Broadening the scope of photostimulated SmI2 reductions

Article abstract of DOI:10.1002/chem.200902198

A study was conducted to demonstrate that the exploitation of the biomolecular process significantly broadened the scope of photostimulated SmI₂ reduction. Two UV cells containing naphthalene, SmI₂, and MeOH were set up under a preliminary test and one of them was exposed to a simple UV lamp, while the other was placed on the bench as a control under ordinary laboratory fluorescent lighting. It was observed that the reaction in the cell under the UV lamp progressed slowly, while significant reaction took place in the cell on the bench. Significant reaction was also observed when the same reaction was carried out with a mercury lamp and a filter that cut off wavelengths below 600 nm. Two methods were employed for the assessment of the photostimulated reactions. One of the methods was based on the fact that stopped flow irradiated the sample continuously in the diode array mode.

Full text of DOI:10.1002/chem.200902198

COMMUNICATION
DOI: 10.1002/chem.200902198
Broadening the Scope of Photostimulated SmI2 Reductions
Mazal Amiel-Levy and Shmaryahu Hoz*[a]
The electron-transfer process is the pivotal step in SmI2
mate mechanistic details of SmI2 reactions may assist in the
design of a strategy to broaden its scope.
reductions.[1] In most cases, this step seems to be endother-
mic and slow. It could be enhanced by increasing the reduc-
tion potential of SmI2, for example, by using additives such
as hexamethylphosphoramide (HMPA).[2] Another way to
increase the reduction potential of SmI2 is by irradiation.[3]
However, in contradistinction to the HMPA method, the
range of applicability of this method is surprisingly limited.
To the best of our knowledge, the reaction types shown in
Scheme 1 are the only examples of the application of this
method.[3]
The common denominator for the photostimulated reac-
tions shown in Scheme 1 is that there is no bimolecular rate-
determining step following the electron transfer. These reac-
tions involve either a dissociative electron attachment[4] pro-
cess, such as in the mesolytic cleavage[5] of alkyl halides, or,
as may be the case for C6H5X leaving groups, the radical
anion may have a short lifetime before it either donates the
electron back to the SmI2 or undergoes cleavage. None of
these reactions necessitate a bimolecular step, which should,
for successful reaction, effectively compete with back elec-
tron transfer.
Once the reason for the limited applicability of the photo-
stimulation technique is understood, knowledge of the inti-
It should be noted that in the ground state there is no ap-
preciable electron transfer from SmI2 to the substrate in
these cases. It is only after the SmI2 is excited that the pro-
cess becomes viable (excitation at around 600 nm is equiva-
lent to approximately 2.2 eV, which is much more than the
0.72 V increase effected by HMPA (4 equiv).[6]) Yet, once
electron transfer has taken place Sm3+ is no longer excited
and, therefore, back electron transfer from the radical anion
to the ground state of Sm3+ has a very high thermodynamic
driving force. Hence, rapid back electron transfer prevents
any bimolecular encounter, such as the protonation of the
radical anion in the Birch-type reduction sequence[7] (elec-
tron–proton–electron–proton transfer).
The easiest way to overcome the deficiency of the ex-
tremely short lifetime of the radical anion caused by rapid
back electron transfer is to convert the consequent bimolec-
ular process, for example, protonation, into a unimolecular
one. We have shown that the fact that MeOH complexes to
[8]
SmI2 enables protonation to occur from within the ion pair
generated by electron transfer.[9] Protonation within the ion
pair is a unimolecular process, which may therefore effec-
tively trap the radical anion as it is formed and, therefore,
the need for a bimolecular encounter between the short-
lived radical anion and a proton donor from the bulk solu-
tion is rendered unnecessary. Herein, we show that exploit-
ing this proximity effect[10] significantly broadens the scope
of photostimulated SmI2 reduction.
Scheme 1. SmI2 reduction under irradiation conditions. Ts=tosyl.
[a] Dr. M. Amiel-Levy, Prof. S. Hoz
Department of Chemistry, Bar-Ilan University
Ramat-Gan 52900 (Israel)
E-mail: shoz@mail.biu.ac.il
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200902198.
