Modulating chemical reactivity using a photoresponsive molecular switch

Article abstract of DOI:10.1016/j.tet.2008.05.050

The kinetics of an alkylation reaction are used to probe the effect of an electron-withdrawing pyridinium group on the nucleophilicity of a free pyridine in a photoresponsive dithienylcyclopentene (DTCP) derivative in order to demonstrate effective and reversible control over chemical reactivity. The kinetic data support the hypothesis that the ring-open isomer of the DTCP (1o) is more reactive than its ring-closed counterpart (1c) due to electronic communication between the two pyridine groups existing only in the latter isomer. The rates of the alkylation reactions of bis(pyridine) versions of the photochromic compounds are also evaluated to provide a better understanding of the through-bond and the through-space effects of the groups located at the ends of the linearly conjugated π-electron backbone. The use of the two DTCP isomers (1o and 1c) as nucleophilic catalysts is also suggested in preliminary investigations.

Full text of DOI:10.1016/j.tet.2008.05.050

Tetrahedron 64 (2008) 8292–8300
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Modulating chemical reactivity using a photoresponsive molecular switch
*
Hema D. Samachetty, Vincent Lemieux, Neil R. Branda
4D LABS, Department of Chemistry, Simon Fraser University, 8888 University Drive, Burnaby, BC, Canada V5A 1S6
a r t i c l e i n f o
a b s t r a c t
Article history:
The kinetics of an alkylation reaction are used to probe the effect of an electron-withdrawing pyridinium
group on the nucleophilicity of a free pyridine in a photoresponsive dithienylcyclopentene (DTCP) de-
rivative in order to demonstrate effective and reversible control over chemical reactivity. The kinetic data
support the hypothesis that the ring-open isomer of the DTCP (1o) is more reactive than its ring-closed
counterpart (1c) due to electronic communication between the two pyridine groups existing only in the
latter isomer. The rates of the alkylation reactions of bis(pyridine) versions of the photochromic com-
pounds are also evaluated to provide a better understanding of the through-bond and the through-space
Received 10 March 2008
Received in revised form 6 May 2008
Accepted 9 May 2008
Available online 14 May 2008
effects of the groups located at the ends of the linearly conjugated
p-electron backbone. The use of the
two DTCP isomers (1o and 1c) as nucleophilic catalysts is also suggested in preliminary investigations.
Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction
this photoresponsive architecture is the fact that the ring-open (A)
and ring-closed (B) isomers tend to exhibit excellent thermal sta-
The incorporation of photoresponsive molecular backbones into
chemical reagents and catalysts offers the possibility to turn a re-
action ‘on’ and ‘off’ on command using light as an external stimulus.
This approach to introduce regulation has the potential to advance
synthetic methods and polymerization techniques as well as the
controlled delivery of biochemical reagents. At the beginning of the
new millennium, catalysis-based chemical synthesis accounted for
approximately 60% of the current chemical products and 90% of the
chemical processes1 representing a rich area to introduce control
at the molecular level. In the life-science realm, the controlled
delivery of biochemical reagents to induce an in vitro or in vivo
cellular response continues to be a versatile and important experi-
mental technique.2 Site-selective drug-delivery will also benefit
from the regulation of molecular structure and function by con-
verting a therapeutic from its initially inactive (or significantly less
active) form to its active form after it reaches the target tissue.3 This
unmasking can be achieved through processes such as metabolism,
hydrolysis or by using an external trigger such as light, assuming
photoresponsive architectures are incorporated into the molecular
design.4
bility across a wide range of temperatures providing potential
candidates for use as the ‘on’ and ‘off’ functions to start and stop
chemical reactions using light. Several research groups have dem-
onstrated this potential use of dithienylethene derivatives by taking
advantage of the geometric and electronic changes to regulate
substrate binding affinities,6 the stereochemical outcome of a cat-
alytic reaction,7 relative acidity,8 retro-Diels–Alder reactions9 and
the Bergmann cyclization.10
F F
F F
F
F
F
F
F
F
F
F
UV
visible
lone pair
EWG
S
S
S
S
lone pair
EWG
A
B
electronically insulated
more reactive lone pair
electronically connected
less reactive lone pair
Scheme 1. The reversible photocyclization reaction of the dithienylcyclopentene
backbone.
Scheme 1 also illustrates a specific approach to regulate chem-
ical reactivity, in this case nucleophilicity. The two thiophenes in
the ring-open isomer A are electronically insulated from each other
such that any nucleophilic lone-pair electrons located on one side
of the molecular backbone will not sense the electronic effects of an
electron-withdrawing group located on the other side. Photo-
cyclization of A to the ring-closed isomer (B) creates a linearly
Photoresponsive compounds containing the dithienylcyclo-
pentene (DTCP) backbone are particularly well suited to modulate
chemical reactivity because they undergo rapid and reversible
cyclization reactions, when exposed to UV and visible light, be-
tween two isomers (A and B in Scheme 1) that have distinct geo-
metric and electronic properties.5 Another appealing property of
conjugated p-electron pathway connecting the two groups (shown
in bold in Scheme 1), allowing the lone-pair electrons to be sub-
jected to the electronic ‘pull’ of the electron-withdrawing group,
* Corresponding author. Tel.: þ1 778 782 8061; fax: þ1 778 782 3675.
E-mail address: nbranda@sfu.ca (N.R. Branda).
0040-4020/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2008.05.050

H.D. Samachetty et al. / Tetrahedron 64 (2008) 8292–8300
8293
thus lowering their nucleophilic strength. Consequently, the ring-
2. Results and discussion
open isomer A is expected to react faster with an electrophile than
its ring-closed counterpart B. We have already applied this strategy
to photoregulate the coordinating ability of a pyridine ring deco-
rated onto a DTCP backbone and have shown that ring-open isomer
1o (Scheme 2) forms a more stable coordination complex with
a ruthenium porphyrin than ring-closed counterpart 1c.11
In the current paper, we use the kinetics of an alkylation re-
action to probe the effect of the electron-withdrawing pyridinium
group on the nucleophilicity of the free pyridine in photoresponsive
dithienylcyclopentene (DTCP) 1o and demonstrate effective and
reversible control over chemical reactivity. This pyridine compound
is particularly pertinent owing to the ubiquitous role of pyridines as
nucleophilic catalysts in synthesis and biochemical processes.12 The
focus of this paper is the model alkylation reaction of both DTCP
isomers with 4-bromobenzyl bromide to produce derivatives 2o
and 2c as also shown in Scheme 2. Through an examination of the
rates of this model reaction, the difference in chemical reactivity of
this interesting compound is demonstrated.
2.1. Synthesis of the monobenzylated DTCP derivatives
1o and 1c
The monobenzylated DTCP 1o is prepared in four steps as
shown in Scheme 3 via the ring-closed isomer of the bis(pyridine)
4o by lithiating the known pyridylthiophene 313 followed by
quenching with octafluorocyclopentene. The direct treatment of
the bis(pyridine)’s ring-open isomer (4o) with 1 mol equiv of
4-bromobenzyl bromide yields an unacceptable amount of the
undesired dibenzylated DTCP (2o) as a side product. Higher yields
of the monobenzylated derivative can be obtained by alkylating the
ring-closed bis(pyridine) 4c, followed by ring-opening of 1c to 1o
using visible light. This observation already suggests that the two
pyridines in the ring-open isomer (4o) react independently from
each other while they communicate to a larger extent when the
DTCP backbone exists in its ring-closed form. This claim will be
further substantiated later in this paper.
F
F
F F
F
F
F
F
Br
F
F
F
F
Br
faster reaction?
S
S
S
S
N
N
N
N
1o
2o
stronger
coordinating
ligand
365 nm > 490 nm
Br
Br
Br
F
F
F F
F
F
F
F
Br
F
F
F
F
Br
slower reaction?
