Clean photochemical synthesis mediated by radical-radical reactions: Radical buffer or the persistent free radical effect?

Article abstract of DOI:10.1021/ol051932g

(Chemical Equation Presented) A new synthetic methodology is reported that takes advantage of the persistent free radical effect (PFRE), where clean products can be obtained in good yields from radical cross-combination reactions, despite their reputation

Full text of DOI:10.1021/ol051932g

ORGANIC
LETTERS
2005
Vol. 7, No. 22
4979-4982
Clean Photochemical Synthesis
Mediated by Radical Radical Reactions:
−
Radical Buffer or the Persistent Free
Radical Effect?
Kathy-Sarah Focsaneanu, Carolina Aliaga, and J. C. Scaiano*
Department of Chemistry, UniVersity of Ottawa, Ottawa, Ontario, Canada K1N 6N5
tito@photo.chem.uottawa.ca
Received August 10, 2005
ABSTRACT
A new synthetic methodology is reported that takes advantage of the persistent free radical effect (PFRE), where clean products can be
obtained in good yields from radical cross-combination reactions, despite their reputation for being of little synthetic value and for resulting
in complex mixtures; these problems can be avoided when the PFRE is used as a synthetic tool.
Primary, secondary, and even tertiary benzylic radicals are
considered to be stabilized but transient radicals: while these
C-H bonds are much weaker than in a corresponding alkane,
benzylic radicals usually undergo self-termination reactions
at close to the diffusion-controlled limit.1 A classic example
is the benzyl radical 1, which follows a second-order decay
with a half-life of a few microseconds under conditions of
laser excitation.2 However, the diphenylacetonitrile radical
2 is at least 1000 times longer lived. Its remarkable
persistence (and lack of reactivity toward O2) is attributed
to other factors, notably the presence of an R-cyano group.3
Spin density calculations show that for 1, the unpaired
electron exhibits mild delocalization through the aromatic
moiety. Interestingly, 2 shows ∼22% contribution of the
unpaired electron on the nitrogen atom.
dissociation energy for the central C-C bond for the dimer
of 2 was measured as 26.2 ( 0.5 kcal/mol. Thus, even at
room temperature, a small fraction of the radical 2 will be
present.
The dimer formed from 2 (1, 1, 2, 2-tetraphenyl succino-
nitrile, TPS) is stable at room temperature; we have used
the term dynamic stability to describe its behavior before,
since TPS is in fact in equilibrium with the corresponding
radical, as shown in Scheme 1.
Dimers of persistent carbon-centered radicals have recently
been prepared from their corresponding monomers; their
UV-vis spectra revealed that, in toluene solution, they are
in thermal equilibrium with the radical species.4 The bond
TPS owes its apparent stability to the fact that the radicals
are unreactive toward O2 and virtually unreactive toward
Scheme 1. Thermal Dissociation of TPS
(1) Griller, D.; Ingold, K. U. Acc. Chem. Res. 1976, 9, 13-18.
(2) Chatgilialoglu, C. In Handbook of Organic Photochemistry; Scaiano,
J. C., Ed.; CRC Press: Boca Raton, FL, 1989; Vol. II.
(3) Font-Sanchis, E.; Aliaga, C.; Focsaneanu, K.-S.; Scaiano, J. C. Chem.
Commun. 2002, 1576-1577.
(4) Frenette, M.; Aliaga, C.; Font-Sanchis, E.; Scaiano, J. C. Org. Lett.
2004, 6, 2579-2582.
10.1021/ol051932g CCC: $30.25
© 2005 American Chemical Society
Published on Web 09/27/2005

