Mass spectrometry of benzyne and cyclopentadienylideneketene

Article abstract of DOI:10.1071/CH09640

The formation of cyclopentadienylideneketene 2 and benzyne 1 in flash vacuum thermolysis reactions is investigated by on-line mass spectrometry. Compounds 13, 14, and 15 all afford ketene 2, which decomposes to benzyne and CO in the high-temperature regime. Cyclopentadienylideneketene 2 is stable on the microsecond time-scale of neutralization-reionization experiments. Collisional activation mass spectrometry of m/z 76 from 14, 15, and 5 indicates that the C₆H₄^·+ ions most likely undergo ring opening in the mass spectrometer.

Full text of DOI:10.1071/CH09640

RESEARCH FRONT
Full Paper
CSIRO PUBLISHING
Aust. J. Chem. 2010, 63, 1076–1083
www.publish.csiro.au/journals/ajc
Mass Spectrometry of Benzyne and
Cyclopentadienylideneketene
A
A
B
Thomas Monsandl, Graham Macfarlane, Robert Flammang,
and Curt Wentrup
A,C
ASchool of Chemistry and Molecular Biosciences, The University of Queensland,
Brisbane, Qld 4072, Australia.
B
Laboratoire de Chimie Organique, Centre de Spectrométrie de Masse,
Université de Mons-Hainaut, B-7000 Mons, Belgium.
Corresponding author. Email: wentrup@uq.edu.au
C
The formation of cyclopentadienylideneketene 2 and benzyne 1 in flash vacuum thermolysis reactions is investigated
by on-line mass spectrometry. Compounds 13, 14, and 15 all afford ketene 2, which decomposes to benzyne and CO in
the high-temperature regime. Cyclopentadienylideneketene 2 is stable on the microsecond time-scale of neutralization-
reionization experiments. Collisional activation mass spectrometry of m/z 76 from 14, 15, and 5 indicates that the C6H•
+
4
ions most likely undergo ring opening in the mass spectrometer.
Manuscript received: 9 December 2009.
Manuscript accepted: 20 April 2010.
Introduction
by photoelectron spectroscopy of benzyne produced by super-
sonic jet flash pyrolysis of benzocyclobutenedione 3 at 1300 C
◦
The two reactive intermediates o-benzyne 1 (in the following
called benzyne) and cylopentadienylideneketene 2 (Scheme 1)
are connected in an intricate and fascinating way, and we wish
in this paper to unravel a part of their relationship. Calculations
(Scheme 1); it was ascribed to the ionization of the π molecu-
lar orbital containing the out-of-plane portion of the CC triple
bond and termed IPπ.[ The ionization of the in-plane portion
of the triple bond appeared at 9.77 ± 0.03 eV. At the same time,
an ion of m/z 76 was observed in a time-of-flight (TOF) mass
spectrometer.[ When higher pyrolysis temperatures were used,
a peak at m/z 50, assumed to be due to butadiyne, appeared in
the MS.[ It has been demonstrated that benzyne decomposes
thermally to butadiyne and acetylene by using high temper-
ature supersonic jet flash pyrolyses of o-diiodobenzene and
benzocyclobutenedione 3 in conjunction with TOF photoion-
8]
indicate a strongly bent structure of ketene 2 as shown.[1] Apart
from IR spectroscopy described below, benzyne has been char-
8]
acterized by microwave and NMR spectroscopy.[2] The C1C2
bond length of 125.5 pm is between the values for ethyne and
ethene, being closer to the ethyne value.
8]
There have been several reports of the observation of ther-
mally or photochemically produced benzyne by mass spectro-
metry.Amassspectrumofbenzyne 1(ormostlikelyarearranged
[9]
•
+
ization mass spectrometry (Scheme 1). The barrier for this
C6H structure of m/z 76) was obtained by Fisher and Loss-
4
−1
◦
reaction is ∼55 kcal mol (1 cal = 4.184 J). A peak at m/z 104
ing by pyrolysis of o-diiodobenzene at 960 C in a reactor
(the mass corresponding to cyclopentadienylideneketene 2) was
installed in the inlet system of a mass spectrometer (residence
−3
[3]
ascribed to dissociative ionization of the benzocyclobutene-
time ∼10 s). Grützmacher and Lohmann generated C6H4
◦
[4]
dione, not to thermal generation of the ketene. The m/z 50 ion
by pyrolysis of bis(o-iodophenyl)mercury at 800 C. Gen-
zel and Osberghaus generated neutral C6H4 by bombardment
of benzene, fluorobenzene, and toluene with 200 eV electrons;
the neutrals were then investigated by electron-ionization mass
spectrometry. Interestingly, C6H2 was also formed.[ It is
reasonable to assume that C6H2 is hexatriyne, HC≡C-C≡C-
(
butadiyne• ) is also a fragment ion of C6H4 , as has been
+
•+
observed in the mass spectrum of benzyne generated from o-
dibromobenzene in an integrated femtosecond laser-molecular
[
10]
5]
beam system (Scheme 2).
