Electron photodetachment spectroscopy of (E)- and (Z)-propionaldehyde enolate anions. Electron affinities of the stereoisomers of propionaldehyde enolate radicals

Full text of DOI:10.1021/ja962482d

2054
J. Am. Chem. Soc. 1997, 119, 2054-2055
Electron Photodetachment Spectroscopy of (E)- and
(Z)-Propionaldehyde Enolate Anions. Electron
Affinities of the Stereoisomers of Propionaldehyde
Enolate Radicals
Bettina C. Ro¨mer and John I. Brauman*
Department of Chemistry, Stanford UniVersity
Stanford, California 94305-5080
ReceiVed July 18, 1996
Stereoisomers can have rather different thermodynamic,
kinetic, and spectroscopic properties, so determining the proper-
ties of such isomers has been the subject of many studies.1
Relatively little, however, is known about the properties of
stereoisomers of gas-phase ions because it is difficult to
independently generate and analyze the ions. Chou and Kass2
have reported the synthesis of the (E)- and (Z)-1-propenyl anions
in the gas phase and showed that the two anions have different
reactivities. Ho and Squires3 have distinguished the diastere-
omeric products of gas-phase hydride reductions of alkyl-
substituted cyclohexanones by their reactivities. Chyall, Brick-
house, Schnute, and Squires4 have addressed the stability of
the two stereoisomeric isomers of secondary 2-butanone enolate
and calculated their basicities.
We report here the electron photodetachment spectra of the
(E)- and (Z)-isomers of propionaldehyde enolate anion. These
ions have different stabilities with respect to their neutral enolate
radicals. The anions also show different spectroscopic behavior,
consistent with their structures, that allows us to differentiate
between them.
Experiments were performed with an ion cyclotron resonance
spectrometer (ICR), in which ions were continuously generated
and detected, as described previously.5 The light source was a
Ti:sapphire laser (Lexel 479, bandwidth of (1 cm-1). We
synthesized (E)- and (Z)-trimethylsilyl enol ethers of propional-
dehyde, as outlined by House et al.,1d and confirmed their
structures and purities (>98%) by NMR. Fluoride ion (gener-
ated from NF3) was used to displace the trimethylsilyl (TMS)
group, producing the corresponding anions (eq 1a,b).
Figure 1. High-resolution photodetachment spectrum of (E)-propi-
onaldehyde enolate anion (top panel) and (Z)-propionaldehyde enolate
anion (bottom panel). Arrows indicate assignment of the electron
affinity.
concentrated in the enolate product.6 It is possible that each
“pure” enolate is contaminated with small amounts of the other
isomer (see below). However, the spectra are sufficiently
different to assure us that the ions have been characterized.
The (predominantly) (E)-isomer shows a fairly sharp onset
at 773 nm, with two clearly distinguishable, narrow (4 nm)
resonances at 766 and 749 nm.7 Above 745 nm, the cross
section continues to rise smoothly, with a very slight slope
change around 720 nm.8 The photodetachment cross section
of the (predominantly) (Z)-isomer rises above zero at 784 nm.
It increases slowly and displays a very small, reproducible
resonance at 766 nm, which is broader (10 nm) than the lower
energy resonance (766 nm) in the spectrum of the (E)-enolate.
Because the signal to noise is relatively low, it is difficult to
discern any other resonance at higher energies. A dramatic slope
change in the cross section is observed at 717 nm; the cross
section continues to increase smoothly toward shorter wave-
lengths. A photodetachment spectrum of the enolate generated
by deprotonation of propionaldehyde has been recorded previ-
ously over the range 771-743 nm.9 The electron affinity
determined in that experiment was 37.38 ( 0.14 kcal/mol. The
photodetachment spectrum9 resembles the spectrum of the (E)-
enolate anion; on the basis of the results reported here, the cross
section features observed in the previously recorded spectrum
can be assigned to the (E)-enolate anion. The photoelectron
spectrum of the enolate formed by deprotonation of propional-
dehyde, possibly a mixture of isomers, shows vibrational
structure that was described as complex.10 The electron affinity
(6) (a) Froehlicher, S. W.; Freiser, B. S.; Squires, R. R. J. Am. Chem.
Soc. 1986, 108, 2853. (b) The activation energy necessary to interconvert
(E)- and (Z)-propionaldehyde enolate anions is related to the barrier of
rotation around the π-bond, which we calculated to be about 28 kcal/mol.
