Relative Solution Electron Affinities of Selectively Deuteriated Pyrenes: Correlations between Voltammetric, Electron Paramagnetic Resonance, and Semiempirical PM3 Data

Article abstract of DOI:10.1021/jp952098n

The equilibrium isotope effects (EIE) for the one-electron transfer between pyrene and seven regioselectively deuteriated pyrene isotopic isomers in dimethylformamide with 0.1 M tetrabutylammonium hexafluorophosphate were measured electrochemically.These data correlate linearly with the free energies (ΔG^o) obtained in tetrahydrofuran using electron paramagnetic resonance (EPR) techniques.However, the slope of the resulting line is not unity, and it indicates that the EIE in the DMF system is only two-thirds of that in the THF system.PM3 calculated ΔG^o's, which would correspond to the gas phase electron transfers, also correlate linearly with both sets of experimental data, but the predicted magnitudes of the EIE's are smaller than those observed experimentally by either technique.The nonunity slopes probably reflect slight differences in ion solvation and/or ion association parameters between the anion radicals of the isotopic isomers.No general relationship between the EIE and the charge on the hydrogen/deuterium substituted carbon atom was found.

Full text of DOI:10.1021/jp952098n

3454
J. Phys. Chem. 1996, 100, 3454-3462
Relative Solution Electron Affinities of Selectively Deuteriated Pyrenes: Correlations
between Voltammetric, Electron Paramagnetic Resonance, and Semiempirical PM3 Data
Ole Hammerich* and Merete F. Nielsen
Department of Chemistry, UniVersity of Copenhagen, Symbion Science Park, FruebjergVej 3,
DK-2100 Copenhagen, Denmark
Han Zuilhof, Patrick P. J. Mulder, and Gerrit Lodder*
Leiden Institute of Chemistry, Gorlaeus Laboratories, Leiden UniVersity, P.O. Box 9502,
2300 RA Leiden, The Netherlands
Richard C. Reiter,* David E. Kage, Charles V. Rice, and Cheryl D. Stevenson*
Department of Chemistry, Illinois State UniVersity, Normal, Illinois 61790-4160
ReceiVed: July 25, 1995; In Final Form: October 5, 1995X
The equilibrium isotope effects (EIE) for the one-electron transfer between pyrene and seven regioselectively
deuteriated pyrene isotopic isomers in dimethylformamide with 0.1 M tetrabutylammonium hexafluorophosphate
were measured electrochemically. These data correlate linearly with the free energies (∆G°) obtained in
tetrahydrofuran using electron paramagnetic resonance (EPR) techniques. However, the slope of the resulting
line is not unity, and it indicates that the EIE in the DMF system is only two-thirds of that in the THF
system. PM3 calculated ∆G°’s, which would correspond to the gas phase electron transfers, also correlate
linearly with both sets of experimental data, but the predicted magnitudes of the EIE’s are smaller than those
observed experimentally by either technique. The nonunity slopes probably reflect slight differences in ion
solvation and/or ion association parameters between the anion radicals of the isotopic isomers. No general
relationship between the EIE and the charge on the hydrogen/deuterium substituted carbon atom was found.
Introduction
radical anion yields smaller Keq values.7 This conclusion was
reinforced by the observation of a linear free energy relationship
for reaction 2.8
The effect of deuterium substitution on the reduction potential
of aromatic compounds has recently been studied in great detail,
both experimentally and theoretically. Electron paramagnetic
resonance (EPR), electrochemical, and other studies have shown
that the electron-accepting ability of polycyclic aromatic
hydrocarbons decreases upon deuteriation.1 That is, the standard
free energy change of reaction 1, where A and *A represent
the isotopically light and heavy analogues, respectively, is
positive.
(2)
When X represents an electron-donating substituent, Keq for
reaction 2 is smaller than it is when X represents an electron-
withdrawing substituent. The substituent causes variations in
the total excess charge density from the antibonding electron
in the part of the molecule where isotopic substitution has taken
place. With electron-withdrawing substituents this excess is
smaller, and the influence of isotopic substitution is therefore
smaller.
Two reports have been published that give some theoretical
clarification of the effect.9,10 Marx, Kleinhesselink, and Wolfs-
berg performed calculations using empirical values of force
constants and concluded that the deuterium isotope effects,
where X ) NO2 and H, are explicable in terms of standard
isotope theory.9 Zuilhof and Lodder later reported analogous
calculations of the isotope effects on the reaction enthalpy in
which the force constants were quantum mechanically derived.10
The latter authors concluded that the overall isotope effect was
the sum of many small, and often oppositely directed, vibrational
changes. For the case where A represents benzene or pyrene
(1a) and *A represents perdeuteriated benzene or perdeuteriated
pyrene (1h) in reaction 1, they found no obvious relationship
between the amount of charge residing on a carbon atom and
the magnitude of ∆G°. This ∆G° was assumed to be identical
to ∆H° as ∆S° ) 0.
A
•- + *A a A + *A•-
(1)
The existence of this deuterium isotope effect on the electron
affinity has been confirmed by a variety of techniques, among
them NMR,2 ion cyclotron resonance,3 mass spectrometry,4 and
cyclic voltammetry.1b,5 The endothermicity has been explained
in terms of the extra electron being added to an antibonding
π*-orbital, which reduces the total bonding.1a,4 In agreement
with the general rule that “the lighter isotope prefers the looser
species”, the equilibrium of reaction 1 lies to the left. The
magnitude of the effect is dependent on the number of
isotopically substituted sites.6 For example, the more deuteriums
on the benzene ring, the more positive is the reaction enthalpy
for electron transfer from the benzene anion radical to deute-
riated benzene.6
The magnitude of the effect has also been linked to the spin
and charge density at the isotopically substituted carbon atom.
From the study of a series of perdeuteriated polycyclic aromatic
hydrocarbons, Stevenson and Sturgeon concluded that deuterium
substitution on carbon atoms with larger charge densities in the
X Abstract published in AdVance ACS Abstracts, February 1, 1996.
0022-3654/96/20100-3454$12.00/0 © 1996 American Chemical Society

Electron Affinities of Deuteriated Pyrenes
J. Phys. Chem., Vol. 100, No. 9, 1996 3455
In view of the widespread interest regarding this equilibrium
isotope effect,1b-e,5,9,10 we were motivated to further investigate
the correlation of experimental Keq’s for reaction 1 with
theoretically computed Keq’s and with charge distributions in
the anion radicals. Seven specifically deuteriated pyrene
isotopic isomers (structures 1b-1g and perdeuteriated pyrene
(1h), as a precursor to 1f) were synthesized, and the ∆G° values
for reaction 1, where A ) pyrene and *A ) 1b-1h, were
measured electrochemically and by EPR, a technique that has
frequently been used to determine ∆G° values for electron
transfer reactions.11 This was complimented by the calculation
Figure 1. Upper: low-field half (a 12 G scan is shown) of the EPR
spectrum of our sample of 2,4,5,7,9,10-d6-pyrene at 173 K reduced
with K in THF. Middle: computer simulation of the EPR spectrum
for 2,4,5,7,9,10-d6-PY•- (coupling constants shown in Table 2). The
presence of the isotopic impurities is not accounted for. Lower:
computer simulation as above but including 76% 2,4,5,7,9,10-d6-PY•-
in the presence of 18.6% 1,2,4,5,7,9,10-d7-PY•-and 6.8% 2,4,5,7,9-d5-
PY•- (Table 1). Note the improved agreement with the real spectrum
(see, for example, the area indicated by the vertical arrow). The peak-
to-peak line width (PPW) is 0.07 G.
of the reaction enthalpies using semiempirical PM3 MO
calculations.12
Results
TABLE 1: Molar Percentages (Normalized to 100%) of
Various Deuteriated Species in Samples 1b-1g Used in
Optimized EPR Simulations
EPR Experiments. The EPR technique for the determination
of relative solution electron affinities,11a-e which can also be
utilized for H and D isotopic isomers,13 was used to study the
competitive electron transfer equilibria between pyrene and the
partially deuteriated pyrenes. This technique obviously relies
upon the detailed analysis of the EPR spectral patterns of the
individual anion radicals involved, including those of any
significant isotopic isomeric impurities.
