Use of external radical sources in flash vacuum pyrolysis to facilitate cyclodehydrogenation reactions in polycyclic aromatic hydrocarbons

Article abstract of DOI:10.1071/CH14289

A new process to facilitate the cyclodehydrogenation of polycyclic aromatic hydrocarbons (PAHs) in flash vacuum pyrolysis (FVP) using an external radical source is described. Using hexanes as an external radical source the conversion of various PAHs to their cyclodehydrogenated products is vastly increased. Various other volatile organic compounds were also examined to determine their ability to act as external radical sources in FVP.

Full text of DOI:10.1071/CH14289

RESEARCH FRONT
CSIRO PUBLISHING
Aust. J. Chem. 2014, 67, 1338–1343
Full Paper
http://dx.doi.org/10.1071/CH14289
Use of External Radical Sources in Flash Vacuum
Pyrolysis to Facilitate Cyclodehydrogenation Reactions
in Polycyclic Aromatic Hydrocarbons
A B
,
A
Aaron W. Amick
and Sara E. Martin
A
Department of Chemistry, Washington College, 300 Washington Ave.,
Chestertown, MD 21620, USA.
B
Corresponding author. Email: aamick2@washcoll.edu
A new process to facilitate the cyclodehydrogenation of polycyclic aromatic hydrocarbons (PAHs) in flash vacuum
pyrolysis (FVP) using an external radical source is described. Using hexanes as an external radical source the conversion
of various PAHs to their cyclodehydrogenated products is vastly increased. Various other volatile organic compounds
were also examined to determine their ability to act as external radical sources in FVP.
Manuscript received: 11 May 2014.
Manuscript accepted: 12 May 2014.
Published online: 5 June 2014.
Introduction
mechanisms that produces PAHs in fuel-rich flames showed
that ꢀOH radical is a key reactive species in the construction of
these complex structures.[38] Their studies found that fullerene
growth begins with the abstraction of a hydrogen from a PAH by
ꢀOH, followed by acetylene addition; this process continues for
many cycles to yield C60 and other higher fullerenes.
Flash vacuum pyrolysis (FVP)[1–16] has become an integral part
of the synthesis of both flat and curved polycyclic aromatic
hydrocarbons (PAHs), most notably in the synthesis of
circumtrindene (1),[17,18] Buckminsterfullerene (C60),[19,20] and
the recent chemical synthesis of a [5,5] carbon nanotube end-
cap.[21] The current strategy to create PAHs by FVP relies on
using pyrolytic precursors that will lose one or more halogens at
elevated temperatures (500–13008C) by the rupture of Ar–X
bonds to create aryl radicals. These aryl radicals can then attack
an adjacent aromatic ring forming a carbon–carbon bond through
an overall ‘cyclodehydrohalogenation’ (see highlighted bonds
in Scheme 1).
The impetus for our study was to examine the feasibility of
adding an external radical source to the carrier gas of FVP to
mimic the conditions present in fuel-rich flames. We avoided
the use of ꢀOH radicals as their possible attachment could
ꢀ
ꢀ
ꢀ
lead to the oxidation of the PAH, while Cl, alkyl, and OR
radicals will facilitate the cyclodehydrogenation of the PAH
without this unwanted oxidation pathway.
This strategy has been very successful; however, the synthe-
sis of these halogenated pyrolytic precursors can be arduous, as
polyhalogenated PAHs are notoriously insoluble and difficult to
purify. The ideal situation would be to use the non-halogenated
parent PAH that is easier to create and in some cases is
commercially available; however, with no carbon–halogen
bonds to easily rupture to produce radicals, the FVP of these
parent molecules are doomed to extremely low yields of the
desired product by simple brute force ‘cyclodehydrogena-
tion.[15] The issues outlined above are illustrated in the two
syntheses of circumtrindene (1) by Scott and co-workers. In
their first reported synthesis of 1, Scott and co-workers pyrolyze
commercially available decacyclene (2) to produce 1 in 0.3 %
yield. To increase the yield of 1 Scott and co-workers carried out
a five-step synthesis of halogenated trichlorodecacyclene 3,
which upon pyrolysis provided a 25–27 % yield of circumtrin-
dene (1); however, the final step to produce compound 3 was
problematic, yielding only 25 % of the desired product.