Moini and Leroi found that the same C6H• ion (m/z 76) was
+
4
•
+
formed by photoionization of the three isomeric dibromoben-
zenes and concluded that substituent scrambling occurred in
the molecular cations before dissociation.[ In a study of the
C≡CH. As shown below, C6H is an important fragment in
2
the collisional activation and neutralization-reionization mass
11]
•+
spectra of C6H . The ionization energy of the C6H4 molecule
4
believed to be benzyne was evaluated as ∼9.5 eV in these three
[
3–5]
[6]
studies,
but corrected to ∼9 eV by Rosenstock et al. The
C
C
O
most reliable ionization energy was determined by Guo and
Grabowski, IE(benzyne) = 9.03 ± 0.1 eV, by using ion/molecule
reactions of the benzyne radical anion in a flowing afterglow
1
2
[
7]
apparatus. The same value, 9.03 ± 0.05 eV, was measured
Scheme 1.
10.1071/CH09640
©
CSIRO 2010
0004-9425/10/071076

RESEARCH FRONT
Mass Spectrometry of Benzyne and Cyclopentadienylideneketene
1077
O
O
1300°C
N2
ϩ
O
O
Ϫ2 CO
O
O
X
3
O
High T
1
800°C
hn
O
I
hn
hn
or hn
X
Hg
hn
O
ϩ
X2 ϭ I , Br , or F Mg
I
2
2
2
10
1
Scheme 2.
hn
hn
O
C
O
hn
Ϫ
COO
hn
Δ
hnЈ
C
O
O
ϩ
3
9
N2
4
1
Δ
7
8
FVT
FVT
Scheme 4.
FVT
O
O
O
O
hn
3
O
3
O
O
O
08 nm
5
6
10
O
CO
hn
308 nm
3
2
08 or 248 nm
O
O
5
C
C
48 or 282 nm
O
O
ϪCO
38 K
2
1
Scheme 3.
7
distonic o-, m-, and p-C6H• ions generated by photodissoci-
ation of the corresponding Mg -difluorobenzene complexes it
+
Scheme 5.
4
+
was concluded that these ions preserve their individuality, at
Early work by Chapman and coworkers suggested that
•+
least partially. The C6H ions dissociate further into C2H2 and
−1
4
the benzyne CC triple bond absorbed at 2085 cm in an Ar
•
+
C4H when sufficient energy is available (≤250 nm photon
[18]
2
matrix at 8 K.
In this work, benzyne was generated by
wavelength).The whole ionization-dissociation process requires
photolysis of several precursors (Scheme 4), including benzo-
cyclobutenedione 3, which was found to exist in a photochemical
equilibrium with the cyclohexadienebisketene (5,6-dicarbonyl-
only one photon. The C4H• ions were most prominent in the
+
2
ortho series and were thought to result from metastable decay
•
+ [12]
of C6H
. However, threshold photoelectron photoion coin-
−1
4
1,3-cyclohexadiene) 9 (2091 cm ). An intermediate absorbing
•+
−
1
cidence experiments indicated that the o-, m-, and p-C6H4F2
at 1838 cm was tentatively identified as benzocyclopropenone
radical cations undergo ring opening before fragmentation (not
10 (Scheme 4).
•
+
[13]
13
15
The assignment of the 2085 cm 1 band to benzyne was
refuted by a photodetachment experiment, in which removal of
an electron from the benzyne radical anion generated a photo-
electron spectrum featuring a vibrational progression, from
−
yielding C6H ions).
Labelling with C and N demon-
4
strated that isomerization and possibly ring opening occurred
before elimination of HCN from the molecular ion of benzo-
nitrile (forming C6H• ), and the collisional activation mass
+
4
−
1
spectra (CAMS) of benzonitrile and 2-ethynylpyridine were
which a stretching vibration of 1860 ± 15 cm was determined
[
14]
•+
[19]
identical.
Further work on photodissociation of the C6H4
for benzyne 1 in the gas phase.
ions led to the conclusion that at least a major fraction of these
A new matrix IR spectroscopic investigation of the photolysis
ions, as well as the C6H• ions derived from o-diiodobenzene,
+
of phthalic anhydride 5 revealed that the benzyne CC stretch is
4
[14b]
−1 −1
in fact at 1846 cm (Ne matrix), and the 2085 cm peak does
have a ring-opened structure.
[20]
Berry and coworkers reported the mass spectrum of a
species (m/z 76) formed within 50 microseconds of flash photo-
lysis of the classical benzyne precursor, benzenediazonium
not belong to benzyne.
Instead, it was assigned to ketene 2,
and an investigation of the kinetics of the matrix photolyses
indicated that 2 is reversibly formed from benzyne + CO; it is not
an intermediate in the photochemical reaction 5 → 1.A separate
photochemical process converted 5 to benzocyclopropenone 10
2
-carboxylate 4 (Scheme 3). Fragment ions at m/z 50, 39, 38,
and others appeared during the first 200 microseconds. The
signal at m/z 76 (ascribed to benzyne, but probably due to a ring-
opened ion) disappeared on a 250–300 microsecond timescale,
with concomitant appearance of m/z 152 (biphenylene 7).[15]
Biphenylene 7 was found to be formed in a bimolecular reaction
−
1
(1852 cm ; Ne matrix). Benzyne 1 was found to dimerize to
biphenylene 7 at 38 K (Scheme 5).
