Photonic amplification by a singlet-state quantum chain reaction in the photodecarbonylation of crystalline diarylcyclopropenones

Article abstract of DOI:10.1021/ja9043449

The photochemical decarbonylation of diphenylcyclopropenone (DPCP) to diphenylacetylene (DPA) proceeds with remarkable efficiency both in solution and in the crystalline solid state. It had been previously shown that excitation to the second electronic excited state (S₂) of DPCP in solution proceeds within ca. 200 fs by an adiabatic ring-opening pathway to yield the S₂ state of DPA, which has a lifetime of ca. 8 ps before undergoing internal conversion to S₁ (Takeuchi, S.; Tahara, T. J. Chem. Phys. 2004, 120, 4768). More recently, we showed that reactions by excitation to S₂ in crystalline solids proceed by a quantum chain process where the excited photoproducts transfer energy to neighboring molecules of unreacted starting material, which are able to propagate the chain. Quantum yields in crystalline suspensions revealed values of Φ_DPCP = 3.3 = 0.3. To explore the generality of this reaction, and recognizing its potential as a photonic amplification system, we have synthesized nine crystalline diarylcyclopropenone derivatives with phenyl, biphenyl, naphthyl, and anthryl substituents. To quantify the efficiency of the quantum chain in the crystalline state, we determined the quantum yields of reaction for all of these compounds both in solution and in nanocrystalline suspensions. While the quantum yields of decarbonylation in solution vary from Φ = 0.0 to 1.0, seven of the nine new structures display quantum yields of reaction in the solid that are above 1. The chemical amplification that results from efficient energy transfer in the solid state, analyzed in terms of the quantum yields determined in the solid state and in solution (Φ_cryst/Φ_soln), reveals quantum chain amplification factors that range from 3.2 to 11.0. The remarkable mechanical response of the solid-to-solid reaction previously documented with macroscopic crystals, where large single-crystalline specimens turn into fine powders, was investigated at the nanometer scale. Experiments with dry crystals of DPCP analyzed by atomic force microscopy showed the formation of DPA in the form of isolated crystalline specimens ca. 35 nm in size.

Full text of DOI:10.1021/ja9043449

Published on Web 07/28/2009
Photonic Amplification by a Singlet-State Quantum Chain
Reaction in the Photodecarbonylation of Crystalline
Diarylcyclopropenones
Gregory Kuzmanich, Matthew N. Gard, and Miguel A. Garcia-Garibay*
Department of Chemistry and Biochemistry, UniVersity of California,
Los Angeles, California 90024-1569
Received May 28, 2009; E-mail: mgg@chem.ucla.edu
Abstract: The photochemical decarbonylation of diphenylcyclopropenone (DPCP) to diphenylacetylene
(DPA) proceeds with remarkable efficiency both in solution and in the crystalline solid state. It had been
previously shown that excitation to the second electronic excited state (S2) of DPCP in solution proceeds
within ca. 200 fs by an adiabatic ring-opening pathway to yield the S2 state of DPA, which has a lifetime
of ca. 8 ps before undergoing internal conversion to S1 (Takeuchi, S.; Tahara, T. J. Chem. Phys. 2004,
120, 4768). More recently, we showed that reactions by excitation to S2 in crystalline solids proceed by a
quantum chain process where the excited photoproducts transfer energy to neighboring molecules of
unreacted starting material, which are able to propagate the chain. Quantum yields in crystalline suspensions
revealed values of ΦDPCP ) 3.3 ( 0.3. To explore the generality of this reaction, and recognizing its potential
as a photonic amplification system, we have synthesized nine crystalline diarylcyclopropenone derivatives
with phenyl, biphenyl, naphthyl, and anthryl substituents. To quantify the efficiency of the quantum chain
in the crystalline state, we determined the quantum yields of reaction for all of these compounds both in
solution and in nanocrystalline suspensions. While the quantum yields of decarbonylation in solution vary
from Φ ) 0.0 to 1.0, seven of the nine new structures display quantum yields of reaction in the solid that
are above 1. The chemical amplification that results from efficient energy transfer in the solid state, analyzed
in terms of the quantum yields determined in the solid state and in solution (Φcryst/Φsoln), reveals quantum
chain amplification factors that range from 3.2 to 11.0. The remarkable mechanical response of the solid-
to-solid reaction previously documented with macroscopic crystals, where large single-crystalline specimens
turn into fine powders, was investigated at the nanometer scale. Experiments with dry crystals of DPCP
analyzed by atomic force microscopy showed the formation of DPA in the form of isolated crystalline
specimens ca. 35 nm in size.
Scheme 1
Introduction
Quantum chain reactions are characterized by the formation
of more than one product molecule per photon of light absorbed
(Scheme 1). Unlike normal chain reactions involving radicals1
and other ground-state reactive intermediates,2 quantum chain
reactions involve chain carriers in the excited state.3,4 The
conditions required for a quantum chain reaction were first
(1) (a) Moad, G.; Solomon, D. H. The Chemistry of Radical Polymeri-
zation, 2nd ed.; Elsevier: Oxford, UK, 2006. (b) Griesbeck, A. G.;
Hoffmann, N.; Warzecha, K.-D. Acc. Chem. Res. 2007, 40, 128.
(2) (a) Marquez, C. A.; Wang, H.; Fabbretti, F.; Metzger, J. O. J. Am.
Chem. Soc. 2008, 130, 17208. (b) Maruyama, T.; Mizuno, Y.; Shimizu,
I.; Suga, S.; Yoshida, J.-I. J. Am. Chem. Soc. 2007, 129, 1902. (c)
Norton, J. E.; Olson, L. P.; Houk, K. N. J. Am. Chem. Soc. 2006,
128, 7835. (d) Baciocchi, E.; Bietti, M.; Putignani, P.; Steenken, S.
J. Am. Chem. Soc. 1996, 118, 5952. (e) Matmour, R.; More, A. S.;
Wadgaonkar, P. P.; Gnanou, Y. J. Am. Chem. Soc. 2006, 128, 8158.
(f) Waddell, W. H.; Lee, C. H. J. Am. Chem. Soc. 1982, 104, 5804.
(3) A quantum chain reaction was first suggested in the isomerization of
1,5,9-cyclodecatriene, but it was discarded on the basis of energetic
arguments. Crandall, J. K.; Mayer, C. F. J. Am. Chem. Soc. 1967, 89,
4374.
documented by Hammond.4 Quantum chain reactions start by
formation of an excited state (Step 1, Scheme 1) and are
followed by a relatively uncommon propagation step (Step 2)
involving an adiabatic photochemical reaction that gives rise
to the photoproduct in the excited state.5 This is followed by
energy transfer to a new ground-state reactant that propagates
the chain (Step 3). Termination occurs when either the excited-
(4) The first documented quantum chain reaction involves the triplet
sensitized photoisomerization of 2,4-hexadiene. Hyndman, H. L.;
Monroe, B. M.; Hammond, G. S. J. Am. Chem. Soc. 1969, 91, 2852.
