Rearrangement of 1-noradamantyl and 1-adamantylcarbene to bridgehead alkenes: Lifetimes of two bridgehead carbenes in solution

Article abstract of DOI:10.1021/jp010907z

B3LYP density functional calculations with the 6-31G* basis set predict that 3-noradamantylcarbene 7 rearranges to adamantene 9 and protoadmant-3-ene 10 over barriers of 0.35 and 7.84 kcal/mol, respectively, after zero-point energy correction. The same level of theory predicts that 1-adamantylcarbene 8 rearranges to homoadamantene 11 over a barrier of 6.06 kcal/mol after zero-point energy correction. To test these predictions, the photochemistry of 3-noradamantyldiazirine 5 and 1-adamantyldiazirine 6 was studied. Photolysis of 5 and 6 in cyclohexane produces the known dimers of the strained alkenes. 1-Adamantylcarbene 8 can be trapped in solution with cyclohexane and piperidine. 1-Adamantylcarbene 8 can be trapped with pyridine to form an ylide upon laser flash photolysis (LFP) of 6. 3-Noradamantylcarbene 7 cannot be intercepted with pyridine in LFP experiments. 3-Noradamantylcarbene 7 is not trapped with cyclohexane and is trapped with piperidine in only miniscule yield. The data indicates that 1-adamantylcarbene 8 is formed more efficiently from its diazirine precursor than is 3-noradamantylcarbene 7 and that the solution-phase lifetime of 1-adamantylcarbene 8 is at least 10 times longer than that of 3-noradamantylcarbene 7.

Full text of DOI:10.1021/jp010907z

10146
J. Phys. Chem. A 2001, 105, 10146-10154
Rearrangement of 1-Noradamantyl and 1-Adamantylcarbene to Bridgehead Alkenes:
Lifetimes of Two Bridgehead Carbenes in Solution
Eunju Lee Tae, Celine Ventre, Zhendong Zhu, Igor Likhotvorik, Francis Ford,
Eric Tippmann, and Matthew S. Platz*
Department of Chemistry, The Ohio State UniVersity, Columbus, Ohio 43210
ReceiVed: March 8, 2001; In Final Form: July 18, 2001
B3LYP density functional calculations with the 6-31G* basis set predict that 3-noradamantylcarbene 7
rearranges to adamantene 9 and protoadmant-3-ene 10 over barriers of 0.35 and 7.84 kcal/mol, respectively,
after zero-point energy correction. The same level of theory predicts that 1-adamantylcarbene 8 rearranges to
homoadamantene 11 over a barrier of 6.06 kcal/mol after zero-point energy correction. To test these predictions,
the photochemistry of 3-noradamantyldiazirine 5 and 1-adamantyldiazirine 6 was studied. Photolysis of 5
and 6 in cyclohexane produces the known dimers of the strained alkenes. 1-Adamantylcarbene 8 can be
trapped in solution with cyclohexane and piperidine. 1-Adamantylcarbene 8 can be trapped with pyridine to
form an ylide upon laser flash photolysis (LFP) of 6. 3-Noradamantylcarbene 7 cannot be intercepted with
pyridine in LFP experiments. 3-Noradamantylcarbene 7 is not trapped with cyclohexane and is trapped with
piperidine in only miniscule yield. The data indicates that 1-adamantylcarbene 8 is formed more efficiently
from its diazirine precursor than is 3-noradamantylcarbene 7 and that the solution-phase lifetime of
1-adamantylcarbene 8 is at least 10 times longer than that of 3-noradamantylcarbene 7.
I. Introduction
successfully modeled by the calculations of Hadad and Sziemies7
and their co-workers. Cubylphenylcarbene cannot be chemically
trapped. Either it rearranges to homocubene too rapidly to permit
efficient bimolecular chemistry or it is not formed efficiently
from its diazo precursor.8,9 Related rearrangements are known
in the parent system and with homocubylcarbene.9
Chemists have been fascinated with bridgehead alkenes ever
since Bredt promulgated his famous rule in 1924.1 Carbene
chemists have attempted to form bridgehead alkenes from
bridgehead carbenes with considerable success, as illustrated
below with 1-norbornylcarbene 1.2 In this case, the unstable
bridgehead alkene rearranges to form a mixture of dideutero
compounds where the labels ultimately reside in two possible
positions. The analogous bridgehead nitrene-to-imine rearrange-
ment has also been studied.3
tert-Butylcarbene (TBC) has been studied in great detail. It
isomerizes to trimethylethylene and 1,1-dimethylcyclopropane
in the gas phase and in solution.10 Ruck and Jones have shown
that solvent modulates the ratio of products formed.11
TBC has no known bimolecular chemistry because the
migration reactions traverse very small barriers and are ex-
tremely rapid. Armstrong et al. calculated that the reactions of
TBC surmount barriers of 3.7 and 0.1 kcal/mol to form
trimethylethylene and 1,1-dimethylcyclopropane, respectively.12
It can be posited that the strain developing in the bridgehead
alkene products might raise the barrier to bridgehead carbene
rearrangement relative to TBC, reduce the speed of the
rearrangement, and increase the chances of detecting or trapping
the bridgehead carbenes 7 and 8. This led us to study the
photochemistry of diazirines 5 and 6. These diazirines were
expected to provide entry to carbenes 7 and 8 which are known
Wilt attempted to generate the hydrocarbon variant of 1 from
the tosylhydrazone salt precursor but was unable to identify
carbene- or alkene-related products.4 Kirmse et al. prepared the
7,7-dimethyl derivative of 1 in solution and isolated the dimers
of the strained alkenes.5 The Jones group prepared 1-d2 in the
gas phase and isolated 4-d2, which arose by retro-Diels-Alder
fragmentation of alkene 2.6 This result implies that the short
bond of 1 migrates selectively, a result which has been
10.1021/jp010907z CCC: $20.00 © 2001 American Chemical Society
Published on Web 10/11/2001

Rearrangement of Carbenes to Bridgehead Alkenes
J. Phys. Chem. A, Vol. 105, No. 44, 2001 10147
The mixture was allowed to warm to 0 °C and was then stirred
for an additional 30 min. A solution of freshly prepared tert-
butylhypochlorite (0.45 mL, 4 mmol) in tert-butanol (0.5 mL)
was then added via syringe. For small-scale reactions, tert-
butylhypochlorite was added in one portion, but for larger-scale
(10 mmol or more) reactions, the addition was performed
dropwise to prevent explosive decomposition of the diazirine.
The solution was stirred at 0 °C for 1 h more, poured over
crushed ice, and then extracted with pentane. The combined
organic solutions were washed several times with water and
brine and then dried over sodium sulfate. After removal of the
solvent in vacuo, the crude diazirine (yellow liquid) was purified
by column chromatography on silica gel with pentane as the
eluant. The purification process was monitored by the UV
absorption at 350 nm of the fractions collected. In this manner,
0.29 g (42%) of a colorless liquid, which was pure by GC-
MS, was collected. The diazirine was stored in the freezer, in
to rearrange to strained alkenes 9-11.2 Furthermore, diazirine
precursors are convenient for laser flash photolysis (LFP), matrix
isolation, and chemical trapping studies. Herein, we are pleased
to report our results.