Similar conclusions were derived from IR studies of the Ar
matrix photolyses of benzocyclobutenedione 3, where ketene 2
9
−1 −1 [15b]
−1 [21]
Here, two photochemical
witharateconstantof ∼10 L mol
s
.
Biphenylene7 and
was observed at 2087 cm
.
triphenylene 8 are the dimerization and trimerization products
routes from benzocyclobutenedione 3 were proposed: (i) via
−
1
of benzyne generated by gas phase pyrolysis of phthalic anhy-
bisketene 9 (now claimed to absorb at 2125 and 2064 cm ) and
[
16]
[17]
−1
dride 5
or indanetrione 6.
Indanetrione also extrudes one
benzocyclopropenone 10 (now absorbing at 1857 cm and later
17]
−1 [22]
molecule of CO to yield benzocyclobutenedione (Scheme 3).
changed again to 1869 cm
), and (ii) via the unobserved

RESEARCH FRONT
1
078
T. Monsandl et al.
O
O
O
O
O
O
hn Ͼ280 nm
hn
Ͼ300 nm
O
CF3
O
O
hn
3
O
hn
O
O
C
C
C
FVT
FVT
FVT
Δ
1
20–140 K
13
14
C
1
1
9
C
O
hn
hn
hn
hn 180–300 nm
2
O
O
O
O
O
FVT
FVT
O
O
ϩ
O
ϩ
C
C
O
O
10
1
2
hn
CO
hn
hn
ϪCO
15
hn
O
O
1
O
O
12
16
Scheme 6.
Scheme 7.
carbenoid ketene 11. Cyclopentadienylidenecarbene 12 was
proposed as an intermediate between 1 and 2 (Scheme 6).[
In a third matrix-isolation IR study, the photolysis of
benzocyclobutenedione 3 in Ar matrix at >280 nm afforded
23]
0.045
0.040
246
32
0.035
1
−
1
the signals of the ketene 2 (2085 cm ) and benzyne 1 (1054,
0
0
.030
.025
−
1
1
038, 848, 736, and 469 cm ) as well as the ring-opening
104
76
−
1
product cyclohexadiene-1,2-bisketene 9 (2138 and 2077 cm
;
[
24]
0.020
0.015
0.010
Scheme 6).
pound at 2091
confirmed. Benzyne was formed by photolysis of the bisketene
at 280–300 nm. The structure of the bisketene was ascertained
by the observation that it reverted to 3 at λ > 320 nm, and also
The previously stated absorptions of this com-
[
18]
[21]
−1
or 2125 and 2064 cm could not be
1
52
9
218
0.005
[24]
0
2
on heating to 120–140 K (Scheme 6).
00
300
400
500
600
700
800
−
1
The assignment of the 2085 cm peak to ketene 2 in the
matrix photolyses described above followed the earlier identifi-
cation of this ketene in flash vacuum thermolysis (FVT)/matrix
Temperature [°C]
Fig. 1. Relative ion intensities as a function of temperature in the FVT-
MS of 13. Ordinate: ratio of ion current versus total ion current for m/z
−
1
isolation experiments: A weak peak at 2085 ± 5 cm in the Ar
+
7
6 (C6H• ), 104 (cyclopentadienylideneketene 2), 132 (ketene 17), 152
matrix of the product resulting from pyrolysis of phthalic anhy-
4
◦
−5
(biphenylene 7), 218, and 246 (starting material 13). M/z 218 is a fragment
ion of 13• ([M-C2H4] ); it is not formed as a neutral on FVT.
dride 5 at 900–920 C (10 hPa) was assigned to cyclopenta-
+
•+
[
25]
dienylideneketene 2.
The yield of the ketene is very low
under these conditions, as most of the starting material remains
unchanged even at this temperature because of the high vacuum
and short contact times. The other observable products in the Ar
matrix are CO, CO2, and benzyne 1. Ketene 2 was also gener-
ated by FVT of the Meldrum’s acid derivatives 14 and 15, which
◦
4
00 C, when the FVT of 13 is essentially complete. The struc-
ture 17 is assigned on the basis of the exact mass measurement
and the concordance with the IR spectroscopic observations
Scheme 8). Presumably, compound 17 has a bent structure
like other methyleneketenes. The alternative elimination of
(
−
1
both afforded a strong peak at 2085 ± 5 cm ascribed to ketene
[1]
[
25]
2
(Scheme 7).
Detailed FVT-IR studies with precursors 13–16 confirmed
C2H4 from 13 is not a thermally important process, but m/z 218
[M-C2H4]• ) is a fragment ion of 13 (Fig. 1).
+
•+
(
−
1
[26,27]
the formation of ketene 2 (2089 cm ) (Scheme 7).
Mass
The intermediate ketene 17 disappeared on further heating
above 400 C, as the new ketene 2 was formed and reached a
maximum intensity at 600 C (Fig. 1; Scheme 8). The struc-
ture 2 is assigned to this compound (m/z 104) due to the exact
spectrometric confirmation of this assignment will be given
below.
