2,3-Dimethyl-2,3-diphenylcyclopropanone
present in higher concentration, while a lower concentration of
oxyallyl 5 should be seen. Indeed, we observed only cyclopro-
panone 2 in the TRIR spectra (Figure 6) and no signal
corresponding to oxyallyl 5 was found. [The absence of the
carbonyl band of oxyallyl 5 in the TRIR spetra can also be
explained by two additional reasons: (a) The UDFT calculations
predict a low IR intensity (ca. 150) for the 1536 cm-1 band. As
a result, the corresponding transient signal will be weak and
masked by the baseline noise in the TRIR spectra (Figure 6).
(b) The formation time of cyclopropanone 2 within the time
resolution of our TRIR setup (30 ns) suggests that the time scale
of interconversion between 5 and 2 is less than 30 ns. Therefore,
oxyallyl 5 is too short-lived to be observed in the TRIR spectra.]
Experimental Section
trans-1 and trans-8 were synthesized according to literature
procedures.9a Phenylmethylketene (4) was synthesized according
to the method described by Dehmlow.19
The detailed description of our time-resolved FTIR (TRIR) setup
is available elsewhere.20 Briefly, the third harmonic of a YAG:
Nd3+ laser (354.7 nm, 10 ns, 1.5-5 mJ per pulse, 10 Hz repetition
rate) was used as an excitation source in all experiments. The sample
solutions were pumped through a 1 mm thick CaF2 flow cell with
a flow rate up to 150 mL per minute. All spectra were recorded in
the 1500-2800 cm-1 spectral window with 8 cm-1 resolution every
20 ns. The raw data were processed and visualized with custom
written LabView based software.
Isolation of cis- and trans-7. Starting trans-1 (0.128 g,
0.76 mmol) was dissolved in acetonitrile (50 mL) in a 125 mL
Erlenmeyer flask. The flask was fitted with a rubber septum, argon
inlet, and magnetic stir bar. The solution of trans-1 was purged
with argon for 15 min and irradiated in a photochemical reactor at
350 nm for 12 h. After irradiation was complete, the solution
became bright yellow. Solvent was removed under reduced pressure
to afford crude product (0.100 g) as light-yellow oil. Column
chromatography on silica gel with hexanes-ethyl acetate (98:2)
as eluant afforded cis- and trans-7 (0.061 g, 57%) as a light-yellow
oil.
Because it is in dynamic equilibrium with cyclopropanone 2
oxyallyl 5 also forms cyclohexadiene 6. The comparison of
calculated free energy difference and equilibrium constant for
oxyallyl 5-cyclohexadiene 6 equilibrium (Table 2, entries 3 and
4) with free energy difference and equilibrium constant for
oxyallyl 5-cyclopropanone 2 equilibrium (Table 2, entries 1 and
2) indicates that cyclohexadiene 6 should be much more stable
than cyclopropanone 2 and the overall equilibrium should be
shifted toward cyclohexadiene 6. Another driving force toward
the formation of cyclohexadiene 6 is aromatization to form
indanone 7, which is an irreversible process. The calculations
agree with our experimental observations. Fast ring closure of
oxyallyl 5 leads to cyclopropanone 2 observed in TRIR spectra
on a shorter time scale, while on a longer time scale oxyallyl 5
forms cyclohexadiene 6.
1
trans-7: H NMR (300 MHz, CDCl3) δ 7.34-7.41 (m, 3H),
7.28-7.34 (m, 2H), 7.20-7.28 (m, 1H), 7.13-7.20 (m, 3H), 3.57
(q, J ) 7.2 Hz, 1H), 1.73 (s, 3H), 1.43 (d, J ) 7.2 Hz, 3H); 13C
NMR (75.5 MHz, CDCl3) δ 14.2, 24.2, 45.1, 57.8, 123.9, 125.0,
127.0, 127.3, 128.0, 128.1, 128.6, 141.6, 142.3, 144.9, 220.0; MS,
m/z (I) 236 (44) (M+), 221 (68), 208 (7), 193 (77), 178 (100), 165
(33), 152 (13), 128 (14), 103 (41), 89 (23), 77 (32), 63 (14), 51
(19); IR (neat) 3063, 2968, 2926, 2869, 1749, 1597, 1447, 1373,
1290, 1151, 1075, 1026, 964, 909, 731, 697 cm-1; HRMS (EI)
m/z calcd for C17H16O 236.1201, found 236.1204.