II. Experimental Section
1
General Methods. H NMR and 13C NMR spectra were
1
recorded on a Bruker AC-200 MHz or on a Bruker DRX-500
MHz spectrometer. Chemical shifts are reported in ppm with
tetramethylsilane as an internal standard. Infrared spectra were
recorded using a Perkin-Elmer 1710 Fourier transform spec-
trometer interfaced with a Perkin-Elmer 3700 data station. UV-
vis spectra were recorded on a Milton-Roy Spectronic 3000
diode array spectrophotometer. GC-MS mass spectral analyses
were performed on a Hewlett-Packard 6890 GC spectrometer
with an HP-1 methylsiloxane capillary column (i.d. 100 µm,
length 40 m, phase film thickness 0.25 µm) and a HP 5973
mass selective detector. Three kinds of lasers were used to
perform LFP experiments: a Lambda Physik LPX-100 excimer
laser (308 nm, 120 mJ, 17 ns), a Lumonics TE-861 excimer
laser (351 nm, 60 mJ, 17 ns), and a Continuum PY62C-10 Nd:
YAG laser (355 nm, 30 mJ, 2 ns).
All reagents were purchased from commercial sources and
used without purification unless noted otherwise. All reactions
were carried out in flame-dried flasks under an argon atmo-
sphere. 1-Adamantanecarboxylic acid was purchased from
Aldrich. Freon-113 was purified by passage through a neutral
alumina column just prior to use. Diethyl ether, tetrahydrofuran,
and cyclohexane were distilled over sodium and benzophenone
and stored under argon. Methanol was distilled from sodium
methoxide. Pyridine was refluxed over potassium hydroxide and
then distilled. It was stored protected from light in a brown bottle
with potassium hydroxide under an atmosphere of dry argon.
Pentane was distilled over P2O5. Dichloromethane was distilled
from CaH2 and stored protected from bright light in a brown
bottle with Linde type 4 Å molecular sieves and under an argon
atmosphere. Piperidine was dried by distillation prior to use.
All distillations were performed under an inert argon atmo-
sphere.
the dark, and under an argon atmosphere. H NMR (CDCl3):
δ 0.41 (s, 1H), 1.29 (m, 6H), 1.51 (m, 3H), 1.60 (m, 3H), 1.86
(m, 3H). 13C NMR (CDCl3): δ 28.32, 29.79, 31.80, 36.94,
39.64. IR: 1588 cm-1. UV (pentane): max 340 and 352 nm.
Anal. GC-MS: [M+] calcd for C11H16N2, 176.1317; found, 1
peak, 176.13040.
Isolation of the Dimers of Homoadamantene (11).16
1-Adamantyldiazirine 6 (250 mg) was dissolved in 15 mL of
cyclohexane. The mixture was photolyzed overnight at 350 nm
in a cold room at 4 °C. A GC-MS study revealed one major
peak for the cyclohexane insertion adduct (M+ ) 232), four
peaks (two intense and two smaller ones) for the dimers of
homoadamantene (M+ ) 296), and other unidentified impurities.
The solution was concentrated in vacuo. The crude mixture was
purified by preparative TLC using UV-vis active silica plates
and hexane as the eluant. The top layer of silica was extracted
with hexane. GC-MS analysis showed the presence of only
the cyclohexane adduct and the major dimers of homoadaman-
tene. It was possible to obtain a mixture of the two major dimers
by successive recrystallization from 2-propanol, but this mixture
could not be separated further. The dimers did not contain
resonances due to alkenyl carbons or protons.
N-(3-Noradamantyl)piperidine (14). 3-Noradamantyl car-
boxylic acid (104 mg, 0.626 mmol) in 5 mL of dicloromethane
was stirred with 258 mg (2.0 equiv) of dicyclohexylcarbodiimide
(DCC), 10 mg of p-(dimethylamino)pyridine (DMAP), and 213
mg (4 equiv) of piperidine. The mixture was stirred for 2 h,
and the amide product was purified by column chromatography
(elution with 30% ethyl acetate in hexanes). The amide was
reduced by 24 mg (0.626 mmol) of lithium aluminum hydride
(LAH) in refluxing ether for 2 h, quenched with a 5% aqueous
sodium hydroxide solution, filtered, and concentrated. The
mixture was column chromatographed (elution with 50% EtOAc
Synthesis. 3-Noradamantyldiazirine (5)13a-c was synthesized
by the method of Likhotvorik et al.13d and Tae et al.14 Methyl
ethers 15 and 1715 were synthesized and had spectral properties
consistent with those described in the literature. An authentic
sample of Diels-Alder adduct 23 was generously provided by
Professor Jones of Princeton University.
1
in hexanes).The yield of the amine was 91%. H NMR (500
MHz, CDCl3): δ 2.42 (br s, 6H), 2.16 (s, 2H), 1.97 (m, 1H),
1.71-1.10 (m, 16H). 13C NMR (126 MHz, CDCl3): δ 67.2,
55.9, 49.1, 48.5, 43.83, 43.80, 37.4, 35.5, 26.0, 24.2. MS (EI)
m/z (rel intensity): 219 (M+, 18), 135 (5), 98 (100), 84 (20).
HRMS (EI): [M+] calcd for C15H25N, 219.1987; found,
219.1981.
1-Adamantyldiazirine (6).13 The diazirine was synthesized by
the procedure of Likhotvorik et al.13d 1-Adamantane carboxal-
dehyde (0.65 g, 4 mmol) was dissolved in 12 mL of anhydrous
THF. The solution was cooled to 0 °C with an ice bath, placed
under an argon atmosphere, and protected from direct light with
foil paper. A lithium bis(trimethylsilyl)amide (1 M) solution in
THF (8 mL, 8 mmol) was added dropwise via syringe. The
reaction mixture was stirred for 30 min at 0 °C and then cooled
to -30 °C. A solution of hydroxylamine-o-sulfonic acid (0.48
g, 4 mmol) in dry diglyme (5 mL) was then added dropwise.
1-Methoxyadamantane (15).15 1-Adamantanol (0.15 g, 1
mmol) was mixed with 60% of NaH (0.08 g, 2 mmol) and CH3I
(2 mL) in tetrahydrofuran (THF). It was stirred overnight. The
mixture was chromatographed (10% ethyl acetate in hexane).
1
The yield was 99%. H NMR (500 MHz, CDCl3): δ 3.20 (s,
3H), 2.12 (s, 3H), 1.71 (s, 6H), 1.62 (d, 3H, J ) 12.2 Hz), 1.57
(d, 3H, J ) 12.2 Hz). 13C NMR (126 MHz, CDCl3): δ 71.8,

10148 J. Phys. Chem. A, Vol. 105, No. 44, 2001
Tae et al.
