Heteroatomic Deprotonation of Substituted Methanes and Methyl

Dec 5, 2012 - Effect of Protonation State and Interposed Connector Groups on Bond Dissociation Enthalpies of Alcohols and Related Systems. Michael ...
0 downloads 0 Views 365KB Size
Download PDF
Article pubs.acs.org/JPCA

Heteroatomic Deprotonation of Substituted Methanes and Methyl Radicals: Theoretical Insights into Structure, Stability, and Thermochemistry Michael Morris, Bun Chan,* and Leo Radom* School of Chemistry and ARC Centre of Excellence for Free Radical Chemistry and Biotechnology, University of Sydney, Sydney, NSW 2006, Australia S Supporting Information *

ABSTRACT: High-level W1w ab initio calculations have been used to investigate the structural and thermochemical changes that result from heteroatomic deprotonation of CH3YH molecules and •CH2YH radicals (YH = BH2, CH3, NH2, OH, AlH2, SiH3, PH2, and SH). The thermochemical quantities considered include gas-phase acidities, various bond dissociation enthalpies, and heats of formation. The high-level ab initio results are compared with available experimental data and generally show excellent agreement. In a small number of cases in which we find discrepancies that persist at even higher theoretical levels (e.g., W4), we suggest that the experimental data should be re-examined. We find that the C−Y bond lengths of •CH2YH contract upon deprotonation, whereas for CH3YH, the predicted effect, in general, is a lengthening of the C−Y bond. These structural changes, for the most part, are reflected in the changes to the C−Y bond dissociation enthalpies. The CH3YH molecules are calculated to be 50−200 kJ mol−1 less acidic in the gas phase than the corresponding •CH2YH radicals, indicating relative stabilization of the •CH2Y− radical anions. The structural and thermochemical changes are rationalized using a combination of resonance and orbital interaction arguments.

■

INTRODUCTION Deprotonated molecules are of considerable interest to chemists, not the least because of the ubiquitous nature of proton-transfer reactions. An important quantity related to the deprotonation of a molecule in the gas phase is the gas-phase acidity (hereafter referred to simply as the acidity). It is defined as the enthalpy difference (ΔacidH°) between a species (AH) and its deprotonation products (A− + H+): AH → A− + H+

whereas that for deprotonation at the more electronegative nitrogen is 1687 kJ mol−1 (see below).7 A substituted methyl radical •CH2YH can likewise be deprotonated at either C or Y. In this case, the resulting species is either the radical anion of a carbene or the radical anion of a stable, closed-shell molecule. Unlike their closedshell counterparts, the radical anions resulting from deprotonation of •CH2YH are, in general, not particularly stable.8 Therefore, theory has an important role to play in understanding the properties of these species. In this paper, we seek to examine the effects of heteroatomic deprotonation on the structure, stability, and thermochemistry of carbon-centered radicals using the high-level W1w procedure9,10 and to compare the results with those for their closed-shell counterparts. We have chosen for this study a set of substituted methanes, CH3YH, and methyl radicals, •CH2YH, where YH represents a range of substituents that includes πdonors (NH2, OH, PH2, and SH), π-acceptors (BH2 and AlH2), and hyperconjugative donors (CH3 and SiH3). This study of the effect of heteroatomic deprotonation of substituted methyl radicals complements previous investigations that have focused on deprotonation of substituted methyl radicals at carbon11 and of substituted methanes at carbon and at the heteroatom.12 It also serves as a basis for our current study of the effect of partial

(1)

The acidities of a large number of molecules have been determined and collected into compendia such as the NIST WebBook.1 In addition, Berkowitz et al.2 and, more recently, Ervin3 have written extensive reviews on the experimental techniques used to obtain acidities. As a supplement to experimental determinations of ΔacidH°, ab initio quantum chemistry has been used to calculate enthalpies of the deprotonation of molecules.4 Weizmann-1 (W1) theory5 has proven particularly successful in generating accurate ΔacidH° values.6 For a substituted methane CH3YH, deprotonation can occur at either the carbon or the heteroatom. The first of these processes results in a carbon-centered anion, whereas the second leads to an anion centered on Y. In most cases, proton removal at the heteroatom is favored. This is because the heteroatom is generally more electronegative than the carbon and can therefore better stabilize the resultant charge. For example, our calculated W1w value of ΔacidH° corresponding to deprotonation at the carbon in CH3NH2 is 1748 kJ mol−1, © XXXX American Chemical Society

Received: October 15, 2012 Revised: November 14, 2012

A

dx.doi.org/10.1021/jp3101927 | J. Phys. Chem. A XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry A

Article

heteroatomic protonation and deprotonation of substituted methyl radicals.