(5) Turro, N. J.; McVey, J.; Ramamurthy, V.; Lechtken, P. Angew. Chem.,
Int. Ed. Engl. 1979, 18, 572–586.
9
11606 J. AM. CHEM. SOC. 2009, 131, 11606–11614
10.1021/ja9043449 CCC: $40.75  2009 American Chemical Society

Photodecarbonylation of Crystalline Diarylcyclopropenones
A R T I C L E S
Scheme 2
state reactant or product undergoes thermal or radiative decay
(Step 4 and/or Step 5).
As recently suggested by Gillmore et al., quantum amplified
reactions with long propagation lengths would be ideal for the
amplification of photonic signals, with potential in lithographic
and sensing applications.6 Most signal amplification strategies
developed to date deal with the optimal capture and use of every
photon absorbed, as exemplified by Swager’s molecular wire
approach7 and several related strategies.8 In contrast, quantum
chain reactions generate more than one chemical event per
photon, and harnessing their potential would be highly desirable.
Recent examples of quantum chain processes in materials
science include the use of polymeric materials undergoing
changes in refractive index as a result of radical ion chains,6
super-high-spin magnetic polycarbenes,8h or triplet quantum
chain9 reactions of Dewar benzene to Hu¨ckel benzene.
Despite their promising nature, the control of quantum chain
reactions represents a significant challenge, especially as it
pertains to the propagation events in Scheme 1. Limitations arise
from the relatively small number of adiabatic reactions devel-
oped at this time (Step 2) and from the challenges involved in
designing an efficient energy-transfer process (Step 3). If one
considers reactions that occur in solution, long chain reactions
will require excited states with long lifetimes and high reactant
concentrations. It is therefore not surprising that almost all
known quantum chain reactions occur in triplet states. Examples
include cis-trans isomerization of alkenes and polyenes,4,10,11
decomposition of dioxetanes to ground- and excited-state
ketones,12,13 and several bond-valence isomerizations.5,9,14 In
fact, singlet-state quantum chains are severely limited by the
short excited-state lifetimes, which make energy transfer by
diffusion mechanisms very unlikely. Adiabatic reactions that
take place in the singlet state have been established by detecting
the fluorescence15 or transient absorption of the excited pho-
toproduct rather than by observation of a quantum chain with
quantum yields of product formation that are greater than 1. It
should be noted that some adiabatic reactions, such as the
fragmentation of dioxetanes and the bond-valence isomerization
of Dewar naphthalene, may take place along both the singlet
and triple manifolds. Other reactions have been reported to occur
specifically along the singlet or triplet state.5
Recognizing that energy transfer in crystalline solids may take
place on a sub-picosecond time scale,16 we recently proposed
that crystalline solids should be ideal candidates for quantum
chain processes, increasing the efficiency of Step 3 in Scheme
1. It seems reasonable that ultrafast energy transfer will make
quantum chains possible even for the shortest-lived singlet states.
On the basis of our recent experience with the photodecarbo-
nylation of crystalline ketones17 and reports of an adiabatic
photodecarbonylation of diphenylcyclopropenone (DPCP) to
diphenylacetylene (DPA),18 we decided to investigate this
reaction as a possible test system.19 As discussed below, an
adiabatic reaction along an upper excited singlet state represents
one of the most challenging systems for quantum chains, which
can set the stage for rapid future progress.
(6) Gillmore, J. G.; Neiser, J. D.; McManus, K. A.; Roh, Y.; Dombrowski,
G. W.; Brown, T. G.; Dinnocenzo, J. P.; Farid, S.; Robello, D. R.
Macromolcules 2005, 38, 7684–7694.
(7) Swager, T. M. Acc. Chem. Res. 1998, 31, 201.
(8) Photon capture and transport by conjugated polymers has been applied
in many different contexts. (a) Swager, T. M. Acc. Chem. Res. 2008,
41, 1181. (b) Mitrovics, J.; Ulmer, H.; Weimar, U.; Go¨pel, W. Acc.
Chem. Res. 1998, 31, 307. (c) Albert, K. J.; Lewis, N. S.; Schauer,
C. L.; Sotzing, G. A.; Stitzel, S. E.; Vaid, T. P.; Walt, D. R. Chem.
ReV. 2000, 100, 2595. (d) Vetrichelvan, M.; Nagarajan, R.; Valiyaveet-
til, S. Macromolecules 2006, 39, 8303. (e) Liu, Y.; Mills, R. C.;
Boncella, J. M.; Schanze, K. S. Langmuir 2001, 17, 7452. (f) Hong,
J. W.; Gaylord, B. S.; Bazan, G. C. J. Am. Chem. Soc. 2002, 124,
11868. (g) Jones, R. M.; Bergstedt, T. S.; McBranch, D. W.; Whitten,
D. G. J. Am. Chem. Soc. 2001, 123, 6726. (h) Matsuda, K.; Nakamura,
N.; Takahashi, K.; Inoue, K.; Koga, N.; Iwamura, H. J. Am. Chem.
Soc. 1995, 117, 5550.
Transient absorption studies in solution, starting in the early
1990s by Hirata et al.,18c suggested an adiabatic reaction from
the second excited state, S2-DPCP, which has been recently
estimated to occur with a time constant of ca. 200 fs, as
illustrated in Scheme 2.18a While there has been some debate
(15) (a) Carr, R. V.; Kim, B.; McVey, J. K.; Yang, N. C.; Gerhartz, W.;
Michl, J. Chem. Phys. Lett. 1976, 39, 57. (b) Yang, N. C.; Carr, R. V.;
Li, E.; McVey Rice, J. Chem. Phys. Lett. 1976, 39, 57.
(9) (a) Merkel, P. B.; Roh, Y.; Dinnocenzo, J. P.; Robellow, D. R.; Farid,
S. J. Phys. Chem. A 2007, 111, 1188. (b) Ferrar, L.; Mis, M.;
Dinnocenzo, J. P.; Fardid, S.; Merkel, P. B.; Robello, D. R. J. Org.
Chem. 2008, 73, 5683.
(16) (a) Powel, R. C; Soos, A. G. J. Lumin. 1975, 11, 1. (b) Swenberg,
C. E.; Gaecintov, N. E. In Organic Molecular Photophysics; Birks,
J. B. Ed.; John Wiley and Sons: London, 1973. (c) Yoon, Z. S.; Yoon,
M.-C.; Kim, D. J. Photochem. Photobiol. C 2005, 6, 249. (d) Ahrens,
M. J.; Sinks, L. E.; Rybtchinski, B.; Liu, W.; Jones, B. A.; Giaimo,
J. M.; Gusev, A. V.; Goshe, A. J.; Tiede, D. M.; Wasielewski, M. R.
J. Am. Chem. Soc. 2004, 126, 8284.
(10) (a) Saltiel, J.; Dmitrenko, O.; Pillai, Z. S.; Klima, R.; Wang, S.;
Wharton, T.; Huang, Z.-N.; van de Burgt, L. J.; Arranz, J. Photochem.
