39 30 21 11 9
K Li Li Li Li Li Li Li Li Li Li
none LiCl LiCl LiCl LiCl LiCl LiCl LiCl LiCl LiCl
1.01 103 1.51 103 2.01 103 3.21 103 6.05 103 1.11 102 2.12 102 3.46 102 5.14 102 7.65 102
3.42[a] 2.47 2.52 2.47 2.18 1.87 1.43 1.04 0.820 0.706 0.508
100 72 74 72 64 55 42 30 24 21 15
K K K K K K K K K K K K
none KBPh4 KBPh4 KBPh4 KBPh4 KBPh4 KBPh4 KBPh4 KBPh4 KBPh4 KBPh4
6.58 8.68 1.08 1.64 2.41 3.46 5.00 7.89 1.08 1.37 2.08
282[a] 280 260 241 210 175 136 111 82.1 65.2 55.0 41.9
&
Chem. Eur. J. 2014, 20, 1 – 11
www.chemeurj.org
M
Additive
[Alkali + ]total [mol L1]
kobs [s1]
100 krel
c = 6.56 104 mol L1
K K K K K K K K K K K
none KOTf KOTf KOTf KOTf KOTf KOTf KOTf KOTf KOTf
6.56 104 8.63 104 1.07 103 1.69 103 3.15 103 4.40 103 6.32 103 1.01 102 1.39 102 1.77 102
281[a] 273 236 219 178 131 107 82.9 61.0 50.0 43.1
100 97.2 84.0 77.9 63.3 46.6 38.1 29.5 21.7 17.8 15.3
K Na Na Na Na Na Na Na Na Na
NaBPh4 NaBPh4 NaBPh4 NaBPh4 NaBPh4 NaBPh4 NaBPh4 NaBPh4 NaBPh4
6.81 104 8.10 104 1.20 103 1.85 103 3.15 103 7.72 103 1.49 102 2.21 102 3.28 102
229[a] 114 95.0 64.3 41.9 26.9 23.4 15.7 11.8 11.0
100 49.8 41.5 28.1 18.3 11.7 10.2 6.9 5.2 4.8
K Li Li Li Li Li Li Li Li
LiCl LiCl LiCl LiCl LiCl LiCl LiCl LiCl
6.73 104 8.48 104 1.02 103 1.38 103 2.08 103 3.48 103 4.91 103 7.46 103
580[a] 58.1 33.5 28.7 24.7 18.0 11.7 9.56 7.04
100 10 5.8 4.9 4.3 3.1 2.0 1.6 1.2
c = 5.34 104 mol L1
c = 6.48 104 mol L1
[a] Calculated as kobs = k2 [Nu] from the second-order rate constants for the corresponding reactions of the free anions (potassium salts or potassium salts in presence of 18-crown-6) given in Table 4.
100 99.3 92.2 85.5 74.5 62.1 48.2 39.4 29.1 23.1 19.5 14.9
Figure 2. Plots of the relative first-order rate constants for the reactions of the anions derived from b-diketones and a monoketone with electrophiles versus the total concentration of potassium (varied by addition of KBPh4) in DMSO solution at 20 8C.
shown in Figures 2 and 3). The observation that the addition of KOTf to solutions of 1-CO2Et and 1-COMe affects the kinetics in the same way as the addition of KBPh4 (Table 2 and Sup&
Reaction
100 krel
K Na Na Na Na Na Na Na Na
104 104 103 103 103 103 103 103 102 102 102
Table 2. (Continued)
4
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!
Full Paper porting Information) indicates that the observed changes of the reaction rates are due to carbanion-potassium interactions and not caused by interactions of the [BPh4] anion with the electrophiles. In Figure 2–4, we have plotted the dependence of the relative rate constants kobs for the reactions of various carbanions with reference electrophiles on the concentration of the potassium ions ([K + ]), which was varied by addition of different amounts of KBPh4. Figure 2 shows that the reactivities of the anion of the monoketone 1-Ph as well as that of the cyclic diketone 7 b (fixed W configuration) were affected relatively little when up to 12 mmol L1 K + was added (80 and 73 % of the reactivity of the free carbanion). In contrast, the reactivities of the b-diketone-derived carbanions 1-COPh and 7 a decreased noticeably with increasing potassium concentration, and reached a plateau at about 40 % of the original value, probably due to a change from E,Z (S-shape) to Z,Z (U-shape) configurations which can interact more strongly with the potassium ions. The decrease of reactivity due to the addition of KBPh4 was smaller for the b-diketone-derived anion 1-COMe. Figure 3 compares the reaction rates of monoester-, b-ketoester-, diester-, and b-ketoamido derived carbanions at variable concentrations of potassium ions. As for the anion of the
Figure 4. Plots of the relative first-order rate constants for the reactions of the anions of the ketones 1-Ph, 1-CN, 1-SO2Ph, and 1-NO2 with electrophiles versus the total concentration of potassium (varied by addition of KBPh4) in DMSO solution at 20 8C.
tassium reduces the nucleophilic reactivities to a larger extent (lower level of the plateaus). The rate constants for the monocarbonyl-substituted carbanions, which are plotted in Figure 4, depend less on the potassium ion concentration than those of the dicarbonyl-substituted carbanions (Figures 2 and 3). While the reactivities of all diketones and ketoesters decrease to 60–20 % of their original values (except the cyclic diketone 7 b) at a potassium concentration of [K + ] = 5 mmol L1, the reactivity of the anion of benzoyl acetonitrile (1-CN) is the only one in Figure 4 that is affected significantly and is reduced to 60 % at a 5 mmol L1 concentration of potassium. Figure 5 compares the influence of different alkali ions on the relative reactivities of 1-Ph and 1-CO2Et. The reactivity of the anion of deoxybenzoin 1-Ph is only slightly affected when the potassium concentration is increased (Figure 5 a). When sodium is added, the reactivity decreases to 50 % of the value of the free carbanion. Addition of lithium chloride or lithium tetrafluoroborate decreases the reactivities to 12 % and 10 %, respectively, at [Li + ] 50 mmol L1 showing a subordinate effect of the anion of the lithium salt. Further increase of the Li + concentration leads to a further decrease of reactivity without reaching a plateau at a concentration of 60–70 mmol L1. For the b-ketoester anion 1-CO2Et, similar plots were observed, albeit with more pronounced effects (Figure 5 b). Addition of KBPh4 decreases the reactivity to 20 % of the value of the free carbanion, while plateaus are reached at about 5 % of the initial reactivity of the free carbanion with 20 mmol L1 NaBPh4 and at 1–2 % with 5 mmol L1 LiCl. Figures 2–4 illustrate that at concentrations of 1 mmol L1 in DMSO, counter-ion effects by K + are negligible, which is confirmed by the observation that the measured pseudo-firstorder rate constants for (1,2)K in the presence and absence of 18-crown-6 fit the same kobs versus [(1,2)K] plots (example shown in Figure 1). Only in the reactions of the anions of the
Figure 3. Plots of the relative first-order rate constants for the reactions of the anions of the ester 2-CO2Et, the b-ketoamide 1-CONEt2, the b-ketoesters 1-CO2Et and 8-COMe, and the diester 8-CO2Et with electrophiles versus the total concentration of potassium (varied by addition of KBPh4) in DMSO solution at 20 8C.
monoketone 1-Ph (Figure 2), the reactivity of the monoesterderived carbanion 2-CO2Et depends only slightly on the concentration of K + . However, in contrast to the behavior of the b-diketones (Figure 2), all anions from b-ketoesters and related compounds (Figure 3) show a significant decrease of reactivity with increasing potassium ion concentration, not only indicating that these carbanions interact with the counter ion K + at lower concentrations (steeper decrease of kobs shows higher equilibrium constants) but also that the coordination with poChem. Eur. J. 2014, 20, 1 – 11
www.chemeurj.org
These are not the final page numbers! ÞÞ
5
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
Full Paper
Figure 6. First-order rate constants for the reaction of (1-CO2Et)Li with 1 e (1.35 105 mol L1, DMSO, 20 8C) and their correlation with the nucleophile concentration.
1 e do not increase linearly with the concentrations of the carbanion, probably because (1-CO2Et)Li forms dimers or higher aggregates in solution, as previously reported for other lithium enolates.[5, 6b,e,f] As the equilibrium between monomer and aggregates shifts towards aggregation by increasing concentration, a nonlinear increase of kobs with increasing [(1CO2Et)Li] was observed. The comparison of the second-order rate constants for the different alkali derivatives of 1-Ph, 1-CO2Et, and 1-CONEt2 (Table 3) shows that the reactivity of the potassium salt of deoxybenzoin (1-Ph)K corresponds to that of the free anion, whereas 1-CO2Et and 1-CONEt2 interact with K + . While the sodium and potassium derivatives of 1-Ph are similarly reactive (slight deviation: reaction with 3 p), the corresponding lithium compound reacts 0.7–0.9 times as fast as the potassium salt (corresponds to free carbanion). In contrast, the second-order rate constants of the sodium derivative of 1-CO2Et are only 30–40 % of those of the free carbanion, almost independent of the nature of the electrophile.