S
S
S
S
N
N
N
N
1c
2c
weaker
coordinating
ligand
Br
Br
Br
Scheme 2. Reversible photocyclization modulates the ability of the pyridine in 1o and 1c to act as a coordinating ligand. The alkylations of 1o and 1c to generate 2o and 2c are the
model reactions described in this report.
F
F
F
F
F
F
F
F
F
F
F
F
Br
CH3
1) n-BuLi, Et2O
312 nm
CH3CN
F
F
2)
S
N
F
F
S
S
S
S
N
N
F
F
3
4o
N
N
4c
(98 %)
(60 %)
F
F
F
F
F
F
F
F
F
F
Br
F
F
F
F
Br
> 490 nm
CH3CN
CH3CN
S
S
S
S
N
N
N
N
1o
(quant)
1c
(45 %)
Br
Br
Scheme 3. Synthesis of ring-closed and ring-open isomers 1c and 1o. The counterions are NO3ꢀ
.

8294
H.D. Samachetty et al. / Tetrahedron 64 (2008) 8292–8300
The ring-closed isomer 4c is generated by irradiating an anhy-
2.3. Differences in reactivity of monocations 1o and 1c
drous CH3CN solution (5.8ꢁ10ꢀ3 M) of bis(pyridine) 4o with
312 nm light14 and monitoring aliquot amounts of the solution by
1H NMR spectroscopy until no further changes are observed. The
progress of the photocyclization reaction is best assessed by
monitoring the peak corresponding to the methine C–H of the
thiophene rings, which shifts from 7.63 to 7.02 ppm as the solution
of 4o is converted into one containing 98% of the ring-closed isomer
4c. This photostationary state can then be alkylated with
1.0 mol equiv of 4-bromobenzyl bromide to afford monobenzylated
bis(pyridinium) 1c after purification by column chromatography
(silica gel, CH3CN/H2O/saturated KNO3) as its nitrate salt. The ring-
open isomer 1o of the monobenzylated bis(pyridinium) is finally
produced by irradiating a CH3CN solution (2.7ꢁ10ꢀ3 M) of 1c with
visible light14 to trigger the ring-opening reaction and generate 1o.
Again, the progress of the photoreaction can be monitored using
the methine peaks of the thiophene rings, which in this case shift
from 7.30 and 7.06 ppm for the ring-closed isomer 1c to 8.00 and
7.62 ppm for 1o. The ring-open monobenzylated DTCP (1o) can be
collected by evaporating the solvent without the need for further
purification.
The rate constants for all alkylation reactions of the compounds
reported in this paper can be estimated using pseudo-first-order
kinetic analysis17 by treating CD3CN solutions of the DTCP com-
pounds with a large excess of 4-bromobenzyl bromide (30 mol -
equiv)18 and monitoring the changes in the concentration of each
component during the course of the reaction using 1H NMR spec-
troscopy at controlled temperatures. The most convenient signals
to monitor in the 1H NMR spectra are those corresponding to the
a-protons of the pyridine and pyridinium rings (protons Ha in
Fig. 2), which decrease in intensity during a 24-h period at 22 ꢂC
when a solution (2.4ꢁ10ꢀ3 M) of the ring-open isomer of mono-
benzylated DTCP (1o) is treated with the excess of the alkylating
reagent.19 Graphical treatment of the change in concentration of 1o
over time gives the expected linear relationship as illustrated in
Figure 220 and an apparent pseudo-first-order rate constant (k01o
)
of (2.9ꢃ0.1)ꢁ10ꢀ5 s
.
ꢀ1 21 Treating an equivalent solution of the ring-
closed isomer 1c22 under identical conditions generates an appar-
ent pseudo-first-order rate constant (k01c) that is just over 3 times
smaller, (9.0ꢃ0.5)ꢁ10ꢀ6 sꢀ1
.
It is clear from these experiments that the ring-open isomer is
just over 3 times more reactive as a nucleophile than its ring-closed
counterpart. Although these differences appear small, they corre-
spond to a difference in a reaction that is 95% complete in just over
1 day (for 1o) compared to one that takes just under 4 days (1c)
to reach the same level of completion. The magnitude of the
2.2. Photochromic properties of monobenzylated DTCPs
1o and 1c
Irradiating a CH3CN solution of compound 1o (2.0ꢁ10ꢀ5 M)
with 365 nm light14 triggers the photocyclization reaction and
generates a solution of the ring-closed isomer 1c. The visual
demonstration of this photoreaction is the change in colour of the
solution from colourless to greenish-blue due to the formation of
the extended
p-conjugated backbone in the ring-closed isomer.
The corresponding UV–vis absorption spectra show trends that are
typical for ring-closing reactions of dithienylethene derivatives
(Fig. 1a). The high-energy bands in the spectra become less intense
as a broad band centred at 649 nm appears. At a concentration of
2.0ꢁ10ꢀ3 M (in CD3CN), a photostationary state is reached within
10 min, at which point only the proton signals corresponding to the
ring-closed isomer 1c are observed in the 1H NMR spectrum.15 Ir-
radiating the coloured solution with light of wavelengths greater
than 490 nm14 results in the complete regeneration of the 1H NMR
and UV–vis spectra corresponding to the ring-open isomer (1o)
without observable formation of side-products. The solutions also
show no apparent signs of degradation even after prolonged irra-
diation with either type of light and after 10 cycles of alternating
the irradiation with 365 nm light and wavelengths greater than
490 nm to toggle the compound between its ring-open and ring-
closed isomers.16
Figure 2. Changes in concentration of monocations 1o (open symbols) and 1c (filled
symbols) when CD3CN solutions of them (2.4ꢁ10ꢀ3 M) are treated with excess
4-bromobenzyl bromide. The experiments were performed in triplicate.
Figure 1. Changes in the UV–vis absorption spectra of CH3CN solutions (2.0ꢁ10ꢀ5 M) of (a) 1o and (b) 2o as they are irradiated with 365 nm light until the photostationary states
are reached.

H.D. Samachetty et al. / Tetrahedron 64 (2008) 8292–8300
8295
difference in rate constants is also expected and can be ascribed, in
part, to the fact that the nucleophilic pyridine ring in the ring-
closed isomer 1c is twisted 20–30ꢂ out of co-planarity from the
(k0A¼(3.0ꢃ0.1)ꢁ10ꢀ5 sꢀ1) as previously measured. After approxi-
mately 3 h, the solution can be irradiated with 365 nm light until
both the ring-open reactant 1o and the ring-open product 2o are
completely converted to their respective ring-closed forms as
attested by 1H NMR spectroscopy.23 The rate of the alkylation re-
action slows and the apparent pseudo-first-order rate constant
(k0B¼(1.0ꢃ0.1)ꢁ10ꢀ5 sꢀ1) is similar to that measured for pure 1c
described previously. Irradiating this reaction with visible light
(greater than 490 nm) after another 3 h regenerates both the ring-
open species and the original rate constant (k0C¼(3.0ꢃ0.1)ꢁ
10ꢀ5 sꢀ1). The system can be modulated again using 365 nm light
(k0D¼(9.9ꢃ0.1)ꢁ10ꢀ6 sꢀ1). There is no observable degradation of
any DTCP species during these photomodulating experiments.
DTCP’s linearly
p-conjugated backbone as has already been sug-
gested to account for similar differences in the coordinating ability
of the two isomers.11 This structural distortion may reduce the
electronic communication between the two ends of the photo-
responsive backbone. We can speculate on another factor that may
be contributing, in this case, to reduce the reactivity of isomer 1o.
The conformational flexibility of the ring-open isomer allows the
compound to adopt a structure where the pyridine and the posi-
tively charged pyridinium groups are in close proximity to each
other due to electrostatic attraction. This conformation would re-
sult in through-space effects detrimental to the reactivity of 1o
such as sterical crowding of the reaction site, a reduction in the
electron-rich nature of the pyridine and the creation of charge–
charge repulsion in the transition state leading to the formation of
the dication 2o.