Photobiol. Sci. 2008, 7, 566. (b) Saltiel, J.; Dmitrenko, O.; Reischl,
W.; Bach, R. D. J. Phys. Chem. A 2001, 105, 3934. (c) Saltiel, J.;
Wang, S.; Ko, D.-H.; Groming, D. A. J. Phys. Chem. A 1998, 102,
5383. (d) Saltiel, J.; Townsend, D. E.; Skyes, A. J. Am. Chem. Soc.
1973, 95, 5968.
(17) (a) Campos, L. M.; Garcia-Garibay, M. A. In ReViews of ReactiVe
Intermediate Chemistry; Platz, M. S., Jones, M., Moss, R., Eds.; Wiley-
Interscience: Hoboken, NJ, 2007. (b) Mortko, C. J.; Garcia-Garibay,
M. A. Top. Stereochem. 2006, 25, 205–253. (c) Garcia-Garibay, M. A.;
Campos, L. In The Handbook for Organic Photochemistry and
Photobiology; Horspool, W., Ed.; CRC Press: Boca Raton, FL, 2003.
(18) (a) Takeuchi, S.; Tahara, T. J. Chem. Phys. 2004, 120, 4768. (b)
Terazima, M.; Hara, T.; Hirota, N. Chem. Phys. Lett. 1995, 246, 577.
(c) Hirata, Y.; Okada, T.; Mataga, N.; Nomoto, T. J. Phys. Chem.
1992, 193, 287.
(11) (a) Karatus, T.; Tsuchiya, M.; Arai, T.; Sakuragi, H.; S.; Tokumaru,
K. Bull. Chem. Soc. Jpn. 1994, 67, 3030. (b) Mo¨llerstedt, H.;
Wennerstro¨m, O. J. Photochem. Photobiol. A 2001, 139, 37.
(12) Turro, N. J.; Waddell, W. H. Tetrahedron Lett. 1975, 25, 2069.
(13) (a) Zharinova, E. V.; Voloshin, A. I.; Kazakov, V. P. J. Mol. Liq.
2001, 91, 237–243. (b) Kazakov, V. P.; Voloshin, A. I.; Shavaleev,
N. M. J. Photochem. Photobiol. A 1998, 119, 177.
(14) (a) Turro, N. J.; Ramamurthy, V.; Katz, T. J. NouV. J. Chim. 1977, 1,
363. (b) Lechtken, P.; Yekta, A.; Shore, N. E.; Steinmetzer, H. C.;
Waddell, W. H.; Turro, N. J. Z. Phys. Chem. 1976, 101, 79.
(19) Kuzmanich, G.; Natarajan, A.; Chin, K.; Veerman, M.; Mortko, C.;
Garcia-Garibay, M. A. J. Am. Chem. Soc. 2008, 130, 1140.
9
J. AM. CHEM. SOC. VOL. 131, NO. 32, 2009 11607

A R T I C L E S
Kuzmanich et al.
over the detailed spectral assignments,20 the time constant for
internal conversion from S2-DPA to S1-DPA was estimated to
be 8 ps. This is followed by intersystem crossing to T1-DPA in
ca. 210 ps and by decay to the ground-state product in ca. 20
µs. The same studies concluded that excitation to S1-DPCP
proceeds by a conventional non-adiabatic reaction to the ground
state of DPA.18 In addition to forming an excited-state photo-
product, the feasibility of a quantum chain process depends on
the energetics of energy transfer. In the cases of DPCP and DPA,
one can estimate from the spectrum that the second excited-
state energies (S2-DPCP and S2-DPA) are 88 and 95 kcal/mol,
respectively, which makes energy transfer from the excited-
state product to the ground-state reactant exothermic, and
therefore viable. Notably, the greater S2-S0 energy gap for DPA
is determined by the lower energy content of its ground state,
which is ca. 10 kcal/mol more stable than that of the strained
ketone (Scheme 2). It is interesting to note that these values
make the adiabatic reaction exothermic by ca. 3.0 kcal/mol.
Evaluation of the factors required for a quantum chain in the
case of DPCP shows that the main limitation should come from
the short lifetime of S2-DPA, which gives only ca. 8 ps to
transfer its excitation energy to a nearby reactant molecule. If
reacted as a molecular solid, where singlet excitons have
hopping times of ca. 1-2 ps, one may expect that every photon
initiating the ring-opening reaction of a DPCP molecule should
be able to “tag” as many as 4-8 reactant molecules before one
of them undergoes internal conversion to the first excited state
(S1), which no longer satisfies the energetic requirements for
energy transfer. Alternatively, the mechanism could involve an
excited state delocalized throughout the crystal, which could
cause several molecules to react.8h To demonstrate this, one
must measure the quantum yield of the solid-state reaction and
unequivocally show that is greater than one (Φrxn > 1.0). In
agreement with these expectations, we recently showed that
macroscopic specimens of DPCP undergo a remarkably efficient
reaction that transforms single crystals into fine powders of DPA
within minutes.19 More importantly, quantum yield determina-
tions with nanocrystalline suspensions using dicumyl ketone
(DCK, Φrxn ) 0.2) as an internal actinometer revealed an
average quantum yield of Φrxn ) 3.3 when DPCP is excited to
S2 in the crystalline state.19
deduced from the observed photostability of the 9-anthryl (1g)
and p-iodophenyl (1h) derivatives, significant quantum chain
processes were determined for all the other structures.
Results and Discussion
Synthesis of Diarylcyclopropenone Derivatives. Derivatives
1b-1h were prepared by Friedel-Crafts alkylation of the
corresponding aromatic moiety with tetrachlorocyclopropenone
in DCM at -78 °C.22 The resulting products were purified by
column chromatography (silica gel, 4:1 hexane:acetone) fol-
lowed by one or more recrystallizations in hexane with ca. 5%
acetone or dichloromethane. As indicated in Table 1, isolated
yields of the pure product varied from 40% to 65%. Samples
of the bisphenol 1i were prepared from the p-methoxy analogue
1b, as shown in Scheme 3, and were purified in a similar
manner. The crystalline salt 1j was obtained by dissolving 1i
in triethylamine, consistent with a remarkably high acidity for
the latter, which was later confirmed by spectrophotometric
methods and by titration of 1i. All cyclopropenone derivatives
were characterized by having the 1H NMR signal of the
hydrogens meta to the electron-withdrawing cyclopropenone at
δ > 8 ppm and the 13C NMR signal of the carbonyl carbon
distinct at ca. 165 ppm. While these compounds were all
characterized by conventional spectroscopic methods that agree
with literature reports,23 we reassigned the structure of 1f as
illustrated in Table 1. A lower symmetry structure with an
o-methoxy linkage, rather than the suggested 4-(p-anisyl)phenyl
derivative, is determined by the 1H and 13C NMR spectra, which
1
include the presence of an isolated H resonance at 8.21 ppm
that integrates for only one hydrogen, and by the 15 rather than
10 signals in the 13C NMR (Supporting Information). All of
the cyclopropenone derivatives are crystalline powders with high
melting points, which help ensure that the solid-state reaction
does not take place in the melt. The parent DPCP has the lowest
melting point, 87 °C. All known derivatives had melting points
identical to those reported in the literature, typically between
130 and 160 °C, as shown in Table 1. The 9-dianthryl derivative
1e and the crystalline salt 1j decomposed before melting at T
g 350 °C.
