Skip to main content
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2021 Oct 22;27(64):15928–15935. doi: 10.1002/chem.202103027

Inherent Reactivity of Spiro‐Activated Electrophilic Cyclopropanes

Patrick M Jüstel 1, Alexandra Stan 1, Cedric D Pignot 1, Armin R Ofial 1,
PMCID: PMC9298281  PMID: 34569669

Abstract

The kinetics of the ring‐opening reactions of thiophenolates with geminal bis(acceptor)‐substituted cyclopropanes in DMSO at 20 °C was monitored by photometric methods. The determined second‐order rate constants of the SN2 reactions followed linear relationships with Mayr nucleophilicity parameters (N/s N) and Brønsted basicities (pK aH) of the thiophenolates as well as with Hammett substituent parameters (σ) for groups attached to the thiophenolates. Phenyl‐substituted cyclopropanes reacted by up to a factor of 15 faster than their unsubstituted analogues, in accord with the known activating effect of adjacent π‐systems in SN2 reactions. Variation of the electronic properties of substituents at the phenyl groups of the cyclopropanes gave rise to parabolic Hammett relationships. Thus, the inherent SN2 reactivity of electrophilic cyclopropanes is activated by electron‐rich π‐systems because of the more advanced C1−C2 bond polarization in the transition state. On the other hand, electron‐poor π‐systems also lower the energetic barriers for the attack of anionic nucleophiles owing to attractive electrostatic interactions.

Keywords: cyclopropanes, kinetics, linear free energy relationships, nucleophilic substitution, thiolates


Ring, ring! The reactivity of spiro‐activated electrophilic cyclopropanes was studied by following the kinetics of their SN2‐type ring‐opening reactions with thiophenolate ions in DMSO at 20 °C. The experimentally determined second‐order rate constants (k 2) correlated linearly with Mayr nucleophilicities N, basicities (pK aH), and Hammett substituent constants of the thiophenolates, but parabolic Hammett plots revealed variable electrophile‐dependent effects on the stabilization of the transition states.

graphic file with name CHEM-27-15928-g008.jpg

Introduction

Cyclopropanes are unique building blocks for organic synthesis that react as electrophiles in polar reactions when substituted with electron‐withdrawing groups.[ 1 , 2 ] In particular, spiro‐activated cyclopropanes have been shown to react with nucleophiles under relatively mild conditions because the orthogonal orientation of the cyclopropane moiety and the plane of activating carbonyl groups facilitates delocalization of negative charge in the transition state (TS). [3] The synthetic versatility of electrophilic cyclopropanes (eCPs) was further enhanced by installing electron‐donating groups at C2 (Scheme 1A). In this way, difunctionalized products can be synthesized from so‐called donor−acceptor cyclopropanes (DACPs, Scheme 1A).[ 4 , 5 ] The polarization of the C1−C2 bond in the DACPs not only favorably enhances reactivity toward nucleophiles but also steers regioselectivity of nucleophilic attack toward the already substituted C2 position. [1]

Scheme 1.

Scheme 1

A) General structures for electrophilic and donor−acceptor cyclopropanes. B) Reactions of nucleophiles with electrophilic cyclopropanes generate methylene‐extended Michael adducts.

Nowadays eCPs are regularly employed as precursors for carbo‐ [6] and heterocycles,[ 7 , 8 ] and asymmetric cyclopropanations have made DACPs useful tools for the synthetic chemist.[ 9 , 10 ] Ring opening, [5e] (3+2), (3+3), and (4+3) cycloadditions [11] as well as rearrangements [12] of functionalized cyclopropanes offer access to a multitude of building blocks. [13]

Our interest was triggered, however, by the simple ring‐opening reactions of nucleophiles with electrophilic cyclopropanes because of their relation to Michael additions to electron‐deficient olefins and nucleophilic substitutions at sp3‐hybridized carbon centers (Scheme 1B). It can be anticipated that kinetic studies will provide fundamental insights into the factors that control polar cyclopropane reactivity.

Though the synthetic potential of electrophilic cyclopropanes has widely been explored, kinetic studies to characterize their reactivities are limited to a few solitary reactions of dimedone‐ and Meldrum's acid derived cyclopropanes with pyridines.[ 14 , 15 ] Lately, the Werz group systematically investigated substituent effects on the reactivity of SnCl4‐complexed 1,1‐di(alkoxycarbonyl)‐substituted cyclopropanes in (3+2)‐cycloadditions with p‐fluorobenzaldehyde by in situ NMR kinetics. [16] It is not straightforwardly possible to derive information on the inherent reactivity of electrophilic cyclopropanes from these kinetic studies, however, because cyclopropane reactivity is dominated by Lewis acid activation and the (3+2)‐cycloadditions follow a complex mechanism. [17]

The currently most comprehensive reactivity scales for electrophile‐nucleophile reactions have been constructed by Mayr. Within this framework, the electrophilic property of an electron‐deficient substrate is described by the electrophilicity parameter E in the Mayr−Patz equation 1. [18]

logk2(20C)=sN(N+E) (1)

