Photogeneration of a diene template for surface Diels-Alder reactions: Photoenolization of an ortho-methyl-benzophenone-modified Au cluster

Article abstract of DOI:10.1139/v03-031

A series of monolayer-protected clusters (MPCs) modified with a photoreactive [4-(11-mercaptoundecyl)-phenyl](2-methylphenyl)methanone (1) moiety have been prepared where 1 is co-absorbed to the MPC surface with dodecanethiol, octadecanethiol, or 11-merca

Full text of DOI:10.1139/v03-031

484
Photogeneration of a diene template for surface
Diels–Alder reactions: Photoenolization of an
ortho-methyl-benzophenone-modified Au cluster
Arnold J. Kell, Christopher C. Montcalm, and Mark S. Workentin
Abstract: A series of monolayer-protected clusters (MPCs) modified with a photoreactive [4-(11-mercaptoundecyl)-
phenyl](2-methylphenyl)methanone (1) moiety have been prepared where 1 is co-absorbed to the MPC surface with
dodecanethiol, octadecanethiol, or 11-mercaptoundecanoic acid methyl ester. Upon irradiation the MPC-anchored 1 re-
acts efficiently through its triplet excited states, yielding 1,4-biradicals that collapse to synthetically useful, long-lived
photodienol intermediates, which can be efficiently trapped in Diels–Alder type chemistry by dienophiles — namely,
dimethyl acetylenedicarboxylate (DMAD). In all cases the Diels–Alder trapping of the dienol occurred efficiently re-
sulting in >60% conversion to the Diels–Alder adduct. This indicates that the local environment surrounding 1 did not
influence its ability to react via the Diels–Alder reaction; however, the reaction could not be taken to completion. The
inability to react completely is attributed to 1 binding to distinct sites on the MPC core; there are edge, vertice, and
terrace sites. Selective population of these specific sites and the subsequent irradiations show that MPCs with 1 an-
chored predominantly at edge and vertice sites results in an extent of reaction of 85 3%, whereas selectively populat-
ing the terrace sites results in an extent of reaction of 36 2%. These results suggest that 1 anchored to edge and
vertice sites is more reactive to the Diels–Alder reaction than that involving terrace sites.
Key words: monolayer protected cluster, site selective reactivity, Diels–Alder, photochemistry.
Resume : On a préparé une série d’agrégats protégés en monocouches (« MPC »), modifiés par une portion photoréac-
tive de [4-(11-mercaptoundécyl)phényl](2-méthylphéyl)méthanone (1) cooabsorbée sur la surface des « MPC » avec du
dodécanethiol, de l’octadécanethiol ou du 11-mercaptoundécanoate de méthyle. Par irradiation, le composé 1 attaché au
« MPC » réagit de façon efficace, par le biais de ses états excités, pour donner des 1,4-biradicaux qui se décomposent
en intermédiaires photodiénoliques utiles d’un point de vue synthétique et qui peuvent être piégés d’une façon efficace
par des diénophiles tel l’acétylènedicarboxylate de diméthyle (ADCM), dans des réactions de type Diels–Alder. Dans
tous les cas, le piégeage de type Diels–Alder du diénol se fait d’une façon efficace conduisant à plus de 60 % de
conversion en adduit de Diels–Alder. Ce résultat indique que l’environnement local autour du composé 1 n’influence
pas sa facilité à réagir par le biais de la réaction de Diels–Alder; toutefois, il n’a pas été possible d’obtenir une réac-
tion complète. On attribue cette inhabilité du composé 1 à réagir complètement à la nature des sites auxquels il est at-
taché sur les « MPC » qui comportent des sites en bordures, aux sommets et sur des terrasses. Une population
sélective de chacun de ces sites spécifiques suivie d’irradiations montrent que le degré de réaction s’élève à 85 3 %
lorsque le composé 1 s’est fixé sur des sites en bordures ou aux sommets des « MPC » mais qu’il n’est que de 36
2 % lorsque le composé 1 s’est fixé sur des sites des terrasses des « MPC ». Ces résultats suggèrent que le composé 1
fixé sur des sites en bordure ou aux sommets sont plus réactifs vis-à-vis de la réaction de Diels–Alder que ceux des
terrasses.
Mots clés : agrégat protégé en monocouche, réactivité sélective d’un site, Diels–Alder, photochimie.
[Traduit par la Rédaction] Kell et al. 494
Introduction
straction from the ortho-alkyl substituent by the n,π* excited
state of the carbonyl group, which generates a 1,4-biradical
that subsequently collapses to the respective E and Z-dienols
(Scheme 1). The lifetime of the Z-dienol is very short (0.03–
1 µs) whereas that of the E-dienol is significantly longer
(3 s). The Z-dienol has a short lifetime because its orienta-
The irradiation of ortho-alkylated benzophenones gener-
ates synthetically useful ortho-quinodimethane enol interme-
diates (1, 2). This reaction, first described by Yang and
Rivas (3), involves an intramolecular γ-hydrogen atom ab-
Received 7 March 2003. Published on the NRC Research Press Web site at http://canjchem.nrc.ca on 29 April 2003.
Dedicated to Professor Don Arnold for his many contributions to chemistry in Canada.
A.J. Kell, C.C. Montcalm, and M.S. Workentin.1 Department of Chemistry, The University of Western Ontario, London, ON
N6A 5B7, Canada.
1Corresponding author (e-mail: mworkent@uwo.ca).
Can. J. Chem. 81: 484–494 (2003)
doi: 10.1139/V03-031
© 2003 NRC Canada

Kell et al.
485
Scheme 1. The generation of the photodienol and its subsequent Diels–Alder reaction.
tion allows reketonization and re-aromatization to occur eas-
ily through a 1,5-hydrogen shift; the reketonization process
for the E-dienol requires acid or base catalysis. The long-
lived E-dienol provides an intermediate that can be effi-
ciently trapped by Diels–Alder dienophiles to yield, exclu-
sively, the endo products, with excellent regioselectivity (1,
2). In this report we extend this photoinitiated Diels–Alder
reaction to monolayer-protected gold clusters or MPCs.
While there are a variety of reported photoinduced reactions
on planar metal surfaces (4–15), there are surprisingly few
reports of photochemical reactions (16, 17) or photophysical
properties (18–20) of organic molecules anchored to MPCs
even though these substrates are emerging as important new
materials (21, 22). In recent studies we reported the
intramolecular Norrish–Yang Type II photoreaction of an
aryl ketone in this type of MPC environment (23, 24); those
studies illustrated, among other things, that unimolecular
carbonyl photoreactions can be efficient in MPCs and can
lead to surface modifications. Here we report the photo-
dienolization of an ortho-methyl-benzophenone-modified
gold MPC and the subsequent trapping of the dienol by a
model dienophile (Scheme 1). To the best of our knowledge,
this study represents the first example of a bimolecular pho-
tochemical reaction on an MPC. Mrksich and co-workers re-
cently reported elegant studies of an electrochemically
generated quinone-immobilized dienophile on planar gold
using Diels–Alder chemistry and showed that the reaction
can be used for selective modification with proteins (25–28).
its reactivity. The Diels–Alder reaction between the
photodienol and a dienophile is a versatile probe of the MPC
environment. This reaction allows for the study of (i) the
ability of the ketone to be excited to the dienol in the MPC
environment, (ii) the ability to trap the dienol, and (iii) the
effects that co-absorbed substrates on the MPC surface have
on the ability of the photodienol to react with a dienophile
(will the photodienol in the monolayer environment be too
sterically hindered to allow the subsequent Diels–Alder re-
action?). Additionally, because the MPC core contains a
number of distinct sites, selectively populating the different
sites with 1 will help elucidate where these bimolecular re-
actions occur most efficiently on the MPC surface.