While derivatives 1a, 1b, 1h, and 1j form white microcrys-
talline powders, those of the more conjugated 1c, 1d, and 1f
are pale yellow, and those of 1g are vibrant red. All crystalline
powders, except for 1g, were soluble in methanol, chloroform,
and acetone. The dianthryl compound 1g was highly insoluble
in all solvents tested, and we decided to acquire its 13C NMR
spectra by solid-state CPMAS NMR to make sure that it
conforms to the other structures (Supporting Information). The
CPMAS clearly shows the presence of the distinct carbonyl
carbon at 159 ppm along with the other cyclopropene carbons
at ca. 146 ppm. Additionally, the only other carbon signals in
the spectrum are in the aromatic region, 116-127 ppm, similar
to those of the other compounds in solution. Judging by its
Recognizing the potential of a photoinduced amplification
system that unveils highly emissive and charge-transporting
diarylacetylenes,21 we decided to explore the scope of the
quantum chain reaction. In this paper, we describe the synthesis,
nanocrystal formation, nanocrystal photochemistry, and solution
and solid-state quantum yield determinations of nine symmetric
diarylcyclopropenones, 1b-1j. The formation and characteriza-
tion of nanocrystals for all DPCP derivatives was deemed
essential to determine reliable solid-state quantum yields.
Included in Table 1 are structures with aryl substituents that
include para-substituted benzenes (1b, 1h-1j), bibenzyl (1e),
biphenyl (1f), two naphthalenes (1c and 1d), and a 9-anthracene
derivative (1g). While efficient quenching mechanisms were
1
UV-vis and H NMR spectra, solutions of bisphenol 1i in
methanol, water, or acetone exist in equilibrium between the
neutral and the mono-dissociated forms. The addition of
triethylamine results in formation of a quaternary ammonium
salt, characterized by a downfield shift of 0.13 and an upfield
(20) Poloukhtine, A.; Popik, V. V. J. Phys. Chem. A 2006, 110, 1749.
(21) (a) Bunz, W. H. F. Chem. ReV. 2000, 100, 1605. (b) Nesterov, E. E.;
Zhu, Z.; Swager, T. M. J. Am. Chen. Soc. 2005, 127, 10083. (c) Meier,
H. Angew. Chem., Int. Ed. 2005, 44, 2482. (d) Levitus, M.; Schmieder,
K.; Ricks, H.; Shimizu, K. D.; Bunz, W. H. F.; Garcia-Garibay, M. A.
J. Am. Chen. Soc. 2001, 123, 4259. (e) Stapleton, J. J.; Harder, P.;
Daniel, T. A.; Reinard, M. D.; Yao, Y.; Price, D. W.; Tour, J. M.;
Allara, D. L. Langmuir 2003, 19, 8245. (f) Samori, S.; Tojo, S.;
Fujitsuka, M.; Spitler, E. L.; Haley, M. M.; Majima, T. J. Org. Chem.
2007, 72, 2785.
1
shift of 0.11 ppm in the H NMR for the aromatic signals at
(22) West, R.; Zecher, D. C.; Goyert, W. J. Am. Chem. Soc. 1970, 92,
149.
(23) (a) Urdabayev, N. K.; Poloukhtine, A; Popik, V. V. Chem. Commun.
2006, 4, 454. (b) Poloukhtine, A.; Popik, V. V. J. Org. Chem. 2003,
68, 7833.
9
11608 J. AM. CHEM. SOC. VOL. 131, NO. 32, 2009

Photodecarbonylation of Crystalline Diarylcyclopropenones
A R T I C L E S
Table 1. Synthetic Isolated Yields, Melting Points, UV-Vis Data, Photochemical Quantum Yields in Solution and in Crystalline Suspension,
and Quantum Yield Amplification Values for Several DPCP Derivatives
a Commercially available. b Reference 18. c Reference 11. d Reference 23b. e Reference 23a. f Reference 12. g See Scheme 3 for synthesis and yields.
Scheme 3
compares favorably with that of 2,4-dinitrophenol, which has
pKa ) 4.09,24 but is not as acidic as picric acid (2,4,6-
trinitrophenol), with pKa ) 0.3.25 A ca. 30 nm red-shift was
observed on going from the spectrum of the phenol to that of
the double salt both in solution and in the crystalline salt.
Preparation and Characterization of Nanocrystalline
Suspensions. We have shown that suspensions of nanocrystalline
solids offer significant advantages over large single crystals and
dry powders for photochemical studies.26 As particle sizes
approach the wavelength of individual photons, samples can
be excited in a nearly homogeneous manner, avoiding problems
6.91 and 7.80 ppm, respectively. The dianionic nature of 1j was
also confirmed by showing that it has a UV-vis spectrum that
is identical to that of a sample prepared by deprotonation with
2 equiv of LHMDS. The UV-vis spectrum of the dianion is
significantly red-shifted relative to that of 1i. The onset of
absorption is at 410 nm for 1j and 360 nm for 1i (Supporting
Information). Determinations of the pKa of 1i by methanolic
titrations with 0.1 M KOH revealed a single value of pKa )
5.48 for the apparent double deprotonation (see Supporting
Information for details). With this value, the acidity of 1i
of penetration depth and possible filtering effects by strongly
absorbing photoproducts.27 Nanocrystalline suspensions are also
ideal for large-scale solid-state photochemical reactions using
(24) Barcza, L.; Buva´ri-Barcza J. Chem. Educ. 2003, 80, 822.
(25) Jin, C.-M.; Ye, C.; Piekarshi, C. Eur J. Inorg. Chem. 2005, 18, 3760.
(26) Veerman, M.; Resendiz, M. J. E.; Garcia-Garibay, M. A. Org. Lett.
2006, 8, 2615–2617.
(27) Resendiz, M. J. E.; Taing, J.; Garcia-Garibay, M. A. Org. Lett. 2007,
9, 4351–4354.
