Laser-induced carbene-carbene rearrangement in solution: The diphenylcarbene-fluorene rearrangement

Article abstract of DOI:10.1021/jo401607m

Diphenylcarbene (DPC) generated by high-intensity laser photolysis of diphenyldiazomethane rearranges to fluorene (FL) by two distinct mechanisms as revealed by methyl-group labeling. Thus, excimer laser irradiation of p,p′-dimethyldiphenyldiazomethane generates 3,6-dimethylfluorene (3,6-DMF) and 2,7-dimethylfluorene (2,7-DMF), which were identified by fluorescence measurements as well as GC-MS and comparison with authentic materials. 3,6-DMF corresponds to direct bond formation between ortho positions in DPC, referred to as ortho,ortho′ coupling. 2,7-DMF corresponds to a carbene-carbene rearrangement, whereby DPC undergoes ring expansion to phenylcycloheptatetraene (PhCHT) followed by ring contraction to o-biphenylylcarbene (o-BPC), which then cyclizes to FL. The carbene-carbene rearrangement dominates over the ortho,ortho′ coupling under all conditions employed. The ortho,ortho′ coupling must take place in a higher excited state (most likely S₂ or T₁) of DPC, because it is not observed at all under thermolysis conditions, where only S₁ and T₀ are populated. The carbene-carbene rearrangement may take place either in a hot S₁ state or more likely in a higher excited state (S₂ or T₁).

Full text of DOI:10.1021/jo401607m

Article
pubs.acs.org/joc
Laser-Induced Carbene−Carbene Rearrangement in Solution: The
Diphenylcarbene−Fluorene Rearrangement
Michele J. Regimbald-Krnel and Curt Wentrup*
̀ ́
School of Chemistry and Molecular Biosciences, The University of Queensland, Brisbane, Queensland 4072, Australia
*
S Supporting Information
ABSTRACT: Diphenylcarbene (DPC) generated by high-
intensity laser photolysis of diphenyldiazomethane rearranges
to fluorene (FL) by two distinct mechanisms as revealed by
methyl-group labeling. Thus, excimer laser irradiation of p,p′-
dimethyldiphenyldiazomethane generates 3,6-dimethylfluorene
(
3,6-DMF) and 2,7-dimethylfluorene (2,7-DMF), which were
identified by fluorescence measurements as well as GC-MS
and comparison with authentic materials. 3,6-DMF corre-
sponds to direct bond formation between ortho positions in
DPC, referred to as ortho,ortho′ coupling. 2,7-DMF corre-
sponds to a carbene−carbene rearrangement, whereby DPC
undergoes ring expansion to phenylcycloheptatetraene
(
PhCHT) followed by ring contraction to o-biphenylylcarbene (o-BPC), which then cyclizes to FL. The carbene−carbene
rearrangement dominates over the ortho,ortho′ coupling under all conditions employed. The ortho,ortho′ coupling must take place
in a higher excited state (most likely S or T ) of DPC, because it is not observed at all under thermolysis conditions, where only
2
1
S and T are populated. The carbene−carbene rearrangement may take place either in a hot S state or more likely in a higher
1
0
1
excited state (S or T ).
2
1
8
INTRODUCTION
T gap and slower intersystem crossing of S to T . Thus, the S
1 0 1
■
lifetime is ∼120 ps in cyclohexane and ∼340 ps in
acetonitrile. The S singlet state is formed from DPDM
within 30 ps of a 266 nm laser pump pulse and decays to T in
In 1980, Turro and co-workers reported the rearrangement of
diphenylcarbene (DPC), 2, to fluorene, 3, in the high-intensity
8
,9
1
0
excimer laser photolysis of diphenyldiazomethane (DPDM), 1.
1
0,11
1
300 fs.
3
S and T have UV absorption maxima at ∼370 and
1
0
(
Scheme 1). In addition, tetraphenylethene (TPE), 4,
11,12
20 nm, respectively.
The excited T state of DPC absorbs
1
diphenylphenanthrene (DPP), 5, and diphenylanthracene
DPA), 6, were also formed.
While TPE is a “normal” carbene reaction product, formed
by dimerization of the ground state T carbene, fluorene (3),
DPP (5), and DPA (6) have never been observed in
conventional UV lamp photolysis of 1, where benzophenone
azine (BPA) and TPE are major products (Scheme 2). It is
known that photolysis of TPE followed by air oxidation can
1
3
at ca. 360 nm in CH CN, fluoresces at ∼500 nm, and decays
3
(
to the ground state triplet, T , on a time scale of 4−9 ns in
0
6
,13
2
hydrocarbon solution or 30−140 ns time scale in an ethanol
0
14
glass at 77 K. The T state is formed under the high-intensity
1
laser conditions used by Turro and co-workers, who observed
1
2
,3
its emission at 507 nm.
