Reversible O -ylide formation in carbene/ether reactions

Article abstract of DOI:10.1021/jp2098119

p-Nitrophenylchlorocarbene reacted reversibly with diethyl ether, di-n-propyl ether, or tetrahydrofuran (THF) to form O-ylides, which were visualized by their UV-visible spectroscopic signatures. Equilibrium constants (K_eq) were determined spectroscopically and ranged from 0.10 M ^-1 (di-n-propyl ether) to 7.5 M^-1 (THF) at 295 K. Studies of K_eq as a function of temperature afforded δH^o, δS^o, and δG^o for the di-n-propyl ether and THF/O-ylide equilibria. δH^o was favorable for ylide formation, but δS^o was quite negative, so that δG^os for the equilibria were small. Electronic structure calculations based on density functional theory provided structures, spectroscopic signatures, and energetics for the carbene/ether O-ylides.

Full text of DOI:10.1021/jp2098119

ARTICLE
pubs.acs.org/JPCA
Reversible O-Ylide Formation in Carbene/Ether Reactions
Pablo A. Hoijemberg, Robert A. Moss,* and Karsten Krogh-Jespersen*
Department of Chemistry and Chemical Biology, Rutgers, The State University of New Jersey, New Brunswick, New Jersey 08903,
United States
S Supporting Information
b
ABSTRACT: p-Nitrophenylchlorocarbene reacted reversibly with diethyl
ether, di-n-propyl ether, or tetrahydrofuran (THF) to form O-ylides, which
were visualized by their UVÀvisible spectroscopic signatures. Equilibrium
constants (Keq) were determined spectroscopically and ranged from 0.10 MÀ1
(di-n-propyl ether) to 7.5 MÀ1 (THF) at 295 K. Studies of Keq as a function of
temperature afforded ΔHo, ΔSo, and ΔGo for the di-n-propyl ether and THF/O-ylide equilibria. ΔHo was favorable for ylide
formation, but ΔSo was quite negative, so that ΔGos for the equilibria were small. Electronic structure calculations based on
density functional theory provided structures, spectroscopic signatures, and energetics for the carbene/ether O-ylides.
evidence7 for equilibration between carbene 2, dioxane, and a
I. INTRODUCTION
derived O-ylide.8 Here we provide direct spectroscopic evidence
We recently used laser flash photolysis (LFP) combined with
and computational support for reversible O-ylide formation
UVÀvisible spectroscopy to demonstrate equilibration between
between PNPCC and THF, diethyl ether, or di-n-propyl ether.
π-complexes of carbenes and 1,3,5-trimethoxybenzene (TMB)
and the individual reactants. Thus, phenylchlorocarbene and
TMB equilibrate with a π-complex with K = 1250 MÀ1 at 294 K
2. EXPERIMENTAL DETAILS AND COMPUTATIONAL
STUDIES
and ΔHo = À7.1 kcal/mol.1 An analogous equilibrium operates
between pentafluorophenylchlorocarbene, TMB, and the de-
2.1. Experimental Details. The preparation and purification
rived π-complex, with K = 3.2 Â 105 MÀ1 at 294 K and
of p-nitrophenylchlorodiazirine (3) has been fully described.4
ΔHo = À10.2 kcal/mol.2 A Hammett analysis of a family of
Solvents and ethers were commercial materials, used as received:
equilibria between para-substituted phenylchlorocarbenes,
pentane (>99%, anhydrous), heptane (99%, distilled), diethyl
ether (anhydrous, certified ACS), di-n-propyl ether (99%, dis-
tilled), and THF (99%, distilled).
TMB, and their π-complexes afforded F = 2.48, indicating that
complexation is favored by electron-withdrawing, destabilizing
substituents on the arylhalocarbene.3
Reversible π-complexation of arylhalocarbenes with very
electron-rich substrates like TMB is thus reasonably general.
However, with less electron-rich partners, such as anisole,
O-ylide formation occurs competitively with π-complexation.
