Intramolecular Reactions of Tethered Furan-Based Bis(p-quinodimethanes)

Article abstract of DOI:10.1002/ejoc.201600995

Tethered bis-2,5-dimethylene-2,5-dihydrofurans were generated in the gas phase by flash vacuum pyrolysis (FVP) of diester precursors. These furan-based bis(p-quinodimethanes) are shown to undergo reactions leading to macrocycles. The observed products strongly support a mechanism involving cyclic diradical intermediates. Formation of the furan-based p-quinodimethane and the corresponding cyclization chemistry was studied by high-level ab initio calculations. The theoretical studies reveal the importance of the entropy-driven elimination steps in the high-temperature FVP chemistry.

Full text of DOI:10.1002/ejoc.201600995

A Journal of
Accepted Article
Title: Intramolecular Reactions of Tethered Furan-based bis-(p-Quinodimethanes)
Authors: Douglas Klumpp; Thomas Gilbert; Walter Trahanovsky
This manuscript has been accepted after peer review and the authors have elected
to post their Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by using the Digital
Object Identifier (DOI) given below. The VoR will be published online in Early View as
soon as possible and may be different to this Accepted Article as a result of editing.
Readers should obtain the VoR from the journal website shown below when it is pu-
blished to ensure accuracy of information. The authors are responsible for the content
of this Accepted Article.
To be cited as: Eur. J. Org. Chem. 10.1002/ejoc.201600995
Link to VoR: http://dx.doi.org/10.1002/ejoc.201600995
Supported by

European Journal of Organic Chemistry
10.1002/ejoc.201600995
FULL PAPER
Intramolecular Reactions of Tethered Furan-based bis-(p-
Quinodimethanes)
Douglas A. Klumpp *[a] Thomas M. Gilbert,[a] and Walter S. Trahanovsky,*[b]
Abstract: Tethered bis-2,5-dimethylene-2,5-dihydrofurans have
been generated in the gas-phase by flash vacuum pyrolysis (FVP) of
diester precursors. These furan-based bis-(p-quinodimethanes) are
shown to undergo reactions leading to macrocycles. The observed
products strongly support a mechanism involving cyclic diradical
Scheme 1. p-Quinonedimethanes.
intermediates. Formation of the furan-based p-quinodimethane and
the corresponding cyclization chemistry was studied by high-level ab
initio calculations. The theoretical studies reveal the importance of
Both compounds 2 and 3 are well known heterocycle-
the entropy-driven elimination steps in the high temperature FVP
chemistry.
based p-quinodimethanes. These heterocyclic systems have
been generated by pyrolysis of varied precursors,14 silyl-induced
elimination reactions,6a and Hofmann elimination.15 Both the
furan- and thiophene-based p-quinodimethanes (2 and 3) have
been characterized by spectroscopic methods.14c Like the p-
xylylene (1), compounds 2 and 3 are known to form cyclophane
products by intermolecular reactions. For example, the furan-
based system (2,5-dimethylene-2,5-dihydrofuran, 2) gives the
dimeric cyclophane (5) in 73% yield (Scheme 2).6a The
mechanism for this conversion is proposed to involve the
diradical species (4), although the diradical intermediate could
not be trapped by hydrogen atom donors or nitroxyl radical
traps.14a
Introduction
p-Quinodimethanes are an important class of organic reactive
intermediates. They are involved in reactions such as in the
pyrolysis of organic compounds, cyclophane syntheses, and in
polymerizations. This type of intermediate was first proposed by
Szwarc in a report describing the pyrolysis of p-xylene.1 At
1000°C, p-xylene leads to the formation of poly-p-xylylene,
oligomers, and several cyclophane products. The conversions
may be understood by formation of the parent p-quinodimethane
reactive intermediate (p-xylylene, 1, Scheme 1) via
dehydrogenation of p-xylene. p-Xylylene (1) and its substituted
derivatives have been studied extensively since Szwarc’s initial
report, in part due to the commercial importance of poly-p-
xylylene macromolecules.2,3 The unique electronic properties p-
quinodimethanes have also been the subject of many theoretical
investigations.4 A variety of methods have been used to
generate 1, including pyrolysis of organomanganese
substrates,5 fluoride-induced silyl elimination,6 Hofmann
elimination,7 photolysis of ,’-dichloroxylenes,8 thermal and
Scheme 2. Reaction of furan-based p-quinodimethane (2).
photochemical
dissociation
of
[2,2’]paracyclophane,9
In the following manuscript, we describe a pyrolytic method
of generating two furan-based p-quinodimethanes tethered by
hydrocarbon chains of varying length. These furan-based bis(p-
quinodimethanes) are shown to undergo several characteristic
reactions of diradical species, including disproportionation,
coupling, and cleavage processes – providing strong evidence
for the role of diradicals in the chemistry of furan-based p-
electrochemical reactions,10 dehydrohalogenation,11 elimination
with bis(sulfonium ion) salts,12 p-Xylylene (1) has also been
thoroughly characterized by spectroscopic methods.13
quinodimethanes.
disproportionation provides
The reaction path involving diradical
novel route to macrocyclic
[a]
[b]
D. A. Klumpp, T. M. Gilbert
Department of Chemistry and Biochemistry
Northern Illinois University
DeKalb, IL 60115 USA
E-mail: dklumpp@niu.edu
W. S. Trahanovsky
a
products having 15 to 22 carbons. The proposed mechanism is
also studied by ab initio computational methods.
Department of Chemistry
Iowa State University
Results and Discussion
Ames, IA 50011-3111 USA
As described previously, the furan-based p-quinodimethane (2)
may be generated by flash vacuum pyrolysis (FVP) of (5-
methylfuran-2-yl)methyl benzoate and related esters.14 The
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved

European Journal of Organic Chemistry
10.1002/ejoc.201600995
FULL PAPER
conversion is thought to occur by a pair [3,3] sigmatropic bond
migrations and -elimination of benzoic acid.14a We sought to
utilize this chemistry to generate a pair of furan-based p-
quinodimethanes linked by a hydrocarbon chain or other tether.
Thus, a series of precursors were synthesized with varied tether
lengths between the furan rings (Scheme 3). 2-Furyllithium is
prepared and reacts with the dibromoalkanes to give the di(2-
furyl)alkanes (6a-g, 62-90% yields). Subsequent formylation
(7a-g, 50-70% yields), reduction, and acylation, gives the FVP
precursors (8a-g, two steps, 80-91% yields). Because these
compounds were expected to be considerably less volatile than
(5-methylfuran-2-yl)methyl benzoate, the acetate esters were
prepared rather than the benzoate esters. In FVP experiments,
both the acetate and benzoate esters have been shown to
generate the desired 2,5-dimethylene-2,5-dihydrofuran 2.14
trap. Chloroform was subsequently distilled into the trap for
isolation of the products (see Supporting Information). The first
compound in the series (8a) provides an unexpected
fragmentation product 9a in 33% yield from FVP as the only
isolable major product (Scheme 4). It is unlikely that product 9b
is formed via the furan-based bis(p-quinodimethane). However,
FVP of the butanediyl derivative 8b gives product 10 as the only
major product, though two other isomeric products could be
detected by GCMS. The furan-based bis(p-quinodimethane) is
likely involved in the formation of compound 10 (vide infra).
With FVP of larger substrates, macrocycles are formed in
reasonable yields (Table 1). For example, compound 8c gives
compound 11c (cis) as the only major product in 46% yield.
Table 1. FVP products and yields from diester substrates 8c-f.
Scheme 3. Synthesis of diester substrates 8a-g.
Compounds 8a-g were subjected to FVP, using
a
procedure similar to that employed in the pyrolysis of the mono-
ester (5-methylfuran-2-yl)methyl benzoate.14 The substrates
were place in a Pyrex sample chamber and the FVP apparatus
evacuated to 10-5 Torr. With mild heating of the sample
chamber (25–150 °C), the precursors were volatilized, passed
through a heated quartz pyrolysis tube packed with quartz
helices, and the products deposited in a liquid nitrogen-cooled
Scheme 4. Pyrolysis products from 8a-b.
[a] Product yields determined by GC-FID analysis using diphenyl ether as an
internal standard (see Supporting Information)
This article is protected by copyright. All rights reserved

