Diphenylnitrenium ion: Cyclization, electron transfer, and polymerization reactions

Article abstract of DOI:10.1021/jo050598z

Reactions of diphenylnitrenium ion were examined using laser flash photolysis (LFP), product analysis, and computational modeling using density functional theory (DFT). In the absence of trapping agents, diphenylnitrenium ion cyclizes to form carbazole. On the basis of laser flash photolysis experiments and DFT calculations it is argued that this process is a concerted cyclization/ proton transfer that forms the H-4a tautomer of carbazole. Additional LFP experiments and product studies show that diphenylnitrenium ion reacts with electron-rich arenes (e.g., N,N-dimethylaniline, diphenylamine, and carbazole) through an initial one-electron transfer. The radical intermediates formed in this step then couple to form dimeric products. Secondary reactions between the diphenylnitrenium ion and these dimers results in the formation of oligomeric materials.

Full text of DOI:10.1021/jo050598z

Diphenylnitrenium Ion: Cyclization, Electron Transfer, and
Polymerization Reactions
Andrew C. Kung, Sean P. McIlroy, and Daniel E. Falvey*
Department of Chemistry and Biochemistry, University of Maryland, College Park, Maryland 20742-2021
falvey@umd.edu
Received March 24, 2005
Reactions of diphenylnitrenium ion were examined using laser flash photolysis (LFP), product
analysis, and computational modeling using density functional theory (DFT). In the absence of
trapping agents, diphenylnitrenium ion cyclizes to form carbazole. On the basis of laser flash
photolysis experiments and DFT calculations it is argued that this process is a concerted cyclization/
proton transfer that forms the H-4a tautomer of carbazole. Additional LFP experiments and product
studies show that diphenylnitrenium ion reacts with electron-rich arenes (e.g., N,N-dimethylaniline,
diphenylamine, and carbazole) through an initial one-electron transfer. The radical intermediates
formed in this step then couple to form dimeric products. Secondary reactions between the
diphenylnitrenium ion and these dimers results in the formation of oligomeric materials.
Introduction
sulted in a recent increase in interest in these species.1
2-14
The present report concerns the reactions of N,N-diphen-
ylnitrenium ion (1). Previous studies using chemical
Nitrenium ions are highly reactive transient species
that contain a positively charged and dicoordinate nitro-
trapping1 laser flash photolysis (LFP), time-resolved
4,15
16
1
-3
gen atom.
These short-lived intermediates are of
17
infrared spectroscopy, and time-resolved Raman spec-
troscopy18 have established that this species is generated
upon photolysis of 1-(N,N-diphenylamino)-2,4,6-trimeth-
interest due to their involvement in carcinogenic DNA-
damaging reactions4 and their postulated involvement
-7
8
in the synthesis of polyaniline (PANI). There is also some
ylpyridinium ions (2), as shown in Scheme 1. It was also
recent interest in using these species in synthetic
+
shown that Ph
2
N , like most simple aromatic nitrenium
9
-11
chemistry.
The development of methods for directly
ions,1
9-24
is a ground-state singlet.
15,25
Its known decay
detecting nitrenium ions using fast laser methods re-
*
To whom correspondence should be addressed. Fax: 301-314-9121.
(12) Anderson, G. B.; Falvey, D. E. J. Am. Chem. Soc. 1993, 115,
9870-9871.
(
1) Abramovitch, R. A.; Jeyaraman, R. Nitrenium Ions. In Azides
and Nitrenes: Reactivity and Utility; Scriven, E. F. V., Ed.; Academic:
Orlando, FL, 1984; pp 297-357.
(13) McClelland, R. A. Tetrahedron 1996, 52, 6823-6858.
(14) Moran, R. J.; Falvey, D. E. J. Am. Chem. Soc. 1996, 118, 8965-
8966.
(
2) Falvey, D. E. Nitrenium Ions. In Reactive Intermediate Chem-
istry; Moss, R. A., Platz, M. S., Maitland Jones, J., Eds.; Wiley-
Interscience: Hoboken, NJ, 2004; Vol. 1; pp 593-650.
(15) Moran, R. J.; Cramer, C. J.; Falvey, D. E. J. Org. Chem. 1996,
61, 3195-3199.
(16) McIlroy, S.; Moran, R. J.; Falvey, D. E. J. Phys. Chem. A 2000,
104, 11154-11158.
(17) Srivastava, S.; Toscano, J. P.; Moran, R. J.; Falvey, D. E. J.
Am. Chem. Soc. 1997, 119, 11552-11553.
(18) Zhu, P.; Ong, S. Y.; Chan, P. Y.; Poon, Y. F.; Leung, K. H.;
Phillips, D. L. Chem. Eur. J. 2001, 7, 4928-4936.