47.7, 40.9, 36.4, 30.4. MS (EI) m/z (rel intensity): 166 (M+,
42), 135 (9), 109 (100), 94 (9), 79 (10).
was not degassed. The vial was exposed to 350 nm radiation
for 26 h. The progress of the reaction was monitored by
observing the disappearance of the diazirine absorption at 340
nm. The cyclohexane was evaporated to give a crude orange
mixture. The crude mixture was chromatographed on silica gel
with pentane elution. The solvent was evaporated to yield 47-
61 mg of white crystals. This mixture was sublimed for 1.5 h
at 55 °C and 2.3 mmHg. This afforded 9 mg of pure cyclohexane
insertion product 24. GC-MS traces were used to estimate that
an additional 5 mg of 24 did not sublime for a total of 14 mg
(11% yield). 13C NMR (CDCl3): δ 2608, 27.1, 2902, 32.6, 33.5,
3-Noradamantyl Methyl Ether (16). 3-Noradamantyl car-
boxylic acid (0.166 g, 1 mmol) was refluxed with LAH (0.15
g, 4 mmol) in THF overnight. It was worked up using a 5%
aqueous sodium hydroxide solution. 3-Noradamantyl alcohol
was isolated by column chromatography (33% ethyl acetate in
hexane). The yield was 99%. The alcohol (0.15 g, 1 mmol)
was mixed with 60% of NaH (0.08 g, 2 mmol) and CH3I (2
mL) in THF. It was stirred overnight. The mixture was
chromatographed (10% ethyl acetate in hexane). 1H NMR (300
MHz, CDCl3): δ 3.37 (s, 2H), 3.36 (s, 3H), 2.20 (s, 2H), 2.12
(t, 1H, J ) 6.7 Hz), 1.72 (m, 2H), 1.60 (m, 8H). 13C NMR (75
MHz, CDCl3): δ 79.9, 59.5, 49.8, 46.8, 44.1, 41.4, 37.7, 35.7.
MS (EI) m/z (rel intensity): 166 (M+, 17), 134 (52), 123 (37),
93 (100), 79 (74). HRMS (EI): calcd for C11H18O, 166.1358;
found, 166.1356.
1
36.7, 43.4, 53.1. H NMR (CDCl3): δ 0.86-0.961 (m, 5H),
1.1 (t, 1H, J ) 12 Hz), 1.25 (q, 2H, J ) 12 Hz), 1.46 (td, 6H,
J ) 2.2, 19.9 Hz), 1.56-1.68 (m, 11H), 0.90 (t, 3H, J ) 20
Hz). MS (EI) m/z (rel intensity): 232(5), 135(100), 107(8),
93(14), 79(14). The material which did not sublime also
contained 43 mg of homoadamantene dimers (28%).
2-Methoxyadamantane (17).15 2-Adamantanone (0.75 g, 5
mmol) was stirred with LAH (0.19 g, 1 mmol) in ether overnight
and quenched with a 5% aqueous sodium hydroxide solution.
2-Adamantyl alcohol was mixed with 60% of NaH (0.3 g, 1.2
equiv) and CH3I (2 mL) in THF and stirred overnight. The
mixture was chromatographed (10% EtOAc in hexane). The
1-Adamantylpiperidine (25). 1-Adamantyldiazirine 6 (100 mg,
0.57 mmol) was dissolved in 2.1 mL of piperidine, freshly
distilled over sodium. It was exposed (without degassing) to
350 nm radiation for 26 h to provide a clear solution. The solvent
was evaporated to provide 130 mg of a yellowish oil. Preparative
TLC was performed with MeOH/CH2Cl2 in a ratio of 1/18. The
compounds were extracted from the silica gel and chromato-
graphed again with CH2Cl2, and the portion of the plate not
absorbing UV-vis light provided 25 mg of the piperidine
insertion product. There was an additional 2 mg present in other
fractions estimated from GC-MS for a 21% overall yield. The
major byproducts were 11 mg of homoadamantene 11 dimers
1
yield was 99%. H NMR (500 MHz, CDCl3): δ 3.33 (s, 3H),
3.32 (s, 1H) 2.01 (m, 4H), 1.9-1.6 (m, 8H), 1.45 (d, 2H, J )
11.1 Hz). 13C NMR (126 MHz, CDCl3): δ 83.2, 55.2, 37.6,
36.5, 31.4, 31.3, 27.5, 27.4. MS (EI) m/z (rel intensity): 166
(M+, 18), 134 (100), 119 (23), 92 (50), 79 (36).
1-Adamantyl Acetate (18). 1-Adamantanol (0.154 g, 1.01
mmol) was mixed with acetic anhydride (0.144 mL, 1.53 mmol),
pyridine (0.165 mL, 2.04 mmol), and DMAP (20 mg) in 5 mL
of dichloromethane. The mixture was stirred for 2 h and then
was chromatographed (20% ethyl acetate in hexanes). 1H NMR
(200 MHz, CDCl3): δ 2.07 (m, 9H), 1.96 (s, 3H), 1.59 (m,
6H). 13C NMR (50 MHz, CDCl3): δ 169.5, 79.9, 41.1, 36.0,
30.6, 22.4. MS (EI) m/z (rel intensity): 194 (M+, 1), 134 (100),
119 (17), 105 (15), 92 (90), 79 (35).
(7%) and a 20-30 mg fraction containing azine (11-16%). 13
C
NMR of 25 (CDCl3): δ 8.1, 20.9, 23.7, 31.7, 34.7, 41.8, 60.3,
78.2. 1H NMR: δ 1.24 (m, 2H), 1.36 (d, 6H, J ) 2.5 Hz), 1.49
(q, 4H, J ) 5.7 Hz), 1.62 (m, 10H), 1.86 (s, 2H), 2.38 (t, 3H,
J ) 5.0 Hz). MS: [M + H] calcd, 234.2144; found, 234.2149.
Chemical Trapping Studies. Constant volumes of a stock
solution of diazirine (300 µL) were added to each cuvette.
Varying amounts (0-100 µL) of a trapping agent were added,
and the total volume of the solution was brought to 500 µL
with the addition of solvent. All of the samples were placed in
Pyrex cuvettes (0.5 mL thin glass cuvette), and the solutions
were degassed for 2 min with a strong argon flow. The solutions
were photolyzed at 350 nm using a Ray-o-net reactor equipped
with 16 RPR-3500 bulbs. The experiment was performed
overnight in a cold chamber at 4 °C. The temperature inside
the reactor was approximately 15 °C. The yields of the products
obtained upon decomposition of the precursors were measured
by integration of the peaks obtained in the GC spectra. The
compounds were identified by comparing their mass spectral
fragmentation patterns and NMR spectra with data from the
literature or from authentic samples. In a typical experiment, a
stock solution of diazirine in cyclohexane was prepared with
an optical density of 1.8 at 355 nm. The stock solution of
diazirine (100 µL) was placed into several cuvettes. A different
volume of the same trapping agent was added to each cuvette
and the total volume adjusted to 200 µL with solvent (cyclo-
hexane).
3-Protoadamantyl Alcohol (19).17 Photolysis of 3-norada-
mantyldiazirine 5 in neat acetic acid produced almost equal
amounts of adamantyl acetate 18 and protoadamantyl acetate
21. The mixture was reduced by 2 equiv of LAH and purified
by column chromatography (30% ethyl acetate in hexane). The
mixture of adamantanol and protoadamantanol was separated
1
using MPLC. H NMR (300 MHz, CDCl3): δ 2.50-2.30 (m,
2H), 2.10-1.5 (m, 14H). 13C NMR (75 MHz, CDCl3): δ 79.9,
48.2, 44.6, 40.1, 39.4, 36.4, 33.4, 32.4, 27.8, 27.7. MS (EI) m/z
(rel intensity): 191 (M+, 1), 155 (2), 131 (3), 108 (100), 55
(29). 1H NMR (300 MHz, CDCl3): δ 2.24-2.15 (m, 2H), 2.07-
1.27 (m, 14H). 13C NMR (75 MHz, CDCl3): δ 79.6, 48.0, 44.4,
39.8, 39.2, 36.2, 33.2, 32.1, 27.6, 27.5.