The nucleophilicity parameter N and the nucleophile‐specific slope s N are specific for the nucleophile in a certain solvent and calibrated towards benzhydrylium ions and structurally related quinone methides (reference electrophiles). In this way, electrophilic and nucleophilic reactivities in the Mayr scales refer to reactions, in which only one new σ‐bond is formed. [18d] With this constraint, it has been demonstrated that Equation (1) usually predicts rate constants with an accuracy better than a factor of 102 for structurally diverse electrophile‐nucleophile combinations. [19] It cannot be expected that SN2 reactions, in which bond formation and bond breaking are coupled events, follow the simple three‐parameter Equation (1). Nevertheless, several examples show that the relative reactivity ordering of nucleophiles, originally calibrated towards one‐bond reference electrophiles, also holds for reactions with electrophilic substrates of SN2 reactions, in particular if the type of atom at the nucleophilic center is kept constant. [20]

Advantageously for the intended kinetic studies at 20 °C, sodium thiophenolate was reported to react with the Meldrum's acid derived spirocyclopropane already at room temperature. [3] In general, thiophenolates are potent nucleophiles that can be fine‐tuned in reactivity by electronic effects but allow maintaining a constant steric demand in the vicinity of the nucleophilic sulfur atom. Because we recently determined the Mayr nucleophilicity parameters N and s N of several UV‐light absorbing ArS ions in DMSO, [21] the scene was set to use thiophenolates as the nucleophiles for investigating the reactivity of electrophilic cyclopropanes by photometric methods (Scheme 2).

Scheme 2.

Scheme 2

Ring‐opening reaction of cyclopropanes 1 with thiophenolates 2 (counterion M+=Na+ or K+).

In this work, we show that the inherent electrophilicity of 1,1‐bis(acceptor)‐substituted cyclopropanes 1 aj can be assessed through kinetic data for reactions of 1 with thiophenolates 2. The experimentally determined second‐order rate constants (k 2) correlated linearly with physicochemical properties of the nucleophiles 2 but parabolic Hammett plots revealed so far undiscovered electrophile‐dependent effects on the stabilization of the transition states of ring‐opening reactions at electrophilic cyclopropanes.

Results and Discussion

Preparation of cyclopropanes 1

The Meldrum's acid derived spirocyclopropane 1 a was purchased. Corey‐Chaykovsky cyclopropanations of active methylene compounds furnished 1 ce, 1 g, and 1 j in yields of 36–84 % according to described procedures. [22] When we applied the analogous synthetic protocol to prepare p‐methyl or p‐methoxy‐substituted derivatives of 1 e, spontaneous Cloke–Wilson rearrangements yielded either solely the corresponding dihydrofuran or mixtures of cyclopropanes and dihydrofurans, which could not be used for subsequent kinetic studies (Scheme 3). [23]

Scheme 3.

Scheme 3

Cloke‐Wilson rearrangement of Meldrum's acid‐derived spirocyclopropanes with electron‐rich aryl groups at C2 (Ar=p‐tolyl or p‐anisyl).

As reported in ref. [24], light induced, NIS‐initiated cyclizations that start from styrenes and cyclic 1,3‐dicarbonyl compounds yielded the spirocyclopropanes 1 b, 1 f, 1 h, and 1 i (in 59–92 % yields).

For details of the synthetic procedures and the spectroscopic characterization of cyclopropanes 1, see the Supporting Information.

Reactions of the cyclopropanes 1 with thiophenolates 2

Products: Reactions of cyclopropanes 1 and sodium thiophenolates 2, generated by deprotonation of thiophenols with sodium hydride, in [D6]DMSO furnished the ionic adducts 3, which were characterized directly by NMR spectroscopy and HRMS. Mixing thiophenols, potassium tert‐butoxide, and the cyclopropanes 1 in DMSO yielded, after aqueous workup, methylene‐extended Michael adducts 4, which were isolated in moderate to excellent yields (Scheme 4). Consistently, product analysis showed that reactions of ArS (2) with cyclopropanes 1 occurred chemoselectively by nucleophilic attack of the thiophenolate at the cyclopropyl ring. Furthermore, all reactions of 2 with aryl‐substituted cyclopropanes 1 b and 1 dj proceeded regioselectively, and carbon‐sulfur bond‐formation was observed only at the already substituted cyclopropane carbon C2, irrespective of whether aryl groups with electron‐withdrawing or electron‐donating substituents were attached to this position.

Scheme 4.

Scheme 4

Reactions of cyclopropanes 1 with sodium or potassium thiophenolates Na/K‐2 furnished ring‐opened products 4 via the initial adducts 3 (yields are given for isolated products). [a] Reactions in [D6]DMSO. [b] Counterion: K+.

Kinetics: The kinetics of the reactions of cyclopropanes 1 ae and 1 j with thiophenolates 2 in DMSO at 20 °C were monitored photometrically by following the fading of the UV absorptions of the thiophenolates at or close to their absorption maximum (stopped‐flow method, λ max=302–329 nm in DMSO), [21] To simplify the kinetic evaluation of the second‐order reactions, we used the cyclopropanes 1 ae or 1 j in at least tenfold excess relative to the initial thiophenolate concentrations [2]0. Under these conditions, the fading out of the thiophenolate absorptions could be fitted by the mono‐exponential decay function A=A 0 exp(−k obs t)+C to determine the (pseudo)‐first‐order rate constants k obs [s−1]. For each cyclopropane/thiophenolate pair k obs was determined at four different concentrations of the (excess) cyclopropane, which allowed us to calculate the second‐order rate constants k 2 [M−1 s−1] from the slope of the linear correlation of k obs with [1]. Figure 1 visualizes this procedure for the ring‐opening reaction of 1 e with the parent thiophenolate PhS (2 c).