The formation of fluorene (FL), 3, under conditions of high
photon density has been proposed to occur via the exited states
or T1. Turro and co-workers stated that the
formation of FL is dependent on the laser intensity and
suggested that it might be produced from S . Wilson and
Schnapp proposed further excitation of S to a higher excited
singlet state, presumably S . This is referred to as DPC* in
Scheme 3. It was suggested that this state might have a π-
system similar to the diphenylmethyl cation and undergo
photochemical cyclization to 7 followed by hydrogen migration
to give FL, 3 (Scheme 3, route a). However, FL was not
produced under the laser-jet photolysis conditions used by
4
lead to DPP formation, but this is apparently not the major
1
,2,15
16
1
S1
source of DPP in the excimer laser photolysis. Under high-
intensity photolysis conditions, a relatively high concentration
of diphenylcarbene (DPC), 2, is obtained, thus making
multiphoton chemistry and intertransient reactions possible,
here leading to FL, DPP, and DPA.
1
2
1
1
2
DPC has a triplet ground state, T , which has been well
0
characterized by ESR, UV (absorption and emission), and IR
5
spectroscopy. Experiments and calculations indicate that the
2
6
,7
first excited singlet state, S , lies ca. 3 kcal/mol above T . The
1
0
small energy gap causes fast S → T conversion in fluid
1
0
2
,6
10
−1
these authors.
solution (k ≈ 10 s ), and the reverse T → S reaction is ca.
0
1
6
3
orders of magnitude slower. The S−T conversion is solvent
dependent: a polar solvent such as acetonitrile stabilizes the
Received: July 24, 2013
singlet, probably by ylide formation. This leads to a smaller S−
©
XXXX American Chemical Society
A
dx.doi.org/10.1021/jo401607m | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Scheme 1. Products of High-Intensity Laser Photolysis of DPDM
Scheme 2. Products of Conventional Photolysis of DPDM
Scheme 3. Two Proposed Mechanisms for ortho,ortho′-
(FVT) conditions (Scheme 4). Here, it is well established that
the reaction proceeds via carbene−carbene rearrangement
Coupling in Diphenylcarbene
Scheme 4. The Carbene−Carbene Mechanism for Fluorene
Formation from Diphenylcarbene under FVT Conditions
Despres et al. proposed another mechanism (Scheme 3,
route b) whereby the first excited triplet state, T , was held
1
16
responsible for the production of FL, DPP, and DPA. In
particular, it was expected that FL, 3, was formed by the
ortho,ortho′ bond formation (Scheme 3).
through ring expansion to phenylcycloheptatetraene
(PhCHT) and ring contraction to o-biphenylylcarbene (o-
17
BPC). An ortho,ortho′ bond formation is not involved at all.
Both of the proposed mechanisms for the formation of FL in
Scheme 3 involve bond formation between the ortho,ortho′
positions of DPC. Quite a different mechanism has been
established for the thermal isomerization of DPC to FL, which
takes place in the gas phase under flash vacuum thermolysis
Therefore, in order to determine whether the carbene−
carbene rearrangement is involved in the laser-induced reaction,
we have investigated the photolysis of p,p′-dimethyldiphenyl-
diazomethane (DMPDM), 8 (Scheme 5). The direct
ortho,ortho′-coupling would produce 3,6-dimethylfluorene
B
dx.doi.org/10.1021/jo401607m | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
(
3,6-DMFL), 9, whereas the carbene−carbene rearrangement
would produce 2,7-dimethylfluorene (2,7-DMFL), 10.
Scheme 5
RESULTS
■
18
The laser-drop (LD) photolysis technique has proven very
valuable for preparative multiphoton laser photochemistry. It
has the advantage of directing the laser pulse to a single drop of
solution, favoring multiphoton chemistry to take place. Laser
excitation of the hanging drop of solution causes it to burst into
smaller droplets that are splattered onto the walls of the
photolysis cell. The photolyzed drop is then collected in the
bottom part of the cell, out of the path of the next laser pulse
directed at the new drop that is forming. Therefore, we used
this technique in the photolyses of diphenyldiazomethanes with
Figure 2. Emission spectrum (λexc = 250 nm) of hexane solutions of
(a) fluorene (FL), (b) 9,10-diphenylphenanthrene (DPP), and (c)
9
,10-diphenylanthracene (DPA).