To address our interests in MPC reactivity we have prepared
a series of MPCs designed to probe a variety of properties
within the MPC environment (Fig. 1). A [4-(11-mercaptoun-
decyl)phenyl](2-methylphenyl)methanone (1) (Scheme 2) sub-
strate was “place-exchanged” onto the surface of previously
prepared dodecanethiolate (C12MPC), octadecanethiolate
(C18MPC), and 11-mercaptoundecanoic acid methyl ester
(MeO2CC10MPC) MPCs. The corresponding MPCs, defined
as 1-C12MPC, 1-C18MPC, and 1-MeO2CC10MPC, respec-
tively, where, for example, 1-C12MPC represents the place-
exchange of 1 onto an original C12MPC, were then irradiated
in the presence of
a dienophile, namely dimethyl
acetylenedicarboxylate (DMAD). The majority of this report
makes use of DMAD as the dienophile, but the reaction was
also shown to occur efficiently with other dienophiles such
as dimethyl fumarate (DMFr) and dimethyl maleate (DMM).
The dienophiles were selected because they were symmetric,
making product identification more straightforward, and be-
cause the solution photochemistry involving ortho-methyl
benzophenone with these dienophiles is well known to occur
efficiently and regioselectively (i.e., where possible, only the
endo product is generated) (3, 31–33).
Little has been done to elucidate the dynamics (29) and
reactivity associated with substrates anchored to MPC sur-
faces, considering that the MPC core is well known to be in-
homogeneous (21, 30): there are edge, vertice, and terrace
sites on each MPC. These distinct sites may display different
reactivity; we have found evidence for this in our earlier
studies (23). It is of interest to observe how efficiently a
photo-induced bimolecular reaction, such as the Diels–Alder
trapping of the E-dienol, can occur on the MPC surface and
what factors associated with the MPC environment influence
The base MPCs were chosen so that a variety of effects
associated with the Diels–Alder reaction between the photo-
dienol and a dienophile could be studied. If we assume that
© 2003 NRC Canada

486
Can. J. Chem. Vol. 81, 2003
Scheme 2. The probe molecule incorporated into the MPC.
spectrometry. Details of the synthetic transformations are
provided in the experimental section.
Generally MPCs are prepared by the reduction of hydro-
gen tetrachloroaurate in the presence of a thiol (34, 35). This
protocol is incompatible with 1 because the presence of a re-
ducing agent will convert the carbonyl functionality to an al-
cohol. So, to prepare the desired 1-MPC, we employ the
place-exchange reaction (36, 37) developed by Murray and
co-workers starting from well-defined base MPCs (35). The
place-exchange reaction is an equilibrium-based reaction in-
volving the replacement of free thiol from solution onto the
MPC surface (36, 37). It is accomplished by stirring an ex-
cess of a thiol that is to be exchanged onto the MPC surface
in the presence of an already prepared base MPC for 4–
5 days. As mentioned earlier, the MPC core consists of dif-
ferent sites. It has been suggested that the place-exchange
reaction populates the edge and vertice sites quickly (within
1 h) and population of the terrace occurs over a much longer
timescale (36). This means that depending on the timescale
of the place-exchange reaction, 1 can populate predomi-
nantly the edge and vertice sites (1 h), or all sites (the ter-
race, edge, and vertice sites) can be populated over much
longer exchange times (4–5 days). This is an important
property of the place-exchange reaction that was exploited
and will be addressed later. The base MPCs were prepared
according to procedures outlined by Murray and co-workers,
as they are known to produce relatively monodisperse MPCs
with a core diameter of 2.0 nm and general stoichiometry of
Au314(X)108, where there are 314 gold atoms comprising the
MPC core and X is the thiolate surrounding the MPC core
(35).
the methylene chains of both 1 and the dilutant chain (or the
substrate on the original MPC, namely C12, C18, or
MeO2CC10) pack similarly, then the mixed 1-C12MPC will
allow for the photochemical generation of the dienol directly
at the interface. In the case of 1-C18MPC, the resulting
photodienol is expected to be embedded within the
monolayer. This is of interest because embedding the dienol
may hinder its ability to react with a dienophile. Both of
these mixed MPCs bear nonpolar terminal methyl groups on
the dilutant chain. Because the terminal groups on MPCs are
known to influence some of their physical properties (21) —
MPCs containing terminal carboxylic acid groups are solu-
ble in methanol and water, whereas MPCs containing termi-
nal methyl groups are soluble only in nonpolar solvents such
as chloroform, dichloromethane, benzene, and hexanes —
we were also interested in employing a dilutant chain con-
taining a terminal ester. The terminal ester would serve to
change the local polarity at the interface and may allow the
polar dienophile to be incorporated into the interface (be-
cause both are polar), possibly allowing the Diels–Alder re-
action to occur more efficiently. We were also interested in
determining if the reactivity of 1 is influenced by its location
on the MPC surface. As mentioned earlier, MPCs have dis-
tinct sites. Selective population of the different sites of the
MPC surface with 1 will allow for the determination of dif-
ferences in the efficiency of the Diels–Alder reaction when
it occurs at edge and vertice sites, as compared with terrace
sites. The extent of this bimolecular reaction within the
MPC environment will be a valuable tool, employed in de-
termining how steric and environmental effects introduced
by the monolayer affect the ability of the dienophile to react
with the surface-bound dienol.
Proton NMR spectroscopy was used to determine the pu-
rity and the extent of exchange (stoichiometry) of the result-
ing mixed MPCs (1-C12MPC, 1-C18MPC, and 1-
MeO2CC10MPC, respectively). The purity of the MPC is
judged by the amount of free thiol or disulfide remaining in
solution after work-up and isolation of the MPC. Any free
thiol or disulfide would appear as sharp resonances in the 1H
NMR spectrum. Figures 2a and 3a are typical 1H NMR
spectra, recorded in benzene-d6, of the 1-C12MPC and 1-
MeO2CC10MPC, respectively, isolated after place-exchange;
of particular note is the absence of sharp resonances assign-
1
able to free thiol (or disulphide in solution). The H NMR
Results and discussion
spectrum of 1-C18MPC is not shown, but is similar to these
(see supplemental information).2 All of the spectra exhibit
distinctive broad resonances (38, 39) at the chemical shifts
measured for pure 1 in benzene-d6 solution, confirming that
the place-exchange has occurred. These resonances include
those of the aromatic protons on the phenyl rings ortho to
the carbonyl at 7.85 (2H) and 7.22 (1H) ppm, the protons on
the methyl function ortho to the carbonyl at 2.29 ppm, and
the CH2 protons of the alkyl spacer α to the benzophenone
function at 2.57 ppm (Figs. 2a and 3a). The remaining meth-
ylene protons of the alkyl chain for 1 appear between 1.15
and 1.90 ppm along with those of the dilutant chain, either
C12, C18, or MeO2CC10. The resonance signal at 0.9 ppm in
Fig. 2a is due to the terminal methyl protons of the remain-
ing dodecanethiolate present on the MPC after the place-
The 1-MPCs were prepared from a [4-(11-mercaptoun-
decyl)phenyl](2-methylphenyl)methanone (1) precursor
(Scheme 2). The synthesis of 1 started from (11-bromoun-
decyl)benzene, prepared by the addition of phenyl lithium to
a large excess of 1,11-dibromoundecane. A Friedel–Crafts
acylation reaction between (11-bromoundecyl)benzene gen-
erated in the mixture and o-toluoyl chloride in the presence
of AlCl3 generated 4-(11-bromoundecyl)phenyl(2-methyl-
phenyl)methanone. The bromide was then converted to thiol
through the reaction of potassium thioacetate and the hydro-
lysis of the resulting thioester with ethanolic K2CO3. The re-
sulting thiol (1) was purified via column chromatography
using 3:1 dichloromethane:hexanes as eluant and character-
1
ized by H NMR, 13C NMR, and IR spectroscopy and mass
2 Supplementary data may be purchased from the Depository of Unpublished Data, Document Delivery, CISTI, National Research Council
Canada, Ottawa, ON K1A 0S2, Canada (http://www.nrc.ca/cisti/irm/unpub_e.shtml for information on ordering electronically).
© 2003 NRC Canada

Kell et al.