Figure 5. Plots of the relative first-order rate constants (100 krel) for the reactions of a) 1-Ph with 3 n and b) 1-CO2Et with 3 e (Li + ) or 3 f (Na + and K + ) versus the total concentration of alkali metal ions in DMSO solution at 20 8C.
b-ketoamide 1-CONEt2 and of the b-ketoester 1-CO2Et, which showed the strongest counter-ion effects in Figure 3, the addition of 18-crown-6 (18-c-6) increased the second-order rate Table 3. Second-order rate constants k2 for the reaction of different alkali derivatives of 1-Ph, 1-CONEt2 and 1constants by factors of 1.4 and CO2Et with the reference electrophiles 3 in DMSO at 20 8C. 1.2, respectively (Table 3). Analogously, second-orderk2(K + /18-c-6)/k2(K + ) Anion Electrophile k2[a] [L mol1 s1] + + + + Na K K /18-c-6 Li rate constants for the reactions [b] of the sodium salts (1-Ph)Na 3h 9.12 105 1.10 106 1.05 106 4 4 4 [b] 3 p 7.83 10 7.79 10 9.68 10 and (1-CO2Et)Na with electro3 3 3 [b] 3 m 8.89 10 9.59 10 9.88 10 philes could be derived because [b] 3n 2.74 103 3.39 103 3.59 103 the corresponding pseudo-first3 3 3 [b] 3o 1.66 10 1.90 10 1.83 10 order rate constants correlated 3g 3.26 104 4.59 104 1.41 linearly with the concentrations 3 3 4.61 10 1.34 3 h 3.43 10 of the carbanions at concentra2.46 103 1.46 3i 1.69 103 3 1 tions 10 mol L . A linear re3l 50.3 69.7 1.38 lationship between the first3e 4.27 105 8.14 105 8.95 105 1.10 order rate constants and the [c] 5 5 5 1.84 10 3.63 10 4.28 10 1.18 3 f — concentrations of the carbanion 7.18 103 8.67 103 1.21 3g 2.63 103 was also found for the reactions 7.18 102 8.92 102 1.24 3h 2.45 102 of (1-Ph)Li thus giving rise to 3j 5.64 17.2 19.9 1.16 3k 3.67 8.84 11.5 1.30 the second-order rate constants 3 1 listed in Table 3. Figure 6, in con[a] Derived from first-order rate constants determined at [1-X] < 2.3 10 mol L . [b] Rate constants in presence and in absence of 18-crown-6 (18-c-6) are identical. [c] k2 could not be determined because the dependtrast, reveals that the pseudoence of the pseudo-first-order rate constants on the concentration of the carbanion followed a nonlinear relafirst-order rate constants for the tion (Figure 6). reactions of (1-CO2Et)Li with &
&
Chem. Eur. J. 2014, 20, 1 – 11
www.chemeurj.org
6
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!
Full Paper Table 4. Second-order rate constants k2 for the reactions of the carbanions 1-X with the reference electrophiles 3 in DMSO at 20 8C. 1-X
k2 [L mol1 s1]
N (sN)[a]
Electrophile
23.15 (0.60)
3h 3p 3m 3n 3o
1.05 106 9.68 104 9.88 103 3.59 103 1.83 103
3g 3h 3i 3l
4.59 104 4.61 103 2.46 103 6.97 101
17.52 (0.74)
3e 3f 3g 3h 3j 3k
8.95 105 4.28 105 8.67 103 8.92 102 1.99 101 1.15 101
16.55 (0.78)
3e 3f 3g 3h 3j 3k
3.21 105 1.28 105 3.65 103 2.60 102 3.81 2.02
16.03 (0.86)
3e 3f 3g 3h
4.79 105 1.22 105 1.77 103 1.98 102
17.46 (0.65)
3c 3d 3e 3f
1.06 106 4.75 105 1.95 105 6.50 104
17.19 (0.56)
3a 3b 3d 3e 3f
5
5.08 10 2.27 105 4.71 104 2.14 104 1.10 104
13.91 (0.76)
3a 3b 3d 3e 3f
1.49 105 6.83 104 6.10 103 2.33 103 9.49 102
19.28 (0.65)
Table 5. Second-order rate constants k2 for the reactions of the carbanions 2-X with the reference electrophiles 3 in DMSO at 20 8C. 2-X
3n 3o
(3.89 106)[c] (1.96 106)[c]
24.99 (0.60)
3m 3n 3o
1.44 105 4.48 104 1.69 104
25.77 (0.56)
3m 3n 3o
1.95 105 5.84 104 2.66 104
Structure–reactivity relationships Figure 8 compares the influence of the substituents X on the nucleophilicities of benzoyl- (1-X), ethoxycarbonyl- (8-X) and phenyl-substituted (2-X) carbanions. As the relative reactivities of these carbanions depend on the nucleophile-specific sensitivity sN, Table 6 also reports relative reactivities within these reaction series towards a common electrophile. The following discussion shows that the qualitative conclusions drawn from Figure 8 (N-values) and Table 6 (relative rate constants) are identical.
Table 6. Relative reaction rates of benzoylmethyl anions 1-X, ethyl acetate anions 8-X and benzyl anions 2-X towards electrophiles.
X
Ph CONEt2 CO2Et CN COMe COPh SO2Ph NO2
Correlation analysis Linear correlations were obtained for log k2 of the reactions of the carbanions 1 and 2 with the reference electrophiles 3 with their electrophilicity parameters E, as depicted for some repre-
These are not the final page numbers! ÞÞ
k2 [L mol1 s1]
sentative examples in Figure 7. As documented in the Supporting Information, all reactions studied in this work followed analogous linear correlations, indicating that Equation (1) is applicable to these reactions. From the slopes of these correlations, the nucleophile-specific parameters sN were derived, and the negative intercepts on the abscissa (log k2 = 0) correspond to the nucleophilicity parameters N (Tables 4 and 5).
The second-order rate constants for the reactions of the free carbanions 1-X and 2-X, derived at concentrations of 103 mol L1 or in presence of 18-crown-6, with the reference electrophiles 3 in DMSO at 20 8C are summarized in Tables 4 and 5. In the following part, the reactivities of the free carbanions are discussed.
www.chemeurj.org
Electrophile
[a] Nucleophile-specific reactivity parameters N (and sN) as defined by Equation (1); for determination, see below. [b] Generated by deprotonation with dimsyl potassium. [c] Approximate second-order rate constant obtained from a plot of [2-CN] vs. kobs that shows a large negative intercept. [d] Generated by deprotonation with KOtBu.
[a] Nucleophile-specific reactivity parameters N (and sN) as defined by Equation (1); for determination, see below.
Chem. Eur. J. 2014, 20, 1 – 11
N (sN)[a]
+3 f
+3k
7.74 104[a] 1.04 103[a] 4.51 102 1.35 102 1.29 102 6.85 101 1.16 101 1.0
1.44 107[a,b] 2.28 103[c] 9.38 102[c] 3.52 102[c] 5.17 101 1.0[a,d]
+3n
9.95 104[b] 7.02 105 8.09 103 6.48 102 1.05 104 1.0[e]
[a] Rate constants calculated by Equation (1). [b] From reference [2g]. [c] From reference [2a]. [d] From reference [8]. [e] From reference [2b].
7
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
Full Paper In all three series, the nitrosubstituted carbanions (1-NO2, 2-NO2, 8-NO2) are by far the least nucleophilic compounds, whereas the phenyl-substituted carbanions are the most reactive ones; the benzhydryl anion 2-Ph has such a high nucleophilicity that it could not be measured with the methods employed in this work. Figure 8, furthermore, shows that all benzoylmethyl anions 1-X are less reactive than analogously substituted ethoxycarbonylmethyl anions 8-X, which are, in turn, less reactive than analogously substituted benzyl anions 2-X. It is obvious that the substituent effects are not additive: while the cyanosubstituted benzyl anion 2-CN is significantly more reactive than Figure 7. Correlation of the second-order rate constants log k2 for the reactions of the nucleophiles 1 with the all other acceptor-substituted electrophiles 3 in DMSO with their electrophilicity parameters E. benzyl anions (Table 6), the cyano-substituted carbanions 1CN and 8-CN are less reactive than their ester-substituted counterparts 1-CO2Et and 8-CO2Et, respectively. Even when the highly reactive phenylacetonitrile anion 2-CN is disregarded, the much smaller substituent effects in the 1 and 8 series reflect the saturation effect, that is, the reduced demand for delocalization of negative charge in benzoyl- (1-X) and ethoxycarbonyl-substituted (8-X) carbanions compared to the benzyl anions 2-X. There is a moderate, nonlinear correlation (Figure 9) between the reactivities of the benzyl anions 2-X towards the quinone methide 3 n and the p electron densities (qCH) of the
Figure 8. Structure–reactivity relationships for benzoyl-substituted carbanions 1-X (left), ethoxycarbonyl-substituted carbanions 8-X (middle), and benzyl anions 2-X (right). [a] From reference [3c]; [b] From reference [2b]; [c] From reference [8].
&
&
Chem. Eur. J. 2014, 20, 1 – 11
www.chemeurj.org
Figure 9. Correlation of log k2 for the reactions of the benzyl anions 2-X with the quinone methide 3 n vs. the p-electron densities qCH of their carbanionic center.[9] [a] The qCH value for the methyl ester is used for the ethyl ester 2-CO2Et.
8
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!