9
J. AM. CHEM. SOC. VOL. 131, NO. 32, 2009 11609

A R T I C L E S
Kuzmanich et al.
nanocrystals obtained by the reprecipitation protocol were
analyzed by atomic force microscopy (AFM) and scanning
electron microscopy (SEM) to verify the particle size and the
crystallinity of the nanospecimens. Both isolated crystals and
aggregates were observed from suspensions deposited on an
alkane-terminated monolayer on a silicon wafer.34 Figure 1a
shows isolated DPCP nanocrystals with sizes ranging between
ca. 100 and 200 nm, in general agreement with the DLS results
from the fluid suspension. The SEM image also shows that the
particles are clearly faceted and crystalline.
Photochemical Reactivity in Polycrystalline Powders. We
have suggested that dissociative photochemical reactions in
crystalline solids can be engineered in a reliable manner by
selecting high-melting substrates with exothermic bond-cleavage
reactions17,35 that have no easy radical recombination and no
quenching processes. It seemed reasonable to us that the
photodecarbonylation of simple DPCP derivatives should be
quite general. Photochemical experiments with cyclopropenones
1a-1j were carried out at 298 K with a mercury arc Hanovia
lamp using a λ > 290 nm (Pyrex) filter or with the 312 nm
bulbs of a Rayonet reactor. Samples irradiated in solution by
1H NMR and in bulk polycrystalline powders revealed that most
compounds reacted to completion to give the corresponding
diaryl acetylene powders within 60 min. The only exceptions
were compounds 1g and 1h. Samples of bis(p-iodophenyl)cy-
clopropenone 1h were photochemically stable in methanol,
chloroform, nanocrystalline suspension, and dry powders. We
interpret this result as indicative of a very efficient singlet-state
deactivation by the iodine heavy-atom effect.36 Similarly,
samples of bis(9-anthryl)cyclopropenone 1g irradiated with λ
g 290 nm were stable in suspensions and in dry solids, but
they are modestly reactive in solution. Reaction quantum yields
of 0.02 and 0.01 have been previously reported in dilute toluene
and methylene chloride, respectively.37,38
Fluorescence measurements with 1g in CH2Cl2 solutions and
nanocrystalline suspensions yielded similar spectra. Identical
absorption (Supporting Information) and fluorescence excitation
(Figure 2) spectra are indicative of a sample with no other
fluorescent impurities. The two spectra are characterized by a
structureless low-energy transition with λmax) 465 nm, having
two shoulders at lower (445 nm) and higher (515) wavelengths
in the case of the solid suspension. As expected, photochemical
excitation in this band results in no decarbonylation reaction in
solution. Previous studies with several dianthryl ethenes by
Becker et al., including 1g, suggested that the lowest-energy
transition in these chromophores is due to an intramolecular
excimer, which is dominant in solution and in the solid state.37,38
Thus, it is likely that the solution reactivity observed upon
excitation of 1g with short wavelengths is due to a very fast
adiabatic reaction from an upper excited state largely localized
in the DPCP chromophore. The lack of reactivity in the solid
state under similar conditions suggests a very fast internal
conversion within the bis(anthracene)-centered lowest excited
state. We considered that the lack of reactivity in crystals of 1g
Figure 1. (a) SEM image of DPCP isolated nanocrystals showing crystals
of ca. 50-200 nm on an alkane monolayer on a Si surface. (b) 4.6×
magnification of the right portion of panel a.
flow reactors and for spectrophotometric analysis by transmis-
sion methods.28
The simplest and most efficient method for the preparation
of nanocrystalline suspensions is the reprecipitation method
initially pioneered by Kasai et al.,29 which produces nanocrystals
between ca. 50 and 600 nm in size for a variety of substrates.
In order to determine the solid-state reactivity and quantum
yields of reaction for the DPCP derivatives in Table 1, we
typically used 1.6-2.6 mg (0.064 mmol) of each compound
dissolved in a minimal amount of acetone or methanol. These
solutions were added rapidly via a micropipet into a test tube
containing 9 mL of vortexing H2O with 1/20th of the critical
micelle concentration (cmc) of either cetyltrimethylammonium
bromide (CTAB) or sodium dodecyl sulfate (SDS).30 The
resulting milky suspensions were sonicated for 40 min at 60
°C to ensure that no acetone remained in the sample. Given
that analogous suspensions were obtained from all diarylcyclo-
propenone derivatives, we decided to carry out a thorough
characterization with the parent compound, assuming that it
would be a reasonable model for the others in the set. Using
X-ray powder diffraction of centrifuged nanoparticles, we
confirmed that the nanocrystalline suspensions adopt one of the
polymorphs observed with bulk DPCP (see Supporting Informa-
tion). Notably, while it is well known that DPCP forms
hydrated31 and anhydrous32 forms by slow evaporation from
water-saturated and dry benzene, respectively, it seemed surpris-
ing that the anhydrous form would be obtained upon precipita-
tion in water. As in other cases reported in the literature, we
hypothesize that this apparent anomaly simply indicates both
that nucleation and growth under these conditions are kinetically
controlled and that the incorporation of water requires thermo-
dynamic equilibrium.33
Dynamic light scattering (DLS) studies performed on a DPCP
suspension with a 4 × 10-5 M loading revealed average particle
sizes between 100 and 250 nm. As illustrated in Figure 1, DPCP
(28) Chin, K. K.; Natarajan, A.; Gard, M.; Campos, L. M.; Johansen, E.;
Shepherd, H.; Garcia-Garibay, M. A. Chem. Commun. 2007, 4266–
4268.
(29) Kasai, H.; Nalwa, H. S.; Oikawa, H.; Okada, S.; Matsuda, H.; Minami,
N.; Kakuta, A.; Ono, K.; Mukoh, A.; Nakanishi, H. Jpn. J. Appl. Phys.
1992, 31, 1132.
(34) Sieval, A. B.; Vlemming, V.; Zuilhof, H.; Sudholter, E. J. R. Langmuir
1999, 15, 8288.
(35) (a) Garcia-Garibay, M. A. Acc. Chem. Res. 2003, 36, 491–498. (b)
Garcia-Garibay, M. A.; Shin, S.; Sanrame, C. Tetrahedron 2000, 56,
6729–6737.
(30) Medium-sensitive spectral measurements can be used to distinguish
between solid suspension and micelles before irradiation.19
(31) Tsukada, H.; Shimanouchi, H.; Sasada, Y. Tetrahedron Lett. 1973,
14, 2455.
(36) Ihmels, H.; Patrick, B. O.; Scheffer, J. R.; Trotter, J. Tetrahedron 1999,
2171.
(32) Tsukada, H.; Shimanouchi, H.; Sasada, Y. Chem. Lett. 1974, 3, 639.
(33) Kumar, V.; Wang, L.; Riebe, M.; Tung, H.-H. ; Pud’homme, R. K.
Mol. Pharm., doi: 10.1021/np.90002t.
(37) Becker, H.-D.; Sandros, K.; Skelton, B. W.; White, A. H. J. Phys.
Chem. 1981, 85, 2930.
