Transient resonance Raman and density functional theory investigation of 4-methoxyphenylnitrenium and 4-ethoxyphenylnitrenium ions

Article abstract of DOI:10.1021/jp0224261

This paper presents a transient resonance Raman and density functional theory study of 4-methoxyphenylnitrenium and 4-ethoxyphenylnitrenium ions in a largely aqueous environment. The transient Raman bands observed in conjunction with the (U)BPW91/cc-PVDZ

Full text of DOI:10.1021/jp0224261

J. Phys. Chem. A 2003, 107, 8067-8074
8067
Transient Resonance Raman and Density Functional Theory Investigation of
4-Methoxyphenylnitrenium and 4-Ethoxyphenylnitrenium Ions†
Pik Ying Chan, Shing Yau Ong, Peizhi Zhu, Cunyuan Zhao, and David Lee Phillips*
Department of Chemistry, The UniVersity of Hong Kong, Pokfulam Road, Hong Kong S.A.R., P. R. China
ReceiVed: NoVember 13, 2002; In Final Form: February 25, 2003
This paper presents a transient resonance Raman and density functional theory study of 4-methoxyphenylni-
trenium and 4-ethoxyphenylnitrenium ions in a largely aqueous environment. The transient Raman bands
observed in conjunction with the (U)BPW91/cc-PVDZ calculation results indicates these nitrenium ions are
singlet state species. Both the 4-methoxyphenylnitrenium and the 4-ethoxyphenylnitrenium ions have significant
iminocyclohexadienyl character to a degree very similar to that previously found for the 2-fluorenylnitrenium
ion. The 4-methoxyphenylnitrenium and the 4-ethoxyphenylnitrenium ions also have significant oxocarbocation
character and this influences both their properties and chemical reactivity. We compare the properties of the
4-methoxyphenylnitrenium and the 4-ethoxyphenylnitrenium ions to those previously found for other
arylnitrenium ions.
Introduction
of bands for both nitrenium ions in a largely aqueous solution
at room temperature. These Raman bands have vibrational
Arylnitrenium ions are believed to be key intermediates in
chemical carcinogenesis and there has been a great deal of
interest in their properties and chemical reactions.1-18 Aromatic
amines (such as 2-acetyl-aminofluorene) can be enzymatically
converted into sulfate esters of the analogous N-hydroxylamines.
These types of sulfate esters in aqueous media will spontane-
ously form an arylnitrenium ion and a sulfate anion.13,17
Arylnitrenium ions such as the 2-fluorenylnitrenium ion can be
selectively trapped by guanine bases in DNA and form adducts.
These types of reactions are thought to result in carcinogenic
mutations.8-10,14,18 Arylnitrenium ions are reactive short-lived
species that are difficult to study. Several photochemical
methods have been developed to enable laser spectroscopy
techniques to be used to directly probe arylnitrenium ions in
room-temperature solutions.19-33 The lifetimes of arylnitrenium
ions and their rate constants for reactions with other molecules
have been measured using time-resolved transient absorption
experiments.19-27,29,30 Recently, time-resolved infrared (TRIR)
absorption spectroscopy and time-resolved resonance Raman
(TR3) spectroscopy have been used to investigate the structure
and properties of several arylnitrenium ions.28,31-33 These studies
showed that the arylnitrenium ions have significant iminocy-
clohexadienyl character to varying degrees depending on their
structure and substituents.28,31-33 Several alkoxyphenylnitrenium
ions including 4-methoxyphenylnitrenium and 4-ethoxyphen-
ylnitrenium ions have been examined using time-resolved
transient absorption spectroscopy and found to have relatively
long lifetimes on the order of a millisecond.34,35 We note that
the 4-ethoxyphenylnitrenium ion is of particular interest because
it would be formed from the carcinogen phenacetin that had
found use as an analegesic until it was banned for use.
frequencies that agree noticeably better with the (U)BPW91/
cc-PVDZ computed vibrational frequencies for the singlet state
than the triplet state for the 4-methoxyphenylnitrenium and the
4-ethoxyphenylnitrenium ions. This and the relatively large
singlet-triplet energy gap from the (U)BPW91/cc-PVDZ
calculations indicates these nitrenium ions are singlet state
species. Both 4-methoxyphenylnitrenium and the 4-ethoxyphen-
ylnitrenium ions display significant iminocyclohexadienyl char-
acter. We compare the properties of the 4-methoxyphenylni-
trenium and the 4-ethoxyphenylnitrenium ions to those previ-
ously found for other arylnitrenium ions characterized by time-
resolved vibrational spectroscopy.
Experimental and Computational Section
4-Methoxyphenyl azide and 4-ethoxyphenyl azide were
prepared following a previously reported method for the
synthesis of azide compounds.36,37 The 4-methoxyphenyl azide
was prepared as follows. In a 200 mL round-bottom flask,
4-methoxyaniline (4.92 g, 40 mmol) was dissolved in water (10
mL) that contained concentrated HCl (20 mL). This solution
was cooled to below 5 °C in an ice bath and diazotized with a
solution of NaNO2 (4.14 g, 60 mmol) with distilled water (10
mL). The mixture was stirred in an ice bath for 1 h. A solution
of NaN3 (5.2 g, 80 mmol) in water (30 mL) was added at 0 °C
and stirred for 30 min. Then, the mixture was allowed to warm
to room temperature and stirred for an additional hour. The
resulting suspension of solid was filtered to isolate product (∼3.5
g). The 4-methoxyphenyl azide was characterized as follows:
1H NMR (400 MHz, CD3CN): δ 7.01 (d, J ) 8.9 Hz, 2H), δ
6.94 (d, J ) 8.9 Hz, 2H), δ 3.77 (s, 3H): IR (film): 2106 cm-1
:
MS (EI): m/z: 149 [C7H7N3O+], 121 [C7H7NO+]; UV (CH3-
CN): λ max ) 256 nm.