For example, p-nitrophenylchlorocarbene (1, PNPCC) forms
both O-ylidic and π-type complexes with anisole, although the
latter are more stable and persistent.4 On the other hand,
PNPCC and simple ethers such as diethyl ether and 18-crown-6
form spectroscopically well-defined O-ylides.4 PNPCC also forms
carbonyl ylides with acetone and benzaldehyde5 and an O-ylide
with THF.6
LFP experiments employed a XeF2 excimer laser emitting
42À56 ns light pulses (full width at half-maximum) at 351 nm
with 55À65 mJ power.9 For these experiments, diazirine 3 was
dissolved in pentane or heptane and the selected ether was added
to the desired concentration. The final absorbance of 3 was
0.2À0.5 at 360 nm.
2.2. Computational Methods. Electronic structure calcula-
tions were carried out using density functional theory (DFT)10
methodologies implemented in the Gaussian 09 suite of pro-
grams.11 Ground state geometry optimizations of carbene and
solvent monomers and dimeric carbene-solvent complexes
Is O-ylide formation between PNPCC and simple ethers
reversible? Does a demonstrable equilibrium exist between
PNPCC, an alkyl ether, and its derived O-ylide? In an extensive
study of halocarbene amides, Platz et al. presented indirect
Received: October 12, 2011
Revised:
December 7, 2011
Published: December 16, 2011
r
2011 American Chemical Society
358
dx.doi.org/10.1021/jp2098119 J. Phys. Chem. A 2012, 116, 358–363
|

The Journal of Physical Chemistry A
ARTICLE
Figure 1. Calibrated UVÀvis spectrum of PNPCC in heptane (circles)
and in 0.049 M THF in heptane (triangles),20,21 100 ns after the laser
pulse. PNPCC absorptions at 308 and 628 nm; the PNPCC/THF
O-ylide absorption is at 452 nm.
Figure 2. Time dependence of the decay of absorbance for (a) PNPCC
in heptane at 316 nm24 in the absence of THF (green), (b) PNPCC
in heptane at 316 nm24 in the presence of 0.049 M THF (blue), and
(c) ylide 4 (red) at 452 nm. The black lines superimposed on the experi-
mental traces correspond to the exponential fits.
(O-ylides) were carried out using the dispersion-corrected B97D,12
M06,13 M06-2X,13 or wB97XD14 exchange-correlation func-
tionals and 6-311+G(d)15 basis sets (B97D/6-311+G(d), etc.).
General solvent effects were incorporated via the polarizable
conductor self-consistent reaction field model (CPCM)16 with
the default parameters for n-pentane or heptane provided in
Gaussian 09. Geometry optimizations of both idealized gas and
simulated solution phase structures were conducted with tight
convergence criteria (opt=tight) and enhanced numerical inte-
gration grid sizes (integral(grid=ultrafine)). Vibrational frequen-
cies (unscaled), evaluated at the optimized structures with
enhanced integration grids, formed the basis for the calculation
of zero-point energy (ZPE) contributions. Standard thermody-
namic corrections (based on harmonic oscillator/rigid rotor
approximations and ideal gas behavior) were then applied to
the computed energies to derive (standard) enthalpies (H; T =
298.15 K) and free energies (G; T = 298.15 K, P = 1 atm).17 We
equate the raw differences in free energies obtained from the
CPCM calculations with reaction enthalpies and then apply ZPE
and thermal entropic corrections. The reaction free energies
presented in Table 3 correspond to a reference state of 1 M
concentration for each species participating in the reaction and
T = 298.15 K.
Table 1. Kinetics of Decay of PNPCC and O-Ylides
PNPCC,
PNPCC
O-ylidea
b
b
b
no ether kdecay
ether
(conc., M)
kdecay
kdecay
0.48 (0.02)
THFc
Et2Od
n-Pr2Oc
(0.049)
(2.33)
(2.05)
0.84 (0.06)
0.71 (0.06)
0.79 (0.04)
1.0 (0.1)
0.9 (0.1)
0.92 (0.07)
a Kinetics are in the presence of added ether. b All rate constants are in
units of 107 sÀ1. Errors are shown in parentheses. c Solvent is heptane.
d Solvent is pentane.
