N-methyl-N-phenylnitrenium ion from photolysis of N-(methylphenylamino)-2,4,6-trimethylpyridinium tetrafluoroborate

Article abstract of DOI:10.1002/(SICI)1099-1395(199712)10:12<917::AID-POC954>3.0.CO;2-6

The photochemical reactions of N-(methylphenylamino)-2,4,6-trimethylpyridinium tetrafluoroborate were studied to find evidence of photodecomposition to a nitrenium ion reactive intermediate. Stable products were formed that were consistent with a singlet-state methylphenylnitrenium ion precursor. The methoxy and chloro adducts. N-methyl-p-anisidine and 4- (and 2-)chloro-N-methylaniline, were identified and quantified by high-performance liquid Chromatographic analysis. A hydride shift from the N-methyl group of the nitrenium ion is also proposed based on the detection of aniline which would result from hydrolysis of the iminium ion rearrangement product. The rate constant for this rearrangement is estimated to be 10⁸s^-1. The reduction product, N-methylaniline, is produced, and is believed to form, at least in part, from the hydrogen atom abstractions of the triplet nitrenium ion. This is supported by the results of triplet sensitized irradiations. Laser flash photolysis studies yielded the transient spectrum of a long-lived intermediate absorbing at 470 nm. This transient species is believed to be the cation radical of N-methylaniline.

Full text of DOI:10.1002/(SICI)1099-1395(199712)10:12<917::AID-POC954>3.0.CO;2-6

JOURNAL OF PHYSICAL ORGANIC CHEMISTRY, VOL. 10, 917–924 (1997)
N-METHYL-N-PHENYLNITRENIUM ION FROM PHOTOLYSIS OF N-
(METHYLPHENYLAMINO)-2,4,6-TRIMETHYLPYRIDINIUM
TETRAFLUOROBORATE
D. CHIAPPERINO AND D. E. FALVEY
Department of Chemistry, University, of Maryland, College Park, Maryland 70742, USA
The photochemical reactions of N-(methylphenylamino)-2,4,6-trimethylpyridinium tetrafluoroborate were studied to
find evidence of photodecomposition to a nitrenium ion reactive intermediate. Stable products were formed that were
consistent with a singlet-state methylphenylnitrenium ion precursor. The methoxy and chloro adducts. N-methyl-p-
anisidine and 4- (and 2-)chloro-N-methylaniline, were identified and quantified by high-performance liquid
chromatographic analysis. A hydride shift from the N-methyl group of the nitrenium ion is also proposed based on the
detection of aniline which would result from hydrolysis of the iminium ion rearrangement product. The rate constant
for this rearrangement is estimated to be 108 s؊1. The reduction product, N-methylaniline, is produced, and is believed
to form, at least in part, from the hydrogen atom abstractions of the triplet nitrenium ion. This is supported by the
results of triplet sensitized irradiations. Laser flash photolysis studies yielded the transient spectrum of a long-lived
intermediate absorbing at 470 nm. This transient species is believed to be the cation radical of N-methylaniline. © 1997
John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 10, 917–924 (1997) No. of Figures: 5 No. of Tables: 1 No. of References: 29
Keywords: nitrenium ion; pyridinium salts; photochemistry; singlet states; triplet states
Received 3 January 1997; revised 5 May 1997; accepted 7 May 1997
INTRODUCTION
behaves as an electrophilic species and the triplet behaves
more like a radical.12, 20–22
Nitrenium ions are short-lived reactive intermediates that
contain a cationic divalent nitrogen atom.1 There has been
an interest in these species owing to the possible role of
aryl-substituted nitrenium ions in chemical carcinogene-
sis.2–5 The chemistry of carbenes and nitrenes, which are
closely related isoelectronic species, is fairly well charac-
terized.6–8 Although much less is known about the chemical
reactivity of nitrenium ions, the last decade of research has
greatly illuminated some of the trends in reactivity unique to
arylnitrenium ion chemistry.9–12 Theoretical investigations
have examined the electronic structure of nitrenium ions.