The nucleophilicity of the pyridine in DTCP 1 can be modulated
while the alkylation reaction progresses by cycling the compound
between its ring-open (1o) and ring-closed (1c) isomers by alter-
nately irradiating solutions of it with UV and visible light (Fig. 3).
When a CD3CN solution (2.4ꢁ10ꢀ3 M) of 1o is treated with an
excess of 4-bromobenzyl bromide (7.3ꢁ10ꢀ2 M, 30 mol equiv) at
22 ꢂC, the alkylation reaction to form 2o proceeds at a similar rate
2.4. Differences in reactivity of bis(pyridine)s 4o and 4c
As already mentioned in Section 2.1 of this paper, higher yields
of the monobenzylated DTCP are produced when the bis(pyridine)
is alkylated in its ring-closed form (4c/1c). The fact that the re-
action of the ring-closed isomer 4c with 4-bromobenzyl bromide
affords monoalkylated 1c as the predominant product while its
ring-open counterpart (4o) affords predominantly the dialkylated
product (2o) suggests that the pyridines in ring-closed 4c are more
reactive than the remaining pyridine in monocation 1c (this is
expected) and that the pyridines in ring-open 4o are not neces-
sarily more reactive than that in monobenzylated 1o. An analysis of
the rate of the alkylation reactions of 4o/1o/2o and 4c/1c/2c
using consecutive first-order kinetics should shed light on the
differences in reactivity of the pyridines in the entire series of DTCP
derivatives.
When a CD3CN solution of the ring-open bis(pyridine) 4o
(4.6ꢁ10ꢀ3 M) is treated with an excess of 4-bromobenzyl bromide
(1.2ꢁ10ꢀ1 M, 27 mol equiv) the change in concentration of each
species (4o, 1o and 2o) during the reaction at 22 ꢂC can be moni-
tored by 1H NMR spectroscopy (for the signals corresponding to the
a
-protons as described previously).25 The results are presented in
Figure 4a and b.
The plots of the relative concentrations of 4o, 1o and 2o against
time are consistent with consecutive pseudo-first-order kinetics.
The concentration of the bis(pyridine) 4o shows the predicted ex-
ponential decay indicative of a pseudo-first-order trend. The trend
for monobenzylated 1o is consistent with a situation where the
concentration increases to a maximum and then decreases sup-
porting the expectation that the concentration of 1o is dependent
on both the concentration of bis(pyridine) 4o and the rate of con-
version to the bis(pyridinium) 2o. The shape of the plot for the 2o
Figure 3. Modulation of the rate of the alkylation reaction of
a
CD3CN solution
(2.4ꢁ10ꢀ3 M) of DTCP
1 with excess 4-bromobenzyl bromide (30 mol equiv) by
alternatively irradiating with UV and visible light at 22 ꢂC.24
Figure 4. a) Changes in concentration of bis(pyridine) 4o (,), monocation 1o (B) and dication 2o (6) when a CD3CN solution of 4o (4.6ꢁ10ꢀ3 M) is treated with excess
4-bromobenzyl bromide. (b) Changes in concentration of bis(pyridine) 4c (-), monocation 1c (C) and dication 2c (:) when a CD3CN solution of 4c (4.6ꢁ10ꢀ3 M) is treated with
excess 4-bromobenzyl bromide.

8296
H.D. Samachetty et al. / Tetrahedron 64 (2008) 8292–8300
Table 1
O
CO2CH3
Average apparent pseudo-first-order rate constants for the alkylation reactions of
CD3CN solutions of DTCP compounds 1 and 4 with an excess of 4-bromobenzyl
bromide at 22 ꢂ
CHO
CO2CH3
CO2CH3
pyridine catalyst
DME
+
C
CO2CH3
NO2
Compound
Apparent rate
NO2
constant (k0) (sꢀ1
)
5
Bis(pyridine) 4o
Bis(pyridine) 4c
Mono(pyridine) 1o
k01
k03
k02
k01o
(1.2ꢃ0.1)ꢁ10ꢀ4
(4.6ꢃ0.3)ꢁ10ꢀ5
(5.4ꢃ0.3)ꢁ10ꢀ5
(2.9ꢃ0.1)ꢁ10ꢀ5
(3.0ꢃ0.1)ꢁ10ꢀ5
(1.5ꢃ0.1)ꢁ10ꢀ5
(9.0ꢃ0.5)ꢁ10ꢀ6
(1.0ꢃ0.1)ꢁ10ꢀ5
(9.9ꢃ0.1)ꢁ10ꢀ6
Scheme 4. Pyridine-catalyzed addition of DMAD to 3-nitrobenzaldehyde.
a
The test reactions are carried out by adding 1.0 and 0.2 mol-
equiv of the appropriate DTCP isomer (1o or 1c) to a mixture
containing 3-nitrobenzaldehyde and DMAD (1:1) in anhydrous
dimethylether (DME). After 2 days, the amount of product 5 formed
for each reaction is assessed by analyzing the ratios of the areas
under the peaks corresponding to 5 and 3-nitrobenzaldehyde by 1H
NMR spectroscopy.16 This analysis reveals that 20% of 5 is generated
when 1.0 mol equiv of ring-open 1o is present as compared to 11%
when the amount of 1o is reduced to 0.2 mol equiv, indicating the
activity of the DTCP as a catalyst. The background reaction of
equimolar amounts of the 3-nitrobenzaldehyde and DMAD in an-
hydrous DME in the absence of the monobenzylated DTE ligands 1o
and 2o did not lead to any product formation after 2 days. In the
case of ring-closed 1c, only 4% of product 5 is observed when
1.0 mol equiv of 1c is present. No observable peaks corresponding
to product 5 appear in the spectrum when only 0.2 mol equiv of 1c
is used. These results show that the ring-open isomer 1o is more
effective than the ring-closed isomer 1c as predicted from the trend
observed in the previous studies.
b
k0A and k0C
Mono(pyridine) 1c
k04
a
k01c
b
k0B
b
k0D
a
Obtained from the kinetic analysis of the reaction of 1o and 1c as described in
Figure 2.
b
Obtained from the kinetic analysis of the cycling reaction as described in
Figure 3.
shows the characteristic initial delay in its formation, which is
governed by the time required for the monocation 1o to be gen-
erated. Treatment of the data using non-linear least squares re-
gression analyses26 affords apparent pseudo-first-order rate
constants of k01¼(1.2ꢃ0.1)ꢁ10ꢀ4 sꢀ1 and k02¼(5.4ꢃ0.3)ꢁ10ꢀ5 sꢀ1 as
averages of three experiments (Table 1).
A CD3CN solution of ring-closed bis(pyridine) 4c27 treated under
identical conditions shows similar changes for the signals corre-
sponding to the a-protons of the pyridine or the pyridinium groups
in the 1H NMR spectra.16 The plots of the relative concentrations of
4c, 1c and 2c against time are shown in Figure 4b and are consistent
with consecutive pseudo-first-order kinetics. Treatment of the
data using non-linear least squares regression analyses26 affords
apparent pseudo-first-order rate constants of k03¼(4.6ꢃ0.3)ꢁ
10ꢀ5 sꢀ1 and k04¼(1.5ꢃ0.1)ꢁ10ꢀ5 sꢀ1 as averages of three experi-
ments (Table 1).