Excited state calculations (transition wavelengths (λ) and
oscillator strengths (f)) at optimized B97D/6-311+G(d) ground
state geometries utilized the time-dependent DFT formalism18
and the hybrid B3LYP functionals (TD-B3LYP/6-311+G(d)//
B97D/6-311+G(d)).19 Assignment of a particular electronic
transition (π f p, σ f p, or π f π*) was based on inspection
of the largest transition amplitudes for the excitation and by
visualization of the contributing MOs.
Figure 3. A316/A460 absorbance ratios for PNPCC and O-ylide 4 vs
time (ns) after the laser pulse. The THF concentrations in heptane are
0.07 M (4), 0.14 M (0), 0.28 M (]), and 1.06 M (Â).26
strong absorption at 456 nm for 4 in simulated heptane; cf.
Table 3, below.
3. RESULTS AND DISCUSSION
3.1. PNPCC and THF. In the absence of added ethers, LFP of
diazirine 3 in heptane generates PNPCC (1). Its calibrated20
UVÀvis spectrum (Figure 1) displays a dominant π f p
absorption at 308 nm and a weaker σ f p absorption at
628 nm.21 The computed π f p absorbance of PNPCC is at
309 nm (f = oscillator strength = 0.5784), whereas the energy of
the carbenic center localized σ f p transition is significantly
underestimated, λcalc = 897 nm (f = 0.0013).22
With 0.049 M added THF, there is a decrease in the carbene
absorbance at 308 nm, coupled with the appearance of a signal
assigned to PNPCC/THF O-ylide 4 at 452 nm; in pure THF, the
ylide absorption is red-shifted to ∼490 nm.23 We calculate a
Is there an equilibrium established between PNPCC, THF,
and ylide 4? We note that the decay of the carbene and ylide
signals conforms to similar kinetics; cf. Figure 2.24 The rate
constants derived from exponential fitting of the decays in
Figure 2 are collected in Table 1. The rise of the absorptions in
each trace (with apparent rate constants of ∼1.5 Â 107 sÀ1) are a
359
dx.doi.org/10.1021/jp2098119 |J. Phys. Chem. A 2012, 116, 358–363

The Journal of Physical Chemistry A
ARTICLE
function of the laser pulse width; they do not represent rate
constants for formation of the carbene or ylide.25
In the THF experiments, the rate constants for the decay of
PNPCC and its THF O-ylide (4) are comparable. The kinetic
linkages between kdecay for PNPCC and ylide 4 are consistent
with equilibration of these species. Analogous linkages are observed
for the O-ylides formed from PNPCC and diethyl ether (5) or di-n-
propyl ether (6); cf. Table 1.
To determine Keq for the PNPCC/THF system, we require
the relative intensities of the carbene absorption at 316 nm24 and
of ylide 4. The latter absorbs at 490 nm in pure THF, but its
maximum is found at 452À468 nm at the THF concentrations of
0.070À1.06 M used in our determination of Keq. Therefore, we
measured the ylide absorbance at 460 nm, the average wave-
length of the absorption maxima in the considered THF con-
centration range.
As illustrated in Figure 3, A316/A460, the PNPCC/ylide 4
absorbance ratio is relatively constant at several THF concentra-
tions between about 80 and 160 ns after LFP generation of the
carbene. During this time interval, the carbene and its THF ylide
are at their maximum concentrations and in relatively stable
equilibrium.26
Figure 4. Average values of A316/A460 vs 1/[THF] (MÀ1) for PNPCC
and O-ylide 4 in heptane at 295 K, 80À160 ns after LFP of diazirine 3.
The slope of the correlation line is 0.135 ( 0.005 M, r = 0.998. See Table
S-1 in the Supporting Information for the numerical values of the
experimental points.