(38) Becker, H.-D.; Andersson, K. J. Org. Chem. 1987, 52, 5205.
9
11610 J. AM. CHEM. SOC. VOL. 131, NO. 32, 2009

Photodecarbonylation of Crystalline Diarylcyclopropenones
A R T I C L E S
Figure 3. AFM images of the edge of a DPCP crystal on a alkene-
functionalized Si surface (a) before exposure to light and (b) after 10 min
of irradiation with a Xe lamp with λ > 300 nm.
Figure 2. Excitation and emission spectra of dianthryl cyclopropenone 1g
in CH2Cl2 (C, black) and in an aqueous suspension (B, red). Excitation and
emission spectra of dianthryl acetylene 2g in an aqueous suspension (A,
blue).
size in a given sample decreased from ca. 1200 nm before
irradiation to ca. 200 nm after 40% conversion. However,
further irradiation and continued conversion did not result
in a further decrease in the average particle size. The large
crystal size of the initial suspension in this case was selected
to show a greater size change and was obtained by preparing
a sample with a higher DPCP loading. In conclusion, while
it is clear that the reaction results in a remarkable process
that leads to recrystallization of the product in specimens of
very small size, it is not known at this time whether the final
dimension is determined by initial crystal size or by differ-
ences arising from the sample being suspended in water
versus at a dry air-crystal interface.
could be due to trace amounts of dianthryl acetylene 2g acting
as a very efficient quencher. However, the absorption spectrum
of 2g has a poor spectral overlap with emission of 1g.
Furthermore, samples of 1g irradiated (at 266, 350, or 480 nm)
for 2 h did not show changes in the emission envelope or
intensity that could be associated with the involvement of 2g
by direct excitation or as an energy acceptor. It should be pointed
out that a modest increase in intensity could have been expected
on the basis of the fact that the fluorescence quantum yield of
2g (ΦF ) 0.04)39 is about 4 times greater than that determined
here for 1g (ΦF ) 0.01). We conclude that the solid-state
reaction of 1g is not prevented by product quenching.
Quantum Yield Determinations in Solution. The UV spec-
trum of DPCP is characterized by a very weak S0-S1
transition between ca. 330 and 390 nm, with λmax ≈ 370 nm
and log ꢀ ≈ 3.3 M-1cm-1, and a very strong S0-S2 transition
Nanometer-Scale Mechanical Response of DPCP Crystals
to the Solid-State Reaction. We previously reported in situ
microscopic observation of large specimens of DPCP crystals
exploding and crumbling into a fine powder of DPA after a
few minutes of UV irradiation.19 We suggested that such a
remarkable mechanical response may be the combined result
of the very fast reaction, which does not allow for local
relaxation, and the large changes in volume that take place at
the molecular level when bent molecules of DPCP transform
into linear molecules of DPA and CO.4 Given that the
mechanical response of several photoreactive crystals has been
shown to depend strongly on their initial size,40,41 we became
interested in documenting such effects at the nanometer scale
when dry DPCP crystals are irradiated and in determining
whether a similar effect would take place with nanocrystalline
specimens in water suspensions.
To explore the nature of the reaction at the nanometer
scale, we drop-cast a sample of DPCP on an alkane-
terminated Si wafer and observed the edges of the relatively
large crystal specimens by AFM in a tapping mode (Figure
3a). The AFM image obtained after the same sample was
irradiated for 10 min with a Xe lamp with λ > 300 nm
revealed a remarkable transformation (Figure 3b). The
resulting DPA is seen as a collection of individual nano-
crystals with an average diameter of ca. 40 nm. Comple-
mentary experiments to investigate changes in crystal size
for nanocrystalline suspensions as a function of conversion
were analyzed by DLS with a sample of DPCP analyzed
before and after exposure to UV light for several time
intervals. The DLS results showed that the average crystal
between 240 and 325 nm, with λmax ≈ 300 nm and log ꢀ )
23
4.28 (284 nm), 4.33 (295 nm), and 4.16 (307 nm) M-1cm-1
.
Although the derivatives possess different shapes and λmax
values, it appears that differences in the intensity of the
corresponding transitions also occur in the other DPCP
derivatives. As indicated in Table 1, the solution quantum
yields for derivatives 1c, 1d, 1f, and 1g after excitation at
350 nm, corresponding mainly to S0-S2 transition, have been
reported in the literature. The quantum yields of 1b, 1e, 1i,
and 1j were determined in this study by standard potassium
ferrioxalate actinometry42 with excitation to the S2 state (or
higher) by irradiation at 312 nm. As shown in the table, the
quantum efficiency for the parent DPCP is quantitative, with
Φsol ) 1.0. Other compounds in the table can be classified
in three groups. Compounds with 1-naphthyl (1c), 2-methoxy-
1-naphthyl (1c), and 5-phenyl-2-methoxyphenyl (1f) groups
have solution quantum yields that are Φsol g 0.5. The second
group comprises compounds with p-MeO (1b), p-(2-phenyl)-
ethyl (1e), p-hydroxy (1i), and p-phenoxy (1j) groups, which
have Φsol ≈ 0.2. The third group includes the 9-anthryl-
substituted analogue 1g, which has unreactive low-energy
states, and the p-iodo derivative 1i, which provides a
spin-orbit coupling mechanism that quenches the reaction.
Compound 1g has a very low reaction quantum yield of Φsol
(41) (a) Al-Kaysi, R. O.; Mu¨ller, A. M.; Bardeen, C. J. J. Am. Chem. Soc.
2006, 128, 15938. (b) Al-Kaysi, R. O.; Bardeen, C. J. AdV. Mater.
2007, 19, 1276.
(39) Makino, T.; Toyota, S. Bull. Chem. Soc. Jpn. 2005, 78, 917.
(40) Garcia-Garibay, M. A. Angew Chem., Int. Ed. 2007, 46, 8945.
(42) Murov, S. L; Carmichael, I.; Hug, G. L. Handbook of Photochemistry,
2nd ed.; Marcel Dekker, Inc.: New York, 1993.
9
J. AM. CHEM. SOC. VOL. 131, NO. 32, 2009 11611

A R T I C L E S
Kuzmanich et al.
) 0.02, and 1i is completely photostable. The reduced
chemical efficiency of the substituted DPCPs in solution is
indicative of slower decarbonylation reactions, faster internal
conversion, and/or decay to the ground state. While informa-
tion to assign the contributions to reactions from the S2 and
S1 states is not available, suppression of the adiabatic reaction
would completely eliminate a solid-state quantum chain. In
contrast, even a fraction of molecules reacting along S2 may
be able to initiate a chain reaction that can be propagated by
energy transfer from the excited product to the reactant. This
was tested by quantum yield determinations in the solid state.