In this paper, we report the first transient resonance Raman
spectra for the 4-methoxyphenylnitrenium and the 4-ethoxyphen-
ylnitrenium ions. The transient Raman spectra display a number
4-Ethoxyphenyl azide was prepared as follows. In a 200 mL
round-bottom flask, 4-ethoxyaniline (5.15 mL, 40 mmol) was
dissolved in water (10 mL) that contained concentrated HCl
(20 mL). This solution was cooled to below 5 °C in an ice bath
and diazotized with a solution of NaNO2 (4.14 g, 60 mmol)
† Part of the special issue “A. C. Albrecht Memorial Issue”.
* To whom correspondence should be addressed. E-mail: phillips@
hkucc.hku.hk. Fax: 852-2857-1586.
10.1021/jp0224261 CCC: $25.00 © 2003 American Chemical Society
Published on Web 04/17/2003

8068 J. Phys. Chem. A, Vol. 107, No. 40, 2003
Chan et al.
with distilled water (10 mL). The mixture was stirred in an ice
bath for 1 h. A solution of NaN3 (5.2 g, 80 mmol) in water (30
mL) was added at 0 °C and stirred for 30min. Then, the mixture
was allowed to warm to room temperature and stirred for an
additional hour. Anhydrous potassium hydrocarbonate is added
to neutralize the mixture. The product is then extracted by
diethyl ether (∼ 3 mL) and characterized as follows: 1H NMR
(400 MHz, CD3CN): δ 6.97 (d, J ) 9.1 Hz, 2H), δ 6.90 (d, J
) 9.1 Hz, 2H), δ 3.99 (q, J ) 7.0 Hz, 2H), δ 1.34 (t, J ) 7.0
Hz, 3H): IR (liquid): 2109 cm-1
: MS (EI): m/z: 163
[C8H9N3O+], 135 [C8H9NO+]; UV (CH3CN): λ max ) 256
nm.
Samples of the precursor compound (4-methoxyphenyl azide
or 4-ethoxyphenyl azide) were prepared with concentrations of
about 2.5 mM in a 60% water/40% acetonitrile mixed solvent
by volume with an acetate buffer (2 mM) and a pH of 3.5. The
transient resonance Raman spectra were obtained using an
experimental apparatus and methods detailed previously so only
a brief description will be presented here.32,33,38 The harmonics
of a pulsed Nd:YAG laser and their hydrogen Raman shifted
laser lines were used for the pump (266 nm) and probe (320
nm) excitation wavelengths in the transient Raman experiments.
An optical time-delay of approximately 10 ns between the pump
and probe pulses was used in the experiments. The pump and
probe laser beams were loosely focused onto a flowing liquid
stream of sample using a near collinear and backscattering
geometry. The Raman light was collected using reflective optics
and imaged through a depolarizer and entrance slit of a 0.5 m
spectrograph equipped with a liquid-nitrogen-cooled CCD. The
grating of the spectrograph dispersed the Raman light onto the
CCD that accumulated signal for about 300-600 s before being
readout to an interfaced PC computer. Approximately 10-20
of these readouts were summed to obtain a spectrum. Pump
only, probe only, and pump-probe spectra were acquired. A
background scan was also obtained. The known Raman bands
of the water/acetonitrile solvent were employed to calibrate the
Raman shifts of the spectra. Subtraction of a probe only
spectrum from the pump-probe spectrum was done to remove
solvent and precursor Raman bands, and then the pump-only
spectrum and background scan were also subtracted to finally
obtain the transient Raman spectrum.
Figure 1. Examples of a typical 266 nm pump only spectrum in the
probe wavelength region (A), a 320 nm probe only Raman spectrum
(B), a pump-probe spectrum (C), and the resulting transient resonance
Raman spectrum (D) of the 4-methoxyphenylnitrenium ion. “S” mark
solvent bands (in B and C), whereas the asterisks mark solvent
subtraction artifacts in the transient spectrum (D).
All of the density functional theory calculations presented
here made use of the Gaussian 98W program suite39 Complete
geometry optimization and vibrational frequency computations
were done analytically using the (U)BPW91 method40,41 and
the cc-PVDZ basis set.42 Density functional theory (DFT)
calculations at random-phase approximation43 (RPA) were done
to estimate the electronic transition energies and oscillator
strengths for the 4-methoxyphenylnitrenium and 4-ethoxyphen-
ylnitrenium ions.
Figure 2. Expanded view of the transient resonance Raman spectrum
of the 4-methoxyphenylnitrenium ion obtained using a 320 nm probe
and 266 nm pump excitation wavelengths. The assignments of the larger
Raman bands are labeled (see text and Table 1 for more details). The
asterisks mark solvent subtraction artifacts.
pump only spectrum in the probe wavelength region (A), a probe
only Raman spectrum (B), a pump-probe Raman spectrum (C),
and a transient Raman spectrum (D) obtained by subtraction of
probe only and pump only spectra from the pump-probe
spectrum. The spectra in Figure 1 were obtained after 266 nm
photolysis of 4-methoxyphenyl azide in 60% water/40% aceto-
nitrile solvent (by volume). Figure 2 presents an expanded view
of the transient resonance Raman spectrum due to 4-methoxy-
phenylnitrenium ion shown in Figure 1. Figure 3 presents the
transient resonance Raman spectrum obtained for the 4-ethoxy-
phenylnitrenium ion obtained under similar conditions as the
4-methoxyphenylnitrenium ion in Figures 1 and 2. Comparison
of experimental vibrational frequencies to those predicted from
density functional theory (DFT) or ab initio calculations for
Results and Discussion
Flash photolysis of alkoxyphenyl azides in a largely aqueous
environment gives rise to strong transient absorption bands with
maxima around 290-305 nm on the ns to ms time scale.34,35
These transient absorption bands are attributed to alkoxyphen-
ylnitrenium ions and are noticeably different than other arylni-
trenium ions such as 4-biphenylylnitrenium and 2-fluorenylni-
trenium ions that have transient absorption bands with maxima
in the 430-460 nm region.20,21,23-25,29 The 320 nm probe
wavelength used in the transient resonance Raman experiments
was chosen to be resonant with the red side of the transient
absorption bands attributed to the 4-methoxyphenylnitrenium
and 4-ethoxyphenylnitrenium ions. Figure 1 presents a typical

4-Methoxyphenylnitrenium and 4-Ethoxyphenylnitrenium
J. Phys. Chem. A, Vol. 107, No. 40, 2003 8069
TABLE 1: Experimental Raman Vibrational Frequencies Observed in the Time-Resolved Resonance Raman Spectra of the
4-Methoxyphenylnitrenium Ion Shown in Figure 2a
triplet 4-methoxyphenylnitrenium ion
singlet 4-methoxyphenylnitrenium ion
experiment
vibrational
mode, possible
description
UBPW91 cc-PVDZ
vibrational
mode, possible
description
BPW91
10 ns time-
resolved Raman
freq. shift (in cm-1
)
calc. value
cc-PVDZ calc.