European Journal of Organic Chemistry
10.1002/ejoc.201600995
FULL PAPER
elimination of two acetic acid molecules from 8c leads to the
bis(furan-based p-quinodimethane) 13c (Scheme 6). Analogous
to the intermolecular
reactions
of
furan-based
p-
quinodimethanes, the initial cyclization reaction leads to the
diradical 14. The final macrocyclic product 11c (cis) is formed
by a transannular hydrogen atom abstraction involving the
diradical 14. The crude pyrolysis mixtures were isolated at low
temperature and analyzed by NMR. No bis(furan-based p-
quinodimethane) 13c was observed in the product mixture –
suggesting the cyclization and hydrogen migration occurs in the
gas-phase. Although the heated zone of the FVP apparatus
showed considerable amounts of charring, there were no dimers,
trimers, or oligomers observed in the product mixture.
Scheme 5. FVP products.
1
Based on H NMR data, the alkene group can be assigned as
having cis stereochemistry.
Compound 8d also gives
macrocyclic products, both 11d (cis) and 11d (trans), in 47%
and 11% yields, respectively. In addition to these products, the
ring-fused cyclophane 12d is isolated in 4% yield (Scheme 5).
1
Although H NMR suggests a single stereoisomer is formed, we
were not able to definitely assign the stereochemistry of the ring
junction. Additionally, a small amount of compound 9d could be
detected by GCMS and observed in the NMR spectrum of the
crude product mixture. This type of compound (i.e., 9b-f) is
observed as a minor product for all of the FVP experiments. Like
diester 8d, compound 8e gives the cis and trans alkenes – 11e
(cis) and 11e (trans) – as the major FVP products. The
proportion of trans stereoisomer increases with ring size and this
trend is also seen in the FVP of 8f. These 22 carbon
macrocycles – 11f (cis) and 11f (trans) – are formed in 23%
overall yield with the trans stereoisomer being favored.
Although 8g was subjected FVP, this substrate primarily forms
char and minor amounts of decomposition products.
Presumably, the large diester 8g is not sufficiently volatile for
clean pyrolytic reactions and it decomposes in the FVP hot zone.
As noted above, the macrocyclic products are formed with
a preference for the cis double bond for products 11c, 11d, and
11e, but with a preference for the trans double bond for products
11f. To determine if the alkene stereochemistry is controlled by
Scheme 6. Proposed mechanism for the fprmation of macrocycle 11c (cis).
The proposed general mechanism also explains some of
the other observed products. For example, compound 8b
provides the bis(2-vinylfuranyl) product 10 from FVP. This
product may arise by either of two routes involving the bis(furan-
based p-quinodimethane) 13b from 8b, either an allowed 14-
electron sigmatropic rearrangement or through 1,4-diradical 15
(Scheme 7). The cleavage of 1,4-diradicals is known to produce
two alkenyl groups.16 In the case of 8d, the ring-fused
cyclophane product 12d was isolated. This product can be
explained by formation of the corresponding bis(furan-based p-
quinodimethane) 13d from 8d and closure to provide a 1,6-
diradical 16. Subsequent bond forming leads to cyclophane 12d
from the 1,6-diradical intermediate 16.
a
kinetic or thermodynamic effect, we evaluated the
thermodynamic ratio of products 11e (cis) and 11e (trans) for
comparison with the product mixture from FVP of 8e (Table 2).
The thermodynamic mixtures were obtained by isolation of
purified 11e (cis) and 11e (trans) and subjecting these
compounds to pyrolysis at ca. 750 °C. From both purified 11e
(cis) and 11e (trans), the observed product mixture consisted of
an averaged 74:26 ratio of 11e (cis):11e (trans) – a value
assumed to be the thermodynamic ratio. Since this ratio differs
considerably from that obtained by FVP of 8e, we conclude the
product ratio is determined by kinetic control.
Table 2. Equilibration of 11e (cis) and 11e (trans) at ca. 750°C.
The observed FVP chemistry is consistent with the
Scheme 7. Diradical reactions from 8b and 8d.
formation of heterocyclic-based p-quinodimethanes linked by the
carbon-chain tether.
previous work with heterocyclic-based p-quinodimethanes - that
subsequent reaction steps involve diradical intermediates. Thus,
Product formation is in accord with
In order to further understand the mechanism of the
conversions to macrocycles, the furan-based p-quinodimethanes
were examined computationally. While multiple model
This article is protected by copyright. All rights reserved

European Journal of Organic Chemistry
10.1002/ejoc.201600995
FULL PAPER
chemistries were employed, for simplicity only MN12SX/6-
311+G(d,p) results are discussed hereafter (see Computational
Methods section and Supporting Information for reasons behind
this choice). Our initial studies examined the conversion of (5-
methylfuran-2-yl)methyl acetate 19 to the furan-based p-
Table 3. Relative energies (kJ mol–1
conversion of 19 to 2 (plus CH3CO2H).
,
MN12SX/6-311+G(d,p)) for the
19
0
19ts
182
184
188
184
194
20
58
57
63
59
71
20ts
163
162
166
163
171
21
38
37
44
39
56
21ts (2 + CH3CO2H)
E, 0K
H298
G298
H853
G853
164
168
169
169
172
74
73
quinodimethane (2,5-dimethylene-2,5-dihydrofuran, 2.
For
0
comparison, energetics were determined at 0K (ZPE-corrected
electronic energy), 298K, and 853K (experimental conditions)
using the 0K-optimized structures. As described previously, the
furan-based p-quinodimethane 2 is formed from the starting
ester by a pair of [3,3] sigmatropic bond migrations and a -
elimination of the carboxylic acid (Figure 1).14a This chemistry
leads to intermediates 20 and 21, and in due course, to the
product furan-based p-quinodimethane 2. The computational
results for this conversion are presented in Table 3 and Figure 1.
The data show that the energetics are quite consistent
regardless of temperature or thermodynamic parameter, save
for the last step where the acetic acid molecule is eliminated. At
all temperatures, formation of the first intermediate 20 is found to
involve a significant barrier of ca. 190 kJ mol–1 (energies
0
28
0
71
0
–56
Finally, -elimination through transition state 21ts leads to
the furan-based p-quinodimethane 2 and a molecule of acetic
acid. As above, the potential energy surface around 21ts is
quite flat, so distances should be viewed with caution. It is
plausible that two close stepwise transition states (one for
breaking of the C–O bond and one for transfer of the hydrogen
to the carbonyl oxygen/dissociation of acetic acid) are
concatenated as a single concerted one here. While -
elimination is generally assumed to be a concerted reaction,
Birney and coworkers have described a number of similar
pseudopericyclic rearrangements and ester elimination
reactions.17 The same dynamics may be operating in the
conversion to the furan-based p-quinodimethane 2. The effects
of temperature associated with formation of 2 are rather
dramatic: at all temperatures, the -elimination is endothermic;
at 298K, it is modestly endergonic; at 853K, it is substantially
exergonic.
The comparison reflects the contribution and
importance of entropy in the free energy of this conversion; one
sees that enthalpies H298 and H853 differ little, denoting that
differences in G are engendered by entropy. Thus, the only
overall calculated exergonic transformation (19  2 + AcOH) is
found at the FVP temperature. Since the steps prior to this
exhibit high energy barriers and involve high-energy
intermediates, this is consistent with the fact that no
intermediates – such as 20 or 21 – have been observed from
FVP experiments.14
Figure 1. MN12SX/6-311+G(d,p) optimized structures for the conversion of 19
to 2 (relative free energies at 580 °C/853 K, kJ mol–1).
The transformation of 8c to 11c (cis) was also studied
computationally (Table 4 and Figures 2-3). As in the case of
furan-based p-quinodimethane 2, formation of the furan-based
bis(p-quinodimethane) 13c is found to be a strongly entropy-
driven process. Of the three possible stereoisomers of 13c, all
were characterized as minima on the potential energy surface
and geometry optimizations indicated that the 13c-Z,Z
stereoisomer (Scheme 8) is 10 kJ mol–1 more stable than 13c-
E,E and 13 kJ mol–1 more stable than the 13c-E,Z. Nevertheless,
the three stereoisomers of 13c are expected to rapidly
interconvert at the pyrolysis temperature. The most stable
stereoisomer, 13c-Z,Z, is used in subsequent calculations and
discussion.
associated
with
19ts)
and
to
be
significantly
endothermic/endergonic. This is clearly a consequence of the
loss of aromatic stabilization within the furan ring. During this
first [3,3]-sigmatropic bond migration, the incipient C–O bond
forms from the carbonyl oxygen in 19 via the six-membered
cyclic transition state 19ts. The transferring acetate is poorly
held by the molecular core, as exemplified by the long distance
for the breaking C–O bond (2.492 Å). The potential energy
surface around the transition state is quite flat, so this distance
value is likely approximate. A degree of stabilization is achieved
with the second sigmatropic bond shift to form intermediate 21
but the intermediate is still significantly less stable than the
starting material 19. Transition state 20ts provides a slightly
smaller barrier than does 19ts. This may arise because
transition state 20ts is earlier (requires less breaking of the C–O
bond) than is 19ts, as suggested by the breaking C–O distances
(2.304 vs 2.492 Å).
Scheme 8. Isomers of furan-based bis(p-quinodimethane) 13.
This article is protected by copyright. All rights reserved

