Channeling the SmI2 reactions to the radical path: Radicals resisting reduction by SmI2

Article abstract of DOI:10.1021/ol501490f

Studies on the reaction of 4-(2,2-diphenylvinyl)pyridine with SmI ₂ revealed that the intermediate radical strongly resists further reduction to the corresponding anion. The resistance of the radical to accepting another electron is traced to its stabilization by the nitrogen lone pair. The literature suggests that oxygen may also play a role similar to that of nitrogen in directing the course of the reaction toward radical rather than to anionic chemistry.

Full text of DOI:10.1021/ol501490f

Letter
pubs.acs.org/OrgLett
Terms of Use
Channeling the SmI2 Reactions to the Radical Path: Radicals Resisting
Reduction by SmI2
Ramesh Yella and Shmaryahu Hoz*
Department of Chemistry, Bar-Illan University, Ramat-Gan 52900, Israel
S
* Supporting Information
ABSTRACT: Studies on the reaction of 4-(2,2-diphenylvinyl)pyridine
with SmI2 revealed that the intermediate radical strongly resists further
reduction to the corresponding anion. The resistance of the radical to
accepting another electron is traced to its stabilization by the nitrogen
lone pair. The literature suggests that oxygen may also play a role similar
to that of nitrogen in directing the course of the reaction toward radical
rather than to anionic chemistry.
However, arriving at the radical intermediate may lead not
ne of the major reasons for the popularity of samarium
1
Odiiodide as a reducing agent is its versatility. While in only to P2 but also to P3. The present paper identifies the
conditions for an effective prevention of the conversion of the
radical (leading to P2) into the anion (leading to P3).
We have recently studied the reduction by SmI2 of substrates
containing nitrogen in the π system.4,5 We suggest that using
these and similar types of substrates can slow the electron
transfer from SmI2 to the radical, thereby increasing the
likelihood of P2 as the dominant reaction path (Scheme 1).
The nitrogen-containing substrates BAI, BMI and BPI,
studied previously by us, display a very colorful and unique
behavior in their reactions with SmI2.4a
hydride-transfer reagents reduction of a general double bond
AB will result in an anionic intermediate HA−B−, reduction
by SmI2 may result in three different reactive intermediates,
each of which may undergo a different inter- and intramolecular
reaction leading to various products (P1−P3; Scheme 1).
Scheme 1
The reactions of these substrates display, among other
unique features, autocatalysis and zero-order kinetics which are
atypical of other substrates.
We have shown in the past that one can apply some control
over the reaction paths by an educated choice of the proton
donor in the reaction.2 Proton donors which complex to SmI2,
such as MeOH or water, will protonate the radical anion
efficiently leading to P2. That is because the protonation takes
place within the ion pair in a unimolecular fashion. Therefore,
the production of P1 will be largely reduced. However, when a
proton donor which is incapable of forming a complex with
SmI2 is used, the radical anion will not be protonated and the
formation of P1 is enabled at the expense of the production of
P2 and P3. This model allows us to explain2a the enigma posed
by the elegant work of Procter et al., who discovered that
MeOH and t-BuOH induced different cyclization modes (eq
1).3 It was suggested that the t-BuOH-derived product involves
When the nitrogen atom was moved from the double bond
undergoing the reduction to a remote position, yet still in
conjugation with the central bond as in 4SP, the aforemen-
tioned phenomena repeated themselves.5 The most profound
feature observed for 4SP was the resistance of the intermediate
radical to accepting another electron. Namely, even when
excess SmI2 was used, the radical dimerized as shown in eq 2
and no reduced monomer was obtained.
The limited tendency of the radical to accept an electron is
surprising in light of the rate constant measured for electron
transfer from SmI2, which is 5 × 105 M−l s−1 to alkyl radical6
and 5.3 × 107 M−l s−1 to benzyl radical.7
Following this discovery, we studied BSP, a derivative of 4SP
in which a hydrogen atom was replaced by a phenyl group in
the action of the negative charge of the radical anion (P1) while
the product obtained in the presence of MeOH seems to be
formed by the radical path (P2).