Chem. Eur. J. 2010, 16, 805 – 809
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
805

In a preliminary test, we set up two UV cells containing
naphthalene, SmI2, and MeOH and exposed one to a simple
UV lamp (UVSL-25; 254/366 nm) such as those used to
locate spots on TLC plates, while the other was placed on
the bench as a control under ordinary laboratory fluorescent
lighting. In the event, the reaction in the cell under the UV
lamp hardly progressed, whereas significant reaction took
place in the cell on the bench. Moreover, significant reaction
was observed when the same reaction was carried out with a
mercury lamp and a filter that cuts off wavelengths below
600 nm. Since most of the synthetic SmI2 reductions are car-
ried out under laboratory lighting, chemists should be aware
of the possibility that their reactions may, at least in part, be
photostimulated.
We have employed two methods for the assessment of the
photostimulated reactions. The first method was based on
the fact that, in the diode array mode, the stopped flow
spectrophotometer irradiates the sample continuously (in
the single wavelength (SWL) mode, the cell is exposed to
only a minute amount of irradiation).[11] Figure 1 shows the
reactions of naphthalene with SmI2 in the presence of
MeOH or trifluoroethanol (TFE), an alcohol that does not
complex to SmI2.[9,12] In the presence of MeOH, a reaction
time of 200 s corresponds to approximately two half-lives,
whereas with TFE, following the disappearance of the SmI2
after 800 s, revealed only natural decay.
The second method involves placing the UV cell with the
reaction mixture (prepared in the glove box) 7.5 cm from
the tip of a 450 W low-pressure mercury lamp for intervals
of 20 s to 2 min (depending on the reaction rate) and meas-
uring the absorption at the end of each interval.
Three substrates were studied: naphthalene, diphenylace-
tylene, and methylbenzoate. None of the three substrates re-
acted in the dark or in the light in the presence of TFE as
the proton donor. With MeOH as the proton donor, naph-
thalene gave the two isomers a and b (Scheme 2a) in >95%
yield. These were obtained in a ratio of a/b=1.3. Reduction
by the Hilmersson method preferentially yields b.[13] The re-
duction of diphenylacetylene yielded cis and trans stilbene
(Scheme 2b). The overall yield was 90% and the cis isomer
was obtained, probably as a result of the irradiation of the
trans isomer. The product from methylbenzoate was benzyl
alcohol (80% yield; Scheme 2c).
Figure 1. Diode array monitoring of the reaction of naphthalene (0.2m)
and SmI2 (2.5 mm) in the presence of a) MeOH (1m, 200 s) and b) TFE
(1m, 800 s).
The photostimulated reactions were performed at high
optical density (OD), since under these conditions all of the
light is absorbed and the concentration of the reactive re-
ductant (excited SmI2) is independent of the total concentra-
tion of SmI2. Therefore, as demonstrated in Figure 2, which
shows the kinetics for the naphthalene reduction, the reac-
tions were zero order in SmI2.
Scheme 2. Photostimulated SmI2 reductions.
A plot of the pseudo-first-order rate constants versus the
concentration of naphthalene (Figure 3 and Figures S1 and
S2 in the Supporting Information) show that increasing the
concentration of the substrate causes a rate increase, which
levels off at high substrate concentrations.
a moderate decrease. This is exemplified in Figure 4 for
naphthalene and in Figures S3 and S4 in the Supporting In-
formation for the other substrates. These results are consis-
tent with the mechanism shown in Scheme 3.
The rate as a function of MeOH concentration shows an
increase to a maximum at around 1.5m MeOH followed by
In the first step, excited SmI2 may either transfer an elec-
tron to the substrate or decay to its ground state. The proba-
806
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 805 – 809

Photostimulated SmI2 Reductions
COMMUNICATION
Figure 2. Change of absorption as a function of the irradiation time in
Scheme 3. Photostimulated reactions in the presence of SmI2 coordinated
~
^
the reaction of naphthalene (0.2m) and 2.17 ( ), 3.28 (&), 4.28 ( ), or
with proton donors (A=substrate).