We plotted calibrated20 average A316/A460 absorbance ratios
in the 80À160 ns time interval vs the inverse of the THF
concentrations between 0.07 and 1.06 M, and we obtained the
linear correlation shown in Figure 4. To determine the equilib-
rium constant, Keq, we use computed oscillator strengths (f) in
place of the experimentally unknown extinction coefficients (ε) of
PNPCC and 4. Thus, we calculate Keq from the following rela-
tion: Keq = (εcarbene/εylide)(1/slope) = (fcarbene/fylide)(1/slope),1,2
where fcarbene (0.5784) is the computed oscillator strength of
PNPCC at 309 nm, and fylide (0.5688) is the computed oscillator
strength of ylide 4 at 456 nm (see Table 3 below and the
Supporting Information). The slope of the correlation line in
Figure 4 is 0.135 ( 0.005 M, r = 0.998, and we obtain Keq
=
(0.5784/0.5688)(1/0.135 M) = 7.5 ( 0.3 MÀ1 at 295 K for the
equilibrium [PNPCC + THF h 4]. We estimate that the ratio
fcarbene/fylide accurately reproduces the desired (εcarbene/εylide
ratio to within a factor of 2.
)
Figure 5. Average values of A316/A476 vs 1/[Et2O] (MÀ1) for PNPCC
and O-ylide 5 in pentane at 295 K, 80 À160 ns after LFP of diazirine 3.
The slope of the correlation line is 6.65 ( 0.11 M, r = 0.999. See Table
S-1 in the Supporting Information for the numerical values of the
experimental points.
3.2. PNPCC and Diethyl Ether or Di-n-propyl Ether. Similar
analyses are carried out for the reactions of PNPCC with diethyl
ether, affording O-ylide 5, and of PNPCC with di-n-propyl ether,
affording O-ylide 6. For PNPCC and diethyl ether (Et2O),
O-ylide 5 absorbs at 476 nm,4 and a plot of calibrated average values
of A316/A476 vs 1/[Et2O] gives the correlation shown in Figure 5,
where the slope = 6.65 ( 0.11 M and r = 0.999. The computa-
tions (cf. Table 3) predict absorption of ylide 5 at 444 nm
(f = 0.6138). Therefore, Keq = (0.5784/0.6138)(1/6.65 M) =
0.142 ( 0.002 MÀ1 at 295 K for the equilibrium [PNPCC +
Et2O h 5].
Diethyl ether is volatile, raising concern about the accuracy of
its putative concentrations. Therefore, we conducted a parallel
series of experiments with its less volatile analogue, di-n-propyl
ether (Pr2O). O-Ylide 6, from PNPCC and Pr2O, absorbs at
468 nm. A plot of calibrated average values of A316/A468 vs
1/[Pr2O] affords the correlation shown in Figure 6, where the
slope is 9.16 ( 0.38 M and r = 0.997. The computations
(cf. Table 3) predict absorption of ylide 6 at 446 nm (f = 0.6107).
Figure 6. Average values of A316/A468 vs 1/[Pr2O] (MÀ1) for PNPCC
and O-ylide 6 in heptane at 295 K, 80À160 ns after LFP of diazirine 3.
The slope of the correlation is 9.16 ( 0.38, r = 0.997. See Table S-1 in the
Supporting Information for the numerical values of the experimental
points.
Table 2. Equilibria of PNPCC and Alkyl Ethers
a,b
ether
ylide
λmax (nm)
Keqa (MÀ1
)
Krel
ΔHo (kcal/mol)
À11 ( 1
ΔSo (eu)
À32 ( 3
ΔGoa(kcal/mol)
À1.19 ( 0.02
THF
Et2O
Pr2O
4
5
6
460
476
468
7.5 ( 0.3
1.0
0.142 ( 0.002
0.103 ( 0.004
0.019
0.014
À6.5 ( 0.9
À27 ( 3
1.33 ( 0.02
a At 295 K. b Relative to Keq for [THF + PNPCC h 4].