Received: May 23, 2014
Published: July 17, 2014
© 2014 American Chemical Society
3876
dx.doi.org/10.1021/ol501490f | Org. Lett. 2014, 16, 3876−3879

Organic Letters
Letter
order to block the radical combination path. As will be shown
below, this led to the formation of a relatively stable radical
which could be observed by UV−vis spectroscopy.
As expected, the replacement of a hydrogen atom by a
phenyl ring indeed quenched the dimerization process, and the
only product obtained was the reduced monomer (eq 3).
Figure 2. Diode array monitoring of the reaction of BSP (5 mM) and
SmI2 (0.5 mM) in THF in the presence of MeOH.
However, as will be shown later on, this reduced monomer
was NOT obtained via P3, but rather via the P2 path.
Monitoring the kinetics of the reaction in the presence of
MeOH or trifluoroethanol (TFE) at the absorption maximum
of the SmI2 (619 nm) under pseudo-first-order conditions
showed that instead of the expected decrease in the absorption
due to the disappearance of the SmI2 an unexpected rapid
increase of the absorption is observed. This absorption is then
followed by a slow decay (Figure 1 and Figures S1, Supporting
Information).
Ab initio calculations at the B3P86/6-311++G(d,p) and
MO62X/6-311++G(d,p)8 levels showed absorption maxima
with varying intensities in the range 450−650 nm, in agreement
with the observed absorption (see the Supporting Information).
Thus, the chemical rationale, and the agreement between the
spectroscopic observation and the quantum mechanical
calculations, support the conclusion that the observed
intermediate is indeed BSP-R.
One of the most important features of this radical is that the
alcohol concentration has hardly any effect on the rate of its
decay. This is evident from Figure 1 and is unambiguously
demonstrated in the reactions where 1 and 100 mM of acetic
acid were used (Figure S2, Supporting Information). Since
there is no dimerization and in light of the lack of rate
dependence on the proton donor type or concentration, the
question is what is the mechanism of the BSP-R disappearance?
It should be pointed out that the observed kinetic traces for the
decay clearly do not follow first-order kinetics. On the other
hand, starting with the analysis after ca. the first 20% of the
reaction, assuming that at this point there is no additional
production of the radical and the only reaction observed is the
decay of the radical, the general second-order equation for A +
A→ P (eq S1, Supporting Information)9 gives a good straight
line for the plot of 1/O.D. vs t (Figure S3, Supporting
Information).
In these plots, A was replaced by the optical density (O.D.)
which is proportional to the concentration (A) of the radical.
The average slope for the bimolecular reactions of BSP-R in the
presence of MeOH (0.5−2 M), TFE (0.5−2 M), or AcOH
(0.001, 0.1 M) is 231 29 with r2 ≥ 0.997. Despite the lack of
dependence of the rate of the radical disappearance on the
proton donor concentration, we have studied the kinetic H/D
effect using MeOH and MeOD. Surprisingly, using the MeOH/
D concentrations of 0.5, 1, and 2 M, we have found an H/D
isotope effect of 3.5 0.3 (Figure 3 and Figure S4, Supporting
Information). It should be noted the maximum absorbance in
the kinetic trace is higher for MeOD than for MeOH because
the competing annihilation reaction of the radical is slower.
The absence of proton donor effect on the rate, on the one
hand, and the existence of a H/D isotope effect on the other,
Figure 1. Absorption at 619 nm vs time for the reaction of BSP (5
mM) and SmI2 (0.5 mM) in THF in the presence of variable amounts
of TFE.
We have shown for 4SP,5 that in the first step, a hydrogen
bond is formed between the proton donor and the lone pair on
the nitrogen. In the second step, concomitant with the electron
transfer, the hydrogen bond tightens to give a N−H bond, and
the corresponding radical is formed. In the case of 4SP, the
radical undergoes rapid combination to give the dimer. In the
present case, because the dimerization step is inhibited by the
added phenyl group, the increase in the absorbance is attributed
to the formation of the corresponding radical (BSP-R, eq 4).