(19) Cramer, C. J.; Dulles, F. J.; Falvey, D. E. J. Am. Chem. Soc.
1994, 116, 9787-9788.
(
3) Novak, M.; Rajagopal, S. Adv. Phys. Org. Chem. 2001, 36, 167-
2
54.
(
4) Famulok, M.; Bosold, F.; Boche, G. Angew. Chem., Int. Ed. Engl.
1
989, 28, 337-338.
(
5) Kadlubar, F. F. DNA Adducts of Carcinogenic Amines. In DNA
Adducts Identification and Biological Significance; Hemmink, K.,
Dipple, A., Shuker, D. E. G., Kadlubar, F. F., Segerb a¨ ck, D., Bartsch,
H., Eds.; University Press: Oxford, UK, 1994; pp 199-216.
(
6) Hoffman, G. R.; Fuchs, R. P. P. Chem. Res. Toxicol. 1997, 10,
(20) Sullivan, M. B.; Brown, K.; Cramer, C. J.; Truhlar, D. G. J.
Am. Chem. Soc. 1998, 120, 11778-11783.
(21) McIlroy, S.; Cramer, C. J.; Falvey, D. E. Org. Lett. 2000, 2,
2451-2454.
3
3
47-359.
(
7) Novak, M.; Kazerani, S. J. Am. Chem. Soc. 2000, 122, 3606-
616.
(
8) Wei, Y.; Jang, G.-W.; Chan, C. C.; Hsueh, K. F.; Hariharan, R.;
(22) Ford, G. P.; Herman, P. S. J. Am. Chem. Soc. 1989, 111, 3987-
3996.
Patel, S.; Whitecar, C. K. J. Phys. Chem. 1990, 94, 7716-7721.
9) Rodrigues, J. A. R.; Abramovitch, R. A.; de Sousa, J. D. F.; Leiva,
(
(23) Li, Y.; Abramovitch, R. A.; Houk, K. N. J. Org. Chem. 1989,
54, 2911-2914.
G. J. Org. Chem. 2004, 69, 2920-2928.
(10) Wardrop, D. J.; Zhang, W. Org. Lett. 2001, 3, 2353-2356.
11) Wardrop, D. J.; Birge, M. S. Chem. Commun. 2004, 1230-1231.
(24) Falvey, D. E.; Cramer, C. J. Tetrahedron Lett. 1992, 33, 1705-
1708.
(
1
0.1021/jo050598z CCC: $30.25 © 2005 American Chemical Society
Published on Web 05/26/2005
J. Org. Chem. 2005, 70, 5283-5290
5283

Kung et al.
SCHEME 1. Decay Pathways for
Diphenylnitrenium Ion (1)
TABLE 1. Yields of 8 and 9 from Photolysis of Various
Initial Concentrations of 2 in Various Solvents
solvent
concn of 2 (mM)
8 (%)
9 (%)
PhCH3
PhCH3
CH3CN
CH3CN
CH3CN
CH2Cl2
CH2Cl2
0.097
4.6
0.097
0.34
4.5
0.064
4.85
26
35
10
7
27
10
67
22
36
36
1.4
5
26
54
aromatic compounds. This was occasioned by our interest
in the polymerization mechanism, but it is also relevant
to newly emerging synthetic applications of nitrenium
ion chemistry and to the issue of DNA damage by
nitrenium ions. Our LPF and product studies indicate
that with easily oxidized arenes, the reaction with the
nitrenium ion can proceed through an initial electron
transfer, followed by coupling to yield dimeric adducts.
reactions include (1) addition of simple nucleophiles (e.g.,
halides, alcohols) to the ortho and para positions on the
1
4,15,17
phenyl rings,
(2) addition of electron-rich aromatic
compounds (e.g., 1,3,5-trimethoxybenzene) to the nitre-
Results and Discussion
26
nium nitrogen atom via a σ complex intermediate, and
3) abstraction of hydride from a suitable donor (e.g., 1,4-
(
A. Carbazole Formation. Three lines of evidence
support our conclusion that carbazole (9) formation
represents the primary decay pathway for 1 in the
absence of trapping agents.
(1) The yields of carbazole relative to both dipheny-
lamine and the polymeric materials (see below) increase
when photolysis of 2 is carried out in samples at high
dilution. Table 1 shows the yields of carbazole and
diphenylamine when 2 was photolyzed with room light
in various solvents at various starting concentrations. In
each case, decreasing starting material concentration
results in a higher yield of carbazole. This is consistent
with, but does not require, that both 8 and the polymeric
materials be derived from accumulated photoproducts.
16
cyclohexadiene). These reaction pathways are sum-
marized in Scheme 1.