3
08 and 248 nm pulsed excimer lasers delivering 75−90 and
Poorly resolved emission bands with maxima at 355, 375,
395, and 420 nm are also observed in Figure 1. These are due
75−180 mJ pulses, respectively.
−5
A concentration of 5 × 10 M of DPDM in hexane was used
to 9,10-diphenylphenanthrene (DPP), 5, and 9,10-diphenylan-
for the laser-drop experiments, since dilution will favor the
unimolecular process(es) responsible for fluorene formation.
Figure 1 gives the emission spectrum of a photolyzate recorded
using an excitation wavelength of 250 nm. The spectrum is in
1
thracene (DPA), 6. Figure 2b,c gives the emission spectra of
authentic samples of DPP and DPA in hexane for comparison.
A second cycle of LD irradiation did not result in an increase
in the fluorene emission. Moreover, green emission corre-
sponding to the relaxation of T to T of DPC was only
1
good agreement with the report by Turro et al. and dominated
1
0
by an emission band arising from fluorene with maxima at 305
and 315 nm. This was confirmed by comparison with the
emission spectrum obtained for an authentic sample of FL in
hexane (Figure 2a).
observed during the first cycle of LD photolysis. Accordingly,
one cycle of LD photolysis was deemed sufficient to
decompose all of the diphenyldiazomethane.
As mentioned above, the first excited state of DPC (S ) is
1
longer-lived in polar solvents (∼340 ps in CH CN) than in
3
6
,8
hydrocarbon solvents (∼120 ps in cyclohexane). A laser-drop
−
5
photolysis of DDM in CH CN (5 × 10 M) revealed that
3
fluorene was no longer a major product. In fact, shoulders at
3
05 and 315 nm suggest that only a trace of FL was formed
(
See Figure S1, Supporting Information). A reversible
formation of an acetonitrile−diphenylcarbene ylide may be
6
responsible. The enhanced stabilization of the singlet state of
DPC not only slows intersystem crossing but may also allow
vibrational deactivation before any chemical reactions can take
place.
A preparative scale laser-drop photolysis of DPDM was
carried out in order to make fluorene detectable by GC-MS.
For this, 7.8 mg of 1 in 750 mL of hexane was photolyzed, and
the resulting mixture was purified by preparative TLC using
hexane as the eluent. Fluorescence analysis was used to identify
the extracts containing fluorene. GC-MS analysis of the
fluorene fraction confirmed the presence of FL by direct
comparison with an authentic sample.
Hexane solutions of DPDM (5 × 10− M or 10−1 M) were
also photolyzed at 254 nm using a low-pressure mercury lamp
for 90, 110, 135, or 180 min. The concentrated photolysis
mixtures were then purified by microcolumn chromatography
prior to emission analysis. As shown in Figure S2, Supporting
Information, although there is emission in the 300−320 nm
region of the spectra, the characteristic band shape of FL is not
obvious. If any fluorene was formed, it was only a trace amount.
5
Figure 1. Emission spectrum (λ = 250 nm) after laser-drop
photolysis of a 5 × 10 M solution of DPDM in hexane at 308 nm
exc
−5
(
90 mJ/pulse).
C
dx.doi.org/10.1021/jo401607m | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Laser-Drop Photolysis of p,p′-Dimethyldiphenyldia-
zomethane (DMPDM). Having determined the necessary
conditions to obtain detectable amounts of FL, it was necessary
to be able to distinguish between the two possible
dimethylfluorene isomers: 3,6-DMFL (9) and 2,7-DMFL
Chart 1
(10). As shown in Figure S3, Supporting Information, the
emission spectra of authentic samples of thee compounds as
well as FL are very similar, although they are slightly shifted
with respect to one another.
However, it was possible to distinguish between 3,6-DMFL
and 2,7-DMFL by gas chromatography on either a BP21
(poly(ethylene glycol) phase, 50 m) column or a BP10 (14%
cyanopropylphenyl siloxane, 86% dimethyl siloxane phase, 50
m) column, with 3,6-DMFL eluting before 2,7-DMFL in both
cases. Both isomers had the same intensity response. A laser-
drop photolysis of DMPDM, 8, on a small scale demonstrated
dimethylfluorene formation. A green emission was again
emission spectra of the products clearly showed that fluorenes
were obtained (Figure S5, Supporting Information).