Continuous monitoring of the reaction using a diode array
showed that λmax of the radical is around 606 nm (Figure 2).
3877
dx.doi.org/10.1021/ol501490f | Org. Lett. 2014, 16, 3876−3879

Organic Letters
Letter
Scheme 3
and since this orbital is shifted upward in energy more than the
bonding orbital is shifted downward, the net outcome is the
destabilization known as the 4-electron destabilization effect.
Additional support for our suggestion that the lone pair is
responsible for the resistance of BSP-R to accept another
electron comes from ab initio calculations. We have calculated
at the B3LYP/6-31+G* level with the polarizable continuum
model (dielectric constant of THF) the energy of both the
radical and its corresponding anion. In the second step, we have
rotated the pyridine ring out of plane so that its π-system will
not interact with the radical/carbanion (the two conformations
are shown for the radical in Figure S5, Supporting
Information). This out-of-plane rotation causes an increase in
energy of 12.6 kcal/mol for the radical and only 5.3 kcal/mol
for the anion. Thus, the pyridine ring differential effect on the
two species amounts to ca. 7 kcal/mol. This value translates to
a rate retardation of ca. 6 orders of magnitude for an electron
transfer to BSP-R.
It is also clear from Scheme 3 that electron transfer to the
SOMO will be more difficult than to a radical, which does not
enjoy the lone pair stabilization because of the difference in
energy of the two orbitals.
Figure 3. Kinetic isotope effect for the reaction of BSP (5 mM)
andSmI2 (0.5 mM) in THF in the presence of MeOH/D (1 M): (a)
kinetic traces; (b) fit to second-order equation.
In conclusion, the resistance of the radical to further
reduction by SmI2 is derived from the lone pair induced
stabilization of the radical as well as the concomitant
destabilization of the corresponding anion.
clearly point at hydrogen radical transfer as the rate-
determining step. This, combined with the bimolecular nature
of the reaction, leads to a mechanism in which one BSP-R
transfers a hydrogen radical to the other (Scheme 2). The final
step is prototropy with rearomatization of the pyridine ring.
In principle, this effect can be exploited using other lone-pair-
carrying atoms such as oxygen. Ab initio calculations at the
B3LYP/6-31+G* level8 on simple model radicals for oxygen
and nitrogen atoms at a position α to the radical center
(MeCHNH2 and MeCHOH, see Figure S6 and data in the
Supporting Information) in their optimal geometry and an
enforced orthogonal geometry show that nitrogen stabilizes the
neighboring radical by more than 4 kcal/mol relative to oxygen
(in the optimal geometry, the π lone pair of the oxygen atom is
overlapping the singly occupied orbital whereas in the
orthogonal geometry it is the σ lone pair which, due to its
low energy, is not expected to contribute significantly to the
radical stabilization).
Despite the lesser efficiency of oxygen, the remarkable and
exhaustive work of Procter et al. on cyclic lactones nicely
demonstrates this principle for oxygen.12−19 It is possible that
in lactones this lesser efficiency is compensated for by having
two oxygen atoms flanking the radical center (eq S2,
Supporting Information). It was shown that (a) the electron
transfer to the carbonyl group is favored when the ring oxygen
is suitably situated (six membered ring) to stabilize the odd
electron on the carbonyl carbon12,13 and (b) after protonation
on the oxygen, the radical which is now stabilized by two
oxygen atoms, undergoes predominantly typical radical
reactions16,17 rather than accepting an additional electron,
although the reactions are sometimes carried out in the
presence of a large excess (6−8 equiv) of SmI2.15
Scheme 2
A thought-provoking question is why BSP-R and the
corresponding radical of 4SP resist further reduction by SmI2
to generate the corresponding anion. We suggest that this
resistance results from the differential interaction between the
nitrogen lone pair and the single electron in the radical relative
to that with the two electrons in the anion.