360
dx.doi.org/10.1021/jp2098119 |J. Phys. Chem. A 2012, 116, 358–363

The Journal of Physical Chemistry A
ARTICLE
Therefore, Keq = (0.5784/0.6107)(1/9.16 M) = 0.103 ( 0.004 MÀ1
at 295 K for the equilibrium [PNPCC + Pr2O h 6].
Plots of ln Keq vs 1/T are shown in Figure 7 for [THF +
PNPCC h4] and in Figure 8 for [Pr2O + PNPCC h6]. The
slopes and Y-intercepts of these correlations afford ΔHo and ΔSo
for the equilibria; these values are recorded in Table 2. Both
equilibria are enthalpically favorable for the formation of the
O-ylides, but the associated entropies (ΔSo ∼ À30 eu) raise
ΔGo considerably. As shown in Table 2, ΔGo remains thermo-
dynamically favorable for PNPCC + THF (ΔGo = À1.2 kcal/mol)
but is unfavorable for Pr2O (ΔGo = +1.3 kcal/mol).
The measured equilibrium constants for PNPCC forming
O-ylides with THF, Et2O, and Pr2O are collected in Table 2. The
ordering of Keq, by ether substrate, is THF > Et2O ≈ Pr2O, with
relative Keqs of 1.0, 0.019, and 0.014, respectively. THF affords
the most favorable carbene/O-ylide equilibrium, 50À70 times
more favorable than Et2O and Pr2O. The [THF + PNPCC h 4]
equilibrium is thermodynamically favorable and lies modestly on
the product side, whereas the equilibria with Et2O or Pr2O lie on
the reactant side.
3.3. Enthalpy and Entropy of O-Ylide Formation. In order
to extract the thermodynamic parameters governing these equi-
libria, we examined the temperature dependence of Keq for the
THF and Pr2O substrates; we excluded Et2O because of its
volatility. We monitored A316/A460 vs 1/[THF] or A316/A468 vs
1/[Pr2O] to determine Keq at five or six different temperatures in the
range 285 e T e 323 K. The data obtained in these determinations
appear as Figures S-1 and S-2 in the Supporting Information,
while the slopes, Y-intercepts, and Keq values derived from these
correlations at each temperature are collected in Table S-2.
3.4. Computational Studies. Electronic structure calcula-
tions based on density functional theory provide structures and
energetics for PNPCCÀether O-ylides and rationalize their
electronic absorption spectra. We have applied four exchange-
correlation functional combinations (B97D,12 M06 and M06-
2X,13 and wB97XD14) and 6-311+G(d)15 basis sets toward the
computation of ground state properties (idealized gas phase
conditions). The thermodynamic parameters (ΔHo, ΔSo) for
O-ylide formation produced by the chosen functionals are quite
similar (Table S-3 in the Supporting Information); the enthalpies
generated by the B97D set of functionals show, by a small margin,
the overall best agreement with experimental observations. At the
B97D/6-311+G(d) level, the computed O-ylide formation enthal-
pies are ΔHo = À10.6 kcal/mol for 4 and À9.1 kcal/mol for 6,
whereas the measured values are À11 and À6.5 kcal/mol,
respectively (Table 2). When soluteÀsolvent interactions
(continuum dielectric model, model heptane solvent) are in-
cluded in the geometry optimizations, the resulting enthalpies for
formation become more favorable by ca. 2.7 kcal/mol (ΔHo =
À13.3 kcal/mol for 4 and À11.7 kcal/mol for 6; Table 3). The
increased thermodynamic stability of the O-ylides is accompa-
nied by a reduction in the length of the polar carbeneÀether
CÀO bond by ca. 0.03 Å (Table 3). All the tested functionals
include dispersion corrections, but only B97D does not include
exact (HartreeÀFock) exchange. Inclusion of exact exchange
appears to increase the computed energies of O-ylide formation,
and all the tested hybrid functionals (M06, M06-2X, and
wB97XD) produce too large enthalpies (Table S-3). The effect
is particularly evident in the enthalpies obtained with the M06-
2X functionals, which incorporate approximately twice as much
exact exchange as M06 or wB97XD.