3-Protoadamantyl-p-nitrobenzoate (20). 3-Protoadamantyl
alcohol (30 mg) was mixed with 0.2 mL of pyridine, 50 mg of
DMAP, and 100 mg of p-nitrobenzoyl chloride in 2 mL of
dichloromethane. The mixture was refluxed for 7 h and then
chromatographed (10% ethyl acetate in hexane). The yield was
72%. Mp: 182-184 °C. 1H NMR (300 MHz, CDCl3): δ 8.25
(d, 2H, JAB ) 8.7 Hz), 8.16 (d, 2H, JAB ) 8.7 Hz), 2.55-1.35
(series of m, 15Η). 13C NMR (75 MHz, CDCl3): δ 164.2, 150.5,
137.5, 130.7, 123.6, 91.5, 46.2, 41.7, 39.8, 39.1, 35.8, 33.1, 27.6,
27.4, 26.7. HRMS (EI): calcd for C17H19NO4, 301.1314; found,
301.1315.
LFP Experiments. The LFP spectrometer has been de-
scribed.18 In all of the LFP experiments, the sample cells were
fitted with a rubber septum, and the solutions were deoxygenated
by passing a stream of argon through the sample for 2 min.
The diazirine solutions utilized in LFP studies were contained
in quartz cells for excitation at 308 or 351 nm. The cells were
fabricated from square tubing purchased from VitroDynamics.
Excitation with the Nd:YAG laser required Suprasil quartz
1-Adamantylcyclohexane (24). 1-Adamantyldiazirine 6 (100
mg, 0.57 mmol) was dissolved in 2.1 mL of cyclohexane
(freshly distilled over sodium) in a 3 mL glass vial. The solvent

Rearrangement of Carbenes to Bridgehead Alkenes
J. Phys. Chem. A, Vol. 105, No. 44, 2001 10149
fluorescence-free static cells purchased from Scientific products.
An almost perfectly flat surface as in this type of quartz cell
results in the minimum scattering of laser light. Low-temperature
LFP experiments were performed using a fluorescence-free cell
in a variable-temperature sample holder and a NESLAB RTE-
110 proportional temperature controller to regulate the temper-
ature.
Stock solutions of diazirines were prepared just prior to the
LFP experiments. For experiments with the Nd:YAG laser, the
different samples prepared had an absorbance close to 1.5 at
355 nm. For those experiments performed with the excimer
lasers, the absorbance of the samples was close to 0.5-0.7 at
308 or 351 nm. The pyridine trapping agent was purified by
distillation and stored over KOH pellets in a dark-brown bottle.
Constant volumes (350 µL) of stock solutions of diazirine were
placed into each cuvette. Varying amounts (0-150 µL) of
pyridine were added. The total volume of the solution was
maintained constant at 500 µL with the addition of solvent.
Matrix Isolation Spectroscopy. The colorless adamantyl-
diazirine 6 was placed in a U-shaped tube submerged in an
acetone bath kept at -25 °C. The tube was connected to a closed
cryogenic system cooled by helium (purchased from Air
Products). Argon gas was flowed over the sample, and the
resulting gaseous mixture of diazirine and argon was deposited
onto the surface of a CsI window. The argon matrix was
maintained at 14 K during the entire experiment. Ray-o-net
lamps (254 and 350 nm) were used to photolyze the sample in
the argon matrix. After exposure to light for a specified period
of time, both the IR and UV-vis spectra were recorded. The
UV-vis spectrum was measured with a Lambda 6 UV-vis
spectrophotometer and the IR spectrum with an FTIR 2000
Perkin-Elmer spectrometer with 2 cm-1 resolution.
Figure 1. Relative energies of singlet 3-noradamantylcarbene 7,
adamantene 9, and protoadamant-3-ene 10 and the barriers to carbene
rearrangement calculated at the B3LYP/6-31G* level.
Calculations. All calculations were performed using the
Gaussian94 suite of programs19 available on four workstations
in the Chemistry Department computer room. Other software,
such as Spartan and Molecule, were also used during this study.
The geometries of the carbenes and olefinic products were
fully optimized at the density functional level of theory B3LYP20
using the 6-31G* basis set. C1 symmetry was maintained during
the optimizations. All of the stationary points were confirmed
by harmonic-frequency calculations. The transition state in-
volved in the rearrangement of the bridgehead carbene to the
alkene product was found to have one imaginary frequency.
For the modeling of the electronic spectra obtained during matrix
experiments, the geometries optimized with the B3LYP/6-31G*
level were employed. The IR analysis was based on the B3LYP/
6-31G* level of theory. Modeling of the electronic spectra
obtained in matrix isolation experiments was based on the
INDO/S method of the ZINDO program.21
Figure 2. Relative energies of singlet 1-adamantylcarbene 8 and
homoadamantene 11 and the barrier to their interconversion calculated
at the B3LYP/6-31G* level.
level calculations of Armstrong et al. [QCISD(T)/6-31+G-
(2d,p))//MP2/6-31G(d)+ZPC calculations]12 indicate that the
singlet and triplet states of TBC are very close in energy, with
the triplet state of TBC believed to be favored by 1-2 kcal/
mol. We make no prediction of the ground-state multiplicities
of 7 and 8 but expect that the singlet-triplet gap will be small
in these carbenes and that rearrangements will proceed via the
singlet states of carbenes 7 and 8.
DFT calculations indicate that the energy barrier to rear-
rangement of singlet 1-adamantylcarbene 8 is 6.06 kcal/mol.
Thus, theory predicts that the singlet state of this carbene should
be sufficently long-lived to be detectable by LFP experiments.
The only bridgehead alkene formed upon ring expansion of
1-adamantylcarbene 8 (homoadamant-3-ene 11) contains a trans-
cycloheptene ring structure. 3-Noradamantylcarbene 7 contains
two carbon-carbon bonds which can migrate to the carbenic
center. Calculations predict that the kinetic product is adaman-
tene 9 formed over a 0.35 kcal/mol barrier. Protoadamant-3-
ene 10 is 11 kcal/mol more stable than adamantene 9. The
III. Results
III.1. Computational Chemistry. Calculations were per-
formed at the density functional B3LYP/6-31G* levels of theory
(DFT). We optimized the geometries of carbenes 7 and 8 in
their closed-shell singlet states and the geometries of the
transition states that correspond to the rearrangements of these
carbenes to the requisite strained alkenes. The stationary points
were verified by harmonic-frequency calculations. This dem-
onstrated that all minima had positive frequencies and that the
transition states had only one imaginary frequency. The results
are summarized in Figures 1 and 2.