Full Paper carbanionic center, which Pagani et al.[9] derived from the 13 C NMR shifts under consideration of the shielding effect. The behavior of the phenylsulfonyl group (2-SO2Ph) is peculiar. Although it concentrates slightly more electron density on the carbanionic center than the cyano-substituted anion, it is considerably less reactive than 2-CN and has a nucleophilicity similar to the acetyl-substituted benzyl anion 2-COMe. Figures S5 and S6 in the Supporting Information show that the secondorder rate constants of the reactions of 1-X and 2-X with electrophiles do not correlate with the 13C NMR shifts of the anionic centers. While there is a fair correlation between the nucleophilic reactivities of the benzoylmethyl anions 1-X and Hammett’s substituent constants sp of the substituents X (Figure 10), the corresponding correlation with sp is of lower quality (see the Supporting Information, Figure S7). In contrast, the nucleophilic reactivities of the analogously substituted benzyl anions 2-X and Figure 10. Plot of the rate constants for the reactions of the nucleophiles 1ethyl acetate anions 8-X correlate neither with sp nor with sp X with the electrophile 3 f vs. Hammett’s sp substituent constants.[10] [a] Calculated with Equation (1). (see the Supporting Information, Figures S8 and S9). Figure 11 shows a fair correlation of most nucleophilicity parameters N with the corresponding pKaH values in DMSO,[4] from which several compounds deviate significantly. As different electrophiles had to be employed for determining the N parameters for the phenacyl series 1-X and the benzyl series 2-X, one cannot construct a single Brønsted plot based on experimental rate constants for the carbanions of both series. However, when this correlation is split up in ordinary Brønsted plots (Figure 12), correlations of moderate quality with Brønsted Figure 11. Correlation of the nucleophilicities N of the anions 1-X and 2-X with the pKaH (DMSO) values.[4] a values of 0.40 and 0.41 are obtained.
Conclusion Stabilized carbanions have played a key-role as reference nucleophiles for the construction of our comprehensive reactivity scales,[3c,g] though the negligible role of ion pairing has only been demonstrated in isolated cases.[2a,d–g, 3c, 11] The systematic investigation of the kinetics of the reactions of the potassium salts of 1, 2, 7, and 8 with benzhydrylium ions and quinone methides in this work showed that at carbanion concentrations 103 mol L1 in DMSO, with Chem. Eur. J. 2014, 20, 1 – 11
Figure 12. Brønsted plots of the second-order rate constants for the reactions of the anions 1-X with 3 f and of the anions 2-X with 3 n versus the pKaH (DMSO) values.[4] [a] Calculated with Equation (1).
www.chemeurj.org
These are not the final page numbers! ÞÞ
9
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
Full Paper only two exceptions, none of the reaction rates depended significantly on the concentration of K + , which justifies these rate constants being assigned to the reactivities of the free carbanions. At higher potassium ion concentration, different sensitivities of the reaction rates on variation of [K + ] were observed. At [K + ] 102 mol L1, for example, the reactivities of the monoketo- and monoester stabilized carbanions 1-Ph and 2-CO2Et decreased only slightly to about 80 % of the value of the free carbanion, whereas the benzoyl ester-substituted carbanion 1CO2Et decreased to 20 % and the benzoyl amido-substituted carbanion 1-CONEt2 to 10 % of the value of the free carbanion. In all other cases, the reactivities at [K + ] 102 mol L1 were approximately 40–60 % of those of the free carbanions. Although Na + and Li + had a larger effect on the reactivities of the carbanions, the trends were analogous to those observed with K + . Thus, in a 0.01 mmol L1 solution of Li + , the reactivity of 1-Ph was 39 % of that of the free carbanion, whereas the reactivity of 1-CO2Et was only 1 % of that of the free carbanion. The reactivities of the free carbanions (log k2) correlate linearly with the electrophilicity parameters E, showing that Equation (1) is applicable. Therefore it is possible to characterize the nucleophilic reactivities of the carbanions by the reactivity parameters N and sN, to integrate these carbanions into our comprehensive nucleophilicity scale[3i] and to predict potential electrophilic reaction partners. The reactivities correlate moderately with Bordwell’s pKaH values and show Brønsted plots with typical a values of 0.40 (carbanions 1-X) and 0.45 (carbanions 2-X). While the substituent effects are not additive, the reactivities of the anions 1 (PhCO-CH-X), but not those of 2 (Ph-CH-X), correlate linearly with Hammett’s sp parameters. As the UV absorption maxima of the carbanions 1-X and 2-X are in the range of 300–400 nm, they can be used as reference nucleophiles for the rapid photometric determination of the reactivities of further synthetically important electrophiles.
[2]
[3]
[4]
[5] [6] [7]
Acknowledgements [8]
We thank the Deutsche Forschungsgemeinschaft (SFB 749, Project B1) for support of this work, Vasily Mezhnev for experimental support, and Dr. A. R. Ofial for help during the preparation of this manuscript.
[9]
Keywords: carbanions · ion pairs · kinetics · linear free-energy relationships · nucleophilicity
[10]
[1] a) R. B. Bates, C. A. Ogle, Carbanion Chemistry, Springer, Berlin, 1983; b) L. Brandsma, Preparative Polar Organometallic Chemistry, Vol. 2, Springer, Berlin, 1990, pp. 1 – 11; c) Advances in Carbanion Chemistry, Vol. 1 (Ed.: V. Snieckus), JAI Press, Greenwich, CT, 1992; d) E. Buncel, J. M. Dust, Carbanion Chemistry, Oxford University Press, Oxford, 2003;
&
&
Chem. Eur. J. 2014, 20, 1 – 11
www.chemeurj.org
[11]
e) M. B. Smith, March’s Advanced Organic Chemistry, 7th ed., Wiley, Hoboken, 2013, pp. 221 – 234. Reactivity of carbanions: a) R. Lucius, H. Mayr, Angew. Chem. 2000, 112, 2086 – 2089; Angew. Chem. Int. Ed. 2000, 39, 1995 – 1997; b) T. Bug, T. Lemek, H. Mayr, J. Org. Chem. 2004, 69, 7565 – 7576; c) T. Binh Phan, H. Mayr, Eur. J. Org. Chem. 2006, 2530 – 2537; d) S. T. A. Berger, A. R. Ofial, H. Mayr, J. Am. Chem. Soc. 2007, 129, 9753 – 9761; e) F. Seeliger, H. Mayr, Org. Biomol. Chem. 2008, 6, 3052 – 3058; f) O. Kaumanns, R. Appel, T. Lemek, F. Seeliger, H. Mayr, J. Org. Chem. 2009, 74, 75 – 81; g) F. CorralBautista, H. Mayr, Eur. J. Org. Chem. 2013, 4255 – 4261. Selected publications: a) H. Mayr, M. Patz, Angew. Chem. 1994, 106, 990 – 1010; Angew. Chem. Int. Ed. Engl. 1994, 33, 938 – 957; b) H. Mayr, T. Bug, M. F. Gotta, N. Hering, B. Irrgang, B. Janker, B. Kempf, R. Loos, A. R. Ofial, G. Remennikov, H. Schimmel, J. Am. Chem. Soc. 2001, 123, 9500 – 9512; c) R. Lucius, R. Loos, H. Mayr, Angew. Chem. 2002, 114, 97 – 102; Angew. Chem. Int. Ed. 2002, 41, 91 – 95; ; d) H. Mayr, B. Kempf, A. R. Ofial, Acc. Chem. Res. 2003, 36, 66 – 77; e) H. Mayr, A. R. Ofial, Pure Appl. Chem. 2005, 77, 1807 – 1821; f) D. Richter, N. Hampel, T. Singer, A. R. Ofial, H. Mayr, Eur. J. Org. Chem. 2009, 3203 – 3211; g) H. Mayr, A. R. Ofial, J. Phys. Org. Chem. 2008, 21, 584 – 595; h) H. Mayr, S. Lakhdar, B. Maji, A. R. Ofial, Beilstein J. Org. Chem. 2012, 8, 1458 – 1478; i) For a comprehensive listing of nucleophilicity parameters N, sN and electrophilicity parameters E, see http://www.cup.lmu.de/oc/mayr/DBintro.html. a) W. S. Matthews, J. E. Bares, J. E. Bartmess, F. G. Bordwell, F. J. Cornforth, G. E. Drucker, Z. Margolin, R. J. McCallum, G. J. McCollum, N. R. Vanier, J. Am. Chem. Soc. 1975, 97, 7006 – 7014; b) F. G. Bordwell, M. Van der Puy, N. R. Vanier, J. Org. Chem. 1976, 41, 1883 – 1885; c) . S. Petrov, E. N. Tsvetkov, S. P. Mesyats, A. N. Shatenshtein, M. I. Kabachnik, Izv. Akad. Nauk SSSR 1976, 782 – 787; d) J. M. Kern, P. Federlin, Tetrahedron 1978, 66, 661 – 670; e) F. G. Bordwell, J. E. Bares, J. E. Bartmess, G. J. McCollum, M. Van der Puy, N. R. Vanier, W. S. Matthews, J. Org. Chem. 1977, 42, 321 – 325; f) W. N. Olmstead, F. G. Bordwell, J. Org. Chem. 1980, 45, 3299 – 3305; g) F. G. Bordwell, J. A. Harrelson Jr., Can. J. Chem. 1990, 68, 1714 – 1718. S. T. A. Berger, F. H. Seeliger, F. Hofbauer, H. Mayr, Org. Biomol. Chem. 2007, 5, 3020 – 3026. M. Raban, E. A. Noe, G. Yamamoto, J. Am. Chem. Soc. 1977, 99, 6527 – 6531. For an overview concerning ion-pairing effects of alkali enolates of b-dicarbonyls: a) H. E. Zaugg, A. D. Schaefer, J. Am. Chem. Soc. 1965, 87, 1857 – 1866; b) M. Raban, D. P. Haritos, J. Am. Chem. Soc. 1979, 101, 5178 – 5182; c) S. M. Esakov, A. A. Petrov, B. A. Ershov, J. Org. Chem. USSR (Engl. Trans.) 1975, 11, 679 – 688; d) A. A. Petrov, S. M. Esakov, B. A. Ershov, J. Org. Chem. USSR (Engl. Trans.) 1976, 12, 774 – 778; e) A. Facchetti, A. Streitwieser, J. Org. Chem. 2004, 69, 8345 – 8355; f) A. Streitwieser, J. Mol. Model. 2006, 12, 673 – 680 and references mentioned there. For determination of the nucleophilic reactivity of 8-NO2 see the Appendix of the Supporting Information. a) S. Bradamante, G. A. Pagani, J. Org. Chem. 1984, 49, 2863 – 2870; b) S. Bradamante, G. A. Pagani, J. Chem. Soc. Perkin Trans. 2 1986, 1035 – 1046; c) E. Barchiesi, S. Bradamante, R. Ferraccioli, G. A. Pagani, J. Chem. Soc. Perkin Trans. 2 1990, 375 – 383; d) E. Barchiesi, S. Bradamante, R. Ferraccioli, G. A. Pagani, J. Chem. Soc. Chem. Commun. 1987, 1548 – 1549. C. Hansch, A. Leo, D. Hoeckman, Exploring QSAR: Hydrophobic, Electronic, and Steric Constants, American Chemical Society, Washington, DC, 1995; and references therein. R. Appel, R. Loos, H. Mayr, J. Am. Chem. Soc. 2009, 131, 704 – 714.