Quantum Yield Determinations in Nanocrystalline Suspen-
sions. Quantum yield determinations in the solid state were
carried out by chemical actinometry with nanocrystalline
suspensions. Particle loadings were adjusted so that all the
photons from the UV source are absorbed and can be
accounted for. Having shown that dicumyl ketone (DCK) is
a good nanocrystalline actinometer with a quantum yield of
decarbonylation of ΦDCK ) 0.20,43 we set out to determine
the quantum yield of decarbonylation for all the reactive
DPCPs in Table 1 using mixed suspensions. Taking advantage
of this method, we previously reported that the quantum yield
of the parent DPCP in a nanocrystalline suspension is ΦCP
) 3.3. The extent of reaction in each case was determined
Quantum Yield Amplification. It is well known that the
crystalline environment can change excited-state deactivation
pathways, such as internal conversion, intersystem crossing,
and fluorescence, potentially affecting the reaction quantum
yields.44 For that reason, changes in the quantum yields of
decarbonylation in nanocrystalline suspensions (ΦSus) should
be analyzed in terms of chain propagation by energy transfer
and also in terms of possible changes in reaction efficiencies
within the crystal lattice, the latter determined by an intrinsic
quantum yield ΦCL
.
We define the quantum yield amplification (QYA) as the ratio
of the quantum yield in nanocrystalline suspension (Φsusp
relative to the quantum yield in solution (Φsol),
)
QYA ) Φsusp/Φsol
(2)
Starting with equal moles of reactant as dissolved and solid
samples, QYA represents the equivalents of additional
photons needed in solution to produce the same moles of
product as in the solid state. The QYA values for all the
compounds in this study are included in Table 1 and are as
high as 11.4 in the case of 1b.
In general, the propagation length of a quantum chain reaction
is determined by the fraction of molecules that proceed to the
product in the first adiabatic reaction (e.g., ΦCL) and which are
involved in the subsequent energy-transfer step (Scheme 1).
Notably, a quantum chain reaction does not imply a quantum
yield greater than 1, even with an infinite number of energy-
transfer steps. The quantum yield of a general chain reaction
(Φchain) with n energy-transfer steps can be described by a
geometric series of the form
1
by H NMR for the DPCP analogues and by gas chroma-
tography for DCK. Every sample was analyzed in triplicate
or quadruplicate, and results were obtained with standard
deviations less than 10%. Using co-suspended DCK nano-
crystals as an internal actinometer, the quantum yield of
decarbonylation is given by the following equation,
Φsusp ) (ACP/ADCK)(NDA/NDC)ΦDCK
(1)
Φchain ) (Φ)1 + (Φ)2 + (Φ)3 + .....(Φ)n
(3)
where ACP is the absorbance of the cyclopropenone, ADCK
the absorbance of dicumyl ketone, NDA the number of moles
of diarylacetylene formed from the corresponding cyclopro-
penone, and NDC the number of moles of dicumene formed
from DCK with a quantum yield ΦDCK ) 0.20. After more
than 80 experiments with variations that included suspension
loadings and with or without surfactants at pre-micellar
concentrations, we determined that Φsusp ) 3.30 ( 0.35 at
312 nm. Quantum yields of reactions carried out in solution
or upon excitation to S1 between λ ) 350 and 390 nm gave
values of 1.0. The solid-state quantum yield strongly supports
the mechanism in Scheme 1, with an average of 3.3 molecules
per photon in the chain. Analogous quantum yield studies
with over 25 independent measurements each were carried
out with the other diaryl cyclopropenone derivatives in Table
1. Notably, independent quantum yield measurements using
DCK or the parent DPCP as the chemical actinometer gave
identical results, making the data set internally consistent.
The quantum yields determined in this manner range from
0.87 to 2.74, which are higher than the values measured in
solution, and in all cases but for 1j, the values are greater
than 1.0, as required for the sought-after quantum chain.
Notably, the reaction is tolerant to several changes in the
chromophore, including oxygen, naphthyl substituents, and
extended linear conjugation. A quantum yield less than 1.0
in the case of diphenolate salt 1j was confirmed with
inorganic sodium and potassium salts suspended in hexane,
removingsuspicionsofquenchingbysomeneutraltriethylamine.
with the value of n determined by the relative rates of energy
transfer and excited-state decay,
n ) kET/kdec
(4)
For a hypothetical solid-state reaction with an infinite number
of energy-transfer steps (n ) ∞) and an intrinsic quantum yield
ΦCL, the series converges to the limit
Φchain ) ΦCL/(1 - ΦCL
)
(5)
which results in observed efficiencies in the ranges of
Φchain < 1.0 for ΦCL < 0.5
Φchain ) 1.0 for ΦCL ) 0.5
Φchain > 1.0 for 0.5 < ΦCL < 1.0
Intrinsic quantum yields of 0.2, 0.4, 0.7, 0.9, and 1.0 would
give maximum Φchain values of 0.25, 0.66, 2.33, 9.0, and ∞,
respectively. The suspension quantum yields in Table 1 cor-
respond to the quantum yield of the chain, i.e., Φsusp ) Φchain
.
In the case of 1b, it can be seen that Φsol ) 0.24 is not large
enough to produce Φsusp ) 2.74; therefore, ΦCL * Φsol. For
Φsusp > 1, ΦCL must be greater than 0.5. At the same time, the
quantum yield of suspensions of 1a, Φsusp ) 3.3, may imply
that ΦCL < Φsol or, more likely, that the number of energy-
(43) Veerman, M.; Resendiz, M. J. E.; Garcia-Garibay, M. A. Org. Lett.
2006, 8, 2615.
(44) Kol’chenko, M. A.; Kozankiewicz, B.; Nicolet, A.; Orrit, M. Opt.
Spectrosc. 2005, 98, 681.
9
11612 J. AM. CHEM. SOC. VOL. 131, NO. 32, 2009

Photodecarbonylation of Crystalline Diarylcyclopropenones
A R T I C L E S
transfer steps (n) is limited. The suggested increase in the
efficiency of the solid-state decarbonylation (ΦCL) of 1b-1f
and 1h-1i may be due to an acceleration of the adiabatic
reaction or to a deceleration of the rate of internal conversion
and decay, the latter being more probable.
expected to give rise to millions of product molecules per
photon of light absorbed by the crystalline reactant.
Experimental Section
General Methods. Solid-state 13C NMR spectra were acquired
on a Bruker DRX300 instrument. IR spectra were obtained with
a Perkin-Elmer Spectrum instrument equipped with a universal
attenuated total reflectance (ATR) accessory. Gas chromatog-
raphy data were acquired on a Hewlett-Packard 5890 series II
gas chromatograph equipped with an HP3396 series II integrator
and an HP-5 capillary column of dimensions 25 m × 0.2 mm
with a film thickness of 0.11 mm. The excitation and emission
spectra of all samples were obtained on a Photon Technology
International Quanta Master spectrofluorimeter. Dynamic light
scattering measurements were taken on a Beckman-Coulter N4
Plus submicron particle size analyzer. Silica used for purification
was Silica-P flash silica gel (40-62 Å), purchased from SiliCycle
Inc. UV-vis spectra were taken on a Beckman DU-650
spectrometer.