Figure 1.

Figure 1

A) Kinetics of the reaction of 1 e with 2 c. B) Monoexponential decay of the absorbance A at 310 nm in the reaction of 2 c (0.247 mM, M+=K+) with 1 e (2.47 mM). C) Linear correlation of the observed rate constants k obs with the concentration of cyclopropane 1 e.

Reactions of the 1,3‐indandione‐derived cyclopropanes 1 fi with the thiophenolates 2 generated stable solutions of enolate ions of structural type 3, which absorb light at 390–396 nm. By obeying the same conditions for (pseudo)first‐order kinetics as before, that is [1]0/[2]0>10, the increasing absorption of the ring‐opened enolate ions could be fitted with the increasing mono‐exponential function A t=A 0[1−exp(−k obs t)]+C to yield the rate constants k obs. In analogy to the evaluation of the kinetics with decreasing absorptions, the experiments were repeated at four different concentrations of 1 fi. The resulting linear correlations of k obs with [1] were of comparable quality as those obtained from absorption decay kinetics and, consequently, the determination of k 2 values for the absorption increase kinetics could be determined analogously, as exemplified in Figure 2.

Figure 2.

Figure 2

A) Kinetics of the reaction of 1 h with 2 d. B) Monoexponential increase of the absorbance A at 396 nm in the reaction of 2 d (0.250 mM, M+=K+) with 1 h (3.75 mM). C) Linear correlation of the observed rate constants k obs with the concentration of cyclopropane 1 h.

Data of the individual kinetic measurements for the nucleophilic attack of thiophenolates 2 at the cyclopropanes 1 are given in the Supporting Information. The determined second‐order rate constants k 2 are compiled in Table 1.

Table 1.

Second‐order rate constants k 2 for the reactions of cyclopropanes 1 with thiophenolates 2 (counterion: K+) in DMSO (20 °C).

k 2 [M−1 s−1]

2 a

2 b

2 c

2 d

2 e

2 f

N=24.97

N=24.35

N=23.36

N=22.80

N=22.50

N=21.30

Electrophile

s N=0.68

s N=0.69

s N=0.74

s N=0.78

s N=0.78

s N=0.86

1 a

30.7[a]

18.0[a]

11.3[a]

n.d.

n.d.

n.d.

1 b

332[a]

242[a]

170[a]

n.d.

n.d.

n.d.

1 c

11.2[a]

8.89[a]

5.52[a]

n.d.

n.d.

n.d.

1 d

84.7

59.2

40.3

25.7

n.d.

n.d.

1 e

30.9

28.2

19.7

16.6

14.3

7.74

1 f

47.7

32.9

20.5

12.3

n.d.

n.d.

1 g

31.5

21.9

15.0

8.38

6.90

3.68

1 h

33.7

26.1

20.3

13.4

n.d.

n.d.

1 i

93.9

65.3

44.9

33.6

n.d.

n.d.

1 j

17.4

11.3

8.54

4.30

n.d.

n.d.

[a] Sodium instead of potassium used as counterion for the thiophenolate 2.

Phenyl effect in reactions of 1 with thiophenolates 2 : A significant phenyl effect was reported for the kinetics of ring‐opening reactions of 1 a,b with pyridine. In acetonitrile at 25 °C, 1 b reacted by a factor of 80 faster with pyridine than 1 a (for 1 a+pyridine: k 2=1.04×10−5 M−1 s−1; for 1 b+pyridine: k 2=8.27×10−4 M−1 s−1). [14]

In this work, the second‐order rate constants k 2 in Table 1 enable a comparison of the reactivities of the parent spirocyclopropanes 1 a and 1 c with those of the 2‐phenyl‐substituted analogues 1 b and 1 e, respectively, which are DACPs (Figure 3). The thiophenolates 2 ac underwent by a factor of 11 to 15 faster reactions with the Meldrum's acid‐derived 2‐phenyl‐cyclopropane 1 b than with the unsubstituted 1 a (Figure 3A), in accord with the known rate enhancing ‘benzyl effect’ on SN2 reactivity. [25] Analogous additions of thiophenolates 2 ac to the dimedone‐derived 1 e were by a factor of 3–4 faster than for the unsubstituted 1 c (Figure 3B).

Figure 3.

Figure 3

Phenyl effect on the reactivities of A) Meldrum's acid‐ and B) dimedone‐derived spriocyclopropanes in reactions with thiophenolates (k 2 in DMSO at 20 °C from Table 1).

In both series of reactions, a consistent trend towards a stronger phenyl effect for the slower reactions was observed, which may explain the phenyl effect of 80 reported by McKinney and colleagues for the reactions of pyridine with 1 a,b, [14] which are more than 5000‐fold slower than the reactions studied in this work.

Comparison of electrophilic cyclopropanes with Michael acceptors: The activating phenyl effect on electrophilic cyclopropanes (Figure 3) is in obvious contrast to the significant deactivation of olefinic Michael acceptors by β‐phenyl groups. For comparison, Figure 4A illustrates that the reactivities of typical Michael acceptors towards neutral and anionic nucleophiles are significantly retarded, as expressed by the 5 to 6 units difference in the logarithmic Mayr E parameters, when β‐hydrogen is replaced by a phenyl group. [26]

Figure 4.