Like their isoelectronic species, carbenes and nitrenes,
nitrenium ions can exist in either a singlet state, with two
non-bonding electrons occupying the same orbital, or a
triplet state, with two non-bonding electrons of parallel spin
in different orbitals. The simplest possible nitrenium ion,
NH2+, has been shown to be a ground-state triplet.13–15
However, aryl substitution is predicted to make the singlet
nitrenium ion the ground state.16–19 Recent experimental
research has demonstrated that these two possible spin
states have very different chemical reactivities: the singlet
Photochemical generation of nitrenium ions has proven
advantageous in that the triplet state, which is typically the
excited state of arylnitrenium ions, is more easily acces-
sible. Use of triplet sensitizers can lead to triplet nitrenium
ion directly, whereas intersystem crossing from the singlet
manifold would have to be somewhat efficient to study the
triplet nitrenium ion under any other reaction conditions.
Photochemical generation also allows for the use of laser
flash photolysis methods to obtain evidence for nitrenium
ion intermediacy and to measure rate constants for their
various reactions. The photochemical ring opening reaction
of 3-methyl-N-alkylanthranilium ions, which leads to aryl-
akylnitrenium ion intermediates, has been valuable in that it
allows for the direct observation of these intermediates by
laser flash photolysis methods.12, 20–22 However, this chem-
istry only allows for the characterization of
2-acetyl-substituted arylnitrenium ions.
Abramovitch and co-workers23, 24 have shown that aryl-
acyl- and diacylnitrenium ions can be generated photo-
chemically from appropriate N-aminopyridinium salts due
to the photoinduced heterolysis of the N—N bond in the
pyridinium cation. Recent work in our laboratories involved
pyridinium salt substrates used to generate diarylnitrenium
* Correspondence to: D. E. Falvey.
Contract grant sponsor: National Science Foundation.
© 1997 John Wiley & Sons, Ltd.
CCC 0894–3230/97/120917–08 $17.50

918
D. CHIAPPERINO AND D. E. FALVEY
ions photochemically.20 However, few results have been
reported for the photochemical generation of unsubstituted
arylalkylnitrenium ions. The present work involves the
photochemical reactions of a pyridinium salt substrate,
N-(methylphenyl)amino-2,4,6-trimethylpyridinium tetra-
fluoroborate (MPAP). Product distributions obtained from
the photolysis of MPAP under various conditions have been
examined. These results are consistent with the formation of
a methylphenylnitrenium ion intermediate (PhMeN+ ). A
comparison of direct and triplet sensitized irradiation shows
that the singlet decays via nucleophilic attack on the phenyl
ring or by rearrangement involving the N-methyl group
while the triplet decays by radical pathways to form the
reduction product, N-methylaniline. The photo-decomposi-
tion of MPAP was also studied by laser flash photolysis
methods, but the transient absorbance detected was not
attributed to a nitrenium ion intermediate. More likely, the
transient species detected was an intermediate product of
nitrenium ion reactions.
Photolysis of MPAP. Photolyses of MPAP for product
studies were carried out using a 150 W Hg–Xe lamp with a
295 nm cut-off filter. Standard conditions involved using
3 ml of N2-purged samples of stock solution in a quartz
cuvette. To investigate the reactivity of MPAP with
methanol, stock solutions were prepared that were typically
2 mg mlϪ1 of MPAP in 10% (v/v) MeOH–MeCN solvent.
Reactivity with chloride ion was investigated using stock
solutions of 2 mg mlϪ1 of MPAP in MeCN with added
Bun4NCl as the source of ClϪ . Photolysis times for various
experiments ranged from 2 to 10 min to minimize secon-
dary photoprocesses unless the primary intent was to
observe the results of these secondary processes. Triplet
sensitized photolyses with 9-fluorenone were carried out
with a 420 nm cut-off filter to prevent the occurrence of any
direct photolysis. The sensitizer was used in a ca 4:1 molar
excess with the pyridinium salt. Longer photolysis times
(20–30 min) were necessary for this slow reaction.
Analytical results for all photolyses were obtained using
a Rainin Instruments high-performance liquid chromato-
graphic (HPLC) system with a C18 reversed-phase column
and a UV detector. All injections were 20 ␮l and the solvent
system used was water–acetonitrile (1:1). Absorbance was
measured at 254 nm. HPLC peaks were identified by co-
injecting photolysis mixtures with authentic products,
which were purchased from Aldrich Chemical or synthe-
sized as necessary (e.g. 2-chloro-N-methylaniline25).