Table 1. Comparison of Calculated and Experimental Ionization Energies of Y and YHa

■

COMPUTATIONAL DETAILS Standard ab initio molecular orbital theory calculations were carried out using the Gaussian 0913 and MOLPRO 201014 programs. Geometries of all species were optimized with the B3-LYP/aug-cc-pV(T+d)Z procedure. Following each geometry optimization, harmonic frequency analysis was performed to confirm the nature of the stationary point as an equilibrium structure. Energies were obtained using the high-level W1w composite procedure.9 In short, W1w employs a series of CCSD and CCSD(T) calculations to approximate the allelectron relativistic CCSD(T)/complete-basis-set level. Full details of this method can be found in the original reference.9 To obtain the zero-point vibrational energies and thermal corrections for enthalpies at 298 K, we used B3-LYP/aug-ccpV(T+d)Z harmonic vibrational frequencies scaled by factors of 0.9884 and 0.9987, respectively.15 Gas-phase acidities (ΔacidH°) at 0 K were calculated as the enthalpy change for the deprotonation reaction 1. Heats of formation (ΔfH°) at 0 K were calculated according to the atomization method outlined by Nicolaides et al.16 Acidities and enthalpies of formation were corrected to 298 K with the use of the scaled B3-LYP/aug-ccpV(T+d)Z vibrational frequencies for the species under consideration and experimental H°298 − H°0 corrections for the atoms in their standard states.17

species

W1w

experimentb

BH BH2 CH2 CH3 NH NH2 O OH AlH AlH2 SiH2 SiH3 PH PH2 S SH

9.81 8.24 10.37 9.84 13.48 11.17 13.59 13.01 8.32 7.12 9.13 8.11 10.17 9.82 10.33 10.39

9.77 ± 0.05 9.8 ± 0.2 10.35 ± 0.15 9.84 ± 0.01 13.49 ± 0.01c 11.14 ± 0.01 13.618 13.017 ± 0.002

9.15 ± 0.02 8.135 ± 0.005 10.149 ± 0.008 9.824 ± 0.002 10.36 10.422 ± 0.001

a

IE values in eV at 0 K. bExperimental values from ref 1. cVertical value.18 The deviation between adiabatic and vertical W1w values of IE(NH) is calculated to be less than 0.02 eV.

Table 2. Comparison of Calculated and Experimental Electron Affinities of Y and YHa

■

RESULTS AND DISCUSSION To facilitate understanding of the quantitative structural and thermochemical data, it is first useful to consider qualitatively how deprotonation affects the bonding in CH3YH and • CH2YH. Effects of Deprotonation on CH3YH. The bonding in substituted methanes and their deprotonated analogs can be described in terms of a series of single-bond (covalent (a)) and no-bond (ionic (b and c)) resonance structures:

In the neutral molecule (1), the ionization energies (IEs) of both CH3 and YH (Table 1) are considerably larger than the electron affinities (EAs) of either CH3 or YH (Table 2), and hence, the covalent resonance structure 1a dominates the description of the C−Y bond. In the deprotonated molecule (2), however, the EA of CH3 is more comparable to the EA of the substituent Y (Table 2), so the no-bond resonance structure 2b plays a more important role in describing the C−Y bond. It can be expected that as the EA of Y decreases, the resonance structure 2b will increase in significance. We note as an aside that the calculated IEs generally agree with experimental values1 to within 0.05 eV (∼ 5 kJ mol−1) (Table 1). For IE(BH2), however, the disagreement is much larger (1.6 eV), and we have carried out very high level W4 calculations19 to try to ascertain the origin of the discrepancy. The W4 value (8.25 eV) for IE(BH2) is very close to the W1w value (8.24 eV).20 This finding suggests that the source of the disagreement lies with the reported experimental value.21 A possible explanation is that the B2H6/BH3+ appearance energy (AE), used to calibrate the experimental results, is inaccurate;

a

species

W1w

BH BH2 CH2 CH3 NH NH2 O OH AlH AlH2 SiH2 SiH3 PH PH2 S SH

0.10 0.26 0.64 0.06 0.34 0.75 1.45 1.82 0.18 1.11 1.08 1.40 1.01 1.26 2.09 2.34

experimentb 0.30 ± 0.25 0.652 ± 0.006 0.08 ± 0.03 0.370 ± 0.004 0.771 ± 0.005 1.461 ± 0.001 1.829 ± 0.001

1.123 ± 0.022 1.405 ± 0.026 1.027 ± 0.006 1.263 ± 0.006 2.07 ± 0.13 2.317 ± 0.002

EA values in eV at 0 K. bExperimental values from ref 1.

there is some disagreement in the literature for AE(B2H6/ BH3+), with values ranging from 12.1 eV22 up to 14.9 eV.23 We believe that the experimental IE for BH2 warrants reexamination. In a similar manner, the calculated EAs for the most part agree with experimental values to better than 0.05 eV (Table 2). The only notable disagreement is for EA(BH), for which the predicted and experimental values differ by 0.20 eV. Once again, high-level W4 computations were used to determine the source of the discrepancy. We find that W4 and W1w essentially predict the same EA (0.11 eV for W4 compared with 0.10 eV for W1w). We therefore suggest that the experimental EA for BH should be reassessed. A consideration of orbital interactions provides further insights into the bonding in CH3YH and CH3Y−. As previously outlined,11,24 when YH is a π-donor substituent (e.g., NH2, OH, PH2, or SH), hyperconjugative electron donation in B

dx.doi.org/10.1021/jp3101927 | J. Phys. Chem. A XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry A

