Characterizing Nitrilimines with Nuclear Magnetic ... - ACS Publications

650

J. Phys. Chem. B 2008, 112, 650-655

Characterizing Nitrilimines with Nuclear Magnetic Resonance Spectroscopy. A Theoretical Study† Robert C. Mawhinney,‡ Gilles H. Peslherbe,* and Heidi M. Muchall* Centre for Research in Molecular Modeling and Department of Chemistry and Biochemistry, Concordia UniVersity, 7141 Sherbrooke Street West, Montreal, Quebec H4B 1R6, Canada ReceiVed: October 12, 2007; In Final Form: NoVember 6, 2007

The 13C chemical shifts in selected nitrilimines, nitriles, acetylenes, allenes, and singlet carbenes have been calculated using density-functional theory [PBE0/6-311++G(2df,pd)] and the gauge including atomic orbital (GIAO) method. The effects of substitution on the 13C chemical shifts in nitrilimines, R1-CNN-R2, have been examined. The carbon nucleus is generally found to be deshielded by substituents in the order CH3 < NH2 < OH < F. Comparison with nitriles, acetylenes, and allenes shows that this effect is related to the presence of the cumulated functionality, CdNdN. Terminal N-substitution is found to have a larger effect than C-substitution due to a large increase in chemical shielding anisotropy. The electronic structure of nitrilimines has recently been shown to possess a carbene component whose resonance contribution varies widely with substitution, and, as previously reported, insight into the electronic structure can be gained by an analysis of the shielding tensor, especially for carbenes. Accordingly, the components of the diagonalized 13 C shielding tensor for nitrilimines and stable singlet carbenes have been examined. This analysis suggests that diaminonitrilimine, H2N-CNN-NH2, may be a stable carbene, and, to the best of our knowledge, it would be the first acyclic, unsaturated stable carbene ever reported. Finally, a detailed analysis of the 13C chemical shifts shows that an increase in the dipolar character of nitrilimines induces a shielding at the carbon nucleus, while an increase in allenic or carbenic character tends to cause a deshielding.

1. Introduction

CHART 1 (R1-CNN-R2)

The importance of nitrilimines as reagents in the synthesis of natural products is well documented.1 Typically formed in situ as reactive intermediates, nitrilimines are unstable compounds that tend to rearrange2 or dimerize.3 Even relatively stable nitrilimines are known to dimerize, and dimerization is accompanied by the formation of the more stable diazomethane isomer.3 Since the carbon nucleus in diazomethanes is usually 20 ppm more shielded than that in nitrilimines, 13C nuclear magnetic resonance (NMR) spectroscopy has been used to differentiate between the two isomers.3,4 The bulk of experimental data on nitrilimines naturally stems from measured properties of stable species.3 Depending on the geometry at the carbon atom, the electronic structure of nitrilimines is commonly described as being either propargylic (RS-P) or allenic (RS-A) (Chart 1).3,5,6 For instance, the crystal structure of N-phosphino-thiophosphino-nitrilimine, (Pri2N)2P(S)-CNN-P(NPri2)2, reveals a bent RCN moiety, and the nitrilimine has been described as prototypical allenic.3 Conversely, N-phosphonio-thiophosphino-nitrilimine, (Pri2N)2P(S)-CNN-P+(Me)(NPri2)2, exhibits a linear geometry at the RCN carbon and is referred to as propargylic.3 As a sensitive probe of the local electronic environment, NMR spectroscopy usually provides a means of assessing subtle changes in electronic structure. However, the observed 13C chemical shift for the CNN carbon in these two compounds is †

Part of the “James T. (Casey) Hynes Festschrift”. * Authors to whom correspondence may be addressed. Tel: (514) 848-2424. Fax: (514) 848-2868. E-mail: [email protected] (H.M.M.); [email protected] (G.H.P.). ‡ Current address: Department of Chemistry, Lakehead University, Thunder Bay, Ontario P7B 5E1, Canada.

relatively closes61 ppm for the former versus 70 ppm for the latter3sdespite the significant change in the local electronic environment implied by the allenic (RS-A) and propargylic (RSP) descriptions (Chart 1). The electronic structure of nitrilimines is more complex, though, as one can write at least six fundamental canonical resonance structures.7 As reported earlier, the carbenic resonance form (RS-C) is significantly more important than previously thought, and its contribution can be increased dramatically by substituents such as NH2 and F.8 More recently, we have shown that the electronic structure of nitrilimines is sufficiently described with allenic (RS-A), dipolar (RS-D), and carbenic (RS-C) contributions, and that allenic and dipolar structures arise from the stabilization of the empty p-orbital and the electron lone pair, respectively, on the carbene carbon (Figure 1).9 Furthermore, on the basis of a large carbenic resonance contribution to its electronic structure and a calculated molecular geometry that is typical of a carbene, we previously suggested that diaminonitrilimine (H2N-CNN-NH2) may be a stable carbene.8

10.1021/jp709968d CCC: $40.75 © 2008 American Chemical Society Published on Web 12/07/2007

NMR Characterization of Nitrilimines

J. Phys. Chem. B, Vol. 112, No. 2, 2008 651

Figure 1. The three main resonance structures of nitrilimines (allenic RS-A, carbenic RS-C, and dipolar RS-D), and the orbital alignment necessary for resonance.