These experiments illustrate that the pyridine in the bis(pyr-
idine) 4o reacts approximately 2.1 times faster than the pyridine
in monobenzylated 1o and support the initial argument that
through-space interactions between the nucleophile and the elec-
tron-deficient pyridinium in 1o contribute to reduce the nucleo-
philic strength of 1o. In the case of the ring-closed forms, the two
pyridine rings in both bis(pyridine) 4c and monocation 1c are in
3. Conclusions
In this paper, we have demonstrated the photomodulation of
chemical reactivity by analyzing the kinetics of an alkylation re-
action. The results indicate that the free pyridine in the ring-open
form 1o reacts approximately 3 times faster than the ring-closed
form 1c under pseudo-first-order conditions. Consecutive pseudo-
first-order kinetics have been used to probe the differences in
reactivity between the ring-open (1o and 4o) and the ring-closed
(1c and 4c) isomers of the bis(pyridine) and the monobenzylated
DTCP to better understand the through-bond and the through-
space effects of having two pyridines, and a pyridine and an
electron-deficient pyridinium at the external positions of the
DTCP backbone. The results suggest that the through-bond com-
munication between the pyridine and the pyridinium of the
monobenzylated 1c and the two pyridine groups of the bis(pyr-
idine) 4c lowers the reactivity of the pyridine nitrogen to a greater
extent than the through-space interaction between the pyridine
and the pyridinium in monobenzylated 1o. Preliminary tests
have demonstrated the use of the ring-open and the ring-closed
DTCP isomers as nucleophilic catalysts by investigating the
reaction of 3-nitrobenzaldehyde with dimethylacetylene
dicarboxylate.
electronic communication owing to the linear
p-conjugated path-
way running along the backbone of the DTCP. As expected the
pyridine in ring-closed bis(pyridine) 4c reacts faster (approxi-
mately 3.1 times) than the pyridine in ring-closed, monobenzylated
1c. The fact that the ring-open bis(pyridine) 4o reacts twice as fast
as its ring-closed counterpart (4c) can be attributed to the extended
delocalization in the latter isomer.
2.5. Photoregulation of catalysis: preliminary observations
As mentioned in Section 1 of this paper, photoresponsive com-
pounds such as DTCP 1 are particularly pertinent candidates for
photoregulating chemical reactivity owing to their popularity as
nucleophilic catalysts in a wide range of synthetic systems. One
only has to be reminded of the widespread use of compounds such
as N,N-dimethylaminopyridine (DMAP) in coupling reactions to be
convinced.12 Preliminary studies using DTCPs 1o and 1c demon-
strate the potential use of photoresponsive compounds in these
applications. The addition of dimethylacetylene dicarboxylate
(DMAD) to 3-nitrobenzaldehyde to generate diester 5 is one of
many model reactions catalyzed by pyridine compounds and is the
one used to show the validity of the approach in this report
(Scheme 4) due to the ease with which the product can be analyzed
using 1H NMR spectroscopy.28
4. Experimental
4.1. Materials
All solvents for synthesis were dried and degassed by passing
them through steel columns containing activated alumina under
nitrogen using a solvent purification system (MBraun). All other
solvents were used as received including those for NMR analysis
(Cambridge Isotope Laboratories). Column chromatography was
performed using silica gel 60 (230–400 mesh) from Silicycle Inc. All
reagents and catalysts except for Pd(PPh3)4 (purchased from Strem)
and octafluorocyclopentene (supplied by the Nippon Zeon Corpo-
ration) were purchased from Aldrich and used as received.

H.D. Samachetty et al. / Tetrahedron 64 (2008) 8292–8300
8297
3-Bromo-2-methyl-5-thiopheneboronic acid13 was prepared fol-
lowing literature procedures.
119.3, 110.6, 15.1. FTIR (KBr-cast,
n
(cmꢀ1)): 3056, 3027, 2910, 1598,
1494, 1412, 1324, 1222, 1159, 1008, 988, 862, 812, 760, 718. LRMS (CI
isobutane): m/z 254, 256 [MþH]þ.
4.2. Techniques
4.4.2. 1,2-Bis(20-methyl-50-(pyrid-400-yl)thien-30-yl)perfluoro-
cyclopentene (4o)13
A solution of 3-bromo-2-methyl-5-pyridylthiophene 3 (199 mg,
0.787 mmol) in anhydrous THF (30 mL) was cooled to ꢀ78 ꢂC using
a dry ice/acetone bath. The solution was treated with n-BuLi
(0.32 mL, 2.5 M in hexanes, 0.79 mmol) dropwise under a N2 at-
mosphere. The resulting deep red solution was then treated with
Melting points (mp) were measured on a Gallenkamp capillary
melting point apparatus. 1H and 13C NMR characterizations were
performed on a Varian Mercury 400 instrument working at
400.1 MHz for 1H NMR and 100.6 MHz for 13C NMR or a Bruker TCI
600 instrument working at 600.3 MHz for 1H NMR and 150.5 MHz
for 13C NMR. The 13C NMR spectra are 1H-decoupled but 19F-cou-
pled. Assignments were confirmed using 1H–1H COSY and selective
perfluorocyclopentene (53 mL, 0.40 mmol) through a cooled gas
1D NOE experiments when necessary. Chemical shifts (
d
) are
tight syringe whereby the solution turned green. The cooling bath
was removed and the solution was allowed to slowly warm up to
ambient temperature and quenched with a saturated solution of
NH4Cl (30 mL). The aqueous layer was removed and extracted with
CH2Cl2 (3ꢁ30 mL), the combined organic layers were washed
with a saturated solution of NaHCO3 (100 mL), dried over anhy-
drous Na2SO4, filtered and the solvent was removed using a rotary
evaporator. The product was purified by flash chromatography
(neutral alumina: activity II–III, hexanes/EtOAc¼2:1). Re-
crystallization of the resulting green solid from EtOAc afforded the
bis(pyridine) 4o (123 mg, 60%) as a white powder. Mp 179–181 ꢂC
reported in parts per million relative to tetramethylsilane using the
residual solvent peak as an internal reference. Coupling constants
(J) are reported in hertz. Relaxation measurements (T1) were per-
formed on the Bruker TCI 600 instrument, with the temperature set
to 22 ꢂC and the values are reported in seconds. FTIR measurements
were performed using a Perkin Elmer 599B IR spectrophotometer.
UV–vis absorption spectroscopy was performed using a Varian Cary
300 Bio spectrophotometer. Low resolution mass spectrometry
(LRMS) measurements were performed using a matrix assisted
laser desorption/ionization source (MALDI) using a PerSeptive
Biosystems Voyager-DE instrument and the MALDI-TOF mass
spectra were obtained using 2,5-dihydroxybenzoic acid as the
matrix. Microanalyses (Anal.) were performed on a Carlo Erba
Model 1106 CHN analyzer.
(lit. 181 ꢂC).13 1H NMR (400 MHz, CD2Cl2,
d (ppm)): 8.58 (d,
J¼6.0 Hz, 4H), 7.50 (s, 2H), 7.42 (d, J¼6.0 Hz, 4H), 2.01 (s, 6H). 13C
NMR (100 MHz, CD2Cl2,
d (ppm)): 150.4, 143.4, 140.0, 139.3, 126.1,
124.6, 119.4, 14.5 (8 of 11 carbons found). FTIR (KBr-cast, n
(cmꢀ1)):
3070, 3032, 2964, 2856, 1630, 1596, 1552, 1504, 1441, 1413, 1339,
1274, 1222, 1193, 1116, 1056, 989, 964, 896, 815, 740, 691. LRMS (CI
isobutane): m/z 523 [MþH]þ.
4.3. Photochemistry
Standard hand-held lamps (312 and 365 nm) used for visualiz-
ing TLC plates (Spectroline E-series, 470
m
W cmꢀ2) were used to
4.4.3. Photochemical synthesis of the ring-closed isomer of the
bis(pyridine) (4c)
carry out the ring-closing reactions. The power of the light source is
given based on the specifications supplied by the company when
the lamps were purchased. A light detector was not used to mea-
sure the intensity during the irradiation experiments. A 312 nm
light source was used for the bis(pyridine) 4o and a 365 nm light
source was used for the monobenzylated bis(pyridinium) 1o and
the dibenzylated bis(pyridinium) 2o. The ring-opening reactions
of the bis(pyridine) 4c, the monobenzylated bis(pyridinium) 1c and
the dibenzylated bis(pyridinium) 2c were carried out using the
light of a 300 W halogen photo-optic source that was passed
through a 490 nm cut-off filter to eliminate higher energy light.