Article

results in a relatively weak π-bond. For Y = BH or AlH, the presence of two singly occupied p(Y) orbitals leads to a twoelectron π-bond as well as a singly occupied orbital of sp2 character at Y. Changes in C−Y Bond Length Due to Deprotonation at Y. An important structural change that accompanies deprotonation of CH3YH and •CH2YH occurs for the C−Y bond lengths. The C−Y bond lengths of the substituted methanes (CH3YH), methyl radicals (•CH2YH), and their deprotonated forms (CH3Y− and •CH2Y−) are given in Table 3. It can be seen that the theoretical values are in good

CH3YH can take place from a lone pair on Y (n(Y)) to a pseudo-π*-orbital of the methyl group (π*(CH3)). When YH is a π-acceptor group (e.g., BH2 or AlH2), hyperconjugative electron donation can take place from a pseudo-π-orbital on CH3 (π(CH3)) to the formally vacant p-orbital on the heteroatom in CH3YH, p(Y). For YH = SiH3 (hyperconjugative donor), hyperconjugative donation occurs from π(YH) to π*(CH3). In CH3Y−, the relevant interaction for all substituents is two-electron donation from n(Y−) to π*(CH3); that is, negative hyperconjugation.24a,25,26 This interaction imparts increased partial-double-bond character to the C−Y bond. For example, in the case of CH3OH, deprotonation at oxygen increases the energy of the oxygen lone pair orbitals. The interaction of the lone pair orbitals with the π* orbitals of the methyl group is consequently increased because of the reduced energy gap (see Figure 1).

Table 3. Optimized C−Y Bond Lengths in Substituted Methanes (CH3YH), Methyl Radicals (•CH2YH), and Their Deprotonated Forms (CH3Y− and •CH2Y−)a,b r(C−Y) YH BH2 CH3 NH2 OH AlH2 SiH3 PH2 SH

Figure 1. Orbital interaction diagram showing increased lone pair donation from the oxygen lone pair orbitals to the methyl π* orbitals following deprotonation, as a result of the reduced energy gap between the interacting orbitals.

a

CH3YH 1.552 1.528 1.464 1.423 1.954 1.878 1.862 1.825

(1.535) (1.471) (1.429) (1.869) (1.858) (1.814)

CH3Y− 1.599 1.526 1.435 1.336 2.061 1.977 1.909 1.840 (1.845)

B3-LYP/aug-cc-pV(T+d)Z values, in Å. parentheses from ref 28.

•

CH2YH

1.523 1.484 (1.492) 1.391 1.365 1.926 1.844 1.768 1.717 b

•

CH2Y− 1.433 1.327 1.331 1.270 1.846 1.817 1.756 1.712

Experimental values in

agreement with the experimental data.28 The mean absolute deviation between theoretical and experimental bond lengths is 0.007 Å, and the largest deviation (for CH3SH) is 0.011 Å. For CH3YH, the effect of deprotonation on the C−Y bond lengths (Δr(C−Y) = r(C−Y − ) − r(C−YH)) can be rationalized using both resonance and orbital interaction arguments. As noted above, deprotonation of CH3YH leads to an enhanced contribution of the no-bond resonance structure 2b, which would result in longer bonds. On the other hand, the negative hyperconjugation between n(Y−) and π*(CH3) in CH3Y− contributes to a shortening of the C−Y bond. When YH = BH2, CH3, AlH2, SiH3, PH2, or SH, the net effect is a lengthening of the C−Y bond, indicating that the contribution of the resonance structure 2b is more important than the negative hyperconjugation. For YH = NH2 or OH, this situation is reversed. These results are in agreement with earlier studies by Boyd et al.12 It is useful to note the variation in Δr(C−Y) with substituent. As one moves across the periodic table, Δr(C− Y) becomes more negative. This arises because of increasing substituent EAs (Table 2): larger substituent EAs diminish the influence of the no-bond resonance structure 2b. Movement downward in a group finds Δr(C−Y) generally becoming more positive. This observation can be attributed to decreasing efficiency in the overlap of the n(Y−) and π*(CH3) orbitals, which appears to be sufficient to counter a slight increase in the substituent EAs. Deprotonation in •CH2YH is accompanied in all cases by a contraction of the C−Y bond. A simple explanation of this behavior for YH = CH3, NH2, OH, SiH3, PH2, or SH is that the interaction between the electron-deficient carbon and the heteroatom is stronger in •CH2Y− than in •CH2YH because of the greater electron-donating capacity of the heteroatom in the radical anion. For YH = BH2 or AlH2, the shortening of the C−