TABLE 1: R

13C

Chemical Shifts (ppm) in Nitrilimines, Nitriles, Acetylenes, and Allenesa RCNNH

HCNNR

RCNNR

H

66

BH2 CH3

93 66

64 69

94 72

NH2 OH F

86 100 112

117 140 168

241 170 199

RCN

RCCH

110 (110.6)b 130 118 (118.9)b 115 114 111

73 (71.9)c 98 82 (79.8)c 82 89 93

RHCCCH2

RHCCCH2 75 (74.8)c

95 88

66 76

106 123 135

86 93 96

a Chemical shifts for the italicized carbon atom with reference to TMS, calculated with GIAO-PBE0/6-311++G(2df,pd). Experimental chemical shifts given in parentheses. b From ref 33. c From ref 34.

The prediction of 13C NMR chemical shifts by theoretical techniques has become common.10 Not only can the effects of ordinary substituents on organic compounds be routinely examined,11 but the NMR spectrum of unknown species can also be determined with some confidence.12,13 In addition, since theoretical assessments include the calculation of the full shielding tensor, a more comprehensive analysis is possible,14 which has proven to be invaluable in characterizing stable carbenes.15-18 In this article, the calculated 13C chemical shielding properties for a series of mono- and disubstituted nitrilimines are reported. To the best of our knowledge, this is the first theoretical account of the chemical shielding properties of nitrilimines, except for a recent article that reported the 13C chemical shift for N-methylmethoxycarbonyl-nitrilimine,19 although isomeric forms of H-CNN-H have previously been investigated.20 The effect of substitution on the 13C chemical shifts in nitrilimines is first assessed and compared to substituent effects in three known systems, i.e., in nitriles, acetylenes, and allenes. This is followed by an evaluation of the components of the shielding tensor responsible for the chemical shifts. Finally, the relationship between the 13C chemical shifts and the contributions from the three major resonance structures (Figure 1) is examined. 2. Computational Methods All results presented were obtained with the Gaussian suite of programs21 using the hybrid density-functional of Perdew, Burke, and Ernzerhof (PBE0)22-24 and the 6-311++G(2df,pd) split-valence triple-ζ basis set with polarization and diffuse functions for all atoms. All geometries were optimized and characterized by vibrational analysis, and the stability of all wave functions was confirmed. Shielding tensors were calculated for the optimized geometries using the gauge including atomic orbitals (GIAO) method.25-27 The combination of densityfunctional theory (DFT) and the GIAO method is reportedly one of the most efficient and reliable approaches for assessing chemical shielding properties.10,13,28,29 All chemical shifts are reported with respect to the isotropic chemical shielding of carbon in tetramethylsilane (TMS), which has a calculated PBE0/6-311++G(2df,pd) value of 188.0 ppm, in good agreement with the experimental value (188.1 ppm,30 186.4 ppm31). The natural chemical shielding (NCS) analysis,32 in which the