GC-MS analyses revealed that a mixture of both dimethyl-
fluorene isomers was obtained in all experiments (Table 1 and
observed, indicating that the T state of p,p′-dimethyldiphe-
1
nylcarbene was formed. The fluorescence of this first excited
Table 1. Fluorenes Formed in High-Intensity Laser-Drop
Photolyses of p,p′-Dimethyldiphenyldiazomethane in
Hexane (5 × 10− M)
19
triplet state is known to peak at 517 nm in acetonitrile. The
5
spectrum of the photolyzate had an emission band in the 310−
320 nm region, which is compatible with a mixture of 3,6-
a
laser wavelength (nm) energy (mJ/pulse) ratio of 3,6-DMFL/2,7-DMFL
DMFL and 2,7-DMFL (Figure 3 and Figure S4, Supporting
Information), although fluorescence measurements do not
permit an evaluation of the amounts of the two fluorenes.
3
2
2
08
48
48
85−90
85−95
75−180
1:15
1:8.5
1:6.5
a
Ratios are based on integrated peak areas of GC traces obtained with
the BP21 capillary column.
Figure 4). A ratio of 1:15 for 3,6-DMFL vs 2,7-DMFL was
obtained at 308 nm, and 1:8.5 or 1:6.5 at 248 nm. It appears
that higher energy causes formation of relatively more 3,6-
DMF.
Co-injections of authentic samples of 3,6-DMFL and 2,7-
DMFL confirmed the assignments and did not result in the
appearance of new peaks in the GC traces. This was confirmed
on the two different GC columns (Figure 4).
Control LD photolysis experiments at 248 nm confirmed
that there was no photochemical interconversion of 2,7-DMFL
and 3,6-DMFL.
DISCUSSION
■
The C H energy surface connecting phenylcarbene, cyclo-
7
7
hepta-1,2,4,6-tetraene, and other isomers has been the subject
of extensive calculations. Very similar calculated activation
barriers were found at the B3LYP, G2(MP2), CCSD(T), and
CASPT2N levels of theory. The energies of the correspond-
ing species on the energy surface for the diphenylcarbene
rearrangement at the BLYP level indicate that the highest
energy barrier, TS4, lies only ca. 23 kcal/mol above the singlet
20
20
Figure 3. Emission spectrum (λ = 250 nm) obtained after laser-drop
exc
−5
DPC, 2 (S ) (Scheme 6). This is similar to the barriers
1
photolysis of a 5 × 10 M solution of DMPDM in hexane at 308 nm
calculated for the ring expansion of phenylcarbene itself and
(
90 mJ/pulse).
21
other substituted phenylcarbenes. The formation of fluorene,
, of course, is highly exothermic, by 75 kcal/mol. Nevertheless,
3
fluorene formation has never been observed in solution
thermolysis because other, faster bimolecular reactions take
place. The carbene−carbene rearrangement to fluorene takes
place only in the low-pressure flash vacuum thermolysis, where
bimolecular reactions are disfavored. The cyclization of o-
biphenylylcarbene to fluorene involves addition to the benzene
Other emission bands present at 360, 380, and 415 nm in
Figure 3 are ascribed to phenanthrene and anthracene
derivatives 11 and 12 (Chart 1) analogous to the photo-
products DPP and DPA in Figure 1, but these compounds were
not investigated further.
Preparative scale laser-drop experiments on DMPDM were
performed as above using the two different lasers, and each
photolysis mixture was purified as described above. The
22
ring to give 8aH-fluorene, followed by a H-shift.
The calculations in Scheme 6 pertain to the lowest singlet
state of DPC, S . We do not know that the same path is
1
D
dx.doi.org/10.1021/jo401607m | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Scheme 6. Diphenylcarbene-Fluorene Rearangement in the
a
S1 State
a
Relative energies + ZPVE in kcal/mol at the B3LYP/6-311+G**//
B3LYP/6-31G* level.
relaxation of the S state of DPC takes place on a time scale of
1
12
∼
10 ps in cyclohexane. Generally, vibrational relaxation of hot
polyatomic molecules takes place in tens to hundreds of
23
picoseconds in solution, and molecular rearrangements may
take place on a similar time scale. Although chemical
activation is usually restricted to the gas phase, it is sometimes
invoked to explain unexpected reactivity in solution.
24
25
26
The formation of 3,6-dimethylfluorene from DMPD was
observed in all the laser-drop experiments. This is the
ortho,ortho′ bond forming reaction. The proportion of this
route to fluorene appears to increase with increasing laser
energy (Table 1), but it is always less than the carbene−carbene
rearrangement route. The spin state responsible for ortho,ortho′
coupling cannot be T or S because this route is not followed
0
1
at all under FVT conditions.