(in cm-1
)
value (in cm-1
)
ν45, ring def.
89
156
175
213
305
317
370
403
453
494
510
594
676
724
768
790
821
ν45, ring def.
ν44, ring def.
ν43, CH3 bend
80
154
167
220
273
349
381
413
510?
513
594
732
736
759
798
856
890
939?
946
ν44, ring def.
ν43, CH3 bend
ν42, C-O bend
ν42, C-O bend
235
ν41, N-H bend
ν41, C-O bend + C-N bend
ν40, ring def. + N-H bend
ν39, C-N bend
ν40, N-H bend
ν39, ring def.
ν38, CCC bend
ν38, CCC bend
428
521
ν37, N-H bend
ν37, C-H bend
ν36, ring def. + N-H bend
ν35, CCC bend + C-O-C bend
ν34, CCC bend
ν36, CCC bend
ν35, ring def.
ν34, ring def.
ν33, ring def. + N-H bend
ν32, CCC bend
ν33, C-H bend + N-H bend
ν32, ring def.
ν31, C-H bend
ν31, ring def.
ν30, N-H bend + C-H bend
ν29, ring def.
ν30, C-H bend + N-H bend
ν29, C-H bend + N-H bend
ν28, ring def.
ν28, N-H bend + C-H bend
ν27, C-H bend
849
939
952
956
ν27, ring def.
960
ν26, C-H bend
ν26, C-H bend + N-H bend
ν25, C-H bend
983
988
ν25, ring def. + N-H bend
ν24, C-O stretch + N-H bend
ν23, CH3 bend
965
ν24, C-H bend + N-H bend
ν23, CH3 bend
1069
1097
1135
1146?
1150?
1271
1316
1374
1382
1413
1424?
1435
1485
1549
1571
1639
3002
1096
1098
1148
1155
1240
1284
1302
1351
1393
1416
1422
1440
1467
1478
1606
2990
ν22, C-H bend
ν22, N-H bend + C-H bend
ν21, C-H bend
ν21, CH3 bend + C-H bend
ν20, C-H bend
ν20, C-H bend+N-H bend+CH3 bend
ν19, C-H bend + N-H bend
ν18, C-H bend + CH3 bend
ν17, N-H and C-H bend
ν16, C-H bend + N-H bend
ν15, CH3 bend
1169
ν19, C-H bend
ν18, C-C str.
ν17, C-H bend + CH3 bend
ν16, C-N str. + C-O str.
ν15, CH3 bend
1350
1408
ν14, CH3 bend
ν14, CH3 bend
ν13, C-C str.
ν13, CH3 bend
ν12, C-C str.
ν12, C-O str. + CH3 bend
ν11, C-C str.
1510
ν11, C-C str. + C-O str.
ν10, C-C str.
ν10, C-N str.
1564
1632
ν9, C-C str.
ν8, C-H str.
ν9, C-C str.
ν8, C-H str.
a Possible vibrational band assignments are also shown based on comparison to calculated vibrational frequencies from UBPW91/cc-PVDZ or
BPW91/cc-PVDZ computations in the 200 to 1700 cm-1 fingerprint region for the ground singlet and triplet state of 4-methoxyphenylnitrenium
ion(see text). The experimental vibrational frequencies are compared to those from (U)BPW91/cc-PVDZ computations for the singlet and triplet
sates of the 4-methoxyphenylnitrenium ion. See text for more details
probable photoproduct species has been used successfully to
identify and assign time-resolved infrared (TRIR) and time-
resolved resonance Raman (TR3) spectra to arylnitrenium ions,
arylnitrenes, and arylnitrene photoproducts.28,31-33,38 A similar
methodology will be used to assign the TR3 spectra observed
in Figures 1-3. We note that the lifetimes of arylnitrenium ions
measured previously by transient absorption or TRIR experi-
ments showed no discernible difference when done in O2- and
N2-purged solutions.31,35 Thus, we do not expect that removal
of O2 is likely to make any difference in the observed transient
Raman spectra shown in Figures 1-3.
(U)BPW91/cc-PVDZ calculations were performed for the
singlet and triplet states of the 4-methoxyphenylnitrenium and
4-ethoxyphenylnitrenium ions to find their total energy, opti-
mized geometry, and vibrational frequencies. Tables 1 and 2
compare the experimental transient resonance Raman vibrational
frequencies observed in Figures 2 and 3 to those predicted from
the DFT calculations for the singlet and triplet states of the
4-methoxyphenylnitrenium and 4-ethoxyphenylnitrenium ions,
Figure 3. Expanded view of the transient resonance Raman spectrum
of the 4-ethoxyphenylnitrenium ion obtained using a 320 nm probe
and 266 nm pump excitation wavelengths. The assignments of the larger
Raman bands are labeled (see text and Table 2 for more details). The
respectively. Figure 4 presents simple schematic diagrams of
the singlet and triplet states of the 4-methoxyphenylnitrenium
asterisks mark solvent subtraction artifacts, and the daggers label small
stray light or ambient light artifacts.