Figure 7. Plot of ln Keq (MÀ1) vs 1/T (KÀ1) for the equilibrium
[PNPCC + THF h 4]. The slope (5312) affords ΔHo = À10.6 (
1.0 kcal/mol, and the Y-intercept (À15.9) gives ΔSo = À31.6 ( 3 eu.
The correlation coefficient is r = 0.986.
However, the computed entropies for O-ylide formation are
far more negative (∼À43 eu, Table 3) than observed (∼À30 eu,
Table 2), independent of the functional combinations used
(Table S-3) or the inclusion of continuum solvent (Table 3).
Due to the overestimation of the formation entropies, the
computed Gibbs free energies are strongly dominated by the
large positive ÀTΔSo term and are generally unfavorable
(Table 3); consequently, the computed equilibrium constants
are too small in magnitude. However, the [PNPCC + THF h4]
equilibrium is computed to be most favorable with an equilibrium
constant approximately 10À60 times larger than the equilibrium
Figure 8. Plot of ln Keq (MÀ1) vs 1/T (KÀ1) for the equilibrium
[PNPCC + Pr2O h 6]. The slope (3250) affords ΔHo = À6.46 (
0.9 kcal/mol, and the Y-intercept (À13.4) gives ΔSo = À26.7 ( 3 eu.
The correlation coefficient is r = 0.964.
Table 3. Computed Properties of O-Ylidesa
ether/ O-ylide
CÀO (Å)
ΔHo (kcal/mol)
ΔSo (cal/(mol K))
ΔGo (kcal/mol)
λmaxb,c (nm)
f b,c
K (rel)
1.0(1.0)
THF/4
Et2O/5
Pr2O/6
1.476(1.450)
1.477(1.450)
1.481(1.451)
À10.6(À13.3)
À8.3(À11.1)
À9.1(À11.7)
À43.1(À43.1)
À43.5(À42.3)
À43.1(À42.3)
2.3(À0.5)
4.7(1.5)
3.8(0.9)
456
444
446
0.5688
0.6138
0.6107
0.016(0.036)
0.076(0.094)
a Calculations are B97D/6-311+G(d) (thermodynamics) and TD-B3LYP/6-311+G(d)//B97D/6-311+G(d) (UVÀvis). Values in parentheses refer to
calculations in which geometries were optimized including solvent (heptane) corrections; see computational details for the computation of reaction
enthalpies and free energies. b In simulated heptane. c The π f p absorption of PNPCC is computed at 309 nm, f = 0.5784.
361
dx.doi.org/10.1021/jp2098119 |J. Phys. Chem. A 2012, 116, 358–363

The Journal of Physical Chemistry A
ARTICLE
’ ACKNOWLEDGMENT
We thank Dr. Lei Wang for helpful discussions. We are pleased
to acknowledge that several preliminary experiments were
carried out in the laboratory of Professor N. J. Turro at
Columbia University, with the assistance of Dr. Steffen Jockusch.
We are grateful to the National Science Foundation for financial
support.
Figure 9. (a) B97D/6-311+G(d) computed structure for PNPCCÀ
Pr2O O-ylide 6. (b) HOMO of 6. (c) LUMO of 6.
’ REFERENCES
(1) Moss, R. A.; Wang, L.; Odorisio, C. M.; Krogh-Jespersen, K.
J. Am. Chem. Soc. 2010, 132, 10677–10679.
(2) Wang, L.; Moss, R. A.; Krogh-Jespersen, K. J. Phys. Chem A 2011,
115, 8113–8118.
constants for species 5 and 6, in qualitative good agreement with
experimental observations.
Figure 9a illustrates the conformer of lowest energy for
PNPCCÀPr2O O-ylide 6 (analogous computed structures of 4
and 5 are available as Figure 4 in ref 4). The computed electronic
transitions (Table 3) are blue-shifted by only 10À20 nm relative
to the experimental peaks and thus compare favorably (Table 2).