Concentrations of photoproducts were measured based on
concentration curves generated from authentic products.
Yields of various photoproducts (e.g. in Figure 2) were
measured by calculating the concentration of the product as
a percentage of the initial concentration of starting material.
In addition to identification of products by HPLC, 1H NMR
spectroscopy was used to confirm the presence of various
photoproducts in crude photolyzed solutions.
EXPERIMENTAL
Laser flash photolysis studies with N-(methylphenyla-
mino)-2,4,6-trimethylpyridinium
tetrafluoroborate
(MPAP). Laser flash photolysis experiments were carried
out with an excimer laser using Xe–HCl reagent gas in an
He buffer. This supplies 308 nm pulses of 10 ns duration.
The gas mixture and pressure were adjusted in order to
create pulses of 30–70 mJ. However, during a given
experiment the pulse energy varied by ca 5%. The transient
behavior was monitored using a probe beam from a CW 300
W Xe arc lamp passed through the sample cuvette
perpendicular to the excitation beam. Transient waveforms
were recorded with a LeCroy 9420 digital oscilloscope,
which digitizes at a rate of one point per 10 ns with a
bandwidth of 350 MHz. Sample concentrations were
adjusted to an optical density of 0·5–1·5 cmϪ1 at 308 nm.
Using 100 mg of MPAP in 50 ml of freshly distilled
acetonitrile in a standard flow cell set-up with N2 purging,
waveforms were collected for wavelengths ranging from
300 to 800 nm in increments of 5 or 10 nm. The system was
then purged with O2 and waveforms were collected at
460 nm for comparison.
Synthesis of MPAP. 2,4,6-Trimethylpyrylium tetra-
fluoroborate was prepared as outlined by Balaban and
Boulton.26 HBF4 (7·0 ml of 48 wt% solution) was added
slowly to 50 ml of acetic anhydride containing 4 ml of tert-
butyl alcohol. The rate of addition of HBF4 solution was
controlled such that the temperature remained below 80°C
for this very exothermic process. The reaction mixture was
cooled on an ice-bath and the product, 2,4,6-trimethylpyr-
Scheme 1
JOURNAL OF PHYSICAL ORGANIC CHEMISTRY, VOL. 10, 917–924 (1997)
© 1997 John Wiley & Sons, Ltd.

N-METHYL-N-PHENYLNITRENIUM ION
919
ylium tetrafluoroborate, was precipitated by the addition of
diethyl ether. Filtration yielded 3·9 g (44% yield). A 2·31 g
amount of this pyrylium salt was stirred in 30 ml of ethanol
and combined with an additional 30 ml of a solution of
ethanol containing 2·0 g of 1-methyl-1-phenylhydrazine
(1·5 molar equivalents). This mixture was stirred at room
temperature for 2 h, then concentrated by rotary evaporation
at 40 °C. Addition of 40 ml of diethyl-ether precipitated
1·8 g of MPAP product (m.p. 141–142 °C, 52% yield): IR
(CH2Cl2), 3680 (vw), 3040 (m), 2950 (w), 2370 (w), 1635
(s), 1600 (m), 1550 (w), 1500 (s), 1380 (w), 1300 (m), 1060
(vs), 860 cmϪ1 (w); 1H NMR (CD3CN), ␦ 7·72 (s, 2 H), 7·34
(t, J=8 Hz, 2 H), 7·00 (t, J=8 Hz, 1 H), 6·42 (d, J=8 Hz,
2 H), 3·51 (s, 3 H), 2·59 (s, 3 H), 2·51 (s, 6 H); 13C NMR
(CD3CN), ␦162·2, 159·6, 144·9, 131·3, 130·6, 122·3, 111·8,
39·3, 22·2, 19·5; HRMS, m/z 227·1537 (M+ ϪBF4; calcu-
lated for C15H19N2, m/z 227·1548).
nucleophiles. The para addition of methanol resulting in
NMPA was observed and quantified by HPLC analysis. In
solutions of MPAP and acetonitrile with tetrabutylam-
monium chloride added, ClϪ addition occured at both ortho
and para positions on the ring of the proposed PhMeN+
intermediate [equation (4)]. The para position was con-
sistently favored by approximately 2:1 at various
concentrations of ClϪ . Again, detection and yield measure-
ments were carried out by HPLC analysis.