hydrogen abstraction from solvents such as toluene. Thus,
the usual fate of the radicals is to recombine to reform TPS.
If a new species, capable of reacting with the radicals, is
introduced in the system, one can anticipate reactions with
radical 2 leading to new products. We considered exploiting
this characteristic buffer-like behavior via a kinetic phenom-
enon known as the Fischer-Ingold persistent free radical
effect (PFRE).
Therefore, by combining the photoreaction of Scheme 3
and the thermal equilibrium of Scheme 1, we obtain a simple,
yet effective, one-pot synthesis of 2,2,3-triphenylpropionitrile,
TPP, as illustrated in Scheme 4. Although other examples
Scheme 4. Synthesis of 2,2,3-Triphenylproprionitrile.
The PFRE is normally examined in systems where two
different radicals are generated simultaneously, frequently
from a single precursor.5 As shown in Scheme 2, the fact
of synthesis employing the PFRE are described in the
literature,8-16 we believe this method to be the first to involve
C-C bond formation through independent control of per-
sistent and transient radicals.
Scheme 2. Operating Mechanism for the PFREa
Lamp irradiation of a toluene solution (27 °C) containing
3.0 mM TPS and 1.2 mM DBK led to TPP as the main
product, along with a considerable amount of DPE (formation
of the cross-product was optimal using an excess of TPS
starting material). When the steady-state irradiation was
carried out at ∼50 °C, the amount of DPE was significantly
reduced, consistent with the more efficient trapping of 1 by
2 as the steady-state concentration of the latter increases when
the equilibrium of Scheme 1 is displaced at higher temper-
atures. After 40 min of UVB irradiation, 75% of the DBK
was converted to TPP, with the remainder being DPE, formed
initially in the reaction (Figure 1a).
a P ) persistent radical, T ) Transient radical. Usually kPP ) 0.
that the transient radicals (T•) undergo self-reaction, while
the persistent radicals (P•) do not, leads to an accumulation
of the latter. If the cross-reaction to form T-P is reasonably
fast, it soon takes over as the dominant form of decay for
T•. The system is “self-adjusting” in that the accumulation
of P• continues until the kinetic condition favoring cross-
reaction is established. After this point, T• and P• are
consumed in stoichiometric amounts.
Subsequently, a large-scale photolysis was carried out
using 300 mg of DBK and 500 mg of TPS in 250 mL of
Inspired by the PFRE, a novel synthetic method was
devised: produce persistent radicals thermally from their
corresponding dimers, while generating transient radicals in
situ photochemically. The persistent radical-dimer equilib-
rium behaves as an effective “radical buffer”, with the former
trapping the transient radicals to form the cross-reaction
product, while the dimer, through the equilibrium of Scheme
1, keeps the “supply” of radicals available. While the radicals
used here are not strictly “persistent”, since they react to
form the dimer, their combination is effectively “incomplete”
as a result of the radical T dimer equilibrium. As a
consequence, TPS will appear to have lost its stability, when
in reality it is not directly involved in any new reactions.
The photochemistry of dibenzyl ketone (DBK) is well-
known:6,7 steady-state photolysis in toluene using UVB lamps
leads to 1, 2-diphenylethane (DPE) as the only significant
product, formed by combination of two benzyl radicals
according to Scheme 3.
Scheme 3. Photochemistry of DBK
Figure 1. (a) Lamp photolysis of 3.0 mM TPS and 1.2 mM DBK;
(b) pulsed laser photolysis of 2.0 mM TPS and 2.0 mM DBK.
Both irradiations performed at room temperature in deareated
toluene and followed by GC-MS: TPS (b), DBK (9), DPE (O),
TPP (0).
4980
Org. Lett., Vol. 7, No. 22, 2005