In general, a radical interacting with a lone pair derives
stability from this interaction. In the MO terminology, this
stabilization stems from the fact that in the MO generated
(Scheme 3), two electrons are stabilized while only one
electron goes up in energy.10 This stabilization of the radical is
contrasted by a destabilization of the anion. The latter
destabilization stems from the fact that the denominator of
the interaction energy term is 1 S (where S is the overlap
integral). Consequently, the antibonding orbital is destabilized
slightly more than the bonding orbital is stabilized.11 Namely,
in the anion, the antibonding orbital is now doubly occupied,
3878
dx.doi.org/10.1021/ol501490f | Org. Lett. 2014, 16, 3876−3879

Organic Letters
Letter
2010. (d) Concellon
́
, J. M.; Rodríguez-Solla, H.; Concellon
́
, C.; Amo,
It should be emphasized that acyclic esters normally do not
undergo reduction by SmI2. This is because in order to benefit
from the stabilization of the odd electron by the π-lone pair on
the oxygen, the free rotation must be frozen, and apparently,
the entropic cost is too high to permit this option.
It is worth noting that the resistance of the radical to accept
the second electron is not an absolute property but is always
relative to the other options. Thus, in the absence of a more
attractive reaction path, the radical will accept an electron. An
example of this is the lactone carrying an additional ester group
where a deuteriolactol was obtained.18,19 In this case and in the
Meldrum acid reactions,14 the inductive effect of the
neighboring group probably contributes some driving force to
the radical conversion into a carbanion. In addition, the ring
may be flexible enough to avoid to some degree the overlap
which results in the four-electron destabilization.
Finally, the reactivity of a lone-pair-stabilized radical should
be discussed. In principle, one may argue that because of its
stabilization, the reactivity of this type of radical may be
relatively low. A more detailed analysis shows that this is not
necessarily the case. In symmetric radical combination reactions
its reactivity will not differ much from that of the unperturbed
radical. In additions to double bonds, its reactivity depends on
the nature of the transition state. In early transition states, this
radical is expected to be nucleophilic and more reactive than
the unperturbed radical. However, in the case of a late
transition state, it will resemble the reactivity of the
unperturbed radical. In terms of valence bond this is because
the stabilization stemming from the partial double bond
between the atoms carrying the three electrons is lost at this
late transition state. In electrophilic radical additions to double
bonds, in an early transition state, the stabilized radical will be
more sluggish than the normal radical, and in a late transition
state it will display a similar reactivity.
V. Chem. Soc. Rev. 2010, 39, 4103. (e) Honda, T. Heterocycles 2010, 81,
2719. (f) Nakata, T. Chem. Soc. Rev. 2010, 39, 1955. (g) Nicolaou, K.
C.; Ellery, S. P.; Chen, J. S. Angew. Chem., Int. Ed. 2009, 48, 7140.
(h) Gopalaiah, K.; Kagan, H. B. New J. Chem. 2008, 32, 607.
(i) Ichikawa, S. Chem. Pharm. Bull. 2008, 56, 1059. (j) Faugeroux, V.;
Genisson, Y. Curr. Org. Chem. 2008, 12, 751. (k) Jung, D. Y.; Kim, Y.
H. Synlett 2005, 3019. (l) Concellon
Soc. Rev. 2004, 33, 599. (m) Dahlen
́
, J. M.; Rodrıgu
́
ez-Solla, H. Chem.
́
, A.; Hilmersson, G. Eur. J. Inorg.
Chem. 2004, 3393. (n) Edmonds, D. J.; Johnston, D.; Procter, D. J.
Chem. Rev. 2004, 104, 3371. (o) Szostak, M.; Fazakerley, N. J.; Parmar,
D.; Procter, D. J. Chem. Rev. 2014, 114, 5959. (p) Choquette, K. A.;
Flowers, R. A. Sm and Yb Reagents. In Comprehensive Organic
Synthesis, 2nd ed.; Molander, G. A., Knochel, P., Eds.; Elsevier: Oxford,
2014; Vol. 1, pp 279−343.
(2) (a) Amiel-Levy, M.; Hoz, S. J. Am. Chem. Soc. 2009, 131, 8280.
(b) Amiel-Levy, M.; Hoz, S. Chem.Eur. J. 2010, 16, 805.