◦
◦
In this paper we wish to report on-line FVT-MS studies
of precursors 13–15, which allow a correlation between the
mass measurement, the agreement with the IR spectroscopic
−
1
thermolysis product absorbing at 2085 ± 5 cm in the IR and
the product with m/z 104 in the MS, both being ascribed
to ketene 2. Furthermore, benzyne generated by FVT of 14,
observations,[
25,26]
and the trapping with methanol.
[26b]
The intensity curve for m/z 76 (C6H4 ) follows that of 2•
•+
+
◦
closely until ∼700 C; under these conditions m/z 76 is wholly or
1
5, and 5 is investigated by collisional activation (CA) and
•+
◦
in part due to fragmentation of 2 (Fig. 1).Above 700 C there is
a thermal source of benzyne; the m/z 76 peak is now too intense
neutralization-reionization (NR) mass spectrometry.
to be wholly due to fragmentation of 2• . At the same time, the
yield of biphenylene 7 (m/z 152) increases due to dimerization
of thermally produced benzyne (Fig. 1). These observations are
in full agreement with the IR-spectroscopic investigations.[
The exact masses of 1, 2, 17, and 7 were measured.
+
Results and Discussion
Norbornene Derivative 13
25,26]
FVT of 13 was monitored by on-line mass spectrometry (Fig. 1).
A reaction product with m/z 132 reaches a maximum intensity at

RESEARCH FRONT
Mass Spectrometry of Benzyne and Cyclopentadienylideneketene
1079
O
O
0.007
0.006
0.005
O
CF3
104
O
C
C
76
Δ
1
3
ϪCF CO H
3
2
0
.004
152
170
Δ
ϪC H4
0.003
0.002
Δ
17
2
ϪCO
2
2
O
O
ϪMe CO
O
O
O
O
O
O
Δ
Δ
C
0.001
ϪC H4
ϪCO2
ϪMe CO
C
O
2
2
2
0
200
14
18
Δ
Δ, IR
Δ
300
400
500
600
700
800
ϪCO
Temperature [°C]
ϪC H6
5
MS
O
O
O
C
Fig. 3. Relative ion intensities as a function of temperature in the FVT-
MS of 15. Ordinate: ratio of ion current versus total ion current for m/z
76 (C H ), 104 (cyclopentadienylideneketene 2), 152 (biphenylene 7), and
Δ
C
O
ϪCO2
ϪMe CO
•+
O
1
6
4
2
170 (ketene 19).
15
19
7
identical with the collisional activation (CA) spectrum. The CA
•+
spectrum shows a strong fragmentation to m/z 76 ([M – CO] );
the NRMS shows enhanced fragmentation to m/z 74 and 50 in
addition to m/z 76 (Figs 4, 5, S1, and S2). It is well recognized
that energy deposition in NR is higher than in CA; the slow,
metastable, processes are therefore of lower amplitude in the NR
spectra.[ Another small but significant difference between the
CA and NR mass spectra is the non-observation of the charge
stripping peak (M/2 = m/z 52) in the NR spectrum, most prob-
Scheme 8.
0
.040
.035
.030
.025
.020
.015
.010
.005
.000
2
34
0
0
0
0
0
0
0
0
29]
132
104
2+
ably due to a lower probability of the M → M process as
compared with the M• → M process. The observation of
a peak at m/z 28 in the NR spectrum is ascribed to reioniza-
tion of CO lost in the metastable decomposition of the m/z 104
ion. Even less energy is deposited in a mass analyzed ion kinetic
energy (MIKE) mass spectrum, which therefore exhibits only the
thermodynamically most favourable fragmentations. The MIKE
mass spectrum of m/z 104 features m/z 76 as the only fragment
ion (not shown).
+
2+
[30]
7
6
1
2
52
06
200
300
400
500
600
700
800
Temperature [°C]
The CA and NR mass spectra of m/z 76 are very similar with
fragment ions at m/z 75, 74, 61, 50, 49, 38, 37, 36, 26, and
Fig. 2. Relative ion intensities as a function of temperature in the FVT-
MS of 14. Ordinate: ratio of ion current versus total ion current for m/z
25. Examples of such CA and NR spectra are shown in Figs 6
+
7
(
6 (C6H• ), 104 (cyclopentadienylideneketene 2), 132 (ketene 17), 152
and 7. Several more CA and NR mass spectra are shown in the
Accessory Publication. The CA spectra show a much stronger
m/z 50 fragment ion compared with the NR spectrum. This can
be rationalized in the same way as described for m/z 104 above.
The same CAMS and NRMS of m/z 76 are obtained from the
precursors 14, 15, and 5, with and without pyrolysis. These CA
spectra are identical within experimental error with the CAMS
of the m/z 76 ions obtained by loss of HCN from benzonitrile and
2-ethynylpyridine molecular ions, which most likely are due to
4
biphenylene 7), 206 (18), and 234 (starting material 14).
Norbornene-Meldrum’s Acid 14
FVT of 14 gives rise to two intermediates, at m/z 132 (17) and
m/z 206 (18), the former being the most intense (Scheme 8;
Fig. 2). Both species rise immediately on thermolysis at 250–
00 C, then fall off rapidly on further thermolysis to give m/z
04 (ketene 2), which reaches a maximum intensity at 600 C.