Figure 4

A) Phenyl effect on Michael acceptors expressed by the decrease in electrophilicities E. B) Relative rate constants k rel for reactions of the thiophenolate ion (2 c) with cyclopropanes 1 and structurally related Michael acceptors (20 °C, DMSO, with kinetic data for 1 b, g, and j from Table 1; rate constants for Michael acceptors calculated by applying N and s N for 2 c from Table 1 and E parameters from ref. [18d] in Equation (1)).

Nevertheless, when electrophilic cyclopropanes and Michael acceptors with identical acceptor groups are compared, relative rate constants for the reactions with the parent thiophenolate 2 c show that Michael acceptors are still by 8 to 9 orders of magnitude stronger electrophiles than the related cyclopropanes (Figure 4B). Thus, breaking a carbon–carbon σ‐bond in the reactions of the cyclopropanes causes systematically higher activation barriers than breaking a weaker carbon−carbon π‐bond in structurally related Michael acceptors.

Spiro activation: Proper orientation of the electron‐withdrawing carbonyl groups is assumed to enhance the electrophilicity of the spiro compounds 1 af compared to analogous non‐spirocyclopropanes. Interestingly, cyclopropane 1 j is only slightly less electrophilic than 1 b, 1 e, or 1 g (Figure 5A). This observation demonstrates that spiro‐activation can conceptually be replaced by activation through rod‐like EWGs, such as cyano groups, whose electron‐withdrawing effect does not require optimum orientation of the electron‐poor π‐system to the σ‐bonds of the cyclopropane moiety to become effective.

Figure 5.

Figure 5

A) Relative reactivities of cyclopropanes 1 towards the thiophenolate ion 2 c (in DMSO, 20 °C). B) Meldrum's acid with two unfavorable s‐(E) ester conformations is more acidic than dimedone (pK a in DMSO, from ref. [27a]).

Reactivity of dimedone‐ versus Meldrum's acid‐derived cyclopropanes: Furthermore, the comparison in Figure 5A illustrates that the Meldrum's acid‐derived cyclopropanes 1 a/1 b are more electrophilic towards the thiophenolates 2 than the dimedone‐derived 1 c/1 e. This observation is counterintuitive, because ketones are known to be more C−H acidic than their ester analogues. For example, pK a=14.1 of pentan‐2,4‐dione is almost two orders of magnitude lower than the pK a=15.9 of dimethyl malonate (Figure 5B, left).[ 27a , 27b ] An inverse ordering is found, however, when the acidity of a cyclic ketone is compared with that of a structurally analogous lactone (Figure 5B, right): Meldrum's acid (pK a=7.3) is the stronger C−H acid than dimedone (pK a=11.2).[ 27a , 27b ]

In an ester group with s‐(Z) conformation, favorable nO→σ*(CO) interaction (known as ‘ester resonance’) transfers electron density from the oxygen lone pair into the antiperiplanar antibonding orbital. [27b] As a consequence, s‐(Z) methyl acetate is preferred by 36 kJ mol−1 if compared to s‐(E) methyl acetate. In the corresponding ester enolate structures, the s‐(Z) geometry remains preferred by 17 kJ mol−1 over the s‐(E) form (Figure 5B, center). [27c] Both ester groups in Meldrum's acid are locked in the s‐(E) geometry and nO→σ*(CO) interactions are ineffective. Besides the impact of the locked s‐(E) geometry of ester groups in lactones on C−H acidity (Figure 5B) and nucleophilicities, [27d] also the electrophilic reactivities of lactones are affected. The lack of nO→σ*(CO) donation gives rise, for example, to enhanced hydrolysis rates of lactones with four to eight ring members. [27e] Furthermore, lactone‐derived Michael acceptors react faster with pyridinium and sulfonium ylides than structurally related open‐chain α,β‐unsaturated esters or cyclic enones. [27f] The observed reactivity ordering of 1 a/1 b being more electrophilic than 1 c/1 e is, thus, in accord with previous observations and can be explained by stereoelectronic effects.[ 27b , 27c , 27d , 27e , 27f ]

Correlation Analysis for the Reactions of 1 with Thiophenolates 2

  1. Correlations with nucleophilicities N of thiophenolates: Variation of the substituents in the thiophenolates 2 af modulated the second‐order rate constants in reactions with 1 (Figure 6

    Figure 6.

    Figure 6

    A) Nucleophilic attack of thiophenolates 2 at electrophilic cyclopropanes 1. B) Applying the Mayr‐Patz equation [Eq. (1)] to reactions of 1 with 2 results in a linear increase of (log k 2)/sN with the nucleophilicity descriptors N of the thiophenolates 2 (with N, s N from ref. [21]). C) Brønsted plots for the cyclopropanes 1 d, 1 g, and 1 j show linear correlations of the second‐order rate constants (log k 2) for the ring‐opening reactions of 1+2 with the basicities of the thiophenolates 2 ae (pK aH in DMSO, from ref. [28]). D) Linear correlation of the second‐order rate constants (log k 2) for reactions of 1 fi+2 with the Hammett substituent constants σ p or σ m of the nucleophiles 2 af. For clarity, data for reactions of 2 with 1 h are not shown. E) Curved relationship of the second‐order rate constant (log k 2) for the reactions of 1 fi+2 with the Hammett substituent constants σ of the electrophiles 1 fi.