CONCLUSION AND OUTLOOK
■
8
Laser-drop photolyses of p,p′-dimethyldiphenyldiazomethane,
, revealed that both 3,6-dimethylfluorene, 9, and 2,7-
dimethylfluorene, 10, are formed. Formation of 3,6-DMFL
corresponds to the ortho,ortho′ bond forming reaction, but
formation of 2,7-DMFL corresponds to the carbene−carbene
rearrangement. The latter is the dominant reaction under all
laser photolysis conditions investigated, but higher laser energy
allows proportionately more ortho,ortho′ coupling to occur.
This is a normal outcome for competing reactions. The ratio of
the two processes depends on the wavelength and the light
intensity. A priori, three possible scenarios for the carbene−
carbene rearrangement can be envisaged: (i) a hot singlet state
Figure 4. GC traces (BP21 column) for the laser-drop photolysis of
DMPDM at 248 nm (75−180 mJ/pulse). (a) Purified photolysis
mixture after laser-drop photolysis of a 5 × 10 M solution of
DMDPD in hexane at 248 nm (75−180 mJ/pulse); (b) co-injection of
photolysis mixture with 3,6-DMFL; (c) co-injection of photolysis
mixture with 2,7-DMFL; (d) pure 3,6-DMFL; (e) pure 2,7-DMFL.
−5
followed in higher excited states of DPC, and only more
elaborate calculations can reveal this. It may well be that not all
the species shown in Scheme 6 are discrete intermediates on
excited state energy surfaces.
The laser-drop photolysis experiments demonstrate that
high-intensity laser irradiation does cause carbene−carbene
rearrangement to occur in solution, nominally at room
temperature. There are three possible explanations: (i) a hot
S of the diphenylcarbene undergoes a rapid thermal carbene−
1
carbene rearrangement, although vibrational relaxation is
expected to take place on a time scale of tens of picoseconds.
(ii) A higher excited singlet state S rearranges in a manner
n
similar to S to give fluorene. (iii) An excited triplet state
1
initiates the reaction, which may proceed via a surface crossing.
25
singlet state S of the diphenylcarbene is formed, which then
Chemical activation (hot molecules) in solution does not
seem very likely in the present case, which involves a double
carbene−carbene rearrangement. The advantage of the laser-
drop method with high-intensity excimer lasers is that it allows
two-photon photochemistry. Thus, it is likely that the
rearrangement occurs in a higher excited state (S or T ).
1
undergoes a very fast thermal carbene−carbene rearrangement;
(
ii) a higher excited singlet state S of DPC (most likely S )
n
2
rearranges in a manner similar to S to give fluorene; (iii)
1
because of the possibility of surface crossings, the reaction
could also be initiated from the excited triplet state, T1.
Conversion between upper excited states is expected to take
place on a time scale of tens of picoseconds. Vibrational
2
1
The ortho,ortho′ coupling must be a reaction in a higher excited
state, because it does not take place under normal FVT
E
dx.doi.org/10.1021/jo401607m | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
conditions, where only S and T are likely to be populated.
Calculations of the reactivities of excited states (particularly S2
mL) were photolyzed using a 40 W low-pressure Hg lamp (254 nm).
Solutions were magnetically stirred during photolysis. The progress of
the reaction was monitored by emission spectroscopy or UV−visible
spectroscopy. Photolyzates were then concentrated on a rotary
evaporator. The residue obtained was chromatographed on silica
using a Pasteur pipet and eluted with hexane. All chromatography
fractions collected (1−2 mL) were analyzed by emission spectroscopy.
Laser-Drop Photolyses. The general procedure described in the
1
0
and T ) of these and other carbenes are now being planned.
1
EXPERIMENTAL SECTION
Diphenyldiazomethane (DPDM), 1, tetraphenylethylene (TPE),
■
27
1,4
1,4
4
, 9,10-diphenylphenanthrene (DPP), 5, 9,10-diphenylanthracene
1
8
1
23
literature was followed with some modifications. The beam from
either a Lambda Physik EMG 101 MSC excimer laser (Xe/HCl/Ne,
(
DPA), 6, p,p′-dimethyldiphenyldiazomethane (DMPDM), 8, 3,6-
17a,28
DMFL, 9, and 2,7-DMFL, 10,
were prepared according to
3
08 nm, ∼15 ns) or a Lumonics Excimer-600 laser (Kr/F /Ne, 248
literature procedures, and fluorene (FL), 3, was a commercial sample.