Effects of Deprotonation on •CH2YH. The bonding in • CH2YH radicals and •CH2Y− radical anions can also be described in terms of a series of resonance structures:

For •CH2YH, the C−Y bond is dominated by the single covalent resonance structure 3a; the IEs of both CH2 and YH (Table 1) are significantly larger than the EAs of either CH2 or YH (Table 2). For •CH2Y−, however, both available resonance contributors are needed to describe the C−Y bond. This is because the EAs of most Y are much closer to the EA of CH2 (Table 2), and for Y = BH, NH, and AlH, they are even smaller. It is useful to note that, in contrast to the closed-shell anion (2), the two resonance contributors for the radical anion (4) are of covalent character. Therefore, from a valence-bond perspective, deprotonation in •CH2YH is not necessarily expected to result in a weakening of the C−Y bond. There are further significant orbital interactions present in both •CH2YH and •CH2Y−. When YH is a π-donor group, a three-electron stabilizing interaction in •CH2YH can take place between p(C•) and n(Y).24,27 When YH is a π-acceptor group, a one-electron stabilizing interaction can take place between the radical carbon p(C•) orbital and the formally vacant p(Y) orbital on the heteroatom. For YH = CH3 or SiH3, the pertinent interaction is hyperconjugative donation from π(YH) to p(C•). Interaction between the p(C•) and p(Y) orbitals leads to formation of a π-bond in •CH2Y−. The strength of the π-bond formed is dependent on the substituent Y. When Y = CH2, NH, O, SiH2, PH, or S, partial occupation of the π*(C−Y) orbital C

dx.doi.org/10.1021/jp3101927 | J. Phys. Chem. A XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry A

Article

carbon and the heteroatom in the radical anion. For secondrow substituents, which can better accommodate additional charge, the relative stabilities of the dissociation products must also be considered. Indeed, if one assumes that a shorter bond generally corresponds to a stronger bond, then the observation that the C−Y bond lengths of •CH2AlH2 and •CH2SiH3 decrease upon deprotonation, despite homolytic BDEs also decreasing, must be a consequence of the product stabilities.29 Stabilization Energies of the •CH2YH Radicals and Their Deprotonated Forms •CH2Y−. A commonly used measure of relative stabilities of radicals is the radical stabilization energy (RSE).27,30−32 We have calculated the RSEs for •CH2YH and •CH2Y− radicals, defined as the enthalpic changes for the isodesmic reactions:

Y bond can be attributed to the formation of a conventional C−Y double bond in •CH2Y−, as opposed to the partial double bond that is present in •CH2YH. It is notable that, although the magnitude of Δr(C−Y) is considerable for all first-row substituents, the value of Δr(C−Y) becomes significantly smaller for all second-row substituents with the exception of Y = AlH. This result is a reflection of the two different mechanisms outlined above. That is, the effect of the increased electron-donating capacity of the heteroatom in the case of the π-donors and hyperconjugative donors in •CH2Y− is expected to reduce as atomic size increases. On the other hand, the effect induced by a transition from a partial to a complete double bond for YH = BH2 or AlH2 appears to be less influenced by movement down the periodic table. Changes in C−Y BDE Due to Deprotonation at YH. The homolytic and heterolytic C−Y bond dissociation enthalpies (BDEs) for CH 3 YH, CH 3 Y − , • CH 2 YH, and • CH2Y− are listed in Table 4. The most prominent feature in

•

CH 2YH + CH4 → CH3YH + •CH3

•

CH 2Y − + CH4 → CH3Y − + •CH3

homolytic BDE CH3BH2 • CH2BH2 CH3CH3 • CH2CH3 CH3NH2 • CH2NH2 CH3OH • CH2OH CH3AlH2 • CH2AlH2 CH3SiH3 • CH2SiH3 CH3PH2 • CH2PH2 CH3SH • CH2SH a

heterolytic BDE

C−Y−

C−Y

C−Y−

441.6 507.9 378.3 417.9 354.9 429.3 386.6 447.2 347.0 399.9 372.6 410.3 300.7 352.2 314.1 379.1

337.2 552.2 335.2 495.5 339.9 478.3 389.4 538.6 268.8 340.4 279.5 373.5 272.5 405.9 283.5 397.8

1230.1 1240.5 1321.4 1305.0 1231.7 1358.0 1159.6 1272.2 1028.1 1025.0 1149.7 1131.4 1128.3 1231.9 1036.9 1153.9

341.1 604.3 391.1 495.5 366.6 507.5 523.4 616.6 280.2 384.9 377.7 415.8 363.9 441.4 478.9 537.3

(3)

The calculated RSEs are compared with experimental values33 in Table 5. It is evident from Table 5 that the agreement between theory and experiment is good: the deviation between the two sets of values does not exceed 6 kJ mol−1.