The identifying absorption near 450 nm arises as the result of a
very pure (>95%) HOMO f LUMO transition, which contains
significant charge-transfer character. The HOMO is a relatively
well-localized lone pair (essentially of π-type symmetry) on the
carbene carbon with some admixture from the phenyl ring; the
LUMO is a phenyl π* orbital with considerable contributions
from the p-NO2 substituent (parts b and c of Figure 9, respectively).
Neither the HOMO nor the LUMO contains much O(ether)
character, which rationalizes the insensitivity of the PNPCCÀ
ether O-ylide signature transition, in energy and intensity, to the
specific nature of the alkyl ether.
(3) Wang, L.; Moss, R. A.; Thompson, J.; Krogh-Jespersen, K. Org.
Lett. 2011, 13, 1198–1201.
(4) Moss, R. A.; Wang, L.; Weintraub, E.; Krogh-Jespersen, K.
J. Phys. Chem. A 2008, 112, 4651–4659.
(5) Bonneau, R.; Liu, M. T. H. J. Am. Chem. Soc. 1990, 112, 744–747.
(6) (a) Celebi, S.; Tsao, M.-L.; Platz, M. S. J. Phys. Chem. A 2001,
105, 1158–1162. (b) Tsao, M.-L.; Zhu, Z.; Platz, M. S. J. Phys. Chem. A
2001, 105, 8413–8416.
(7) The spectroscopic signature of the O-ylide was not observed.
Equilibration was deduced from the LFP-UV monitored kinetics of the
parallel formation of the N-ylide of pyridine.8
(8) Tippmann, E. M.; Platz, M. S.; Svir, I. B.; Klymenko, O. V. J. Am.
Chem. Soc. 2004, 126, 5750–5762.
(9) For more details, see: Moss, R. A.; Tian, J.; Sauers, R. R.; Ess,
D. H.; Houk, K. N.; Krogh-Jespersen, K. J. Am. Chem. Soc. 2007,
129, 5167–5174. The current pulse width is broader than reported in
the previous study.
(10) Parr, R. G.; Yang, W. Density-Functional Theory of Atoms and
Molecules; University Press: Oxford, U.K., 1989.
4. CONCLUSIONS
(11) Frisch, M. J. et al. Gaussian 09, revision A.02; Gaussian, Inc.:
Pittsburgh, PA, 2009. See reference 1 in the Supporting Information for
the complete reference to Gaussian 09.
(12) Grimme, S. J. Comput. Chem. 2006, 27, 1787–1799.
(13) (a) Zhao, Y.; Truhlar, D. G. J. Phys. Chem. A 2006, 110,
5121–5129. (b) Zhao, Y.; Truhlar, D. G. Theor. Chim. Acc. 2008, 120,
215–241.
p-Nitrophenylchlorocarbene reacts reversibly with diethyl
ether, di-n-propyl ether, or tetrahydrofuran (THF) to form
O-ylides. These can be visualized by their UVÀvis spectroscopic
signatures. Equilibrium constants (Keq) could be determined
spectroscopically and ranged from 0.10 MÀ1 (di-n-propyl ether)
to 7.5 MÀ1 (THF) at 295 K. Studies of Keq as a function of tem-
perature afforded ΔHo, ΔSo, and ΔGo values for the di-n-propyl
ether and THF/O-ylide equilibria. ΔHo was favorable for ylide
formation, but ΔSo was quite negative (∼À30 eu). Therefore
ΔGos for the equilibria were small, ∼ À1.2 kcal/mol for THF
and +1.3 kcal/mol for di-n-propyl ether. Electronic structure
calculations based on density functional theory provided struc-
tures, spectroscopic signatures, and energetics for the carbene/
ether O-ylides. Comparisons of the computed and experimental
data were generally satisfactory.
(14) Chai, J.-D.; Head-Gordon, M. Phys. Chem. Chem. Phys. 2008,
10, 6615–6620.