We considered the possibility that these addition products
did not form via a nitrenium ion intermediate and were, in
fact, the result of an SN2' reaction involving the excited state
MPAP and the nucleophile. If an SN2' mechanism were
operative, then the photodecomposition of MPAP to form
these adduct products would be sensitive to the concentra-
tion of nucleophile present. To test for this alternative
mechanism, experiments were performed in which the
decay of the UV–visible spectrum of MPAP was monitored
as a function of photolysis time and methanol concentration.
Generally, UV measurement proved to be the most
convenient method for monitoring the extent of MPAP
photodecomposition owing to its characteristic band at
365 nm. It was found that the absorbance by MPAP decays
with photolysis time, showing little or no difference in the
extreme cases of 100% methanol as solvent and 100%
acetonitrile. We therefore concluded that photodissociation
of MPAP to PhMeN+ was the operative mechanism, and
addition occurred via a bimolecular reaction of PhMeN+
and nucleophile [equation (5)].
RESULTS AND DISCUSSION
Photolysis of MPAP in acetonitrile solutions with nucleo-
philes present resulted in the formation of the three major
products shown in Scheme 1. These products were detected
and quantified by HPLC based on comparison with the
authentic compounds. Identification of products was further
1
confirmed by H NMR spectroscopy of crude photolysis
mixtures. The results depicted in Scheme 1 are not
unexpected for reactions proceeding via a nitrenium ion
intermediate. The ring-substituted N-methylaniline deriva-
tives, such as N-methyl-p-anisidine (NMPA), are consistent
with other reported results involving nucleophilic attack on
an arylnitrenium ion.12, 20, 21 Arylnitrenium ions exist as
delocalized cations and react with nucleophiles at the para
and ortho ring positions [e.g. para addition illustrated in
equation (1)]. Also, 1,2-sigmatropic shifts have been
observed previously, and serve to explain the formation of
aniline as one of the photoproducts.
The N-tert-butyl substituent in other reported nitrenium
ion chemistry undergoes a 1,2-methyl shift on to the
nitrogen center, leading to an iminium ion intermediate
which upon work-up is hydrolyzed to the N-methylaniline
derivative.12, 21, 22 In the present system, we believe a 1, 2
hydride shift from the N-methyl group followed by
hydrolysis of the iminium ion is the mechanism for the
formation of aniline [equation (2)]. Finally, the detection of
N-methylaniline, the reduction product, is consistent with
previous work,22 and can be attributed to hydrogen atom
abstraction reactions of the radicaloid triplet nitrenium ion
[equation (3)].
Rearrangement products
Another reaction pathway available to the singlet methyl-
phenylnitrenium ion is its rearrangement leading ultimately
to the formation of aniline. This observed product is
attributed to a hydride shift from the N-methyl group on to
the nitrogen atom followed by subsequent hydrolysis of the
resulting iminium ion either during work-up or by the
water–acetonitrile solvent system used in HPLC analysis.
We considered that an imine intermediate may form as
the result of an E2 mechanism involving the concerted
deprotonation of the N-methyl group and elimination of
2,4,6-collidine from excited-state MPAP. An E1 mechanism
involving deprotonation of the N-methyl group of a
nitrenium ion intermediate was also possible as an alter-
native to the hydride shift mechanism. Both the E1 and E2
mechanisms would be sensitive to the concentration of base,
and an enhancement of the yield of aniline at the expense of
other photoproducts should be observed as the concentra-
tion of base is increased. The photolysis of MPAP was
repeated in the presence of 0·5 molar equivalents of
2,4,6-collidine, which is present in these solutions anyway
as a byproduct of these photolyses. With methanol present
at nucleophile, the added collidine caused no enhancement
of the aniline yield relative to NMPA yield (Table 1). The
insensitivity to added base lends support to the hydride shift
mechanism.
Various conditions were employed to measure the effect
on the relative yields of these three product types and to test
the proposed mechanisms for their formation.