When diphenylnitrenium ion (1) is generated in the
absence of trapping agents, stable products include
diphenylamine (8) and carbazole (9), which are formed
in relatively low yields.14 Along with these two products,
large amounts of colored materials were formed. The
latter had not been characterized, but they were pre-
sumed to be high molecular weight oligomers related to
poly(diphenylamine).
The present study was undertaken with three goals.
First, we wanted to elucidate the primary decay pathway
for diphenylnitrenium ion in the absence of traps. It was
not clear at the outset of this study whether the formation
of diphenylamine and the other products occurred con-
currently with the carbazole pathway or resulted from
secondary decay reactions. Through a combination of
LFP, product analysis, and computational modeling, we
provide evidence that cyclization is the primary decay
pathway and that the other products result from second-
ary reactions.
(
2) The decay rate of the diphenylnitrenium ion shows
clean first-order kinetics, and the rate constant does not
vary significantly with solvent. This was established by
LFP experiments, wherein diphenylnitrenium ion was
generated by pulsed laser (355 nm, 20-40 mJ/pulse, 4-8
3 2 2
ns) photolysis of 2 in CH CN, toluene, and CH Cl . The
decay of 1 was measured at its 440 nm absorption
maximum, and the rate constants were determined to
5
-1
A second goal was to determine the nature of the
uncharacterized colored materials as well as the mech-
anism of their formation. Obviously, this serves the first
goal by differentiating primary and secondary decay
processes. Increasing commercial importance of PANI
and its derivatives also encouraged us to examine this
process. Formation of poly(diphenylamine) in these reac-
tions may provide additional insight into the possible
involvement of nitrenium ions in PANI synthesis, as well
as to help identify new methods by which these important
polymers could be prepared. Below, we demonstrate by
MALDI-TOF-MS that polymers are formed in these
reactions and LFP experiments provide some insights
into the mechanism by which this occurs.
be (7.7 ( 1.4) × 10 s in each of these solvents. The
origin of diphenylamine through H atom (or hydride)
abstraction from the solvent would require that each of
these solvents react at equal rates, or more precisely, that
any change in H atom transfer rates would be balanced
by a commensurate change in cyclization rate. Such a
mechanism is highly improbable. We also note that the
clean first-order kinetics observed in these experiments
are also inconsistent with the polymeric materials form-
ing directly from coupling of two nitrenium ions. It should
also be noted that the LFP experiments are carried out
under conditions where accumulation of the photoprod-
ucts is expected to be negligible.
(3) Likewise, the decay rate constant shows no signifi-
The third goal was to determine the mechanism by
which diphenylnitrenium ion reacts with electron-rich
cant variation with the concentration of 2. The decay rate
constants measured over the concentration range of 15-
2
40 µM and show variation that is indistinguishable from
(
25) Cramer, C. J.; Falvey, D. E. Tetrahedron Lett. 1997, 38, 1515-
the experimental uncertainty.
1
518.
The formation of carbazole from diphenylnitrenium ion
formally requires the loss of one aryl proton and the net
(26) McIlroy, S.; Falvey, D. E. J. Am. Chem. Soc. 2001, 123, 11329-
1
1320.
5284 J. Org. Chem., Vol. 70, No. 13, 2005

Diphenylnitrenium Ion
SCHEME 2. Nazarov-Type Cyclization of Diphenylnitrenium Ion to Carbazole
TABLE 2. Significant UV-vis Absorption Bands (in nm)
and Oscillator Strengths (i parentheses) Predicted for 1,
migration of one proton from the phenyl ring to the
nitrogen. One possible mechanism for this process would
an electrocyclization pathway, analogous to the Nazarov
cyclization as shown in Scheme 2. The initial ring closure
would form the protonated carbazole tautomer 10. Con-
version of this intermediate into stable products would
require a deprotonation to give the carbazole tautomer
1
0, and 11 Compared to Assignments from LFP
Experiments
method
1
10
11
TD-DFT
ZINDO
expt
645 (0.14)
679 (0.26)
660
449 (0.22)
455 (0.33)
n.d.
373 (0.16)
365 (0.28)
340
11 followed by a net 1,7 proton shift to give the observed
product. Although the experiments and calculations
described below cause us to favor a modification of this
mechanism, Scheme 2 serves as a useful starting point
for these discussions. This is because Scheme 2 identifies
two limiting intermediates, 10 and 11, that should be
considered in light of any data.
in concentrations too low to intercept significant fractions
of 1, also quenches this new intermediate with a rate
8
-1 -1
constant of 7 × 10 M s .