487
Fig. 1. Cartoon depictions of the various MPCs studied, showing that the photodienol will be (a) at the interface, (b) embedded within
the monolayer, or (c) near a more polar co-absorbed substrate.
Fig. 2. The irradiation of Au314(C12S)50(1)58 in benzene-d6 (*)
before irradiation (a) and after an irradiation period of 80 h in
the presence of a 3-times molar excess of DMAD (b). The filled
arrows indicate the decrease in intensity for the resonances asso-
ciated with Au314(C12S)50(1)58 while the hollow arrows indicate
an increase in the resonances associated with the product,
namely Au314(C12S)50(1)26(1-DMAD)32. Note: the DMAD has
Fig. 3. The irradiation of Au314(MeO2CC10S)59(1)49 in benzene-
d6 (*) before irradiation (a) and after an irradiation period of
~96 h in the presence of a 3-times molar excess of DMAD (b).
The filled arrows indicate the decrease in intensity for the reso-
nances associated with Au314(MeO2CC10S)59(1)49 while the hol-
low arrows indicate an increase in the resonances associated with
the product, namely Au314(MeO2CC10S)59(1)15(1-DMAD)34.
1
been washed away before the final H NMR spectra was ac-
1
quired. The H NMR spectrum of an authentic sample of the
model compound (2-DMAD) in benzene-d6 is shown in (c) (ac-
tual molecule pictured).
(C12S)50(1)58 and Au314(C18S)74(1)34 for the 1-C12MPC and
the 1-C18MPC, respectively. These stoichiometries were de-
termined in deuteriochloroform solution by comparison of
the integrations for the resonance signal at 7.72 ppm attrib-
uted to the aromatic protons α to the carbonyl in 1 and the
resonance signal at 0.86 ppm attributed to the terminal
methyl group of either C12 or C18 after I2 decomposition of
the appropriate MPC.
In Fig. 3a, the resonance signal at 3.35 ppm and the
shoulder at 2.28 ppm are due to the terminal methyl protons
of the ester and the methylene protons α to the ester, respec-
tively, on the MeO2CC10 moiety. The stoichiometry of the
mixed 1-MeO2CC10MPC was determined directly from the
1H NMR spectrum of the MPC in benzene-d6 by taking the
ratio of the integrations for the resonance at 3.35 ppm (from
MeO2CC10) and 7.85 ppm (from 1) and was found to be
Au314(MeCO2C10S)59(1)49. This was possible because there
was no overlap in these resonance signals, allowing accurate
integration directly from the MPC.
exchange reaction (or in the case of 1-C18MPC, the reso-
nance at 0.9 ppm is due to the terminal methyl group of
octadecanethiolate). As expected, the resonances associated
with the protons α, β, and γ to the thiolate moiety are not ob-
served: because the substrates are anchored to the MPC sur-
face they are unable to rotate freely, and the signal broadens
into the baseline. There is some overlap of the broad reso-
nances attributed to the terminal methyl group of the dilutant
chain and those of the methylene groups of both 1 and the
dilutant chain in the 1-C12 and 1-C18MPCs. This necessitates
the use of an I2 decomposition reaction that quantitatively
liberates any thiolate bound to the MPC surface as disulfide
and reduces the MPC core to elemental gold, allowing accu-
rate integration to determine their stoichiometries (40). As
mentioned above, the MPCs employed in the study have a
general stoichiometry of Au314(X)108, where X is either
C12S, C18S, or MeO2CC10S. The total number of substrates
bound to the MPC surface is expected to remain constant af-
ter place-exchange, so the stoichiometry is based on the 1:X
Irradiations were carried out in argon- or N2-purged ben-
zene-d6 solutions in Pyrex NMR tubes using a Rayonet pho-
tochemical reactor fitted with 350 nm bulbs and a merry-go-
round apparatus. Generally, the solutions contained ~15–
20 mg of 1-C12MPC, 1-C18MPC, or 1-MeO2CC10MPC dis-
ratio. The stoichiometries of the MPCs were Au314
-
© 2003 NRC Canada

488
Can. J. Chem. Vol. 81, 2003
Scheme 3. The photochemical generation of 2-DMAD from 2,
which serves as the model for this reaction on the MPC surface.
solved in 0.5 mL benzene-d6 (~0.008–0.012 M with respect
to 1), with a three-times molar excess of the dienophile,
namely dimethyl acetylenedicarboxylate (DMAD). Irradia-
tion under these conditions produced MPCs consistent with
the trapping of the photodienol (Scheme 1). The reaction can
be monitored directly by NMR spectroscopy; this is illus-
trated in Figs. 2 and 3 for 1-C12MPC and 1-MeO2CC10MPC,
respectively, where the MPCs were irradiated in the pres-
ence of DMAD as the dienophile. The 1H NMR and IR
spectra of the MPCs irradiated in the presence of DMAD
can be compared with those of an authentic sample of the
Diels–Alder products made from irradiation of [4-
the MPC with 1. The extent of reaction was also not affected
by the relative concentration of DMAD: the concentration
was varied from 1 to 10 times the molar excess of 1. Though
the physical environment of the MPC was different in each
of these MPCs studied, the reaction only proceeds to ~65%
conversion in all cases. Assuming that the substrates bound
to the MPC are distributed over the entire MPC surface be-
cause the place-exchange reaction employed was carried out
over 5 days, the extent of reaction may be related to the
number of 1 bound at the edge and vertice positions of the
MPC as compared with the number bound to the terrace.
That is, there may be a site-dependent reactivity associated
with the substrates bound to the MPC surfaces.
As mentioned above, in the MPCs studied 1 should be po-
sitioned in each of the three distinct sites (i.e., on the edge,
vertice, or terrace) (Fig. 4). It is assumed that every 1 bound
to the MPC is capable of forming the photodienol. If this is
the case, there must be certain sites on the MPC surface
where the photodienol can be more efficiently trapped as the
Diels–Alder adduct and others where the reaction is less ef-
ficient. It is intriguing to propose that there is a correlation
between the position of substrates on the MPC core and the
extent of reaction. To investigate this, the place-exchange re-
action was exploited to selectively populate either the edge
and vertice sites or the terrace sites with 1.
(dodecyl)phenyl](2-methylphenyl)methanone
(2),
with
DMAD as dienophile (Scheme 3). The irradiation product of
1
2 with DMAD was characterized via H NMR, 13C NMR,
1
and IR spectroscopy, as well as mass spectrometry. The H
NMR spectrum of 2-DMAD is shown in Fig. 2c. The key
spectral changes that occur upon irradiation of 1-C12MPC,
1-C18MPC, and 1-MeO2CC10MPC with DMAD are consis-
tent with the formation of the Diels–Alder product. These
include the appearance of two overlapping, broad aromatic
peaks at 7.6 and 7.5 ppm and another aromatic resonance
signal at 6.87 ppm, as well as broad peaks at 4.15 (the
hydroxyl H), 3.62 (the methylene H’s), and 3.35 ppm (the
methyl ester H’s), concomitant with a decrease in the inten-
sity of broad aromatic peaks at 7.85, 7.22, and 7.05, along
with the broad resonance at 2.39 ppm (the ortho methyl
1
group of 1) (Figs. 2 and 3). The H NMR spectra comparing
1-C12MPC, 1-DMADC12MPC, and the model compound 2-
DMAD are shown in Figs. 2a–c, respectively. The IR spec-
tra of the irradiated MPCs are consistent with the product as
well, evidenced by the growth of peaks between 1715–
1730 cm–1 (consistent with an ester functionality) and be-
tween 3434–3468 cm–1 (consistent with a hydroxyl moiety).
Product formation was generally complete after 48–60 h
with no sign of further reaction on extended irradiation. The
success of these trapping experiments allows us to formulate
some ideas about the reactivity and steric constraints of sub-
strates confined to the MPC surface.