Results and Discussion
To introduce an external radical source into our FVP apparatus,
we designed a new setup that would allow us to introduce a
known amount of solvent vapour into the carrier gas stream
(Fig. 1). The carrier gas is split into two streams, each of which
is regulated by a flow meter. One flow meter regulates the
amount of unaltered nitrogen, while the other regulates the
flow of nitrogen through the reservoirs that contain the solvent
being used as the external radical source. The latter nitrogen
stream is bubbled through the solvent to maximize the amount
of solvent vapour in the gas phase. Gas diffusers (sparge tips)
were also utilized to increase the surface area of the nitrogen
bubbles passing through the solvent to maximize the amount of
solvent vapour in the nitrogen stream. In a third chamber both
the unaugmented nitrogen stream and the nitrogen stream that
contains solvent vapour are mixed, and this gas mixture is used
as the carrier gas. This new system was calibrated so that a set
amount of solvent vapour (in mole-%) could be inserted into
the carrier gas.
Howard and co-workers have shown that fuel-rich flames
produce fullerenes,[22–32] nanotubes,[33,34] geodesic polyar-
enes,[35] and flat PAHs[36,37] that allow for the industrial scale
synthesis of these molecules. Their investigations into the
To test this new process we chose o-terphenyl (4)[39] as the
substrate for our proof-of-concept study. In previous pyrolysis
Journal compilation Ó CSIRO 2014
www.publish.csiro.au/journals/ajc

Use of External Radical Sources in Flash Vacuum Pyrolysis
1339
N2-hexane vapour
ϩ
1200–1300ЊC
1100ЊC, 0.67–1.33 hPa
0.3 %
4
5
6
1
2
Scheme 2. Pyrolysis of o-terphenyl (4) with various amounts of hexanes at
11008C and 0.67–1.33 hPa.
C
Њ
1100
Cl
25–27 %
Table 1. Pyrolysis of o-terphenyl (4) with various amounts of hexanes,
at 11008C and 0.67–1.33 hPa
Cl
Cl
Entry
mol-%
4A
5A
6A
1
2
3
4
5
6
0
98 %
33 %
54 %
48 %
72 %
55 %
2 %
33 %
46 %
28 %
20 %
11 %
0 %
34 %
0 %
3
2.3
3.2
5.9
6.9
12.2
Scheme 1. Scott and co-workers’ pyrolyzes of decacyclene (2) and
trichlorodecacyclene (3) to produce circumtrindene (1).
24 %
8 %
34 %
AThe percent product ratios were calculated from the integration of peaks for
4, 5, and 6 in the gas chromatogram of the crude pyrolysate of each reaction.
R
H
1,2-shift
of H
4
1,2-shift
of H
–H
Fig. 1. Apparatus for the introduction of an external radical source to a
H
FVP setup.
5
studies using 4, only a trace amount of the cyclodehydrogenated
product triphenylene (5) was produced. We initially examined
the use of carbon tetrachloride[40] and hexanes[41] as a source of
ꢀCl and ꢀalkyl radicals, respectively. We chose to use hexanes in
comparison to other smaller alkanes for ease of use, as the
introduction of ethane, propane, and butane would prove to be
slightly more difficult and potentially more hazardous. Initial
experiments demonstrated that both hexanes and carbon tetra-
chloride showed promise as they did increase the amount of
cyclodehydrogenated product 5; however, hexanes was chosen
over carbon tetrachloride as it did not produce noxious or solid
byproducts such as HCl and hexachloroethane.
The pyrolysis of o-terphenyl (4) with various amounts of
hexanes vapour was studied systematically (Scheme 2). When
the FVP was conducted on 4 in the absence of hexanes, only a
trace amount of triphenylene (5) was observed. As the concen-
tration of hexanes vapour increased so did the amount of 5. The
amount of 5 reached its maximum of 46 % at 3.2 mol-% of
hexanes vapour (Table 1). Hexanes vapour concentrations
above 3.2 mol-% showed a decrease in the amount of 5 being
produced and reached a low of 11 % as the amount of hexanes
increased to 12.2 mol-% (Table 1).
Scheme 3. Proposed mechanism for the cyclodehydrogenation of 4 to
yield 5.
This process is not expected to be selective and therefore this
aryl radical may need to do multiple 1,2-shift of hydrogen to be
in a position that is adjacent to the neighbouring aromatic ring
for cyclization to occur (Scheme 3).[42]
The introduction of a large concentration of hexanes was
observed to have a deleterious effect on the production of 5 from
4. Spectroscopic analysis revealed that a phenyl group was lost
from 4 to form biphenyl (6). Formation of biphenyl (6) shows a
degradation pathway in which a phenyl group is lost by the
attachment of a hydrogen atom to one of the ipso-carbons of 4
(Scheme 4).[43–46] If the addition of the hydrogen atom occurs on
the ipso-carbon of the phenyl ring (Pathway 1 below) the
molecule can fragment to yield benzene and a biphenyl radical.