% deuteriated
desired compound
deuteriated species
2,7-d2-PY
1,2,7-d3-PY
species
2,7-d2-PY
72.4 1b
4.1
2,4,7-d3-PY
2-d1-PY
2.7
4.7
2,4,5,7,9-d5-PY
2,4,5,7,9,10-d6-PY
misc d4-PY’s
1,3,6,8-d4-PY
1,3,6-d3-PY
4,5,9,10-d4-PY
2,4,5,9,10-d5-PY
4,5,9-d3-PY
1,4,5,9,10-d5-PY
1,2,3,6,7,8-d6-PY
1,2,3,4,6,7,8-d7-PY
1,2,3,6,7-d5-PY
1,2,3,4,6,7-d6-PY
1,2,3,6,8-d5-PY
1,2,3,4,6,8-d6-PY
1,2,3,4,5,6,7,8-d8-PY
1,2,3,4,5,6,7,8,9-d9-PY
d10-PY
5.5
3.1
7.5
2,4,5,7,9,10-d6-Pyrene (1f). The potassium reduction of our
synthesized 2,4,5,7,9,10-d6-pyrene (pyrene ) PY) in THF yields
an anion radical solution that exhibits the expected EPR pattern
from the desired 2,4,5,7,9,10-d6-PY•-, but this anion radical only
accounts for 74.6% of the spectral intensity. It is necessary to
include an 18.6% 1,2,4,5,7,9,10-d7-PY•- and a 6.8% 2,4,5,7,9-
d5-PY•- contribution (in accordance with Table 1) to obtain a
simulation that is a good representation of the real EPR spectrum
(see Figure 1; the aH’s and aD’s are given in Table 2).
Although the EPR spectrum of our potassium reduced sample
of 2,4,5,7,9,10-d6-PY is quite complex, it is very well simulated
as can be seen in Figure 1. The hyperfine lines due to
2,4,5,7,9,10-d6-PY•- appear at magnetic fields where the hy-
perfine lines due to PY•- are relatively intense. Thus, the
mixtures of PY and 2,4,5,7,9,10-d6-PY were biased heavily
toward the deuteriated system in terms of relative concentrations.
This allows good resolution of the 2,4,5,7,9,10-d6-PY•- spectrum
while unperturbed hyperfine components from PY•- can be
found outside of the spectral region of 2,4,5,7,9,10-d6-PY•-
(Figure 2).
1,3,6,8-d4-PY
4,5,9,10-d4-PY
83.0 1c
17.0
74.7 1d
14.3
9.4
1.6
58.8 1e
3.0
8.6
5.8
4.3
2.9
6.9
4.4
2.3
2.9
74.6 1f
18.6
6.8
61.7 1g
17.4
15.5
5.4
1,2,3,6,7,8,-d6-PY
misc d4-PY’s
2,4,5,7,9,10-d6-PY
2,4,5,7,9,10-d6-PY
1,2,4,5,7,9,10-d7-PY
2,4,5,7,9-d5-PY
1,3,4,5,6,8,9,10-d8-PY
1,2,3,4,5,6,8,9,10-d8-PY
1,3,4,5,6,8,9-d7-PY
1,3,4,5,6,9,10-d7-PY
1,3,4,5,6,8,9,10-d8-PY
A mixture of 79.3 mg of our (isotopically impure) 2,4,5,7,9,10-
d6-PY sample (79.3 × 0.746/208 ) 0.284 mmol of 1f) and 1.1
mg (0.005 45 mmol) of PY in 20 mL of THF was reduced with
a very deficient amount of potassium metal and further diluted
with THF. Upon EPR analysis (173 K), the resulting anion
radical solution exhibited both expected anion radicals along
with the impurity anion radicals mentioned above. The EPR
spectrum (Figure 2) is best simulated with the ratio of the PY
and 2,4,5,7,9,10-d6-PY anion radicals being [2,4,5,7,9,10-d6-

3456 J. Phys. Chem., Vol. 100, No. 9, 1996
Hammerich et al.
TABLE 2: Coupling Constants in Gauss Used To Generate
Simulations of the EPR Spectra
a
compound(s)
aH1,3,6,8 aH4,5,9,10 aH2,7 aD1,3,6,8 aD4,5,9,10 aD2,7
1a, 1b, 1d-f, 1h 4.818 2.130 1.012
0.746 0.335 0.152
1c, 1g 4.818 2.130 0.9875 0.746 0.335
a The perturbation of the 2,7-proton coupling constant caused by
deuteriation in the 1-, 3-, 6-, and 8-positions is real and is the subject
of a separate article.31
Figure 3. Upper: low-field half (9 G scan) of our reduced 1,2,3,6,7,8-
d6-PY product. Lower: computer simulation of this spectrum that is
generated assuming the isotopic impurity levels given in Table 1 and
the coupling constants shown in Table 2. PPW ) 0.138 G.
Figure 2. Upper: an 8 G section, starting just before the second PY•-
peak, of the EPR spectrum resulting from the K reduction of a mixture
of our 2,4,5,7,9,10-d6-pyrene sample mixed with pyrene ([2,4,5,7,9,10-
d6-PY]/[PY] ) 52.1) at 173 K in THF. Lower: computer simulation
assuming [2,4,5,7,9,10-d6-PY•-]/[PY•-] ) 27.3 and isotopic impurities
at levels indicated in Table 1. PPW ) 0.08 G.
PY•-]/[PY•-] ) 27.3. The isotopic impurity anion radicals were
included in the simulation in relative abundancies as described
in Table 1. Electron transfer between various deuterio isomers
may have altered these ratios somewhat. However, the presence
of both more and less deuteriated isotopic “impurities” tends
to average out any observable effect of this equilibration. In
any event, the ∆EA between pyrene and a given deuteriated
isomer is significantly larger than those between the given
isomer and its major cosynthesized impurities. The Keq for
reaction 3 implied by this particular experiment is 27.3/(0.284/
0.005 45) ) Keq(1f) ) 0.52.
Figure 4. Upper: a 6 G scan of the EPR spectrum at 173 K of a
mixture of 50.5 mg of our (impure) 1,2,3,6,7,8-d6-PY sample containing
0.143 mmol of 1e and 0.007 92 mmol of PY that has been partially
reduced with potassium metal. This scan begins 3 G downfield from
the first line corresponding to the anion radical of 1e, indicated by the
vertical arrow. Lower: computer simulation generated implying an
equilibrium constant for reaction 1 (A ) pyrene and *A ) 1e) of 0.43.