    A) only moderately. For example, exchanging p‐methoxy in 2 a for p‐trifluoromethyl in 2 f reduced the second‐order rate constants k 2 in the reactions with 1 e or 1 g, respectively, only within one order of magnitude (Table 1).

  2. By using a rearranged form of Equation (1), a linear relationship of good quality is observed when the ratio (log k 2)/sN is correlated with the recently reported nucleophilicity parameters N of the thiophenolates in DMSO, [21] as exemplified for the reactions of 1 e (r 2=0.9907) and 1 g (r 2=0.9888) in Figure 6B. However, reactivity parameters N and s N for 2 were calibrated by utilizing kinetics of reactions, in which only one new σ‐bond was generated. [18] As a consequence, it is expected that the slopes of the correlations in Figure 6B deviate from unity because SN2‐type ring‐opening reactions of electrophilic cyclopropanes are comprised of coupled bond‐breaking and bond‐making processes. An assignment of Mayr E parameters to cyclopropanes 1 is, therefore, not possible at the moment. Nevertheless, linear correlations as shown in Figure 6B (and Section 5.11 in the Supporting Information) can be used to predict reaction times at least for the reactions of 1 with further thiophenolates in polar, aprotic solvents.

  3. Brønsted correlations: Reported pK aH values in DMSO show that the Brønsted basicity of thiophenolates 2 increases when going from acceptor‐ to donor‐substituted derivatives. [28] If substituents in 2 are varied in the meta and para positions, that is, remote from the reacting sulfur atom, the logarithmic second‐order rate constants (log k 2) of the ring‐opening reactions with the cyclopropanes 1 increase linearly with the increase of pK aH of 2 (Figure 6C). Brønsted β Nuc values, that is, the slope of the correlations in Figure 6C, do not show significant differences between the β Nuc values of 0.22 and 0.24 for the spiro‐activated cyclopropanes 1 d and 1 g, respectively, and that for the 1,1‐dicyano‐activated 1 j (β Nuc=0.26), which is the only electrophilic cyclopropane in this study without a spiro motif. Similar β Nuc values have been reported for the ring‐opening reactions of pyridines (in MeCN) with 1 b (β Nuc=0.26) [14] and a 1,2‐di(dimedone)‐substituted cyclopropane (β Nuc=0.24). [15]

  4. Hammett correlations for thiophenolates: As shown for the kinetics of a series of ring‐opening reactions of thiophenolates 2 with indandione‐derived spirocyclopropanes 1 f, 1 g, and 1 i (Figure 6D), the second‐order rate constants log k 2 follow a linear correlation with Hammett substituent constants σ [29] of the thiophenolates 2. The linear relationships in Figure 6D provide another possibility to straightforwardly predict rate constants for further thiophenolates that have not been included in this work.

  5. Hammett correlations for cyclopropanes: The relative order of the correlation lines in Figure 6D readily indicates that indandione‐derived cyclopropanes with substituents at the 2‐phenyl ring undergo faster reactions with the thiophenolates 2 than the unsubstituted 1 g. Accordingly, systematic variation of the substituents at the 2‐phenyl group of 1 fi in reactions with a given thiophenolate 2 did not result in linear but in U‐shaped correlations of log k 2 with Hammett σ constants (Figure 6E). In each reaction series, log k 2 goes through a minimum for the parent 2‐phenyl‐substituted cyclopropane 1 g. Both the installation of electron‐donating (1 h, 1 i) and electron‐withdrawing substituents (1 f) at the 2‐phenyl group in 1 increase the electrophilic reactivity of the cyclopropanes 1 relative to the parent 1 g. Analogous parabolic Hammett plots were observed by Hudson and Klopman in reactions of lithium thiophenolates with benzyl bromides in methanol at 20 °C, which follow an SN2 mechanism. [30]

The required high degree of organization in transition states of SN2 reactions causes a mix of stabilizing effects, highly dependent on the participating electrophiles, nucleophiles, leaving groups and solvents. The mutual interactions of at minimum six atoms plus solvent molecules in the SN2 TS leave enough freedom for the ensemble to use different effects to lower the energetic barrier.

Attractive electrostatic interactions between the electrophile, the incoming nucleophile, and the leaving group have been identified to be the main factors that influence the activation barriers of SN2 reactions at electrophilic carbon centers with adjacent π‐systems, that is, substitutions at benzylic or allylic substrates.[ 31 , 32 , 33 ] In the kinetics summarized in Figure 6E, nucleophile, leaving group, and solvents are kept constant. Hence, the TS are only influenced by the structural features of the electrophile.

In accord with previous interpretations,[ 31 , 32 ] we observed that EWGs (as in 1 f) stabilize the TS in reactions with highly basic, negatively charged nucleophiles by favorable electrostatic interactions. With EDGs in para‐position at the phenyl ring in (hyper)conjugation with the electrophilic carbon (as in 1 h and 1 i), however, cyclopropane C−C bond polarization is enhanced. This gives rise to a more electron‐deficient electrophilic center and promotes a slightly more advanced bond‐breaking in the TS. [25] In this way, also EDGs at the 2‐phenyl ring lower the energetic barrier for nucleophilic attack. In short, EDGs and EWGs provoke a slight deviation from a synchronous SN2 mechanism and both enable decreased activation barriers. The underlying reasons for the SN2‐accelerating effect of EWG‐ and that of EDG‐substituted phenyls at the electrophilic center of cyclopropanes are different, however, as illustrated in the More O'Ferrall‐Jencks diagram in Figure 7.