Again, the yield of the m/z 76 ion follows that of 2 closely, and
is probably due largely to fragmentation of 2 (see below). The
yield of biphenylene 7 increases above 600 C.The identification
ring-opened C6H• ions.
+
[14]
The CAMS and NRMS show very
◦
3
1
4
◦
•+
pronounced peaks at m/z 74 (C6H2 ) as seen, for example, in
Fig. 6. Strong C6H2 signals were also observed by Genzel and
Osberghaus in their study of the bombardment of benzene with
•+
◦
[5]
200 eV electrons. A ring-opened C6H4 cation or neutral may
of ketene 2 is based on the exact mass measurement and the
stabilize itself by loss of 2H to form the hexatriene (HC≡C-
agreement with the IR-spectroscopic investigation.[
25,26]
C≡C-C≡CH) molecular ion or neutral.
The m/z 104 species generated by FVT of 14 at 600–
The m/z 50 ion has a strong recovery signal in the NRMS.
The main product ions correspond to loss of H, CH, and C2H in
agreement with the diacetylene structure, HC≡C-C≡CH. The
same CA and NR mass spectra of m/z 50 are obtained from
precursors 14, 15, and 5 (Figs S5 and S6).
◦
8
(
00 C survives a neutralization-reionization mass spectrometry
[28]
NRMS) experiment,
which indicates that the corresponding
neutral (MW 104) is at least stable on the microsecond timescale
of this experiment. The NR mass spectrum is similar to, but not

RESEARCH FRONT
1
080
T. Monsandl et al.
100
ϫ3.00
76
9
9
8
8
7
7
6
6
5
5
4
4
3
3
2
2
1
1
5
0
5
0
5
0
5
0
5
0
5
0
5
0
5
0
5
0
5
0
5
0
74
37
62
60
4
8
7
2
10
15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
m/z
◦
Fig. 4. CAMS of m/z 104 from FVT of 14 (essentially unchanged 200–900 C). The same CAMS is obtained in the FVT of
15 (ordinate: relative % intensity). Very weak peaks at m/z 25–26 are barely visible on this scale.
74
100
50
9
9
8
8
7
7
6
6
5
5
4
4
3
3
2
2
1
1
5
0
5
0
5
0
5
0
5
0
5
0
5
0
5
0
5
0
5
0
37
6
1
4
6
8
72
3
5
28
59
104
26
24
85
83
4
10
15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
m/z
◦
Fig. 5. NRMS of m/z 104 from FVT of 14 (600–700 C). The same NRMS is obtained in the FVT of 15 (ordinate: relative %
intensity).
Cage Compound 15
◦
◦
at 350 C, to be replaced by ketene 2 (m/z 104) at 580 C
(Fig. 3). Again, ketene 19 is assumed to have a bent struc-
Compound 15 features only a weak molecular ion in the MS
ture like other methyleneketenes.[ The strong m/z 76 (C6H4
peak also reached a maximum at 580 C, but it remained above
700 C after the ketene has disappeared, thus indicating that ben-
zyne is a thermolysis product at this temperature. The yield of
1]
•+
)
(
m/z 272). The IR-spectroscopic investigation indicated that
◦
[25,26]
◦
thermolysis was complete at 500 C.
not observed as a thermal product (m/z 206 absent), but the
intermediate ketene 19 (m/z 170) reached a maximum intensity
Compound 18 was
◦

RESEARCH FRONT
Mass Spectrometry of Benzyne and Cyclopentadienylideneketene
1081
100
ϫ8.00
50
ϫ8.00
9
9
8
8
7
7
6
6
5
5
4
4
3
3
2
2
1
1
5
0
5
0
5
0
5
0
5
0
5
0
5
0
5
0
5
0
5
0
3
7
7
4
3
8
6
1
36
6
0
7
3
4
9
4
8 51
10
15
20
25
30
35
40
m/z
45
50
55
60
65
70
Fig. 6. CAMS of m/z 76 from the FVT of 14. The same spectrum is obtained from 15 and is constant over the temperature
◦
◦
range 600–700 C. The same spectrum is also obtained by dissociative ionization of 5 (200 C).
74
100
9
9
8
8
7
7
6
6
5
5
4
4
3
3
2
2
1
1
5
0
5
0
5
0
5
0
5
0
5
0
5
0
5
0
5
0
5
0
5
0
3
6
7
73
49
3
76
5
2
6
7
48
6
0
1
25
6
10
15
20
25
30
35
40
45
m/z
50
55
60
65
70
75
◦
◦
◦
Fig. 7. NRMS of m/z 76 from FVT of 14 at 600 C. The same NRMS is obtained from 15 (600–800 C) and 5 (200 C)
(
ordinate: relative % intensity).
◦
Phthalic Anhydride 5
biphenylene 7 also increased above 600 C (Fig. 3). The exact
masses of 1, 2, and 7 were re-confirmed.