The relative concentrations of 1e and isotopic impurities are included
as shown in Table 1. PPW ) 0.105 G.
tamination at the 4-, 5-, 9-, and 10-positions of compound 1e.
Despite this, an excellent computer simulation of the EPR
spectrum of the anion radical of this material was obtained
(Figure 3) assuming the product to be 58.9% isotopically pure
1,2,3,6,7,8-d6-PY and including the isotopic impurities at the
levels indicated in Table 1.
(3)
When a mixture of 50.5 mg of our (isotopically impure)
1,2,3,6,7,8-d6-PY sample (50.5 × 0.589/208 ) 0.143 mmol of
1e) and 1.6 mg (0.007 92 mmol) of PY is partially reduced, the
resulting EPR spectrum is best simulated (Figure 4) utilizing a
ratio [1,2,3,6,7,8-d6-PY•-]/[PY•-] of 7.83. This implies an
equilibrium constant for reaction 1, where A and *A represent
Repetition of the experiment yielded an average Keq of 0.57 (
0.06 at 173 K.
1,2,3,6,7,8-d6-Pyrene (1e). The synthetic procedure used (see
Experimental Section) resulted in a significant isotopic con-

Electron Affinities of Deuteriated Pyrenes
J. Phys. Chem., Vol. 100, No. 9, 1996 3457
Figure 6. Upper: low-field halves of the EPR spectra of solutions
resulting from the reduction of our 2,7-d2-PY (left, 15 G scan) and
1,3,4,5,6,8,9,10-d8-PY (right, 8 G scan). Lower: computer simulations
of these spectra that are generated by including the (essential) isotopic
impurities given in Table 1 and utilizing the coupling constants shown
in Table 2. PPW ) 0.095 G on the left, and PPW ) 0.140 G on the
right.
Figure 5. Upper: low-field halves of the EPR spectra of solutions
resulting from the reduction of our 1,3,6,8-d4-PY (left, 9 G scan) and
4,5,9,10-d4-PY (right, 13.5 G scan). Lower: computer simulations of
these spectra that are generated by including the isotopic impurities
given in Table 1 and utilizing the coupling constants shown in Table
2. PPW ) 0.125 G on the left and PPW ) 0.110 G on the right.
TABLE 3: Potential Differences and Electrochemical (in
DMF), EPR (in THF), and Theoretical Relative Solution
Electron Affinities (∆G° ) ∆EA for Reaction 1) of the
Various Deuteriated Pyrenes Relative to That of Pyrene
pyrene and 1,2,3,6,7,8-d6-PY respectively, of 7.83/(0.143/
0.007 92) ) 0.43. Repetition yielded an average value for Keq(1e)
of 0.42 ( 0.05 at 173 K.
∆EAEL, ∆EAEPR, ∆EAROHF
,
∆Ep, mV J/mol
J/mol
J/mol
*A
1c and 1d. Each of these materials was treated as described
for 1e and 1f in a manner that is consistent with the results in
Tables 1 and 2 and Figure 5. The equilibrium constants at 173
K for reaction 1 are 0.43 ( 0.04 and 0.63 ( 0.04 when *A )
1,3,6,8-d4-PY and 4,5,9,10-d4-PY, respectively.
1g and 1b. Each of these materials was treated as described
for 1e and 1f in a manner that is consistent with the values in
Tables 1 and 2. The EPR analysis of mixtures of our
1,3,4,5,6,8,9,10-d8-pyrene sample and pyrene reveals that the
equilibrium constant for reaction 1 is 0.37 ( 0.04 when *A )
1,3,4,5,6,8,9,10-d8-PY (Figure 6). This system represents our
most highly substituted synthesized deuterio pyrene. The least
deuterium-substituted synthesized pyrene is represented by
structure 1b, and Keq at 173 K for reaction 1 when *A ) 2,7-
d2-PY is 0.80 ( 0.04, (Figure 6).
1h. The perdeuteriated PY (used as supplied from Aldrich
Chemical Co.) was assumed to be 100% isotopically pure. The
equilibrium constant at 173 K for reaction 1 when *A ) 1h
was found to be 0.25 ( 0.04. This value is lower than that
previously reported (0.37 ( 0.06)7 and is based upon more data
and a more sophisticated analysis.
Voltammetric Experiments. Any difference in the reduction
potentials of A and *A (reaction 1) means that the equilibrium
constant deviates from unity. In N,N-dimethylformamide
(DMF) containing 0.1 M tetra-n-butylammonium hexafluoro-
phosphate, the deuteriated pyrenes were invariably found to be
more difficult to reduce than pyrene itself. The largest
difference observed, 12.4 mV, was that between perdeuteriated
pyrene (1h) and pyrene (1a), which translates (∆G° ) -nF∆E°)
into a difference in solution electron affinities of 1197 J/mol.
Identical experiments carried out on compounds 1a-1h yield
potential differences, tabulated in Table 3, ranging from 1.5 to
12.4 mV with standard deviations between 0.2 and 0.4 mV.
MO Calculations. The isotope effects on the one-electron
affinity were calculated as perturbations (due to isotopic
0
0
145
656
309
724
521
965
0
320
1240
668
1210
813
1430
1990
0
17
242
84
268
105
334
355
PY (1a)
2,7-d2-PY (1b)
1.5 ( 0.2
6.8 ( 0.2
3.2 ( 0.2
7.5 ( 0.4
5.4 ( 0.3
10.0 ( 0.2
1,3,6,8-d4-PY (1c)
4,5,9,10-d4-PY (1d)
1,2,3,6,7,8-d6-PY (1e)
2,4,5,7,9,10-d6-PY (1f)
1,3,4,5,6,8,9,10-d8-PY (1g)
d10-PY (1h)
12.4 ( 0.3 1197
substitution) upon the zero-point energy differences between
the various neutral and radical anionic species. These isotope
effects have been calculated using the PM3 parametrization12
with three different formalisms: restricted open-shell Hartree-
Fock (ROHF), unrestricted Hartree-Fock (UHF), and an-
nihilated unrestricted Hartree-Fock (AUHF) theories.14 Pre-
vious calculations on benzene/benzene radical anions using the
UHF formalism for the radical anions accounted for 50-60%
of the experimentally observed ∆EA values and reproduced
geometrical distortions (symmetry lowering from D6h to C2h on
reduction) correctly as well. However, for the one-electron
reduction of cyclooctatetraene the UHF calculations did not
predict the experimental geometry of the radical anion correctly,
while the ROHF formalism did. Therefore, results based upon
all three available formalisms are reported (Table 4).
The charge distributions in pyrene and the radical anion of
pyrene have been calculated as well and are given for the carbon
positions where hydrogen atoms are attached in Table 5. The
calculated values of the charges depend on where the electron
cloud is “cut” between the atoms that share these electrons.
Therefore, the reported data must be considered to be charge-
indicators rather than absolute values. Many more or less
elaborate schemes for obtaining the “correct” atomic charge have
been published, but disputes are still going on concerning which
procedure is best in which case.15 We have therefore used the
standard data as obtained from the VAMP and MOPAC
programs.16 In parentheses are added the so-called light-in-
heavy charges in which the charge on the hydrogen atom is

3458 J. Phys. Chem., Vol. 100, No. 9, 1996
Hammerich et al.