Nucleophilic addition to methylphenylnitrenium ion
Nucleophilic addition to the proposed nitrenium ion inter-
mediate was studied using methanol and ClϪ as
© 1997 John Wiley & Sons, Ltd.
JOURNAL OF PHYSICAL ORGANIC CHEMISTRY, VOL. 10, 917–924 (1997)

920
D. CHIAPPERINO AND D. E. FALVEY
Since only two major photoproducts are identified from
Formation of reduction product
the proposed singlet nitrenium ion intermediate, the follow-
ing equality can be derived:
N-Methylaniline (NMA), the reduction product, is detected
in relatively large amounts in these photolyses. Its forma-
tion is believed to be at least in part the result of the triplet
nitrenium ion undergoing two consecutive hydrogen atom
abstractions followed by deprotonation. This has been the
typical source of reduction product in previous work in
which a triplet nitrenium ion was indicated as an inter-
mediate species.22 The presence of the reduction product is
taken as confirmation that the triplet nitrenium ion is
produced to some extent along with the singlet nitrenium
ion, despite the fact that it is the higher energy species. Our
general model is based on the assumption that intersystem
[NMPA]/[PhNH2]=kMeOH[MeOH]/khyd
(6)
The [NMPA]/[PhNH2] ratio from photolysis of MPAP was
measured as a function of [MeOH]. This experiment gave
the linear function with slope kMeOH/khyd of approximately
0·6±0·1 (Figure 1). The rate constant kMeOH for nucleophilic
attack of methanol on arylnitrenium ions can be estimated
as 1ϫ108 Ϫ1 sϪ1 on the basis of previous work.12 The rate
M
constant for rearrangement to the iminiumion intermediate
is therefore estimated to be on the same order of magnitude,
108 sϪ1
.
© 1997 John Wiley & Sons, Ltd.
JOURNAL OF PHYSICAL ORGANIC CHEMISTRY, VOL. 10, 917–924 (1997)

N-METHYL-N-PHENYLNITRENIUM ION
Table 1. Relative yields of photoproducts
921
singlet nitrenium ion only if intersystem crossing of triplet
to singlet can compete with other reactions which consume
the triplet nitrenium ion. Since triplet to singlet intersystem
crossing would be energetically favorable based on the
theoretical calculations of the singlet triplet energy gap, our
model would predict an enhanced yield of the presumed
triplet product but also some products derived from the
singlet nitrenium ion intermediate following triplet to
singlet intersystem crossing.
Conditions
PhNH2
NMPA
NMA
Standarda
Collidineb
NMPAc
1
1
0·2
<0·1
1
2·3
3·0
n.d.
<0·1
2·0
0·4
<0·1
10
10
1·1
Tripletd
CHDe
Extendedf
1
0·6
3·3
We conducted triplet sensitized photolyses of MPAP
using 9-fluorenone as our triplet sensitizer. Direct photolysis
of MPAP is presumably eliminated by use of a 420 nm cut-
off filter in this experiment. These triplet sensitized
photolyses did result in NMA as the major photoproduct,
with only trace amounts of singlet products detected (Table
1), and we take this result as confirmation that NMA is a
photoproduct derived from the triplet nitrenium ion inter-
mediate. The trace yield of singlet products seems to
suggest that there is little intersystem crossing of triplet to
singlet. However, it may be that singlet products aniline and
NMPA react with the triplet excited state of MPAP to be
consumed by secondary photoprocesses, and are in fact
under-represented in the results shown in Table 1. The
nature of these secondary photoprocesses is discussed in
more detail below.
a Standard conditions unless specified otherwise were 2 min
photolysis with a 295 nm cut-off filter, initial concentration of
6·4 mM MPAP and 2·5 M (10%) MeOH in acetonitrile solutions.
b 5·0 m
c 4·4 m
M
collidine present prior to photolysis.
NMPA present prior to photolysis.
M
d Triplet sensitized iradiation with a 420 nm cut-off filter; 9-fluor-
enone present at its solubility limit (ca 4 molar equivalents).
e 0·11
M cyclohexadiene present prior to photolysis.
f Standard conditions, but photolyzed for 20 min.
crossing of the singlet nitrenium ion to the triplet would be
unlikely since the singlet is predicted to be the ground-state
species for this arylnitrenium ion. Therefore, intersystem
crossing of the singlet excited state of MPAP to the triplet
state prior to dissociation is believed to be the source of
triplet nitrenium ion.