Figure 7.

Figure 7

SN2‐accelerating effects in the TS of reactions of 2‐aryl‐substituted cyclopropanes with thiophenolates.

Variation of substituents in SnCl4‐catalyzed (3+2)‐cycloadditions of 2‐aryl‐1,1‐di(alkoxycarbonyl)cyclopropanes with p‐fluorobenzaldehyde recently studied by the Werz group resulted in a linear Hammett relationship, in which EDG‐substituted substrates reacted faster than EWG‐substituted analogues. [16a] Accordingly, Werz reported that relaxed force constants calculated for the C1−C2 bonds, which undergo cleavage in the course of the reactions, correlated with the relative reaction rates: lower force constants for EDG‐substituted cyclopropanes were associated with faster reactions. We assume that complexation with the Lewis acid SnCl4 enhanced the electron‐deficiency of the 1,1‐di(alkoxycarbonyl)‐substituted cyclopropanes such that their reactivity was entirely governed by the degree of the C−C bond polarization, consistent with the intimate ion pair mechanism suggested for this reaction by Johnson and coworkers. [17]

Conclusion

The kinetics of the SN2‐type ring‐opening reactions of thiophenolates at Meldrum's acid‐, dimedone‐, and indandione‐derived spiro‐activated cyclopropanes were investigated by photometric methods. Product studies indicated selective nucleophilic substitutions at the C2 position of the cyclopropane ring, and kinetics revealed that SN2 reactions occurred three to 15 times faster when the C2 carried an aryl group. Linear correlations of the determined second‐order rate constants in DMSO with Mayr nucleophilicity parameters (N/s N), Brønsted basicities (pK aH), and Hammett substituent parameters (σ) of thiophenolates facilitate the rational prediction of reaction rates for so‐far‐unexplored cyclopropane−thiophenolate combinations.

When the reactivity of the electrophiles was varied by installing EWG and EDG groups at the phenyl rings attached to the 2‐position of the cyclopropanes, a parabolic Hammett relationship was observed and rationalized by a gradual change from favorable electrostatic interactions (with EWGs) to an enhanced dissociative character of the C−C bond (with EDGs) in the transition states of the SN2 reactions. Computations by the Aggarwal group showed that the rate‐enhancing effect of EWG on SN2 reactions might be limited to reactions with negatively charged nucleophiles. [32] It remains to be tested, therefore, whether the accelerating effect of EWGs at the electrophilic cyclopropanes vanishes when neutral nucleophiles are used instead of the anionic thiophenolates.

The kinetic studies in this work provide insight into the inherent reactivity of the cyclopropanes 1, which is not shifted by complexation of the carbonyl groups with acids. The astonishing observation that reactions under neutral or basic conditions are accelerated by both EDGs and EWGs at the 2‐phenyl groups of the cyclopropanes opens interesting perspectives to enhance the efficiency and synthetic versatility of ring‐opening nucleophilic reactions at cyclopropanes. Not only vicinally donor−acceptor substituted cyclopropanes but also those with the corresponding acceptor−acceptor motifs will facilitate nucleophilic attack.

Currently, we are investigating whether the ordering of electrophilic reactivity of the cyclopropanes, that we referenced toward thiophenolates in this work, also enables a rational design of reactions with other types of nucleophiles.

Conflict of interest

The authors declare no conflict of interest.

Supporting information

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer reviewed and may be re‐organized for online delivery, but are not copy‐edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

Supporting Information

Acknowledgements

We thank the Department Chemie (LMU München) and the German Academic Exchange Service DAAD (fellowship to A.S.) for financial support. Open Access funding enabled and organized by Projekt DEAL.