A solution of the ring-open isomer of the bis(pyridine) (4o)
(61 mg, 0.12 mmol) was dissolved in CH3CN (20 mL) and irradiated
with 312 nm light. Aliquots (2 mL) of the reaction mixture were
removed via a pipette at regular intervals, concentrated to dryness
in vacuo, re-dissolved in CD3CN and monitored by 1H NMR spec-
troscopy until a solution containing 98% of the ring-closed form 4c
was obtained. No attempts to isolate the pure ring-closed were
made and the remaining 2% was assigned to the ring-open form 4o.
1H NMR (400 MHz, CD3CN,
J¼6.0 Hz, 4H), 7.02 (s, 2H), 2.21 (s, 6H).
d
(ppm)): 8.67 (d, J¼6.0 Hz, 4H), 7.52 (d,
4.4. Synthesis
4.4.4. Nitrate salt of the ring-closed isomer of 1-[50-(pyrid-400-yl)-
20-methylthien-30-yl]-2-{2%-methyl-5%-[N-(4&-bromobenzyl-
pyrid)-4&-yl]thien-3%-yl}perfluorocyclopentene (1c)
4.4.1. 3-Bromo-2-methyl-5-(40-pyridyl)thiophene (3)13
A mixture of THF (50 mL) and a saturated aqueous Na2CO3 so-
lution (50 mL, 2 M) was deoxygenated by bubbling N2 through the
solution for 45 min, treated with 4-bromopyridine hydrochloride
(2.81 g, 14.4 mmol) and N2 was bubbled through the solution for
an additional 5 min. 3-Bromo-2-methyl-5-thiopheneboronic acid13
(2.90 g, 13.1 mmol) and Pd(PPh3)4 (35 mg, 0.030 mmol) were im-
mediately added and the reaction mixture was heated at reflux for
20 h under a N2 atmosphere. The heat source was removed and the
reaction mixture was allowed to slowly cool to room temperature
at which time, the aqueous layer was removed and extracted with
CHCl3 (3ꢁ50 mL). The combined organic layers were washed with
brine (200 mL), dried over anhydrous Na2SO4, filtered and con-
centrated to dryness in vacuo. The product was purified using
a short column (silica, hexanes/EtOAC¼6:1, containing 1% of Et3N)
yielding an off-white solid. Recrystallization from EtOAc afforded
pyridylthiophene 3 (2.35 g, 70%) as colourless crystals. Mp 80–
The ring-open isomer of the bis(pyridine) (4o) (61 mg,
0.12 mmol) was dissolved in anhydrous CH3CN (20 mL) and the
solution was irradiated with 312 nm light. Aliquots (2 mL) of the
reaction mixture were removed via a pipette at regular intervals,
concentrated to dryness in vacuo, re-dissolved in CD3CN and
monitored by 1H NMR spectroscopy until a solution containing 98%
of the ring-closed form 4c was obtained. The resulting dark blue
solution was treated with 4-bromobenzyl bromide (29 mg,
0.12 mmol) and left to stir at room temperature for 7 days under
a N2 atmosphere. The solvent was evaporated to dryness in vacuo
and purification by flash chromatography (silica, CH3CN to remove
any unreacted 4c followed by a mixture of CH3CN/H2O/a saturated
solution of KNO3¼100:1:0.01) afforded the nitrate salt of ring-
closed monobenzylated 1c (39 mg, 45%) as a dark blue solid. Mp
138–140 ꢂC. 1H NMR (500 MHz, CD3CN,
d
(ppm)): 8.72 (d, J¼6.5 Hz,
2H), 8.70 (d, J¼6.0 Hz, 2H), 8.05 (d, J¼6.5 Hz, 2H), 7.65 (d, J¼8.5 Hz,
2H), 7.54 (d, J¼6.0 Hz, 2H), 7.39 (d, J¼8.5 Hz, 2H), 7.30 (s, 1H), 7.06
(s, 1H), 5.74 (s, 2H), 2.24 (s, 3H), 2.23 (s, 3H). 13C NMR (125 MHz,
82 ꢂC (lit. 83 ꢂC).13 1H NMR (400 MHz, CD2Cl2,
d (ppm)): 8.56 (d,
J¼5.6 Hz, 2H), 7.38 (d, J¼5.6 Hz, 2H), 7.34 (s, 1H), 2.44 (s, 3H). 13C
NMR (100 MHz, CD2Cl2,
d
(ppm)): 150.7, 140.3, 138.1, 136.8, 128.0,
CD3OD, d (ppm)): 160.3, 154.5, 151.7, 151.3, 149.5, 149.3, 146.1, 141.5,

8298
H.D. Samachetty et al. / Tetrahedron 64 (2008) 8292–8300
133.9, 133.7, 132.2, 126.3, 125.3, 124.5, 122.4, 118.8, 70.0, 67.8, 64.5,
assessed by monitoring the methine peaks of the thiophene rings
(8.00 ppm for the ring-open isomer 2o as opposed to 7.35 ppm for
the ring-closed form 2c) until the peak corresponding to the ring-
open isomer 2o was no longer observed. 1H NMR (400 MHz, CD3CN,
25.8, 25.5 (21 of 26 carbons found). UV–vis (CH3CN, lmax (nm)
(log 3 n
/Mꢀ1 cmꢀ1)): 649 (4.12). FTIR (KBr-cast, (cmꢀ1)): 3124, 2930,
2866, 1635, 1596, 1561, 1504, 1474, 1384, 1341, 1276, 1217, 1197, 1129,
1091, 1054, 979, 933, 844, 818, 748. LRMS (MALDI-TOF): m/z 691,
693 [MꢀNO3]þ. Anal. Calcd for C32H22S2F6N3BrO3: C, 50.94; H, 2.94;
N, 5.57. Found: C, 50.62; H, 3.15; N, 5.30.
d
(ppm)): 8.80 (d, J¼6.4 Hz, 4H), 8.12 (d, J¼6.4 Hz, 4H), 7.67 (d,
J¼8.0 Hz, 4H), 7.41 (d, J¼8.0 Hz, 4H), 7.35 (s, 2H), 5.72 (s, 4H), 2.28 (s,
6H). UV–vis (CH3CN, lmax (nm) (log 3
/Mꢀ1 cmꢀ1)): 670 (4.03).
4.4.5. Photochemical synthesis of the nitrate salt of the ring-open
isomer of the monobenzylated DTCP (1o)
4.5. Fitting of the collected data
The nitrate salt of the ring-closed isomer of monobenzylated
DTCP (1c) (29 mg, 38 mmol) was dissolved in anhydrous CH3CN
Linear least squares regression analysis of the data for the in-
dependent reaction of the ring-open and the ring-closed isomers of
the monobenzylated dithienylcyclopentene (DTCP) (1o and 1c)
with an excess of 4-bromobenzyl bromide and the in situ reaction
of both the ring-open and the ring-closed isomers of the mono-
benzylated DTCP (1o and 1c) with an excess of the 4-bromobenzyl
bromide were performed using the commercially available Excel
Data Analysis Tool Package. Non-linear least squares regression
analysis of the data for the reaction of both the ring-open (4o) and
the ring-closed (4c) isomers of the bis(pyridine) with an excess of
the 4-bromobenzyl bromide were completed using a free trial
version of GraphPad Prism Software.