TABLE 4: Difference in Electron Affinity with Respect to
All-H Pyrene (∆EA in J/mol), as Obtained from PM3
Calculations
compound substitution pattern ∆EAROHF ∆EAUHF ∆EAAUHF
1a
1b
1c
1d
1e
1f
d0
2,7-d2
0
17
242
84
268
105
334
355
0
79
401
75
485
159
485
560
0
29
251
84
276
109
330
355
1,3,6,8-d4
4,5,9,10-d4
1,2,3,6,7,8-d6
2,4,5,7,9,10-d6
1,3,4,5,6,8,9,10-d8
d10
1g
1h
TABLE 5: Calculated and Light-in-Heavy (in Parentheses)
Charges at the Carbons in Pyrene and Pyrene Radical
Anion by the Various Methods
position
PY
PY•- (ROHF) PY•- (UHF) PY•- (AUHF)
C1
C2
C4
-0.09 (+0.02) -0.25 (-0.16) -0.22 (-0.13) -0.25 (-0.16)
-0.10 (+0.00) -0.05 (+0.01) -0.07 (-0.01) -0.05 (+0.01)
-0.09 (+0.02) -0.16 (-0.08) -0.16 (-0.08) -0.16 (-0.08)
Figure 7. Plot of ∆EA determined electrochemically in DMF with
tetrabutylammonium hexafluorophosphate at 294 K vs ∆EA determined
via EPR in THF with K+ serving as the counterion at 173 K. The linear
correlation constant is 0.987, and the line fits the equation ∆EAELECT
) 0.61∆EAEPR - 8.99.
TABLE 6: Calculated Charge Differences with the
Corresponding Carbon Atoms in Neutral Pyrene and the
Light-in-Heavy Charge Differences (in Parentheses)
position
PY•- (ROHF)
PY•- (UHF)
PY•-(AUHF)
C1
C2
C4
-0.16 (-0.18)
+0.05 (+0.01)
-0.07 (-0.10)
-0.13 (-0.15)
+0.03 (-0.01)
-0.07 (-0.10)
-0.16 (-0.18)
+0.05 (+0.01)
-0.07 (-0.10)
1d) of the experimental Keq values. For compounds with
deuterium atoms at several distinct positions, to predict the
expected contribution of the deuterium atoms at a position, one
must compensate for the contribution of the deuterium atoms
at other positions. For example, Keq(EPR,1e) ) 0.42. This leads
to a value of Keq(EPR,1e)/Keq(EPR,1b) ) 0.42/0.80 ) 0.52 for the
contribution to the Keq value of the deuterium atoms at the
1-positions, which implies Keq ) (0.52)1/4 ) 0.85 for a single
deuterium at that position. If all experimental data are treated
this way, this yields four Keq values for monodeuteriation at
each position, and averaging leads to Keq values of 0.85, 0.94,
and 0.91 for monodeuteriation at the 1-, 2-, and 4-positions,
respectively. In the case of the electrochemical experiments
these values are 0.94, 0.97, and 0.93, respectively. The
experimental (EPR and less clearly, electrochemical) data
suggest that the isotope effect is largest for deuteriation at the
1-positions. Monodeuteriation at the 2- and 4-positions leads
to isotope effects that are similar in size, with the effect at the
4-positions perhaps just slightly larger. The computational data
(Table 4) lead to the same conclusion.
added to that of the carbon atom to which it is attached. In
any case this has the advantage that one less arbitrary boundary
has to be assumed (that between the electron clouds of a carbon
atom and the attached hydrogen).
The data in Table 5 give rise to those in Table 6, which
presents differences in the charge densities between the pyrene
anion radical and the neutral pyrene for each of the three
formalisms. The charge differences are presumably slightly
more reliable as charge-indicators because of a partial cancel-
lation of errors.
Discussion
The data in Table 3 show a gradual increase in ∆EA with
increasing deuteriation of pyrene and that the CV and EPR data
are within experimental error of being collinear (Figure 7). The
smallest value for ∆EA (Keq closest to 1.00) is observed for
2,7-d2-pyrene. For vibrationally uncoupled L atoms (L ) H,
D), the measured values of Keq are expected to follow the
isotopic product rule.17 The rule applies reasonably well for
the 2,4,5,7,9,10-d6 pyrene (1f) system. The product of Keq(EPR,1b)
× Keq(EPR,1d) ) 0.50, which is slightly smaller than the
experimentally observed 0.57 for Keq(EPR,1f). For the electro-
chemical experiments, Keq(CV,1b) × Keq(CV,1d) ) 0.83, which
agrees well with 0.80 measured for Keq(CV,1f). The deviation
from the product rule is somewhat larger for isotopic isomers
in which both the 1- and 2-positions are deuteriated. Deviations
from the isotopic product rule are to be expected because of an
accumulation of factors such as imperfect EPR simulations of
mixtures, incomplete knowledge regarding isotopic impurities,
and vibrational coupling between L atoms at adjacent carbon
atoms. We estimate that all of these to lead to about (10%
uncertainty in Keq values. Propagated ranges in product Keq
values overlap those of averaged, measured Keq values, and the
isotopic product rule17 holds reasonably well.
For comparison of the experimental and theoretical (multi-
deuteriated) data, the ∆G°’s (∆G° ) ∆EA) for the electron
transfer reaction (at 173 K) from the ROHF data are graphically
presented in Figure 8. The AUHF data are almost identical to
the ROHF data and will not be discussed separately.
The plot shown in Figure 8 and the data in Table 3 lead to
two conclusions. First, a correlation between the experimental
and calculated values is observed. This means that the effects
of isotopic substitution on the one-electron-accepting ability of
pyrene are reflected well by the PM3 calculations, irrespective
of the method (ROHF or UHF) used. The underlying assump-
tion that ∆EA is an intrinsic property of the pyrene/pyrene
radical anion couple, and dominated by ZPE differences,
therefore passes the experimental test. Second, Figure 8 shows
that the calculations significantly underestimate the values of
∆EA (the slopes are not unity). However, the observed linear
correlation means that theoretically obtained data can be usefully
compared to the experimental data since the relevant properties
are underestimated theoretically by a constant proportion.
Assuming this rule holds, one can predict the experimental
Keq value for monodeuteriation on each of the three distinct
positions in the molecule. For compounds 1b-1d these values
follow from taking the square root (1b) or fourth root (1c and
The calculations, of course, do not attempt to account for
the effects of solvation or ion association. Thus, in order for
the slopes of the lines shown in Figure 8 to be unity, the

Electron Affinities of Deuteriated Pyrenes
J. Phys. Chem., Vol. 100, No. 9, 1996 3459
By use of the hydrogen-summed-into-carbon charges, the charge
on C2 stays about constant upon reduction of pyrene. Although
the isotope effect of monodeuteriation is largest at the 1-posi-
tions, which carry the highest negative charge, such a correlation
is not evident for the isotope effects at the other two positions.
