5070 J. Am. Chem. Soc., Vol. 118, No. 21, 1996
Lee et al.
dideuterated quadricyclane (8). This reaction proceeds by 63%/
25.0 cal/mol K ) 2.5 cal/mol K, we obtain ∆rxnH ) -0.9 kcal/
mol for reaction 7. ∆fH(C7H7-) is determined to be 117.0 kcal/
mol using ∆fH(NH2-) ) 26.9 kcal/mol and ∆fH(NH3) ) -11.0
kcal/mol. In these estimations, ∆fH(C7H8) is assumed to be 80
kcal/mol; however, this value is not needed in determining the
upper limit of the EA, since it cancels out in the calculation.
Hence EA(C7H7) is less than 21.6 kcal/mol (0.937 eV). Another
way to estimate the EA is from bracketing reactions (Table 4b);
electron transfer from quadricyclanide ion to SO2 (EA ) 1.107
eV)17,18 is fast; however, no electron transfer was observed to
O2 (EA ) 0.451 eV).19 These reactions indicate that the EA
of quadricyclyl radical is greater than 0.451 eV and less than
1.107 eV. Electron transfer to CS2 (EA ) 0.51 eV)20 was not
observed but the presence of other rapid reaction channels
precludes any firm conclusions from this system. These
estimates were helpful in the analysis of spectra in a concurrent
photoelectron spectroscopy (PES) study. From the PES study,11
two different EAs for quadricyclane are deduced, 0.868(6) and
0.962(6) eV. These two values probably correspond to two
different isomers of structure 5 and 6.
To estimate the relative energies of the quadricyclanide ions
(5) and (6), we have performed ab initio calculations: MP3
(frozen core)/6-31++G** single point energy with fully
optimized geometries and vibrational frequencies at the RHF/
6-31++G** level (Table 2). The HF energy, MP3 energy, and
zero point energy of quadricyclanide ion are -268.96181,
-269.97073, and 0.12109 Hartrees for 5 and -268.96156,
-269.97084, and 0.12074 Hartrees for 6, respectively. The
energy of 5 is greater than that of 6 by less than 0.3 kcal/mol
including zero point energies; thus the stability of these two
isomers is almost identical. The calculations show that structure
7 is a transition state.
For the electron affinities, calculations of the two quadricyclyl
radical structures have been performed: MP3 (frozen core)/6-
31++G** single point energy with fully optimized geometries
and vibrational frequencies at the UHF/6-31++G** level. The
HF energy, MP3 energy, and zero point energy of quadricyclyl
radicals are -268.98626, -269.95620, and 0.12422 Hartrees
for 5 and -268.98624, -269.95390, and 0.12412 Hartrees for
6, respectively. The calculated electron binding energies (eBE)
are eBE(5) ) 0.40 eV, eBE(6) ) 0.46 eV without zero point
energies, and eBE(5) ) 0.48 eV, eBE(6) ) 0.54 eV with 90%
zero point energy corrections. The 0.06 eV energy difference
is almost entirely due to higher correlation energy of the
quadricyclyl radical 5 compared to that of the radical 6. These
calculated eBEs are much lower than the experimental values.
Another approach for evaluating eBEs is Koopmans' theorem
which states that the electron binding energy of a closed-shell
molecule is equal to the negative of the highest occupied orbital
energy. Using this theorem, we obtain 1.61 and 1.74 eV for
the eBEs of anions 5 and 6, respectively. In this case, the eBEs
are considerably higher than the experimental values, but the
order is the same as from the previous estimation using the total
energy. These estimates suggest that radical 6 has a higher EA
than radical 5.