(15 mL) and irradiated with light of wavelengths greater than
490 nm until a colourless solution was observed. Aliquots (2 mL) of
the reaction mixture were removed via a pipette, concentrated to
dryness in vacuo, re-dissolved in CD3CN and monitored by 1H NMR
spectroscopy until complete disappearance of the peaks corre-
sponding to the ring-closed isomer 1c (methine peaks of the
thiophene rings at 7.30 and 7.06 ppm) was observed. The solvent
was evaporated to dryness yielding the nitrate salt of the ring-open
monobenzylated DTCP 1o (28 mg, 97%) as an off-white solid. 1H
NMR (500 MHz, CD3CN,
d
(ppm)): 8.60 (d, J¼7.0 Hz, 2H), 8.57 (d,
J¼6.0 Hz, 2H), 8.06 (d, J¼7.0 Hz, 2H), 8.00 (s, 1H), 7.65 (d, J¼8.5 Hz,
2H), 7.62 (s, 1H), 7.51 (d, J¼6.0 Hz, 2H), 7.36 (d, J¼8.5 Hz, 2H), 5.69
(s, 2H), 2.11 (s, 3H), 2.04 (s, 3H). 13C NMR (150 MHz, CD3CN,
4.6. Kinetic analysis
d
(ppm)): 151.2, 150.7, 148.9, 145.7, 145.3, 145.2, 140.7, 140.3, 136.1,
All data were treated using pseudo-first-order kinetic analysis
by adding an excess of 4-bromobenzyl bromide and using the
following equation:
134.0, 133.1, 132.7, 132.2, 127.9, 126.6, 126.5, 126.4, 126.3, 124.2,
123.7, 62.8, 15.3, 15.0 (23 of 26 carbons found). UV–vis (CH3CN, lmax
(nm) (log 3
/Mꢀ1 cmꢀ1)): 356 (4.37).
d½DTCPꢄ
¼ k0½DTCPꢄ
dt
4.4.6. The bis(hexafluorophosphate) salt of 1,2-bis(20-methyl-50-
[N-(4%-bromobenzyl)-pyrid-400-yl]thien-30-yl)perfluorocyclo-
pentene (2o)
where k0¼k[4-bromobenzyl bromide].
Graphical treatment of the data by plotting the natural log of the
concentration of [DTCP] against time gives a straight line with
a slope of ꢀk0 where k0 is the pseudo-first-order rate constant.
The ring-open isomer of the bis(pyridine) (4o) (25 mg, 48
was dissolved in anhydrous CH3CN (15 mL), treated with 4-bro-
mobenzyl bromide (13 mg, 96 mol) and then heated under reflux
mmol)
m
for 18 h under a N2 atmosphere. The heating source was removed
and the mixture was allowed to slowly cool down to room tem-
perature. The solvent was removed in vacuo and the reaction
mixture was sonicated with Et2O (3ꢁ10 mL) to remove any
unreacted 4-bromobenzyl bromide. The product was collected by
vacuum filtration, washed with Et2O (3ꢁ5 mL) and left to dry
yielding the dibenzylated DTCP as the bromide salt. The green solid
was dissolved in the minimum amount of EtOH (1 mL) and a satu-
rated solution of NH4PF6 was added. The resulting pale green pre-
cipitate was collected by vacuum filtration, washed with copious
amounts of water and left to dry yielding the bis(hexafluoro-
phosphate) salt of the dibenzylated DTCP 2o (52 mg, 96%) as a very
pale green solid. Mp 230–232 ꢂC. 1H NMR (400 MHz, CD3CN,
ln½DTCPꢄt ¼ ꢀk0t þ ln½DTCPꢄ0
4.6.1. Experiment to ensure pseudo-first-order kinetics
The validity of using pseudo-first-order kinetics was determined
by monitoring three samples of different concentrations (1.9, 2.8
and 3.8 mM) of DTCP 1o. The apparent pseudo-first-order rate
constants (k0) were calculated to be (2.9ꢃ0.1)ꢁ10ꢀ5
,
(2.9ꢃ0.1)ꢁ10ꢀ5 and (2.8ꢃ0.1)ꢁ10ꢀ5 sꢀ1
.
A solution of the ring-open isomer (1o) of the monoalkylated
DTCP (1.0 mg, 1.3
transferred into an NMR tube. The solution was treated with an
excess of 4-bromobenzyl bromide (13 mg, 52 mol) and the sample
mmol) in CD3CN (0.70 mL) was prepared and
d
(ppm)): 8.67 (d, J¼6.8 Hz, 4H), 8.08 (d, J¼6.8 Hz, 4H), 8.00 (s, 2H),
7.64 (d, J¼8.4 Hz, 4H), 7.38 (d, J¼8.4 Hz, 4H), 5.63 (s, 4H), 2.11 (s,
6H). 13C NMR (100 MHz, CD3CN,
(ppm)): 151.7, 149.3, 145.5, 136.3,
133.3, 133.4, 132.5, 132.0, 127.7, 124.4, 123.9, 63.5, 15.4 (13 of 16
carbons found). UV–vis (CH3CN, lmax (nm) (log
/Mꢀ1 cmꢀ1)): 360
(4.59). FTIR (KBr-cast,
(cmꢀ1)): 3139, 1638, 1553, 1518, 1470, 1441,
m
was immediately placed in the NMR instrument, which was already
set to 22 ꢂC. The sample was kept in the probe throughout the re-
action and the progress of the reaction was monitored over a period
of 24 h. By measuring the relative integrals of the areas under the
peaks corresponding to the starting monobenzylated DTCP 1o and
the dibenzylated DTCP 2o generated, the mole fractions c1o and c2o
(corresponding to the monobenzylated DTCP 1o and the dibenzy-
lated DTCP 2o, respectively) were obtained. From these values,
a plot of the natural log of the concentration of 1o against time was
obtained allowing the apparent pseudo-first-order rate constant, k0,
to be determined using linear least squares regression analysis of
the data. The alkylation reaction was repeated twice more follow-
ing the described procedure using a solution of the ring-open iso-
d
3
n
1407, 1339, 1276, 1227, 1193, 1149, 1115, 1056, 988, 964, 844, 815,
748. LRMS (MALDI-TOF): m/z 691, 693 [1oꢀPF6]þ. Anal. Calcd for
C
39H28Br2F6N2S2(PF6)2: C, 40.64; H, 2.45; N, 2.43. Found: C, 40.74;
H, 2.42; N, 2.23.
4.4.7. Photochemical synthesis of the ring-closed isomer of the
dibenzylated DTCP (2c)
The ring-open isomer of the dibenzylated DTCP (2o) (2 mg) was
dissolved in CD3CN (0.7 mL) and transferred into an NMR tube.
The solution was irradiated with 365 nm light and monitored by
1H NMR spectroscopy. The progress of the photocyclization was
mer (1o) of the monoalkylated DTCP (1.5 mg, 2.0
(0.70 mL) and a solution of the ring-open isomer (1o) of the
monoalkylated DTCP (2.0 mg, 2.7 mol) in CD3CN (0.70 mL), all the
mmol) in CD3CN
m

H.D. Samachetty et al. / Tetrahedron 64 (2008) 8292–8300
8299
while keeping the concentration of the 4-bromobenzyl bromide
prepared and a known volume of that solution (0.70 mL, 1.67
m
mol)
(13 mg, 52
m
mol, 7.3ꢁ10ꢀ2 M) constant. The calculated apparent
was transferred into a quartz NMR tube. The solution was treated
with an excess of 4-bromobenzyl bromide (13 mg, 52 mol) fol-
pseudo-first-order rate constants were very similar for all three
samples regardless of the initial concentration of the mono-
benzylated DTCP 1o consistent with a pseudo-first-order pattern.
m
lowing the same procedure used for the ring-open isomer 1o and
the ring-closed isomer 1c and the progress of the reaction was
monitored under the same conditions at regular intervals of time
by 1H NMR spectroscopy. After about 3 h, the NMR tube was
irradiated with 365 nm light until the peaks corresponding to the
ring-open reactant 1o (monobenzylated DTCP) and the ring-open
product 2o (dibenzylated DTCP) were no longer observed. The
sample was quickly placed in the NMR probe and the progress of
the reaction was monitored under the same conditions at regular
intervals of time. After another 3 h, the sample was irradiated with
light of wavelengths greater than 490 nm until the peaks corre-
sponding to the ring-closed species were no longer observed. Such
in situ switching by alternating irradiation at 365 and >490 nm
over 3 h intervals was repeated. Graphical treatment of the data in a
similar fashion to the method already described for the reaction of
independent solutions of the ring-open (1o) and the ring-closed
(1c) isomers showed that the apparent rate constants for the
two ring-opening steps are very close in value (k0A¼(3.0ꢃ0.1)ꢁ
10ꢀ5 sꢀ1 and k0C¼(3.0ꢃ0.1)ꢁ10ꢀ5 sꢀ1) and that of the two ring-
closing steps are also similar (k0B¼(1.0ꢃ0.1)ꢁ10ꢀ5 sꢀ1 and
k0D¼(9.9ꢃ0.1)ꢁ10ꢀ6 sꢀ1).