If the hydrogen atom adds to the ortho-substituted benzene ring
of 4 (Pathway 2 below) the molecule will fragment to produce
biphenyl (6) and a benzene radical. Both Pathways 1 and 2 will
lead to the production of biphenyl, which is observed in the
crude pyrolysate. As this degradation is catalyzed by ꢀH radical,
it is no surprise that a larger concentration of hexanes vapour
will increase the amount of biphenyl (6) and benzene produced.
The mechanism for the cyclodehydrogenation reaction cata-
lyzed by an external radical source will begin with the abstrac-
tion of a hydrogen atom on the aromatic ring by an alkyl radical.

1340
A. W. Amick and S. E. Martin
H
H
N2-hexane vapour
H
H
ϩ
1100ЊC, 0.67–1.33 hPa
Pathway 1
Pathway 2
10
11
13
12
H
ϩ
Scheme 6. Pyrolysis of benzo[c]phenanthrene (10) with various amounts
of hexanes vapour, at 11008C and 0.67–1.33 hPa.
ϩ
H
Scheme 4. Two mechanistic pathways for the degradation of 4 into
Table 3. Pyrolysis of benzo[c]phenanthrene (10) with various amounts
of hexanes, at 11008C and 0.67–1.33 hPa
biphenyl and benzene.
Entry
mol-%
10A
11A
12A
13A
N2-hexane vapour
1
2
3
0
88 %
53 %
47 %
8 %
32 %
33 %
3 %
11 %
16 %
1 %
4 %
4 %
3.2
5.9
ϩ
1100ЊC, 0.67–1.33 hPa
AThe percent product ratios were calculated from the integration of known
signals for 10, 11, 12, and 13 in the 1H NMR of the crude pyrolysate of each
reaction.
7
9
8
Scheme 5. Pyrolysis of 1,10-binaphthyl (7) with various amounts of
hexanes at 11008C and 0.67–1.33 hPa.
The addition of hexanes produced a modest increase in the
amounts of 8 and 9 that was produced (combined yield of 38 %)
as compared with a pyrolysis in the absence of hexanes (com-
bined yield of 22 %). The data also illustrate that there was no
preference for ring size formation; the six-membered ring
product 8 and five-membered ring product 9 were formed in
equal amounts at higher hexane concentrations.
Table 2. Pyrolysis of 1,19-binaphthyl (7) with various amounts of
hexanes, at 11008C and 0.67–1.33 hPa
Entry
mol-%
7A
8A
9A
1
2
3
0.0
3.2
5.9
78 %
70 %
63 %
13 %
20 %
19 %
9 %
10 %
19 %
With no inherent difference in the energy barrier for the
formation of a five-membered ring versus a six-membered
ring, we decided to examine the ability of hexanes to form other
five-membered rings in FVP, and the substrate that was chosen
was benzo[c]phenanthrene (10). This molecule is only able to
form a five-membered ring to produce benzo[ghi]fluoranthene
(11) (Scheme 6). Previous FVP studies revealed a low conver-
sion of 10 to 11, and when 11 is formed it can rearrange to form
cyclopenta[cd]pyrene (12).[48,49] Benzo[c]phenanthrene (10)
also rearranges under these conditions to afford chrysene (13),
which was discovered by Necula and Scott in unpublished work.
Using our new technique, the pyrolysis of compound 10 was
investigated at various hexanes vapour concentrations.
As with other substrates, when no hexanes vapour was
present, the conversion of 10 to 11 and 12 was low (a combined
yield of 11 %). When 3.2 mol-% of hexanes vapour was added,
the combined ratio of 11 and 12 greatly increases to 43 %, with
the amount of the desired benzo[ghi]fluoranthene (11) at 32 %.
A maximum combined product ratio of 11 and 12 of 49 % was
obtained at 5.9 mol-% hexanes, with the amount of the desired
11 at 33 % (Table 3). These results demonstrate the power of this
new technique to increase the production of cyclodehydroge-
nated products in FVP from substrates that typically show
limited reactivity.
AThe percent product ratios were calculated from the integration of known
signals for 7, 8, and 9 in the 1H NMR of the crude pyrolysate of each reaction.
Our current FVP apparatus limits our ability to accurately
determine the amount of benzene being produced by the
degradation of 4, which we believe would account for the erratic
quantities we observe for biphenyl (6).