European Journal of Organic Chemistry
10.1002/ejoc.201600995
FULL PAPER
With elimination of two molecules of acetic acid at 853K,
the furan-based bis(p-quinodimethane) intermediate 13c-Z,Z
forms in a highly exergonic process (G853 = –192 kJ mol–1),
contrasted with the endothermicities predicted by the E and H
values (ca. 90 kJ mol–1at all temperatures) and the essentially
neutral G298 value. (Table 4). This value is notably greater than
that for elimination of two molecules of acetic acid from two
molecules of monoester 19 (G853 = 2 x 56 = 112 kJ mol–1) to
give two molecules of the furan-based p-quinodimethane 2.
formation of the cyclophane structure in 14 might prevent
efficient stabilization of the benzylic radical centers (contributing
to the unfavorable enthalpy change); however, this does not
seem to be the case. The optimized structure of 14 exhibits
C(ring)–C(ring)–C(ring)–C(radical) torsion angles > 175°, while
the C(ring)–C(ring)–C(radical)–C(methylene) torsion angles are
> 170°. Moreover, the H-C-C-H bonds are effectively staggered
throughout the structure. The energetically dominant entropic
change presumably arises mostly from loss of degrees of
freedom associated with cyclization, including the torsion angles
mentioned. Despite residing in a significant potential energy well,
none of the furan-based bis(p-quinodimethane) intermediates,
such as 13c-Z,Z, were observed (low temperature NMR) and
oligomeric by-products were not found in the product mixtures.
This suggests the subsequent reaction steps occur rapidly in the
gas-phase.
Table 4. Relative energies (kJ mol–1
conversion of 8c to 11c (plus CH3CO2H).
, MN12SX/6-311+G(d,p)) for the
8c 13c-Z,Z
14
14ts 11c (cis)
+2 CH3CO2H
E, 0K
H298
G298
H853
G853
0
0
0
0
0
89
89
104
107
31
153
155
82
–42
–47
The final reaction step leads to the formation of 11c (cis)
by transannular hydrogen atom abstraction. The transition state
14ts for this process was located through a combination of
potential energy surface scans and optimizations and it exhibits
"folding" of the macrocycle to lessen the distance the hydrogen
atom travels (Figure 3). Specifically, the optimized structure of
–10
84
–123
–59
99
144
–47
–192
–106
–257
Figure 3. Optimized structures for the transformation of diradical 14 to the
alkene 11c (cis) via transition state 14ts (UMN12SX/6-311+G(d,p) level).
14 shows a distance of 2.66 Å between the radical center and
the reactive hydrogen – well beyond the sum of covalent atomic
radii (1.04 Å) but within the sum of van der Waals radii (2.9 Å).
Folding the ring (and stretching the C–H bond) to reach the
transition state shortens this to 1.68 Å. The pairing of folding and
hydrogen transfer motions makes it difficult to characterize the
transition state as early or late from a dynamic perspective; that
said, one sees that the breaking carbon-hydrogen bond is 1.217
Å at the transition state, elongated by only ca. 0.1 Å from the
equilibrium value. Employing the methodology of Manz and
Sholl,18 we find y = 0.20, characteristic of an early transition state.
The barrier for this step is predicted to remain in the range of
50–60 kJ mol–1 (Table 4) as the reaction temperature increases
from 0-853 K. These represent only modest additional energy
inputs compared to those required to form 14 (Figure 2).
The hydrogen atom abstraction leads to a pair of new,
stable bonds, a C-H -bond and C-C -bond. As such, this
contributes to a strongly exergonic step (G853 = –294 kJ mol–1)
from the diradical 14. Overall, the formation of 11c (cis) from 8c
is favorable by 269 kJ mol–1, much of which is due to the
elimination of two acetic acid molecules at the elevated pyrolysis
temperature. The conversion of 8c to 11c (cis) occurs by a total
Figure 2. Graphical representation of relative energies (MN12SX/6-
311+G(d,p) level, kJ mol–1).
Following
generation
of
the
furan-based
bis(p-
quinodimethane) intermediate 13c–Z,Z, cyclization gives the
singlet diradical intermediate 14. Calculations reveal this to be
an energetically unfavorable step at all temperatures, despite
the formation of two aromatic furan rings (Figure 2). At 853K,
the free energy change for the step is +86 kJ mol–1, which can
be partitioned as +15 kJ mol–1 of enthalpic change (Table 4,
H853 values) and +71 kJ mol–1 of entropic change,
corresponding to an entropy loss of –83 J/mol • K. The
unfavorable enthalpy change appears to arise mostly from the
formation of an open-shell diradical. It is conceivable that
This article is protected by copyright. All rights reserved