Once prepared, diazo compounds were stored in the freezer. Due to
slow decomposition over time, they were purified by column
chromatography prior to photolysis using basic Al O and hexane as
2
nm, ∼15 ns) was focused by means of a quartz lens (F = 200) into a
drop of the photolysis solution suspended from a 2-in. syringe needle
(
20 gauge). The flow rate of the solution was controlled by a syringe
2
3
pump. The lasers were run at a constant repetition rate of 10 Hz in
eluent.
order to obtain high laser power. The N -purged solution (ca. 45 mL)
Steady-state fluorescence spectra were measured with a lumines-
cence spectrometer using spectral grade hexanes. A spectral bandwidth
of 5 nm in excitation and emission was used. Excitation wavelengths of
2
was syringed out of the purging vessel using a 50 mL gastight syringe
with Teflon-tipped plunger just prior to photolysis. A volume of 250
mL of solution to be photolyzed was prepared at a time and stored in
2
4
50 nm were applied, and the spectra were recorded in the range 280−
the dark prior to N -purging.
80 nm. Cross-contamination of solutions was avoided at all times by
2
Laser-Drop Photolysis of Diphenyldiazomethane, 1. A solution of
washing the quartz cuvette with hexane, then measuring emission of
hexane before proceeding to the next solution.
GC-MS analyses were performed on a quadrupole mass
spectrometer connected to a gas chromatograph equipped with a
BP-5 capillary column (30 m × 0.25 mm with 0.25 μm phase
thickness; He carrier at 20 psi head pressure; injector 200 °C; detector
−5
7
.8 mg of DPDM in 750 mL of hexane (5 × 10 M) was photolyzed
at 308 nm (75−85 mJ/pulse over 4 days). The photolyzate was
concentrated on a rotary evaporator and purified by preparative TLC
(see details above). GC-MS analysis permitted the identification of
fluorene by direct comparison with authentic material.
Laser-Drop Photolyses of p,p′-Dimethyldiphenyldiazomethane,
2
80 °C; column temperature 100−250 °C, programmed at 16 °C/
min).
Purification of laser photolysis mixtures was first done by thin-layer
8
(
. A solution of 17.4 mg of p,p′-dimethyldiphenyldiazomethane
−5
DMPDM) in 1.5 L of hexane (5 × 10 M) was photolyzed at 308
nm (85−90 mJ/pulse over 5 days). After purification of the
photolyzate, GC analysis using the BP21 capillary column revealed
the presence of both 3,6-DMFL and 2,7-DMFL in a ratio of 1.0:15.
DMPDM, 12.4 mg in 1.0 L of hexane, was photolyzed by LDP at 248
nm (75−180 mJ/pulse over 3 days). A mixture of 3,6-DMFL and 2,7-
DMFL in a ratio of 1.0:6.6 was obtained. DMPDM, 11.6 mg in 1.0 L
of hexane, was photolyzed by LDP at 248 nm (85−95 mJ/pulse) over
chromatography (SiO 60, F-254, 1, or 0.25 mm thickness). The plates
2
were normally eluted 2 to 3 times using spectral grade hexanes. All
fractions recovered from the plates were extracted once with spectral
grade hexane. The hexane extracts were then analyzed by fluorescence
spectroscopy for fluorene/dimethylfluorene emission. All fractions
containing fluorene/dimethylfluorene were then extracted several
times with distilled chloroform or hexane until no more fluorene/
dimethylfluorene emission was detected. The extracts were then
combined and concentrated. The residue obtained was purified once
or twice more by preparative TLC prior to GC analysis.
Purifications by HPLC used a Si-80-125-C5 normal phase analytical
column (eluent, hexane; flow rate, 1 mL/min; fluorescence detection
λexc 250 nm, λem 320 nm).
Characterization of the dimethylfluorene isomers in the various
laser photolysis mixtures was done by gas chromatographic analysis on
either a BP21 or a BP10 capillary column (50 m × 0.22 mm with 0.25
μm phase thickness; He carrier at 42 psi head pressure; injector 200
2
days. A mixture of 3,6-DMFL and 2,7-DMFL in a ratio of 1.0:8.4 was
obtained. The presence of both isomers was confirmed by co-injection
on the BP21 column.
Laser-Drop Photolysis of 2,7-Dimethylfluorene, 10. A solution of
.0 mg of 2,7-DMFL 10 in 250 mL of hexane (2 × 10− M) was
5
1
photolyzed by LDP at 248 nm (115−120 mJ/pulse). The photolyzate
was analyzed by emission spectroscopy and by capillary GC (BP21
column). No 3,6-DMFL was detected. Only 2,7-DMFL was present.