8070 J. Phys. Chem. A, Vol. 107, No. 40, 2003
Chan et al.
TABLE 2: Experimental Raman Vibrational Frequencies Observed in the Time-Resolved Resonance Raman Spectra of the
4-Ethoxyphenylnitrenium Ion Shown in Figure 3a
triplet 4-ethoxyphenylnitrenium ion
singlet 4-ethoxyphenylnitrenium ion
experiment
vibrational
mode, possible
description
UBPW91/cc-PVDZ
vibrational
mode, possible
description
BPW91/cc-PVDZ
10 ns time-
resolved Raman
freq. shift (in cm-1
)
calc. value
calc. value
(in cm-1
)
(in cm-1
)
ν48, C-O-C bend
308
314
365
412
456
495
524
592
677
734
766
792
796
816
851
881
939
ν48, CCC bend
304
350
369
428
508
523
592
730
731
761
792
794
864
867
884
939
965
326
ν47, N-H bend
ν47, ring def. + N-H bend
ν46, C-N bend + C-O bend
ν45, C-N bend
ν46, ring def. + N-H bend
ν45, N-H bend + C-O bend
ν44, N-H bend
ν44, C-H bend
ν43, C-H bend
ν43, CCC bend
532
598
ν42, CCC bend
ν42, ring def.
ν41, CCC bend
ν41, C-H bend + N-H bend
ν40, C-O bend
ν40, ring def. + N-H bend
ν39, C-O bend
ν39, ring def.
ν38, ring def.
ν38, C-H bend
793
881
ν37, N-H bend + C-H bend
ν36, C-H bend + CH3 bend
ν35, ring def.
ν37, C-H bend + CH3 bend
ν36, C-H bend
ν35, C-O str.
ν34, C-H bend + N-H bend
ν33, ring def. + C-O str.
ν32, C-H bend
ν34, C-H bend + N-H bend
ν33, ring def.
ν32, C-C str.
952
982
ν31, C-H bend
953
957
978
ν31, C-H bend + N-H bend
ν30, C-H bend
979
991
ν30, ring def.
ν29, C-C str.
ν29, C-H bend + N-H bend
ν28, C-C str.
1068
1098
1103
1136
1151
1248
1269
1305
1346
1361?
1376
1415
1417
1432
1436?
1469
1553
1572
1640
2996
ν28, C-C str + C-H bend
ν27, C-H bend
1096
1102
1103
1155
1240
1247
1280
1301
1333
1355
1367
1416
1418
1433
1440
1459
1479
1605
2981
ν27, CH bend + CH3 bend
ν26, C-H bend + N-H bend
ν25, C-H bend + N-H bend
ν24, C-H bend + CH3 bend
ν23, C-H bend + N-H bend
ν22, C-H bend + CH2 bend
ν21, C-H bend+CH2 bend+CH3 bend
ν20, CH2 bend+CH3 bend
ν19, C-C str.+N-H bend
ν18, CH2 bend+CH3 bend
ν17, CH2 bend
ν26, C-H bend + CH3 bend
ν25, C-H bend
1169
1328
ν24, C-C str.
ν23, C-H bend + CH3 bend
ν22, C-C str.
ν21, C-H bend + CH2 bend
ν20, C-N str.+C-H bend+CH3 bend
ν19, C-N str. + CH3 bend
ν18, C-N str.+C-H bend+CH2 bend
ν17, CH3 bend
ν16, C-H bend
ν16, C-C str. + CH3 bend
ν15, C-C str. + CH3 bend
ν14, C-O str. + C-N str.
ν13, C-C str.
1452
1492
ν15, C-C str. + CH3 bend
ν14, C-C str. + CH3 bend
ν13, C-C str.
ν12, C-C str.
ν12, C-N str.
1564
1636
ν11, C-C str.
ν11, C-C str.
ν10, C-H str.
ν10, C-H str.
a Possible vibrational band assignments are also shown based on comparison to calculated vibrational frequencies from UBPW91/cc-PVDZ or
BPW91/cc-PVDZ computations in the 200 to 1700 cm-1 fingerprint region for the ground singlet and triplet state of 4-ethoxyphenylnitrenium
ion(see text). The experimental vibrational frequencies are compared to those from (U)BPW91/cc-PVDZ computations for the singlet and triplet
sates of 4-ethoxyphenylnitrenium ion. See text for more details.
and 4-ethoxyphenylnitrenium ions with selected DFT optimized
geometry bond lengths indicated next to the appropriate bonds.
Inspection of Figure 4 shows that the singlet states of the
4-methoxyphenylnitrenium and 4-ethoxyphenylnitrenium ions
have significantly more imine character (stronger CdN bonds)
accompanied by more cyclohexadienyl (greater carbon-carbon
bond length alternation) and oxo (stronger CdO bonds)
character compared to their respective triplet states. These
differences in structure lead to noticeable differences in the
vibrational frequencies of the singlet and triplet states and enable
them to be distinguished from one another using vibrational
spectroscopy.