European Journal of Organic Chemistry
10.1002/ejoc.201600995
FULL PAPER
of eight independent reaction steps: the elimination of each
acetic acid to generate the furan-based p-quinodimethanes
involves three steps each and the cyclization is a two-step
process proceeding through the macrocyclic diradical.
The transformation of 8b to 10 was also examined by DFT
calculations. This study sought to compare the viability of a
mechanism involving the furan-based bis(p-quinodimethane)
13b and an allowed 14-electron sigmatropic rearrangement
versus one proceeding through the 1,4-singlet diradical 15
at the experimental temperature of 853 K. Nevertheless, the
G853 values for the transition states differ so little as to be
indistinguishable. As all other steps are exothermic/exergonic,
the computational modeling does not allow differentiation
between the two mechanisms. Consequently, DFT calculations
suggest both mechanistic pathways are viable for conversion of
furan-based bis(p-quinodimethane) 13b to the observed product
10.
Regarding products 9a-f, it is not exactly clear how these
compounds are formed from the starting diesters. In most cases,
they are observed as minor byproducts, but with 9a it is the
dominant product (formed in 33% isolated yield from 8a). It is
plausible that they arise from radical homolysis and/or
disproportionation reactions.
(Schemes 7 and 9).
As above, the furan-based bis(p-
quinodimethane) 13b-Z,Z proved 5-6 kJ mol–1 more stable than
the (E,E) and (E,Z) stereoisomers. Scans of the potential
energy surfaces associated with cyclizing the three
stereoisomers of 13b suggested several possible transition state
structures. Energetic and wavefunction stability considerations
suggested plausible diradical and sigmatropic pathways. These
are shown in Scheme 9, along with E(0 K) data. The top process
is the sigmatropic rearrangement, while the bottom one is the
biradical process. Complete energetic data appear in Table 5
while a full description of the computational methods is found in
the Supporting Information section.
Conclusions
In summary, we showed that furan-based bis(p-
quinodimethane) structures may be generated by FVP of
suitable heterocyclic esters.
The furan-based bis(p-
Scheme 9. The 14 electron sigmatropic and singlet diradical pathways for the
formation of 10 from the furan-based bis(p-quinodimethane) 13b.
quinodimethanes) are shown to form macrocyclic products by
reaction of the tethered furan-based p-quinodimethanes. This
represents a rare example of macrocyclic synthesis based on
gas-phase chemistry. In accord with previous studies involving
p-quinodimethanes, the observed products are consistent with
formation of diradical intermediates. Theoretical studies have
demonstrated
a
strong thermodynamic driving force for
elimination of the carboxylic acid molecule (to form the p-
quinodimethane) at elevated temperatures as the key driving
force for the reactions. It is also shown that formation of the
macrocyclic diradical is an energy intensive step, but radical
disproportionation leads to a sizable stabilization of the system,
providing the final macrocyclic product. Additionally, minor
products were observed in most of the FVP experiments –
acyclic aldehydes 9a-f – which appear to be formed by a novel
bond homolysis cascade.19
Table 5. Relative energies (kJ mol–1, MN12SX/6-311+G(d,p)) for the two
pathways for conversion of 13b to 10.
Experimental Section
13b- 13b-ts1
Z,Z
15
30
13b- 13b-ts2
E,E
10
Some general methods have been described previously.14c
Gas
E,
0
86
6
77
–75
chromatographic analysis was performed on a gas chromatograph (GC),
employing a 30 meter DB-1 capillary column, helium carrier gas, and
flame ionization detector (FID). Elemental analyses were done by a
commercial analytical lab and accurate mass MS data was obtained from
a time-of-flight MS instrument operating in EI mode. For the FVP product
mixtures, yield percentages were calculated from GC integration by
comparison to an internal standard. Phenyl ether was used as the
internal standard and the FID detector response factor was previously
determined by preparing a solution of measured amounts of phenyl ether
and 11c (cis). The isolated yield of 11c (cis) was found to be
comparable to the yield calculated from GC analysis. When flash
chromatography was used to purify products, standard methods were
used with 230-400 mesh silica gel.20
0K
H298
G298
H853
G853
0
0
0
0
81
109
77
25
51
6
14
6
71
103
66
–76
–61
–78
–33
23
101
163
29
167
Based on the calculated E (0K) and H values, the
sigmatropic rearrangement is somewhat favored kinetically, as
gauged by the 10 kJ mol–1 greater stability for concerted
transition state (13b-ts2) versus the stepwise transition state
(13b-ts1). However with entropy considered, the diradical
pathway becomes more favorable at increasing temperature.
The stepwise, diradical pathway (via 13b-ts1) is slightly favored
This article is protected by copyright. All rights reserved