Further verification of the assignment comes from
theoretical calculations. Density functional theory (B3LYP/
6-31G(d,p)) was used to optimized geometries for 10 and
11. Similar calculations on 1 have been previously
reported. From these geometries, UV-vis absorption
bands were computed using both time-dependent density
To characterize the primary decay processes, we un-
dertook more detailed LFP experiments on 2, aimed at
detecting the possible intermediates in this process.
Figure 1 shows transient UV-vis spectra taken various
times following pulsed laser photolysis (355 nm, 20 mJ,
27
functional theory (TD-DFT) as well as the semiempiri-
-8 ns) of 2 in toluene. As previously reported,11 nitre-
28
4
cal ZINDO method. The presumably more accurate
nium ion 1 is formed concurrently with the excitation
pulse and can be detected through its strong signals at
B3LYP geometries were used in the ZINDO calculations.
The predicted UV-vis signals are shown in Table 2. In
general, good agreement between the two computational
predictions and the experimental values was observed.
In particular, we note that both TD-DFT and ZINDO
predictions are more consistent with the assignment of
the 340 nm intermediate to 11.
4
40 and 660 nm. The decay of these signals is ac-
companied by a first-order growth of a new species with
max ) 340 nm. The new intermediate lives for ca. 15 µs.
Similar behavior was observed in both CH Cl and
CH CN.
λ
2
2
3
Following Scheme 2, the carrier for this 340 nm signal
could be either the initial cyclization product 10 or the
carbazole tautomer 11. The former species decays through
loss of a proton, a reaction that would be accelerated by
the addition of base and slowed by addition of a strong
acid. The latter species decays by a formal proton shift,
a reaction that could be catalyzed by either an acid or a
base. Two observations cause us to assign the signal to
The electocyclization pathway was further examined
by modeling the transition state leading to 10. With the
(B3LYP/6-31G(d,p) structures for 1 and 10, the transition
state for the reaction was located by means of the Berny
algorithm at the AM1 level and then reoptimized at the
DFT level. Analytic vibrational frequency calculations
confirmed that structures 1 and 10 (Figure 2) are local
minima, having no imaginary frequencies and that the
transition state structure (12) is a saddle point, having
a single imaginary frequency. In the absence of solvation,
the conversion of 1 to 10 is predicted to be endothermic
by ca. 13 kcal/mol, having a barrier of 16.6 kcal/mol. On
the other hand, when the structures were reoptimized
using a polarized-continuum solvation model, the reac-
tion is predicted to be ca. 7 kcal/mol endothermic, with a
barrier of 10 kcal/mol. The energies of 1, 10, and 12 are
listed in Table 3, along with some selected geometric
parameters, including C-N-C bond angle and the
distance between the two carbons that become covalently
attached in 10. By these measures, product 10 is very
similar in geometry to the transition state.
1
1. First, the decay rate of the observed species is
accelerated both by added acid as well as added base.
Specifically, we find that addition of 48% aqueous HClO
quenches the 340 nm signal with a rate constant of 8.2
4
8
-1 -1
10 M s . Likewise, addition of 1,3,5-collidine as a base,
Given that the neutral intermediate 11 is observed to
appear immediately following the decay of the nitrenium
ion and that formation of 10 is likely to be endothermic
with a significant barrier, it occurred to us that 11 could
(
27) Marques, M. A. L.; Gross, E. K. U. Annu. Rev. Phys. Chem.
2
004, 55, 427-455.
(
28) Zerner, M. C. In Reviews in Computational Chemistry; Boyd,
FIGURE 1. Transient UV-vis absorption spectra derived
from pulsed photolysis of 2 in toluene.
D. B., Lipkowitz, K. B., Eds.; VCH: New York, 1991; Vol. 2; pp 313-
366.
J. Org. Chem, Vol. 70, No. 13, 2005 5285

Kung et al.
FIGURE 3. Thermodynamic cycle used to estimate the acidity
of hypothetical intermediate 10.
especially facile. Moreover, it suggests that 10 might not
exist as a discrete intermediate, but that proton transfer
to the counterion and/or the solvent would be concerted
with its formation.
∆
G - ∆G
9 10
pK10 ) pK +
(1)
as -4.9. Even
9
2.3RT
Bessiere30 estimates the pK
of HBF
a
4
allowing for some rather large (>16 kcal/mol) discrep-
ancies in the estimation of ∆G - ∆G10, proton transfer
to the counterion should be highly favored. In fact, the
9
+
a 4
pK ’s of both HBF and 9H were determined in polar
protic media, which would favor the formation of ions. It
FIGURE 2. Calculated (B3LYP/6-31G(d,p) geometries for the
conversion of 1 to 10.
is expected in the aprotic media employed here both that
-
the favorability of proton transfer from 10 to the BF
4
TABLE 3. Energies and Geometric Parameters for
Intermediates and the Transition State in the
Conversion of 1 to 10
counterion is underestimated as a result. Thus, we
suggest that the mechanism for carbazole formation in
solution is the cyclization reaction coupled to a transfer
<
C1NC1′
(deg)
R(C2-C2′)
<H2C2C2′H2′
E
-
of the C2 proton to the BF
4
counterion as shown in
structure
(Å)
(deg)
(hartrees)
Scheme 3. We note that proton transfer to the 2,4,6-
trimethylpyridine leaving group is also possible. Further
computational and experimental studies aimed at deter-
mining the degree of proton transfer in the transition
state would certainly be interesting but beyond the scope
of the present investigation.