The negative charge on C4 is calculated to be substantially larger
than that on C2, but the isotope effect per deuterium at those
two positions is about equal according to results from both
theory and experiment. According to the experimental and
ROHF data ∆EA(2-d1) e ∆EA(4-d1), while the UHF results
predict ∆EA(2-d1) g ∆EA(4-d1). It is therefore concluded that
no evidence for a general relation between ∆EA and the charge
on the carbon atom at which the deuterium atom resides is
observed.
Conclusions
The experimental (electrochemical and EPR) and theoretical
(PM3) data on the isotope effect on the electron affinity for a
series of partially deuteriated pyrene isotopic isomers show a
gradual increase in ∆EA with increasing deuterium substitution.
The correlation between the electrochemical data and the EPR
data is excellent. The experimental and theoretical data correlate
well, which is strong evidence for the applicability of the
theoretical method used. No evidence is found for a general
relation between the value of ∆EA and the charge on the carbon
atom that carries a hydrogen or deuterium atom. The nonunity
slopes of the lines shown in Figures 7 and 8, as well as MOPAC
7 calculations, suggest that ion association and ion solvation
are slightly different for the deuteriated and nondeuteriated anion
radicals of pyrene.
Figure 8. Plots of the experimental (EPR, upper, and electrochemical,
lower) free energy changes for the electron transfer (∆EA) from the
pyrene anion radical to the various deuteriated pyrenes vs the ROHF
calculated values. The lines fit the equations ∆EAEPR ) 4.8∆EAROHF
+ 49 and ∆EAELECT ) 3.0∆EAROHF + 18. The correlation constants
are 0.96 (EPR) and 0.97 (ELECT).
enthalpies (and free energies) of solvation of the products and
the reactants of reaction 1 must be identical. The enthalpy of
solvation of the pyrene anion radical plus K+ in THF (∆H°sol
for reaction 4) is -765 kJ/mol.18
Experimental Section
Synthesis. Six newly described isotopic isomers of pyrene
were synthesized, namely 2,7-di- (1b), 1,3,6,8-tetra- (1c),
4,5,9,10-tetra- (1d), 1,2,3,6,7,8-hexa- (1e), 2,4,5,7,9,10-hexa-
(1f), and 1,3,4,5,6,8,9,10-octadeuteriopyrene (1g). The strategy
used was that of repeated acidic exchange of the arene hydrogens
or basic exchange of the benzylic hydrogens in 4,5,9,10-
tetrahydropyrene (2) and 1,2,3,6,7,8-hexahydropyrene (3) in
deuteriated solvents. These procedures followed by dehydro-
(4)
If deuteriation perturbed the enthalpy for reaction 4 by as little
as several hundred J/mol (or a few hundredths of 1%), it would
be sufficient to result in the slopes of the lines shown in Figure
8 deviating from unity to the degree that they do. Indeed,
preliminary MO calculations using the program MOPAC 7
indicate that this is the case and yield a difference in the
solvation of the pyrene and perdeuteriated pyrene anion radicals
in DMF of about 1 kJ/mol. Furthermore, the fact that the slope
of the line shown in Figure 7 is only 0.6 probably reflects the
fact that these ion association and solvation effects are greater
in THF with K+ serving as the counterion than in DMF with
(C4H9)4N+.
The hypothesis that a relation exists between the observed
equilibrium isotope effect and the charge developed on the
hydrogen/deuterium substituted carbon atom was tested by
comparing the experimental results to the calculated atomic
charges in Table 5. All three formalisms used indicate the order
of negative charge on the L-atom-substituted carbon atoms in
the radical anion to be C1 > C4 > C2, irrespective of taking
the calculated charges or the light-summed-in-heavy charges
as charge-indicators. The charge differences (Table 6) also show
the same order: the development of negative charge is largest
on C1, and the next largest is on C4. For C2, development of
a slightly less negative charge is calculated with all methods.
2
3
genation yielded 1d, 1e, and 1g. Selective replacement of
deuteriums by hydrogens via a two-step procedure of bromin-
ation and subsequent debromination yielded 1b, 1c, and 1f.
Deuterium exchange reactions in acidic media on polycyclic
aromatic hydrocarbons are often performed using deuteriotri-
fluoroacetic acid19 or concentrated deuteriosulfuric acid.20 For
preparative purposes it appeared advantageous to use another
deuterium-exchange medium: deuteriated poly(phosphoric acid)
(PPA-d).21 The advantages of PPA-d as a deuterium-donating
agent are the following: (1) it can be easily prepared, at a large
scale, from D2O and P2O5, which are relatively cheap materials;
(2) it can be heated to relatively high temperatures (>200 °C),
which is useful in slow exchange reactions; (3) it is not an
oxidizing agent and does not enter into aromatic substitutions.
Deuterium oxide (99.9% D) and DMSO-d6 (99.5% D) were
purchased from Aldrich. Pyrene (99%) was obtained from
Janssen Chimica. 1,2,3,6,7,8-Hexahydropyrene (3) was pre-
pared by reduction of pyrene with Na in refluxing isoamyl

3460 J. Phys. Chem., Vol. 100, No. 9, 1996
Hammerich et al.
alcohol,22 and 4,5,9,10-tetrahydropyrene (2) was prepared
analogously from 4,5-dihydropyrene in refluxing toluene.23 4,5-
Dihydropyrene was purified by column chromatography prior
to reaction, in order to remove a small amount of pyrene. In
this way the formation of 3 during the reduction was avoided.
All solvents were distilled before use and dried if necessary.
Silica gel (230-400 mesh) was obtained from Merck and used
in all column chromatographic purifications. Petroleum ether
with a boiling range 40-60 °C was used as an eluant.
57%). 1H and 2H NMR show a deuterium incorporation of 98%
at positions 2, 4, 5, 7, 9, and 10 and of 5% at positions 1, 3, 6,
and 8. IR (KBr): 1580, 1433, 1018, 889, 675 cm-1. MS (m/
z): M+ ) 208.1196 (C16H4D6 requires 208.1159).
1,2,3,6,7,8-Hexadeuterio-4,5,9,10-tetrahydropyrene (2e).
4,5,9,10-Tetrahydropyrene (2a) (10.0 g; 48.4 mmol) was deu-
teriated 3 times with PPA-d (90 g) in the manner described for
pyrene. The reaction was carried out at 160 °C for 3 h. After
purification (silica; petroleum ether) 2e was obtained as white
plates (6.5 g; 63%). According to NMR, the deuterium
incorporation at the aromatic positions is 98%. On the benzylic
positions approximately 10% deuterium is incorporated. IR
(KBr): 2915, 2880, 2820, 2245, 1427, 1377, 1210, 1194, 833,
The 300 MHz 1H NMR and 46.9 MHz 2H NMR spectra were
1
recorded on a Bruker WM-300 spectrometer. In the H NMR
spectra TMS was used as an internal standard (0 ppm) and
CDCl3 (99.8% D) as a solvent. Concentrations were typically
2 mg/mL. In the 2H NMR spectra CDCl3 was used as a standard
(7.26 ppm) and CHCl3 or CCl4 as a solvent. Concentrations
were typically 40 mg/mL. Deuterium incorporations were
calculated by intergration of the proton and deuterium signals.
IR spectra were recorded on a Pye-Unicam SP3-200. Accurate
molecular masses were determined by electron impact mass
spectrometry using a V. G. Micromass ZAB-HFqQ mass
spectrometer coupled to a V. G. 11/250 data system. The
samples were introduced via the direct insertion probe into the
ion source.