Table 4. Calculated Homolytic and Heterolytic BDEs of CH3−YH and •CH2−YH and Their Deprotonated Formsa C−Y

(2)

Table 5. Comparison of Calculated and Experimental Radical Stabilization Energies for •CH2YH Radicals and Their Deprotonated Forms, •CH2Y−a RSE(•CH2YH)

a b

YH

W1w

BH2 CH3 NH2 OH AlH2 SiH3 PH2 SH

41.9 15.1 49.9 36.1 28.4 13.3 27.0 40.5

experiment

RSE(•CH2Y−) b

18.8 ± 1.7 46.4 ± 8.8 37.4 ± 1.0

46.4 ± 8.8

W1w 242.6 135.9 143.2 124.7 91.7 69.6 109.0 89.9

Enthalpies for reactions 2 and 3, respectively, in kJ mol−1 at 298 K. Experimental values from ref 33.

In kJ mol−1 at 298 K.

The factors affecting the stabilization of •CH2YH radicals have been outlined in a number of previous works.24b,27,32 To summarize, the •CH2YH radicals with π-donor substituents are stabilized by the n(Y) → p(C•) interaction, the radicals with πacceptor substituents are stabilized by delocalization of the unpaired electron to YH, and the interaction between p(C•) and π(YH) confers stability to radicals with hyperconjugative donor substituents. The RSEs for the deprotonated species are all strongly positive, indicating that the stabilization afforded by the substituent is greater in •CH2Y− than in CH3Y−. The greatly enhanced stability may be attributed to the additional πcharacter present in the C−Y bond of •CH2Y−. The largest RSE is obtained for Y = BH, with RSE(•CH2BH−) being greater than all the other RSEs by ∼100 kJ mol−1 or more. The particularly large value for this RSE arises because •CH2BH− has a full C−Y double bond, as opposed to the partial double bonds present with group 4, 5, and 6 substituents. In addition, Y = BH has the smallest EA of all substituents, thereby reducing the contribution of the favorable bonded resonance structure 2a in CH3BH−. Because of a more efficient overlap between the

Table 4 is the substantial reduction in the heterolytic BDEs following deprotonation, as found previously by Boyd et al.12 This reduction can be attributed to the absence of unfavorable charge separation in the dissociation products of CH3Y− and • CH2Y− Nevertheless, there are no cases in which the heterolytic BDE drops below the homolytic BDE. For the closed-shell species, the drop follows the trend in the electron affinities of Y (Table 2): as the EA of Y decreases, the contribution from the no-bond resonance structure 2b increases, making heterolytic cleavage easier.. Deprotonation also generally leads to a decrease in the homolytic BDEs of CH 3 YH. This behavior has been rationalized by Boyd et al.12 in terms of substituent electronegativity. That is, homolytic cleavage of the C−Y bond is easier in CH3Y− than in CH3YH because the electronegativity difference between CH3 and Y− is smaller than that between CH3 and YH. The •CH2Y− radical anions with first-row substituents display larger C−Y homolytic BDEs than •CH2YH. For the most part, this is due to the stronger interaction between the D

dx.doi.org/10.1021/jp3101927 | J. Phys. Chem. A XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry A

Article

p(C•) and p(Y) orbitals, RSEs with first-row substituents are larger than those with second-row substituents. For the sake of completeness, the C−H bond dissociation enthalpies for CH3YH and CH3Y−, from which the stabilization energies are calculated, are listed in Table 6. For all the substituents, the C−H BDEs are smaller for the deprotonated species than for the neutral species.

deprotonation than CH3YH is. The relative stabilization of the open-shell anions can be attributed to the increased electron donation from the heteroatom to the electron-deficient carbon, as well as to the increased π-character of the C−Y bond for YH = BH2 or AlH2, when •CH2YH is deprotonated. Both of these interactions are more stabilizing than the two-electron donation from n(Y−) to π*(CH3) that occurs when CH3YH is deprotonated. Resonance arguments also support the observation of larger acidities for CH3YH. Whereas all the contributing resonance structures for the •CH2Y− radical anion are bonded, CH3Y− has a no-bond structure 2b, which acts to destabilize the molecule. Thermochemical Cycle. The acidities and bond dissociation enthalpies are linked in a thermochemical cycle (Scheme 1), which indicates that:

Table 6. Comparison of Calculated and Experimental C−H Bond Dissociation Enthalpies in CH3YH and CH3Y−a BDE(H−CH2YH)

a

YH

W1w

H BH2 CH3 NH2 OH AlH2 SiH3 PH2 SH

439.3 397.4 424.2 389.4 403.2 410.9 426.0 412.3 398.8

experimentb 439.3 ± 0.4 420.5 ± 1.3 392.9 ± 8.4 401.92 ± 0.63

392.9 ± 8.4

BDE(H−CH2Y−) W1w 407.8 196.7 303.4 296.2 314.6 347.7 369.8 330.4 349.4

Δacid H °(CH3YH) + BDE(CH3Y −) = BDE(CH3YH) + Δacid H °( •CH 2YH)

Scheme 1. Thermochemical Cycle Relating CH3YH, • CH2YH, CH3Y−, and •CH2Y−

In kJ mol−1 at 298 K. bExperimental values from ref 33.