(15) (a) Ditchfield, R.; Hehre, W. J.; Pople, J. A. J. Chem. Phys. 1971,
54, 724–728. (b) Hariharan, P. C.; Pople, J. A. Mol. Phys. 1974,
27, 209–214. (c) Krishnan, R.; Binkley, J. S.; Seeger, R.; Pople, J. A.
J. Chem. Phys. 1980, 72, 650–654. (d) McLean, A. D.; Chandler, G. S.
J. Chem. Phys. 1980, 72, 5639–5648. (e) Clark, T.; Chandrasekhar, J.;
Spitznagel, G. W.; Schleyer, P. v. R. J. Comput. Chem. 1983, 4, 294–301.
(16) Barone, V.; Cossi, M. J. Phys. Chem. A 1998, 102, 1995–2001.
(17) McQuarrie, D. A. Statistical Thermodynamics; Harper and Row:
New York, 1973.
(18) Casida, M. E.; Jamorski, C.; Casida, K. C.; Salahub, D. R.
J. Chem. Phys. 1998, 108, 4439–4449.
’ ASSOCIATED CONTENT
(19) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648–5652. (b) Lee,
C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785–789.
(20) Calibration corrects the raw UVÀvis absorptions for wave-
length-dependent variations in sample absorptivity (including precursor
diazirine), xenon monitoring lamp emission, and detector sensitivity.
(21) In the uncalibrated spectrum in pentane, reproduced in ref 4,
the π f p absorption appears at 316 nm, and the σ f p absorption at
628 nm is much more pronounced.
S
Supporting Information. Figures S-1ÀS-2; Tables
b
S-1ÀS-3; complete reference to Gaussian 09; B97D/6-311+G-
(d) optimized geometries and absolute energies of THF; Et2O,
Pr2O, 4, 5, and 6; TD-B3LYP/6-311+G(d)//B97D/6-311+G-
(d) electronic excitation energies and oscillator strenths for THF,
Et2O, Pr2O, 4, 5, and 6. This material is available free of charge
via the Internet at http://pubs.acs.org.
(22) Thus, the interaction between the phenyl π-type orbitals and
the (formally empty) carbene p orbital leads to a computed σÀp
(HOMOÀLUMO) gap in PNPCC, which is too small. The tendency
of DFT to underestimate the separation between occupied and unoccupied
levels for weakly interacting systems, and thus of TD-DFT to underestimate
the electronic excitation energies when local, time-independent functionals
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: moss@rutchem.rutgers.edu (R.A.M); krogh@rutchem.
rutgers.edu (K.K.-J.).
362
dx.doi.org/10.1021/jp2098119 |J. Phys. Chem. A 2012, 116, 358–363

The Journal of Physical Chemistry A
ARTICLE
are employed, has been documented; see, e.g., Dreuw, A.; Weisman, J. L.;
Head-Gordon, M. J. Chem. Phys. 2003, 119, 2943–2946.
(23) The UVÀvis absorption of THF ylide 4 is reported at ∼510 nm
in pure THF; see Figure 3 in ref 6a.
(24) The absorbance vs time traces in Figure 2 were determined at
316 nm, where our detector system is 40% more sensitive than at
308 nm.
(25) Carbene 1 is very likely formed from excited diazirine 3 on the
picosecond time scale, faster than the time resolution of our nanosecond
LFP system. See, for example, Zhang, Y.; Burdzinski, G.; Kubicki, J.;
Vyas, S.; Hadad, C. M.; Sliwa, M.; Poizat, G.; Buntinx, G; Platz, M. S.
J. Am. Chem. Soc. 2009, 131, 13784–13790. O-ylide 4 also appears to
form from carbene 1 and THF faster than our ∼60 ns time resolution.
(26) At times >200 ns, scatter increases due to decay of the species
and a diminished signal/noise ratio. At times <60 ns, the carbene and
ylide absorptions are within the laser pulse width.
363
dx.doi.org/10.1021/jp2098119 |J. Phys. Chem. A 2012, 116, 358–363