&
5.4 mm ( ) of SmI2 in the presence of MeOH (1m).
of MeOH increases. However, if, as we assume, the proton
is transferred from MeOH coordinated to the SmI2, then
once the coordination sphere is fully occupied with MeOH
molecules, a plateau should be reached. However, the plots
for the three substrates show maxima, which are then fol-
lowed by a decrease in rate. The rate attenuation by MeOH
in concentrations over 1.5m can be explained in two ways.
The first explanation is that the electron transfer in our re-
actions is of the inner-sphere type.[13,15] A tight methanol sol-
vation shell around SmI2 hinders inner-sphere electron
transfer and the reactions at high MeOH concentrations
may either take place with the small number of SmI2 mole-
cules that have an incomplete methanol coordination shell;
or the outer-sphere mechanism can also operate, but at a
rate lower than that of the inner-sphere mechanism. Alter-
natively, Scaiano and co-workers reported that the lifetime
of SmI2 in the excited state at 355 nm is lowered,[16] by coor-
dinating HMPA, molecules from 125 to 16 ns in the pres-
ence of HMPA (29 mm). It is possible that MeOH coordina-
tion has a similar effect on the lifetime of the SmI2 in the
excited state in the visible range. This possibility was tested
in the following way: If indeed the lifetime of the excited
SmI2 is shortened by MeOH coordination, then at high
MeOH concentration more substrate would be needed to
trap the short-lived SmI2 before it spontaneously decays. As
a result, the plateau in the plot of the rate versus substrate
concentration would be reached at a higher substrate con-
centration. We have examined this hypothesis by plotting
the rates of the reactions of SmI2 with naphthalene versus
the naphthalene concentration in the presence of 1 and 6.2m
MeOH (Figure 5).
Figure 3. Rate versus naphthalene concentration in its reaction with SmI2
(4.7 mm) and MeOH (1m).
Figure 4. Rate versus MeOH concentration in the reduction of naphtha-
The results show that the plateau is reached at about the
same concentration of naphthalene and, therefore, we con-
clude that the reason for the rate decline observed at
MeOH concentrations greater than one is due to the hin-
dered approach of the substrate to the MeOH-engulfed ex-
cited SmI2. This inner–outer sphere argument could be re-
fined further. The maximum in the plot of rate versus sub-
strate concentration is around 1.5m MeOH. However, this
concentration is well below the saturation concentration in
which the SmI2 is fully coordinated to methanol molecules.
This is evidenced from the spectrum of SmI2 as a function of
MeOH (Figure S5 in the Supporting Information). The
lene (0.2m) with SmI2 (5 mm).
bility for electron transfer increases linearly with the con-
centration of the substrate. The plateau is achieved when all
of the excited SmI2 interacts with the substrate at a rate
faster than it decays.[14] Once the radical anion of the sub-
strate is formed it may be protonated by the complexed
MeOH, which prevents back electron transfer.
The rate dependence on MeOH is less straightforward. It
is expected that the rate will increase as the concentration
Chem. Eur. J. 2010, 16, 805 – 809
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
807

S. Hoz and M. Amiel-Levy
have shown that the applicability range of photostimulated
SmI2 reactions could be vastly extended by converting the
bimolecular protonation step to a unimolecular one. It
should be noted that ethylene glycol and water also form
strong complexes with SmI2 and may serve the same pur-
pose as MeOH.[18] This effectively prevents the rapid back
electron transfer, which was found to be the cause for the
limited applicability of photostimulated SmI2 reductions.
Thus, we demonstrate once again the importance of under-
standing the mechanistic details of a reaction to design a
successful strategy for its employment.
Figure 5. Rate versus naphthalene concentration in its reaction with SmI2
&
^
(4.7 mm) in the presence of 1 ( ) and 6.2m ( )MeOH, left and right axes,
Keywords: electron transfer · photochemistry · radical ions ·
reduction · samarium iodide
respectively.
double-humped spectrum (560 and 618 nm) typical of SmI2
in THF coalesces into a single absorption only at MeOH
concentrations of around 4m. Thus, at 1.5m MeOH, not all
the sites around the SmI2 are taken by methanol molecules
and the substrate can form an intimate contact with the
SmI2. Yet, in spite of this incomplete solvation shell, the re-
action rate drops. A possible explanation is that there are
(at least) two different coordination sites around the SmI2
and inner-sphere electron transfer occurs preferentially from
one of these sites. At MeOH concentrations higher than
1.5m, MeOH molecules start to occupy this site as well, ob-
structing the inner-sphere electron transfer and decreasing
the reduction rate.