We wanted to confirm our assumption that N-methylani-
line was formed via the triplet manifold, and conducted
triplet sensitized irradiation experiments toward this goal.
Triplet sensitized irradiations, which produce the triplet
excited state of MPAP directly through energy transfer,
should produce only triplet nitrenium ion as the result of
dissociation. We would therefore expect products of the
It is worthwhile to consider the results of direct
photolysis experiments that were conducted with added
1,4-cyclohexadiene (CHD). This efficient hydrogen atom
donor present at a concentration of 0·11
M
is apparently able
to trap triplet nitrenium ion to increase the yield of NMA
slightly relative to singlet products. This result implies that
intersystem crossing of triplet nitrenium ion to singlet
nitrenium ion will occur when the concentration of
Figure 1. [NMPA]/[PhNH2] derived from product yields in photolyses of MPAP solutions of various [MeOH]
© 1997 John Wiley & Sons, Ltd. JOURNAL OF PHYSICAL ORGANIC CHEMISTRY, VOL. 10, 917–924 (1997)

922
D. CHIAPPERINO AND D. E. FALVEY
hydrogen atom donors is sufficiently low. Taken together,
the results of the CHD experiment and the triplet sensitized
photolysis lead us to conclude that the singlet nitrenium ion
can be accessed from the triplet nitrenium ion but this
process does not compete well with the fast hydrogen atom
abstractions.
the triplet sensitized irradiation experiments make us
reasonably certain that at least some NMA forms via the
triplet nitrenium ion. However, the greater portion is
probably due to secondary photoprocesses that may not
involve a nitrenium ion species. An explanation that is
consistent with our observations is that photoinduced single
electron transfer occurs from the relatively electron-rich
aniline derivatives which are formed from the reactions of
singlet nitrenium ion. N,N-Dimethylaniline is a well known
photochemical electron donor.27 It is likely that NMPA,
another electron-rich aniline derivative, will behave sim-
ilarly although perhaps not as efficiently. If excited-state
MPAP accepts an electron from NMPA, the result of this
electron transfer process would be p-methoxy-N-methyl-
anilino radical cation and neutral radical MPAP. The neutral
radical MPAP species can then undergo homolytic N—N
bond cleavage to yield neutral collidine and N-methyl-
anilino radical. The latter would abstract a hydrogen atom to
form NMA. Thus a process which does not involve a
nitrenium ion intermediate can produce NMA and would
explain the eventual disappearance of the NMPA being
produced from the nitrenium ion pathway.
This possibly was investigated by conducting the photol-
ysis of MPAP with added NMPA (0·5 molar equivalents)
under identical conditions. We expected an enhanced yield
of NMA at the expense of the photoproducts of the
nitrenium ion pathway since these photoinduced SETs
consume MPAP by a different pathway, reducing the
quantum yield of the nitrenium ion intermediate. The results
did indeed show a greatly enhanced yield of NMA and a
diminished yield of aniline (Table 1). (The change in the
yield of NMPA could not be quantified because of the
relatively large amount added prior to photolysis). These
results are consistent with our proposed electron transfer
mechanism.
Secondary photoprocesses
We had reason to believe that significant secondary
photoprocesses were occurring. What is inferred regarding
the chemical reactivity of the nitrenium ion intermediate is
necessarily based on short photolysis times and very low
percentage conversion of MPAP in our attempt to minimize
these secondary processes. However, we were curious about
the nature of the secondary photochemistry and concerned
about any impact it might have on our interpretation of the
data. The evidence of and nature of these secondary
processes are now discussed in more detail. It was observed
that with 2·5
M
MeOH solutions and short photolysis times,
NMPA is produced in higher yield relative to other
photoproducts. However, experiments conducted using
longer photolysis times show that NMPA is actually
consumed in the photolysis until it is no longer detected.
The yield of aniline also decayed over time, although its
decay was not as pronounced as that for NMPA. Chloro-
substituted N-methylaniline products also showed slow
decay but seemed more impervious to secondary photo-
chemical processes.