B. Secondary Products. We next turned our atten-
tion to the formation of diphenylamine (8) from photolysis
of 2. The former results from a net two-electron reduction
of nitrenium ion 1. However, the source of the reducing
equivalents has never been elucidated. The possibility
that they might derive from the solvent (via either H
atom or hydride transfer) seems unlikely, given the
invariance of the decay rate with the different solvents
employed. Therefore, we considered the possibility that
this might be derived from a reaction of 1 with ac-
cumulated photoproducts.
1
1
1
126.5
109.3
107.5
3.077
2.037
1.621
-97.0
-165.7
-166.3
-517.8558
-517.8325
-517.8450
2 (TS)
0
SCHEME 3. Concerted Proton Transfer/
Cyclization Mechanism
form through a concerted process. Therefore, we consid-
ered a pathway whereby the cyclization step is coupled
to the proton loss (Scheme 3).
The acidity of structure 10 is not known, but the lack
of aromaticity in any of the rings implies that it would
be considerable. We can estimate its acidity (pK10) using
the thermodynamic cycle indicated in Figure 3. The latter
To identify possible accumulated photoproducts, the
photolyzates from 2 were analyzed in more detail. In
particular we scaled up the reaction and carried out
chromatographic separations on the mixture. These
experiments reveal that, in addition to the major photo-
products 8 and 9, two dimeric species (13 and 14) were
also formed (Figure 4). Product 13 clearly derives from
coupling between a diphenylnitrenium ion and carbazole.
Interestingly, this adduct forms from addition of the
nitrogen of carbazole to the 4-position of the nitrenium
ions arylsubstituent. This adduct was distinguished from
9
relates pK10 to the acidity of carbazole (pK ) and the free
energy changes converting carbazole to its tautomer 11
+
(
(
∆G
9
) and N-protonated carbazole (9H ) to its isomer 10
) 28 kcal/mol and ∆G10 ) 46 kcal/
∆G10). Values of ∆G
9
mol were obtained from AM1 calculations, assuming that
zero-point vibrational energies, entropy changes, as well
as solvation differences for these reactions to be negli-
1
3
its isomers by the number of signals in the C NMR
spectrum. Assuming rapid rotation about single bonds,
product 13 is predicted to give 14 unique signals, which
is exactly what is observed. We also considered the
possible coupling of the nitrenium center with the 3-posi-
tion on carbazole, which would have 16 unique carbons.
gible. From the studies of Chen et al.29 we obtain pK
)
9
-
6. These values are applied to eq 1 to estimate the pK
a
of 10 to be -19. This value is certainly consistent with
intuitive expectation that deprotonation of 10 should be
(
29) Chen, H. J.; Hakka, L. E.; Hinman, R. L.; Kresge, A. J.;
Whipple, E. B. J. Am. Chem. Soc. 1971, 93, 5102-5107.
(30) Bessiere, J. Anal. Chim. Acta 1970, 52, 55-63.
5286 J. Org. Chem., Vol. 70, No. 13, 2005

Diphenylnitrenium Ion
FIGURE 4. Structures of adducts obtained from photolysis
of 2 in the presence of carbazole yielding 13, diphenylamine
yielding 14, and N,N-dimethylaniline yielding 15.
TABLE 4. Yields of Diphenylamine (8) and Adducts 13
and 14 from Photolysis of 2 in CH3CN with Varying
Concentrations of Added Carbazole
yield (µM)
carbazole (µM)
66
8
13
14
2
52
78
99
110
320
830
26
tr
tr
1
700
800
FIGURE 6. A portion of the MALDI-TOF-MS derived from
photolysis of 1 in CH CN.
4
3
TABLE 5. Trapping Rate Constants and Product Yields
for the Reaction of Various Arenes with 1
ktrap (M- s-1
1
)
8 (%) adduct (%) LFP signal (nm)
arene
9
PhN(CH3)2
Ph2NH
Ph3N
7.3 × 10
57
n.d.
76
39
51
15 (17)
14 (16)
n.d.