Gas-Phase Acidities of CH3YH and •CH2YH. The calculated and experimental 1 acidities of CH 3 YH and • CH2YH are shown in Table 7. In general, the W1w acidities match the experimental acidities very well, with only one calculated value (CH3SiH3) lying outside the limits of experimental uncertainty. It can be seen from Table 7 that the CH3YH molecules are ∼50−200 kJ mol−1 less acidic in the gas phase than the corresponding •CH2YH radicals, as reflected in the larger magnitudes of the calculated acidities. This behavior indicates that •CH2YH is stabilized to a greater extent following

This shows immediately that the difference between the acidities of CH3YH and •CH2YH is equal to the difference between the BDEs of CH3YH and CH3Y−: Δacid H °(CH3YH) − Δacid H °( •CH 2YH) = BDE(CH3YH) − BDE(CH3Y −)

Table 7. Comparison of Calculated and Experimental GasPhase Acidities of Substituted Methanes and Substituted Methyl Radicalsa,b CH4 • CH3 CH3BH2 • CH2BH2 CH3CH3 • CH2CH3 CH3NH2 • CH2NH2 CH3OH • CH2OH CH3AlH2 • CH2AlH2 CH3SiH3 • CH2SiH3 CH3PH2 • CH2PH2 CH3SH • CH2SH

W1w

experimentc

1746.3 1714.8 1745.3 1544.5 1758.0 1637.1 1686.8 1593.6 1601.0 1512.4 1574.4 1511.1 1592.2 1535.9 1566.2 1484.3 1496.5 1447.1

1743.6 ± 2.9 1711.7 ± 1.7

(4)

(5)

Heats of Formation. The W1w heats of formation for CH3YH, CH3Y−, •CH2YH, and •CH2Y− are compared with experimental values1,33 in Table 8. The agreement between the theoretical and experimental heats of formation is good, with all differences being smaller than 6 kJ mol−1. The largest discrepancy (5.6 kJ mol−1) occurs for CH3SH, and we have again carried out higher-level calculations, this time at the W4lite level,19 to see whether it is theory or experiment that is more likely to be at fault. The calculated W4-lite heat of formation (−27.6 kJ mol−1) is, in fact, quite close to the W1w value (−28.4 kJ mol−1) and suggests that the experimental heat of formation for CH3SH (−22.8 kJ mol−1) may be slightly too small in absolute magnitude. This would also account for the small discrepancies between theory and experiment in the derived RSE (Table 5) and BDE (Table 6) values.

1758.0 ± 8.4 1687.0 ± 3.4 1597.0 ± 6.0

■

1579.0 ± 8.8

CONCLUDING REMARKS High-level ab initio quantum chemical calculations have been used to examine the heteroatomic deprotonation of substituted methanes and methyl radicals. The following important points emerge from the present work: 1. Deprotonation of •CH2YH results in a contraction of the C−Y bond. For π-donor (NH2, OH, PH2, and SH) and hyperconjugative donor (CH3 and SiH3) substituents,

1496.0 ± 8.4

a In kJ mol−1 at 298 K. bValues correspond to deprotonation at the heteroatom. cExperimental values from ref 1.

E

dx.doi.org/10.1021/jp3101927 | J. Phys. Chem. A XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry A

Article

Table 8. Comparison of Calculated and Experimental Heats of Formationa W1w CH3BH2 CH3BH− • CH2BH2 • CH2BH− CH3CH3 CH3CH2− • CH2CH3 • CH2CH2− CH3NH2 CH3NH− • CH2NH2 • CH2NH− CH3OH CH3O− • CH2OH • CH2O− CH3AlH2 CH3AlH− • CH2AlH2 • CH2AlH− CH3SiH3 CH3SiH2− • CH2SiH3 • CH2SiH2− CH3PH2 CH3PH− • CH2PH2 • CH2PH− CH3SH CH3S− • CH2SH • CH2S−

31.6 245.9 210.8 224.4 −88.2 138.8 117.8 224.0 −24.1 131.8 147.2 209.8 −205.2 −135.2 −20.2 −38.8 61.3 104.7 253.9 234.1 −36.0 25.3 171.9 176.8 −22.1 13.1 172.0 125.3 −28.4 −62.8 152.2 68.4

experiment

■

−84.0 ± 0.4b

Optimized geometries of relevant species (Table S1), ZPVEs, thermal corrections to enthalpies, and W1w single-point energies (Table S2), and full citations for references 13 and 14 (Table S3). This material is available free of charge via the Internet at http://pubs.acs.org.

−23.5 ± 0.5b

■

151.9 ± 8.4c −205 ± 10b

4.

5.

AUTHOR INFORMATION

Corresponding Author

*E-mails: [email protected] (B.C.), radom@chem. usyd.edu.au (L.R.).