P. M. Jüstel, A. Stan, C. D. Pignot, A. R. Ofial, Chem. Eur. J. 2021, 27, 15928.

Dedicated to the memory of Professor Klaus Hafner

References

  • 1. Danishefsky S., Acc. Chem. Res. 1979, 12, 66–72. [Google Scholar]
  • 2. Verhé R., de Kimpe N. in The Chemistry of the Cyclopropyl Group, Vol. 1 (Ed.: Rappoport Z.), Wiley, Chichester, 1987, pp. 445–564. [Google Scholar]
  • 3. Danishefsky S., Singh R., J. Am. Chem. Soc. 1975, 97, 3239–3241. [Google Scholar]
  • 4. Reissig H.-U., Zimmer R., Chem. Rev. 2003, 103, 1151–1196. [DOI] [PubMed] [Google Scholar]
  • 5.For selected recent reviews, see:
  • 5a. Cavitt M. A., Phun L. H., France S., Chem. Soc. Rev. 2014, 43, 804–818; [DOI] [PubMed] [Google Scholar]
  • 5b. Talukdar R., Saha A., Ghorai M. K., Isr. J. Chem. 2016, 56, 445–453; [Google Scholar]
  • 5c. Budynina E. M., Ivanov K. L., Sorokin I. D., Melnikov M. Y., Synthesis 2017, 49, 3035–3068; [Google Scholar]
  • 5d. Liu J., Liu R., Wei Y., Shi M., Trends Chem. 2019, 1, 779–793; [Google Scholar]
  • 5e. Pirenne V., Muriel B., Waser J., Chem. Rev. 2021, 121, 227–263; [DOI] [PubMed] [Google Scholar]
  • 5f. Craig A. J., Hawkins B. C., Synthesis 2020, 52, 27–39; [Google Scholar]
  • 5g. Augustin A. U., Werz D. B., Acc. Chem. Res. 2021, 54, 1528–1541; [DOI] [PubMed] [Google Scholar]
  • 5h. Ghosh K., Das S., Org. Biomol. Chem. 2021, 19, 965–982. [DOI] [PubMed] [Google Scholar]
  • 6. 
  • 6a. Grover H. K., Emmett M. R., Kerr M. A., Org. Biomol. Chem. 2015, 13, 655–671; [DOI] [PubMed] [Google Scholar]
  • 6b. Ivanova O. A., Trushkov I. V., Chem. Rec. 2019, 19, 2189–2208. [DOI] [PubMed] [Google Scholar]
  • 7. Carson C. A., Kerr M. A., Chem. Soc. Rev. 2009, 38, 3051–3060. [DOI] [PubMed] [Google Scholar]
  • 8. Reiser O., Isr. J. Chem. 2016, 56, 531–539. [Google Scholar]
  • 9. Pellissier H., Tetrahedron 2008, 64, 7041–7095. [Google Scholar]
  • 10. Wang L., Tang Y., Isr. J. Chem. 2016, 56, 463–475. [Google Scholar]
  • 11. 
  • 11a. Young I. S., Kerr M. A., Angew. Chem. Int. Ed. 2003, 42, 3023–3026; [DOI] [PubMed] [Google Scholar]; Angew. Chem. 2003, 115, 3131–3134; [Google Scholar]
  • 11b. Garve L. K., Petzold M., Jones P. G., Werz D. B., Org. Lett. 2016, 18, 564–567; [DOI] [PubMed] [Google Scholar]
  • 11c. Garve L. K., Pawliczek M., Wallbaum J., Jones P. G., Werz D. B., Chem. Eur. J. 2016, 22, 521–525; [DOI] [PubMed] [Google Scholar]
  • 11d. Augustin A. U., Sensse M., Jones P. G., Werz D. B., Angew. Chem. Int. Ed. 2017, 56, 14293–14296; [DOI] [PubMed] [Google Scholar]; Angew. Chem. 2017, 129, 14481–14485. [Google Scholar]
  • 12. 
  • 12a. Hudlický T., Kutchan T. M., Naqvi S. M., Org. React. 2004, 33, 247–335; [Google Scholar]
  • 12b. Krüger S., Gaich T., Beilstein J. Org. Chem. 2014, 10, 163–193. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13. Schneider T. F., Kaschel J., Werz D. B., Angew. Chem. Int. Ed. 2014, 53, 5504–5523; [DOI] [PubMed] [Google Scholar]; Angew. Chem. 2014, 126, 5608–5628. [Google Scholar]
  • 14. McKinney M. A., Kremer K. G., Aicher T., Tetrahedron Lett. 1984, 25, 5477–5480. [Google Scholar]
  • 15. Ohkata K., Nagai T., Tamura A., Nandate M.-a., Hanafusa T., J. Chem. Soc. Perkin Trans. 2 1982, 1255–1259. [Google Scholar]
  • 16. 
  • 16a. Kreft A., Lücht A., Grunenberg J., Jones P. G., Werz D. B., Angew. Chem. Int. Ed. 2019, 58, 1955–1959; [DOI] [PubMed] [Google Scholar]; Lücht A., Kreft A., Grunenberg J., Jones P. G., Werz D. B., Angew. Chem. 2019, 131, 1975–1979; [Google Scholar]
  • 16b. Kreft A., Jones P. G., Werz D. B., Org. Lett. 2018, 20, 2059–2062. [DOI] [PubMed] [Google Scholar]
  • 17. 
  • 17a. Pohlhaus P. D., Sanders S. D., Parsons A. T., Li W., Johnson J. S., J. Am. Chem. Soc. 2008, 130, 8642–8650; [DOI] [PubMed] [Google Scholar]
  • 17b. Campbell M. J., Johnson J. S., Parsons A. T., Pohlhaus P. D., Sanders S. D., J. Org. Chem. 2010, 75, 6317–6325. [DOI] [PubMed] [Google Scholar]
  • 18. 
  • 18a. Mayr H., Patz M., Angew. Chem. Int. Ed. Engl. 1994, 33, 938–957; [Google Scholar]; Angew. Chem. 1994, 106, 990–1010; [Google Scholar]