The magnitude of the barrier agrees with the barriers to rotation calculated
for similar compounds by Froehlicher et al. (see previous reference). The
exothermicity of a reaction of fluoride ion with a trimethylsilyl enol ether
derivative is approximately 23 kcal/mol. We estimate that the well depth
for the reaction path adds about 10 kcal/mol to the available energy, giving
a total of 33 kcal/mol.
(7) (a) The splitting between the two resonances at 766 and 749 nm is
296 cm-1, corresponding to a vibronic transition. Vibrational excitations
have been observed for other enolate ions and have been discussed
previously. (b) Jackson, R. L.; Hiberty, P. C.; Brauman, J. I. J. Chem. Phys.
1981, 74, 3705. (c) Marks, J.; Brauman, J. I.; Mead, R. D.; Lykke, K. R.;
Lineberger, W. C. J. Chem. Phys. 1988, 88, 6785.
The electron photodetachment spectra are shown in Figure
1. Although the TMS enol ethers are >98% pure, the ions are
produced in a reaction whose exothermicity is sufficient to allow
isomerization of the enolate if all of the reaction energy is
(1) (a) Bingham, R. C. J. Am. Chem. Soc. 1976, 98, 535. (b) Levin, C.;
Moss, R. A. J. Am. Chem. Soc. 1973, 95, 629. (c) Eliel, E. L.; Bailey, W.
F.; Kopp, L. D.; Willer, R. L.; Grant, D. M.; Bertrand, R.; Christensen, K.
A.; Dalling, D. K.; Duch, M. W.; Wenkert, E.; Schell, F. M.; Cochran, D.
W. J. Am. Chem. Soc. 1975, 97, 322. (d) House, H. O.; Czuba, L. J.; Gall,
M.; Olmstead, H. D. J. Org. Chem. 1969, 34, 2324.
(2) Chou, P. K.; Kass, S. R. J. Am. Chem. Soc. 1991, 113, 4357.
(3) Ho, Y.; Squires, R. R. J. Am. Chem. Soc. 1992, 114, 10961.
(4) Chyall, L. J.; Brickhouse, M. D.; Schnute, M. E.; Squires, R. R. J.
Am. Chem. Soc. 1994, 116, 8681.
(5) (a) The continuous-mode detection provided a large signal-to-noise
ratio which allowed detection of small changes (less than 1%) in the ion
population. (b) Marks, J.; Drzaic, P. S.; Foster, R. F.; Wetzel, D. M.;
Brauman, J. I.; Uppal, J. S.; Staley, R. H. ReV. Sci. Instrum. 1987, 58, 1460.
(c) Wetzel, D. M.; Brauman, J. I. Chem. ReV. 1987, 87, 607.
(8) We find that the photodetachment spectrum of (E)-propionaldehyde
enolate anion is best described (neglecting the resonances) by two straight
lines intersecting at about 715 ( 4 nm.
(9) Brinkman, E. A.; Berger, S.; Marks, J.; Brauman, J. I. J. Chem. Phys.
1993, 99, 7586.
(10) Ellison, G. B.; Engelking, P. C.; Lineberger, W. C. J. Phys. Chem.
1982, 86, 4873.
S0002-7863(96)02482-1 CCC: $14.00 © 1997 American Chemical Society

Communications to the Editor
J. Am. Chem. Soc., Vol. 119, No. 8, 1997 2055
determined from the photoelectron spectrum (37.15 ( 0.53 kcal/
mol) was in good agreement with the photodetachment threshold
measured by ICR, suggesting that it can again be assigned to
the (E)-enolate radical.
In the spectrum of the (Z)-isomer, we observe a dramatic
slope change at 717 nm. We expect the two isomers to display
different onsets, as well as different resonances due to dipole-
bound states, and assign the 0-0 transition to the onset of that
slope change. We believe that the long tail to the red of the
slope change is caused by a small amount (∼5%) of the (E)-
isomer present in the sample.14 In the tail, we can distinguish
a small resonance feature, which corresponds to a resonance in
the spectrum of the (E)-isomer (766 nm). The resonance in
the tail is broader than the corresponding resonance in the
spectrum of the (E)-isomer. For dipole-bound states, excess
energy has been shown to cause shorter lifetimes and thus
broader resonances.15 We presume that some of the isomerized
(E)-enolate has excess vibrational energy and has not thermal-
ized completely.16 Given that a larger fraction (than usually
expected for ions trapped in the ICR for extended periods of
time) of the isomerized (E)-enolate is excited, the tail at
wavelengths below 773 nm in the spectrum of the predominantly
(Z)-isomer can be assigned to a hot band of the isomerized (E)-
enolate.