480
1
9
9
9
0
1.3 × 10
7.0 × 10
2.0 × 10
1.3 × 10
690
360, 560
480
680
1
,4-DMB
n.d.
Carbazole
13 (24)
The formation of these oligomeric species is further
confirmed by MALDI-TOF mass spectra of the reaction
products. Figure 6 shows a portion of the latter in the
region corresponding to tetramers having varying amounts
of carbazole and diphenylamine. Similar sets of peaks
are detected at 833 and 1000 amu, corresponding to
pentamers and hexamers, respectively.
FIGURE 5. Steady-state UV-vis absorption spectra taken
at various times following photolysis of 2 in a CH CN solution.
3
A structure derived from coupling the 4-position of the
nitrenium ion fragment with the 3- or 1-positions of
carbazole would each give 20 unique signals.
The other dimeric species, 14, derives from the coupling
of a nitrenium ion with diphenylamine. This was con-
firmed by carrying out photolyses with added diphenyl-
amine. Under these conditions 14 is the major product.
As with 13, this species can be distinguished from its
isomers on the basis of the number of 13C NMR peaks.
For compound 14, 12 unique signals are predicted and
C. Reactions of 1 with Electron Rich Aromatics.
The observations of the oligomers from the nontrap
photolysis lead us to explore the reactions of diphenylni-
trenium ion with diphenylamine and carbazole, as well
as with several related arylamines. In this study, we
focused our attention on arene traps having low oxidation
potentials (Eox < +1.4 V vs SCE). Table 5 shows rate
constants for the reaction of 1 with these traps. These
rate constants were determined from LFP experiments
wherein 1 was generated from photolysis of 2 and the
pseudo-first-order decay rate constant, kobs, of 1 was
determined with various concentrations of trap. These
data were analyzed using eq 2, and in each case a good
linear correlation was obtained. As is apparent from
Table 5, these traps react with diphenylnitrenium ion at
1
2 are observed.
Carbazole is postulated to be the primary photoprod-
uct. In that case, diphenylamine and 13 would be derived
from secondary reactions of 1 with this product. To test
this point, we generated the former in the presence of
excess carbazole and examined product distributions by
GC. As shown in Table 4, the yields of both the adduct
10
-1 -1
3
near the diffusion limit of CH CN (1.9 × 10
M
s ).
1
3 and 8 increase as the carbazole concentration is
raised.
The structures of the two minor products suggested
kobs ) k + ktrap[trap]
(2)
0
Each of the traps in Table 5 react with diphenylnitre-
nium ion via an electron-transfer reaction. Figure 7
shows transient absorption spectra acquired following
pulsed laser photolysis of 2 in the presence of N,N,N-
triphenylamine, N,N-dimethylaniline, 1,4-dimethoxyben-
zene, N,N-diphenylamine, and carbazole. In each case,
following the decay of nitrenium ion 1, we detect radical
that the remaining materials might be derived from
higher order addition reactions creating mixed oligomers
of carbazole and diphenylamine. Two observations sup-
port this. Figure 5 shows changes in the UV-vis spectra
3
resulting from photolysis of 2 in CH CN with a handheld
shortwave UV lamp. The product absorption band at 570
nm signals the formation of an extended π-system.
Similar visible absorption bands are characteristic of both
poly(diphenylamine) and poly(3,6-carbazole).31
(
31) Wen, T.-C.; Chen, J.-B.; Gopalan, A. Mater. Lett. 2002, 57, 280-
290.
J. Org. Chem, Vol. 70, No. 13, 2005 5287

Kung et al.
FIGURE 7. Transient absorption spectra from LFP of 2 in CH
3
CN with (a) 0.15 mM N,N,N-triphenylamine, (b) 0.59 N,N-
4
dimethylaniline, (c) 0.82 mM 1,4-dimethoxybenzene, (d) 0.6 mM N,N-diphenylamine, and (e) 3.0 mM carbazole with HBF .
ion intermediates resulting from electron transfer.32
Specifically, radical cations for N,N,N-triphenylamine
may be obscured by the stronger signal for diphenylamine
cation radical.
(
λ
max ) 373 and 560 nm), N,N-dimethylaniline (λmax
)
The stable products from each of these reactions was
analyzed. Aside from the adducts listed in Table 5,
significant amounts of the polymers were also formed.
In the case of diphenylamine, it was impractical to assess
the yield of any additional diphenylamine, which would
have corresponded to a relatively small increase in the
signal due to unconverted trap. In the case of N,N,N-
triphenylamine and 1,4-dimethoxybenzene, no adducts
formed in sufficient yields to allow for characterization.
In the case of N,N-dimethylaniline, an adduct was
isolated. This compound is derived from coupling the
4-position of the N,N-dimethylaniline with the nitrogen
from diphenylnitrenium ion as seen in Figure 4.