This level of theory (B3LYP/6-31G*) does a poor job of
computing the singlet-triplet separations of carbenes22 and thus
was not used to study the triplet states of 7 and 8. The higher-

10150 J. Phys. Chem. A, Vol. 105, No. 44, 2001
Tae et al.
SCHEME 1
SCHEME 2
thermodynamic product of rearrangement of 3-noradamantyl-
carbene 7 (protoadamant-3-ene 10) is formed over a barrier of
7.84 kcal/mol. Thus, theory predicts that the short bond will
migrate, in good agreement with previously reported experi-
mental and computational work on related bridgehead car-
benes.6,7 Calculations predict that 3-noradamantylcarbene 7, like
TBC,12 will be too short-lived to be detected by LFP experi-
ments. Carbocations behave in exactly the same way as the
carbenes of this work, because it is the short bridge that migrates
when there is a choice.23
The IR and UV-vis spectra of the strained alkenes were
predicted by theory as described in the next section. The
corresponding matrix spectroscopic data for adamantene 9 and
protoadament-3-ene 10 have been reported previously.14
III.2. Matrix Isolation Spectroscopy. 1-Adamantyldiazirine
6 was evaporated directly onto a cold CsI window with a large
excess of argon. The 14 K matrix spectra exhibited IR and UV-
vis bands characteristic of 1-adamantyldiazirine 6.
performed additional CIS and CISD calculations on the ground
state of homoadamantene 11 using SCF MOs by the INDO/S
method with the ZINDO program.21 The strong absorption band
observed at 277 nm is due to electron promotion from the π to
the π* orbitals of the CdC bond of homoadamant-3-ene with
an oscillator strength of 0.241. Thus, the changes in the UV
spectrum induced by photolysis of 1-adamantyldiazomethane
12 in argon are also consistent with the formation of homoada-
mantene 11 in the matrix.
The IR spectrum of 1-adamantyldiazirine 6 contained an
intense band at 1588 cm-1, assigned to the NdN bond.
1-Adamantyldiazirine 6 also had two maxima in the UV
spectrum at 342 and 350 nm. Photolysis of 1-adamantyldiazirine
6 at 350 nm led to a decrease in the intensity of these bands,
which were replaced by a set of new absorption bands. We
observed the appearance of a broad band at 250 nm in the UV
spectrum and a band at 2100 cm-1 in the IR spectrum. These
bands were assigned to diazo compound 12 (Scheme 1), formed
by isomerization of the cyclic diazirine to its linear isomer.
Subsequent photolysis of the diazirine at 350 nm led to an
increased yield of the diazo compound and the complete
disappearance of the bands characteristic of the diazirine moiety
in both the IR and the UV spectra. Continuous irradiation of
the diazo compound with 254 nm light generated a new band
at 1630 cm-1 in the IR spectrum and a band at 260 nm in the
UV spectrum. The spectra are given in the Supporting Informa-
tion. Martella et al.24 have reported that strained alkene 11 (also
formed by the rearrangement of 1-adamantylcarbene 8) has an
IR transition at 1610 cm-1. Density functional theory predicts
that homoadamantene 11 will have a carbon-carbon double
bond stretch at 1644 cm-1. Thus, IR spectroscopy supports the
formation of homoadamantene 11 upon photolysis of 1-ada-
mantyldiazomethane 12 in argon, in accord with our expecta-
tions based on literature reports.2 It was not possible to
confidently assign other vibrational bands of homoadamantene
11 in the matrix because of low signal to noise.
We have previously reported that the photolysis of 3-norada-
mantyldiazirine 5 in argon at 14 K produces adamantene 9, as
identified by IR and UV-vis spectroscopy and upon comparison
of our spectra with the literature.14 The strained alkenes
adamantene 9 and protoadamant-3-ene 10 can be photochemi-
cally interconverted in argon at 14 K.14 Much earlier studies
had previously demonstrated that adamantene 9 is formed by
rearrangement of 3-noradamantylcarbene 7.2,26
III.3. LFP Studies. LFP of 3-noradamantyldiazirine 5 in
pentane or benzene produces adamantene 9 (λmax ) 325 nm).14
LFP of 3-noradamantyldiazirine 5 in the presence of pyridine
fails to produce the transient spectrum of a carbene-pyridine
ylide. Either the carbene is not formed efficiently upon
photolysis of 3-noradamantyldiazirine 5 or it rearranges too
rapidly (τ < 10 ps) for capture by pyridine (Scheme 2).
Rearrangement of 5* (or a related diazirine ring-opened
biradical) to adamantene 9 in concert with nitrogen extrusion
would not be surprising. We have come to similar conclusions
with cyclopropyl27 and homocubyldiazirines.9
LFP of 1-adamantyldiazirine 6 in pentane fails to produce a
UV-vis active transient. In the presence of pyridine, however,
LFP of 1-adamantyldiazirine 6 produces the transient spectrum
shown in Figure 3.
There was no IR evidence for the presence of persistent
1-adamantylcarbene 8 in argon. Yao et al.25 were able to detect
adamantylchlorocarbene and its rearranged product chloro-
homoadamant-3-ene upon photolysis of the appropriate diazirine
precursor. The chloro substituent stabilizes the carbene which
allows its direct detection in an argon matrix.
The transient is formed faster than the resolution of the
spectrometer (Figure 4). The yield of the ylide is not influenced
by the presence of oxygen. The yield (Ay) of ylide 13 is
dependent on the concentration of pyridine (Figure 4). A plot
of 1/Ay versus 1/[pyr] is roughly linear (Supporting Information).
Division of the intercept by the slope of this plot gives kPYRτ,
Excited-state calculations were performed with the B3LYP/
6-31G* geometries in an effort to provide a picture of the
excited-state electronic structure of the strained alkene. We also

Rearrangement of Carbenes to Bridgehead Alkenes
J. Phys. Chem. A, Vol. 105, No. 44, 2001 10151
The other product was neither 16 nor 17, established by
comparison with authentic samples.
Carbenes react rapidly with alcohols to form ethers. Thus,
the absence of 16 in the product mixture is consistent with an
extremely rapid carbene rearrangement or direct formation of
adamantene 9 from the excited 3-noradamantyldiazirine precur-
sor (5*; Scheme 2).
We have not identified the second methyl ether but speculate
that it is the product of the trapping of protoadamant-3-ene 10
based on our work with acetic acid (vide infra).
Figure 3. Transient spectrum of the pyridine ylide produced by LFP
of 1-adamantyldiazirine 6 in pentane at ambient temperature.
Alkene dimers are observed upon photolysis of 3-norada-
mantyldiazirine 5 in methanol. Methanol is obviously only a
moderately effective trap of adamantene (kCH OH ) 8.4 × 104
3
M-1 s-1).14 Acetic acid traps adamantene 9 over 150-fold faster
than does methanol (kCH COOH ) 1.4 × 107 M-1 s-1).14
3
Photolysis of 3-noradamantyldiazirine 5 in neat acetic acid yields
two isomeric adducts. No alkene dimers were formed in the
more reactive solvent. One product was quickly identified as
acetate 18, the product expected from trapping adamantene 9.
Figure 4. Optical yield of the pyridine ylide formed upon LFP of
1-adamantyldiazirine 6 as a function of pyridine concentration.