deaerated toluene in a 300 mL quartz Erlenmeyer flask.
Following 20 h of UVB irradiation, 0.36 g of TPP was
isolated, corresponding to a conversion of 53%.
Under 308 nm laser excitation, the reaction products were
dominated by DPE (∼65% conversion), with only small
amounts of TPP formed (Figure 1b). This suggests conditions
favorable to recombination, despite the presence of the
equilibrium concentration of 2.
Laser flash photolysis (LFP) experiments of 2.7 mM DBK
and 3.2 mM TPS in benzene reveal the characteristic
absorption of benzyl radicals at 317 nm following 308 nm
laser excitation2 (Figure 2a). At 25 °C, the decay of 1 follows
(Figure 2b). Only after ∼90% of the radicals have decayed
the second-order fit of the data [(signal)-1 vs time] (see inset
in Figure 2b) shows some deviations from linearity, sug-
gesting minor involvement of another mode of decay,
presumably the cross combination of Scheme 4 leading to
TPP.
In contrast with the 25 °C results, at higher temperatures
the difference between the decay traces in the presence and
absence of TPS is clear, as shown in Figure 3. The inset
Figure 3. 2.7 mM DBK and 3.2 mM TPS in toluene after 308 nm
laser excitation, monitored at 320 nm and fitted with second-order
or first + second-order expressions at ∼25 °C (O), ∼55 °C (0),
and ∼85 °C (4). Inset: 2.7 mM DBK, fitted with second-order
kinetics at ∼25 °C (O), ∼55 °C (0), and ∼75 °C (4).
graph shows that this is not an artifact of DBK itself; thus,
the effect of TPS is to accelerate the decay rate for 2.
It is interesting to compare the results from laser and lamp
irradiation. Pulsed laser irradiation tends to favor the
formation of T-T (DPE in our example). This reflects the
thermal and light intensity dependence of the supply of the
two radicals. In general, we expect kTT g kTP. Therefore,
cross-reaction can only dominate if the conditions of eq 1
are fulfilled.
Figure 2. (a) Transient absorption spectrum of 2.7 mM DBK and
3.2 mM TPS in toluene, 0.64 µs after 308 nm laser excitation;
(b) 2.7 mM DBK with (O) and without (0) the presence of
3.2 mM TPS, traces taken at 320 nm after 308 nm laser excitation
at 25 °C, fitted with second-order expressions. Inset: DBK and
TPS in toluene, (signal)-1 vs time.
kTP[T•][P•] . kTT[T•]2; i.e., & kTP[P•] . kTT[T•] (1)
Since radical concentrations generated by pulsed lasers can
easily be 4 orders of magnitude higher than those achieved
under lamp illumination, the latter provides an easier way
to meet this criterion.
clean second-order kinetics, indicating that recombination
to give DPE is the only important reaction path, consistent
with the product studies described above. At this temperature,
the decay of 1 is virtually insensitive to the presence of TPS
On the other hand, meeting the criterion of eq 1 under
conditions of laser excitation should be easier at higher
temperatures, since the concentration of P• will follow a van’t
Hoff dependence with temperature, controlled by the bond
dissociation energy for the dimer. The traces of Figure 3
confirm these ideas; consistent with this, the yield of cross-
combination products under conditions of laser excitation
improves at high temperature. In light of these findings, the
PFRE mechanism in Scheme 4 can be modified as shown
in Scheme 5, in which the persistent radical is introduced
from the corresponding dimer.
(5) Fischer, H. J. Am. Chem. Soc. 1986, 108, 3925-3927.
(6) Engel, P. S. J. Am. Chem. Soc. 1970, 92, 6074-6076.
(7) Lunazzi, L.; Ingold, K. U.; Scaiano, J. C. J. Phys. Chem. 1983, 87,
529-530.
(8) Alajarin, M.; Vidal, A.; Ortin, M.-M.; Bautista, D. New. J. Chem.
2004, 28, 570-577.
(9) Fischer, H. Chem. ReV. 2001, 101, 3581-3610.
(10) Magnus, P.; Venable, J. D.; Shen, L.; Lynch, V. Tetrahedron Lett.
2005, 46, 707-710.
(11) Renaud, P., Sibi, M. P., Eds. Radicals in Organic Synthesis; Wiley-
VCH: New York, 2001; Vols. 1 and 2.
(12) Sanjua´n, A.; Aguirre, G.; Alvaro, M.; Garc´ıa, H.; Scaiano, J. C.;
Chre´tien, M. N.; Focsaneanu, K.-S. Photochem. Photobiol. Sci. 2002, 1,
955-959.
(13) Studer, A. Chem. Eur. J. 2001, 7, 1159-1164.
(14) Studer, A. Chem. Soc. ReV. 2004, 33, 267-273.
(15) Wetter, C.; Studer, A. Chem. Commun. 2004, 174-175.
(16) Zard, S. Z. Radical Reactions in Organic Synthesis; Oxford
University Press: Oxford, 2003.
Org. Lett., Vol. 7, No. 22, 2005
4981