On the basis of these LFP and product analysis
experiments, we provide Scheme 4, which is consistent
with all current results. Nitrenium ion 1 forms carbazole
4
40 nm), 1,4-dimethoxybenzene (λmax ) 410 and 425 nm),
and N,N-diphenylamine (λmax ) 690 nm) are detected,
along with a weaker, broad absorption for diphenylamine
neutral radical, which has a maxima near 650 nm.33 In
the case of carbazole the spectra are complicated in CH
CN as a result of the acidity of both the carbazole cation
radical (pK ) 1.5) and the diphenylamine cation radical
3
-
a
3
4
(pK
a
) 2.5). However, when experiments were carried
, the well-resolved spectrum of the
out with added HBF
4
diphenylamine cation radical (λmax ) 680 nm) was
observed. The corresponding carbazole radical was not
detected, but its relatively weak absorption (610 nm)35
(32) Shida, T. Electronic Absorption Spectra of Radical Ions; Elsevi-
er: Amsterdam, 1988.
(
33) Leyva, E.; Platz, M. S.; Niu, B.; Wirz, J. J. Phys. Chem. 1987,
1, 2293-2298.
34) Bordwell, F. G.; Zhang, X.-M.; Cheng, J.-P. J. Org. Chem. 1993,
8, 6410-6416.
9
5
(
(35) G. A. DiLabio; Litwinienko, G.; Lin, S.; Pratt, D. A.; Ingold, K.
U. J. Phys. Chem. A 2002, 106, 11719-11725.
5288 J. Org. Chem., Vol. 70, No. 13, 2005

Diphenylnitrenium Ion
SCHEME 4. Dimerization and Oligomerization of
Intermediates from Diphenylnitrenium Ion
interesting to note that the two mechanisms are really
not all that distinct. Generalizing from Scheme 4, it
would seem that any encounter between a nitrenium ion
and an aniline derivative (or oligomer) would produce the
same intermediates derived from the radical mechanism.
Further investigations into the specific intermediates in
PANI formation are currently being pursued.
Conclusions
The experiments described here further develop the
view of diphenylnitrenium ion chemistry. First, it is clear
that the primary decay pathway (in solution) for diphen-
ylnitrenium ion is a concerted electrocyclization/depro-
tonation reaction that provides carbazole via its H-4a
tautomer (11). The latter can be detected in LFP experi-
ments. Laser flash photolysis experiments and product
analysis show that with relatively easily oxidized arenes
the reaction with diphenylnitrenium ion goes through an
initial electron transfer rather than direct formation of
a σ complex. The electron-transfer reactions lead to
diphenylamine and, in many cases, oligomeric products
derived from oxidation and coupling reactions. This
appears to be the first direct observation of one-electron-
transfer reactions from a singlet nitrenium ion.
Experimental Section.
in the absence of traps. As the carbazole accumulates, 1
reacts with this product via a net H atom abstraction (we
suggest that this is a stepwise electron transfer/proton-
transfer process, but the current data are not conclusive
on this point). Coupling of the 9-carbazolyl radical with
the diphenylamine cation radical forms the dimeric
product 13. The cation radical of 8 can then either couple
with another cation radical of 8 to form 14 or abstract
an additional electron from the various photoproducts to
Calculations. All geometry optimizations and vibrational
frequency calculations were carried out using the Gaussian
4
0
03 suite of programs. The values in Tables 2 and 3, as well
as Figure 2, were calculated using density functional theory
and in particular the hybrid B3LYP functional, composed of
Becke’s B3 three-parameter gradient-corrected exchange func-
4
1,42
tional
with the LYP correlation functional of Lee, Yang,
43
44
and Parr as originally described by Stevens et al. Unless
noted, the 6-31G(d,p) basis set was used for these calcula-
4
5
tions. The thermochemical analysis in Figure 3 used AM1-
•
46
+
form diphenylamine. Likewise, carbazolyl radical (9 ) can
derived energies for 9, 9H , 10, and 11.
also be expected to oxidize (or abstract a H atom from)
any of the oligomeric species. Subsequent oxidation of the
dimeric products by 1 is followed by coupling reactions
creating trimers and tetramers. Higher order oligomers
are generated by analogous oxidation/coupling processes.
Thus, each nitrenium ion is capable of oxidizing two
oligomers, which in turn can couple to form larger
species. The present studies do not permit a definitive
assignment of the structures of any of the oligomers
larger than the dimers 13 and 14 because of the com-
plexity of these mixtures. However, the similarity of the
UV-vis spectrum to that obtained from poly(diphenyl-
amine) suggests that their structures are analogous.