the subsituent increases the singlet state is more stabilized over
the triplet state.31,44 This leads to greater differences in the
structures of the singlet and triplet states for stronger π-donating
para substituents such as phenyl and alkoxy groups. We
performed single-point energy calculations for the 4-methoxy-
nitrenium and 4-ethoxynitrenium ions at the (U)BPW91/cc-
PVDZ optimized geometry and found singlet-triplet gaps of
28.0 and 28.1 kcal/mol, respectively, at the (U)BPW91/cc-PVDZ
level of theory and 35.1 and 35.5 kcal/mol, respectively, at the
(U)MP2/cc-PVDZ level of theory with the singlet state being
more stable in each case. This indicates the singlet states of the
4-methoxynitrenium and 4-ethoxynitrenium ions are substan-
tially more stable than their triplet states. This and the fast
formation of the nitrenium ions by protonation of the initially
produced singlet arylnitrene precursor35 suggests that the
transient resonance Raman spectra observed in Figures 2 and 3
are due to the singlet state 4-methoxynitrenium and 4-ethoxy-
nitrenium ions similar to the assignment of the transient
absorption and resonance Raman spectra of the 2-fluorenylni-
trenium ion to the singlet state.19,22,32
We note that there are circumstances where the ground state
of the arylnitrenium ion may be a triplet state when a bulky
aryl ligand or π-acceptor substituent may substantially desta-
bilize the singlet state relative to the triplet state.14,24 However,
this is uncommon, and there are only a few reports that predict
or observe triplet ground-state arylnitrenium ions.14,24 The
singlet-triplet gaps change as a function of substituents para
to the nitrenium ion moiety and as the π-donating character of

4-Methoxyphenylnitrenium and 4-Ethoxyphenylnitrenium
J. Phys. Chem. A, Vol. 107, No. 40, 2003 8071
Figure 4. Diagrams of the singlet and triplet states of the 4-methoxyphenylnitrenium and 4-ethoxyphenylnitrenium ions with the atoms numbered.
A diagram of the singlet state of the 2-fluorenylnitrenium ion is also given for comparison purposes. Selected bond lengths are shown (in Å) from
the (U)BPW91/cc-PVDZ computations.
Inspection of Table 1 shows that the transient resonance
Raman vibrational frequencies are in noticeably better agreement
with the frequencies predicted for the singlet 4-methoxyphen-
ylnitrenium ion than for the corresponding triplet state. The
singlet state calculated vibrational frequencies are different from
the experimental Raman frequencies by about 13.8 cm-1 on
average compared to differences of about 24.3 cm-1 on average
for the triplet state. Examination of Table 2 shows a similar
agreement of the experimental Raman frequencies of Figure 3
to those predicted for the singlet state of 4-ethoxyphenylnitre-
nium ion (differences of about 11 cm-1 on average) compared
to the corresponding triplet state (differences of about 16.5 cm-1
on average). Figure 5 compares the computed vibrational
frequencies and relative Raman intensities for the singlet and
triplet states of the 4-methoxyphenylnitrenium ion to the
experimental resonance Raman spectrum vibrational frequencies
and relative intensities. Inspection of Figure 5 shows that the
experimental vibrational frequencies and relative Raman intensi-
ties exhibit a good correlation with those computed for the
singlet 4-methoxyphenylnitrenium ion but not for the triplet
state. This and the large computed singlet-triplet gap with the
singlet state being more stable leads us to assign the transient
Raman spectra in Figures 2 and 3 to the singlet states of
4-methoxyphenylnitrenium and 4-ethoxyphenylnitrenium ions,
respectively.
Closer examination of Tables 1 and 2 and Figure 5 shows
that the vibrational bands in the 1450-1650 cm-1 region are
particularly useful to distinguish between the singlet and triplet
states of these arylnitrenium ions. For example, in the 1450-
1650 cm-1 region, the singlet state of the 4-methoxyphenylni-
trenium ion has predicted vibrational bands at 1485, 1549, 1571,
and 1639 cm-1, whereas the triplet state has predicted bands at

8072 J. Phys. Chem. A, Vol. 107, No. 40, 2003
Chan et al.
Figure 5. Comparison of the experimental transient Raman vibrational frequencies and relative intensities (from Figure 2) to those computed for
the singlet and triplet states of the 4-methoxyphenylnitrenium ion. The largest Raman band in the 1600 cm-1 region is scaled to be 100 for all three
diagrams. Only Raman bands with intensity >1 are shown. The dashed lines help indicate the correlation between the spectra.
TABLE 3: Electronic Transition Energies and Oscillator
one large oscillator strength electronic transition at ∼303 and
Strengths Obtained from RPA/cc-PVDZ Calculations
∼292.6 nm respectively in the 200-1000 nm region. These
Using the BPW91/cc-PVDZ Optimized Geometry for the
results are consistent with experimental results that show these
Singlet 4-Methoxyphenylnitrenium and
4-Ethoxyphenylnitrenium Ions
ions have strong transient absorption bands with maxima around
290-305 nm on the ns to ms time scale.34,35 This is consistent
with and provides further support for our assignment of the
transient resonance Raman spectra shown in Figures 2 and 3 to
the singlet states of the 4-methoxyphenylnitrenium and 4-ethoxy-
phenylnitrenium ions.