Received: July 21, 2014 Published online on && &&, 0000
10
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!
Full Paper
FULL PAPER & Reaction Kinetics
Kinetics of the reactions of differently substituted secondary carbanions with benzhydrylium ions and Michael acceptors are determined in DMSO solution in the presence of variable concentrations of potassium, sodium, or lithium salts. The second-order rate constants depend on the nature of the metal and on the structure of the carbanions. Nucleophilicity parameters N of the “free” carbanions are derived from the linear free-energy relationship log k2 = sN(N + E).
Chem. Eur. J. 2014, 20, 1 – 11
www.chemeurj.org
These are not the final page numbers! ÞÞ
F. Corral-Bautista, R. Appel, J. S. Frickel, H. Mayr* && – && Quantification of Ion-Pairing Effects on the Nucleophilic Reactivities of Benzoyl- and Phenyl-Substituted Carbanions in Dimethylsulfoxide
11
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
Full Paper
& Reaction Kinetics
Quantification of Ion-Pairing Effects on the Nucleophilic Reactivities of Benzoyl- and Phenyl-Substituted Carbanions in Dimethylsulfoxide Francisco Corral-Bautista, Roland Appel, Johanna S. Frickel, and Herbert Mayr*[a] Dedicated to Professor Mieczysław Ma˛kosza on the occasion of his 80th birthday
Abstract: Second-order rate constants for the reactions of acceptor-substituted phenacyl (PhCOCHAcc) and benzyl anions (PhCHAcc) with diarylcarbenium ions and quinone methides (reference electrophiles) have been determined in dimethylsulfoxide (DMSO) solution at 20 8C. By studying the kinetics in the presence of variable concentrations of potassium, sodium and lithium salts (up to 102 mol L1), the influence of ion-pairing on the reaction rates was examined. As the concentration of K + did not have any influence on the rate constants at carbanion concentrations in the range of 104–103 mol L1, the acquired rate constants could be assigned to the reactivities of the
free carbanions. The counter ion effects increase, however, in the series K + < Na + < Li + , and the sensitivity of the carbanion reactivities toward variation of the counter ion strongly depends on the structure of the carbanions. The reactivity parameters N and sN of the free carbanions were derived from the linear plots of log k2 against the electrophilicity parameters E of the reference electrophiles, according to the linear-free energy relationship log k2(20 8C) = sN(N + E). These reactivity parameters can be used to predict absolute rate constants for the reactions of these carbanions with other electrophiles of known E parameters.
Introduction Reactions of carbanions with alkyl halides, Michael acceptors, or carbonyl groups are among the most important methods for the formation of carbon–carbon bonds.[1] In previous work, we studied the nucleophilic reactivities of different types of carbanions[2] and found that pKaH values are not a reliable measure for their relative reactivities.[2b,d,f,g] Since these investigations were mostly performed in DMSO solution at low ion concentrations, ion pairing was neglected. We now examine the concentration range in which this assumption is justified and characterize the nucleophilic reactivities of the benzoyl-substituted carbanions 1 and the analogously substituted benzyl anions 2 (Scheme 1) by determining the rate constants of their reactions with the electrophiles shown in Table 1. The secondorder rate constants are then analyzed by the linear freeenergy relationship Equation (1), where E is an electrophilespecific parameter, and N and sN are solvent-dependent nucleophile-specific parameters.[3] log k2 ð20 CÞ ¼ sN ðN þ EÞ
Scheme 1. Benzoyl- (1-X) and benzyl-substituted carbanions (2-X) and their absorption maxima lmax (in nm) in DMSO. [a] pKaH in DMSO solution. From reference [4]; [b] the nucleophilic reactivities of the anions 2-CO2Et and 2NO2 were reported in references [2b] and [2g]; [c] lmax from reference [2g]; [d] 2-COPh = 1-Ph.
Relationships between structures and nucleophilic reactivities of the carbanions 1 and 2 are discussed.
ð1Þ
[a] Dr. F. Corral-Bautista, Dr. R. Appel, J. S. Frickel, Prof. Dr. H. Mayr Department Chemie, Ludwig-Maximilians-Universitt Mnchen Butenandtstraße 5-13 (Haus F), 81377 Mnchen (Germany) Fax: (+ 49) 89-2180-77717 E-mail: [email protected]
Results and Discussion Preparation of the alkali salts The alkali salts (1-X)M were generated by treatment of the corresponding CH acids with alkali tert-butoxide (LiOtBu,
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/chem.201404500. Chem. Eur. J. 2014, 20, 1 – 11
These are not the final page numbers! ÞÞ
1
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
Full Paper Table 1. Benzhydrylium ions 3 a–f and Michael acceptors 3 g–p employed as reference electrophiles in this study. Electrophile
E[a]
lmax[b] [nm]
R = NMe2 R = N(CH2)4
3a 3b
7.02 7.69
613 620
n=2 n=1
3c 3d
8.22 8.76
618 627
Scheme 3. Possible configurations of b-dicarbonyl enolates.
n=2 n=1
R1 = Ph R1 = Ph R1 = tBu R1 = tBu R1 = tBu R1 = OMe R1 = Me R1 = tBu
R2 = OMe R2 = NMe2 R2 = NO2 R2 = Me R2 = OMe R2 = OMe R2 = NMe2 R2 = NMe2
3e 3f
9.45 635 10.04 630
3g 3h 3i 3j 3k 3l 3m 3n
12.18 13.39 14.36 15.83 16.11 16.38 16.36 17.29
3o
17.90 521
3p
14.68 520
Previous NMR investigations of the sodium salt of pentane2,4-dione (R = R’ = CH3 in Scheme 3) in methanol showed the presence of a mixture of the S- and U-configurations.[6] The absence of the W-configuration can be explained by the steric interaction between the coplanar R and R’ groups. The U/S ratio increased when sodium iodide was added, due to the formation of the sodium complex of the bidentate U-configuration. Addition of more than one equivalent of 18-crown-6 led to the disappearance of the U-configuration, showing that the free enolate ion exists in the S-configuration exclusively; obviously, in the free carbanion the U-configuration is destabilized by the repulsive Coulomb interaction between the negatively charged oxygen atoms. Based on these investigations,[6, 7] we assign the two sets of signals for 1-COMe to the E,Z and Z,E configurations, the two nonidentical S-configurations of unsymmetrical b-dicarbonyl compounds. The increasing NMR shielding of the carbanionic carbon and the attached proton from Li + to Na + and K + reflects the increasing electron density with increasing freedom of the carbanion 1-CO2Et (Scheme 2). Due to their low stability, the potassium salts of the anion of deoxybenzoin (1-Ph) and of the benzyl anions 2-CN, 2-COMe, and 2-SO2Ph, were not isolated, but were generated by deprotonation of the corresponding CH acids with KOtBu or dimsyl potassium (typically 1.05 equiv) in DMSO prior to combining them with the electrophiles. As the UV/Vis absorbances of the carbanions 1-Ph, 2-CN, 2-COMe and 2-SO2Ph reached a limiting value when the corresponding CH acids were treated with 1.05 equivalents of base and did not increase further when additional 2–5 equivalents of base were added, one can conclude that 1.05 equivalents of base are sufficient for their complete deprotonation.
422 533 374 371 393 407 490 486
[a] Electrophilicity-parameters E were taken from references [3b,c,f, 5]; [b] in DMSO solution.