shielding tensor is partitioned into magnetic contributions from electron lone pairs and bond orbitals, was employed with the same model chemistry to uncover carbene lone pair contributions. 3. Results and Discussion A. 13C Chemical Shifts. The GIAO-PBE0/6-311++G(2df, pd) calculated 13C chemical shifts for mono- (R-CNN-H and H-CNN-R) and disubstituted (R-CNN-R) nitrilimines, nitriles (R-CtN), acetylenes (R-CtC-H), and allenes (RHCd CdCH2), with substituents H, BH2, CH3, NH2, OH, and F, are given in Table 1 along with available experimental data.33,34 The calculated values are in very good agreement with the experimental 13C chemical shifts, with a maximum error of only 2 ppm. Moreover, the effect of methyl substitution, which causes the carbon nucleus in both nitriles and acetylenes to become deshielded by 8 ppm, is reproduced. These findings further support the recent assessments that the GIAO-DFT combination is a very reliable technique for predicting 13C chemical shifts.10,13,28,29 For nitrilimines, 8 of the 16 systems investigated are predicted to have chemical shifts below 100 ppm. These are comparable to the experimental 13C chemical shifts observed for stabilized nitrilimines, which range from 45 to 86 ppm.3 Ditritylnitrilimine, Ph3C-CNN-CPh3, has the largest experimental value, 86 ppm,3 and our predicted value of 72 ppm for dimethylnitrilimine, H3C-CNN-CH3, (Table 1) is in moderately good agreement with this experimental chemical shift, even though hydrogen is obviously a poor model of a phenyl group. Similarly, our calculated chemical shift for H2B-CNN-BH2, 94 ppm (Table 1), can be compared with the experimentally observed values of 65.8, 66.8, and 68.0 ppm for (iPr2N)2B-CNN-B(NPri2)2, ((c-C6H11)2N)2B-CNN-B(N(c-C6H11)2)2 and ((c-C6H11)2N)2BCNN-B(NPri2)2, respectively.3 The poor agreement can be attributed to the presence of amino groups on the borane substituent in the stabilized nitrilimines, which diminish the role of the empty boron p-orbital and thus leave a more shielded carbon nucleus; but even so, experimental and calculated chemical shifts agree in magnitude. The predicted 13C chemical shift of the parent nitrilimine, H-CNN-H, is 66 ppm, and C-substitution causes a deshielding of the nitrilimine carbon nucleus (Table 1). The deshielding

652 J. Phys. Chem. B, Vol. 112, No. 2, 2008 increases in the order CH3 < NH2 < BH2 < OH < F, and the range of chemical shifts (∆δ), calculated as the difference between the most and the least shielded carbon atom, is 46 ppm. In nitriles, the order of deshielding of the carbon atom is almost entirely reversed (F < OH < NH2 < CH3 < BH2), while the substituted carbon atom in acetylenes exhibits approximately the same order of deshielding as that in nitrilimines (CH3 ) NH2 < OH < F < BH2), with the exception of BH2. Since BH2 causes the largest deshielding in nitriles and acetylenes, and the larger deshielding observed upon NH2, OH, and F substitution in nitrilimines is not present, the range of 13C chemical shifts in the substituted nitriles and acetylenes is relatively small (∆δ ) 19 and 16 ppm, respectively). On the other hand, the substituted carbon atom in allenes (Table 1) exhibits both the same order (CH3 < BH2 < NH2 < OH < F) and magnitude (∆δ ) 47 ppm) of chemical shift changes as that in nitrilimines, indicating that the cumulated framework (CdNdN) is, at least partially, responsible for the observed substituent effect in nitrilimines. This is further supported by the fact that the same order of deshielding is also observed in the calculated 13C chemical shifts of the related ketenimines (RHCdCdNH), although the range is much larger (∆δ ) 79 ppm).11 N-substitution in nitrilimines has the same general effect on 13C chemical shielding as C-substitution, with the exception of the BH2 group, which causes a slight shielding of 2 ppm. Interestingly, the range of chemical shifts (∆δ ) 104 ppm) is more than twice that for C-substitution, despite the substituent being three bonds removed from the carbon atom. The 13C chemical shifts of N-substituted nitrilimines can be compared to those for the gamma carbon atom in substituted allenes, RHCdCdCH2, which appear to follow a similar order of deshielding upon substitution: BH2 < CH3 < NH2 < OH < F (Table 1). However, the range of chemical shifts in the allenes is more modest (∆δ ) 30 ppm), as might be expected from a remote substituent. Therefore, we must conclude that the large 13C chemical shift range in nitrilimines is an indication of a significant change in the local electronic structure at carbon, consistent with the resonance picture (Figure 1).8,9 Naturally, disubstitution in nitrilimines also causes a deshielding of the carbon nucleus, with the order of deshielding CH3 < BH2 < OH < F < NH2. Inspection of the chemical shifts for the methyl-, hydroxyl- and fluorosubstituted nitrilimines (Table 1) reveals that the effect of symmetrical disubstitution (a change of 5, 104, and 133 ppm, respectively, from H-CNN-H) is approximately the sum of the two individual monosubstitution effects (changes of 3, 108, and 148 ppm, respectively). However, the 13C chemical shift in diaminonitrilimine does not follow this trend, with the change due to symmetric disubstitution (175 ppm) being more than twice the sum of the chemical shift changes for monosubstitution (71 ppm). As shown previously, the carbene character of nitrilimines increases upon substitution (Table S1 of the Supporting Information), and the properties of diaminonitrilimine, H2NCNN-NH2, suggest a stable carbene.8,9 For perspective, the 13C chemical shift for the carbon nucleus in a series of acyclic and cyclic (Chart 2) singlet carbenes has been assessed with our chosen model chemistry, and the results are collected in Table 2. The calculated values for the diaminocarbenes are in good agreement with the experimental ranges,35 as well as with those from complete active space self-consistent field (CASSCF) calculations (Table 2).15 On the basis of the 13C chemical shifts for known stable carbenes, a value of 241 ppm (Table 1) for H2N-CNN-NH2