We begin our analysis of the Diels–Alder trapping of the
photodienol by comparing the extents of reaction for the
cases where the photodienol is directly at the interface (1-
C12MPC), when it is embedded within the monolayer (1-
C18MPC), and when the local polarity of the interface is var-
ied (1-MeO2CC10MPC). The reaction proceeds efficiently,
generating only one product on extensive irradiation on all
of the MPCs studied. Irradiation of Au314(C12S)50(1)58,
Au314(C18S)74(1)34, and Au314(MeO2CC10S)59(1)49 produces
MPCs with final stoichiometries of Au314(C12S)50(1)26(1-
DMAD)32 (64 2% conversion to the Diels–Alder adduct),
Selective population of the edge and vertice sites occurs if
short (~1 h) place-exchange reactions are carried out (36).
Using this strategy, C12MPC was stirred in the presence of 1
in toluene for 1 h under a nitrogen atmosphere, generating
1(edge)-C12MPC, where 1(edge) implies that the 1 is posi-
tioned predominantly at the edge and vertice sites of a
C12MPC, as suggested by Murray and co-workers.3 The
MPC was purified and I2 decomposition was utilized to de-
termine the stoichiometry of this MPC, which was
Au314(C12S)71(1)37. The MPC was then weighed out
(~15 mg), dissolved in benzene-d6, and irradiated in the
presence of DMAD (three-times excess) until there were no
1
further change in the H NMR spectrum (Fig. 5). The spec-
tral changes were similar to those explained for the above
MPCs. Upon purification to remove the excess DMAD, the
stoichiometry of the irradiated MPC was found to be
Au314(C12S)71(1)5(1-DMAD)32, which translates to an 85
3% conversion to Diels–Alder adduct based on two irradia-
tions. This is a significant increase in the conversion of 1 to
the Diels–Alder adduct, suggesting that 1 positioned at the
edge and vertice more readily undergo the reaction. How-
ever, this result does not indicate if 1 bound to the terrace is
less reactive towards the Diels–Alder reaction. This can only
be determined by preparing an MPC where 1 has been selec-
tively positioned on the terrace.
Au314(C18S)74(1)13(1-DMAD)21 (60
2%), and Au314-
(MeO2CC10S)59(1)15(1-DMAD)34 (69%), respectively. The
irradiations were carried out at least twice for Au314(C12S)50-
(1)58 and Au314(C18S)74(1)34, whereas Au314(MeO2CC10S)59-
(1)49 was irradiated only once in parallel with Au314(C12S)50-
(1)58 (Table 1).
These conversions are essentially identical, suggesting
that the ability of the reaction to proceed is not affected by
the length or the polarity of the dilutant chain co-absorbed to
3 This is known to populate some of the terrace as well; see ref. 36.
© 2003 NRC Canada

Kell et al.
489
Table 1. Stoichiometries of the original MPCs and the extents of reaction for their irradiations in the presence of DMAD.
Stoichiometry (before
irradiation)
Stoichiometry (after irradiation in
presence of DMAD)b
Conversion
(%)c
MPCa
d
1-C12MPC
1-C18MPC
1-C10CO2MeMPC
1(edge)-C12MPC
1(terrace)-C12MPC
Au314(C12S)50(1)58
Au314(C18S)74(1)34
Au314(C12S)50(1)26(1-DMAD)32
Au314(C18S)74(1)13(1-DMAD)21
Au314(MeO2CC10S)59(1)15(1-DMAD)34
Au314(C12S)71(1)5(1-DMAD)32
Au314(C12S)90(1)11(1-DMAD)7
64
60
69
85
36
2
2
d
e
Au314(MeO2CC10S)59(1)49
d
Au314(C12S)71(1)37
Au314(C12S)90(1)18
3
2
d
aFor example, 1-C12MPC represents 1 place exchanged onto an original dodecanethiolate MPC.
1
bDetermined via ratio integrations for of resonances attributed to 1 (7.85 ppm) and 1-DMAD (7.6–7.5 ppm) in the H NMR spectrum, uncertainty is 5%.
cConversion determined as (the number of 1-DMAD ligands generated)/(number of 1 ligands on starting MPC).
dDetermined via ratio of integrations for the mixed disulfides generated upon I2 decomposition of the MPC; uncertainty is 5%.
eDetermined via ratio of integrations for the substrates on the actual MPC in solution; uncertainty is 5%.
Fig. 4. A cartoon representation of the MPC core and the assign-
ments of the edge, vertice, and terrace sites.
Fig. 5. The irradiation of Au314(C12S)71(1)37 (where 1 is posi-
tioned predominantly at the edge and vertice sites on the MPC
surface) in benzene-d6 (*) before irradiation (a) and after an irra-
diation period fo 96 h in the presence of a 3-times molar excess
of DMAD (b). The filled arrows indicate the decrease in inten-
sity for the resonances associated with Au314(C12S)71(1)37 while
the hollow arrows indicate an increase in the resonances associ-
ated with the product, namely Au314(C12S)71(1)5(1-DMAD)32.
1
Note: the DMAD was washed away before the H NMR spec-
trum was acquired for (b).
Selectively populating the terrace with 1 is not as straight-
forward as populating the edge and vertice sites. The proce-
dure involves initial preparation 1-C12MPC in a place-
exchange reaction carried out over 5 days. This should popu-
late the terrace, edge, and vertice sites with 1. The resulting
MPC (with a stoichiometry of Au314(C12S)50(1)58) is then
subjected to a subsequent place-exchange reaction in the
presence of an excess of dodecanethiol (C12SH) for 1 h,
which should populate predominantly the edge and vertice
sites with dodecanethiol and displace any 1 positioned there.
Conveniently, any 1 on the terrace should be trapped, allow-
ing for the study of the reaction between its photodienol and
DMAD. The stoichiometry of the resulting MPC was deter-
mined as for the other MPCs and found to be Au314
-
(C12S)90(1)18 and will be referred to as 1(terrace)-C12MPC,
where 1(terrace) implies that 1 is predominantly at the ter-
race and the dilutant chain is C12S. Upon irradiation of 1(ter-
race)-C12MPC in the presence of a three-times molar excess
of DMAD, similar spectral changes were observed as for the
previous MPCs, and the final stoichiometry of the MPC was
Au314(C12S)90(1)11(1-DMAD)7, which translates to a 36
2% conversion to the Diels–Alder adduct based on two irra-
diations (Fig. 6). This conversion is much lower than that for
the MPCs containing 1 at the edge and vertice positions,
suggesting that the reactivity on the MPC surface is position
dependent.
It is reasonable to assume that the edge and vertice sites,
which may provide more room for a bimolecular reaction to
occur based on the highly faceted shape of the MPC core
(Fig. 4), are the positions on the MPC surface where the
Diels–Alder reaction is occurring more efficiently. The idea
of the edge and vertice sites anchoring substrates with less
order has been reported recently through an investigation
that described how intracluster hydrogen bonding decreases
the rate of cyanide-induced MPC decomposition (41), pre-
sumably because the highly faceted gold core prohibits ef-
fective intracluster chain interactions directly at the edge and
vertice sites in the absence of substrates capable of hydrogen
bonding. Our own work also suggests that mobility con-
straints imposed by aryl ketones anchored to terrace sites on
MPCs prevent the unimolecular Norrish–Yang Type II reac-
tion from occurring, whereas the reaction occurs more
readily at the edge and vertice sites (23, 24). Also of note is
a study involving the SN2 reaction of amines and MPC-
bound terminal bromides (40). The authors found the SN2
reaction proceeds quite efficiently (at least 80% completion)
regardless of the bulkiness of the amine. Though the product
conversions for this simple reaction are higher than we re-
port, the authors employed MPCs smaller (~140 gold atoms
in the MPC core) than those we used. The smaller MPC
© 2003 NRC Canada

490
Can. J. Chem. Vol. 81, 2003
Fig. 6. The irradiation of Au314(C12S)90(1)18 (where 1 is posi-
tioned predominantly at terrace sites on the MPC surface) in ben-
zene-d6 (*) before irradiation (a) and after an irradiation period fo
96 h in the presence of a 3-times molar excess of DMAD (b).