A substantial amount of information was gained during this
initial study, but most important was the discovery that the
addition of hexanes vapour to the carrier gas in FVP does
increase the amount of cyclodehydrogenated product 5 and that
a large amount of hexanes vapour is not needed to afford the
desired transformation. Using this information we expanded the
scope of this methodology to other substrates.
Next we chose to examine the pyrolysis of 1,10-binaphthyl
(7) with hexanes as a radical source. The cyclodehydrogenation
of compound 7 can close to form a new six-membered ring
affording perylene (8), or it can close a five-membered ring to
create benzo[j]fluoranthene (9) (Scheme 5).[47]
Pyrolysis of 7 was surveyed with 0 to 5.9 mol-% of hexanes
vapour (Table 2). When no hexanes vapour is present the
product ratio of perylene (8) is 13 % and benzo[j]fluoranthene
(9) is 9 %. As the amount of hexanes was increased to 3.2 mol-%,
the amount of perylene (8) increased to 20 %, while 9 was
obtained in 10 % yield. The maximum product ratio of perylene
(8) and benzo[j]fluoranthene (9) was 19 and 19 %, respectively,
at 5.9 mol-% hexanes.
We also examined the ability of other solvents to act as an
external radical source in FVP. The following four solvents were
selected based upon what radicals they would produce under
FVP conditions: dichloromethane (BDEC–Cl 338 kJ molꢁ1
,
where BDE ¼ bond dissociation energy),[50,51] nitromethane
(BDEC–N 256 kJ molꢁ1),[41] trimethyl orthoformate (BDEC–O
,356 kJ molꢁ1),[41,52] and methyl tert-butyl ether (BDEC–O

Use of External Radical Sources in Flash Vacuum Pyrolysis
1341
ppm downfield from tetramethylsilane with deuterochloroform
(d 77.16) as a secondary reference. Analysis and integration of
1H and 13C NMR spectra was accomplished using MestreC
NMR software. Preparative thin layer chromatography was
performed on 20 cm ꢂ 20 cm Analtech Uniplate Taper plates,
Alumina GF. For column chromatography, silica gel 32–63 mm
was used. Gas chromatography–mass spectrometry (GCMS)
analysis was performed on a Thermo Finnigan Trace DSQ
with electron impact ionization with a Restek Rtx-PCB
(30 m ꢂ 0.25 mm ꢂ 0.25 mm film) column, and GC integrations
were accomplished using the Avalon integration program. High-
resolution mass spectrometry (HRMS) was performed by the
Mass Spectrometry Center, Boston College. All FVPs were
done according to the procedure developed by Necula and
Scott.[13] To ensure an accurate view of the product distribu-
tions limited purification was done on each sample. Typically
samples were passed through a silica gel plug with dichlor-
omethane as the eluent. Following the plug, samples were
concentrated under reduced pressure and dried under high
vacuum. These purifications provided better 1H NMR spectra,
however they did not remove higher molecular weight soluble
polymers that were created during the pyrolysis and in some
cases led to mass recoveries greater than 100 %.
N2-solvent vapour
ϩ
1100ЊC, 0.67–1.33 hPa
10
11
13
12
Scheme 7. Pyrolysis of benzo[c]phenanthrene (10) with various solvents,
at a temperature of 11008C and 0.67–1.33 hPa.
Table 4. Pyrolysis of benzo[c]phenanthrene (10) with various solvents
Solvent
10A
11A
12A
13A
Hexanes
53 %
81 %
63 %
69 %
58 %
32 %
19 %
27 %
25 %
29 %
11 %
0 %
4 %
0 %
1 %
2 %
2 %
Dichloromethane
Nitromethane
10 %
4 %
Trimethyl orthoformate
Methyl tert-butyl ether
11 %
AThe percent product ratios were calculated from the integration of known
signals for 10, 11, 12, and 13 in the 1H NMR of the crude pyrolysate of each
reaction.