The greater complication in proposing an accurate
mechanistic model is the relative yield of NMA. This
product, which has been attributed to a triplet nitrenium ion
intermediate, shows an increasing rate of formation coincid-
ing with the decay of other species. Data illustrating this are
presented in Figure 2, which shows the yields of the three
major products as a function of photolysis time when
methanol is present. At very short photolysis times and low
MPAP conversion, only trace amounts of NMA are
detected. Longer photolysis times result in proportionately
greater amounts of NMA formation and NMPA dis-
appearance. Thus, the growth curve of NMA as a function
of time is sigmoidal in shape [Figure 2(B)]. The results of
In contrast to this dramatic increase in the yield of NMA,
we consider again the results of the experiment with 0·11
M
CHD in which the enhanced yield of NMA is much less
pronounced. This compound is a very efficient hydrogen
atom donor but is not likely to behave as a photochemical
electron donor. The difference in the degree to which CHD
and NMPA enhance the yield of NMA is attributed to their
Figure 2. (a) Time-resolved product yields. (b) Expansion of early times
© 1997 John Wiley & Sons, Ltd.
JOURNAL OF PHYSICAL ORGANIC CHEMISTRY, VOL. 10, 917–924 (1997)

N-METHYL-N-PHENYLNITRENIUM ION
923
Scheme 2
Figure 3. Transient absorbance spectrum 1.6 microseconds after laser pulse.
respective mode of reactivity in two distinctly different
Laser flash photolysis
chemical pathways. CHD traps triplet nitrenium ion, while
NMPA participates in secondary electron transfer processes
ultimately to produce NMA. It has been shown that one-
electron oxidations of various aniline derivatives result in
the polymerization of these substrates, particularly for those
with methoxy substituents.28 It is likely that a polyaniline
derivative is the ultimate photoproduct, which would be
consistent with our proposed electron transfer process and
with the observed decline of NMPA for extended photoly-
ses. A detailed mechanism that incorporates the SET
processes we propose is shown in Scheme 2.
Laser studies were carried out to detect the nitrenium ion
intermediate. The transient absorbance was fairly long-lived
and, therefore, not attributed to the nitrenium ion inter-
mediate which would presumably have a much shorter
lifetime.12 The transient absorbance was very weak with a
maximum at about 470 nm (Figure 3). This ␭max is typical of
anilino radical cations,29 and it is assumed that this transient
absorbance is due to the radical cation of NMA which
would form as the product of a hydrogen atom abstraction
by triplet PhMeN+ . The fact that the absorbance is very
© 1997 John Wiley & Sons, Ltd.
JOURNAL OF PHYSICAL ORGANIC CHEMISTRY, VOL. 10, 917–924 (1997)

924
D. CHIAPPERINO AND D. E. FALVEY
175–276, American Chemical Society, Washington, DC
(1984).
5. M. Famulok and G. Boche, Angew. Chem., Int. Ed. Engl. 28,
468–469 (1989).
6. R. A. Moss and M. Jones, Carbenes. Wiley, New York (1983).
7. G. B. Schuster, Adv. Phys. Org. Chem. 22, 311–361 (1986).
8. M. S. Platz (Ed.), Kinetics and Spectroscopy of Carbenes and
Biradicals. Plenum Press, New York (1990).
9. R. A. McClelland, Tetrahedron 52, 6823–6858 (1996).
10. M. Novak, M. J. Kahley, J. Lin, S. A. Kennedy and T. G.
James, J. Org. Chem. 60, 8294–8304 (1995).
11. J. Michalak, H. B. Zhai and M. S. Platz, J. Phys. Chem. 100,
14028–14036 (1996).
weak despite the high extinction coefficients of these radical
cations is in agreement with our belief that the triplet
nitrenium ion forms to a minimal extent in these reactions.
Purging the system with oxygen weakened the absorbance
at this wavelength. Since molecular oxygen can act as a
quencher of the triplet excited state of MPAP, it would
diminish the yield of triplet nitrenium ion and thus diminish
the yield of N-methylanilino radical cation.