810 cm-1
212.1472).
.
MS (m/z): M+ ) 212.1449 (C16H8D6 requires
1,2,3,6,7,8-Hexadeuteriopyrene (1e). To a solution of 2e
(3.10 g; 14.6 mmol) in dry toluene (100 mL) DDQ (7.4 g; 32.8
mmol) was added. The reaction mixture was refluxed under
an argon atmosphere for 2 h. After the mixture had cooled to
room temperature it was filtered over Hyflo. The organic layer
was washed twice with a solution of Na2CO3 and dried over
MgSO4; the solvent was evaporated in vacuo. Column chro-
matography yielded 1e as white crystals (2.65 g; 87%). 1H and
2H NMR give a deuterium incorporation at positions 1, 2, 3, 6,
7, and 8 of 97% and at positions 4, 5, 9, and 10 of about 10%.
IR (KBr): 1560, 1290, 1154, 832, 780, 694 cm-1. MS (m/z):
M+ ) 208.1136 (C16H4D6 requires 208.1159).
2,7-Dideuteriopyrene (1b). Compound 1e (1.80 g; 8.6 mmol)
was brominated and debrominated in the same way as outlined
for perdeuteriopyrene. The yield of 1b after purification was
960 mg (54%). Deuterium incorporation according to NMR is
97% at positions 2 and 7, 2% at positions 1, 3, 6, and 8, and
about 10% at positions 4, 5, 9, and 10. IR (KBr): 1584, 1307,
1323, 1148, 1010, 895, 886, 810, 708, 678, 660 cm-1. MS
(m/z): M+ ) 204.0931 (C16H8D2 requires 204.0908).
Deuteriated PPA was prepared by adding P2O5 (500 g) in
small portions to D2O (100 g) under mechanical stirring. An
alcohol/dry ice bath was used to keep the temperature below
50 °C. When the addition was complete, the coolant was
removed and the reaction mixture stirred for 2 h at 180 °C to
dissolve all remaining P2O5. The prepared amount of PPA-d
was transferred into six 50 mL round-bottom flasks while still
warm. A stirrer bar was added to each flask, and the flasks
were closed, allowed to cool, sealed, and stored.
Perdeuteriopyrene (1h). Perdeuteriopyrene was synthesized
by portionwise addition of pyrene (3.06 g; 15.0 mmol) to a
round-bottom flask containing approximately 90 g of PPA-d
under vigorous stirring at 190-200 °C. After 4 h of stirring
the reaction mixture was allowed to cool to 100 °C, and the
dark mixture was poured over ice (500 g) and extracted twice
with CH2Cl2. The combined organic layers were washed with
a solution of Na2CO3 and dried (MgSO4). The solvent was
evaporated. After the first and second exchange cycle purifica-
tion was confined to a rapid cleanup of the brown product over
a small amount of silica gel (elution with petroleum ether). The
final, third exchange cycle was followed by regular column
chromatography and yielded 2.03 g (60%) of 1h. Deuterium
incorporation at each position as determined by NMR and mass
spectrometry (MS) was >98%. IR (KBr): 1555, 1422, 1331,
1275, 1037, 946, 810, 743 cm-1. MS (m/z): M+ ) 212.1448
(C16D10 requires 212.1410).
1,3,6,8-Tetradeuterio-4,5,9,10-tetrahydropyrene (2c). Com-
pound 2e (2.06 g; 9.7 mmol) was treated with Br2 (3.5 g; 22
mmol) and FeCl3‚6H2O (270 mg; 1.0 mmol) in water (75 mL).25
After recrystallization (cyclohexane) 2,7-dibromo-1,3,6,8-tet-
radeuterio-4,5,9,10-tetrahydropyrene was obtained as white
crystals in 1.45g (40%) yield. A 1.40 g (3.8 mmol) sample of
this compound was reacted with LiAlH4 (320 mg; 8.1 mmol)
and TiCl4 (380 mg; 2.0 mmol) in THF (75 mL) as described
for the preparation of 1f. The mixture was refluxed for 3 h to
complete the reduction. Compound 2c was obtained in 660 mg
(83%) yield. It was dissolved in DMSO (6 mL) and added to
a NaH suspension (500 mg, 50% in paraffin oil, twice washed
with 10 mL of cyclohexane). The reaction mixture was heated
to 90-95 °C at which temperature evolution of H2 took place.
After 2 h the dark-brown suspension was allowed to cool to
room temperature. Water was added, and the mixture was
extracted twice with CH2Cl2. The combined organic layers were
washed with water and dried over MgSO4; the solvent was
evaporated in vacuo. Purification by column chromatography
yielded 2c as white crystals (590 mg; 90%). Deuterium
incorporation (NMR) was 95% at positions 1, 3, 6, and 8 and
e1% at positions 2, 4, 5, 7, 9, and 10. IR (KBr): 2915, 2880,
2,4,5,7,9,10-Hexadeuteriopyrene (1f). Perdeuteriopyrene (0.98
g; 4.6 mmol) was brominated with Br2 (3.3 g; 20.6 mmol) in
nitrobenzene (40 mL).24 1,3,6,8-Tetrabromo-2,4,5,7,9,10-hexa-
deuteriopyrene (2.24 g; 100%) was collected as a light-green
powder and was not purified further. To this solid, tetrahy-
drofuran (100 mL) was added, and the resulting suspension was
cooled with an ice-bath. LiAlH4 (700 mg; 18.4 mmol) was
added under stirring. TiCl4 (900 mg; 4.7 mmol) was cautiously
added, whereupon the reaction mixture turned dark-brown
immediately. After 30 min the coolant was removed, and
stirring was continued for another 90 min at room temperature.
The reaction mixture was poured on ice and extracted twice
with diethyl ether. The combined organic layers were dried
(MgSO4), and the solvent was evaporated. The resulting yellow
residue was purified with column chromatography. 2,4,5,7,9,10-
Hexadeuteropyrene (1f) was isolated as white crystals (550 mg;
2825, 2245, 1426, 1411, 1294, 908 cm-1. MS (m/z): M+
210.1379 (C16D10H4 requires 210.1347).
)
1,3,6,8-Tetradeuteriopyrene (1d). Compound 2c (580 mg;
2.75 mmol) was dehydrogenated with DDQ (1.36 g; 6.0 mmol)
in refluxing toluene as described for the preparation of 1e. The
yield after purification was 425 mg (77%). Deuterium incor-
poration (NMR) was 95% at positions 1, 3, 6, and 8 and e1%
at positions 2, 4, 5, 7, 9, and 10. IR (KBr): 1586, 1385, 1282,

Electron Affinities of Deuteriated Pyrenes
J. Phys. Chem., Vol. 100, No. 9, 1996 3461
1193, 1148, 959, 910, 827, 720 cm-1
206.1073 (C16H6D4 requires 206.1034).
.