European Journal of Organic Chemistry
10.1002/ejoc.201600995
FULL PAPER
Computational Methods. Reactants, intermediates, and products were
fully optimized without constraints initially at either the HF/6-31+G(d,p)
level or the M06-2X/6-31+G(d,p) level, and confirmed to be minima (no
imaginary frequencies) or transition states (1 imaginary frequency) using
the Gaussian 09 suite.21 When appropriate, wavefunctions were tested
for stability with respect to diradical behaviour using the Stable keyword.
Singlet diradicals were treated using the "broken symmetry" approach,
which employs unrestricted "mixed" wavefunctions for singlets. Test
optimizations indicated that this approach gave structures with SCF
15% NaOH, and then 1.8 mL of H2O. The mixture is stirred at 25 °C for
an additional 20 min. and ~2 g of MgSO4 is then added to the solution.
Vacuum filtration removes the precipitated salts, the reaction flask is then
rinsed with 20 mL of ethyl acetate, and this solution is poured through the
filtered salts. Removal of the solvent then provides crude diol.
The resulting diol is not isolated/purified, but dissolved in THF (20 mL).
To the diol (10 mmol) solution, triethylamine (3.5 mL, 25 mmol) and
acetyl chloride (1.8 mL, 25 mmol) is added and the solution stirred at
25 °C. The mixture is allowed to react for at least 5 h, during which the
triethylammonium chloride precipitates. The progress of the reaction is
monitored by thin layer chromatography (eluent, 4:1 hexanes:ethyl
acetate). Upon completion of the reaction, the product mixture is poured
into a separatory funnel containing 50 mL of 1.0 M HCl and 25 mL of
ethyl ether. The acidic extraction was then followed by extraction with
saturated NaHCO3 and saturated NaCl. Drying with MgSO4 and removal
of the solvent gave the crude diacetate 8. The diacetate is then purified
by flash chromatography using 9:1 hexanes:ethyl acetate as the eluent.
From the -di(2-formyl-5-furyl)alkanes (7), typical yields of diacetates
8a-g are 80-91%.
energies within
a
few kJ mol–1 of those from more demanding
CASSCF(6,6) calculations, so the approach was considered appropriate.
Structures were further optimized at the (u)M06-2X/6-311+G(d,p) and
(u)MN12SX/6-311+G(d,p) DFT levels,22,23 with frequency analyses
performed at the former level to provide more accurate thermal
corrections and to confirm that the structures remained minima after
inclusion of DFT-based correlation. We employed multiple model
chemistries (see Supporting Information) to provide thorough analyses of
the structures and to provide a measure of computational variance. We
report here only results from the MN12SX model for clarity and because
it has been shown to perform exceptionally for a range of test sets
encompassing molecules and solid-state materials.24 Electronic energies
for all structures were corrected for zero point energy using data from the
HF/6-31+G(d,p) frequency calculations with the zero point energies
scaled by a factor of 0.93.25 Enthalpies and free energies were corrected
using unscaled thermal corrections from the (u)M06-2X/6-311+G(d,p)
frequency calculations. Complete details are available as Supporting
Information.
1,3-Di(2-acetoxymethyl-5-furyl)propane (8a). Clear oil. 1H NMR
(CDCl3, 300 MHz) , 6.28 (d, J= 3.1 Hz, 2H), 5.96 (d, J= 3.0 Hz, 2H), 4.98
(s, 4H), 2.65 (t, J= 7.5 Hz, 4H), 2.06 (s, 6H), 1.99 (p, J= 7.4 Hz, 2H). 13C
NMR (CDCl3, 75 MHz) , 170.08, 156.94, 147.14, 110.89, 105.62, 57.73,
26.89, 25.90, 20.40. IR (thin film, cm–1) 2949, 2872, 1732, 1560, 1437,
1375, 1227, 1022, 972, 791, 752. High-resolution mass spectrum, calcd.
for C17H20O6 320.1260, found 320.1254.
General procedure for the preparation of -di(2-acetoxymethyl-5-
furyl)alkanes (8a-g). (6a-g) are prepared using a published procedure
for the alkylation of furans.26 To a stirred solution at -78 °C containing
furan (2.0 g, 29 mmol) in THF (18 mL) and HMPA (2 mL), 11.6 mL BuLi
(2.5 M, 29 mmol) is added slowly. The solution is stirred at -78 °C for 1 h,
then the -dibromoalkane (13 mmol, in 5 mL THF) is added and the
mixture is allowed to warm (3 h, -78 to 20 °C). The dark solution is then
quenched with 20 mL 1.0 M HCl, and this mixture is poured into a
separatory funnel containing 50 mL of diethyl ether and 30 mL of 1.0 M
HCl. Acidic extraction is followed by washes with saturated NaHCO3,
brine, and drying with anhydrous MgSO4. Filtration and removal of ether
provides the -di(2-furyl)alkanes which is purified by distillation or by
flash chromatography using 9:1 hexanes:ethyl acetate. Typical yields for
this step are 62 to 90%.
1,4-Di(2-acetoxymethyl-5-furyl)butane (8b). White solid, mp 99-101°C
(ether:hexane). 1H NMR (CDCl3, 300 MHz) , 6.29 (d, J = 3.1 Hz, 2H),
5.95 (d. J = 3.1 Hz, 2H), 4.99 (s, 4H), 2.69-2.60 (m, 4H), 2.08 (s, 6H), 1.
73-1.67 (m, 4H). 13C NMR (CDCl3, 75 MHz) , 170.64, 157.04, 147.60,
111.45, 105.88, 58.29, 27.77, 27.36, 20.93. IR (thin film, cm-1) 2941,
2864, 1734, 1560, 1437, 1375, 1215, 1018, 974, 957, 791. Anal. calcd.
for C18H22O6: C 64.66, H 6.63. Found: C 64.70, H, 6.56.
1,5-Di(2-acetoxymethyl-5-furyl)pentane (8c). Clear oil. 1H NMR (C6D6,
300 MHz) , 6.15 (d, J= 3.1 Hz, 2H), 5.74 (d, J= 3.1 Hz, 2H), 4.94 (s, 4H),
2.34 (t, J= 7.5 Hz, 4H), 1.59 (s, 6H), 1.41–1.32 (m, 4H), 1.12–1.07 (m,
2H). 13C NMR (CDCl3, 75 MHz) , 170.75, 157.37, 147.53, 111.48,
105.79, 58.38, 28.71, 27.98, 27.59, 21.01. IR (thin film, cm–1) 2937, 2862,
1742, 1560, 1435, 1375, 1236, 1018, 974, 795. Anal. calcd. for
C19H24O6: C 65.50, H 6.94. Found: C 66.75, H, 7.42.
(7a-g) were prepared using a published procedure for the formylation of
furans.27 To a solution at -78 °C containing the -di(2-furyl)alkane (10
mmol) in THF (9 mL) and HMPA (1 mL), 8.4 mL of butyllithium (2.5 M, 21
mmol) is added slowly. The solution is stirred at -78 °C for 1 h, DMF
(0.73 g, 0.1 mol) is then added and it is allowed to warm (3 h, -78 °C to
20 °C). This solution is stirred for 12 h, then quenched with enough 10%
HCl to produce a solution of about pH 7, and added to a separatory
funnel containing 50 mL of diethyl ether and 30 mL of 10% HCI. Acidic
extraction is followed by washes with saturated NaHCO3, brine, and
drying with anhydrous MgS04. Filtration and removal of the ether
provides the which is purified by flash chromatography using 3:2
hexanes:ethyl acetate. The -di(2-formyl-5-furyl)alkanes are yellow
solids obtained in 50-70% yields.
1,6-Di(2-acetoxymethyl-5-furyl)hexane (8d).
Clear oil. 1H NMR
(CDCl3, 300 MHz) , 6.25 (d, J = 3.0 Hz, 2H), 5.90 (d. J = 3.0 Hz, 2H),
4.95 (s, 4H), 2.56 (t, J = 7.6 Hz, 4H), 2.03 (s, 6H), 1.62–1.55 (m, 4H),
1.37–1.30 (m, 4H). 13C NMR (CDCl3, 75 MHz) , 170.70, 157.48, 147.46,
111.46, 105.71, 58.34, 28.82, 28.00, 27.73, 20.95. IR (thin film, cm–1
)
2934, 2860, 1745, 1560, 1437, 1375, 1360, 1238, 1018, 974, 798. Anal.
calcd. for C20H26O6: C 66.28, H 7.23. Found: C 67.60, H, 8.22.
1,7-Di(2-acetoxymethyl-5-furyl)heptane (8e). Clear oil. 1H NMR
(CDCl3, 300 MHz) , 6.24 (d, J= 3.1 Hz, 2H), 5.89 (d, J= 3.1 Hz, 2H),
4.95 (s, 4H), 2.56 (t, J= 7.6 Hz, 4H), 2.03 (s, 6H), 1.64–1.52 (m, 4H), 1.30
(s, 6H). 13C NMR (CDCl3, 75 MHz) , 170.68, 157.57, 147.44, 111.43,
105.66, 58.34, 29.04, 28.99, 28.04, 27.81, 20.95. IR (thin film, cm–1) 2932,
2858, 1744, 1560, 1437, 1375, 1360, 1236, 1018, 972, 798. Anal. calcd.
for C21H28O6: C 67.00, H 7.50. Found: C 68.09, H, 8.03.
-Di(2-acetoxymethyl-5-furyl)alkanes (8a-g). The general procedure
used to convert the dialdehydes to the diacetates is based on published
procedures.28 A stirred mixture of LiAIH4 (0.57 g, 15 mmol) in THF (25
mL) is cooled to 0 °C, and to it is slowly added a solution of the -di-(2-
formylfuryl-5)alkane (10 mmol) in THF (10 mL). The ice bath is then
removed and the mixture was stirred for 1 h at 25 °C. The mixture is
cooled to 0 °C and 0.6 mL of H2O is slowly added, followed by 0.6 mL of
1,12-Di(2-acetoxymethyl-5-furyl)dodecane (8f). White solid, mp 43-
46 °C (ether:hexane). 1H NMR (CDCl3, 300 MHz) , 6.29 (d, J= 3.1 Hz,
This article is protected by copyright. All rights reserved