singlet
transition
energies
oscillator
strength
species
4-methoxyphenylnitrenium ion
566.6 nm
405.0 nm
320.7 nm
303.3 nm
260.1 nm
257.3 nm
236.5 nm
217.0 nm
563.4 nm
410.4 nm
361.3 nm
338.2 nm
297.1 nm
292.6 nm
278.5 nm
256.8 nm
0.0006
0.0145
0.0000
0.3327
0.0003
0.0000
0.0095
0.0001
0.0006
0.0244
0.0377
0.0001
0.0000
0.3090
0.0000
0.0000
It is interesting to compare the structural properties of the
singlet 4-methoxyphenyl nitrenium and 4-ethoxyphenylnitre-
nium ions to those of previously investigated arylnitrenium ions
(note all structural comparisons discussed are made based on
the same BPW91/cc-PVDZ level of calculated structures).28,31-33
The imine character of the 4-methoxyphenyl nitrenium and
4-ethoxyphenylnitrenium ions is very strong (CdN bonds of
about 1.3042 and 1.3040 Å respectively) and comparable to
that of the 2-fluorenylnitrenium ion (with a CdN bond of about
1.3077 Å).32,33 The CdN imine bonds in the 4-methoxyphenyl
nitrenium, and 4-ethoxyphenylnitrenium ions are noticeably
stronger than in the 4-biphenylylnitrenium ion (1.3083 Å),33
diphenylnitrenium ion (1.3512 Å),28,33 N-(4-biphenylyl)-N-
methylnitrenium ion (1.321 Å),31 N-(4-methylphenyl)-N-meth-
ylnitrenium ion (1.323 Å),31 N-(4-chlorophenyl)-N-methylni-
trenium ion (1.323 Å),31 and N-(4-methoxyphenyl)-N-meth-
ylnitrenium ion (1.317 Å).31 Similarly, the degree of C-C bond
alternation (cyclohexadienyl character) in the phenyl ring to
which the nitrenium moiety is attached is also very strong in
the two 4-alkoxyphenylnitrenium ions examined and comparable
to that found in the 2-fluorenylnitrenium ion.32,33 For example,
the first phenyl ring in the 2-fluorenylnitrenium ion has C12-
C11 of 1.4732 Å; C11-C10 of 1.3701 Å; and C10-C7 of
1.4373 Å and C12-C13 of 1.4696 Å; C13-C8 of 1.3685 Å;
and C8-C7 of 1.4497 Å (see Figure 4 for numbering system)
compared to C3-C8 of 1.4710 Å; C8-C7 of 1.3688 Å; and
C7-C6 of 1.4413 Å and C3-C4 of 1.4705 Å; C4-C5 of 1.3641
Å; and C5-C6 of 1.4486 Å for the singlet 4-methoxyphenylni-
trenium ion (see Figure 4). Our results clearly indicate that the
alkoxyphenylnitrenium ions have a large degree of iminocy-
clohexadienyl character with an amount similar to the 2-fluo-
renylnitrenium ion. This is consistent with the similarity of the
4-ethoxyphenylnitrenium ion
1467, 1478, and 1605 cm-1. The transient Raman frequencies
observed at 1510, 1564, and 1633 cm-1 in the 1450-1650 cm-1
region in Figures 2 and 5 exhibit substantially better agreement
with those computed for the singlet state (differences on average
of 12.7 cm-1) compared to the triplet state (differences of about
53 cm-1 on average) for these three modes. In addition, the
relative intensity pattern for these vibrational modes are
significantly different for the singlet and triplet states as seen
in Figure 5. The vibrational modes in the 1450 to 1650 cm-1
region are sensitive and diagnostic of the degree of imine,
cyclohexadienyl, and oxo character of the alkoxyphenylnitre-
nium ions.
We used density functional theory (DFT) computations at
the random-phase approximation43 (RPA) to estimate the
electronic transition energies and oscillator strengths for the
singlet state 4-methoxyphenylnitrenium and 4-ethoxyphenylni-
trenium ions, and these results are presented in Table 3.
Inspection of Table 3 reveals that the singlet state 4-methoxy-
phenylnitrenium and 4-ethoxyphenylnitrenium ions have only

4-Methoxyphenylnitrenium and 4-Ethoxyphenylnitrenium
J. Phys. Chem. A, Vol. 107, No. 40, 2003 8073
experimental resonance Raman vibrational frequencies of the
symmetric aromatic CdC stretch mode that is observed at 1633
cm-1 in the 4-methoxyphenylnitrenium ion and at 1636 cm-1
in the 4-ethoxyphenylnitrenium ion compared to 1633 or 1637
cm-1 in the 2-fluorenylnitrenium ion.32,33 The symmetric
aromatic CdC stretch vibrational mode in the 1568-1650 cm-1
region has previously been demonstrated to exhibit a good
correlation with the degree of iminocyclohexadienyl character
of a range of arylnitrenium ions,28,31-33 and the alkoxynitrenium
ions studied here also show this type of correlation.
be due to the 4-alkoxyphenylnitrenium ions reacting like
oxocarbocations and the greater charge delocalization in the
biphenylyl systems leads to lower oxocarbocation character and
a longer lifetime. Our present results showing that 4-alkoxy-
phenylnitenium ions have significant oxocarbocation character
and combined with our previous work on para phenyl substituted
arylnitrenium ions demonstrating effective charge delocalization
into the para substituted phenyl ring32,33 provide support for this
hypothesis. The barriers to reaction with water for oxocarboca-
tions have been suggested to be lower for those with greater
charge on the oxocarbo moiety compared to those in which the
charge is more delocalized.46
The large degree of iminocyclohexadienyl character found
in the 2-fluorenylnitrenium ion was largely attributed to the
substitution of a second phenyl ring para to the nitrenium ion
and the tendency of the biphenyl-like rings to go toward a more
quinoidal-like structure as charge delocalized into the phenyl
rings as well as the locking fluorene bridge that enhances this
tendency.32,33 However, the alkoxyphenylnitrenium ions do not
have this second phenyl ring, and it is the alkoxy substitution
para to the nitrenium ion moiety that appears to cause a similar
effect as a para substituted phenyl group. Instead of charge
delocalization proceeding through a quinoidal-like structure into
two phenyl rings as in the 2-fluorenylnitrenium ion, the charge
in the alkoxyphenylnitrenium ion tends to proceed to the CdO
group leading to a significant charge on the CdO group and
noticeable imine and oxo cyclohexadienyl character.
The oxocarbocation character of the 4-alkoxyphenylnitrenium
ions also appears to influence the chemical reactivity of these
ions toward nucleosides such as 2′deoxyguanosine. For example,
the biphenyl-4-yl-nitrenium and the 4′-methoxybiphenyl-4-
ylnitrenium ions exhibited significant selectivity toward reaction
with 2′-deoxyguanosine even in the nucleophilic solvent water,30
whereas this selectivity was not found for the 4-alkoxyphen-
ylnitrenium ions.34,35 This was suggested to be due to greater
charge at the oxocarbo group in the alkoxyphenylnitrenium ions
and hence a lesser charge and reactivity at the nitrenium
moiety.35 Our present results are reasonably consistent with this
hypothesis in that there does appear to be a noticeable
localization of charge at the oxocarbo moiety at the expense of
the nitrenium ion moiety. However, the nitrenium moiety
appears to have an imine character and charge very similar to
that found in the 2-fluorenylnitrenium and 4-biphenylylnitenium
ions. Thus, we speculate that the different chemical reactivity
of the alkoxynitrenium ions toward nucleosides such as 2′-
deoxyguanosine is not just an effect of charge localization and
is also due to an increase in chemical reactivity toward
nucleosides such as water (via the oxocarbocation site). This
could lead to a situation where reaction with water (via the
oxocarbocation site) has a rate relative to the reaction of the
nitrenium ion moiety with nucleosides such as 2′-deoxygua-
nosine so that the overall alkoxynitrenium ion exhibits little
selectivity toward 2′-deoxyguanosine as found in previous laser
flash studies.34,35 This hypothesis would also be consistent with
the 4′-methoxybiphenyl-4-ylnitrenium ion exhibiting some
selectivity toward 2′-deoxyguanosine because this arylnitrenium
ion would have significantly less charge on the methoxy group
(so its oxocarbocation character and reactivity would be
significantly less than in the 4-alkoxyphenylnitrenium ions),
whereas the nitrenium ion moiety would still retain a reactivity
like the 4-biphenylylnitrenium ion. Further work is needed to
see if this is actually the case. Additional studies of the chemical
reactions of alkoxyarylnitrenium ions using a variety of
experimental techniques should prove useful to more fully
understand the chemical behavior of these interesting arylni-
trenium ions.