−17.0 ± 0.7c

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We gratefully acknowledge the receipt of an Australian Postgraduate Award (to M.M.), funding (to L.R.) from the Australian Research Council (ARC), and generous grants of computer time from the National Computational Infrastructure (NCI) National Facility and Intersect Australia Ltd.

■

REFERENCES

(1) NIST Chemistry WebBook; Linstrom, P. J.; Mallard, W. G., Eds.; NIST Standard Reference Database Number 69; National Institute of Standards and Technology: Gaithersburg, MD, October 2012 (http:// webbook.nist.gov). (2) Berkowitz, J.; Ellison, G. B.; Gutman, D. J. Phys. Chem. 1994, 98, 2744−2765. (3) Ervin, K. M. Chem. Rev. 2001, 101, 391−444. (4) See, for example: (a) Smith, B. J.; Radom, L. Chem. Phys. Lett. 1995, 245, 123−128. (b) Topol, I. G.; Tawa, G. J.; Caldwell, R. A.; Eissenstat, M. A.; Burt, S. K. J. Phys. Chem. A 2000, 104, 9619−9624. (c) Silva, C. O.; Nascimento, M. A. C. Adv. Chem. Phys. 2002, 123, 423−468. (d) Ho, J.; Coote, M. L. WIREs Comput. Mol. Sci. 2011, 1, 649−660. (5) (a) Martin, J. M. L.; de Oliveira, G. J. Chem. Phys. 1999, 111, 1843−1856. (b) Martin, J. M. L.; Parthiban, S. W1 and W2 Theories, and Their Variants: Thermochemistry in the kJ/mol Accuracy Range. In Quantum-Mechanical Prediction of Thermochemical Data; Cioslowski, J., Ed.; Kluwer: Dordrecht, 2001; pp 31−65. (6) Pickard, F. C., IV; Griffith, D. R.; Ferrara, S. J.; Liptak, M. D.; Kirschner, K. N.; Shields, G. C. Int. J. Quantum Chem. 2006, 106, 3122−3128. (7) Note that a smaller value of ΔacidH° corresponds to a more acidic substance. (8) The ionization energies of most of the •CH2Y− radical anions considered are close to zero. (9) Boese, A. D.; Oren, M.; Atasoylu, O.; Martin, J. M. L.; Kallay, M.; Gauss, J. J. Chem. Phys. 2004, 120, 4129−4141. (10) The W1w procedure9 is a slightly modified version of the W1 protocol.5 It makes use of the smaller but somewhat more balanced aug-cc-pV(n+d)Z basis sets instead of the aug-cc-pVnZ+2d1f basis sets used in the W1 procedure. (11) Mayer, P. M.; Radom, L. J. Phys. Chem. A 1998, 102, 4918− 4924. (12) (a) Boyd, S. L.; Boyd, R. J.; Bessonette, P. W.; Kerdraon, D. I.; Aucoin, N. T. J. Am. Chem. Soc. 1995, 117, 8816−8822. (b) Boyd, S.

−22.8 ± 0.6b 151.9 ± 8.4c

In kJ mol−1 at 298 K. bExperimental values from ref 1. cExperimental values from ref 33.

3.

ASSOCIATED CONTENT

S Supporting Information *

118.8 ± 1.3c

a

2.

reflects the greater stabilization of the •CH2Y− radical anions. 6. Examination of a thermochemical cycle shows that the difference in acidities between CH3YH and •CH2YH is equal to the difference in the C−H BDEs between CH3YH and CH3Y−.

this is due to the greater π(Y) → p(C•) electrondonation present with the radical anion. For π-acceptor (BH2 and AlH2) substituents, the C−Y bond contracts because of the formation of a conventional C−Y double bond in •CH2Y−, in contrast to the partial double bond that is present in •CH2YH. Deprotonation of CH3YH leads to a shortening of the C−Y bond when YH = NH2, or OH and a lengthening otherwise. Negative hyperconjugation and an increased contribution of the no-bond resonance structure 2b are the dominant factors responsible for the changes. The C−Y heterolytic BDEs of both CH3YH and • CH2YH drop sharply following deprotonation as a consequence of there no longer being charge separation in the dissociation products of CH3Y− and •CH2Y−. The related homolytic BDEs decrease upon deprotonation for CH3YH but increase for •CH2YH. The •CH2Y− radical anions are all stabilized relative to CH3Y−. This occurs because of the additional π-character present in the C−Y bond of •CH2Y−. The gas-phase acidities of CH3YH are 50−200 kJ mol−1 larger than the corresponding values for •CH2YH. This F

dx.doi.org/10.1021/jp3101927 | J. Phys. Chem. A XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry A