The resonances in the (E)-isomer are similar to those observed
in other simple enolates; we attribute these resonances to tran-
sitions to a dipole-supported state that undergoes autodetach-
ment.7b,9,11 The dipole moments of (E)- and (Z)-propionalde-
hyde enolate radicals were calculated from ab initio molecular
orbital calculations, using Gaussian 94, to be 3.34 and 2.93 D,
respectively.12 The spectroscopic results we observe are in good
agreement with the model we proposed previously,9 in which
the lifetime of a dipole-supported state depends on the relative
directions of the figure axis and dipole moment of the neutral.
The (E)-isomer is a near symmetric top, and the dipole moment
is aligned close to the figure axis. For such a system, in which
it is possible for the molecule to be in a fairly high rotational
state but in which the dipole does not rotate, we expect the
dipole state to be fairly long lived and thus to show sharp
resonances. In contrast, in the (Z)-isomer, which is also a near
symmetric top, the dipole moment is not aligned with the
principal axis. Its dipole-supported state is expected to be
shorter lived and thus to show broader resonances. Indeed, these
resonances may be too broad for us to observe.9
From the photodetachment onset, we can assign the electron
affinity for the isomers. For the (E)-isomer, we can determine
the electron affinity from the observed resonances caused by
autodetachment from the dipole-bound state. The photodetach-
ment spectra of acetaldehyde and acetaldehyde-d3 also show a
number of resonances.7b,11b For acetaldehyde, the lower energy
resonance was assigned to be the 0-0 transition, confirmed by
ultra-high-resolution experiments by Lineberger and co-
workers.11b They measured an electron binding energy for the
dipole-bound state of 6 cm-1 relative to the neutral plus electron
continuum. Because the structure and dipole moment of the
propionaldehyde enolates are roughly similar to those of
acetaldehyde enolate, we can infer that the large lower energy
resonance (766 nm) in the spectrum of the (E)-enolate anion
corresponds closely to the electron binding energy of the anion.
We assign the electron affinity for the (E)-enolate radical to be
37.3 ( 0.2 kcal/mol (1.619 ( 0.007 eV), in good agreement
with the value determined from the photodetachment of depro-
tonated propionaldehyde.7b,9,11b,13 There is evidence of a small
hot band to the red of the assigned onset.
On this basis, we determine the onset for adiabatic electron
detachment at 715 ( 5 nm,17 yielding an electron affinity for
the (Z)-radical of 40.0 ( 0.3 kcal/mol (1.73 ( 0.01 eV).18 The
(E)-isomer appears to also contain small amounts (∼7%) of the
corresponding (Z)-isomer, indicated by the slight slope change
at about 720 ( 5 nm in the spectrum of the (E)-enolate.8
A
slope change at about 725 ( 20 nm was also observed in the
low-resolution spectrum of the enolate anions formed by the
deprotonation of propionaldehyde, indicating that the anions
were again a mixture of (E)- and (Z)-isomers.19
In summary, we have generated the (E)- and (Z)-enolates of
propionaldehyde and have found them to possess different
intrinsic properties. The anion stabilities relative to the neutral
radicals differ noticeably, as does their spectroscopic behavior.
The electron affinity of (E)-propionaldehyde enolate radical was
measured to be 37.3 ( 0.2 kcal/mol and that of the correspond-
ing (Z)-radical to be 40.0 ( 0.3 kcal/mol. Finally, the spectrum
of the (E)-isomer displayed resonances attributed to dipole-
bound states, while the spectrum of the (Z)-isomer does not,
consistent with our model for long-lived dipole states.
Acknowledgment. We are grateful to the National Science Founda-
tion for support of this work.
(11) (a) Lykke, K. R.; Mead, R. D.; Lineberger, W. C. Phys. ReV. Lett.