As the alkoxy group is changed from methoxy to ethoxy for
the singlet alkoxyphenylnitrenium ions in Figure 4, the degree
of imine and oxo character increase somewhat at the same time:
the CdN bond decreases from 1.3042 to 1.3040 Å, and the
CdO bond decreases from 1.3099 to 1.3061 Å. This imine and
oxo character of the 4-alkoxyphenylnitrenium ions can also be
affected by changing the character of the nitrenium moiety from
-NH+ to NCH3+. For example, the singlet 4-methoxyphenylni-
trenium ion had CdN and CdO bond lengths of about 1.3042
and 1.3099 Å, respectively, that become weaker and lengthen
to about 1.317 and 1.316 Å, respectively, in the N-(4-
methoxyphenyl)-N-methylnitrenium ion (1.317 Å).31 This lesser
imine and oxo character is also accompanied by a smaller
cyclohexadienyl character in N-(4-methoxyphenyl)-N-methylni-
trenium ion31 compared to the 4-methoxyphenylnitrenium ion.
The nitrenium ion moiety charge delocalization into the phenyl
ring of 4-alkoxyphenylnitrenium ions leads to greater oxo and
imine character that is accompanied by cyclohexadienyl char-
acter. The charge that would be delocalized further into a second
phenyl ring in systems such as the 2-fluorenylnitrenium ion32
(or other para phenyl substituted arylnitrenium ions such as
4-biphenylylnitrenium ion33 and N-(4-biphenylyl)-N-methylni-
trenium ion31) mainly concentrates at the CdO group and gives
the 4-alkoxyphenylnitrenium ions substantial oxocarbocation
character. This has important implications for the properties and
chemical reactivity of 4-alkoxyphenylnitrenium ions compared
to para phenyl substituted arylnitrenium ions as discussed below.
Acknowledgment. This work was supported by grants from
the Committee on Research and Conference Grants (CRCG),
the Research Grants Council (RGC) of Hong Kong (HKU 7112/
00P), and the Large Items of Equipment Allocation 1993-94
from the University of Hong Kong.
The 4-alkoxyphenylnitrenium ions have been previously noted
to have intriguing properties and chemical reactivity compared
to para phenyl substituted arylnitrenium ions. For example, the
4-alkoxyphenylnitrenium ions have lifetimes similar to para
phenyl substituted arylnitrenium ions such as the biphenyl-4-
ylnitrenium ion34,35,45 even though 4-alkoxy substitution is much
more cation stabilizing on the σ+ scale than the 4-phenyl
substitution. In addition, it was found that adding another phenyl
group between the methoxy moiety and the nitrenium ion moiety
in the 4′-methoxybiphenyl-4-ylnitrenium ion gave an almost 3
orders of magnitude longer lifetime on the order of mil-
liseconds.29 McClelland and Ramlall35 suggested that this could
Supporting Information Available: The Cartesian coor-
dinates, energies, zero-point energy corrections, and predicted
vibrational frequencies and intensities from the (U)BPW91/cc-
PVDZ density functional theory calculations are given for the
singlet and triplet states of the 4-methoxyphenylnitrenium and
4-ethoxyphenylnitrenium ions. Single-point energy calculations
are also given for the singlet and triplet states of the 4-meth-

8074 J. Phys. Chem. A, Vol. 107, No. 40, 2003
Chan et al.
(26) Robbins, R. J.; Laman, D. M.; Falvey, D. E. J. Am. Chem. Soc.
1996, 118, 8127.
(27) Moran, R. J.; Falvey, D. E. J. Am. Chem. Soc. 1996, 118, 8965.
(28) Srivastava, S.; Toscano, J. P.; Moran, R. J.; Falvey, D. E. J. Am.
Chem. Soc. 1997, 119, 11552.
(29) Ren, D.; McClelland, R. A. Can. J. Chem. 1998, 76, 78.
(30) McClelland, R. A.; Gadosy, T. A.; Ren, D. Can. J. Chem. 1998,
76, 1327.
(31) Srivastava, S.; Ruane, P. H.; Toscano, J. P.; Sullivan, M. B.; Cramer,
C. H.; Chiapperino, D.; Reed, E. C.; Falvey, D. E. J. Am. Chem. Soc. 2000,
122, 8271.
(32) Zhu, P.; Ong, S. Y.; Chan, P. Y.; Leung, K. H.; Phillips, D. L. J.
Am. Chem. Soc. 2001, 123, 2645.
(33) Zhu, P.; Ong, S. Y.; Chan, P. Y.; Poon, Y. F.; Leung, K. H.; Phillips,
D. L. Chem. Eur. J. 2001, 7, 4928.
(34) Sukhai, P.; McClelland, R. A. J. Chem. Soc., Perkin Trans. 2 1996,
oxyphenylnitrenium and 4-ethoxyphenylnitrenium ions at the
(U)BPW91/cc-PVDZ and (U) MP2/cc-PVDZ levels of theory
using the (U)BPW91/cc-PVDZ optimized geometry. This mater-
ial is available free of charge via the Internet at http://
pubs.acs.org.