Product studies The benzhydrylium salt 3 eBF4 was added as a solid to equimolar solutions of the potassium salts (1-X)K in DMSO. After decoloration of the blue solutions (typically within one to two minutes), water was added to precipitate the products 4-X. After filtration, the crude products were recrystallized from ethanol or ethanol/CH2Cl2 mixtures and characterized by NMR spectroscopy and mass spectrometry (Scheme 4). Combination of solutions of the potassium salts (1-X)K with solutions of the quinone methides 3 j,k,n in DMSO (with 5–10 % of dichloromethane as cosolvent) and workup with aqueous acetic acid yielded the crude reaction products 5-X which were purified by chromatography and characterized by
Scheme 2. 1H and 13C NMR chemical shifts of the anionic centres of the alkali salts 1-X (0.16 < c < 0.24 mmol L1) in [D6]DMSO. [a] Generated by treatment of the conjugate CH acid with KOtBu in DMSO solution; [b] two isomers in solution (ratio = 1:1.3).
NaOtBu, KOtBu) in ethanol and isolated after washing the precipitated salts with dry diethyl ether (Scheme 2). The anions 1-X gave rise to a single set of NMR signals in [D6]DMSO solution, except for (1-COMe)K, which shows two sets of broad signals in the ratio 1:1.3. For this anion, four possible configurations have to be considered (Scheme 3). &
&
Chem. Eur. J. 2014, 20, 1 – 11
www.chemeurj.org
2
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!
Full Paper
Scheme 4. Reaction of the potassium salts (1-X)K with the benzhydrylium salt 3 eBF4 in DMSO.
Scheme 6. Reactions of the carbanions (2-X)K with the quinone methide 3 n in DMSO at ambient temperature. [a] Determined by 1H NMR spectroscopy after purification by chromatography.
Scheme 5. Reactions of the potassium salts (1-X)K with the quinone methides 3 in DMSO. [a] Determined by 1H NMR spectroscopy after purification by chromatography; [b] could not be determined. Figure 1. Decrease of the absorbance A (at 630 nm) during the reaction of (1-CN)K (1.35 104 mol L1) with 3 f (1.71 105 mol L1) in DMSO at 20 8C. Inset: Correlation of the first-order rate constants kobs with the concentrations of (1-CN)K. Filled circles in absence, empty circles in presence of 2.4 equivalents of 18-crown-6.
NMR spectroscopy and mass spectrometry. The most reactive carbanion 1-Ph was not employed as a stable salt but generated by deprotonation of deoxybenzoin with KOtBu (ca. 1.05 equiv) in DMSO prior to adding the solution of the quinone methide 3 n. In all cases, the products 5-X were obtained in good yields with low diastereoselectivities (Scheme 5). Solutions of the benzyl anions (2-X)K in DMSO were obtained by treatment of the corresponding CH acids with 1.05 equivalents of dimsylK or KOtBu. Addition of the quinone methide 3 n, followed by workup with aqueous acetic acid, gave the products 6-X in good yields as mixtures of two diastereoisomers (Scheme 6).
order rate constants (see below, Tables 4 and 5) were obtained as the slopes of linear correlations of kobs with the concentrations of the nucleophiles (Figure 1). Counter ion effects All carbanions 1-X were obtained as alkali salts by treatment of ketones, b-diketones, b-ketoesters, and related compounds with alkali tert-butoxide. To examine how ion pairing affects the kinetics of their reactions with electrophiles, we determined the first-order rate constants at constant concentrations of the carbanion salts (ca. 104 mol L1) and the electrophiles (ca. 105 mol L1) in the presence of variable concentrations (0– 102 mol L1) of alkali salts (LiCl, LiBF4, NaBPh4, KBPh4, KOTf; representative examples are shown in Table 2; detailed experimental data for further nucleophiles are in the Supporting Information). The measured rate constants were related to the second-order rate constants of the free carbanions (krel = 1), which were taken from Tables 4 and 5 (see below) or from the quoted references. For comparison, we also studied the carbanions derived from the ester 2-CO2Et, the b-diketones 7, the diester 8-CO2Et and the b-ketoester 8-COMe (structures are
Kinetic investigations The kinetics of the reactions of the carbanions 1-X and 2-X with the reference electrophiles 3 were studied in DMSO solution at 20 8C by monitoring the absorptions of the electrophiles with conventional or stopped-flow UV/Vis spectrometers. To simplify the evaluation of the kinetic experiments, the carbanions were used in large excess (> 10 equiv). Thus, their concentrations remained almost constant throughout the reactions, and pseudo-first-order kinetics were obtained in all runs. The first-order rate constants kobs were derived by least-squares fitting of the exponential function At = A0·exp(-kobs t) + C to the time-dependent absorbances At of the electrophiles. SecondChem. Eur. J. 2014, 20, 1 – 11
www.chemeurj.org
These are not the final page numbers! ÞÞ
3
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
Full Paper Table 2. First-order rate constants kobs for the reactions of various carbanions with reference electrophiles in the presence of variable concentrations of alkali salts in DMSO at 20 8C. 100 krel was calculated from the second-order rate constants of the free anions given in Table 4. Reaction
c = 2.12 104 mol L1
c = 3.69 104 mol L1
c = 4.39 104 mol L1
c = 9.53 104 mol L1
c = 6.58 104 mol L1
M
Additive
[Alkali + ]total [mol L1]
K K K K K K K K K
KBPh4 KBPh4 KBPh4 KBPh4 KBPh4 KBPh4 KBPh4 KBPh4
3.51 104 4.80 104 7.36 104 1.48 103 2.79 103 5.27 103 7.79 103 1.16 102
0.761[a] 0.748 0.737 0.729 0.701 0.658 0.631 0.604 0.599
100 98 97 96 92 86 83 79 79 100 71 72 69 62 57 54 49 49
kobs [s1]
NaBPh4 NaBPh4 NaBPh4 NaBPh4 NaBPh4 NaBPh4 NaBPh4 NaBPh4
5.69 104 1.11 103 1.71 103 3.09 103 4.80 103 7.74 103 1.21 102 1.82 102
1.32[a] 0.950 0.935 0.909 0.823 0.757 0.707 0.652 0.649
K Li Li Li Li Li Li Li Li Li Li
none LiBF4 LiBF4 LiBF4 LiBF4 LiBF4 LiBF4 LiBF4 LiBF4 LiBF4
4.65 104 1.37 103 2.27 103 4.08 103 7.69 103 1.13 102 1.74 102 2.66 102 4.30 102 5.93 102
1.58[a] 1.18 1.10 0.976 0.871 0.688 0.616 0.480 0.333 0.178 0.145
100 75 69 62 55 44 39 30 21 11 9
K Li Li Li Li Li Li Li Li Li Li
none LiCl LiCl LiCl LiCl LiCl LiCl LiCl LiCl LiCl
1.01 103 1.51 103 2.01 103 3.21 103 6.05 103 1.11 102 2.12 102 3.46 102 5.14 102 7.65 102
3.42[a] 2.47 2.52 2.47 2.18 1.87 1.43 1.04 0.820 0.706 0.508
100 72 74 72 64 55 42 30 24 21 15
K K K K K K K K K K K K
none KBPh4 KBPh4 KBPh4 KBPh4 KBPh4 KBPh4 KBPh4 KBPh4 KBPh4 KBPh4
6.58 8.68 1.08 1.64 2.41 3.46 5.00 7.89 1.08 1.37 2.08
282[a] 280 260 241 210 175 136 111 82.1 65.2 55.0 41.9
&
Chem. Eur. J. 2014, 20, 1 – 11
www.chemeurj.org
M
Additive
[Alkali + ]total [mol L1]
kobs [s1]
100 krel
c = 6.56 104 mol L1
K K K K K K K K K K K
none KOTf KOTf KOTf KOTf KOTf KOTf KOTf KOTf KOTf
6.56 104 8.63 104 1.07 103 1.69 103 3.15 103 4.40 103 6.32 103 1.01 102 1.39 102 1.77 102
281[a] 273 236 219 178 131 107 82.9 61.0 50.0 43.1
100 97.2 84.0 77.9 63.3 46.6 38.1 29.5 21.7 17.8 15.3
K Na Na Na Na Na Na Na Na Na
NaBPh4 NaBPh4 NaBPh4 NaBPh4 NaBPh4 NaBPh4 NaBPh4 NaBPh4 NaBPh4
6.81 104 8.10 104 1.20 103 1.85 103 3.15 103 7.72 103 1.49 102 2.21 102 3.28 102
229[a] 114 95.0 64.3 41.9 26.9 23.4 15.7 11.8 11.0
100 49.8 41.5 28.1 18.3 11.7 10.2 6.9 5.2 4.8
K Li Li Li Li Li Li Li Li
LiCl LiCl LiCl LiCl LiCl LiCl LiCl LiCl
6.73 104 8.48 104 1.02 103 1.38 103 2.08 103 3.48 103 4.91 103 7.46 103
580[a] 58.1 33.5 28.7 24.7 18.0 11.7 9.56 7.04
100 10 5.8 4.9 4.3 3.1 2.0 1.6 1.2
c = 5.34 104 mol L1
c = 6.48 104 mol L1
[a] Calculated as kobs = k2 [Nu] from the second-order rate constants for the corresponding reactions of the free anions (potassium salts or potassium salts in presence of 18-crown-6) given in Table 4.