External Radical Source Introduction Method
The introduction of an external radical source was accomplished
by splitting the nitrogen carrier gas into two streams that were
regulated by flow meters. One flow meter was used for pure
nitrogen gas, and the other regulated the flow of nitrogen gas
through two reservoirs of the solvent used as the radical gen-
erator, connected in series. The pure nitrogen gas stream and the
stream carrying the external radical source were combined and
used as the carrier gas for FVP. The amount of solvent in the
carrier gas (reported in mol-%) was determined by capturing the
solvent vapours in a liquid nitrogen cooled trap over a period of
time. The solvent was removed, and its volume was measured to
provide a lower limit for the amount of solvent per unit time. The
Scienceware flow meters were calibrated to supply 961 mL
minꢁ1 of gas at a reading of 20 mm. Knowing both the volume of
nitrogen gas and the amount of solvent collected per unit time,
the mole fractions were calculated, assuming 24.47 L molꢁ1 for
nitrogen gas. The design of this setup also allowed for per-
turbations to be made to the apparatus. These had a direct affect
on the amount of solvent vapour in the gas phase (i.e. 1 chamber,
2 chambers, and with or without sparging), and all perturbations
used in our study were calibrated. To determine the effective-
ness of this procedure, we examined the ratio of products to
starting material by the integration of the signals in the chro-
matogram of the GCMS or by the integration of known
356 kJ molꢁ1).[41,53] The substrate chosen for this solvent screen
was benzo[c]phenanthrene (10) (Scheme 7), as this substrate
will not suffer the loss of appended phenyl or naphthyl groups
that o-terphenyl (4) and 1,10-binaphthyl (7) can. The results of
this study are summarized in Table 4.
Each solvent did act as a radical generator in FVP to a varying
extent, with methyl tert-butyl ether having the best products to
starting material distribution among the four new solvents
tested, with 11 being produced in 29 % yield. The best ratio of
10 to 11 was obtained using dichloromethane, but this result is
deceiving, as the pyrolysis of dichloromethane produced
HCl, which likely caused the polymerization of cyclopenta
[cd]pyrene (12). We have demonstrated that all four solvents
studied can be used as external radical sources, with hexanes and
methyl tert-butyl ether providing the best results.
Conclusions
The use of external radical sources in FVP has been a successful
venture, and this new technique shows great promise. We
have shown that the introduction of an external radical source
can greatly increase the reactivity of usually inert PAHs such as
o-terphenyl (4) and benzo[c]phenanthrene (10) at 11008C.
Hexanes were used extensively throughout this study, but our
work has shown that other solvents such as dichloromethane,
methyl tert-butyl ether, nitromethane, and methyl orthoformate
all have the ability to generate radicals at high temperatures and
facilitate the cyclodehydrogenation of PAHs.
1
signals in H NMR of the crude pyrolysate. The calibration
data and all spectra for each experiment are provided in the
Supplemental Material.
General Procedure for the FVP of o-Terphenyl (4)
With and Without an External Radical Source
Experimental
Flash vacuum pyrolysis was performed on 100 mg (0.435 mmol)
of o-terphenyl (4) at 11008C with a steady flow of carrier gas
with a composition of 0–12.2 mol-% hexanes vapour in the
nitrogen gas (final pressure 0.67–1.33 hPa) over 4–6 h utilizing
one or two reservoirs and with or without sparge. The pyrolysis
was accomplished in a quartz tube (length ¼ 105.4 cm and int-
erior diameter ¼ 3.4 cm), in an electric oven (length ¼ 61 cm).
General Experimental Procedures
All solvents and commercial chemicals were used without
purification, unless otherwise stated. Proton NMR chemical
shifts are reported in ppm downfield from tetramethylsilane
with deuterochloroform (d 7.26) as a secondary reference,
unless otherwise specified. Carbon NMR shifts are reported in

1342
A. W. Amick and S. E. Martin
Table 5. Mass recoveries for the pyrolyses of o-terphenyl (4) with
Table 8. The mole percent of the various solvents used in the solvent
screen
various amounts of hexanes
Hexanes vapour
[mol-%]
Starting mass (4)
Mass recovered
Solvent
Solvent
[mol minꢁ1
N2
[mol minꢁ1
Solvent
]
]
[mol-%]
0.0
2.3
3.2
5.9
6.9
12.2
100 mg
100 mg
100 mg
100 mg
100 mg
100 mg
87 mg
127 mg
49 mg
Nitromethane
0.0003
0.0059
0.0061
0.0026
0.0392
0.0392
0.0392
0.0392
0.71 %
13.15 %
13.47 %
6.23 %
Trimethyl orthoformate
Dichloromethane
95 mg
Methyl tert-butyl ether
99 mg
147 mg
Table 9. Mass recoveries for the pyrolyses of benzo[c]phenanthrene
(10) with various solvents
Table 6. Mass recoveries for the pyrolysis of 1,19-binaphthyl (7) with
various amounts of hexanes
Solvent
Starting amount
Mass recovered
Nitromethane
40 mg
40 mg
40 mg
40 mg
32 mg
64 mg
32 mg
69 mg
Hexanes vapour
[mol-%]
Starting mass (7)
Mass recovered
Trimethyl orthoformate
Dichloromethane
0
100 mg
100 mg
100 mg
86 mg
100 mg
92 mg
Methyl tert-butyl ether
3.2
5.9
steady flow of carrier gas with a composition of 0–5.9 mol-%
hexanes vapour in nitrogen gas (final pressure 0.67–1.33 hPa)
over 4–6 h utilizing two reservoirs and with no sparge. The
pyrolysis was accomplished in a quartz tube (length ¼ 105.4 cm
and interior diameter ¼ 3.4 cm), in an electric oven (length ¼
61 cm). The crude pyrolysate was isolated from the cold trap
and the quartz tube and passed through a silica plug with
dichloromethane. 1H NMR analysis of the crude pyrolysate
was used to determine the ratio of 10 to 11, 12, and 13. The
hydrogen atoms that were integrated to determine the product
ratios have been highlighted and colour-coded on the spectra in
the Supplementary Material. Mass recoveries for these reactions
are shown in Table 7.