CONCLUSIONS
Experimental results indicate that methylphenylnitrenium
ion is produced in the photolysis of N-(methylphenyl-
amino)-2,4,6-trimethylpyridinium tetrafluoroborate. The
reactions of singlet and triplet nitrenium ions lead to
different products depending on the spin state that is
formed. Similarly to the previously reported reactivity of
arylalkylnitrenium ions, nucleophilic addition to the aryl
substituent is observed, as is rearrangement via 1,2-shifts
from the alkyl substituent of the nitrenium center. Both of
these product types are attributed to a singlet-state nitrenium
ion intermediate. Parent amines form, and are attributed to
two different pathways that both involve the formation and
reaction of radical species. One of these pathways is
proposed to be a secondary photoprocess. This process
involves photoinduced electron transfer in which the
primary photoproducts, various aniline derivatives, act as
electron donors to the pyridinium ion, causing its decom-
position. The other pathway resulting in N-methylaniline is
the hydrogen atom abstractions of the triplet nitrenium ion.
12. R. J. Robbins, L. L.-N. Yang, G. B. Anderson and D. E. Falvey,
J. Am. Chem. Soc. 117, 6544–6552 (1995).
13. S. D. Peyerimhoff and R. J. Buenker, Chem. Phys. 42, 167–176
(1979).
14. G. P. Ford and P. S. Herman, J. Am. Chem. Soc. 111,
3987–3996 (1989).
15. C. J. Cramer, F. J. Dulles, J. W. Storer and S. E. Worthington,
Chem. Phys. Lett. 218, 387–394 (1994).
16. S. A. Glover and A. P. Scott, Tetrahedron 45, 1763–1776
(1989).
17. G. P. Ford and P. S. Herman, Theochem 82, 269–282 (1991).
18. D. E. Falvey and C. J. Cramer, Tetrahedron Lett. 33,
1705–1708 (1992).
19. C. J. Cramer, F. J. Dulles and D. E. Falvey, J. Am. Chem. Soc.
116, 9787–9788 (1994).
20. R. J. Moran and D. E. Falvey, J. Am. Chem. Soc. 118,
8965–8966 (1996).
21. G. B. Anderson, L. L.-N. Yang and D. E. Falvey, J. Am. Chem.
Soc. 115, 7254–7262 (1993).
22. S. Srivastava and D. E. Falvey, J. Am. Chem. Soc. 117,
10186–10193 (1995).
23. R. A. Abramovitch, J. M. Beckert, P. Chinnasamy, H. Xiaohua,
W. Pennington and A. R. V. Sanjivamurthy, Heterocycles 28,
623–628 (1989).
24. R. A. Abramovitch and Q. Shi, Heterocycles 37, 1463–1466
(1994).
25. R. Stoermer, Chem. Ber. 38, 2523–2534 (1898).
26. A. T. Balaban and A. J. Boulton, in Chemical Syntheses, edited
by H. E. Baumgarten, Coll. Vol. V, pp. 1112–1113. Wiley, New
York (1973).
ACKNOWLEDGEMENTS
We thank the National Science Foundation for funding and
Mr Karsten Hesse for carrying out some preliminary
experiments.
REFERENCES
1. R. A. Abramovitch and R. Jeyaraman, in Azides and Nitrenes:
Reactivity and Utility, edited by E. F. V. Scriven, pp. 297–357.
Academic Press, Orlando, FL (1984).
27. S. L. Murov, Handbook of Photochemistry, p. 119. Marcel
Dekker, New York (1973).
28. G. D’Aprano, E. Proynov, M. Lebœuf, M. Leclerc and D. R.
Salahub, J. Am. Chem. Soc. 118, 9736–9742 (1996).
29. X. Zhang, S.-R. Yeh, S. Hong, M. Freccero, A. Albini, D. E.
Falvey and P. S. Mariano, J. Am. Chem. Soc. 116, 4211–4220
(1994).
2. J. A. Miller, Cancer Res. 30, 559–576 (1970).
3. E. Kriek, Biochem. Biophys. Acta, 355, 177–203 (1974).
4. R. C. Garner, C. N. Martin and D. B. Clayson, in Chemical
Carcinogens edited by C. E. Searle, 2nd ed., Vol. 1, pp.
© 1997 John Wiley & Sons, Ltd.
JOURNAL OF PHYSICAL ORGANIC CHEMISTRY, VOL. 10, 917–924 (1997)