Preparative FVT of 5 afforded biphenylene as the only major
product, triphenylene in ∼160 times smaller amount than
biphenylene, and benzene (absolute yield 4.5 ± 1%), as deter-
•+
•+
The CA and NR mass spectra of m/z 104 (2 ), 76 (C6H4 ),
and 50 (C4H ) were identical with those obtained from 14
•
+
2
•
+
•+
4
(
Figs 4–7 and S1–S12).
mined by GCMS. The ions at m/z 104 (C H O ), 76 (C H ),
7 4 6

RESEARCH FRONT
1
082
T. Monsandl et al.
and 50 (C4H• ) are ionization-dissociation peaks in the mass
+
FVT-MS measurements at low and high resolution. There was
a ∼3 cm path in the high vacuum of the spectrometer from the
exit of the FVT tube (10 cm × 0.8 cm i.d.) to the ion source. A
six-sector VG-Micromass AutoSpec 6F mass spectrometer with
EBEEBE geometry was used for CAMS (identical spectra were
obtained using either O2 or He as collision gas) and NRMS
(using NH3/O2 as collision gases) measurements.[ A colli-
sion cell was inserted between the third and the fourth sectors. A
FVT oven was fitted inside the source housing in immediate con-
2
spectrum of 5. Their compositions were re-confirmed by high-
resolution mass measurements.
The m/z 104 fragment ion in the mass spectrum of 5 (without
pyrolysis) does not survive NRMS (Figs S7 and S8, Accessory
Publication). In contrast, the m/z 104 ion from ketene 2 generated
by FVT of precursors 14 and 15 does survive NRMS as described
above (Fig. 5). Therefore, the ionization-dissociation ion at m/z
04 in the mass spectrum of 5 does not have the structure of
ketene 2. It may have the open carbenoid ketene structure 11
Scheme 5) or a different structure. It must be noted that the
32]
1
[33]
tact with the ion source as previously described.
Compounds
13, 14, and 15[
26b]
were prepared according to the literature.
(
yield of thermolysis of phthalic anhydride to 1 and/or 2 is far
The following exact masses were measured: 1 Calc. for C6H4:
76.03132. Found: 76.03130. 2 Calc. for C7H4O: 104.02622.
Found: 104.02620. 7 Calc. for C12H8: 152.06264. Found:
152.0626. 14 Calc. for C13H14O4: 234.08922. Found: 234.0893.
17 Calc. for C9H8O: 132.05754. Found: 132.0578. 18 Calc. for
C11H10O4: 206.05790. Found: 206.0579.
too low under our experimental conditions, even at the highest
◦
accessible temperatures (1000 C), to permit the direct detection
of the mass spectra of these neutral compounds.The m/z 104 and
7
6 ions observed are due to dissociative ionization of phthalic
anhydride.
The CA and NR mass spectra of m/z 76 (C6H• ) formed by
+
4
dissociative ionization of 5 were identical with those described
above and arising from thermal generation of benzyne from 14
and 15 (Figs 6, 7, and S9–S12).
Accessory Publication
Collision-activation and neutralization-reionization mass spec-
tra of m/z 104, 76, and 50 from precursors 14, 15, and 5. This
material is available free of charge from the Journal’s website.
The m/z 50 ion (C4H• ) has a strong recovery signal in the
+
2
NRMS. The CA and NR mass spectra of m/z 50 are identical
for precursors 14, 15, and 5 with and without pyrolysis and
are in agreement with the butadiyne structure (Figs S5 and S6,
Accessory Publication).
Acknowledgements
This work was supported by the Australian Research Council and the
Deutsche Forschungsgemeinschaft (Postdoctoral Fellowship for T.M.). The
mass spectrometers were purchased with funds provided by The Vice-
Chancellor ofThe University of Queensland (Kratos) and the Fonds National
de la Recherche Scientifique, Brussels (Micromass).
Conclusion
The FVT-MS results corroborate the FVT-IR results[25,26] and
allow the identification of the m/z 104 species as cyclopenta-
dienylideneketene 2. Likewise, the intermediate ketenes 17 and
References
[
1] (a) For calculations on cyclopentadienylideneketene 2 see: J. G.
Radziszewski, P. Kaszynski, A. Friedrichsen, J. Abildgaard, Collect.
Czech. Chem. Commun. 1998, 63, 1094. doi:10.1135/CCCC19981094
1
9 are identified by their mass spectra together with the previ-
[26b]
ouslyreportedIRspectraandtrappingwithMeOH.
Benzyne
1
is generated in the high-temperature FVT of 13–15 and 5.
(
b) A. C. Scheiner, H. F. Schaefer, J. Am. Chem. Soc. 1992, 114, 4758.
[25,26]
The positive identification of benzyne by infrared
and
doi:10.1021/JA00038A045
(c) A. P. Scott, L. Radom, Chem. Phys. Lett. 1992, 200, 15.
doi:10.1016/0009-2614(92)87038-Q
[
31]
photoelectron spectroscopy under similar reaction conditions
makes it highly likely that the neutral C6H4 species gener-
ated in the high-temperature FVT-MS experiments possesses
the benzyne structure initially. It cannot be guaranteed that the
[2] (a) For microwave spectra of benzyne see: P. D. Godfrey,Aust. J. Chem.