1984, 52, 2221. (b) Mead, R. D.; Lykke, K. R.; Lineberger, W. C.; Marks,
J.; Brauman, J. I. J. Chem. Phys. 1984, 81, 4883. (c) Marks, J.; Wetzel, D.
M.; Comita, P. B.; Brauman, J. I. J. Chem. Phys. 1986, 84, 5284. (d) Lykke,
K. R.; Neumark, D. M.; Anderson, T.; Trapa, V. J.; Lineberger, W. C. J.
Chem. Phys. 1987, 87, 6842. (e) Murray, K. K.; Lykke, K. R.; Lineberger,
W. C. Phys. ReV. A 1987, 36, 699. (f) Lykke, K. R.; Murray, K. K.;
Neumark, D. M.; Lineberger, W. C. Perkin. Trans. R. Soc. London A 1988,
324, 179. (g) Wetzel, D. M.; Brauman, J. I. J. Chem. Phys. 1989, 90, 68.
(h) Mullin, A. S.; Murray, K. K.; Schulz, C. P.; Szaflarski, D. M.;
Lineberger, W. C. Chem. Phys. 1992, 166, 207. (i) Mullin, A. S.; Murray,
K. K.; Schulz, C. P.; Lineberger, W. C. J. Phys. Chem. 1993, 97, 10281.
(j) Fermi, E.; Teller, E. Phys. ReV. 1947, 72, 399. (k) Wightman, A. S.
Phys. ReV. 1950, 77, 521. (l) Garrett, W. R. J. Chem. Phys. 1982, 77, 3666
(and references therein). (m) Clary, D. C. J. Phys. Chem. 1988, 92, 3173
(and references therein). (n) Simons, J. J. Chem. Phys. 1989, 91, 6858.
(12) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Robb, M. A.;
Cheeseman, J. R.; Keith, T. A.; Petersson, G. A.; Montgomery, J. A.;
Raghavachari, K.; Al-Laham, M. A.; Zakrzewski, V. G.; Ortiz, J. V.;
Foresman, J. B.; Cioslowski, J.; Stefanov, B. B.; Nanayakkara, A.;
Challacombe, M.; Peng, C. Y.; Ayala, P. Y.; Chen, W.; Wong, M. W.;
Andres, J. L.; Replogle, E. S.; Gomperts, R.; Martin, R. L.; Fox, D. J.;
Binkley, J. S.; Defrees, D. J.; Baker, J.; Stewart, J. P.; Head-Gordon, M.;
Gonzalez, C.; Pople, J. A. Gaussian 94 (Version C); Gaussian, Inc.:
Pittsburgh, PA, 1995.
JA962482D
(14) (a) Generally, we do not expect to see hot bands because molecules
are trapped in the ICR cell for 100-1000 ms and can relax by radiative
and collisional processes. Even though the desilylation reaction is reasonably
exothermic, anions formed in this reaction usually appear to thermalize
before photodetachment, and the fraction of excited anions should be fairly
small relative to the fraction of relaxed anions (see next reference). (b)
Brinkman, E. A.; Berger, S.; Brauman, J. I. J. Am. Chem. Soc. 1994, 116,
8304.
(15) Marks, J. Ph.D. Dissertation, Stanford University, 1986.
(16) It is not clear why this enolate did not fully relax.
(17) We determine the onset of the slope change by taking linear least-
squares fits to the higher and lower energy sections of the photodetachment
curve. The intercept of the two best fit curves then corresponds to the
detachment threshold. We have chosen an error limit of (5 nm to easily
accommodate any possible variations in the intercept of the two lines due
to different fitting parameters.
(18) Our calculational results (Gaussian 94, geometry optimizations using
HF/6-31+G* level followed by MP2 single-point calculations) also predict
the electron affinity of the (Z)-radical to be greater than that of the (E)-
radical by about 2 kcal/mol. We found the anion of the (Z)-isomer to be
more stable than that of the (E)-isomer by 2 kcal/mol and found both radicals
to have similar stabilities.
(19) In the CW ICR experiment, propionaldehyde enolate anions were
formed by the exothermic deprotonation of propionaldehyde with fluoride
ion and may not have reached equilibrium concentration before photode-
tachment.
(13) The electron binding energy in these enolate anions is very small,
as shown for acetaldehyde enolate. The error associated with the assignment
of the electron affinity to the center of the vibrational band cannot exceed
the width of the resonance toward higher and lower wavelengths.