[1] For some recent reviews, see: a) D. J. Edmonds, D. Johnston, D. J.
Procter, Chem. Rev. 2004, 104, 3371–3403; b) H. B. Kagan, Tetrahe-
dron 2003, 59, 10351–10372; c) P. G. Steel, J. Chem. Soc. Perkin
Trans. 1 2001, 2727–2751; d) A. Krief, A.-M. Laval, Chem. Rev.
1999, 99, 745–777; e) G. A. Molander, C. R. Harris, Tetrahedron
1998, 54, 3321–3354; f) G. A. Molander, C. R. Harris, Chem. Rev.
1996, 96, 307–338.
[2] a) H. Farran, S. Hoz, Org. Lett. 2008, 10, 865–867; b) R. A. Flowers,
Synlett 2008, 1427–1439; c) A. Dahlen, G. Hilmersson, Eur. J. Inorg.
Chem. 2004, 3393–3403; d) Z. Hou, Y. Wakatsuki, J. Chem. Soc.
Chem. Commun. 1994, 1205–1206; e) G. A. Molander, Org. React.
1994, 46, 211–367; f) G. A. Molander, J. A. McKie, J. Org. Chem.
1992, 57, 3132–3139; g) J. Inanaga, M. Ishikawa, M. Yamaguchi,
Chem. Lett. 1987, 1485–1486.
To identify the rate-determining step, we measured the ki-
netic H/D isotope effect. Table 1 displays the isotope effect
as a function of MeOH concentration for the reduction of
naphthalene.
[3] a) J. M. Concellꢁn, H. Rodrꢂguez-Solla, C. Simal, D. Santos, N. R.
Paz, Org. Lett. 2008, 10, 4549–4552; b) J. M. Concellꢁn, H. Rodrꢂ-
guez-Solla, C. Simal, M. Huerta, Org. Lett. 2005, 7, 5833–5835; c) E.
Prasad, B. W. Knettle, R. A. Flowers II, Chem. Eur. J. 2005, 11,
3105–3112; d) Y. Sumino, N. Harato, Y. Tomiska, A. Ogawa, Tetra-
hedron 2003, 59, 10499–10508; e) A. Ogawa, S. Ohya, M. Doi, Y.
Sumino, N. Sonoda, T. Hirao, Tetrahedron Lett. 1998, 39, 6341–
6342; f) A. Ogawa, S. Ohya, T. Hirao, Chem. Lett. 1997, 275–276;
g) A. Ogawa, Y. Sumino, T. Nanke, S. Ohya, N. Sonoda, T. Hirao, J.
Am. Chem. Soc. 1997, 119, 2745–2746.
Table 1. H/D isotope effect as a function of methanol concentration in
the reaction of naphthalene (0.2m) and SmI2 (5 mm).
A
RateH/RateD
[4] a) M. Rajappan, L. L. Zhu, A. D. Bass, L. Sanche, C. R. Arumai-
nayagam, J. Phys. Chem. C 2008, 112, 17319–17323; b) T. Clark, G.
Illing, J. Chem. Soc. Chem. Commun. 1985, 529–530; c) T. F.
O’Malley, Phys. Rev. 1966, 1, 14–29; d) J. Bertran, I. Gallardo, M.
Moreno, J.-M. Savꢃant, J. Am. Chem. Soc. 1992, 114, 9576–9583.
[5] P. Maslak, J. N. Narvaez, Angew. Chem. 1990, 102, 302–304; Angew.
Chem. Int. Ed. Engl. 1990, 29, 283–285.
[6] a) J. B. Shotwell, J. M. Sealy, R. A. Flowers, J. Org. Chem. 1999, 64,
5251–5255; b) M. Shabangi, R. A. Flowers, Tetrahedron Lett. 1997,
38, 1137–1140; c) R. J. Enemaerke, T. Hertz, T. Skrydstrup, K.