2010, 63, 1061. doi:10.1071/CH10152
•
+
b) 13C dipolar NMR spectrum of benzyne-1,2-13C2: A. M. Orendt,
(
C6H radicalcationsretainthebenzynestructure.Althoughevi-
4
J. C. Facelli, J. G. Radziszewski, W. J. Horton, D. M. Grant, J. Michl,
dence has been published in favour of retention of the benzyne
[
12]
J. Am. Chem. Soc. 1996, 118, 846. doi:10.1021/JA953417R
structures,
there is strong evidence that extensive ring open-
1
13
(
c) H and C NMR spectra: R. Warmuth, Angew. Chem. Int. Ed.
Engl. 1997, 36, 1347. doi:10.1002/ANIE.199713471
d) H. Jiao, P. von R. Schleyer, B. R. Beno, K. N. Houk,
ing takes place before fragmentation of the molecular radical
[
11,13,14]
cations of dihalobenzenes and benzonitrile.
The col-
(
lisional activation mass spectra of the C6H• radical cations
+
4
R. Warmuth, Angew. Chem. Int. Ed. Engl. 1997, 36, 2761.
doi:10.1002/ANIE.199727611
(e) R. Warmuth, J. Yoon, Acc. Chem. Res. 2001, 34, 95. doi:10.1021/
AR980082K
generated from 5, 14, and 15 are identical (see Fig. 6 and the
Accessory Publication), and they are also identical within exper-
imental error with the CAMS of m/z 76 obtained from the
[14a]
benzonitrile and 2-ethynylpyridine molecular ions,
which
[3] I. P. Fisher, F. P. Lossing, J. Am. Chem. Soc. 1963, 85, 1018.
doi:10.1021/JA00890A054
most likely are due to ring-opened C6H• ions.
+
[14]
4
[
4] H.-F. Grützmacher, J. Lohmann, Liebigs Ann. Chem. 1967, 705, 81.
doi:10.1002/JLAC.19677050110
The neutral C6H4 species and cyclopentadienylideneketene 2
are stable on the microsecond timescale of the NRMS experi-
ments. However, the m/z 104 fragment ion in the mass spectrum
of phthalic anhydride 5 does not survive the NRMS experiment
and does not possess the structure of ketene 2. The CAMS and
[
[
5] H. Genzel, O. Osberghaus, Z. Naturforsch. A 1967, 22a, 331.
6] H. M. Rosenstock, R. Stockbauer, A. C. Parr, J. Chim. Phys. 1980,
77, 745.
[
7] Y. Guo, J. J. Grabowski, J. Am. Chem. Soc. 1991, 113, 5923.
NRMS of the m/z 50 fragment ion in the mass spectrum of C6H•
+
4
doi:10.1021/JA00016A001
indicate the butadiyne structure of this ion.
[8] X. Zhang, P. Chen, J. Am. Chem. Soc. 1992, 114, 3147. doi:10.1021/
JA00034A076
[
9] X. Zhang, A. T. Maccarone, M. R. Nimlos, S. Kato, V. M. Bierbaum,
G. B. Ellison, B. Ruscic,A. C. Simmonett, W. D.Allen, H. F. Schaefer,
J. Chem. Phys. 2007, 126, 44312. doi:10.1063/1.2409927
Experimental
ApparatusandproceduresforFVT, matrixisolation, andIRspec-
troscopy were as previously described.[
25,32,33]
A similar FVT
[10] E. W.-G. Diau, J. Casanova, J. D. Roberts, A. H. Zewail, Proc. Natl.
Acad. Sci. USA 2000, 97, 1376. doi:10.1073/PNAS.030524797
oven was fitted to a Kratos MS25RFA mass spectrometer for

RESEARCH FRONT
Mass Spectrometry of Benzyne and Cyclopentadienylideneketene
1083
[
[
[
11] M. Moini, G. E. Leroi, J. Phys. Chem. 1986, 90, 4002. doi:10.1021/
[23] (a) For calculations on cyclopentadienylidenecarbene 12 lying
−
1
−1
J100408A035
30 ± 4 kcal mol (126 ± 16 kJ mol ) above singlet benzyne in a
−
1
12] H.-C. Liu, C.-S. Wang, W. Guo, Y.-D. Wu, S. Yang, J. Am. Chem. Soc.
shallow minimum with a barrier of 0–5 kcal mol for rearrangement
to benzyne, see (a) ref. 10.
2002, 124, 3794. doi:10.1021/JA0122546
13] A.-M. Boulanger, D. M. P. Holland, D. A. Shaw, P. M. Mayer,
(b) N. J. Burton, G. E. Quelch, M. M. Gallo, H. F. Schaefer, J. Am.
Chem. Soc. 1991, 113, 764. doi:10.1021/JA00003A006
(c) J. Cioslowski, P. Piskorz, D. Moncrieff, J. Am. Chem. Soc. 1998,
120, 1695. doi:10.1021/JA9725662
J.Am. Soc. Mass Spectrom. 2009, 20, 20. doi:10.1016/J.JASMS.2008.