4.6.2. Kinetics experiments for the alkylation reaction of the ring-
open isomer (1o) of the monobenzylated DTCP under pseudo-
first-order conditions
A stock solution (2.4ꢁ10ꢀ3 M) of the ring-open isomer (1o) of
the monoalkylated DTCP (12 mg, 16 mmol) in CD3CN (6.5 mL) was
prepared and a known volume of that solution (0.70 mL, 1.71
was transferred into an NMR tube. The solution was treated with an
excess of 4-bromobenzyl bromide (13 mg, 52 mol) and the sample
mmol)
m
was immediately placed in the NMR instrument, which was already
set to 22 ꢂC. The sample was kept in the probe throughout the re-
action and the progress of the reaction was monitored by 1H NMR
spectroscopy over a period of 24 h. By measuring the relative in-
tegrals of the areas under the peaks corresponding to the starting
monobenzylated DTCP 1o and the dibenzylated DTCP 2o generated,
the mole fractions c1o and c2o (corresponding to the mono-
benzylated DTCP 1o and the dibenzylated DTCP 2o, respectively)
were obtained. From these values, a plot of the natural log of
the concentration of 1o against time was obtained allowing the
apparent pseudo-first-order rate constant, k01o, to be determined
using linear least squares regression analysis of the data. To test the
reproducibility of the results, the alkylation reaction was repeated
twice following the described procedure using samples prepared
from the same stock solution (2.4ꢁ10ꢀ3 M) of the ring-open isomer
1o in CD3CN. From these values, the average apparent pseudo-first-
order rate constant, k01o, for the alkylation reaction of the ring-open
isomer (1o) of the monobenzylated DTCP was calculated to be
4.6.5. Kinetics experiment for the alkylation reaction of the ring-
open isomer (4o) of the bis(pyridine) under consecutive pseudo-
first-order conditions
A stock solution (4.6ꢁ10ꢀ3 M) of the ring-open isomer (4o) of
the bis(pyridine) (5.1 mg, 9.7 mmol) in CD3CN (2.1 mL) was pre-
pared and a known volume of that solution (0.70 mL, 3.3
transferred into an NMR tube. The solution was treated with an
excess of 4-bromobenzyl bromide (22 mg, 87 mol) following the
mmol) was
(2.9ꢃ0.1)ꢁ10ꢀ5 sꢀ1
.
m
same procedure used for the alkylation reactions of the ring-open
(1o) and the ring-closed (1c) isomers of the monobenzylated DTCP
and the progress of the reaction was monitored under the same
conditions for 10 h. By measuring the relative integrals of the areas
under the peaks corresponding to the starting bis(pyridine) 4o, the
generated monobenzylated DTCP 1o (as an intermediate) and
4.6.3. Kinetics experiments for the alkylation reaction of the
ring-closed isomer (1c) of the monobenzylated DTCP under
pseudo-first-order conditions
The same volume (0.70 mL, 1.71 mmol) of the stock solution
(2.4ꢁ10ꢀ3 M) of the ring-open isomer (1o) of the monoalkylated
DTCP (12 mg, 16
mmol) in CD3CN (6.5 mL), used for the previous
the generated dibenzylated DTCP 2o, the mole fractions c4o, c1o,
experiment, was transferred into an NMR tube. The solution was
irradiated with 365 nm light until the methine peaks of the thio-
phene rings corresponding to the ring-open isomer 1o (8.00 and
7.62 ppm for the ring-open isomer 1o as opposed to 7.30 and
7.06 ppm for the ring-closed form 1c) were no longer observed, as
monitored by 1H NMR spectroscopy. The resultant dark bluish
solution was treated with an excess of 4-bromobenzyl bromide
and c2o (corresponding to the bis(pyridine) 4o, the mono-
benzylated DTCP 1o and the dibenzylated DTCP 2o, respectively)
were obtained. From these values, a plot of the relative concen-
trations for all three species (4o, 1o and 2o) against time was
o0btained. The average apparent pseudo-first-order rate constants,
k1o ¼ ð1:2 ꢃ 0:1Þ ꢁ 10ꢀ4sꢀ1 and k10 o ¼ ð5:4 ꢃ 0:2Þ ꢁ 10ꢀ5sꢀ1
,
corresponding to the two consecutive pseudo-first-order alkylation
steps, were obtained after non-linear regression analysis of the
data.
(13 mg, 52 mmol) following the same procedure used for the ring-
open isomer 1o and the progress of the reaction was monitored
under the same conditions for 24 h. Graphical treatment of the data
by plotting the natural log of the concentration of monobenzylated
DTCP 1c against time followed by a linear least squares regression
analysis of0 the data gave the apparent pseudo-first-order rate
constant, k1c. To test the reproducibility of the results, the alkyl-
ation reaction was repeated twice following the described pro-
cedure using samples prepared from the same stock solution
(2.4ꢁ10ꢀ3 M) of the ring-open isomer (1o) in CD3CN. From these
4.6.6. Kinetics experiment for the alkylation reaction of the ring-
closed isomer (4c) of the bis(pyridine) under consecutive pseudo-
first-order conditions
The same volume (0.70 mL, 3.3 mmol) of the stock solution
(4.6ꢁ10ꢀ3 M) of the ring-open isomer (4o) of the bis(pyridine)
(5.1 mg, 9.7 mmol) in CD3CN (2.1 mL), used for the previous experi-
ment, was transferred into an NMR tube. The solution was irradi-
ated with 313 nm light by monitoring the methine peaks of the
thiophene rings (7.63 ppm for the ring-open isomer 4o as opposed
to 7.02 ppm for the ring-closed form 4c) until a solution containing
98% of the ring-closed isomer 4c was observed. Without attempting
to isolate the pure ring-closed isomer 4c, the solution was treated
values, the average apparent pseudo-first-order rate constant, k10 c
,
for the alkylation reaction of the ring-closed isomer (1c) of the
monobenzylated DTCP was calculated to be (9.0ꢃ0.5)ꢁ10ꢀ6 sꢀ1
.
4.6.4. In situ photomodulation of the alkylation reaction of the ring-
open (1o) and the ring-closed (1c) isomers of the monobenzylated
DTCP under pseudo-first-order conditions
with an excess of 4-bromobenzyl bromide (22 mg, 87 mmol) fol-
lowing the same procedure used for the ring-open isomer and
the progress of the reaction was monitored under the same con-
ditions for 17 h. Graphical treatment of the data by plotting the
A stock solution (2.4ꢁ10ꢀ3 M) of the ring-open isomer (1o) of
the monoalkylated DTCP (1.8 mg, 2.4 mmol) in CD3CN (1.0 mL) was

8300
H.D. Samachetty et al. / Tetrahedron 64 (2008) 8292–8300
15. The signals in the 1H NMR spectra of both isomers were assigned using 1H–1H
COSY and selective 1D NOE experiments.