Daasbjerg, Chem. Eur. J. 2000, 6, 3747–3754.
[7] a) H. L. Dryden, G. M. Webber, R. R. Burtner, J. A. Cella, J. Org.
Chem. 1961, 26, 3237–3245; b) N. L. Allinger, M. P. Cava, D.
DeJonj, C. R. Johnson, N. A. Lebel, C. L. Stevens, Organic Chemis-
try, Worth Publishing, New York, 1973, p. 382.
[8] a) H. Farran, S. Hoz, J. Org. Chem. 2009, 74, 2075–2079; b) A. Tar-
nopolsky, S. Hoz, Org. Biomol. Chem. 2007, 5, 3801–3804; c) G.
Kleiner, A. Tarnopolsky, S. Hoz, Org. Lett. 2005, 7, 4197–4200;
d) A. Yacovan, I. Bilkis, S. Hoz, J. Am. Chem. Soc. 1996, 118, 261–
262.
0.1
1
6.2
1.47Æ0.18
1.22Æ0.29
1.1Æ0.2
The isotope effect decreases with the increase in the con-
centration of MeOH, suggesting that at the beginning proto-
nation is rate determining. However, as the MeOH concen-
tration is increased and there are more MeOH molecules
around the SmI2, the capture of the radical anion by a
MeOH molecule from within the ion pair is almost complete
and, therefore, proton transfer ceases to be rate determin-
ing[17] and the isotope effect decreases.
In conclusion, photostimulated SmI2 reduction reactions
enjoy three major benefits: 1) The reducing power of SmI2
is significantly increased; more than by any additive. 2) The
reaction mixture is not contaminated by an additive.
3) Most substrates are inert to photostimulation at 600 nm.
Up to now, and in spite of these advantages, only a very
small group of substrates could be reduced this way. We
[9] M. Amiel-Levy, S. Hoz, J. Am. Chem. Soc. 2009, 131, 8280–8284.
[10] a) L. Mandolini, Adv. Phys. Org. Chem. 1987, 22, 1–111; b) G. Illu-
minati, L. Mandolini, Acc. Chem. Res. 1981, 14, 95–102; c) A.
Kirby, J. Adv. Phys. Org. Chem. 1981, 17, 183–278; d) G. Illuminati,
L. Mandolini, B. Masci, J. Am. Chem. Soc. 1977, 99, 6308–6312.
808
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 805 – 809

Photostimulated SmI2 Reductions
COMMUNICATION
[11] This should also be considered by kineticists who perform measure-
ments in the diode array mode rather than in the SWL mode.
[12] V. Dhandapani, P. K. Sadasivam, S. Antharjanam, E. Prasad, R. A.
Flowers II, J. Am. Chem. Soc. 2008, 130, 7228–7229.
1999, 343–344; d) M. Shabangi, M. L. Kuhlman, R. A. Flowers, Org.
Lett. 1999, 1, 2133–2135.
[16] W. G. Skene, J. C. Scaiano, F. L. Cozens, J. Org. Chem. 1996, 61,
7918–7921.
[13] A. Dahlꢃn, A. Nilsson, G. Hilmersson, J. Org. Chem. 2006, 71,
1576–1580.
[14] It should be emphasized that the quantum yield at the plateau is not
necessarily 1.0, since the encounter between the two reactants may
lead, not only to electron transfer, but also to quenching.
[15] a) E. Prasad R. A. Flowers, J. Am. Chem. Soc. 2002, 124, 6895–
6899; b) R. S. Miller, J. M. Sealy, M. Shabangi, M. L. Kuhlman, J. R.
Fuchs, R. A. Flowers, J. Am. Chem. Soc. 2000, 122, 7718–7722;
c) R. J. Enemaerke, K. Dassbjerg, T. Skrydstrup, Chem. Commun.
[17] Depending on the concentration of the substrate, the rate-limiting
step may be either the electron transfer or the rate may be inde-
pendent of any of the reactants and be affected only by the irradia-
tion intensity.
[18] We chose MeOH because its reaction rates were more suitable for
our measurements.
Received: August 6, 2009
Published online: November 26, 2009
Chem. Eur. J. 2010, 16, 805 – 809
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
809