MS (m/z): M+
)
potentiostat was a PAR Model 173/276 driven by an HP 3314A
function generator. The function generator was activated
periodically (in this study every 30 s) by a specially designed
trigger operated in a mode that allowed every second measure-
ment in a series to be delayed by 10 ms corresponding to a
half-period of the 50 Hz sine wave of the power line.30 In this
way most of the noise from the power line in one measurement
will be out of phase with the noise in the previous measurement
and thus cancels when the average of these two measurements
is taken. In addition, the high-frequency noise was reduced by
filtering the output from the potentiostat by means of a PAR
Model 189 selective amplifier. The current-voltage curves
were recorded in the current-time mode on a 12-bit Nicolet
oscilloscope, Model 2090-3C/206-2, with a horizontal resolution
of 0.5 mV/point. The temperature was continuously monitored
by a Philips PM2519 automatic multimeter equipped with a
temperature probe. The function generator, the oscilloscope,
and the multimeter were interfaced to an HP 9826A desk
computer equipped with an HP 98635A floating point accelera-
tor and an HP 9154A hard disk. The software used for
instrument control and data-handling procedures (see below)
was written in HP Basic 3.0.
4,5,9,10-Tetradeuterio-1,2,3,6,7,8-hexahydropyrene (3d).
1,2,3,6,7,8,-Hexahydropyrene (3) (5.9 g; 28.4 mmol) was
deuteriated twice with PPA-d (90g) as described for pyrene (160
°C, 3h). Compound 3d was collected after column chroma-
tography as white plates (4.7 g; 78%). Deuterium incorporation
(NMR) was 97% at positions 4, 5, 9, and 10, 12% at positions
1, 3, 6, and 8, and 9% at positions 2 and 7. IR (KBr): 2900,
2840, 2195, 1560, 1441, 1380, 1326, 1291, 1256, 1060, 896,
910 cm-1. MS (m/z): M+ ) 212.1537 (C16H12D4 requires
212.1503).
1,1′,3,3′,4,5,6,6′,8,8′,9,10-Dodecadeuterio-1,2,3,6,7,8-hexahy-
dropyrene (3g). Compound 3d (3.0 g; 14.2 mmol) was
deuteriated twice with DMSO-d6 (20 g) and NaH (3.0 g, 50%
dispersion in paraffin oil) for 20 h.26 Column chromatography
yielded 2.10 g (67%) of pure 3g. Deuterium incorporation
(NMR) was 97% at positions 1, 3, 4, 5, 6, 8, 9, and 10 and
12% at positions 2 and 7. IR (KBr): 2895, 2840, 2240, 2190,
2090, 1553, 1450, 1382, 1317, 1171, 1053, 902, 843, 644 cm-1
.
MS (m/z): M+ ) 220.2036 (C16H4D12 requires 220.2005).
1,3,4,5,6,8,9,10-Octadeuteriopyrene (1g). Compound 3g
(1.95 g; 8.9 mmol) was dehydrogenated with DDQ (6.3 g; 27.9
mmol) in refluxing toluene as described for 1e, yielding after
column chromatography 1g (1.56 g; 83%) as white crystals.
Deuterium incorporation (NMR) was 97% at positions 1, 3, 6,
and 8, 95% at positions 4, 5, 9, and 10, and 12% at positions 2
and 7. IR (KBr): 1560, 1278, 922, 840, 789 cm-1. MS (m/z):
M+ ) 210.1261 (C16H2D8 requires 210.1285).
The pyrene (0.02 mmol) was added to the voltammetry cell
and dissolved in 10 mL of DMF containing Bu4NPF6 (0.1 M)
by purging with nitrogen for at least 10 min. This also served
the purpose of removing dissolved oxygen. The same solvent-
supporting electrolyte stock solution was used for all eight
pyrenes.
The undeuteriated pyrene (1a) served as an external standard
during the voltammetric measurements, which were performed
in an AA-BB-AA-BB-AA-BB-AA-CC-AA-CC-AA-
CC-...-AA-HH-AA-HH-AA-HH-AA sequence, where
each single letter refers to a series of measurements carried out
as described below: A on the pyrene solution and B-H on the
solutions of the seven deuteriated pyrenes (1b-1h). Thus, two
series of measurements were carried out on each solution after
which the electrode assembly was transferred to the next solution
in the order described.
4,5,9,10-Tetradeuteriopyrene (1d). Compound 1g (950 mg;
4.5 mmol) was subjected to the bromination/debromination
sequence as outlined for the preparation of 1f. Compound 1d
was obtained after purification in 435 mg (46%) yield.
Deuterium incorporation (NMR) was 95% at positions 4, 5, 9,
and 10, 3% at positions 1, 3, 6, and 8, and 9% at positions 2
and 7. IR (KBr): 1585, 1428, 1157, 890, 804, 725, 675 cm-1
.
MS (m/z): M+ ) 206.1028 (C16H6D4 requires 206.1034).
EPR Measurements. EPR measurements were carried out
as previously described on an IBM (Bruker) ER-200D spec-
trometer equipped with an IBM variable temperature unit11f and
simulated.27 It is important to note that the Keq’s determined
via the EPR measurements lie within the experimental error
reported only if less than 10% of the pyrenes are reduced in
the partial reductions, that is, when [*A•-] and [A•-] , [*A]0
and [A]0, where [*A]0 and [A]0 represent the total concentrations
of deuteriated pyrene and pyrene used in the reaction. The ratio
[A*]0/[A]0 was obtained by careful weighing of the two
materials. Each reported Keq is an average based upon three to
six such isotopic ratios.
Electrochemistry. N,N-Dimethylformamide (LAB-SCAN,
HPLC grade) was distilled under reduced pressure prior to use.
Tetrabutylammonium hexafluorophosphate was prepared by a
procedure analogous to that described earlier for the tetrafluo-
roborate salt.28
The voltammetry cell was a cylindrical tube (50 mL) with a
NS29 joint to accommodate a plastic holder for the three
electrodes and the nitrogen inlet. The working electrode was
prepared by sealing a 0.6 mm platinum wire in glass and
polishing to obtain a planar surface. The electrode was
electrochemically covered with mercury by serving as a cathode
in a solution of mercuric nitrate in dilute aqueous nitric acid.
The reference electrode, consisting of an Ag-wire in DMF
containing Bu4NPF6 (0.1 M) was prepared as described by
Moe29 and stored until the potential was constant. The
counterelectrode was a platinum wire sealed in soft glass. The
A series of measurements consisted of 10 current-time
curves, each of which included an 800 mV negatively swept
linear potential scan (10 V/s) initiated approximately 600 mV
before the peak potential for the reduction of pyrene. The rate
of the scan back to the start potential was 10 times that of the
forward scan, i.e., 100 V/s, in order to diminish the time needed
to record a single curve. The current-time curves were
recorded at the oscilloscope and then transferred to the computer.
Averages were taken for every two current-time curves (see
above), resulting in five averaged curves: 1 and 2; 3 and 4; 5
and 6; 7 and 8; 9 and 10. The rest potential in the time interval
between two measurements (ca. 30 s) was equal to the start
potential.
Each of the averaged curves was processed as follows. First,
an equation for the base line was estimated by linear regression
of the data points recorded well in front of the rising part of
the reduction wave. The base line was extrapolated to the
potential region of the reduction peak and subtracted from the
current-time curve. An approximate position of the peak
located by the operator was used to define a (20 mV data region
around this peak, and the fitting of these data to a third-order
polynomial followed by application of a maximizing procedure
was then used to determine the peak current, Ip, and the peak
potential, Ep. Linear regression of the data points in a (15
mV region of the curve around the point corresponding most
closely to I ) (1/2)Ip was finally used to determine the half-
peak potential, Ep/2, as the value of E for which I ) (1/2)Ip.