Mawhinney et al. CHART 2

TABLE 2: 13C Chemical Shifts (ppm) for the Carbene Carbon Atom in Singlet Carbenes calculateda H-C-NH2 H-C-OH C(NH2)2 C(OH)2 CF2 1 2 3 4

374 541 250 304 314 222 229 248 252

CASSCFb

experimentalc

236-256 287 212

205-230 235 - 245

a Chemical shifts with reference to TMS, calculated with GIAOPBE0/6-311++G(2df,pd). b From ref 15. c Approximate chemical shift ranges for substituted diaminocarbenes from ref 35.

supports our earlier proposal, which was based on the large increase in carbenic resonance contribution upon substitution and the calculated molecular geometry, that diaminonitrilimine might be a stable carbene.8,9 According to an earlier study,15 the shielding tensor components are an even better probe of the stability of carbenes, and we therefore now turn to an analysis of the individual components of the shielding tensor. B. Tensor Components. Chemical shielding is a rank 2 tensor property. Experimentally obtained chemical shifts (δ) represent average values related to the shielding tensor through eqs 1 and 2,

δ(X) ) σX:Ref - σXiso iso

(1)

1 σXiso ) (σ11 + σ22 + σ33) 3

(2)

where X is the nucleus of interest, X:Ref is the corresponding nucleus in a reference compound (in the case of the 13C nucleus, the carbon atom of TMS), σiso is the isotropic chemical shielding, and σii (i ) 1, 2, 3) are the diagonal components of the shielding tensor sorted in increasing order.10 By definition, σ11 is the most deshielded tensor component, whereas a component lacking deshielding, e.g., the parallel component in acetylenes and nitriles (Table S2, Supporting Information), generally has values between +200 ppm and +300 ppm.14,18,36 The anisotropy and asymmetry are two scalar descriptors commonly used in NMR spectroscopy to reflect the deviation and distribution, respectively, of the tensor components about the average value.10 From the many measures that exist for these two properties, we will use the most straightforward span Ω (eq 3) for anisotropy and skew κ (eq 4) for asymmetry.37

Ω ) |σ33 - σ11|

(3)

σiso - σ22 Ω

(4)

κ)3

A tensor obviously contains more information about the local electronic environment than a scalar quantity, such as the

NMR Characterization of Nitrilimines

J. Phys. Chem. B, Vol. 112, No. 2, 2008 653 TABLE 4: Components of the Diagonalized 13C Shielding Tensor (ppm), Associated Scalar Descriptors Ω (ppm) and K, and Carbon Lone Pair (nC) Contribution (ppm) to the Isotropic Shielding of the CNN Carbon Atom in Nitriliminesa

Figure 2. Principal axis system for carbenes.

TABLE 3: Components of the Diagonalized 13C Shielding Tensor (ppm), Associated Scalar Descriptors Ω (ppm) and K, and Carbon Lone Pair (nC) Contribution (ppm) to the Isotropic Shielding of the Carbene Carbon Atom in Singlet Carbenesa H-C-NH2 H-C-OH C(NH2)2 C(OH)2 CF2 1 2 3 4

σxx

σyy

σzz

Ω

κ

nC

-639 -1078 -318 -408 -435 -260 -231 -312 -281

155 92 165 86 35 158 110 151 101

-73 -72 -34 -25 21 0 -2 -19 -12

794 1170 483 494 470 418 340 462 382

-0.43 -0.72 -0.17 -0.55 -0.94 -0.25 -0.35 -0.27 -0.40

-209 -284 -147 -169 -119 -134 -147 -163

a Calculated with GIAO-PBE0/6-311++G(2df,pd). See Figure 2 for the definition of the coordinate system.