ASSOCIATED CONTENT
■
°
C; detector 280 °C; column temperature 100−180 °C, programmed
at 16 °C/min. Separation of 3,6-dimethylfluorene (3,6-DMFL), 9, and
,7-dimethylfluorene (2,7-DMFL), 10, was achieved on both of these
*
S
Supporting Information
2
Figures S1−S5 showing additional emission spectra. This
material is available free of charge via the Internet at http://
pubs.acs.org.
columns. After two to three purifications by preparative TLC,
photolysis mixtures were first analyzed on the BP21 column to
determine the ratio of both isomers present. A control injection of a
1
:1 standard solution of 3,6-DMFL and 2,7-DMFL in hexane
AUTHOR INFORMATION
confirmed that both isomers had the same detector response on the
BP21 column. The dimethylfluorene isomers were identified by co-
injection with authentic materials on the two different capillary
columns described above. To avoid contamination, three different
syringes were used, and blank (solvent only) injections were done after
each GC analysis.
■
*
Corresponding Author
E-mail: wentrup@uq.edu.au.
Notes
The authors declare no competing financial interest.
Photolyses. All photolyses were conducted by using spectral grade
hexane, which was distilled over anhydrous MgSO using a distillation
column (45 cm × 2 mm i.d.) filled with glass O-rings. Solutions of
diazo compounds were purged (in the dark) with ultrahigh purity N2
4
ACKNOWLEDGMENTS
■
This paper is dedicated to the memory of Nicholas Turro. The
Australian Research Council and The University of Queensland
supported this work. We thank Professor Halina Rubinstein-
Dunlop for many discussions and advice on laser work. M.J.R.-
K. is indebted to the Natural Science and Engineering Council
of Canada and to The University of Queensland for
Scholarships.
for at least 20 min prior to irradiation. For the static lamp irradiations,
the purging of solutions was done in the quartz vessels used for
photolysis. For laser-drop irradiations, the purging of solutions was
done in round-bottom flasks capped with rubber septa, which had
been bleached in toluene.
Low-Intensity Irradiations. Solutions of DPDM, 1, in hexane at
−
5
−3
either 5 × 10 M (1.5 mg/150 mL) or 1 × 10 M (29.6 mg/150
F
dx.doi.org/10.1021/jo401607m | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
(
27) Baltzly, R.; Mehta, N. B.; Russell, P. B.; Brooks, R. E.; Grivsky,
REFERENCES
■
E. M.; Steinberg, A. M. J. Org. Chem. 1961, 26, 3669.
28) Chardonnens, L.; Wurmli, A. Helv. Chim. Acta 1946, 29, 922.
(
1) Turro, N. J.; Aikawa, M.; Butcher, J. A.; Griffin, G. W. J. Am.
(
̈
Chem. Soc. 1980, 102, 5127−5128.
2) Wilson, R. M.; Schnapp, K. A. Chem. Rev. 1993, 93, 223−249.
3) (a) Reimlinger, H. Chem. Ber. 1964, 97, 339. (b) Durr, H. Top.
(
(
̈
Curr. Chem. 1975, 55, 87. (c) Closs, G. L.; Rabinow, B. E. J. Am. Chem.
Soc. 1976, 98, 8190. (d) Griller, D.; Majewski, M.; McGimpsey, W. G.;
Nazran, A. S.; Scaiano, J. C. J. Org. Chem. 1988, 53, 1550−1553.
(
4) Bunker, C. E.; Hamilton, N. B.; Sun, Y.-P. Anal. Chem. 1993, 65,
3
(
1
(
460−3465.
5) Sander, W.; Bucher, G.; Wierlacher, S. Chem. Rev. 1993, 93,
583−1621.
6) (a) Eisenthal, K. B.; Turro, N. J.; Aikawa, M.; Butcher, J. A., Jr.;
DuPuy, C.; Hefferon, G.; Hetherington, W.; Korenowski, G. M.;
McAuliffe, M. J. J. Am. Chem. Soc. 1980, 102, 6563−6565.
(
b) Eisenthal, K. B.; Moss, R. A.; Turro, N. J. Science 1984, 225,
1
439−1445. (c) Eisenthal, K. B.; Turro, N. J.; Sitzmann, E. V.; Gould,
I. R.; Hefferon, G.; Langan, J.; Cha, Y. Tetrahedron 1985, 41, 1543−
1
1
(
554. (d) Langan, J.; Sitzman, E. V.; Eisenthal, K. B. Chem. Phys. Lett.