Table 7. Mass recoveries for the pyrolyses of benzo[c]phenanthrene
(10) with various amounts of hexanes
Hexanes vapour
[mol-%]
Starting mass (10)
Mass recovered
0
40 mg
80 mg
80 mg
39 mg
80 mg
3.2
5.9
109 mg
The crude pyrolysate was isolated from the cold trap and the
quartz tube. GCMS analysis of the crude pyrolysate was used to
determine the product ratios of starting material 4 to tripheny-
lene (5) and biphenyl (6). Mass recoveries for these reactions are
shown in Table 5 and the spectra for these reactions are provided
in the Supplementary Material.
General Procedure for the FVP of Benzo[c]phenanthrene
(10) Using Various External Radical Sources
Flash vacuum pyrolysis was performed on 40 mg (0.18 mmol) of
benzo[c]phenanthrene (10) at 11008C with a steady flow of
nitrogen through the solvent reservoirs (,175 mL minꢁ1) and a
flow rate of pure nitrogen in the other stream of 786 mL minꢁ1
(final pressure 0.67–1.33 hPa) over 4–6 h utilizing two reser-
voirs and with sparge. We chose to keep the flow constant
because each solvent has different vapour pressures, making
it difficult to keep the concentration of each the same. The
pyrolysis was accomplished in a quartz tube (length ¼ 105.4 cm
and interior diameter ¼ 3.4 cm), in an electric oven (length ¼
61 cm). Four solvents were investigated: dichloromethane,
nitromethane, trimethyl orthoformate, and methyl tert-butyl
ether. The crude pyrolysate was isolated from the cold trap and
the quartz tube and passed through a silica plug with dichlor-
omethane. 1H NMR analysis of the crude pyrolysate was used to
determine the ratio of 10 to 11, 12, and 13. The hydrogens that
were integrated to determine the product ratios have been
highlighted and colour-coded on the spectra in the Supple-
mentary Material. The mole percent for each solvent was
determined and are listed in the Table 8 as well as the mass
recoveries of each reaction shown in Table 9.
General Procedure for the FVP of 1,10-Binaphthyl (7)
With and Without an External Radical Source
Flash vacuum pyrolysis was performed on 100 mg (0.394 mmol)
of 1,10-binaphthyl (7) at 11008C with a steady flow of carrier gas
with a composition of 0–5.9 mol-% hexanes vapour in the
nitrogen gas (final pressure 0.67–1.33 hPa) over 4–6 h utilizing
two reservoirs and with no sparge. The pyrolysis was accom-
plished in a quartz tube (length ¼ 105.4 cm and interior dia-
meter ¼ 3.4 cm), in an electric oven (length ¼ 61 cm). The crude
pyrolysate was isolated from the cold trap and the quartz tube,
and passed through a silica plug with dichloromethane. 1H NMR
analysis of the crude pyrolysate was used to determine the
product ratio of starting 7 to compounds 8 and 9. The hydrogen
atoms that were integrated to determine the product ratios have
been highlighted and colour-coded on the spectra in the Sup-
plementary Material. Mass recoveries for these reactions are
shown in Table 6.
General Procedure for the FVP of Benzo[c]phenanthrene
(10) With and Without an External Radical Source
Supplementary Material
Flash vacuum pyrolysis was performed on 40–80 mg
(0.35 mmol) of benzo[c]phenanthrene (10) at 11008C with a
All spectroscopic data and specific experimental details for this
study are available on the Journal’s website.

Use of External Radical Sources in Flash Vacuum Pyrolysis
1343
Acknowledgement
[28] K. Das Chowdhury, J. B. Howard, J. B. Vander Sande, J. Mater. Res.