Earlier mechanistic studies of oxidative aniline polym-
erization considered two mechanisms. The first postu-
lated chain elongation via coupling of a nitrenium ion
with a aniline derivative.8 More recent work suggests
that the chain growth occurs through coupling of two
Laser Flash Photolysis Studies. Laser flash photolysis
experiments were performed at 298 K using a Nd(YAG) laser.
Second, third, and fourth harmonic generator crystals were
used to create output wavelengths at 355 and 266 nm. The
UV-vis light source was produced by a CW 350 W Xe arc
lamp. The solution to be photolyzed, unless otherwise noted,
(
40) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; J. A. Montgomery, J.; Vreven, T.; Kudin,
K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone,
V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.;
Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa,
J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene,
M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Adamo, C.;
Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A.
J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma,
K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.;
Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.;
Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui,
Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.;
Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challa-
combe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.;
Gonzalez, C.; Pople, J. A. Gaussian03, Revision B.03.; Gaussian, Inc.:
Pittsburgh, PA, 2003.
,36
aniline (or aniline oligomer) cation radicals and that
arylnitrenium ions are not involved.3
7-39
While the cur-
(
(
(
41) Becke, A. D. Phys. Rev. A 1988, 38, 3098-3100.
rent work does not directly address this issue, it is
42) Becke, A. D. J. Chem. Phys. 1993, 98, 5648-5652.
43) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785-789.
(
36) Wei, Y.; Tang, X.; Sun, Y. J. Polym. Sci., Part A: Chem. 1989,
7, 2385-2396.
37) Ding, Y.; Padias, A. B.; H. K. Hall, J. J. Polym. Sci., Part A:
(44) Stevens, P. J.; Devlin, F. J.; Chabalowski, C. F.; Frisch, M. J.
J. Phys. Chem. 1994, 98, 11632-11627.
2
(
(45) Hehre, W. J.; Radom, L.; Schleyer, P. v. R.; Pople, J. A. Ab Initio
Molecular Orbital Theory; Wiley: New York, 1986.
(46) Dewar, M. J. S.; Zoebisch, E. G.; Healy, E. F.; Stewart, J. J. P.
J. Am. Chem. Soc. 1985, 107, 3902-3909.
Polym. Chem. 1999, 37, 2569-2579.
(
(
38) Lux, F. Polym. Rev. 1995, 35, 2915.
39) Holze, R. Collect. Czech. Chem. Commun. 2000, 65, 899-923.
J. Org. Chem, Vol. 70, No. 13, 2005 5289

Kung et al.
was prepared in an anhydrous N
2
purged flow cell equipped
CH
3
CN. Next, 0.5 mL of sample and 0.5 mL of matrix were
with a magnetic stir-bar. Once prepared the solution was
mixed in a vial, and then 1 µL of acetic acid was added. The
sample plate was spotted with 1 µL of matrix alone and
allowed to air-dry. Then 1 µL of sample and matrix solution
was co-spotted and allowed to air-dry. The samples were then
analyzed using a MALDI-TOF-MS instrument.
purged for 10 min with N
septum.
2
and then sealed with a rubber
Kinetic studies performed on diphenylnitrenium ion used
stock solutions of 20 mg of 1-(N,N-diphenylamino)-2,4,6-
trimethylpyridinium tetrafluoroborate in 60 mL of solvent, the
solvent being CH
the stock solution was taken via syringe and placed in a N
purged quartz cuvette. The cuvette remains under N pressure
2
Cl
2
, toluene, or CH
3
CN. A 3 mL portion of
Acknowledgment. This work was supported by the
Chemistry Division of the National Science Foundation.
We thank Mr. Art Winter for assistance with the
calculations.
2
2
during the LFP experiments. At least five different concentra-
tions, ranging from 0 to 2.2 mM, of each quencher were
obtained through the addition of pure quencher via syringe to
determine the rate constants shown in Table 5.
Supporting Information Available: 1H NMR and 13
C
NMR spectra for adducts 13-15; Cartesian coordinates from
DFT calculations on 1 and 10-12; and detailed descriptions
of preparative photolysis reactions generating compounds 13-
MALDI-TOF-MS. Compound 2 (0.0806 g) was dissolved in
1
0 mL of dry CH
3
CN (0.2149 mmol; 21.49 mM), and then the
solution was purged under N
2
for 15 min. A 1 mL portion of
1
5. This material is available free of charge via the Internet
the solution was removed as a dark standard. The remaining
solution was photolyzed with a Xe lamp for 3 h. For the matrix,
at http://pubs.acs.org.
0.2013 g of 2,5-dihydroxybenzoic acid was dissolved in 10 mL
JO050598Z
5290 J. Org. Chem., Vol. 70, No. 13, 2005