Nanocrystalline Suspensions. Nanocrystalline suspensions for
microscopy and DLS studies were prepared by injecting 50 µL
of a DPCP (8 mM) solution in acetone into 10 mL of vortexing
water (Millipore). The resulting suspension (4 × 10-5 M) was
sonicated three times at room temperature for 4 min, allowing
for 2 min rest between runs. These samples were not made using
surfactant. During drying, the drop-cast suspension will become
more concentrated. If CTAB is present in the suspension, its
concentration will eventually rise above the critical micelle
concentration. The formation of micelles or the observation of
CTAB is undesirable in this situation; therefore, surfactant was
not used during the microscopy studies. AFM studies were
performed in tapping mode with a multi-mode Veeco Nanoscope
IIIa instrument using monolithic silicon probes with a tip radius
reported to be <10 nm (VistaProbes T300R, Nanoscience
Instruments, Inc., Phoenix, AZ). To prepare the AFM sample, a
dilute DPCP suspension (4 × 10-5 M) was drop-cast onto a clean
silicon wafer and allowed to dry in a desiccator. Scanning
electron microscopy studies were performed with a JEOL JSM-
6700F field-emission scanning electron microscope. To prepare
the SEM sample, a dilute DPCP suspension (4 × 10-5 M) was
drop-cast onto carbon tape and allowed to dry in a desiccator.
If one views energy migration in solids as discrete events,
the efficiency of energy transfer from the excited-state photo-
product to unreacted starting material is determined by the time
constant of energy transfer, τET, relative to the time constant of
internal conversion, τIC, to the S1 of DPA. Energy transfer from
DPA to DPCP within a crystal should take place within time
scales typical of singlet exciton hopping, τET ≈ 1-2 ps. The
energetics for energy transfer are favorable (95 and 88 kcal/
mol for the respective S2 states), and translation symmetry of
nearest neighbors guarantees van der Waals contacts and ideally
parallel orientations. Given that the S2 lifetimes of diarylacety-
lenes 2a, 2b, and 2i in hexane solution reported by Hirata and
Okada fall within τIC ≈ 7-8 ps,45 one may expect that the
average number of energy steps will be given by
n ) kET/kIC ≈ 4-8
(6)
Satisfyingly, the observed Φsusp values are consistent with
the estimated number of discrete energy transfers.
Conclusions
The photoinduced decarbonylation of diarylcycloprope-
nones to diarylacetylenes occurs with remarkable efficiency
in the crystalline solid state. With a set of 10 aryl-substituted
derivatives, we have shown that the adiabatic reaction
previously postulated for the second excited state (S2) of the
parent diphenyl compound is capable of initiating a quantum
chain process where the excited-state S2 photoproduct
transfers its energy within a few picoseconds to a ground-
state reactant to propagate the chain. While the quantum
yields of photodecarbonylation in solution vary from ca. 0.2
to 1.0, the values obtained with nanocrystalline suspensions
vary from 0.87 to 3.3. The increase in quantum efficiencies
on going from solution to crystalline solids corresponds to
quantum chain amplification factors varying between 3.2 and
11.4. A remarkable mechanical response of the previously
documented solid-to-solid reaction of diphenyl cycloprope-
none 1a, where large single-crystalline specimens turn into
fine powders, was shown to scale down to the nanometer
scale by formation of isolated crystalline specimens of the
photoproduct of ca. 35 nm in size. Thus, the remarkable
potential of crystalline-state adiabatic photoreactions for
chemical amplification has been unveiled with a very
challenging system where the quantum chain is limited by
the very short lifetime of an upper singlet excited state. Future
femtosecond spectroscopy measurements with nanocrystalline
suspensions of the DPCP analogues 1a-1i should help to
firmly elucidate the kinetics of the reaction and the operative
energy-transfer mechanism and to elucidate the factors that
determine the solution quantum yields. Current work by our
group is aimed at studies involving adiabatic reactions that
occur in the triplet state which, according to triplet-state
lifetimes and known energy-transfer efficiencies, could be
Relative quantum yield determinations were performed with
dicumyl ketone as an internal standard in a Rayonet photochemi-
cal reactor using either 312 nm lamps (BLE-8T312) or 352 nm
lamps (BLE-8T352). Quantum yields were determined with
equimolar, optically dense suspensions. Two independent sus-
pensions of actinometer and substrate were independently
synthesized. They were then combined immediately prior to
irradiation in a 50 mL quartz Erlenmeyer flask and irradiated
for 5 min. Every minute, an aliquot was removed (approximately
one-fifth of the suspension). The suspensions were extracted with
deuterated chloroform (2 mL), washed with brine (2 × 2 mL),
1
and dried over magnesium sulfate. H NMR spectra were taken
immediately to determine the extent of product formation. In
the case of 1g, due to insolubility in chloroform-d3, the
suspensions were evaporated under a stream of argon at room
temperature and ambient pressure. The residue was dissolved
1
in methanol-d4 for H NMR analysis.
The same samples were then subjected to gas chromatography
to determine the extent of dicumyl formation. The conversion
of cyclopropenone derivatives (CPD) could not be monitored
by gas chromatography due to partial thermal decarbonylation
that occurred. Many of the CPD were not thermally stable at
elevated temperatures. DPCP has been shown to thermally
decarbonylate between 130 and 140 °C.46 Thermal decarbony-
lation was noted in all cases for the CPD when subjected to gas
chromatography to give the acetylene exclusively. Therefore,
gas chromatography was not used to detect DPCP/CPD conver-
(45) Hirata, Y.; Okada, T Chem. Phys. Lett. 1993, 209, 397.
9
J. AM. CHEM. SOC. VOL. 131, NO. 32, 2009 11613

A R T I C L E S
Kuzmanich et al.
sion. It was confirmed that, at 60 °C, the cyclopropenone moiety
is stable and does not thermally decarbonylate.
acknowledged. G.K. acknowledges the National Science Foundation
IGERT MCTP Program (DGE0114443).
These experimental yields were reproduced at least in triplicate
with CTAB/H2O. High conversion data were not used to avoid
potential problems caused by the absorption of the products.
Supporting Information Available: Synthetic procedures and
spectral characterization of all compounds; solid-state analysis
(XPD, DSC, and DLS) of 1a as a function of the reaction
progress in the solid state. This material is available free of
charge via the Internet at http://pubs.acs.org.
Acknowledgment. Financial support by the National Science
Foundation (Grants DMR0605688 and CHE0551938) is gratefully
(46) Breslow, R.; Haynie, R.; Mirra, J. J. Am. Chem. Soc. 1959, 81, 247.
JA9043449
9
11614 J. AM. CHEM. SOC. VOL. 131, NO. 32, 2009