1996, 11, 341. doi:10.1557/JMR.1996.0040
[29] V. M. Rotello, J. B. Howard, T. Yadav, M. M. Conn, E. Viani,
L. M. Giovane, A. L. Lafleur, Tetrahedron Lett. 1993, 34, 1561.
doi:10.1016/0040-4039(93)85006-I
[30] J. B. Howard, A. L. Lafleur, Y. Makarovsky, S. Mitra, C. J. Pope,
T. K. Yadav, Carbon 1992, 30, 1183. doi:10.1016/0008-6223(92)
90061-Z
[31] J. B. Howard, J. T. McKinnon, M. E. Johnson, Y. Makarovsky,
A. L. Lafleur, J. Phys. Chem. 1992, 96, 6657. doi:10.1021/
J100195A026
The authors would like to thank Larry Scott for his insights on this project.
Larry has been a great mentor and teacher.
References
[1] R. F. C. Brown, Pyrolytic Methods in Organic Chemistry: Application
of Flow and Flash Vacuum Pyrolytic Techniques 1980 (Academic
Press: New York, NY).
[2] R. F. C. Brown, Recl. Trav. Chim. Pays Bas 1988, 107, 655.
doi:10.1002/RECL.19881071202
[3] R. F. C. Brown, Pure Appl. Chem. 1990, 62, 1981. doi:10.1351/
PAC199062101981
[4] R. F. C. Brown, F. W. Eastwood, Synlett 1993, 9. doi:10.1055/
S-1993-22330
[5] R. F. C. Brown, Eur. J. Org. Chem. 1999, 3211. doi:10.1002/(SICI)
1099-0690(199912)1999:12,3211::AID-EJOC3211.3.0.CO;2-F
[6] C. Wentrup, Chimia (Aarau) 1977, 31, 258.
[7] C. Wentrup, Lect. Heterocycl. Chem. 1984, 7, 91.
[8] H. McNab, Contemp. Org. Synth. 1996, 3, 373. doi:10.1039/
CO9960300373
[32] J. B. Howard, J. T. McKinnon, Y. Makarovsky, A. L. Lafleur,
M. E. Johnson, Nature 1991, 352, 139. doi:10.1038/352139A0
[33] M. J. Height, J. B. Howard, J. W. Tester, J. B. Vander Sande, J. Phys.
Chem. B 2005, 109, 12337. doi:10.1021/JP046021N
[34] M. J. Height, J. B. Howard, J. W. Tester, J. B. Vander Sande, Carbon
2004, 42, 2295. doi:10.1016/J.CARBON.2004.05.010
[35] A. L. Lafleur, J. B. Howard, J. A. Marr, T. Yadav, J. Phys. Chem. 1993,
97, 13539. doi:10.1021/J100153A020
[36] A. L. Lafleur, J. B. Howard, E. Plummer, K. Taghizadeh, A. Necula,
L. T. Scott, K. C. Swallow, Polycycl. Aromat. Comp. 1998, 12, 223.
doi:10.1080/10406639808233840
[9] H. McNab, Aldrichim Acta 2004, 37, 19.
[10] L. T. Scott, Acc. Chem. Res. 1982, 15, 52. doi:10.1021/AR00074A004
[11] L. T. Scott, Chem. Aust. 1987, 54, 298.
[12] L. T. Scott, M. S. Bratcher, S. Hagen, J. Am. Chem. Soc. 1996, 118,
8743. doi:10.1021/JA9621511
[13] A. Necula, L. T. Scott, J. Anal. Appl. Pyrolysis 2000, 54, 65.
doi:10.1016/S0165-2370(99)00085-6
[14] L. T. Scott, Angew. Chem. Int. Ed. 2004, 43, 4994. doi:10.1002/
ANIE.200400661
[37] A. L. Lafleur, J. B. Howard, K. Taghizadeh, E. F. Plummer, L. T. Scott,
A. Necula, K. C. Swallow, J. Phys. Chem. 1996, 100, 17421.
doi:10.1021/JP9605313
[38] H. Richter, W. J. Grieco, J. B. Howard, Combust. Flame 1999, 119, 1.