100 99.3 92.2 85.5 74.5 62.1 48.2 39.4 29.1 23.1 19.5 14.9
Figure 2. Plots of the relative first-order rate constants for the reactions of the anions derived from b-diketones and a monoketone with electrophiles versus the total concentration of potassium (varied by addition of KBPh4) in DMSO solution at 20 8C.
shown in Figures 2 and 3). The observation that the addition of KOTf to solutions of 1-CO2Et and 1-COMe affects the kinetics in the same way as the addition of KBPh4 (Table 2 and Sup&
Reaction
100 krel
K Na Na Na Na Na Na Na Na
104 104 103 103 103 103 103 103 102 102 102
Table 2. (Continued)
4
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!
Full Paper porting Information) indicates that the observed changes of the reaction rates are due to carbanion-potassium interactions and not caused by interactions of the [BPh4] anion with the electrophiles. In Figure 2–4, we have plotted the dependence of the relative rate constants kobs for the reactions of various carbanions with reference electrophiles on the concentration of the potassium ions ([K + ]), which was varied by addition of different amounts of KBPh4. Figure 2 shows that the reactivities of the anion of the monoketone 1-Ph as well as that of the cyclic diketone 7 b (fixed W configuration) were affected relatively little when up to 12 mmol L1 K + was added (80 and 73 % of the reactivity of the free carbanion). In contrast, the reactivities of the b-diketone-derived carbanions 1-COPh and 7 a decreased noticeably with increasing potassium concentration, and reached a plateau at about 40 % of the original value, probably due to a change from E,Z (S-shape) to Z,Z (U-shape) configurations which can interact more strongly with the potassium ions. The decrease of reactivity due to the addition of KBPh4 was smaller for the b-diketone-derived anion 1-COMe. Figure 3 compares the reaction rates of monoester-, b-ketoester-, diester-, and b-ketoamido derived carbanions at variable concentrations of potassium ions. As for the anion of the
Figure 4. Plots of the relative first-order rate constants for the reactions of the anions of the ketones 1-Ph, 1-CN, 1-SO2Ph, and 1-NO2 with electrophiles versus the total concentration of potassium (varied by addition of KBPh4) in DMSO solution at 20 8C.
tassium reduces the nucleophilic reactivities to a larger extent (lower level of the plateaus). The rate constants for the monocarbonyl-substituted carbanions, which are plotted in Figure 4, depend less on the potassium ion concentration than those of the dicarbonyl-substituted carbanions (Figures 2 and 3). While the reactivities of all diketones and ketoesters decrease to 60–20 % of their original values (except the cyclic diketone 7 b) at a potassium concentration of [K + ] = 5 mmol L1, the reactivity of the anion of benzoyl acetonitrile (1-CN) is the only one in Figure 4 that is affected significantly and is reduced to 60 % at a 5 mmol L1 concentration of potassium. Figure 5 compares the influence of different alkali ions on the relative reactivities of 1-Ph and 1-CO2Et. The reactivity of the anion of deoxybenzoin 1-Ph is only slightly affected when the potassium concentration is increased (Figure 5 a). When sodium is added, the reactivity decreases to 50 % of the value of the free carbanion. Addition of lithium chloride or lithium tetrafluoroborate decreases the reactivities to 12 % and 10 %, respectively, at [Li + ] 50 mmol L1 showing a subordinate effect of the anion of the lithium salt. Further increase of the Li + concentration leads to a further decrease of reactivity without reaching a plateau at a concentration of 60–70 mmol L1. For the b-ketoester anion 1-CO2Et, similar plots were observed, albeit with more pronounced effects (Figure 5 b). Addition of KBPh4 decreases the reactivity to 20 % of the value of the free carbanion, while plateaus are reached at about 5 % of the initial reactivity of the free carbanion with 20 mmol L1 NaBPh4 and at 1–2 % with 5 mmol L1 LiCl. Figures 2–4 illustrate that at concentrations of 1 mmol L1 in DMSO, counter-ion effects by K + are negligible, which is confirmed by the observation that the measured pseudo-firstorder rate constants for (1,2)K in the presence and absence of 18-crown-6 fit the same kobs versus [(1,2)K] plots (example shown in Figure 1). Only in the reactions of the anions of the
Figure 3. Plots of the relative first-order rate constants for the reactions of the anions of the ester 2-CO2Et, the b-ketoamide 1-CONEt2, the b-ketoesters 1-CO2Et and 8-COMe, and the diester 8-CO2Et with electrophiles versus the total concentration of potassium (varied by addition of KBPh4) in DMSO solution at 20 8C.
monoketone 1-Ph (Figure 2), the reactivity of the monoesterderived carbanion 2-CO2Et depends only slightly on the concentration of K + . However, in contrast to the behavior of the b-diketones (Figure 2), all anions from b-ketoesters and related compounds (Figure 3) show a significant decrease of reactivity with increasing potassium ion concentration, not only indicating that these carbanions interact with the counter ion K + at lower concentrations (steeper decrease of kobs shows higher equilibrium constants) but also that the coordination with poChem. Eur. J. 2014, 20, 1 – 11
www.chemeurj.org
These are not the final page numbers! ÞÞ
5
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
Full Paper
Figure 6. First-order rate constants for the reaction of (1-CO2Et)Li with 1 e (1.35 105 mol L1, DMSO, 20 8C) and their correlation with the nucleophile concentration.
1 e do not increase linearly with the concentrations of the carbanion, probably because (1-CO2Et)Li forms dimers or higher aggregates in solution, as previously reported for other lithium enolates.[5, 6b,e,f] As the equilibrium between monomer and aggregates shifts towards aggregation by increasing concentration, a nonlinear increase of kobs with increasing [(1CO2Et)Li] was observed. The comparison of the second-order rate constants for the different alkali derivatives of 1-Ph, 1-CO2Et, and 1-CONEt2 (Table 3) shows that the reactivity of the potassium salt of deoxybenzoin (1-Ph)K corresponds to that of the free anion, whereas 1-CO2Et and 1-CONEt2 interact with K + . While the sodium and potassium derivatives of 1-Ph are similarly reactive (slight deviation: reaction with 3 p), the corresponding lithium compound reacts 0.7–0.9 times as fast as the potassium salt (corresponds to free carbanion). In contrast, the second-order rate constants of the sodium derivative of 1-CO2Et are only 30–40 % of those of the free carbanion, almost independent of the nature of the electrophile.
Figure 5. Plots of the relative first-order rate constants (100 krel) for the reactions of a) 1-Ph with 3 n and b) 1-CO2Et with 3 e (Li + ) or 3 f (Na + and K + ) versus the total concentration of alkali metal ions in DMSO solution at 20 8C.
b-ketoamide 1-CONEt2 and of the b-ketoester 1-CO2Et, which showed the strongest counter-ion effects in Figure 3, the addition of 18-crown-6 (18-c-6) increased the second-order rate Table 3. Second-order rate constants k2 for the reaction of different alkali derivatives of 1-Ph, 1-CONEt2 and 1constants by factors of 1.4 and CO2Et with the reference electrophiles 3 in DMSO at 20 8C. 1.2, respectively (Table 3). Analogously, second-orderk2(K + /18-c-6)/k2(K + ) Anion Electrophile k2[a] [L mol1 s1] + + + + Na K K /18-c-6 Li rate constants for the reactions [b] of the sodium salts (1-Ph)Na 3h 9.12 105 1.10 106 1.05 106 4 4 4 [b] 3 p 7.83 10 7.79 10 9.68 10 and (1-CO2Et)Na with electro3 3 3 [b] 3 m 8.89 10 9.59 10 9.88 10 philes could be derived because [b] 3n 2.74 103 3.39 103 3.59 103 the corresponding pseudo-first3 3 3 [b] 3o 1.66 10 1.90 10 1.83 10 order rate constants correlated 3g 3.26 104 4.59 104 1.41 linearly with the concentrations 3 3 4.61 10 1.34 3 h 3.43 10 of the carbanions at concentra2.46 103 1.46 3i 1.69 103 3 1 tions 10 mol L . A linear re3l 50.3 69.7 1.38 lationship between the first3e 4.27 105 8.14 105 8.95 105 1.10 order rate constants and the [c] 5 5 5 1.84 10 3.63 10 4.28 10 1.18 3 f — concentrations of the carbanion 7.18 103 8.67 103 1.21 3g 2.63 103 was also found for the reactions 7.18 102 8.92 102 1.24 3h 2.45 102 of (1-Ph)Li thus giving rise to 3j 5.64 17.2 19.9 1.16 3k 3.67 8.84 11.5 1.30 the second-order rate constants 3 1 listed in Table 3. Figure 6, in con[a] Derived from first-order rate constants determined at [1-X] < 2.3 10 mol L . [b] Rate constants in presence and in absence of 18-crown-6 (18-c-6) are identical. [c] k2 could not be determined because the dependtrast, reveals that the pseudoence of the pseudo-first-order rate constants on the concentration of the carbanion followed a nonlinear relafirst-order rate constants for the tion (Figure 6). reactions of (1-CO2Et)Li with &
&
Chem. Eur. J. 2014, 20, 1 – 11
www.chemeurj.org
6
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!