The filled arrows indicate the decrease in intensity for the reso-
nances associated with Au314(C12S)90(1)18 while the hollow ar-
rows indicate an increase in the resonances associated with the
product, namely Au314(C12S)71(1)11(1-DMAD)7. Note: the DMAD
edge and vertice sites are populated with 1 quickly, resulting
in the displacement of dodecanethiol into solution. Conse-
quently, both dodecanethiol and 1 will be in solution, but
excess 1 is employed in the place-exchange reaction, so
there will always be a higher concentration of 1 in solution.
Because population of the MPC during the place-exchange
reaction is dependent on the concentration of thiol in solu-
tion, and because the edge and vertice sites are populated
most easily, they may be populated to a greater extent with
1
was washed away before the H NMR spectrum was acquired for
(b).
1. This could account for the conversion of the Au314
-
(C12S)50(1)58, Au314(C18S)74(1)34, and Au314(MeO2CC10)59-
(1)49 to Au314(C12S)50(1)26(DMAD-1)32 (64 2% conversion
to Diels–Alder adduct), Au314(C18S)74(1)13(DMAD-1)21
(62 2%), and Au314(MeO2CC10)59(1)15(1-DMAD)34 (69%),
respectively, being slightly higher than expected.
Also of note are the results of irradiations of 1-C12 and 1-
C18 MPCs in the presence of DMFr and DMM. The spectral
1
changes observed in the H NMR spectra are consistent with
the generation of the Diels–Alder adducts, with the final
stoichiometry of the photolysed MPCs being Au314(C12S)50-
(1)21(1-DMFr)37 (64% conversion), Au314(C18S)74(1)13(1-
DMFr)21 (63% conversion), Au314(C12S)50(1)22(1- DMM)36
(61% conversion), and Au314(C18S)74(1)15(1-DMM)19 (58%
conversion), respectively. The average conversion for these
reactions is 62 4%, which is, within experimental error,
what we would expect based on the results of the analogous
reaction with DMAD. These reactions show a variety of
dienophiles can be employed in this reaction, and we are
working on exploiting this aspect of the chemistry. The spec-
tral data for the DMFr- and DMM-modified MPCs are pro-
vided in the supplemental information.
cores may result in more disordered monolayers. There have
also been studies carried out on 2-D self-assembled
monolayers (SAMs) suggesting that if edge sites are pro-
duced within the monolayer, the substrates at the edge sites
are much more mobile and floppy, as evidenced by their
ability to “trap” embedded groups (42, 43).
Experimental
Commercial solvents and reagents used
If the conversion were related to the number of edge and
vertice sites on an MPC one may assume that it would be di-
rectly related to the percentage of edge and vertice sites on
the MPC itself. Our results indicate that the conversion to
Diels–Alder adduct is ~65%. A rough calculation4 indicates
that, assuming equal distribution over all sites on the MPC,
41% of 1 should reside at the edge and vertice sites. If the
substrates at the edge and vertice sites and one atom row ad-
jacent to the edge and vertice sites were able to undergo the
reaction, this would account for 86% of the sites on the
MPC surface. Because the conversion lies somewhere be-
tween these numbers, we offer the following explanation:
there is the possibility that there is not an equal distribution
of 1 over the entire surface of the MPC and slightly more 1
is concentrated at the edge and vertice sites. If this were
true, it could explain why the conversion to Diels–Alder
adduct is higher than the percentage of edge and vertice sites
on the MPC surface. It is also possible that some of the 1 di-
rectly adjacent to the edge and vertice is capable of reacting,
which would increase the conversion as well. We prefer the
former explanation, based on the mechanism of place-
exchange. At the beginning of the place-exchange reaction
there is much more 1 than dodecanethiol in solution. The
The compounds dodecanethiol, octadecanethiol, hydrogen
tetrachloroaurate(III), tetraoctylammonium bromide, sodium
borohydride, 1,11-dibromoundecane, 1.8 M phenyl lithium
in cyclohexane-ether, o-toluoyl chloride, potassium thio-
acetate, dimethyl acetylenedicarboxylate, dimethyl maleate,
and dimethyl fumarate were all purchased from Aldrich and
used as received. Potassium carbonate (Caledon), aluminum
chloride (BHD), benzene-d6 (Cambridge Isotope Labora-
tories), and iodine (BDH) were also used as received. Ace-
tone, dichloromethane, benzene, toluene, methanol, diethyl
ether, and hexanes were purchased from either Caledon or
EM Science and used as received. Tetrahydrofuran was dried
by distillation from sodium/benzophenone. Ethanol (both an-
hydrous and 95%) was purchased from Commercial Alco-
hols Inc. Silica gel was purchased from EM Science.
General instrumentation
1H NMR spectra were recorded on a Varian Mercury 400
(400.087 MHz) spectrometer in either deuteriochloroform or
benzene-d6 solutions and are reported in parts per million
(ppm) with respect to chloroform or benzene peaks at
7.26 ppm or 7.15 ppm, respectively. 13C NMR spectra were
4 Our calculation assumes equal distribution of the substrate over the entire surface and involves dividing the number of atoms directly at
edge and vertice sites and by the total number of surface atoms (41%) or dividing the number of atoms at the edge and vertice sites and
those directly adjacent to the edge and vertice sites by the total number of surface atoms (86%).
© 2003 NRC Canada

Kell et al.
491
recorded on a Varian Mercury 400 (100.602 MHz) spec-
trometer in either deuteriochloroform or benzene-d6 solution
and are reported in parts per million with respect to chloro-
form or benzene peaks at 77.0 or 128.02 ppm. UV–vis ab-
ammonium chloride (30 mL), washed with 3 × 30 mL of
distilled water, and dried over MgSO4 for 20 min. Concen-
tration yielded a yellowish oil that was purified by column
chromatography using silica gel and 3:1 dichloro-
sorption spectra were recorded on
a
Cary 100Bio
methane:hexanes as eluant; a colorless oil (0.5 g,
spectrometer in spectrometry-grade benzene. Mass spectra
and exact masses were recorded on a MAT 8200 Finnigan
high resolution mass spectrometer; the latter employed a
mass of 12.0000 for carbon. IR spectra were recorded on a
Bomem MB-Series or a Bruker Vector 33 spectrometer us-
ing a dropcasting technique on NaCl plates and are reported
in wavenumbers (cm–1).
1.30 mmol) was produced in 76% yield. UV–vis (benzene)
(nm) (ε (M–1cm–1)): 339 (9.790 × 10), 277 (7.832 × 103).
IR (cm–1) (dropcast on NaCl): 2925, 2853, 1661, 1604,
1
1267. H NMR (400 MHz, CDCl3) (ppm) δ: 7.73 (d, J =
7.8 Hz, 2H), 7.38 (m, H), 7.28 (m, 5H), 2.67 (t, J = 7.8 Hz,
2H), 2.52 (quartet, J = 7.0 Hz, 2H), 2.33 (s, 3H), 1.61 (m,
4H), 1.44–1.19 (m, 15H (includes SH proton)). 13C NMR
(400 MHz, CDCl3) (ppm) δ: 198.36, 148.99, 138.90, 136.45,
135.24, 130.82, 130.29, 129.94, 128.51, 128.28, 125.08,
36.06, 34.05, 31.16, 29.51, 29.48, 29.45, 29.40, 29.25,
29.02, 28.34, 24.63, 19.91. EI-MS m/z (%): 382 (20), 364
(6), 223 (8), 195 (100), 119 (23), 84 (20), 49 (33). Exact
mass calcd.: 382.2330; found: 382.2329.