doi:10.1016/S0010-2180(99)00032-2
[39] C. Jacobelli, G. Perez, C. Polcaro, E. Possagno, R. Bassanelli, E. Lilla,
J. Anal. Appl. Pyrolysis 1983, 5, 237. doi:10.1016/0165-2370(83)
80030-8
[15] V. M. Tsefrikas, L. T. Scott, Chem. Rev. 2006, 106, 4868. doi:10.1021/
CR050553Y
[16] C. Wentrup, Aust. J. Chem. 2014, 67, 1150. doi:10.1071/CH14096
[17] L. T. Scott, M. S. Bratcher, S. Hagen, J. Am. Chem. Soc. 1996, 118,
8743. doi:10.1021/JA9621511
[40] G. Glockler, J. Phys. Chem. 1959, 63, 828. doi:10.1021/J150576A013
[41] S. J. Blanksby, G. B. Ellison, Acc. Chem. Res. 2003, 36, 255.
doi:10.1021/AR020230D
[42] M. A. Brooks, L. T. Scott, J. Am. Chem. Soc. 1999, 121, 5444.
doi:10.1021/JA984472D
[18] R. B. M. Ansems, L. T. Scott, J. Am. Chem. Soc. 2000, 122, 2719.
doi:10.1021/JA993028N
[19] M. M. Boorum, Y. V. Vasil’ev, T. Drewello, L. T. Scott, Science 2001,
294, 828. doi:10.1126/SCIENCE.1064250
[43] B. R. Cook, B. B. Wilkinson, C. C. Culross, S. M. Holmes,
L. E. Martinez, Energy Fuels 1997, 11, 61. doi:10.1021/EF9600669
[44] R. Louw, J. W. Rothuizen, R. C. C. Wegman, J. Chem Soc., Perkin
Trans. 2 1973, 1635. doi:10.1039/P29730001635
[20] L. T. Scott, M. M. Boorum, B. J. McMahon, S. Hagen, J. Mack,
J. Blank, H. Wegner, A. de Meijere, Science 2002, 295, 1500.
doi:10.1126/SCIENCE.1068427
[21] L. T. Scott, E. A. Jackson, Q. Zhang, B. D. Steinberg, M. Bancu, B. Li,
J. Am. Chem. Soc. 2012, 134, 107. doi:10.1021/JA209461G
[22] A. Goel, P. Hebgen, J. B. Vander Sande, J. B. Howard, Carbon 2002,
40, 177. doi:10.1016/S0008-6223(01)00170-1
[45] A. Fahr, S. E. Stein, J. Phys. Chem. 1988, 92, 4951. doi:10.1021/
J100328A026
[46] L. W. Vernon, Fuel 1980, 59, 102. doi:10.1016/0016-2361(80)90049-6
[47] X. A. Garcia, K. J. Huettinger, Erdol Kohle Erdgas Petrochem. 1990,
43, 113.
[48] M. Sarobe, L. W. Jenneskens, U. E. Wiersum, Tetrahedron Lett. 1996,
37, 1121. doi:10.1016/0040-4039(95)02320-8
[23] K. Swallow, J. Howard, W. Grieco, T. Benish, K. Taghizadeh,
E. Plummer, A. L. Lafleur, Polycycl. Aromat. Comp. 1999, 14, 201.
doi:10.1080/10406639908019126
[24] W. J. Grieco, J. B. Howard, L. C. Rainey, J. B. Vander Sande, Carbon
2000, 38, 597. doi:10.1016/S0008-6223(99)00149-9
[25] W. J. Grieco, A. L. Lafleur, H. Richter, K. C. Swallow, K. Taghizadeh,
J. B. Howard, Chem. Phys. Process. Comb. 1997, 131.
[26] H. Richter, A. J. Labrocca, W. J. Grieco, K. Taghizadeh, A. L. Lafleur,
J. B. Howard, J. Phys. Chem. B 1997, 101, 1556. doi:10.1021/
JP962928C
[49] L. T. Scott, N. H. Roelofs, Tetrahedron Lett. 1988, 29, 6857.
doi:10.1016/S0040-4039(00)88459-2
[50] W. S. McGivern, A. Derecskei-Kovacs, S. W. North, J. S. Francisco,
J. Phys. Chem. A 2000, 104, 436. doi:10.1021/JP993275D
[51] E. Tschuikow-Roux, S. Paddison, Int. J. Chem. Kinet. 1987, 19, 15.
doi:10.1002/KIN.550190103
[52] Assumed to be comparable to the dissociation of isopropylmethyl
ether.
[53] B. S. Jursic, J. Chem. Soc., Perkin Trans. 2 1999, 369. doi:10.1039/
A803744B
[27] H. Richter, K. Taghizadeh, W. J. Grieco, A. L. Lafleur, J. B. Howard,
J. Phys. Chem. 1996, 100, 19603. doi:10.1021/JP962356W