The mixture of acetates formed on photolysis of 3-norada-
mantyldiazirine 5 was reduced with LAH, and the resulting
mixture of alcohols was separated by chromatography.
where τ is the lifetime of 1-adamantylcarbene 8 in the absence
of pyridine and kPYR is the absolute rate constant of the reaction
of 1-adamantylcarbene 8 with pyridine. If kPYR ) 1.0 × 109
M-1 s-1, as is commonly observed,28 one can deduce lifetimes
of 1-adamantylcarbene of 1.5 ns in cyclohexane and 2.5 ns in
cyclohexane-d12. These results demonstrate that the lifetime of
1-adamantylcarbene 8 in the alkane solvent is limited, at least
in part, by reaction with the solvent. This conclusion will be
confirmed by product analysis (vide infra).
The NMR spectrum of one of the alcohols was identical with
that reported for 3-protoadamantanol 19.16 A crystal structure
of the p-nitrobenzoyl derivative 20 confirmed the structure
(Supporting Information).
From these results, one can posit that protoadamant-3-ene
10 is formed along with adamantene 9 upon photolysis of
3-noradamantyldiazirine 5 in acetic acid and is trapped to give
21.
The ylide yield can be reduced in the presence of typical
carbene quenchers such as methanol. A Stern-Volmer analysis
of the quenching of the ylide can be used to obtain the rate
constant for the reaction of adamantylcarbene with methanol.
s
A plot of Ay /Ay versus [CH3OH] is predicted and found to be
linear with a slope of kQ/kPYR[pyr] (Supporting Information).
Assuming again28 that kPYR is 1 × 109 M-1 s-1 and with constant
[pyr] ) 2.47 M, 1-adamantylcarbene 8 was found to have a
rate constant of reaction with methanol kQ ) 6 × 109 M-1 s-1
.
III.4. Chemical Trapping Experiments. 1-Noradamantyl-
carbene. Photolysis of 3-noradamantyldiazirine 5 in either
cyclohexane or benzene leads to products which have properties
consistent with the previously described dimers of strained
alkenes 9 and 10 (Scheme 2).2,5,24 The dimers lack vinylic proton
resonances in their NMR spectra and have the expected m/z
ratios and fragmentation patterns of strained alkene dimers.
Carbene-benzene and carbene-cyclohexane adducts, if formed,
were produced in no more than trace quantities.
If the calculations are even approximately correct that relaxed
3-noradamantylcarbene 7 has a strong kinetic preference for the
formation of adamantene 9 (∆∆Ea ≈ 7.5 kcal/mol), then the
result in acetic acid indicates either that protoadamant-3-ene
10 is formed from a rearrangement in the excited state of
3-noradamantyldiazirine 5 or that adamantene 9 and protoada-
mant-3-ene 10 interconvert in the presence of an acid catalyst.
Acetate 22, which would arise from trapping 3-noradaman-
tylcarbene 7, was not observed. This is further evidence for
either a very short lifetime of 3-noradamantylcarbene 7 or that
it is simply not formed upon photolysis of 3-noradamantyldiaz-
irine 5. These conclusions are tempered by the possibility of
Photolysis of 3-noradamantyldiazirine 5 in neat methanol
produced two methyl ethers, one of which was positively
identified as 1-methoxyadamantane 15, a product expected from
trapping adamantene, by comparison with an authentic sample.

10152 J. Phys. Chem. A, Vol. 105, No. 44, 2001
Tae et al.
SCHEME 3
of piperidine reduces the yield of both homoadamantene 11
dimers and carbene-cyclohexane adduct 24 (Supporting Infor-
mation), as predicted by Scheme 3.
The absolute yield of the piperidine adduct derived from
trapping 1-adamantylcarbene 8 is at least 10-fold larger than
the yield of piperidine adduct derived from trapping 3-norada-
mantylcarbene 7.
IV. Discussion
On the basis of numerous previous reports, we were confident
that photolysis of diazirines 5 and 6 would produce strained
bridgehead alkenes.2-7,26 Again, given so much experimental
and theoretical precedent, we expected that the 3-noradaman-
tylcarbene 7 would prefer to migrate the short C-C bridge bond
to form adamantene rather than to migrate a long bridge to form
protoadamant-3-ene 10.2-7,26 B3LYP density functional theory
calculations indicated that the preference is substantial. The
calculated barriers to rearrangement of 1-noradamantylcarbene
7 to adamantene 9 and protoadamant-3-ene 10 are 0.35 and 7.84
kcal/mol, respectively. The low barrier of 3-noradamantylcar-
bene 7 rearrangement to form adamantene 9 is reminiscent of
the low barrier (0.1 kcal/mol) to rearrangement of TBC.12 The
thermodynamic product of carbene rearrangement (protoada-
mant-3-ene 10) is not the kinetic product of rearrangement
(adamantene 9). 1-Adamantylcarbene 8, on the other hand, is
predicted to surmount a barrier of 6.06 kcal/mol to form
homoadamantene 11. The key question of this work is whether
3-noradamantylcarbene 7 or 1-adamantylcarbene 8, unlike TBC,
is sufficiently long-lived to be detected by chemical trapping
or by LFP methods.
protonation of 3-noradamantylcarbene 7 (or of the diazo isomer
of 5) followed by fast cationic rearrangement.
To avoid cationic rearrangements, 3-noradamantyldiazirine
5 was photolyzed in the presence of butadiene. Photolysis of
3-noradamantyldiazirine 5 in benzene containing 1,3-butadiene
(273 K) produced five adducts. The major adduct 23, formed
by a Diels-Alder reaction of adamantene 9, was identified by
comparison with an authentic sample kindly provided by
Professor Jones of Princeton University.17
LFP of 1-adamantyldiazirine 6 in the presence of pyridine
produces a carbene-pyridine ylide; LFP of 3-noradamantyl-
diazirine 5 under the same conditions does not. LFP of
3-noradamantyldiazirine 5 produces adamantene as the only
species detectable in LFP experiments.14 Straightforward analy-
sis of the pyridine ylide data yields a lifetime for 1-adaman-
tylcarbene 8 of 1.5 ns (cyclohexane, ambient temperature),
assuming that the absolute rate constant of the reaction of
carbene with pyridine is 1 × 109 M-1 s-1. This is a longer
lifetime than that of perdeuteriomethylcarbene (0.5 ns) but a
shorter lifetime than that of 2-tert-butylcyclopropylcarbene (15
ns) under similar conditions.25
Continuous photolysis of diazirines 5 and 6 in cyclohexane
produces the dimers of adamantene 9 and homoadamantene 11,
respectively. An adduct of 1-adamantylcarbene 8 and cyclo-
hexane is formed in addition to the alkene dimers. This indicates
that the lifetime of 1-adamantylcarbene 8 in cyclohexane is
controlled, at least in part, by reaction with the solvent.
The yield of cyclohexane adduct and the yield of homoada-
mantene dimers decrease in the presence of piperidine, which
traps the 1-adamantylcarbene 8 generated upon photolysis of
1-adamantyldiazirine 6. A plot of the yield of the 1-adaman-
tylcarbene-piperidine adduct/homoadamantene 11 dimers is
linear. This indicates that some homoadamantene is formed by
rearrangement of the trappable carbene and not exclusively from
the excited state of diazirine precursor 6.