Steady-state photolysis experiments
Steady-state irradiation experiments were carried out in
septa-sealed Pyrex NMR tubes using benzene-d6 as solvent
or 5 mL Pyrex photolysis cells using reagent-grade benzene
as solvent. The light source was a Rayonet photochemical
reactor fitted with bulbs that emitted UV light in the 300–
400 nm range, with a maximum at 350 nm and a merry-go-
round apparatus to ensure an equal amount of radiation was
received. In a typical procedure, ~15–20 mg of MPC was
weighed out and placed under vacuum to ensure all solvent
was removed. The MPC was then dissolved in ~0.5 mL ben-
zene-d6, degassed for 15 min with either argon or nitrogen
[4-(11-Dodecyl)phenyl](2-methylphenyl)methanone (2)
To a flame-dried, 10-mL round-bottom flask fitted with an
argon inlet and condenser was added aluminum chloride
(1.8 g, 13.5 mmol), dichloromethane (7 mL), and phenyldo-
decane (2.5 g, 10.0 mmol), and the mixture was cooled in a
salted ice bath. A solution containing o-toluoyl chloride
(1.7 g, 11.0 mmol) in 2 mL of dichloromethane was then
added to the mixture over 1 min. This mixture was left stir-
ring for 3 h while warming to room temperature. The reac-
tion was then quenched by pouring the entire contents of the
flask into a beaker containing 40 mL of distilled water
cooled in an ice bath. The organic layer was diluted with
20 mL dichloromethane, washed with 4 × 50 mL of distilled
water, and dried over MgSO4 for 30 min. Concentration of
the dried organic phase yielded a yellow oil, which was puri-
fied by column chromatography using silica gel and 1:1 hex-
anes:dichloromethane as eluant, generating a clear, colorless
liquid (1.83 g, 5.02 mmol) in 50% yield. IR (cm–1) (dropcast
on NaCl): 2926, 2853, 1662, 1605, 1264. 1H NMR
(400 MHz, CDCl3) (ppm) δ: 7.72 (d, J = 7.8 Hz, 2H), 7.37
(t, H), 7.30–7.21 (m, 5H), 2.65 (t, J = 2H), 2.31 (s, 3H), 1.62
1
gas, and sealed with a rubber septum and Parafilm. An H
NMR spectrum was recorded prior to the addition of
dienophile and at intermittent reaction times until the reac-
tion was complete. The temperature of the solutions was
typically 38 2°C.
[4-(11-Mercaptoundecyl)phenyl](2-methylphenyl)metha-
none (1)
To
a
solution of 1,11-dibromoundecane (5.2 g,
16.5 mmol) in dry THF (25 mL) was added phenyl lithium
(3.7 mL of a 1.8 M solution, 6.7 mmol) dropwise, generat-
ing 11-phenyl-1-bromoundecane in situ. After aqueous
workup and drying, the 11-phenylbromoundecane – 1,11-di-
bromoundecane mixture was redissolved with 25 mL di-
chloromethane and cooled in a salted ice bath. When the flask
was cooled, o-toluoyl chloride (1.12 g, 7.26 mmol) and alu-
minum chloride (1.01 g, 7.59 mmol) were added, and the
mixture was stirred for 4 h, maintaining a temperature of 0°C.
Upon aqueous workup and drying with MgSO4, the resulting
[4-(11-bromoundecyl)phenyl](2-methylphenyl)methanone was
purified via gradient column chromatography on silica gel,
beginning with 10:1 hexanes:dichloromethane to elute the
unreacted dibromoundecane and ending with 2:1 hexanes:di-
chloromethane which eluted [4-(11-bromoundecyl)phenyl](2-
methylphenyl)methanone (0.90 g) as a clear colorless oil.
The bromide was converted to the thioacetate through a re-
action with potassium thioacetate in acetone, quantitatively
generating [4-(11-acetylsulfanylundecyl)phenyl](2-methyl-
phenyl)methanone. The thioacetate (0.72 g, 1.7 mmol) was
then transferred to a 100-mL round-bottom flask fitted with
a reflux condenser, dissolved in absolute ethanol, and the so-
lution degassed with argon for 15 min after which time the
solution was charged with potassium carbonate (0.24 g,
2.07 mmol) and heated to reflux for 3 h. The ethanol was re-
moved via rotary evaporation, and the resulting liquid was
redissolved in dichloromethane and washed with saturated
1
(quintet, 2H), 1.37–1.21 (m, 18H), 0.86 (t, 3H). H NMR
(400 MHz, C6D6) (ppm) δ: 7.84 (d, 2H J = 8.6 Hz), 7.20 (t,
J = 7.8 Hz, H), 7.07 (m, H), 7.01–6.89 (m, 4H), 2.40 (t, J =
7.8 Hz, 2H), 2.28 (s, 3H), 1.45 (broad quintet, 2H), 1.36–
1.16 (m, 18H), 0.91 (t, J = 7.0 Hz, 3H). 13C NMR
(400 MHz, CDCl3) (ppm) δ: 198.36, 149.04, 138.95, 136.48,
135.26, 130.85, 130.30, 129.95, 128.49, 128.25, 125.09,
36.04, 31.90, 31.11, 29.64, 29.61, 29.54, 29.45, 29.33,
29.28, 22.67, 14.11. EI-MS m/z (%): 364 (3.5), 195 (100),
119 (9), 91 (8). Exact mass calcd.: 364.2766; found:
364.2765.
Dodecanethiolate MPC (C12MPC)
Following the procedures of Brust et al. (34) and Murray
and co-workers (35): to a 250-mL round-bottom flask was
added hydrogen tetrachloroaurate(III) trihydrate (0.30 g,
0.768 mmol) dissolved in 28 mL distilled water (resulting in
a bright yellow solution) and tetraoctylammonium bromide
(2.01 g, 0.369 mmol) dissolved in 70 mL toluene (a clear
and colorless solution). The contents were rapidly stirred for
30 min at room temperature to facilitate the phase transfer of
© 2003 NRC Canada

492
Can. J. Chem. Vol. 81, 2003
the hydrogen tetrachloroaurate(III) trihydrate into the tolu-
ene layer, which resulted in the organic layer turning a dark
orange color and the aqueous layer becoming clear and col-
orless. After phase transfer, the aqueous layer was removed
and dodecanethiol (0.15 g, 0.18 mL, 0.762 mmol) was added
via a volumetric pipet to the solution, which was allowed to
stir at room temperature while a fresh solution of sodium
borohydride (0.33 g, 8.68 mmol) in 18 mL water was pre-
pared. The aqueous sodium borohydride was added to the
toluene solution over ~5 s and the mixture was allowed to
stir at room temperature overnight (~18 h). The organic
layer was washed with 3 × 20 mL distilled water, dried with
MgSO4, and concentrated. The concentrated MPC was then
suspended in 200 mL of 95% ethanol and placed in the
freezer overnight, during which time the C12MPC precipi-
tated from solution. The ethanol was then decanted and the
MPC was dissolved in benzene and concentrated, resulting
in the formation of a film in the round-bottom flask. This
film was washed repeatedly with 10 × 15 mL of 95% etha-
nol warmed to 40°C. The MPC was pure, according to the
1H NMR spectrum, which showed no signs of dodecan-
ethiol, dodecyldisulfide, or tetraoctylammonium bromide.
had a stoichiometry of Au314(C12S)71(1)37. Population of the
terrace with 1 was accomplished by stirring the Au314
-
(C12S)50(1)58 (0.050 g) (prepared via the 5 day place-
exchange reaction described above) dissolved in 10 mL tolu-
ene with an excess of dodecanethiol (0.10 g) under a nitro-
gen atmosphere for 1 h. The resulting MPC had a
stoichiometry of Au314(C12S)90(1)18.
[4-(11-Mercaptoundecyl)phenyl](2-methylphenyl)metha-
none – octadecanethiolate MPC (1-C18MPC)
Following the procedure outlined above for Au314(SC12)50-
(1)58, C18MPC (0.05 g, 0.044 mmol octadecanethiolate) was
dissolved in 15 mL toluene in a 100-mL round-bottom flask
fitted with an argon inlet, to which [4-(11-mercap-
toundecyl)phenyl](2-methylphenyl)-methanone (0.0226 g,
0.059 mmol) dissolved in ~3 mL toluene was added. After
stirring for 4 days, the toluene was removed by rotary evapo-
ration, and the mixed MPC was washed with a warmed (35–
40°C) 6:1 ethanol:toluene solution to remove excess [4-(11-
mercaptoundecyl)phenyl](2-methylphenyl)methanone
and
octadecanethiol. The resulting MPC had a composition of
Au314(C18S)74(1)34, as determined via iodine decomposition.