References and Notes
(1) Miller, J. A. Cancer Res. 1970, 20, 559.
(2) Scribner, J. D.; Naimy, N. K. Cancer Res. 1975, 35, 1416.
(3) Miller, E. C. Cancer Res. 1978, 38, 1479.
(4) Miller, E. C.; Miller, J. A. Cancer 1981, 47, 2327.
(5) Singer, B.; Kusmierek, J. T. Annu. ReV. Biochem. 1982, 51, 655.
(6) Miller, J. A.; Miller, E. C. EnViron. Health Perspect. 1983, 49, 3.
(7) Garner, R. C.; Martin, C. N.; Clayson, D. B. In Chemical
Carcinogens, 2nd ed.; Searle, C. E., Ed.; ACS Monograph 182; American
Chemical Society: Washington DC, 1984; Vol. 1, pp 175-276.
(8) Famulok, M.; Boche, G. Angew. Chem., Int. Ed. Engl. 1989, 28,
468.
(9) Meier, C.; Boche, G. Tetrahedron Lett. 1990, 31, 1693.
(10) Humphreys, W. G.; Kadlubar, K. K.; Guengerich, F. P. Proc. Natl.
Acad. Sci. U.S.A. 1992, 89, 8278.
(11) Kadlubar, F. F. In DNA Adducts of Carcinogenic Amines; Hem-
miniki, K., Dipple, A., Shuker, D. E. G., Kadlubar, K. K., Segerba¨ch, D.,
Bartsch, H., Ed.; Oxford University Press: Oxford, U.K., 1994; pp 199-
216.
1529.
(35) Ramlall, P.; McClelland, R. A. J. Chem. Soc., Perkin Trans. 2 1999,
225.
(36) Brown, B. R.; Yielding, L. W.; White, W. E., Jr. Mutat. Res. 1980,
70, 17.
(37) White, W. E., Jr.; Yielding, L. W. In Methods in Enzymology Vol.
XLVI Affinity Labeling; Jakoby, W. B., Wilchek, M., Eds.; Academic
Press: Orlando, FL, 1977; pp 646-647.
(38) Ong, S. Y.; Zhu, P.; Poon, Y. F.; Leung, K. H.; Fang, W.-H.;
Phillips, D. L. Chem. Eur. J. 2002, 8, 2163.
(39) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A., Jr.;
Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A.
D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi,
M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.;
Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Malick,
D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Cioslowski, J.;
Ortiz, J. V.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi,
I.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.;
Peng, C. Y.; Nanayakkara, A.; Gonzalez, C.; Challacombe, M.; Gill, P. M.
W.; Johnson, B. G.; Chen, W.; Wong, M. W.; Andres, J. L.; Head-Gordon,
M.; Replogle, E. S.; Pople, J. A. Gaussian 98; Gaussian, Inc.: Pittsburgh,
PA, 1998.
(12) Dipple, A. Carcinogenesis 1995, 16, 437.
(13) Novak, M.; Kahley, M. J.; Lin, J.; Kennedy, S. A.; James, T. G. J.
Org. Chem. 1995, 60, 8294.
(14) Cramer, C. J.; Falvey, D. E. Tetrahedron Lett. 1997, 38, 1515.
(15) Novak, M.; Kennedy, S. A. J. Am. Chem. Soc. 1995, 117, 574.
(16) Hoffman, G. R.; Fuchs, R. P. P. Chem. Res. Toxicol. 1997, 10,
347.
(17) Novak, M.; Kennedy, S. A. J. Phys. Org. Chem. 1998, 11, 71.
(18) Novak, M.; VandeWater, A. J.; Brown, A. J.; Sanzebacher, S. A.;
Hunt, L. A.; Kolb, B. A.; Brooks, M. E. J. Org. Chem. 1999, 64, 6023.
(19) McClelland, R. A.; Ahmad, A.; Dicks, A. P.; Licence, V. J. Am.
Chem. Soc. 1999, 121, 3303.
(20) Anderson, G. B.; Falvey, D. E. J. Am. Chem. Soc. 1993, 115, 9870.
(21) Davidse, P. A.; Kahley, M. J.; McClelland, R. A.; Novak, M. J.
Am. Chem. Soc. 1994, 116, 4513.
(22) McClelland, R. A.; Davidse, P. A.; Hadialic, G. J. Am. Chem. Soc.
1995, 117, 4173.
(40) Becke, A. J. Chem. Phys. 1986, 84, 4524.
(41) Perdew, J. P.; Burke, K.; Wang, Y. Phys. ReV. B 1996, 54, 16533.
(42) Dunning, T. H. J. Chem. Phys. 1989, 90, 1007.
(43) Bauernschmitt, R.; Ahlrichs, R. Chem. Phys. Lett. 1996, 256, 454.
(44) Sullivan, M. B.; Brown, K.; Cramer, C. J.; Truhlar, D. G. J. Am.
Chem. Soc. 1998, 120, 11778.
(45) Novak, M.; Kahley, M. J.; Lin, J.; Kennedy, S. A.; Swanegan, L.
A. J. Am. Chem. Soc. 1994, 116, 11626.
(46) Richard, J. P.; Amyes, T. L.; Jagannadham, V.; Lee, Y.-G.; Rice,
D. J. J. Am. Chem. Soc., 1995, 117, 5198.
(23) Robbins, R. J.; Yang, L. L.-N.; Anderson, G. B.; Falvey, D. E. J.
Am. Chem. Soc. 1995, 117, 6544.
(24) Srivastava, S.; Falvey, D. E. J. Am. Chem. Soc. 1995, 117, 10186.
(25) McClelland, R. A.; Kahley, M. J.; Davidse, P. A.; Hadzialic, G. J.
Am. Chem. Soc. 1996, 118, 4794.