European Journal of Organic Chemistry
10.1002/ejoc.201600995
FULL PAPER
2H), 5.93 (d, J= 3.1 Hz, 2H), 4.99 (s, 4H), 2.60 (t, J= 7.7 Hz, 4H), 2.07 (s,
6H), 1.67–1.59 (m, 4H), 1.39–1.23 (m, 16H). 13C NMR (CDCl3, 75 MHz)
, 171.71, 157.72, 147.40, 111.43, 105.61, 58.37, 29.59, 29.53, 29.34,
29.20, 28.09, 27.88. IR (thin film, cm–1) 2920, 2853, 1736, 1562, 1472,
1379, 1246, 1204, 1028, 937, 798. Anal. calcd. for C26H38O6: C, 69.93, H,
8.58. Found: C, 70.02, H, 8.66.
(cis). White solid, mp 96-97 °C (ether:hexane), 46% yield. 1H NMR
(CDCl3, 300 MHz)  5.99 (d, J = 11.6 Hz, lH), 5.92 (d, J = 3.1 Hz, lH),
5.88 (d, J= 2.9, lH), 5.84 (d, J= 2.9 Hz, lH), 5.79 (d, J= 3.2 Hz, lH), 5.48
(dt, J= 8.8, 11.6 Hz, lH), 2.60-2.51 (m, 6H), 2.42-2.35 (m, 2H). 1.87–1.75
(m, 2H). 13C NMR (C6D6, 75 MHz)  155.6, 154.6, 153.8, 153.4, 129.1,
118.2, 110.4, 107.9, 107.2, 105.9, 30.2, 28.7, 28.5, 27.7, 27.6. IR 2924,
2854, 1560, 1456, 1423, 1396, 1169, 1018, 789. MS (EI) 228 (M+), 134,
133, 121, 107. High-resolution mass spectrum, calcd for C15H16O2
228.1150, found 228.1148.
1,16-Di(2-acetoxymethyl-5-furyl)hexadecane (8g). White solid, mp 56-
57°C (ether:hexane). 1H NMR (CDCl3, 300 MHz) , CDCl3 6.25 (d, J=
3.1 Hz, 2H), 5.90 (d, J= 3.1 Hz, 2H), 4.90 (s, 4H), 2.56 (t, J= 7.6 Hz, 4H),
2.04 (s, 6H), 1.63–1.54 (m, 4H), 1.34–1.20 (m, 24H). 13C NMR (CDCl3,
75 MHz) , CDCl3 171.70, 157.71, 147.36, 111.43, 105.61, 58.38, 29.67,
29.58, 29.36, 29.23, 28.11, 27.90. IR (thin film, cm–1) 2916, 2851, 1730,
1558, 1472, 1377, 1236, 1207, 1020, 966, 800. Anal. calcd. for
C30H46O6: C 71.68, H, 9.22. Found: C 71.61, H, 9.32.
FVP of 8d. FVP required heating the starting material 11d to 105°C for
volatilization. The crude mixture contained four products in a 18:4:2: 1
ratio and GCMS indicated respective m/z values of 242, 242, 242, and
260. The three most abundant products were isolated and identified as
11d (cis), 11d (trans), and 12d, while one product could not be isolated
but
presumably
was
product
9d.
cis–1,2-
l
Dehydro[6.2](2,5)furanophane, 11d (cis). Clear oil, 47% yield. H NMR
(CDCl3, 300 MHz)  5.99 (d, J= 3.3 Hz, lH), 5.97 (d, J= 3.2 Hz, lH), 5.89
(d, J= 11.7, lH), 5.80 (d, J= 2.9 Hz, lH), 5.75 (d, J= 2.9 Hz, lH), 5.42 (dt,
J= 8.3, 11.6 Hz, lH), 3.07-2.98 (m, 2H), 2.81-2.75 (m, 2H), 2.62 (t, J = 6.3,
2H), 2.30-2.20 (m, 2H), 1. 78- 1.65 (m, 2H), 1.28–1.20 (m, 2H). 13C NMR
(CDCl3, 75 MHz)  155.2, 154.1, 152.4, 152.1, 129.8, 117.0, 110.3, 106.9,
106.9, 106.4, 29.3, 28.9, 27.9, 27.5, 27.4, 27.0. IR (thin film, cm–1) 2930,
2860, 1568, 1435, 1213, 1132, 1015, 984, 783. MS m/e (EI) 242 (M+),
135, 120, 107, 94. High-resolution mass spectrum, calcd for C16H18O2
242.1307, found 242.1309. trans–1,2-Dehydro[6.2](2,5)furanophane,
11d (trans). Clear oil, 11% yield. lH NMR (CDCl3, 300 MHz)  6.02 (d,
J= 3.0 Hz, lH), 5.97 (d, J= 15.9 Hz, lH), 5.96 (d, J= 3.0, lH), 5.91 (d, J=
3.0 Hz, lH), 5.84 (d, J= 2.9 Hz, lH), 5.56 (dt, J= 7.1, 15.9 Hz, lH), 2.84 (s,
4H), 2.57-2.50 (m, 2H), 2.12-2.03 (m, 2H), 1.67–1.50 (m, 4H). 13C NMR
(CDCl3, 75 MHz)  155.1, 154.0, 153.0, 149.5, 132.2, 118.3, 107.5, 106.2,
105.4, 105.4, 31.0, 29.2, 28.7, 28.1, 25.8, 23.5. MS (EI) 242 (M+), 135,
120, 107, 77. High-resolution mass spectrum, calcd for C16H18O2
242.1307, found 242.1306. [1:2]Butano[2.2](2,5)furanophane (12d).
Clear oil, 4% yield. IR (thin film, cm–1) 2918, 2853, 1547, 1447, 1190,
1157, 1016, 1007, 787. lH NMR (CDCl3, 300 MHz)  6.10 (d, J= 3.0 Hz,
2H), 6.03 (d, J= 3.0 Hz, 2H), 2.82 (d, J= 9.4, 2H), 2.61 (d, J = 9.6, 2H),
2.43-2.37 (m, 2H), 1.95–1.84 (m, 2H), 1.70–1.55 (m, 2H), 1.43–1.33 (m,
2H). 13C NMR (CDCl3, 75 MHz)  161.2, 156.0, 107.9, 105.3, 48.7, 30.7,
28.4, 26.0. IR (thin film, cm–1) 2918, 2853, 1547, 1447, 1190, 1157, 1016,
1007, 787. MS (EI) 242 (M+), 135, 120, 107, 94. High-resolution mass
spectrum, calcd for C16H18O2 242.1307, found 242.1311.
General procedure for FVP of the diester substrates 8a-g. The FVP
apparatus is based on a published design.14a The heated zone is first
allowed to equilibrate to the desired FVP temperature. The appropriate
diester (ca. 0.1-0.3 g) is placed in the sample chamber (see Supporting
Information) and the apparatus evacuated to 10-5 to 10-6 Torr. The FVP
trap is cooled with liquid nitrogen (-196 °C) and the sample chamber is
then heated. Products deposit in the trap as white or yellow solids.
Upon complete vaporization of the sample (some residue may remain),
dry CHCl3 or CDCl3 (ca. 10 mL) is distilled into the trap from the side inlet.
The FVP apparatus is filled with argon gas and the trap is removed. The
trap is allowed to warm and anhydrous Na2CO3 (0.5 g) is added to the
product mixture. The mixture is then filtered and the products are
isolated.
FVP of 8a. FVP required heating the starting material 8a to 70 °C for
volatilization. Product composition was somewhat variable, but one major
product was observed by GCMS to have m/z of 218. This product was
isolated and identified as 9a. 1-(2-Methyl-5-furyl)-3-(2-formyl-5-
furyl)propane (9a). Clear oil, 33% yield. 1H NMR (CDCl3, 300 MHz) 
9.50 (s, 1H), 7.15 (d, J= 3.5, 1H), 6.23 (d, J= 3.5 Hz, 1H), 5.85 (s, 1H),
5.82 (s, 1H), 2.75 (t, J= 7.6, 2H), 2.62 (t, J= 7.3, 2H), 2.23 (s, 3H), 2.01 (p,
J = 7.5, 2H). 13C NMR (CDCl3, 75 MHz)  176.8, 163.2, 152.9, 151.7,
150.4, 123.2, 108.8, 105.9, 105.7, 27.5, 27.2, 26.0, 13.4. IR (thin film,
cm–1) 2950, 1677, 1516, 1020, 784. MS (EI) 218 (M+), 123, 122, 109,
108, 107, 96, 95, 81, 53. High-resolution mass spectrum, calcd for
C13H14O3 218.0943, found 218.0943.
FVP of 8e. FVP required heating the starting material 11e to 105 °C for
volatilization. The crude mixture contained three products in a 25:18:1
ratio and GCMS indicated respective m/z values of 256, 256, and 274.
The two most abundant products were isolated and identified as 11e
(trans), and 11e (cis), while the minor product could not be isolated, but
was likely 9e. cis–1,2-Dehydro[7.2](2,5)furanophane, 11e (cis). Clear
oil, 22% yield. lH NMR (CDCl3, 300 MHz)  6.02 (d. J = 3.1 Hz, lH), 6.00
(d, J = 3.1 Hz, lH), 5.92 (d. J = 11.7 Hz, lH), 5.89 (d, J= 11.7, lH), 5.76 (d.
J = 2.9 Hz, lH), 5.71 (d, J = 2.8 Hz, lH), 5.29 (dt, J = 8.1, 11.7 Hz, lH),
3.01-2.93 (m, 2H), 2.85-2.79 (m, 2H), 2.56 (t, J= 6.0 Hz, 2H), 2.32 (dd, J=
6.9, 6.9, 2H), 1.69–1.57 (m, 2H). 1.42–1.31 (m, 4H). 13C NMR (CDCl3. 75
MHz)  155.0, 152.4, 152.0, 130.0, 117.9, 110.3, 106.4, 106.3, 105.7,
29.3, 28.2, 28.0, 28.0, 27.8, 27.3, 27.0. IR (thin film, cm–1) 2920, 2857,
1585, 1429, 1013, 980, 779. MS m/e (EI) 256 (M+), 120, 107, 94, 91.
High-resolution mass spectrum, calcd for C17H20O2 256.1463, found
256.1463. trans–1,2-Dehydro[7.2](2,5)furanophane, 11e (trans).
FVP of 8b. FVP required heating the starting material 8b to 90 °C for
volatilization. The crude mixture contained four products in a 43:10:4:1
ratio and GCMS indicated respective m/z values of 214, 232, 214, and
214. The most abundant product was isolated and identified as 1,2-di(2-
vinylfuryl-5)ethane (10), the second most abundant product was thought
to be 9b, and the two minor products could not be isolated and identified.
1,2-Di(2-vlnyl-5-furyl)ethane (10). Clear oil, 10% yield. lH NMR (CDCl3,
300 MHz)  6.40 (dd, J= 11.3, 17.5 Hz, 2H), 6.10 (d, J= 3.2, 2H), 5.96 (d,
J= 3.2 Hz, 2H), 5.54 (dd, J= 1.2, 17.5 Hz, 2H), 5.04 (dd, J = 1.3, 11.3 Hz,
2H), 2.95 (s, 4H); 13C NMR (CDCl3, 75 MHz)  154.5, 151.8, 125.0, 110.9,
109.0, 108.9, 107.1, 26.9. IR (thin film, cm–1) 2918, 1678, 1639, 1585,
1528, 1254, 1032, 1018, 897, 785. MS (EI) 214 (M+), 108, 107, 77, 55.
High-resolution mass spectrum, calcd for C14H14O2 214.0994, found
214.0997.
FVP of 8c. FVP required heating the starting material 11c to 100°C for
volatilization. The crude mixture contained three products in a 17:2:1
ratio and GCMS indicated respective m/z values of 228, 228, and 246.
The most abundant product was isolated and identified as 11c (cis), the
product of m/z 246 was thought to be 9c, while the other product could
not be isolated/identified. cis–1,2-Dehydro[5,2](2,5)furanophane, 11c
l
Clear oil, 17% yield. H NMR (CDCl3, 300 MHz)  5.96 to 5.87 (m, 5H),
5.83 (d, J= 3.0 Hz, lH), 2.94- 35 2.83 (m, 4H), 2.54 (t, J = 6.1 Hz, 2H),
2.10 (dd, J = 6.0, 5. 7 Hz, 2H), 1. 70–1.61 (m, 2H), 1.49–1.41 (m, 4H).
13C NMR (CDCl3, 75 MHz)  154.1, 153.9, 129.9, 117.9, 106.5, 106.1,
105.8, 105.5, 31.1, 27.9, 27.7, 27.2, 27.0, 26.5, 24.6. IR (thin film, cm–1
)
This article is protected by copyright. All rights reserved