Full Paper Table 4. Second-order rate constants k2 for the reactions of the carbanions 1-X with the reference electrophiles 3 in DMSO at 20 8C. 1-X
k2 [L mol1 s1]
N (sN)[a]
Electrophile
23.15 (0.60)
3h 3p 3m 3n 3o
1.05 106 9.68 104 9.88 103 3.59 103 1.83 103
3g 3h 3i 3l
4.59 104 4.61 103 2.46 103 6.97 101
17.52 (0.74)
3e 3f 3g 3h 3j 3k
8.95 105 4.28 105 8.67 103 8.92 102 1.99 101 1.15 101
16.55 (0.78)
3e 3f 3g 3h 3j 3k
3.21 105 1.28 105 3.65 103 2.60 102 3.81 2.02
16.03 (0.86)
3e 3f 3g 3h
4.79 105 1.22 105 1.77 103 1.98 102
17.46 (0.65)
3c 3d 3e 3f
1.06 106 4.75 105 1.95 105 6.50 104
17.19 (0.56)
3a 3b 3d 3e 3f
5
5.08 10 2.27 105 4.71 104 2.14 104 1.10 104
13.91 (0.76)
3a 3b 3d 3e 3f
1.49 105 6.83 104 6.10 103 2.33 103 9.49 102
19.28 (0.65)
Table 5. Second-order rate constants k2 for the reactions of the carbanions 2-X with the reference electrophiles 3 in DMSO at 20 8C. 2-X
3n 3o
(3.89 106)[c] (1.96 106)[c]
24.99 (0.60)
3m 3n 3o
1.44 105 4.48 104 1.69 104
25.77 (0.56)
3m 3n 3o
1.95 105 5.84 104 2.66 104
Structure–reactivity relationships Figure 8 compares the influence of the substituents X on the nucleophilicities of benzoyl- (1-X), ethoxycarbonyl- (8-X) and phenyl-substituted (2-X) carbanions. As the relative reactivities of these carbanions depend on the nucleophile-specific sensitivity sN, Table 6 also reports relative reactivities within these reaction series towards a common electrophile. The following discussion shows that the qualitative conclusions drawn from Figure 8 (N-values) and Table 6 (relative rate constants) are identical.
Table 6. Relative reaction rates of benzoylmethyl anions 1-X, ethyl acetate anions 8-X and benzyl anions 2-X towards electrophiles.
X
Ph CONEt2 CO2Et CN COMe COPh SO2Ph NO2
Correlation analysis Linear correlations were obtained for log k2 of the reactions of the carbanions 1 and 2 with the reference electrophiles 3 with their electrophilicity parameters E, as depicted for some repre-
These are not the final page numbers! ÞÞ
k2 [L mol1 s1]
sentative examples in Figure 7. As documented in the Supporting Information, all reactions studied in this work followed analogous linear correlations, indicating that Equation (1) is applicable to these reactions. From the slopes of these correlations, the nucleophile-specific parameters sN were derived, and the negative intercepts on the abscissa (log k2 = 0) correspond to the nucleophilicity parameters N (Tables 4 and 5).
The second-order rate constants for the reactions of the free carbanions 1-X and 2-X, derived at concentrations of 103 mol L1 or in presence of 18-crown-6, with the reference electrophiles 3 in DMSO at 20 8C are summarized in Tables 4 and 5. In the following part, the reactivities of the free carbanions are discussed.
www.chemeurj.org
Electrophile
[a] Nucleophile-specific reactivity parameters N (and sN) as defined by Equation (1); for determination, see below. [b] Generated by deprotonation with dimsyl potassium. [c] Approximate second-order rate constant obtained from a plot of [2-CN] vs. kobs that shows a large negative intercept. [d] Generated by deprotonation with KOtBu.
[a] Nucleophile-specific reactivity parameters N (and sN) as defined by Equation (1); for determination, see below.
Chem. Eur. J. 2014, 20, 1 – 11
N (sN)[a]
+3 f
+3k
7.74 104[a] 1.04 103[a] 4.51 102 1.35 102 1.29 102 6.85 101 1.16 101 1.0
1.44 107[a,b] 2.28 103[c] 9.38 102[c] 3.52 102[c] 5.17 101 1.0[a,d]
+3n
9.95 104[b] 7.02 105 8.09 103 6.48 102 1.05 104 1.0[e]
[a] Rate constants calculated by Equation (1). [b] From reference [2g]. [c] From reference [2a]. [d] From reference [8]. [e] From reference [2b].
7
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
Full Paper In all three series, the nitrosubstituted carbanions (1-NO2, 2-NO2, 8-NO2) are by far the least nucleophilic compounds, whereas the phenyl-substituted carbanions are the most reactive ones; the benzhydryl anion 2-Ph has such a high nucleophilicity that it could not be measured with the methods employed in this work. Figure 8, furthermore, shows that all benzoylmethyl anions 1-X are less reactive than analogously substituted ethoxycarbonylmethyl anions 8-X, which are, in turn, less reactive than analogously substituted benzyl anions 2-X. It is obvious that the substituent effects are not additive: while the cyanosubstituted benzyl anion 2-CN is significantly more reactive than Figure 7. Correlation of the second-order rate constants log k2 for the reactions of the nucleophiles 1 with the all other acceptor-substituted electrophiles 3 in DMSO with their electrophilicity parameters E. benzyl anions (Table 6), the cyano-substituted carbanions 1CN and 8-CN are less reactive than their ester-substituted counterparts 1-CO2Et and 8-CO2Et, respectively. Even when the highly reactive phenylacetonitrile anion 2-CN is disregarded, the much smaller substituent effects in the 1 and 8 series reflect the saturation effect, that is, the reduced demand for delocalization of negative charge in benzoyl- (1-X) and ethoxycarbonyl-substituted (8-X) carbanions compared to the benzyl anions 2-X. There is a moderate, nonlinear correlation (Figure 9) between the reactivities of the benzyl anions 2-X towards the quinone methide 3 n and the p electron densities (qCH) of the
Figure 8. Structure–reactivity relationships for benzoyl-substituted carbanions 1-X (left), ethoxycarbonyl-substituted carbanions 8-X (middle), and benzyl anions 2-X (right). [a] From reference [3c]; [b] From reference [2b]; [c] From reference [8].
&
&
Chem. Eur. J. 2014, 20, 1 – 11
www.chemeurj.org
Figure 9. Correlation of log k2 for the reactions of the benzyl anions 2-X with the quinone methide 3 n vs. the p-electron densities qCH of their carbanionic center.[9] [a] The qCH value for the methyl ester is used for the ethyl ester 2-CO2Et.
8
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!
Full Paper carbanionic center, which Pagani et al.[9] derived from the 13 C NMR shifts under consideration of the shielding effect. The behavior of the phenylsulfonyl group (2-SO2Ph) is peculiar. Although it concentrates slightly more electron density on the carbanionic center than the cyano-substituted anion, it is considerably less reactive than 2-CN and has a nucleophilicity similar to the acetyl-substituted benzyl anion 2-COMe. Figures S5 and S6 in the Supporting Information show that the secondorder rate constants of the reactions of 1-X and 2-X with electrophiles do not correlate with the 13C NMR shifts of the anionic centers. While there is a fair correlation between the nucleophilic reactivities of the benzoylmethyl anions 1-X and Hammett’s substituent constants sp of the substituents X (Figure 10), the corresponding correlation with sp is of lower quality (see the Supporting Information, Figure S7). In contrast, the nucleophilic reactivities of the analogously substituted benzyl anions 2-X and Figure 10. Plot of the rate constants for the reactions of the nucleophiles 1ethyl acetate anions 8-X correlate neither with sp nor with sp X with the electrophile 3 f vs. Hammett’s sp substituent constants.[10] [a] Calculated with Equation (1). (see the Supporting Information, Figures S8 and S9). Figure 11 shows a fair correlation of most nucleophilicity parameters N with the corresponding pKaH values in DMSO,[4] from which several compounds deviate significantly. As different electrophiles had to be employed for determining the N parameters for the phenacyl series 1-X and the benzyl series 2-X, one cannot construct a single Brønsted plot based on experimental rate constants for the carbanions of both series. However, when this correlation is split up in ordinary Brønsted plots (Figure 12), correlations of moderate quality with Brønsted Figure 11. Correlation of the nucleophilicities N of the anions 1-X and 2-X with the pKaH (DMSO) values.[4] a values of 0.40 and 0.41 are obtained.