this delocalization with other favorable stereoelectronic
effects leads to a stabilized and persistent benzylic radical.
Charge distributions computed using the ChelpG method
assigns to the carbon radical center in 2 a value that is
noticeably more positive (-0.225) in comparison to the value
assigned to 1 (-0.451). Thus, with respect to each other,
the diphenylacetonitrile radical displays greater electrophilic
character and is activated toward nucleophilic attack from
the benzylic radical 1. Therefore, the resulting product
distribution can be interpreted in terms of charge control:
the significant difference between the charges favors cross-
reaction between the two radicals.
Free-radical reactions, and in particular those involving
radical coupling, have gained a reputation for yielding
complex mixtures of products. However, we can circumvent
this issue if we require the radical intermediates to be
disciplined; that is, rigorous control of the kinetic parameters
defining the radical-radical or radical-molecule reactions.17
We show that reactions involving radicals with different
persistent character can actually result in high yield clean
synthetic procedures whenever a convenient source of the
more persistent radical can be identified, as with radical 2
and a variety of similarly labile radical dimers.4 In our
examples, a dimer effectively behaves as a radical buffer,
maintaining an adequate concentration of radicals. Other
sources may be identified that allow similar concepts to be
applied to synthesis design. For example, we anticipate that
this concept could be further applied to the synthesis of
unsymmetric peroxides, provided the transient radical reacts
with oxygen and the resulting peroxyl can couple with the
persistent radical s a concept similarly explored by Minis-
ci.18 This synthetic approach follows the same reactivity
patterns normally identified with the persistent free radical
effect.
Scheme 5. Mechanism of Formation of Cross-Product, T-P
In typical examples of the PFRE, kPP is near zero and the
radical P• is truly “persistent”; however, this is not a
requirement, as long as kPP , kTT. The example in this paper
is somewhat different from the case in Scheme 2 in that the
two radicals are not formed from a single precursor and are
not produced in a stoichiometric ratio. We previously
described the mechanism of Scheme 2 as “self-adjusting”;
in fact, the mechanism of Scheme 5 also has this charac-
teristic, except that the origin of the concentration adjustment
is different. In Scheme 5, the concentration of persistent
radical is supplied by the dissociation equilibrium for the
dimer P-P. Thus, if P• is consumed by reaction with T•, the
equilibrium will maintain the persistent radical concentration,
effectively acting as a radical buffer. In contrast with Scheme
2, the mechanism of Scheme 5 does not allow the concentra-
tion of P• to grow to any arbitrary value, but rather it is
capped by the equilibrium constant. In effect, P-P is a
“dormant” source of radical P•; however, it is possible to
adjust the concentration by changing the temperature.
Interestingly, the combination of a thermal and a photo-
chemical source for the two radicals allows virtually
independent control of the supply of the two radicals, since
photochemical reactions tend to show low temperature
dependence, but their rate can be adjusted by changing the
light intensity.
Acknowledgment. The Natural Sciences and Engineering
Research Council of Canada (NSERC) has generously
supported this research through a Canada Graduate Scholar-
ship (K.S.F.) and a discovery grant (J.C.S.). We thank M.
Llamazares for assisting with preliminary experiments. We
dedicate this paper to the memory of Prof. Hanns Fischer.
We anticipate that similar syntheses could be performed
using only thermal sources for both T• and P•, although it is
clear that in this case one would lose the independent control
of the two sources that is achieved by using a combination
of thermal and photochemical sources.
To help us rationalize the results obtained, we expanded
our earlier calculation of spin distribution in radicals 1 and
2. The spin density distribution on radical 2 shows that about
50% of the unpaired electron is located on the benzylic
carbon atom, and a 23% contribution of this density is
developed on the nitrogen atom of the cyano group. This
resonance delocalization increases the C-C(N) double bond
character, reducing the bond length from 1.458 Å for the
precursor of 2 to 1.418 Å for radical 2. The combination of
Supporting Information Available: Materials, instru-
ments, and experimental details. This material is available
free of charge via the Internet at http://pubs.acs.org.
OL051932G
(17) Chatgilialoglu, C. In Radicals in Organic Synthesis; Renaud, P.,
Sibi, M. P., Eds.; Wiley-VCH: New York, 2001; Vol. 1: Basic Principles,
p 31.
(18) Bravo, A.; Bjørsvik, H.-R.; Fontana, F.; Liguori, L.; Minisci, F.
J. Org. Chem. 1997, 62, 3849-3857.
4982
Org. Lett., Vol. 7, No. 22, 2005