(3) Hutton, T. K.; Muir, K. W.; Procter, D. J. Org. Lett. 2003, 5, 4811.
(4) (a) Rao, C. N.; Hoz, S. J. Am. Chem. Soc. 2011, 133, 14795.
(b) Rao, C. N.; Hoz, S. J. Org. Chem. 2011, 76, 9438.
(5) Yella, R.; Hoz, S. Org. Lett. 2013, 15, 5262.
(6) Hasegawa, E.; Curran, D. P. Tetrahedron Lett. 1993, 34, 1717.
(7) Skene, W. G.; Scaiano, L. C.; Cozens, F. L. J. Org. Chem. 1996,
61, 7918.
(8) GAUSSIAN 03, revision A.04; Gaussian Inc., Pittsburgh, 2003.
(9) Frost, A. A.; Pearson, R. G. Kinetics and Mechanism; Wiley: New
York, 1961.
(10) Hoz, S. J. Org. Chem. 1982, 47, 3545−3547.
(11) Anslyn, E. V.; Dougherty, D. A. Modern Physical Organic
Chemistry; University Science Books: Sausalito, CA, 2006; p 831.
(12) (a) Kamochi, Y.; Kudo, T. Chem. Lett. 1993, 1495. (b) Duffy, L.
A.; Matsubara, H.; Procter, D. J. J. Am. Chem. Soc. 2008, 130, 1136.
(13) Parmar, D.; Duffy, L. A.; Sadasivam, D. V.; Matsubara, H.;
Bradley, P. A.; Flowers, R. A., II; Procter, D. J. J. Am. Chem. Soc. 2009,
131, 15467.
(14) Guazzelli, G.; De Grazia, S.; Collins, K. D.; Matsubara, H.;
Spain, M.; Procter, D. J. J. Am. Chem. Soc. 2009, 131, 7214.
(15) Collins, K. D.; Oliveira, J. M.; Guazzelli, G.; Sautier, B.; De
Grazia, S.; Matsubara, H.; Helliwell, M.; Procter, D. J. Chem.Eur. J.
2010, 16, 10240.
A real challenge left for the synthetic chemist is to find a lone
pair containing group which can be added and removed in a
way similar to a protecting group to channel the SmI2 reduction
to the appropriate reaction path.
(16) Parmar, D.; Matsubara, H.; Price, K.; Spain, M.; Procter, D. J. J.
Am. Chem. Soc. 2012, 134, 12751.
(17) Sautier, B.; Lyons, S. E.; Webb, M. R.; Procter, D. J. Org. Lett.
2012, 14, 146.
(18) Szostak, M.; Collins, K. D.; Fazakerley, N. J.; Spain, M.; Procter,
D. J. Org. Biomol. Chem. 2012, 10, 5820.
(19) Szostak, M.; Spain, M.; Choquette, K. A.; Flowers, R. A., II;
Procter, D. J. J. Am. Chem. Soc. 2013, 135, 15702.
ASSOCIATED CONTENT
* Supporting Information
Equations S1 and S2, Figures S1−S6, computed spectra,
Gaussian archives, complete ref 8, experimental section, and
NMR and mass spectra. This material is available free of charge
via the Internet at http://pubs.acs.org.
■
S
AUTHOR INFORMATION
Corresponding Author
■
*E-mail: shoz@biu.ac.il.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
We thank Prof. David Procter from the University of
Manchester, UK, for his very helpful comments.
■
REFERENCES
■
(1) For reviews on SmI2 over the past 10 years, see: (a) Szostak, M.;
Procter, D. J. Angew. Chem., Int. Ed. 2012, 51, 9238. (b) Sautier, B.;
Procter, D. J. Chimia 2012, 66, 399. (c) Procter, D. J.; Flowers, R. A.,
II; Skrydstrup, T. Organic Synthesis Using Samarium Diiodide: A
Practical Guide; The Royal Society of Chemistry: Cambridge, U.K.,
3879
dx.doi.org/10.1021/ol501490f | Org. Lett. 2014, 16, 3876−3879