986, 124, 59−62.
7) Woodcock, H. L.; Moran, D.; Brooks, B. R.; Schleyer, P. v. R.;
Schaefer, H. F. J. Am. Chem. Soc. 2007, 129, 3763.
8) Wang, J.; Kubicki, J.; Peng, H.; Platz, M. S. J. Am. Chem. Soc.
008, 130, 6604−6609.
9) Sitzmann, E. V.; Langan, J.; Eisenthal, K. B. J. Am. Chem. Soc.
984, 106, 1868−1869.
10) Portella-Oberli, M. T.; Jeannin, C.; Soep, B.; Zerza, G.; Chergui,
M. Chem. Phys. Lett. 1998, 296, 323−328.
11) Peon, J.; Polshakov, D.; Kohler, B. J. Am. Chem. Soc. 2002, 124,
428−6438.
12) Hadel, L. M.; Maloney, V. M.; Platz, M. S.; McGimpsey, W. G.;
Scaiano, J. C. J. Phys. Chem. 1986, 90, 2488−2491.
13) Horn, K. A.; Allison, B. D. Chem. Phys. Lett. 1985, 116, 114−
18.
14) (a) Scaiano, J. C.; Weir, D. Can. J. Chem. 1988, 66, 491−494.
b) Kohtani, S.; Sugita, C.; Itoh, M. J. Phys. Chem. 1996, 100, 17735−
7739.
15) Wilson, R. M.; Schnapp, K. A. J. Am. Chem. Soc. 1988, 110,
82−983.
16) Despres, A.; Lejeune, V.; Migirdicyan, E.; Admasu, A.; Platz, M.
(
2
(
1
(
(
6
(
(
1
(
(
1
(
9
(
S.; Berthier, G.; Parisel, O.; Flament, J. P.; Baraldi, I.; Momicchioli, F. J.
Phys. Chem. 1993, 97, 13358−13367.
(
17) (a) Wentrup, C.; Wilczek, K. Helv. Chim. Acta 1970, 53, 1459−
1
463. (b) Jones, W. M.; Joines, R. C.; Myers, J. A.; Mitsuhashi, T.;
Krajca, K. E.; Waali, E. E.; Davis, T. L.; Turner, A. B. J. Am. Chem. Soc.
973, 95, 826−835.
18) Banks, J. T.; Scaiano, J. C. J. Am. Chem. Soc. 1993, 115, 6409−
413.
19) Johnston, L. J.; Scaiano, J. C. Chem. Phys. Lett. 1985, 116, 109−
13.
20) (a) Matzinger, S.; Bally, T.; Patterson, E. V.; McMahon, R. J. J.
1
(
6
(
1
(
Am. Chem. Soc. 1996, 118, 1535. (b) Wong, M. W.; Wentrup, C. J. Org.
Chem. 1996, 61, 7022. (c) Schreiner, P. R.; Karney, W. L.; Schleyer, P.
v. R.; Borden, W. T.; Hamilton, T. P.; Schaefer, H. F., III J. Org. Chem.
1
996, 61, 7030.
(
(
21) Geise, C. M.; Hadad, C. M. J. Org. Chem. 2002, 67, 2532−2540.
22) Monguchi, K.; Itoh, T.; Hirai, K.; Tomioka, H. J. Am. Chem. Soc.
2
004, 126, 11900.
23) (a) Elsaesser, T.; Kaiser, W. Annu. Rev. Phys. Chem. 1991, 42, 83.
b) Sukowski, U.; Seilmeier, A.; Elsaesser, T.; Fisher, F. F. J. Chem.
Phys. 1990, 93, 4094. (c) Pigliucci, A.; Vauthey, E. Chimia 2003, 57,
00−203.
24) (a) Sension, R. J.; Repinec, S. T.; Hochsrasser, R. M. J. Chem.
(
(
2
(
Phys. 1990, 93, 9185−9188. (b) Burdzinski, G. T.; Middleton, C. T.;
Gustafson, T. L.; Platz, M. S. J. Am. Chem. Soc. 2006, 128, 14804−
1
4805.
(
25) Wentrup, C. Aust. J. Chem. 2013, DOI: 10.1071/CH13283.
(
26) Hoffmann, R. W.; Suzuki, K. Angew. Chem., Int. Ed. 2013, 52,
2
655−2656.
G
dx.doi.org/10.1021/jo401607m | J. Org. Chem. XXXX, XXX, XXX−XXX