Conclusion Stabilized carbanions have played a key-role as reference nucleophiles for the construction of our comprehensive reactivity scales,[3c,g] though the negligible role of ion pairing has only been demonstrated in isolated cases.[2a,d–g, 3c, 11] The systematic investigation of the kinetics of the reactions of the potassium salts of 1, 2, 7, and 8 with benzhydrylium ions and quinone methides in this work showed that at carbanion concentrations 103 mol L1 in DMSO, with Chem. Eur. J. 2014, 20, 1 – 11
Figure 12. Brønsted plots of the second-order rate constants for the reactions of the anions 1-X with 3 f and of the anions 2-X with 3 n versus the pKaH (DMSO) values.[4] [a] Calculated with Equation (1).
www.chemeurj.org
These are not the final page numbers! ÞÞ
9
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
Full Paper only two exceptions, none of the reaction rates depended significantly on the concentration of K + , which justifies these rate constants being assigned to the reactivities of the free carbanions. At higher potassium ion concentration, different sensitivities of the reaction rates on variation of [K + ] were observed. At [K + ] 102 mol L1, for example, the reactivities of the monoketo- and monoester stabilized carbanions 1-Ph and 2-CO2Et decreased only slightly to about 80 % of the value of the free carbanion, whereas the benzoyl ester-substituted carbanion 1CO2Et decreased to 20 % and the benzoyl amido-substituted carbanion 1-CONEt2 to 10 % of the value of the free carbanion. In all other cases, the reactivities at [K + ] 102 mol L1 were approximately 40–60 % of those of the free carbanions. Although Na + and Li + had a larger effect on the reactivities of the carbanions, the trends were analogous to those observed with K + . Thus, in a 0.01 mmol L1 solution of Li + , the reactivity of 1-Ph was 39 % of that of the free carbanion, whereas the reactivity of 1-CO2Et was only 1 % of that of the free carbanion. The reactivities of the free carbanions (log k2) correlate linearly with the electrophilicity parameters E, showing that Equation (1) is applicable. Therefore it is possible to characterize the nucleophilic reactivities of the carbanions by the reactivity parameters N and sN, to integrate these carbanions into our comprehensive nucleophilicity scale[3i] and to predict potential electrophilic reaction partners. The reactivities correlate moderately with Bordwell’s pKaH values and show Brønsted plots with typical a values of 0.40 (carbanions 1-X) and 0.45 (carbanions 2-X). While the substituent effects are not additive, the reactivities of the anions 1 (PhCO-CH-X), but not those of 2 (Ph-CH-X), correlate linearly with Hammett’s sp parameters. As the UV absorption maxima of the carbanions 1-X and 2-X are in the range of 300–400 nm, they can be used as reference nucleophiles for the rapid photometric determination of the reactivities of further synthetically important electrophiles.
[2]
[3]
[4]
[5] [6] [7]
Acknowledgements [8]
We thank the Deutsche Forschungsgemeinschaft (SFB 749, Project B1) for support of this work, Vasily Mezhnev for experimental support, and Dr. A. R. Ofial for help during the preparation of this manuscript.
[9]
Keywords: carbanions · ion pairs · kinetics · linear free-energy relationships · nucleophilicity
[10]
[1] a) R. B. Bates, C. A. Ogle, Carbanion Chemistry, Springer, Berlin, 1983; b) L. Brandsma, Preparative Polar Organometallic Chemistry, Vol. 2, Springer, Berlin, 1990, pp. 1 – 11; c) Advances in Carbanion Chemistry, Vol. 1 (Ed.: V. Snieckus), JAI Press, Greenwich, CT, 1992; d) E. Buncel, J. M. Dust, Carbanion Chemistry, Oxford University Press, Oxford, 2003;
&
&
Chem. Eur. J. 2014, 20, 1 – 11
www.chemeurj.org
[11]
e) M. B. Smith, March’s Advanced Organic Chemistry, 7th ed., Wiley, Hoboken, 2013, pp. 221 – 234. Reactivity of carbanions: a) R. Lucius, H. Mayr, Angew. Chem. 2000, 112, 2086 – 2089; Angew. Chem. Int. Ed. 2000, 39, 1995 – 1997; b) T. Bug, T. Lemek, H. Mayr, J. Org. Chem. 2004, 69, 7565 – 7576; c) T. Binh Phan, H. Mayr, Eur. J. Org. Chem. 2006, 2530 – 2537; d) S. T. A. Berger, A. R. Ofial, H. Mayr, J. Am. Chem. Soc. 2007, 129, 9753 – 9761; e) F. Seeliger, H. Mayr, Org. Biomol. Chem. 2008, 6, 3052 – 3058; f) O. Kaumanns, R. Appel, T. Lemek, F. Seeliger, H. Mayr, J. Org. Chem. 2009, 74, 75 – 81; g) F. CorralBautista, H. Mayr, Eur. J. Org. Chem. 2013, 4255 – 4261. Selected publications: a) H. Mayr, M. Patz, Angew. Chem. 1994, 106, 990 – 1010; Angew. Chem. Int. Ed. Engl. 1994, 33, 938 – 957; b) H. Mayr, T. Bug, M. F. Gotta, N. Hering, B. Irrgang, B. Janker, B. Kempf, R. Loos, A. R. Ofial, G. Remennikov, H. Schimmel, J. Am. Chem. Soc. 2001, 123, 9500 – 9512; c) R. Lucius, R. Loos, H. Mayr, Angew. Chem. 2002, 114, 97 – 102; Angew. Chem. Int. Ed. 2002, 41, 91 – 95; ; d) H. Mayr, B. Kempf, A. R. Ofial, Acc. Chem. Res. 2003, 36, 66 – 77; e) H. Mayr, A. R. Ofial, Pure Appl. Chem. 2005, 77, 1807 – 1821; f) D. Richter, N. Hampel, T. Singer, A. R. Ofial, H. Mayr, Eur. J. Org. Chem. 2009, 3203 – 3211; g) H. Mayr, A. R. Ofial, J. Phys. Org. Chem. 2008, 21, 584 – 595; h) H. Mayr, S. Lakhdar, B. Maji, A. R. Ofial, Beilstein J. Org. Chem. 2012, 8, 1458 – 1478; i) For a comprehensive listing of nucleophilicity parameters N, sN and electrophilicity parameters E, see http://www.cup.lmu.de/oc/mayr/DBintro.html. a) W. S. Matthews, J. E. Bares, J. E. Bartmess, F. G. Bordwell, F. J. Cornforth, G. E. Drucker, Z. Margolin, R. J. McCallum, G. J. McCollum, N. R. Vanier, J. Am. Chem. Soc. 1975, 97, 7006 – 7014; b) F. G. Bordwell, M. Van der Puy, N. R. Vanier, J. Org. Chem. 1976, 41, 1883 – 1885; c) . S. Petrov, E. N. Tsvetkov, S. P. Mesyats, A. N. Shatenshtein, M. I. Kabachnik, Izv. Akad. Nauk SSSR 1976, 782 – 787; d) J. M. Kern, P. Federlin, Tetrahedron 1978, 66, 661 – 670; e) F. G. Bordwell, J. E. Bares, J. E. Bartmess, G. J. McCollum, M. Van der Puy, N. R. Vanier, W. S. Matthews, J. Org. Chem. 1977, 42, 321 – 325; f) W. N. Olmstead, F. G. Bordwell, J. Org. Chem. 1980, 45, 3299 – 3305; g) F. G. Bordwell, J. A. Harrelson Jr., Can. J. Chem. 1990, 68, 1714 – 1718. S. T. A. Berger, F. H. Seeliger, F. Hofbauer, H. Mayr, Org. Biomol. Chem. 2007, 5, 3020 – 3026. M. Raban, E. A. Noe, G. Yamamoto, J. Am. Chem. Soc. 1977, 99, 6527 – 6531. For an overview concerning ion-pairing effects of alkali enolates of b-dicarbonyls: a) H. E. Zaugg, A. D. Schaefer, J. Am. Chem. Soc. 1965, 87, 1857 – 1866; b) M. Raban, D. P. Haritos, J. Am. Chem. Soc. 1979, 101, 5178 – 5182; c) S. M. Esakov, A. A. Petrov, B. A. Ershov, J. Org. Chem. USSR (Engl. Trans.) 1975, 11, 679 – 688; d) A. A. Petrov, S. M. Esakov, B. A. Ershov, J. Org. Chem. USSR (Engl. Trans.) 1976, 12, 774 – 778; e) A. Facchetti, A. Streitwieser, J. Org. Chem. 2004, 69, 8345 – 8355; f) A. Streitwieser, J. Mol. Model. 2006, 12, 673 – 680 and references mentioned there. For determination of the nucleophilic reactivity of 8-NO2 see the Appendix of the Supporting Information. a) S. Bradamante, G. A. Pagani, J. Org. Chem. 1984, 49, 2863 – 2870; b) S. Bradamante, G. A. Pagani, J. Chem. Soc. Perkin Trans. 2 1986, 1035 – 1046; c) E. Barchiesi, S. Bradamante, R. Ferraccioli, G. A. Pagani, J. Chem. Soc. Perkin Trans. 2 1990, 375 – 383; d) E. Barchiesi, S. Bradamante, R. Ferraccioli, G. A. Pagani, J. Chem. Soc. Chem. Commun. 1987, 1548 – 1549. C. Hansch, A. Leo, D. Hoeckman, Exploring QSAR: Hydrophobic, Electronic, and Steric Constants, American Chemical Society, Washington, DC, 1995; and references therein. R. Appel, R. Loos, H. Mayr, J. Am. Chem. Soc. 2009, 131, 704 – 714.
Received: July 21, 2014 Published online on && &&, 0000
10
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!
Full Paper
FULL PAPER & Reaction Kinetics
Kinetics of the reactions of differently substituted secondary carbanions with benzhydrylium ions and Michael acceptors are determined in DMSO solution in the presence of variable concentrations of potassium, sodium, or lithium salts. The second-order rate constants depend on the nature of the metal and on the structure of the carbanions. Nucleophilicity parameters N of the “free” carbanions are derived from the linear free-energy relationship log k2 = sN(N + E).
Chem. Eur. J. 2014, 20, 1 – 11
www.chemeurj.org
These are not the final page numbers! ÞÞ
F. Corral-Bautista, R. Appel, J. S. Frickel, H. Mayr* && – && Quantification of Ion-Pairing Effects on the Nucleophilic Reactivities of Benzoyl- and Phenyl-Substituted Carbanions in Dimethylsulfoxide
11
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