chemical shift, but requires the knowledge of the orientation of the molecule in the magnetic field, as defined by a principal axis system. The defining features of a carbene, a divalent carbon atom with a lone pair, and an orthogonal empty p-orbital provide such an axis system.15 Figure 2 shows the alignment of the carbene lone pair with the z-axis and the empty p-orbital along the y-axis.38 In fact, any sp2-like atom can be described by such a principal axis system.18 When defined in this way, the 13C shielding tensor of a carbene exhibits certain key features: a strongly deshielded σxx component, a less strongly deshielded σzz component, and a σyy component that is not particularly deshielded.15,17,18 The individual 13C shielding tensor components, as well as the anisotropy and asymmetry, of all carbenes studied are given in Table 3. As expected, all carbenes exhibit the above features: σxx is always strongly deshielded with a large negative value, σyy is shielded (positive), and σzz is weakly deshielded. Changing the substituent in the acyclic carbenes from NH2 to either OH or F increases the amount of deshielding in σxx and σyy and decreases it in σzz. As a result particularly prominent in CF2, σyy and σzz approach each other in value, as indicated by a more negative value of κ, which approaches that of an oblate ellipsoid (κ ) -1.00).37 In carbenes, Ω is controlled by changes in σxx and σyy and therefore displays considerable variation. The exceptional stability of diaminocarbenes such as 1 and 2 (Chart 2) is commonly attributed to an increase in the population of the formally empty p-orbital and a decrease in the lone pair density on carbon, which can be monitored by the magnitude of σxx deshielding on the carbene carbon,15 in that more stable carbenes show less deshielding in σxx. According to this criterion, the order of stability of diaminocarbenes is 2 > 1 > 4 > 3 > C(NH2)2, consistent with the literature.39-42 In accord with this, the deshielding contribution from the nC f p interaction of the carbene lone pair with the empty p-orbital on carbon to the isotropic shielding, calculated through an NCS analysis,32 is less for the more stable carbenes (Table 3), as is expected from partial occupation of the p-orbital. The shielding tensor components and scalar descriptors for the carbon nucleus in nitrilimines are given in Table 4. Since the local orientation about this nucleus can also be described

HCNNH R BH2 CH3 NH2 OH F R BH2 CH3 NH2 OH F R BH2 CH3 NH2 OH F

σxx

σyy

204

128

219 216 173 169 166

20 115 120 91 66

195 194 78 -48 -110

96 131 144 144 141

187 199 -240 -44 -114

50 117 119 96 66

σzz

Ω

34 170 R-CNN-H 46 199 35 181 12 161 6 163 -4 170 H-CNN-R 83 112 31 162 -9 153 47 192 29 251 R-CNN-R 44 144 32 167 -38 359 4 139 14 181

κ

nC

-0.11 0.74 0.11 -0.34 -0.04 0.18

-55

0.76 -0.23 -0.14 0.01 -0.11

-89 -102 -125

0.91 -0.02 -0.12 0.32 -0.42

-137 -95 -116

a Calculated with GIAO-PBE0/6-311++G(2df,pd). See Figure 2 for the definition of the coordinate system.

as sp2-like,43 its principal axis system is defined according to Figure 2. The BH2-substituted systems, however, possess Cs symmetry and have a linear geometry about carbon,8,9 which is reflected in the skew. The κ values, ranging from +0.74 to +0.91 (Table 4), are indicative of a linear geometry at carbon as found in nitriles and acetylenes, which are predicted to have skews ranging from +0.81 to +1.00 (Table S2). Unlike in nitriles and acetylenes, though, none of the tensor components in the BH2-substituted nitrilimines lack deshielding, and therefore the electronic structure of the carbon atom cannot actually be described as sp-like.45 Because of their Cs symmetry, the BH2-substituted nitrilimines will be excluded from the remainder of this discussion. The tensor components for the parent nitrilimine H-CNN-H are consistent with the fact that the majority of the carbene character is removed through its resonance with the allenic and dipolar structures (Figure 1).9 This vastly reduces the amount of σxx deshielding, but has only a minor effect on the other two components, σyy and σzz, which retain the aforementioned carbene-like characteristics. As noted above, substituents cause the carbon nucleus in nitrilimines to become more deshielded (Table 1). This is obviously reflected in the tensor components. C-substitution causes all components of the tensor to become more deshielded (Table 4), except for H3C-CNN-H, where the increase in the values of σxx and σzz is exactly compensated by the decrease in the value of σyy, resulting in a 13C chemical shift for C-methylnitrilimine identical to that of H-CNN-H (Table 1). Overall, σxx and σzz exhibit similar maximal changes (-38 ppm), while σyy has a larger overall change (-62 ppm). As a result, Ω does not change much upon substitution, while κ varies around zero. N-substitution was shown above to have an even more pronounced effect than C-substitution on the 13C chemical shift (Table 1). The maximal change in the three tensor components from H-CNN-H [σxx (-314 ppm), σyy (+16 ppm), and σzz (-43 ppm) (Table 4)] demonstrates that this is due to the particularly large deshielding of σxx. The changes in σxx and σzz are consistent with our assessment of a slight general increase in carbene character with N- over C-substitution (Table S1).9

654 J. Phys. Chem. B, Vol. 112, No. 2, 2008

Mawhinney et al.