Octadecanethiolate MPC (C18MPC)
[4-(11-Mercaptoundecyl)phenyl](2-methylphenyl)metha-
none – 11-mercaptoundecanoic acid methyl ester MPC
(1-MeO2CC10MPC)
The octadecanethiolate MPC was synthesized as described
above for the dodecanethiolate MPC. The procedure in-
volved 0.20 g (0.51 mmol) hydrogen tetrachloroaurate,
1.32 g (2.41 mmol) tetraoctylammonium bromide, 0.20 g
(5.59 mmol) sodium borohydride, and 0.146 g (0.51 mmol)
octadecanethiol.
Following the procedure outlined above for Au314(SC12)50-
(1)58, MeO2CC10MPC (0.15 g, 0.16 mmol 11-merca-
ptoundecanoic acid methyl ester) was dissolved in 40 mL to-
luene in a 100-mL round-bottom flask fitted with an argon
inlet, to which [4-(11-mercaptoundecyl)phenyl](2-methyl-
phenyl)methanone (0.073 g, 0.19 mmol) was added. After
stirring for 4 days, the toluene was removed by rotary
evaporation, and the mixed MPC was washed with ethanol
to remove excess [4-(11-mercaptoundecyl)phenyl](2-methyl-
phenyl)methanone and 11-mercaptoundecanoic acid methyl
11-Mercaptoundecanoic acid methyl ester MPC (MeO2-
CC10MPC)
The MPC was synthesized as described above for the
dodecanethiolate MPC. The procedure involved 0.30 g
(0.76 mmol) hydrogen tetrachloroaurate, 2.04 g (3.42 mmol)
tetraoctylammonium bromide, 0.32 g (8.36 mmol) sodium
borohydride, and 0.177 g (0.76 mmol) 11-mercaptounde-
canoic acid methyl ester. The MPC was purified by washing
with hexanes.
ester. The resulting MPC had a composition of Au314
-
(MeO2CC10S)59(1)49, as determined by direct integration of
the aromatic resonances at 7.85 ppm (resonances from pro-
tons α to carbonyl) and 3.42 ppm (resonances from protons
of the terminal ester) in benzene-d6 solution.
[4-(11-Mercaptoundecyl)phenyl](2-methylphenyl)metha-
none – dodecanethiolate MPC (1-C12MPC)
MPC decomposition procedure
As reported by Murray and co-workers (40), approxi-
mately 10 mg of mixed MPC was placed in a 10-mL round-
bottom flask and dissolved in 2 mL dichloromethane. A
small crystal of iodine (~1 mg) was added, and the solution
was stirred until the originally brown, opaque solution be-
came clear and a light purple colour with a black precipitate
(20 min). The dichloromethane was rotary evaporated, and
0.5 mL deuteriochloroform was added to redissolve the re-
sulting mixed disulfide. This was placed in an NMR tube by
first passing it through a pipette equipped with a Kimwipe
filter to remove the precipitate. The sample was then ana-
Following the procedure outlined by Murray and co-
workers (36, 37), C12MPC (0.11 g, 0.136 mmol of dodecan-
ethiol) was dissolved in 34 mL toluene in a 100-mL round-
bottom flask fitted with an argon inlet. Excess [4-(11-mercap-
toundecyl)phenyl](2-methylphenyl)methanone (0.0629 g,
0.163 mmol) was then added to the flask and the mixture
was stirred for 4–5 days under argon. After 4–5 days the so-
lution was concentrated and the mixed MPC was washed
with warm (35–40°C) 95% ethanol to remove excess [4-(11-
mercaptoundecyl)phenyl](2-methylphenyl)-methanone and
1
dodecanethiol. The H NMR (400 MHz, C6D6) spectra indi-
1
lyzed by H NMR spectroscopy, where the integrations for
cated that the [4-(11-mercaptoundecyl)-phenyl](2-methyl-
phenyl)methanone was incorporated onto the MPC, and
iodine-induced decomposition indicated that the MPC had a
composition of Au314(C12S)50(1)58. Population of the edge
and vertice sites with 1 was accomplished using the same
conditions but with a decreased time period over which the
exchange reaction took place (1 h) (36). The resulting MPC
the appropriate resonance signals were compared.
Generation of 2-methyl-4′-(dodecyl)benzophenone Diels–
Alder adducts
In a typical procedure, [4-(11-dodecyl)phenyl](2-methyl-
phenyl)methanone (2) (0.1 g, 0.41 mmol) was dissolved in
© 2003 NRC Canada

Kell et al.
493
5 mL of benzene, transferred to a 5 mL Pyrex photolysis cell
sealed with a septum, and the solution was degassed with ar-
gon for 20 min. The dienophile (namely dimethyl acetylene-
dicarboxylate, dimethyl fumarate, or dimethyl maleate)
(0.41 mmol) was then added to the degassed solution via sy-
ringe, and the mixture was irradiated for 12 h. The product
was purified via column chromatography, employing 1%
methanol in dichloromethane as eluant.
we have found for unimolecular reactions carried out on
identical MPC cores (23, 24). However, the reasons for the
lowered extents of reaction are not expected to be the same.
Mobility constraints play a definitive role in the efficiency
of unimolecular reactions carried out on MPCs, but in this
investigation, which involves bimolecular reactivity, sterics
and the ability to adopt specific orientations play key roles
in the efficiency of the reaction.
It is intriguing to propose that there are differences in re-
activity and dynamics associated with the distinct sites on
the MPC surface. We are currently investigating the effect
that increasing and decreasing the size of the MPC core has
on the extent of reaction, because the size of the terrace can
be easily manipulated by increasing or decreasing the size of
the metal core. This study will provide more insight on the
physical properties inherent to substrates when they are an-
chored to specific sites on these highly faceted gold cores.
2-DMAD Diels–Alder adduct
IR (cm–1) (dropcast on NaCl): 3474, 2924, 2853, 1732. 1H
NMR (400 MHz, C6D6) (ppm) δ: 7.55 (d, J = 8.6 Hz, 1H),
7.49 (d, J = 8.6 Hz, 2H), 6.9–7.05 (m, 4H), 6.82 (d, J =
7.8 Hz, 1H), 4.26 (s, 1H), 2.61 (s, 2H), 3.34 (s, 3H), 3.28 (s,
3H), 2.95 (t, J = 5.5 Hz, 2H), 2.46 (t, J = 7.0 Hz, 2H), 1.38–
1.60 (m, 4H), 1.0–1.3 (br, 14H) 0.85 (t, J = 7.0 Hz, 3H). 13C
NMR (400 MHz, C6D6) (ppm) δ: 168.06, 166.66, 143.40,
142.09, 141.84, 140.41, 130.78, 128.51, 127.50, 127.36,
126.13, 74.67, 52.17, 36.12, 32.59, 32.01, 31.20, 30.39,
30.36, 30.28, 30.19, 30.09, 30.00, 23.40, 14.69. EI-MS m/z
(%): 554 (45), 552 (42), 303 (6), 223 (8), 195 (100), 119
(23), 84 (20), 49 (33). Exact mass calcd.: 506.3032; found:
506.3028.
Acknowledgments
Financial support for this research from the Natural Sci-
ences and Engineering Research Council of Canada
(NSERC), the Academic Development Fund (UWO), the
Canadian Foundation for Innovation, and the Ontario Re-
search and Development Challenge Fund is gratefully ac-
knowledged. MSW thanks the Government of Ontario for a
Premier’s Research Excellence Award, and AJK thanks
NSERC for a graduate scholarship. We also thank Dr. Rob-
ert Donkers for helpful suggestions.