Article

L.; Boyd, R. J. J. Am. Chem. Soc. 1997, 119, 4214−4219. (c) Boyd, S. L.; Boyd, R. J. J. Phys. Chem. A 1999, 103, 7087−7093. (13) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A. et al. Gaussian 09, Revision A.02; Gaussian, Inc.: Wallingford, CT, 2009. (14) Werner, H.-J.; Knowles, P. J.; Manby, F. R.; Schütz, M.; Celani, P.; Knizia, G.; Korona, T.; Lindh, R.; Mitrushenkov, A.; Rauhut, G. et al. MOLPRO, 2010.1; University College Cardiff Consultants Limited: Cardiff, UK, 2010. (15) Merrick, J. P.; Moran, D.; Radom, L. J. Phys. Chem. A 2007, 111, 11683−11700. (16) Nicolaides, A.; Rauk, A.; Glukhovtsev, M. N.; Radom, L. J. Phys. Chem. 1996, 100, 17460−17464. (17) Curtiss, L. A.; Raghavachari, K.; Redfern, P. C.; Pople, J. A. J. Chem. Phys. 1997, 106, 1063−1079. (18) Dunlavey, S. J.; Dyke, J. M.; Jonathan, N.; Morris, A. Mol. Phys. 1980, 39, 1121−1135. (19) Karton, A.; Rabinovich, E.; Martin, J. M. L.; Ruscic, B. J. Chem. Phys. 2006, 125, 144108/1−17. (20) Lower level CBS-Q calculations also produced similar values for IE(BH2): Jursic, B. S. J. Mol. Struct.: THEOCHEM 2000, 505, 67−80. (21) Fehlner, T. P.; Koski, W. S. J. Am. Chem. Soc. 1964, 86, 2733− 2734. (22) Margrave, J. L. J. Phys. Chem. 1957, 61, 38−40. (23) Wilson, J. H.; McGee, H. A., Jr. J. Chem. Phys. 1967, 46, 1444− 1452. (24) (a) Radom, L. Prog. Theor. Org. Chem. 1981, 3, 1−64. (b) Mayer, P. M.; Glukhovtsev, M. N.; Gauld, J. W.; Radom, L. J. Am. Chem. Soc. 1997, 119, 12889−12895. (25) For a discussion of negative hyperconjugation, see, for example: (a) Schleyer, P. v. R.; Kos, A. J. Tetrahedron 1983, 39, 1141−1150. (b) Tostes, J. R.; Seidl, P. R.; Taft, C. A.; Lie, S. K.; Carneiro, J. W.; de, M.; Brown, W.; Lester, W. A., Jr. J. Mol. Struct.: THEOCHEM 1996, 388, 85−95. (c) Allinger, N. L. Molecular Structure; Wiley: Hoboken, 2010. (26) (a) Glidewell, C.; Thomson, C. J. Comput. Chem. 1982, 3, 495− 506. (b) Hopkinson, A. C.; Lien, M. H. J. Mol. Struct.: THEOCHEM 1983, 92, 153−166. (c) Williams, I. H. J. Mol. Struct.: THEOCHEM 1983, 105, 105−117. (d) Masamura, M.; Ikuta, S. J. Comput. Chem. 1999, 20, 1138−1144. (e) Walrond, L. D.; Mayer, P. M. J. Mol. Struct.: THEOCHEM 2007, 811, 303−312. (27) Henry, D. J.; Parkinson, C. J.; Mayer, P. M.; Radom, L. J. Phys. Chem. A 2001, 105, 6750−6756. (28) Graner, G.; Hirota, E.; Iijima, T.; Kuchitsu, K.; Ramsay, D. A.; Vogt, J.; Vogt, N. Landolt-Börnstein − Group II Molecules and Radicals: Numerical Data and Functional Relationships in Science and Technology; Springer-Verlag: Heidelberg, 1999. (29) The observation that the C−Y bond lengths of •CH2AlH2 and • CH2SiH3 decrease upon deprotonation cannot be attributed to a lower heterolytic BDE, as is the case for systems examined in refs 12 and 24b, since homolysis is the minimum energy dissociation process. (30) For recent reviews, see: (a) Zipse, H. Top. Curr. Chem. 2006, 263, 163−189. (b) Coote, M. L.; Lin, C. Y.; Zipse, H. The Stability of Carbon Centered Radicals. In Carbon-Centered Free Radicals and Radical Cations: Structure, Dynamics and Reactivity; Forbes, M. D. E., Ed.; Wiley: Hoboken, 2010; pp 83−104. (c) Hioe, J.; Zipse, H. Radical Stability − Thermochemical Aspects. In Encyclopedia of Radicals in Chemistry, Biology and Materials; Chatgilialoglu, C., Studer, A., Eds.; Wiley: Chichester, 2012; pp 449−476. (31) Coote, M. L.; Lin, C. Y.; Beckwith, A. L. J.; Zavitsas, A. A. Phys. Chem. Chem. Phys. 2010, 12, 9597−9610. (32) Menon, A.; Henry, D. J.; Bally, T.; Radom, L. Org. Biomol. Chem. 2011, 9, 3636−3657 and references therein. (33) Luo, Y.-R. Comprehensive Handbook of Chemical Bond Energies; CRC Press: Boca Raton, 2007.

G

dx.doi.org/10.1021/jp3101927 | J. Phys. Chem. A XXXX, XXX, XXX−XXX