CHART 3

Unlike for C-substitution, Ω increases significantly upon N-substitution, while κ values are comparable. As mentioned above, C,N-disubstitution can have a dramatic effect on the 13C chemical shift (Table 1), and a general additivity of substituent effects was observed. Diaminonitrilimine, with a very unique 13C chemical shift, is an exception to the general additivity rule and is discussed separately below. The magnitude of the σyy component in the disubstituted species is very similar to that in the C-substituted nitrilimines, while the large changes observed in the σxx and σzz components are also seen upon N-substitution. In general, all three tensor components undergo a deshielding upon disubstitution, changing by as much as -318 (σxx), -62 (σyy), and -72 ppm (σzz) from the corresponding values in H-CNN-H (Table 4). As a result, the tensor components for 13C in F-CNN-F approach those in CF2 (Table 3), reflecting the large carbene character found for this nitrilimine (Table S1).9 The situation is similar for HOCNN-OH, even if it is less pronounced. It was noted above that the carbon nucleus in H2N-CNNNH2 has a chemical shift (Table 1) similar to that calculated for C(NH2)2, 3, and 4 (Table 2). The shielding tensor (Table 4) reveals that the σzz component (-38 ppm) is close to the corresponding value for the acyclic C(NH2)2, (-34 ppm, Table 3). However, the remaining two components, σxx and σyy (-240 and 119 ppm, respectively, Table 4), are more akin to those of a cyclic, unsaturated stable carbene such as 2 (-231 and 110 ppm, Table 3). In fact, with such a σxx value, diaminonitrilimine must be considered a stable carbene. In addition, the span Ω is comparable to that for the stable carbenes 2 and 4. Therefore again, H2N-CNN-NH2 may be best described as an acyclic, unsaturated stable carbene (Chart 3). While the components of the magnetic shielding tensor reveal the increased carbene character of H2N-CNN-NH2 (73% carbene, Table S1), the resemblance (or lack thereof) of the tensor components between the remaining nitrilimines and their respective carbenes is less obvious. The NCS analysis provides information as to the source of the observed deshielding at the nitrilimine carbon, and as is apparent from the data in Table 4, the isotropic shielding of seven of the nitrilimines studied is affected by nC contributions. Six of these molecules possess a carbene character of 35% or more (Table S1), and this is, in hindsight, reflected in the one negative shielding tensor component, σxx. C. 13C Chemical Shifts and Resonance Contributions. Relationships between chemical shifts and molecular or atomic properties, such as charge distribution or nucleus hybridization, are commonly used to explain observed changes in chemical shielding.11 Formula I relates the resonance contributions from our previous work9 (given in Table S1; resonance contributors given in Figure 1) to the 13C chemical shifts, and was obtained through a least-squares linear regression analysis. A plot of the values acquired from Formula I versus the GIAO-PBE0 values is given in Figure 3. As the correlation coefficient for Formula I suggests, this formula reproduces the calculated 13C chemical shifts reasonably well, with nearly all data points close to the one-to-one correspondence line. The correlation is obviously not perfect, and the data point for H-CNN-F seems to be a particular outlier. A closer analysis shows that this is due to the chemical shielding anisotropy, Ω, which is known for its large impact on chemical shifts.34,46 It increases from 153 ppm

Figure 3. Plot of the 13C chemical shift obtained from Formula I versus the GIAO-PBE0/6-311++G(2df,pd) calculated 13C chemical shift. The solid line is the one-to-one correspondence line.

for H-CNN-NH2 to 251 ppm for H-CNN-F (Table 4), leading to an increase in 13C chemical shift (Table 1), despite the similar weights of the three resonance contributors (Table S1).

δ(13C) ) 0.890(%RS-A) - 0.155(%RS-D) + 2.95(%RS-C) r2 ) 0.886 (Formula I) As a final note, according to Formula I, an increase in allenic or carbenic character results in a larger chemical shift of the nitrilimine carbon, while increasing the dipolar character has the opposite effect. In organic synthesis, where carbenic versus dipolar reactivity can lead to different products, knowledge of the behavior of the 13C chemical shift of nitrilimines upon substitution thus might prove a valuable aid for predictive purposes.44 Conclusions The calculated 13C chemical shifts for a series of nitrilimines, R1-CNN-R2, reveal that both C- and N-substitutions deshield the nitrilimine carbon in the order CH3 < NH2 < OH < F. N-substitution is found to have a larger effect than Csubstitution, despite the remoteness of the substituent. As uncovered by an analysis of the components of the shielding tensor, this can be traced back to the role of σyy, the tensor component that does not involve the p-orbital on carbon. Symmetric C,N-disubstitution gives rise to an approximately additive effect on the 13C chemical shift, with the exception of the NH2 substituent, where additivity is surpassed. The chemical shift and shielding tensor components, in fact, suggest that diaminonitrilimine may be an acyclic, unsaturated stable carbene, perhaps the first of its kind. More generally, the present theoretical work not only produces results that are amenable to experimental validation but also further demonstrates that the carbenic character of nitrilimines cannot be ignored, contrary to what has recently been suggested.47 Finally, a formula relating the 13C chemical shifts to the nitrilimine resonance forms shows that an increase in allenic and carbenic resonance contributions deshields the carbon nucleus, whereas an increase in the dipolar contribution causes a shielding, a finding that might be of great practical value to organic chemists. Acknowledgment. This research was supported by a Nouveaux Chercheurs-EÄ quipe grant from the Fonds Que´becois de Recherche sur la Nature et les Technologies (FQRNT) and