3462 J. Phys. Chem., Vol. 100, No. 9, 1996
Hammerich et al.
(9) Marx, D.; Kleinhesselink, D.; Wolfsberg, M. J. Am. Chem. Soc.
1989, 111, 1493.
(10) Zuilhof, H.; Lodder, G. J. Phys. Chem. 1992, 96, 6957.
(11) (a) Allendoerfer, R. D.; Papez, R. J. J. Phys. Chem. 1972, 76, 1012.
(b) Lawler, R. G.; Tabit, C. T. J. Am. Chem. Soc. 1969, 91, 5671. (c) Hirota,
N. J. Phys. Chem. 1967, 71, 127. (d) Stevenson, G. R.; Alegria, A. E. J.
Phys. Chem. 1973, 77, 3100. (e) Kotake, Y.; Jansen, E. G. J. Am. Chem.
Soc. 1989, 111, 5138. (f) Stevenson, C. D.; Halverson, T. D.; Reiter, R. C.
J. Am. Chem. Soc. 1993, 115, 12405.
The entire measurement sequence gave rise to a set of six
potential differences for each deuteriated pyrene, and the values
given in Table 3 refer to the averages of these six values. The
standard deviations were between 0.2 and 0.4 mV. The half-
peak widths, Ep - Ep/2, were in most cases between 57.8 and
58.4 mV, typical for aromatic hydrocarbons at a scan rate of
10 V/s.
Calculations. The MO calculations were performed using
the programs VAMP 4.30 (RHF, ROHF, and AUHF) and
MOPAC 6 (UHF). Computation details have been previously
published.10
(12) (a) Stewart, J. J. P. J. Comput. Chem. 1989, 10, 209. (b) Stewart,
J. J. P. J. Comput. Chem. 1989, 10, 221.
(13) The use of this technique for the heavy isotopically substituted
systems has been criticized and is controversial; see ref 9.
(14) (a) Hurley, A. C. Introduction to the Electron Theory of Small
Molecules; Academic Press: New York, 1976; pp 242-253. (b) Dewar,
M. J. S. The Molecular Orbital Theory of Organic Chemistry; McGraw-
Hill: New York, 1969; pp 250-277.
(15) Bachrach, S. M. ReV. Comput. Chem. 1994, 5, 171.
(16) Zuilhof, H.; Lodder, G. J. Phys. Chem. 1995, 99, 8033.
(17) Bigeleisen, J. J. Chem. Phys. 1955, 23, 2264.
(18) Stevenson, G. R.; Schock, L. E.; Reiter, R. C. J. Phys. Chem. 1983,
87, 4004.
(19) Mackor, E. L.; Hofstra, A.; Van der Waals, J. H. Trans. Faraday
Soc. 1957, 53, 1309.
(20) Renaud, R. N.; Kovachic, D.; Leitch, L. C. Can. J. Chem. 1961,
39, 21.
(21) Gold, V.; Satchell, D. N. P. J. Chem. Soc. 1955, 3622.
Acknowledgment. The authors are grateful to R. Fokkens
and H. Peeters, University of Amsterdam, J. van Houte and R.
A. M. van der Hoeven, Leiden University, for the determination
of the high-resolution mass spectrometric data and deuterium
incorporations, and A. W. M. Lefeber, Leiden University, for
the 2H NMR measurements. C.D.S. thanks the National Science
Foundation (Grant CHE-9412036) for support of this work. The
electrochemical work was supported in part by The Danish
Natural Science Research Council (Grant No. 11-0063). O.H.
and M.F.N. gratefully acknowledge the receipt of funds from
The Carlsberg Foundation and The Danish Natural Science
Research Council for the purchase of the instrumentation.
(22) Van den Braken-Van Leersum, A. M.; Tintel, C.; Van’t Zelfde,
M.; Cornelisse, J.; Lugtenburg, J. Recl. TraV. Chim. Pays-Bas, 1987, 106,
120.
(23) Rodenburg, L.; Floor, M.; Lefeber, A.; Cornelisse, J.; Lugtenburg,
J. Recl. TraV. Chim. Pays-Bas, 1988, 107, 1.
References and Notes
(1) (a) Stevenson, G. R.; Espe, M. P.; Reiter, R. C. J. Am. Chem. Soc.
1986, 108, 532. (b) Morris, D. E.; Smith, W. H. J. Electrochem. Soc. 1991,
138, 1351. (c) Guthrie, R. D.; Shi, B. J. Am. Chem. Soc. 1990, 112, 3136.
(d) Tanko, J. M.; Drumright, R. E. J. Am. Chem. Soc. 1992, 114, 1844. (e)
Wen, X. L.; Liu, Z. L.; Liu, J. M.; Liu, Y. C. J. Chem. Soc., Faraday
Trans. 1992, 88, 3323.
(2) Stevenson, G. R.; Sturgeon, B. E.; Vines, K. S.; Peters, S. J. J.
Phys. Chem. 1988, 92, 6850.
(3) Stevenson, G. R.; Reiter, R. C.; Espe, M. P.; Bartmess, J. E. J.
Am. Chem. Soc. 1987, 109, 3847.
(24) Volmann, H.; Becker, H.; Corell, M.; Streeck, H. Justus Liebigs
Ann. Chem. 1937, 531, 1-159, synthesis on p 81.
(25) Lee, H.; Harvey, R. G. J. Org. Chem. 1986, 51, 2847.
(26) Chen, T.-S.; Wolinska-Macydiarz, J.; Leitch, L. C. J. Labelled
Compd. 1970, 6, 285.
(27) (a) Morse, P. D. EPRWare User Manual; Scientific Software
Services: Bloomington, Il., 1990. (b) Morse, P. D.; Reiter, R. C. EWSim
User Manual; Scientific Software Services: Bloomington, Il., 1990.
(28) Nielsen, M. F.; Hammerich, O.; Parker, V. D. Acta Chem. Scand.,
Ser. B 1986, 40, 101.
(29) Moe, N. S. Anal. Chem. 1974, 46, 968.
(4) Stevenson, G. R.; Wehrmann, G. C., Jr.; Reiter, R. C. J. Phys. Chem.
1991, 95, 6936, and references therein.
(5) Goodnow, T. T.; Kaifer, A. E. J. Phys. Chem. 1990, 94, 7682.
(6) Stevenson, G. R.; Reidy, K. A.; Peters, S. J.; Reiter, R. C. J. Am.
Chem. Soc. 1989, 111, 6578.
(7) Stevenson, G. R.; Sturgeon, B. E. J. Org. Chem. 1990, 55, 4090.
(8) Stevenson, G. R.; Wehrmann, G. C., Jr.; Reiter, R. C. J. Phys. Chem.
1991, 95, 901.
(30) Nielsen, M. F.; Laursen, S. Aa.; Hammerich, O. Acta Chem. Scand.
1990, 44, 932.
(31) Zuilhof, H.; Lodder, G.; van Mill, R. P.; Mulder, P. P. J.; Kage,
D.; Reiter, R. C.; Stevenson C. D. J. Phys. Chem. 1995, 99, 3461.
JP952098N