Conclusion
cis-7: 1H NMR (300 MHz, CDCl3) δ 7.34-7.41 (m, 3H), 7.28-
7.34 (m, 2H), 7.20-7.28 (m, 1H), 7.13-7.20 (m, 3H), 3.59 (q, J
) 7.8 Hz, 1H), 1.74 (s, 3H), 1.36 (d, J ) 7.8 Hz, 3H); 13C NMR
(75.5 MHz, CDCl3) δ 17.6, 24.9, 46.4, 57.9, 124.2, 125.2, 126.6,
126.7, 126.9, 128.0, 128.5, 141.8, 143.0, 145.8, 220.0.
We have found distinct differences in the products of the UV
photolysis of trans-1 depending on the regime of irradiation
(steady state vs pulsed). From the analysis of 1H NMR, IR, and
UV/vis spectra, the steady-state photolysis of trans-1 at 254
A total of 15 rather than the expected 17 signals in the 13C NMR
spectrum of indanone trans-7 were observed. The reason for that
is an overlap of two signals of ortho-carbons of the aromatic ring
(at the C1 carbon of the indanone ring) at 126.7 ppm as well as
two signals of meta-carbons at 128.7 ppm (see the Supporting
Information).
We believe that this apparent mismatch is a common phenom-
enon for 1-phenyl-2-indanone aromatic systems (see, for example,
the previously reported 13C NMR spectrum of 1-phenyl-1,3,3-
trimethyl-2-indanone21).
and 350 nm produces intermediate cyclohexadiene
6
(Scheme 2). The cyclohexadiene undergoes rearrangement to
form cis- and trans-1,3-dimethyl-1-phenyl-2-indanones 7 with
a yield of ca. 60%. On the other hand, the pulsed laser UV
photolysis of trans-1 at 355 nm results in a characteristic
carbonyl absorption peak attributed to 2,3-dimethyl-2,3-diphe-
nylcyclopropanone 2 at 1814 cm-1 in the TRIR spectra.
Cyclopropanone 2 was observed for the first time in our time-
resolved experiments. The other two products, 1-phenyldiazo-
ethane 3 and phenylmethylketene 4, were detected by their
characteristic IR bands in the TRIR spectra as characteristic
absorption bands at 2101 and 2038 cm-1, respectively. Similarly,
the formation of 1-phenyldiazoethane was observed after the
pulsed laser photolysis of another 1-pyrazoline (trans-4,5-
dihydro-3,5-dimethyl-3,5-diphenyl-3H-pyrazol-4-ol, trans-8).
Steady-State Photolysis of trans-8. A 0.05 M solution of trans-1
in deuterated solvent (C6D6, CD3CN, or CD3OD) was placed in a
NMR tube and the tube was capped with a rubber septum. The
solution was purged with argon for 15 min. Then the appropriate
volume of toluene (0.05 M) was added. The solution was photolyzed
for 30 min in a photochemical reactor at 350 nm monitoring by
1
300 MHz H NMR spectroscopy in 2 min intervals (conversion
1
40%). According to H NMR spectra trans-9 was formed with
∼77% yield. The solvent was evaporated under reduced pressure
trans-9: 1H NMR (300 MHz, C6D6) δ 7.43-7.49 (m, 2H), 7.38-
7.42 (m, 2H), 7.32-7.37 (m, 4H), 7.23-7.30 (m, 2H), 3.74 (d, J
We propose that the irradiation of trans-1 leads to zwitterionic
oxyallyl 5 by nitrogen expulsion. The intermediate cyclopro-
panone 2 exists in a dynamic equilibrium with 5, and the latter
closes to cyclohexadiene 6 on a longer time scale. Aromatization
of 6 produces indanones 7. Another decomposition pathway of
trans-1 is retro-1,3-dipolar addition that yields the diazo
compound 3 and ketene 4. Accordingly, our TRIR experiments
conclusively prove, for the first time, that retro-1,3-dipolar
addition is indeed an actual fragmentation pathway of 1-pyra-
zolines (trans-1 and trans-8).
to afford trans-9 as a light-yellow oil (∼0.010 g).
(19) (a) Dehmlow, E. V.; Slopianka, M. Liebigs. Ann. Chem. 1979, 572-
593. (b) Baigrie, L. M.; Seiklay, H. R.; Tidwell, T. T. J. Am. Chem. Soc.
1985, 107, 5391.
(20) Fedorov, A. V.; Danilov, E. O.; Merzlikine, A. G.; Rodgers, M. A.
J.; Neckers, D. C. J. Phys. Chem. A 2003, 107, 3208.
(21) Korth, H. G.; Sustmann, R.; Lommes, P.; Paul, T.; Ernst, A.; de
Groot, H.; Hughes, L.; Ingold, K. U. J. Am. Chem. Soc. 1994, 116, 2767.
J. Org. Chem, Vol. 72, No. 8, 2007 2783