09.020
[
14] (a) T. A. Molenaar-Langeveld, R. H. Fokkens, N. N. M. Nibbering,
Org. Mass Spectrom. 1986, 21, 15. doi:10.1002/OMS.1210210105
[24] T. Mosandl, C. Wentrup, J. Org. Chem. 1993, 58, 747. doi:10.1021/
JO00055A034
(
b) W. J. van der Hart, E. Oosterveld, T. A. Molenaar-Langeveld,
N. M. M. Nibbering, Org. Mass Spectrom. 1989, 24, 59. doi:10.1002/
OMS.1210240112
[25] C. Wentrup, R. Blanch, H. Briehl, G. Gross, J. Am. Chem. Soc. 1988,
110, 1874. doi:10.1021/JA00214A034
[
15] (a) R. S. Berry, J. Clardy, M. E. Schaefer, J. Am. Chem. Soc. 1964, 86,
[26] (a) R. F. C. Brown, N. R. Browne, K. J. Coulston, L. B. Danen, F. W.
Eastwood, M. J. Irvine, A. D. E. Pullin, Tetrahedron Lett. 1986, 27,
1075. doi:10.1016/S0040-4039(86)80052-1
2738. doi:10.1021/JA01067A057
(
b) M. E. Schaefer, R. S. Berry, J. Am. Chem. Soc. 1965, 87, 4497.
doi:10.1021/JA00948A017
16] (a) M. P. Cava, M. J. Mitchell, D. C. DeJongh, R.Y. Van Fossen, Tetra-
hedron Lett. 1966, 2947. doi:10.1016/S0040-4039(01)99893-4
(b) R. F. C. Brown, N. R. Browne, K. J. Coulston, F. W. Eastwood,
M. J. Irvine, A. D. E. Pullin, U. E. Wiersum, Aust. J. Chem. 1989, 42,
1321.
[
(
b) E. K. Fields, S. Meyerson, Tetrahedron Lett. 1967, 8, 571.
[27] R. F. C. Brown, K. J. Coulston, F. W. Eastwood, Aust. J. Chem. 1994,
47, 47.
doi:10.1016/S0040-4039(00)90550-1
[
[
17] R. F. C. Brown, R. K. Solly, Aust. J. Chem. 1966, 19, 1045.
18] (a) O. L. Chapman, C.-C. Chang, J. Kolc, R. N. Rosenquist,
H. Tomiola, J. Am. Chem. Soc. 1975, 97, 6586. doi:10.1021/
JA00855A055
[28] F. Ture cˇ ek, Top. Curr. Chem. 2007, 225, 77. doi:10.1007/3-540-
36113-8_3
[29] (a) P. Gerbaux, R. Flammang, C. Th. Pedersen, M. W. Wong, J. Phys.
Chem. A 1999, 103, 3666. doi:10.1021/JP984121L
(b) D. Harnish, J. L. Holmes, Org. Mass Spectrom. 1994, 29, 213.
doi:10.1002/OMS.1210290502
[30] (a) F. A. Wiedmann, J. Cai, C. Wesdemiotis, Rapid Commun. Mass
Spectrom. 1994, 8, 804. doi:10.1002/RCM.1290081003
(b) P. Gerbaux, R. Flammang, E. H. Mørkved, M. W. Wong,
C. Wentrup, J. Phys. Chem. A 1998, 102, 9021. doi:10.1021/
JP9824454
[31] A. Chrostowska, G. Pfister-Guillouzo, F. Gracian, C. Wentrup, Aust.
J. Chem. 2010, 63.
[32] R. H. Bateman, R. H. Brown, M. Lefevere, R. Flammang, Y. Van
Haverbeke, Int. J. Mass Spectrom. Ion Process. 1992, 115, 205.
doi:10.1016/0168-1176(92)85042-X
(
b) O. L. Chapman, K. Mattes, C. L. McIntosh, J. Pacansky, G. V.
Calder, G. Orr, J. Am. Chem. Soc. 1973, 95, 6134. doi:10.1021/
JA00799A060
[
[
[
[
19] D. G. Leopold, A. E. S. Miller, W. C. Lineberger, J. Am. Chem. Soc.
1
986, 108, 1379. doi:10.1021/JA00267A003
20] J. G. Radziszewski, B.A. Hess, R. Zahradnik, J.Am. Chem. Soc. 1992,
14, 52. doi:10.1021/JA00027A007
21] J. G. G. Simon, N. Münzel, A. Schweig, Chem. Phys. Lett. 1990, 170,
87. doi:10.1016/0009-2614(90)87113-6
1
1
22] (a) For the matrix IR and UV spectra of benzocyclopropenone, see:
J. G. G. Simon, A. Schweig, Y. Xie, H. F. Schaefer, Chem. Phys. Lett.
1992, 200, 631. doi:10.1016/0009-2614(92)80101-G
(
b) For the 1H and 13C NMR spectra of benzocyclopropenone, see:
[33] J. Brown, J. R. Flammang, Y. Govaert, M. Plisnier, C. Wentrup,
Y. Van Haverbeke, Rapid Commun. Mass Spectrom. 1992, 6, 249.
doi:10.1002/RCM.1290060405
J. G. G. Simon, A. Schweig, Chem. Phys. Lett. 1993, 201, 377.
doi:10.1016/0009-2614(93)85086-4