NMR Characterization of Nitrilimines partially by the Natural Sciences and Engineering Research Council (NSERC) of Canada. Calculations were performed at the Centre for Research in Molecular Modeling (CERMM), which was established with the financial support of the Concordia University Faculty of Arts & Science, the Ministe`re de l’EÄ ducation du Que´bec (MEQ) and the Canada Foundation for Innovation (CFI). G.H.P. holds a Concordia University Research Chair. Supporting Information Available: Tables for the normalized contributions of the main resonance structures of nitrilimines, the 13C shielding tensor components for acetylenes and nitriles, the dia- and paramagnetic contributions to the 13C shielding tensor components in acetylenes, nitriles, and BH2substituted nitrilimines, and Cartesian coordinates of nitrilimines. This material is available free of charge via the Internet at http:// pubs.acs.org. References and Notes (1) Sharp, J. T. Nitrile ylides and nitrile imines. In The Chemistry of Heterocyclic Compounds 59: Synthetic Applications of 1,3-Dipolar Cycloaddition Chemistry Toward Heterocycles and Natural Products; Padwa, A., Pearson, W. H., Eds.; John Wiley & Sons: New York, 2002. (2) Maier, G.; Eckwert, J.; Bothur, A.; Reisenauer, H. P.; Schmidt, C. Liebigs Ann. 1996, 1041. (3) Bertrand, G.; Wentrup, C. Angew. Chem., Int. Ed. Engl. 1994, 33, 527. (4) von Locquenghien, K. H.; Reau, R.; Bertrand, G. J. Chem. Soc., Chem. Commun. 1991, 1192. (5) Wong, M. W.; Wentrup, C. J. Am. Chem. Soc. 1993, 115, 7743. (6) Faure, J. L.; Reau, R.; Wong, M. W.; Koch, R.; Wentrup, C.; Bertrand, G. J. Am. Chem. Soc. 1997, 119, 2819. (7) Huisgen, R. Angew. Chem., Int. Ed. 1963, 75, 604. (8) Mawhinney, R. C.; Muchall, H. M.; Peslherbe, G. H. Chem. Commun. 2004, 1862. (9) Mawhinney, R. C.; Muchall, H. M.; Peslherbe, G. H. Org. Biomol. Chem. To be submitted for publication. (10) Facelli, J. C. Concepts Magn. Reson., Part A 2004, 20, 42. (11) Tahmassebi, D. Magn. Reson. Chem. 2003, 41, 273. (12) Wiberg, K. B. J. Comput. Chem. 1999, 20, 1299. (13) Bagno, A.; Rastrelli, F.; Saielli, G. J. Phys. Chem. A 2003, 107, 9964. (14) Wiberg, K. B.; Hammer, J. D.; Zilm, K. W.; Cheeseman, J. R. J. Org. Chem. 1999, 64, 6394. (15) vanWu¨llen, C.; Kutzelnigg, W. J. Chem. Phys. 1996, 104, 2330. (16) Alder, R. W.; Blake, M. E.; Oliva, J. M. J. Phys. Chem. A 1999, 103, 11200. (17) Arduengo, A. J.; Dixon, D. A.; Kumashiro, K. K.; Lee, C.; Power, W. P.; Zilm, K. W. J. Am. Chem. Soc. 1994, 116, 6361. (18) Wiberg, K. B.; Hammer, J. D.; Keith, T. A.; Zilm, K. J. Phys. Chem. A 1999, 103, 21. (19) Jalbout, A. F.; Jiang, Z.; Abou-Rachid, H.; Benkaddour, N. N. Spectrochim. Acta, Part A 2004, 60, 603. (20) Jaszunski, M.; Helgaker, T.; Ruud, K.; Bak, K. L.; Jorgensen, P. Chem. Phys. Lett. 1994, 220, 154. (21) (a) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A., Jr.; Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Rega, N.; Salvador, P.; Dannenberg, J. J.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; Baboul, A. G.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.;