  • 18b. Mayr H., Ofial A. R., SAR QSAR Environ. Res. 2015, 26, 619–646; [DOI] [PubMed] [Google Scholar]
  • 18c. Mayr H., Tetrahedron 2015, 71, 5095–5111; [Google Scholar]
  • 18d.a database of Mayr's reactivity parameters E, N, and s N is freely accessible at www.cup.lmu.de/oc/mayr/DBintro.html.
  • 19. Mayr H., Angew. Chem. Int. Ed. 2011, 50, 3612–3618; [Google Scholar]; Angew. Chem. 2011, 123, 3692–3698. [Google Scholar]
  • 20. 
  • 20a. Phan T. B., Breugst M., Mayr H., Angew. Chem. Int. Ed. 2006, 45, 3869–3874; [DOI] [PubMed] [Google Scholar]; Angew. Chem. 2006, 118, 3954–3959; [Google Scholar]
  • 20b. Mayer R. J., Tokuyasu T., Mayer P., Gomar J., Sabelle S., Mennucci B., Mayr H., Ofial A. R., Angew. Chem. Int. Ed. 2017, 56, 13279–13282; [DOI] [PubMed] [Google Scholar]; Angew. Chem. 2017, 129, 13463–13467. [Google Scholar]
  • 21. Jüstel P. M., Pignot C. D., Ofial A. R., J. Org. Chem. 2021, 86, 5965–5972. [DOI] [PubMed] [Google Scholar]
  • 22. 
  • 22a. Nambu H., Ono N., Hirota W., Fukumoto M., Yakura T., Chem. Pharm. Bull. 2016, 64, 1763–1768; [DOI] [PubMed] [Google Scholar]
  • 22b. Tukhtaev H. B., Ivanov K. L., Bezzubov S. I., Cheshkov D. A., Melnikov M. Y., Budynina E. M., Org. Lett. 2019, 21, 1087–1092. [DOI] [PubMed] [Google Scholar]
  • 23. 
  • 23a. Cloke J. B., J. Am. Chem. Soc. 1929, 51, 1174–1187; [Google Scholar]
  • 23b. Wilson C. L., J. Am. Chem. Soc. 1947, 69, 3002–3004; [Google Scholar]
  • 23c. Zhang M., Li T., Cui C., Song X., Chang J., J. Org. Chem. 2020, 85, 2266–2276. [DOI] [PubMed] [Google Scholar]
  • 24. Qian P., Du B., Song R., Wu X., Mei H., Han J., Pan Y., J. Org. Chem. 2016, 81, 6546–6553. [DOI] [PubMed] [Google Scholar]
  • 25. Streitwieser A. Jr., Chem. Rev. 1956, 56, 571–752. [Google Scholar]
  • 26. Allgäuer D. S., Jangra H., Asahara H., Li Z., Chen Q., Zipse H., Ofial A. R., Mayr H., J. Am. Chem. Soc. 2017, 139, 13318–13329. [DOI] [PubMed] [Google Scholar]
  • 27. 
  • 27a.Internet Bond-energy Databank (pK a and BDE)-iBonD 2.0 Home Page, http://ibond.nankai.edu.cn or http://ibond.chem.tsinghua.edu.cn;
  • 27b. Alabugin I. V., Kuhn L., Medvedev M. G., Krivoshchapov N. V., Vil’ V. A., Yaremenko I. A., Mehaffy P., Yarie M., Terent'ev A. O., Zolfigol M. A., Chem. Soc. Rev. 2021, 50, 10253–10345 (correction: I. V. Alabugin, L. Kuhn, M. G. Medvedev, N. V. Krivoshchapov, V. A. Vil’, I. A. Yaremenko, P. Mehaffy, M. Yarie, A. O. Terent'ev, M. A. Zolfigol, Chem. Soc. Rev. 2021, 50, 10700–10702), and refs cited therein; [DOI] [PubMed] [Google Scholar]
  • 27c. Wiberg K. B., Laidig K. E., J. Am. Chem. Soc. 1988, 110, 1872–1874; [Google Scholar]
  • 27d. Corral-Bautista F., Mayr H., Eur. J. Org. Chem. 2015, 7594–7601; [Google Scholar]
  • 27e. Huisgen R., Ott H., Tetrahedron 1959, 6, 253–267; [Google Scholar]
  • 27f. Mayer R. J., Allihn P. W. A., Hampel N., Mayer P., Sieber S. A., Ofial A. R., Chem. Sci. 2021, 12, 4850–4865. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Bordwell F. G., Hughes D. L., J. Org. Chem. 1982, 47, 3224–3232. [Google Scholar]
  • 29. Hansch C., Leo A., Hoekman D., Exploring QSAR: Hydrophobic, Electronic, and Steric Constants, American Chemical Society, Washington, D. C., 1995. [Google Scholar]
  • 30. Hudson R., Klopman G., J. Chem. Soc. 1962, 1062–1067. [Google Scholar]
  • 31. Wu C.-H., Galabov B., Wu J. I.-C., Ilieva S., Schleyer P. v. R., Allen W. D., J. Am. Chem. Soc. 2014, 136, 3118–3126. [DOI] [PubMed] [Google Scholar]
  • 32. Robiette R., Trieu-Van T., Aggarwal V. K., Harvey J. N., J. Am. Chem. Soc. 2016, 138, 734–737. [DOI] [PubMed] [Google Scholar]
  • 33.For alternative rationalizations within the framework of valence bond theory, see: Shaik S. S., J. Am. Chem. Soc. 1983, 105, 4359–4367. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer reviewed and may be re‐organized for online delivery, but are not copy‐edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

Supporting Information


Articles from Chemistry (Weinheim an Der Bergstrasse, Germany) are provided here courtesy of Wiley

RESOURCES