European Journal of Organic Chemistry
10.1002/ejoc.201600995
FULL PAPER
2920, 2857, 1585, 1429, 1013, 980, 779. MS (EI) 256 (M+), 120, 107, 94,
91. High-resolution mass spectrum, calcd for C17H20O2 256.1463, found
256.1467.
J. Phys. Chem. A 2013, 117, 4709-4715. (d) M. Bobrowski, P. Skurski,
S. Sylwia Chem. Phys. 2011, 382, 20-26.
[5]
[6]
H.-Y. Parker, C. E. Klopfenstein, R. A. Wielesek, T. Koenig, J. Am.
Chem. Soc. 1985, 107, 5276-5277.
(a) Y. Ito, S. Miyata, M. Nakatsuka, T. Saegusa, J. Org. Chem. 1981,
46, 1044-1045. (b) H. Amii, Y. Hatamoto, M. Seo, K. Uneyama, J. Org.
Chem. 2001, 66, 7216-7218.
FVP of 8f. FVP required heating the starting material 8f to 130 °C for
volatilization. The crude mixture contained two major products in a 4:1
ratio and GCMS indicated respective m/z values of 326 and 326. The
most abundant product was isolated and identified as 11f (trans), while a
partial characterization of the other product suggested its identity as 11f
(cis). In addition, a minor product was observed by GCMS which had a
m/z value of 344. This product was thought to be 9f. trans–1,2-
Dehydro[12.2](2,5)furanophane, 11f (trans). Clear oil, 19% yield. lH
NMR (CDCl3, 300 MHz)  6.05 (d, J = 16.1, lH), 5.96-5.86 (m, 3H), 5.80
(d. J = 3.0 Hz, lH), 5. 75 (d, J = 2.9 Hz, lH), 2.99-2.84 (m, 4H), 2.47 (t, J=
7.1 Hz, 2H), 2.14 (dd, J= 6.0, 5.7 Hz, 2H), 1.53–1.12 (m, 16H). 13C NMR
(CDCl3, 75 MHz)  154.6, 153.7, 152.6, 151.7, 128.6, 119.2, 106.8, 106.5,
105.7, 105.1, 31.1, 28.0, 27.6, 27.5, 27.2, 27.2, 26.6, 26.1. MS (EI) 326
(M+), 121, 108, 107, 95. IR (thin film, cm–1) 2926, 2854, 1568, 1533,
1460, 1437, 1281, 1092, 1015, 960, 800, 775. High-resolution mass
spectrum, calcd for C22H30O2 326.2247, found 326.2246. cis–1,2-
[7]
[8]
P. S. Hammond, D. T. Longone, Tetrahedron Lett. 1978, 5, 415-418.
M. A. Miranda, M. J. Pérez-Prieto, E. Font-Sanchis, J. C. Scaiano,
Chem. Commun. 1998, 1541-1542.
[9]
(a) W. F. Gorman, J. Polymer Sci., Part A: Polym. Chem. 1966, 4,
3027-3029. (b) S. Shimizu, N. Nakashima, Y. Sakata, Chem. Phys.
Lett. 1998, 284, 396-400.
[10] J. H. P. Utley, J. Gruber, J. Mater, Chem. 2002, 12, 1613-1624.
[11] (a) H. G. Gilch, W. L. Wheelwright, J. Polym. Sci., Part A 1966, 4,
1337-1341. (b) T. Schwalm, J. Wiesecke, S. Immel, M. Rehahn,
Macromolecules 2007, 40, 8842-8854.
[12] (a) R. A. Wessling, R. G. Zimmerman, Polyxylyidene Articles. U.S.
Patent 3,706,677, December 19, 1972. (b) M. Kanbe, M. Okawara, J.
Polymer Sci., Part A: Polym. Chem. 1968, 6, 1058-1062.
l
[13] (a) D. J. Williams, J. M. Pearson, M. Levy, J. Am. Chem. Soc. 1970, 92,
1436-1438. (b) J. M. Pearson, H. A. Six, D. J. Williams, M. Levy, J. Am.
Chem. Soc. 1971, 93, 5034-5036. (c) W. S. Trahanovsky, S. P. Lorimor, J.
Org. Chem. 2006, 71, 1784-1794.
Dehydro[12.2](2,5)furanophane, 11f (cis). Clear oil, ~ 4% yield. H
NMR (CDCl3, 300 MHz)  6.05-5.95 (m), 5.86 (d), 5.86-5.75 (m), 2.92 (s),
2.55 (t), 2.23 (d), 1.30 (s). MS (EI) 326 (M+), 133, 121, 108, 107, 95, 94,
55.
[14] (a) W. S. Trahanovsky, D. L. Miller, Y. Wang, J. Org. Chem. 1997, 62,
8980-8986. (b) P. S. Chen, C.-H. Chou, J. Chin. Chem. Soc. (Taipei)
1992, 39, 251-255. (c) W. S. Trahanovsky, M.-G Park,. J. Org. Chem.
1974, 39, 1448.
FVP of 8g. FVP required heating the starting material 8g to 150°C for
volatilization. The crude mixture contained no products such as 11g or
9g. The GCMS data suggested that considerable fragmentation occurred
during FVP.
[15] H. E. Winberg, F. S. Fawcett, W. E. Mochel, C. W. Theobald, J. Am.
Chem. Soc. 1959, 82, 1428-1435.
[16] (a) S. G. Cohen, R. J. Zand, J. Am. Chem. Soc. 1962, 84, 586-591. (b)
P. S. Engel, R. A. Hayes, L. Kiefer, S. Szilagyi, J. W. Timberlake, J. Am.
Chem. Soc. 1978, 100, 1876-1883.
[17] (a) S. Sharma, T. Rajale, D. B. Cordes, F. Hung-Low, D. M. Birney, J.
Am. Chem. Soc. 2013, 135, 14438-14447. (b) H. Ji, X. Xu, S. Ham, L.
A. Hammad, D. M. Birney, J. Am. Chem. Soc. 2009, 131, 528-537.
[18] T. A. Manz, D. S. Sholl, J. Comput. Chem. 2010, 31, 15628-15634.
[19] This study is based in part on the Ph.D. Dissertation of D. A. Klumpp,
Iowa State University, 1993. A related conversion involving thiophene-
based bis(p-quinodimethanes) was reported recently, see: W. S.
Trahanovsky, D. A. Klumpp, Tetrahedron Lett. 2016, in press.
[20] Still, W. C.; Kahn, M.; Mitra. A J. Org. Chern. 1978, 43, 2923.
[21] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb,
J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A.
Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F.
Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K.
Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O.
Kitao, H. Nakai, T. Vreven, J. A. Montgomery, Jr., J. E. Peralta, F.
Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N.
Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J.
C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, N. J. Millam, M.
Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R.
Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C.
Pomelli, J. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G.
A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö.
Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, D. J. Fox, Gaussian
09, Revision D.01, Gaussian, Inc., Wallingford CT, 2009.
Acknowledgements
This work was supported in part by the U.S. Department of
Energy, Office of Basic Energy Sciences, Division of Chemical
Sciences, under Contract W-7405-ENG-82 and the National
Science Foundation (1300878). We gratefully acknowledge this
support. We also thank the reviewers of this manuscript for
their comments and suggestions.
Keywords: pyrolysis • furan • quinonedimethane • marcocyle •
diradical
[1]
[2]
[3]
(a) M. Szwarc, Nature 1947, 160, 403. (b) M. Szwarc,. J. Chem. Phys.
1947, 16, 128-136.
(a) H. Sano, J. Nishimura, Sci. Synth. 2009, 45a, 483-506. (b) L. A.
Errede, M. Szwarc, Quart. Revs. (London) 1958, 12, 301-320.
(a) W. F. Beach, “Xylylene Polymers” in Encyclopedia of Polymer
Science and Technology (4th Ed.), (Ed.: H. F. Mark), Wiley, Hoboken,
N.J., 2014; pp. 409-448.
(b) L. Hermosilla, S. Catak, V. Van
Speybroeck, M. Waroquier, J. Vandenbergh, F. Motmans, P.
Adriaensens, L. Lutsen, T. Cleij, D. Vanderzande, Macromolecules
2010, 43, 7424-7433. (c) T. Itoh, Sci. Synth. 2009, 45b, 697-722. (d)
T. Schwalm, J. Wiesecke, S. Immel, M. Rehahn, Macromolecules 2007,
40, 8842-8854.
[22] Y. Zhao, D. G. Truhlar, Theor. Chem. Acc. 2008, 120, 215-241.
[23] R. Peverati, D. G. Truhlar, Phys. Chem. Chem. Phys. 2012, 14, 16187-
16191.
[24] R. Peverati, D. G. Truhlar, Phil. Trans. R. Soc. A 2014 372: 20120476
(1–52).
[4}
(a) K. Fukuda, T. Nozawa, H. Yotsuyanagi, M. Ichinohe, A. Sekiguchi,
M. Nakano, J. Phys. Chem. C 2015, 119, 1188-1193. (b) G. Trinquier,
J.-P. Malrieu, Chem. Eur. J. 2015, 21, 814-828. (c) M. de Wergifosse,
F. Wautelet, B. Champagne, R. Kishi, K. Fukuda, H. Matsui, M. Nakano,
[25] (a) A. P. Scott, L. Radom, J. Phys. Chem. 1996, 100, 16502-16513. (b)
I. M. Alecu, J. Zheng, Y. Zhao, D. G. Truhlar, J. Chem. Theory Comput.
2010, 6, 2872-2887.
This article is protected by copyright. All rights reserved