Photolysis of 3-noradamantyldiazirine 5 in piperidine, where
cation formation is unlikely, leads to the formation of adduct
14 (Scheme 2) in only trace quantities (<1%).
1-Adamantylcarbene. Photolysis of 1-adamantyldiazirine 6
in cyclohexane leads to the formation of carbene-solvent adduct
24 in 11% yield and dimers of homoadamantene 11 (Scheme
3) in 28% yield. This confirms that 1-adamantylcarbene 8 is
produced upon photolysis of 1-adamantyldiazirine 6, that it is
sufficiently long-lived to undergo bimolecular chemistry (in
contrast to TBC), and that its lifetime of 1.5 ns in cyclohexane
is, in part, limited by reaction with solvent.
The mass spectra of all isomers of the dimers have base peaks
at m/z 296 (M+) and strong peaks at m/z ) 239 and 135 with
different intensity ratios for different isomers. The mixture of
dimers, isolated as a white solid by successive recrystallization
with 2-propanol, yielded two major products. The dimers did
not have vinylic resonances in their proton or carbon NMR
spectra. Our results with the dimers of homoadamantene mimic
those of earlier workers.2,5,17
1-Adamantyldiazirine 6 was photolyzed in a 1/1 mixture of
cyclohexane/cyclohexane-d12. The ratio of cyclohexane to
cyclohexane-d2 adduct is equal to 1.3 by GC-MS. The C-H/
C-D insertion ratio is very similar to the isotope effect on
carbene lifetimes measured during the LFP experiments using
pyridine as a kinetic probe.
Photolysis of 1-adamantyldiazirine 6 in neat piperidine
produced the insertion adduct 25 (Scheme 3) in 21% isolated
yield. As expected from Scheme 3, the yield of adduct 25
increases with increasing concentration of piperidine in cyclo-
hexane (Supporting Information). Increasing the concentration
Continuous photolysis of 3-noradamantyldiazirine 5 in cy-
clohexane, methanol, or acetic acid fails to produce an adduct
of 3-noradamantylcarbene 7, although in every solvent, evidence
for the presence of adamantene can be obtained. Photolysis of
3-noradamantyldiazirine 5 in piperidine forms an adduct in very
low yield (<1%). Either 3-noradamantylcarbene 7 is not formed
upon the photolysis of 3-noradamantyldiazirine 5 or it rearranges
very rapidly (τ < 50 ps). The deduced lifetimes of 3-norada-

Rearrangement of Carbenes to Bridgehead Alkenes
J. Phys. Chem. A, Vol. 105, No. 44, 2001 10153
mantylcarbene 7 and 1-adamantylcarbene 8 are consistent with
the rearrangement barriers predicted by DFT calculations. We
propose that substantial amounts of adamantene 9 are formed
by a rearrangement in the excited state of the 3-noradamantyl-
diazirine precursor (5*).
is much longer lived than either TBC or 3-noradamantylcarbene
7, neither of which can be trapped with pyridine in LFP
experiments. 3-Noradamantylcarbene 7 can be trapped with
piperidine but in only trace quantities. The data indicates that
3-noradamantylcarbene 7 has a lifetime in cyclohexane that is
less than 50 ps. The data also indicates that 3-noradamantyl-
carbene 7 is formed very inefficiently from 3-noradamantyl-
diazirine 5. The data is consistent with the rearrangement in
the excited state of 3-noradamantyldiazirine 5 concurrent with
the loss of nitrogen to efficiently form adamantene 9, without
the intermediacy of 3-noradamantylcarbene 7.
Supporting Information Available: Computational chem-
istry data, UV-vis, IR, and mass spectra for the carbenes. This
material is available free of charge via the Internet at http://
pubs.acs.org.
Several diazirine excited-state rearrangements have been
recently postulated.9,24 The most relevant and convincing
precedent can be found with homocubyldiazirine.9 Chen et al.
demonstrated that photolysis of 26 leads to homocubene 27 and
not homocubylidene 28 as the first trappable intermediate. In
Chen’s work, the strained alkene can subsequently isomerize
to a carbene, a process which does not occur with adamantene
9.9 This is sensible because adamantene 9 is calculated to be
References and Notes
(1) (a) Bredt, J.; Thouet, H.; Schmitz, L. Justus Liebigs Ann. Chem.
1924, 437, 1. (b) Warner, P. M. Chem. ReV. 1989, 89, 1067.
(2) Jones, M., Jr. In AdVances in Carbene Chemistry; Brinker, U., Ed.;
JAI Press: Stamford, CT, 1998; Vol. 2, p 77.
(3) (a) Sheridan, R. S.; Ganzer, G. A. J. Am. Chem. Soc. 1983, 105,
6158. (b) Radziszewski, J. G.; Downing, J. W.; Wentrup, C.; Kaszynski,
P.; Jawdosiuk, M.; Kovacic, P.; Michl, J. J. Am. Chem. Soc. 1985, 107,
2799.
(4) Wilt, J. W.; Schneider, C. A.; Dabek, H. F., Jr.; Kraemer, J. F. J.
Org. Chem. 1966, 31, 1543.
(5) See ref 2, citation no. 4.
(6) (a) Wilt, J. W.; Schneider, C. A.; Dabek, H. F., Jr.; Kraemer, J. F.
J. Org. Chem. 1966, 31, 1543. (b) Wolf, A. D.; Jones, M., Jr. J. Am. Chem.
Soc. 1973, 95, 8209. (c) Ruck, R. T.; Jones, M., Jr. Tetrahedron Lett. 1998,
39, 4433.
(7) (a) Geise, C. M.; Hadad, C. M. J. Am. Chem. Soc. 2000, 122, 5861.
(b) Kohl, E.; Stroter, T.; Siedschlag, C.; Polburn, K.; Szeimies, G. Eur. J.
Org. Chem. 1999, 3057.
(8) (a) Eaton, P. E.; Hoffmann, K.-L. J. Am. Chem. Soc. 1987, 109,
5285. (b) Eaton, P. E.; Appell, R. B. J. Am. Chem. Soc. 1990, 112, 4055.
(c) Eaton, P. E.; White, A. J. J. Org. Chem. 1990, 55, 1321.
(9) Chen, N.; Jones, M., Jr.; White, W. R.; Platz, M. S. J. Am. Chem.
Soc. 1991, 113, 4981.
(10) (a) Frey, H. M. AdV. Photochem. 1964, 4, 225. (b) Frey, H. M.;
Stevens, I. D. R. J. Am. Chem. Soc. 1965, 87, 1. (c) Chang, K.-T.; Shechter,
H. J. Am. Chem. Soc. 1979, 101, 5082. (d) Fukushima, M.; Jones, M., Jr.;
Brinker, U. H. Tetrahedron Lett. 1982, 23, 2212.
(11) Ruck, R. T.; Jones, M., Jr. Tetrahedron Lett. 1998, 39 (25), 4433.
(12) Armstrong, B. M.; McKee, M. L.; Shevlin, R. B. J. Am. Chem.
Soc. 1995, 117, 3685.
(13) (a) Sasaki, T.; Eguchi, S.; Ryu, I. H.; Hirako, Y. Tetrahedron Lett.