Conclusions
The Diels–Alder reaction of the photodienol generated
from 1 in the MPC environment is shown to proceed effi-
ciently, with similar extents of reaction regardless of the lo-
cal environment surrounding the reactive group. In addition,
this reaction exhibited evidence suggesting site-dependant
reactivity. It appears as though there are properties inherent
to substrates bound at specific sites on the MPC core that
render them more or less reactive toward the Diels–Alder re-
action. This is evidenced by the general trend observed with
respect to the reactivity of 1 on the MPC surface: in the
comparison of the reactivity of the multi-site-populated
MPC — where all of the sites on the MPC are populated (1-
C12MPC) — placing 1 predominantly at terrace sites (1(ter-
race)-C12MPC) results in significantly lower conversions,
while placing 1 predominantly at edge and vertice sites
(1(edge)-C12MPC) results in significantly higher conver-
sions. In fact, quantitatively speaking, the same amount of 1
reacts efficiently when the entire MPC is populated (1-
C12MPC) as when the majority of 1 is at the edge and
vertice sites (1(edge)-C12MPC); the number of 1 converted
to 1-DMAD is 32 in each case. Qualitatively, we attribute a
more hindered environment within the terrace of the MPC as
the main cause of the lowered extent of conversion. The
most accessible sites — those at the edge and vertice — on
the MPCs can undergo this rather complex bimolecular reac-
tion more efficiently. Those within the terrace (where the
monolayer is very well packed and more accurately mimics
that of a 2-D SAM) are more hindered and less likely to un-
dergo the rather complex bimolecular reaction. It is difficult
to predict if the decreased ability for the terrace-bound
dienol to react is due to steric constraints that do not allow
the dienophile to reach the photodienol or if the dienophile
is able to reach the dienol but cannot achieve the geometry
required for the reaction to occur. The inability for these bi-
molecular reactions to reach completion is similar to what
References
1. A.C. Weedon. In The chemistry of enols. Edited by Z.
Rappoport. John Wiley and Sons, Toronto, Ont., Canada. 1990.
p. 615.
2. R.M. Wilson. In Organic photochemistry. Edited by A. Padwa.
Marcel Dekker Inc., New York. 1985. p. 339.
3. N.C. Yang and C. Rivas. J. Am. Chem. Soc. 83, 2213 (1961).
4. G.J. Jocys and M.S. Workentin. J. Chem. Soc., Chem.
Commun. 839 (1999).
5. J.L. Pitters, M. Kovar, K. Griffiths, P.R. Norton, and M.S.
Workentin. Angew. Chem., Int. Ed. 39, 2144 (2000).
6. L.F. Rozsnyai and M.S. Wrighton. J. Am. Chem. Soc. 116,
5993 (1994).
7. E.W. Wollman, D. Kang, C.D. Frisbie, I.M. Lorkovic, and
M.S. Wrighton. J. Am. Chem. Soc. 116, 4395 (1994).
8. M.O. Wolf and M.A. Fox. J. Am. Chem. Soc. 117, 1845 (1995).
9. M.O. Wolf and M.A. Fox. Langmuir, 12, 955 (1996).
10. I. Willner, A. Doron, and E. Katz. J. Phys. Org. Chem. 11, 546
(1998).
11. T. Kim, K.C. Chan, and R.M. Crooks. J. Am. Chem. Soc. 119,
189 (1997).
12. W. Li, V. Lynch, H. Thompson, and M.A. Fox. J. Am. Chem.
Soc. 119, 7211 (1997).
13. M.A. Fox and M.D. Wooten. Langmuir, 13, 7099 (1997).
14. S. Sortino, S. Petralia, G. Compagnini, S. Conoci, and G.
Condorelli. Angew. Chem. Int. Ed. 41, 1914 (2002).
15. O. Prucker, C.A. Naumann, J. Rühe, W. Knoll, and C.W.
Frank. J. Am. Chem. Soc. 121, 8766 (1999).
16. J. Zhang, J.K. Whitesell, and M.A. Fox. Chem. Mater. 13,
2323 (2001).
17. J. Hu, J. Zhang, F. Liu, K. Kittredge, J.K. Whitesell, and M.A.
Fox. J. Am. Chem. Soc. 123, 1464 (2001).
© 2003 NRC Canada

494
Can. J. Chem. Vol. 81, 2003
18. H. Imahori, M. Arimura, T. Hanada, Y. Nishimura, I.
Yamazaki, Y. Sakata, and S. Fukuzumi. J. Am. Chem. Soc.
123, 335 (2001).
33. E. Block and R. Stevenson. J. Chem Soc. Perkin Trans. 1, 308
(1973).
34. M. Brust, M. Walker, D. Bethell, D.J. Schiffrin, and R.J.
Whyman. J. Chem. Soc., Chem. Commun. 801 (1994).
35. M.J. Hostetler, J.E. Wingate, C.-J. Zhong, J.E. Harris, R.W.
Vachet, M.R. Clark, J.D. Londono, S.J. Green, J.J. Stokes,
G.D. Wignall, G.L. Glish, M.D. Porter, N.D. Evans, and R.W.
Murray. Langmiur, 14, 17 (1998).
19. T. Wang, D. Zhang, W. Xu, J. Yang, R. Han, and D. Zhu.
Langmuir, 18, 1840 (2002).
20. A. Aguila and R.W. Murray. Langmuir, 16, 5949 (2000).
21. A.C. Templeton, W.P. Wuelfing, and R.W. Murray. Acc. Chem.
Res. 33, 27 (2000).
22. V. Chechik, R.M. Crooks, and C.J.M. Stirling. Adv. Mater. 12,
1161 (2000).
36. Y. Song and R.W. Murray. J. Am. Chem. Soc. 124, 7096
(2002).
23. A.J. Kell and M.S. Workentin. Langmuir, 17, 7355 (2001).
24. A.J. Kell, D.L.B. Stringle, and M.S. Workentin, Org. Lett. 2,
3381 (2000).
25. Y. Kwon and M. Mrksich. J. Am. Chem. Soc. 124, 806 (2002).
26. M.N. Yousaf, E.W.L. Chan, and M. Mrksich. Angew. Chem.
Int. Ed. 39, 1943 (2000).
37. M.J. Hostetler, A.C. Templeton, and R.W. Murray. Langmuir,
15, 3782 (1999).
38. R.H. Terrill, T.A. Postlethwaite, C.-h. Chen, C.-D. Poon, A.
Terzis, A. Chen, J.E. Hutchison, M.R. Clark, G. Wignall, J.D.
Londono, R. Superfine, M. Falvo, C.S. Johnson, Jr., E.T.
Samulski, and R.W. Murray. J. Am. Chem. Soc. 117, 12 537
(1995).
27. M.N. Yousaf and M. Mrksich. J. Am. Chem. Soc. 121, 4286
(1999).
39. A. Badia, W. Goa, S. Singh, L. Demers, L. Cicca, and L.
28. E.W.L. Chan, M.N. Yousaf, and M. Mrksich. J. Phys. Chem.
Reven. Langmuir, 12, 1262 (1996).
A, 104, 9315 (2000).
29. A. Badia, R.B. Lennox, and L. Reven. Acc. Chem. Res. 33,
475 (2000).
30. R.L. Whetten, J.T. Khoury, M.M. Alvarez, S. Murthy, I.
Vezmar, Z.L. Wang, P.W. Stephens, C.L. Cleveland, W.D.
Luedtke, and U. Landman. Adv. Mater. 8, 428 (1996).
31. M. Pfau, E.W. Sarver, and N.D. Heindel. Bull. Soc. Chim. Fr.
1, 183 (1973).
40. A.C. Templeton, M.J. Hostetler, C.T. Kraft, and R.W. Murray.
J. Am. Chem. Soc. 120, 1906 (1998).
41. R. Paulini, B.L. Frankcamp, and V.M. Rotello. Langmuir, 18,
2368 (2002).
42. W. Fudickar, J. Zimmermann, L. Ruhlmann, J. Schneider, B.
Röder, U. Siggel, and J.-H. Fuhrhop. J. Am. Chem. Soc. 121,
9539 (1999).
43. M. Lahav, E. Katz, and I. Willner. Langmuir, 17, 7387 (2001).
32. M. Pfau, S. Combrisson, J.E. Rowe, and N.D. Heindel. Tetra-
hedron, 34, 3459 (1978).
© 2003 NRC Canada