European Journal of Organic Chemistry
10.1002/ejoc.201600995
FULL PAPER
[26] M. J. Arco, M. H. Trammell, J. D. White, J. Org. Chem. 1976, 41, 2075-
2083.
Organic Syntheses, 2nd Ed. (Ed: H. E. Baumgarten), Wiley, New York,
1973, Coll. Vol. V, pp. 258-263.
[27] B. L. Feringa, R. Hulst, R. Rikers, L. Brandsma, Synthesis 1988, 316.
[28] (a) L. F. Fieser, M. Fieser, Reagents for Organic Synthesis, Wiley, New
York, 1967, p. 584. (b) C. J. Blankley, F. J. Sauter, H. O. House, in
This article is protected by copyright. All rights reserved

European Journal of Organic Chemistry
10.1002/ejoc.201600995
FULL PAPER
Entry for the Table of Contents (Please choose one layout)
FULL PAPER
Reactive Intermediates
Douglas A. Klumpp*, Thomas M. Gilbert,
and Walter S. Trahanovsky*
Page No. – Page No.
Intramolecular Reactions of Tethered
Furan-based bis-(p-
Flash vacuum pyrolysis (FVP) is used to generate furan-based bis-(p-
quinodimethanes and a gas-phase macrocyclization occurs to give cyclophane
products. A mechanism is proposed involving cyclization of the furan-based bis-(p-
quinodimethane, formation of cyclic diradicals, and transannular hydrogen atom
migration. The chemistry is also studied by computational methods, revealing the
importance of entropy effects in the high-temperature, gas-phase chemistry.
Quinodimethanes)
This article is protected by copyright. All rights reserved