1974, 2011. (b) Kaufmann, G. M.; Smith, J. A.; Vander Stouw, G. G.;
Shechter, H. J. Am. Chem. Soc. 1965, 87, 935. (c) Martella, D. J.; Jones,
M., Jr.; Schleyer, P. v. R.; Maier, W. F. J. Am. Chem. Soc. 1979, 101,
7634. (d) Likhotvorik, I. R.; Tae, E. L.; Ventre, C.; Platz, M. S. Tetrahedron
Lett. 2000, 41 (6), 795-796.
33.96 kcal/mol lower in energy than 3-noradamantylcarbene 7,
whereas 27 and 28 are essentially degenerate in energy.29
Why is the excited state of 3-noradamantyldiazirine 5 more
prone to rearrangement than that of 1-adamantyldiazirine 6? We
speculate that the same structural factor (overlap of the migrating
bond) that favors rapid rearrangement of 3-noradamantylcarbene
7 (0.35 kcal/mol) relative to 1-adamantylcarbene 8 (6.06 kcal/
mol) operates in the analogous diazirine excited state. Thus,
carbenes with small barriers to rearrangement will be very
difficult to detect not only because of their short lifetimes but
also because of facile rearrangements in structurally related
precursors that lead to inefficient carbene formation.
(14) Tae, E. L.; Zhu, Z.; Platz, M. S. J. Phys. Chem. A 2001, 105, 3803.
(15) (a) Kropp, P. J.; Adkins, R. L. J. Am. Chem. Soc. 1991, 113, 2709.
(b) Davidson, R. I.; Kropp, P. J. J. Org. Chem. 1982, 47, 1904.
(16) Farcasiu, M.; Farcasiu, D.; Conlin, R. T.; Jones, M., Jr.; Schleyer,
P. v. R. J. Am. Chem. Soc. 1973, 95, 8207.
(17) Sosnowski, J. J.; Danaher, E. B.; Murray, R. K., Jr. J. Org. Chem.
1985, 50, 2759.
(18) Gritsan, N. P.; Zhai, H. B.; Yuzawa, T.; Karweik, D.; Brooke, J.;
Platz, M. S. J. Phys. Chem. A 1997, 101, 2833.
(19) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Gill, P. M. W.;
Johnson, B. G.; Robb, M. A.; Cheeseman, J. R.; Keith, T.; Petersson, G.
A.; Montgomery, J. A., Jr.; Raghavachari, K.; Al-Laham, M. S.; Zakrzewski,
V. G.; Ortiz, J. V.; Foresman, J. B.; Peng, C. Y.; Avala, P. Y.; Chen, W.;
Wong, M. W.; Andres, J. L.; Replogle, E. S.; Gomperts, R.; Martin, R. L.;
Fox, D. J.; Binkley, J. S.; Defrees, D. J.; Baker, J.; Stewart, J. J. P.; Head-
Gordon, M.; Gonzalez, C.; Pople, J. A. Gaussian94, Revision D.3; Gaussian,
Inc.: Pittsburgh, PA, 1995.
(20) (a) Becke, A. D. Phys. ReV. A: At., Mol., Opt. Phys. 1988, 38,
3098. (b) Becke, A. D. J. Chem. Phys. 1993, 98, 5648. (c) Lee, C.; Yang,
W.; Parr, R. G. Phys. ReV. B: Solid State 1988, 37, 785. (d) Labanowski,
J. W.; Andzelm, J. Density Functional Methods in Chemistry; Springer:
New York, 1991. (e) Parr, R. G.; Yang, W. Density Functional Theory in
Atoms and Molecules; Oxford University Press: New York, 1989.
Conclusions
Photolysis of 1-adamantyldiazirine 6 in argon at 14 K
produces 1-adamantyldiazomethane and homoadamantene 11,
which were characterized by IR spectroscopy. 1-Adamantyl-
carbene 8 was not detected as a persistent species in the argon
matrix. LFP of 1-adamantyldiazirine 6 produces a carbene-
derived ylide in the presence of pyridine. 1-Adamantylcarbene
8 can also be trapped in solution with cyclohexane and with
piperidine. Dimers of homoadamantene 11 are detected upon
photolysis of 1-adamantyldiazirine 6. An adduct of homoada-
mantene 11 and acetic acid was formed upon the photolysis of
1-adamantyldiazirine 6 in acetic acid. The lifetime of 1-ada-
mantylcarbene 8 in cyclohexane is deduced to be ≈1.5 ns. This
is consistent with B3LYP calculations with the 6-31G* basis
set, which predict a barrier to the rearrangement of 1-adaman-
tylcarbene 8 to homoadamantene 11 of 6.06 kcal/mol after zero-
point energy correction. It is clear that 1-adamantylcarbene 8

10154 J. Phys. Chem. A, Vol. 105, No. 44, 2001
Tae et al.
(21) (a) Zener, M. C.; Ridley, J. E. Theor. Chim. Acta 1973, 32, 111.
(b) Edwards, W. D.; Zewer, M. C. Theor. Chim. Acta 1987, 72, 347.
(22) Scott, A. P.; Platz, M. S.; Radom, L. J. Am. Chem. Soc. 2001, 123,
6069.
(23) Slobodin, Y. M.; Ashkinazi, L. A.; Klimchuk, G. N.; Zabuslaeva,
L. A. Russ. J. Org. Chem. 1985, 1953.
(24) Martella, D. J.; Jones, M., Jr.; Schleyer, P. v. R.; Maier, W. F. J.
Am. Chem. Soc. 1979, 101, 7634.
(25) Yao, G.; Rempala, P.; Bashore, C.; Sheridan, R. S. Tetrahedron
Lett. 1999, 40, 17.
Downing, J. W.; Wiberg, K. B.; Walker, F. H.; Miller, R. D.; Kovacic, P.;
Jawdosiuk, M.; Bonacic-Koutecky, V. Pure Appl. Chem. 1983, 55, 313.
(c) Conlin, R. T.; Miller, R. D.; Michl, J. J. Am. Chem. Soc. 1979, 101,
7637. (d) Bian, N.; Jones, M., Jr. J. Am. Chem. Soc. 1995, 117, 8957. (e)
Martella, D. J.; Jones, M., Jr.; Schleyer, P. v. R. J. Am. Chem. Soc. 1978,
100, 2896. (f) Hirai, K.; Tomioka, H.; Okazaki, T.; Tokunaga, K.; Kitagawa,
T.; Takeuchi, K. J. Phys. Org. Chem. 1999, 12, 165.
(27) Huang, H.; Platz, M. S. J. Am. Chem. Soc. 1998, 120, 5990.
(28) Bucher, G.; Scaiano, J. C.; Platz, M. S. Kinetics of Carbene
Reactions in Solution; Landolt-Bornstein, Group II; Springer: Berlin,
Germany, 1998; Vol. 18 E2, p 141.
(26) (a) Michl, J.; Radziszewski, J. G.; Downing, J. W.; Kopecky, J.
Pure Appl. Chem. 1987, 59, 1613. (b) Michl, J.; Radziszewski, G. J.;
(29) Hrovat, D. A.; Borden, W. T. J. Am. Chem. Soc. 1992, 114, 2719.