37% loss of H+/D+; these values are close to the statistical ratios
(67%/33% loss of H+/D+) in a system containing four hydrogens
and two deuteriums of comparable acidity. It should be
mentioned that, for norbornadienide ion reacting with ND3, three
H/D exchanges have been observed.15
To estimate the entropy of acidity, we employed ab initio
calculations. The geometries and energies of quadricyclane (1)
and quadricyclanide ion (5, 6) were calculated using Gaussian-
9216 with MP3 (frozen core)/6-31++G**//RHF/6-31++G**
basis sets. The rotational constants and vibrational frequencies
(uncorrected) were used to calculate the entropy change of
deprotonation. The entropies of 1, 5, and 6 are 67.9, 69.4, and
69.3 cal/mol K, respectively, at 300 K (Table 2). Using the
entropy of the proton, 26.0 cal/mol K, we obtain ∆acidS (5) )
27.5 cal/mol K and ∆acidS (6) ) 27.4 cal/mol K (eq 3).
-
∆
acidS ) S(C7H7 ) + S(H+) - S(C7H8)
(3)
The enthalpy change at 300 K was then obtained using eq 4.
∆
acidH ) ∆acidG + T∆acidS
(4)
The acidity at 0 K (∆acidH0) was calculated using eq 5.
T
-
∆
acidH0 ) ∆acidH -
{C (C H ) + C (H+) -
∫
T
p
7
7
p
0
Cp(C7H8)} dT (5)
where Cp is the heat capacity. The heat capacities can be
expressed as a function of temperature using rotational constants
and vibrational frequencies obtained from ab initio calcula-
tions.14 The integral term is 1.6 kcal/mol for the quadricyclanide
ion 5. The C-H bond dissociation energy (BDE) can be
calculated using eq 6, once EA(A) is known
BDE(A-H) ) ∆acidH(AH) - IP(H) + EA(A)
(6)
where IP is the ionization energy of hydrogen atom (313.6 kcal/
mol).
To estimate the upper limit of the EA of the C7H7 radical,
the energetics of several reactions were compared. When O-
ion (heat of formation ∆fH ) 25.8 kcal/mol) reacts with
quadricyclane, the C7H6- ion as well as HO- ion (∆fH ) -32.8
kcal/mol) are formed. Exothermic formation of the hydroxide
ion and quadricyclyl radical requires that ∆fH(C7H7) be less
than 138.6 kcal/mol (Table 4a). To estimate the heat of
formation of C7H7- ion, consider the enthalpy change of reaction
7.
Since sites C1 and C2 in quadricyclane have similar acidity,
but the electron binding energies of the resulting anions 5 and
6 differ by almost 0.1 eV, the C-H bond dissociation energies
at C1 and C2 must differ by this amount. Table 3 summarizes
the acidities, ∆acidG, ∆acidS, and ∆acidH, and C-H bond
NH2- + C7H8 a C7H7- + NH3
(7)
From the free energy change ∆rxnG ) -1.7 kcal/mol and the
entropy change ∆rxnS ) ∆acidS (C7H8) - ∆acidS (NH3) ) 27.5-
(15) Lee, R. E.; Squires, R. R. J. Am. Chem. Soc. 1986, 108, 5078.
(16) Frisch, M. J.; Tucker, G. W.; Head-Gordon, M.; Gill, P. M. W.;
Wong, M. W.; Foresman, J. B.; Johnson, B. G.; Schlegel, H. B.; Robb, M.
A.; Replogle, E. S.; Gomperts, R.; Andres, J. L.; Raghavachari, K.; Binkley,
J. S.; Gonzalez, C.; Martin, R. L.; Fox, D. J.; DeFrees, D. J.; Baker, J.;
Stewart, J. J. P.; Pople, J. A. GAUSSIAN 92, Revision C; Gaussian Inc.:
Pittsburgh, PA, 1992.
(17) Nimlos, M. R.; Ellison, G. B. J. Phys. Chem. 1986, 90, 2574.
(18) Grabowski, J. J.; Van Doren, J. M.; DePuy, C. H.; Bierbaum, V.
M. J. Chem. Phys. 1984, 80, 575.
(19) Travers, M. J.; Cowles, D.; Ellison, G. B. Chem. Phys. Lett. 1989,
164, 449.
(20) Kebarle, P.; Chowdhury, S. Chem. ReV. 1987, 87, 513.