J. Phys. Chem. B, Vol. 112, No. 2, 2008 655 Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Andres, J. L.; Gonzalez, C.; Head-Gordon, M.; Replogle, E. S.; Pople, J. A. Gaussian 98, revision A.11.4; Gaussian, Inc.: Pittsburgh, PA, 2002. (b) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A., Jr.; Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Salvador, P.; Dannenberg, J. J.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; Baboul, A. G.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Andres, J. L.; Gonzalez, C.; Head-Gordon, M.; Replogle, E. S.; Pople, J. A. Gaussian 03, revision C.02.; Gaussian, Inc.: Wallingford, CT, 2004. (22) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. ReV. Lett. 1996, 77, 3865. (23) Perdew, J. P.; Emzerhof, M.; Burke, K. J. Chem. Phys. 1996, 105, 9982. (24) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. ReV. Lett. 1997, 78, 1396. (25) Schreckenbach, G.; Ziegler, T. J. Phys. Chem. 1995, 99, 606. (26) Ditchfield, R. Mol. Phys. 1974, 27, 789. (27) Wolinski, K.; Hinton, J. F.; Pulay, P. J. Am. Chem. Soc. 1990, 112, 8251. (28) Cimino, P.; Gomez-Paloma, L.; Duca, D.; Riccio, R.; Bifulco, G. Magn. Reson. Chem. 2004, 42, S26. (29) Magyarfalvi, G.; Pulay, P. J. Chem. Phys. 2003, 119, 1350. (30) Jameson, A. K.; Jameson, C. J. Chem. Phys. Lett. 1987, 134, 461. (31) Raynes, W. T.; McVay, R.; Wright, S. J. J. Chem. Soc., Faraday Trans. 2 1989, 85, 759. (32) Bohmann, J. A.; Weinhold, F.; Farrar, T. C. J. Chem. Phys. 1997, 107, 1173. (33) Jackowski, K.; Molchanov, S.; Gryff-Keller, A. J. Mol. Struct. (THEOCHEM) 2000, 497, 189. (34) Beeler, A. J.; Orendt, A. M.; Grant, D. M.; Cutts, P. W.; Michl, J.; Zilm, K. W.; Downing, J. W.; Facelli, J. C.; Schindler, M. S.; Kutzelnigg, W. J. Am. Chem. Soc. 1984, 106, 7672. (35) Bourissou, D.; Guerret, O.; Gabbai, F. P.; Bertrand, G. Chem. ReV. 2000, 100, 39. (36) Wiberg, K. B.; Hammer, J. D.; Zilm, K. W.; Keith, T. A.; Cheeseman, J. R.; Duchamp, J. C. J. Org. Chem. 2004, 69, 1086. (37) Mason, J. Solid State Nucl. Magn. Reson. 1993, 2, 285. (38) We have chosen this particular alignment to be consistent with the presentation in ref. 15. (39) Cheng, M. J.; Hu, C. H. Chem. Phys. Lett. 2000, 322, 83. (40) Cheng, M. J.; Hu, C. H. Chem. Phys. Lett. 2001, 349, 477. (41) Heinemann, C.; Thiel, W. Chem. Phys. Lett. 1994, 217, 11. (42) Heinemann, C.; Mu¨ller, T.; Apeloig, Y.; Schwarz, H. J. Am. Chem. Soc. 1996, 118, 2023. (43) The nitrilimine frontier molecular orbitals have pseudo-π-symmetry, and their orientation around carbon resembles that of a carbene.44 A topological analysis of the electron density reveals a maximum in the valence shell charge concentration in the “carbene plane”, corresponding to a lone pair on carbon. These results will be presented elsewhere. (44) Mawhinney, R. C.; Muchall, H. M.; Peslherbe, G. H. Can. J. Chem. 2005, 83, 1615. (45) For an sp-type nucleus, the tensor components are defined as being either parallel or perpendicular, and the paramagnetic (deshielding) contribution to the parallel component is always negligible,46 ranging from 0 ppm to +7 ppm for the carbon atom in all nitriles and acetylenes studied (Table S3, Supporting Information). In contrast, for BH2-substituted nitrilimines, it ranges from -68 ppm (C-substitution) to -100 ppm (C,N-disubstitution). (46) Pople, J. A. Proc. R. Soc. London, Ser. A 1957, 239, 541. (47) Cargnoni, F.; Molteni, G.; Cooper, D. L.; Raimondi, M.; Ponti, A. Chem. Commun. 2006, 1030.