16. See Supplementary data for details.
relative concentrations of the bis(pyridine) 4c, the monobenzylated
DTCP 1c and the dibenzylated DTCP 2c against time, followed
by non-linear least squares regression analysis of the data,
17. The treatment of all data using pseudo-first-order kinetic analysis should be
valid assuming all substitution reactions of the DTCP derivatives with DTCP
1o proceed by an SN2 pathway, which exhibit overall second-order kinetics.
Adding a sufficiently high concentration of the alkylating reagent so that
its concentration remains constant during the course of the reaction reduces
the reaction to one governed by pseudo-first-order kinetics. However, one
important assumption for this to hold true is that there is no significant back
reaction. The term apparent pseudo-first-order rate constant will be used
throughout this paper in order to take into account any back reaction where
the apparent pseudo-first-order rate constant refers to the sum of the forward
pseudo-first-order rate constant and the backward pseudo-first-order rate
constant.
gave the average apparent pseudo-first-order rate constants, k30 c
¼
ð4:6 ꢃ 0:3Þ ꢁ 10ꢀ5sꢀ1 and k40 c ¼ ð1:5 ꢃ 0:1Þ ꢁ 10ꢀ5sꢀ1
,
corre-
sponding to the two consecutive pseudo-first-order alkylation steps.
Supplementary data
Supplementary data associated with this article can be found in
the online version, at doi:10.1016/j.tet.2008.05.050.
18. The validity of using pseudo-first-order kinetics was determined by monitoring
three samples of different concentrations (1.9, 2.8 and 3.8 mM) of 4-bromo-
benzyl bromide. The apparent pseudo-first-order rate constants (k0) were
References and notes
calculated to be (2.9ꢃ0.1)ꢁ10ꢀ5
,
(2.9ꢃ0.1)ꢁ10ꢀ5 and (2.8ꢃ0.1)ꢁ10ꢀ5 sꢀ1
,
1. Valkenberg, M. H.; Holderich, W. F. Catal. Rev. 2002, 44, 321–374.
2. Dynamic Studies in Biology: Phototriggers, Photoswitches and Caged Biomolecules;
Goeldner, M., Givens, R., Eds.; Wiley-VCH: Weinheim, Germany, 2005.
3. Saltzman, W. M. Drug Delivery: Engineering Principles for Drug Therapy; Oxford
University Press: New York, NY, 2001.
4. (a) McCoy, C. P.; Rooney, C.; Jones, D. S.; Gorman, S. P.; Nieuwenhuyzen, M.
Pharm. Res. 2007, 24, 194–200; (b) Denny, W. A. Eur. J. Med. Chem. 2001, 36,
577–595.
5. (a) Special issue on photochromism: Irie, M. Chem. Rev. 2000, 100, 1685–1716;
(b) Irie, M. In Molecular Switches; Feringa, B. L., Ed.; Wiley-VCH: Weinheim,
Germany, 2001; pp 37–62; (c) Irie, M. In Photochromic and Thermochromic
Compounds; Crano, J. C., Guglielmetti, R. J., Eds.; Plenum: New York, NY, 1999;
Vol. 1, pp 207–222; (d) Tian, H.; Yang, S. Chem. Soc. Rev. 2004, 33, 85–97; (e)
Tian, H.; Wang, S. Chem. Commun. 2007, 781–792.
6. (a) Takeshita, M.; Uchida, K.; Irie, M. Chem. Commun. 1996, 1807–1808; (b)
Kawai, S. H. Tetrahedron Lett. 1998, 39, 4445–4448; (c) Takeshita, M.; Irie, M.
J. Org. Chem. 1998, 63, 6643–6649; (d) Mulder, A.; Jukovic, A.; van Leeuwen,
F. W. B.; Kooijman, H.; Spek, A. L.; Huskens, J.; Reinhoudt, D. N. Chem.dEur. J.
2004, 10, 1114–1123.
7. Sud, D.; Norsten, T. B.; Branda, N. R. Angew. Chem., Int. Ed. 2005, 44, 2019–2021.
8. (a) Kawai, S. H.; Gilat, S. L.; Lehn, J.-M. Eur. J. Org. Chem. 1999, 2359–2366; (b)
Odo, Y.; Matsuda, K.; Irie, M. Chem.dEur. J. 2006, 12, 4283–4288.
9. (a) Lemieux, V.; Branda, N. R. Org. Lett. 2005, 7, 2969–2972; (b) Lemieux, V.;
Gauthier, S.; Branda, N. R. Angew. Chem., Int. Ed. 2006, 45, 6820–6824.
10. Sud, D.; Wigglesworth, T. J.; Branda, N. R. Angew. Chem., Int. Ed. 2007, 46, 8017–
8019.
11. Samachetty, H. D.; Branda, N. R. Chem. Commun. 2005, 2840–2842.
12. (a) Vaidya, R. A.; Mathias, L. J. J. Am. Chem. Soc. 1986, 108, 5514–5520; (b)
Scriven, E. F. V. Chem. Soc. Rev. 1983, 12, 129–161; (c) Bender, M. L.; Turnquest,
B. W. J. Am. Chem. Soc. 1957, 79, 1652–1655.
respectively.
19. The time delay during the 1H NMR acquisition was set 5 times larger than the
largest longitudinal relaxation T1 value to ensure that the peak intensities of
protons Ha in 1o and 1c correspond to the number of protons present and
allows the relative number of different protons to be determined by measuring
the areas under the peaks.
20. Linear regression was performed using the Microsoft Excel Data Analysis Tool
Package.
21. All kinetic experiments were done in triplicate.
22. A CD3CN solution of the photostationary state containing at least 98% of ring-
closed 1c was prepared using the same stock solution (2.4ꢁ10ꢀ3 M) of the ring-
open isomer (1o) used for the kinetic analysis by irradiating it with 365 nm
light until complete ring cyclization was achieved as monitored by 1H NMR
spectroscopy.
23. The photochemical cycling studies on independent solutions of the mono-
benzylated DTCP (1) and the dibenzylated DTCP (2) indicate that both DTCP
compounds can be toggled between their ring-open and their ring-closed
isomers numerous times without observable degradation and that both the
ring-closing and ring-opening reactions proceed with a high photostationary
state. See Supplementary data for details.
24. The data was collected from nine integrations of the areas under the peaks in
the 1H NMR spectra. Error was estimated as 2%, which is less than the size of
the circles plotted for each data point in the figure.
25. These studies were carried out in the absence of an internal standard because
the 1H NMR spectra of the bis(pyridine), the monoalkylated DTCP and the
dialkylated DTCP showed no apparent signs of degradation during the course
of the reaction.
26. The least squares regression analysis was performed using the software
GraphPad Prism.
27. This solution was prepared by irradiating the solution of 4o in an NMR tube
with 312 nm light until complete ring cyclization was achieved as monitored by
1H NMR spectroscopy.
28. Nair, V.; Pillai, A. N.; Menon, R. S.; Suresh, E. Org. Lett. 2005, 7, 1189–1191; (b)
Nair, V.; Shreekanth, A. R.; Abhilash, N.; Biju, A. T.; Remadevi, B.; Menon, R. S.;
Rath, N. P.; Srinivas, R. Synthesis 2003, 1895–1902.
13. Gilat, S. L.; Kawai, T.; Lehn, J.-M. Chem.dEur. J. 1995, 1, 275–284.
14. All ring-closing reactions were carried out using the light source from a lamp
used for visualizing TLC plates at 312 or 365 nm (Spectroline
E series,
470 W cmꢀ2). The ring-opening reactions were carried out using the light of
a 300 W halogen photo-optic source passed through a 490 nm cut-off filter to
eliminate higher energy light.