Mechanistic studies of proton-donor coordination to samarium diiodide

Article abstract of DOI:10.1002/anie.200702571

Three's a crowd: Initial coordination of diethylene glycol to SmI ₂ liberates THF or iodide, thus providing open coordination sites for substrates and enhancing reactivity. Concentrations of diethylene glycol that lead to coordinative saturation of SmI₂ (see structure) reduce its reactivity. Replacement of a hydroxy proton with a methyl group decreases the affinity of the chelate for SmI₂, resulting in a decrease in the reactivity of the complex. (Figure Presented).

Full text of DOI:10.1002/anie.200702571

Communications
DOI: 10.1002/anie.200702571
Reaction Mechanisms
Mechanistic Studies of Proton-Donor Coordination to Samarium
Diiodide**
Joseph A. Teprovich, Jr., Marielle N. Balili, Tomislav Pintauer, and Robert A. Flowers, II*
In memoryof DmitryRudkevich
Alcohols, glycols, and glymes are commonly used as additives
in numerous reductions and reductive coupling reactions of
samarium diiodide (SmI2).[1,2] Typically alcohols and glycols
are used as proton donor sources, whereas glymes are used as
coordinating additives. All of these additives have been
shown to exert an influence on the regiochemical[3,4] and
stereochemical[5,6] outcomes of numerous SmI2-mediated
reactions. The seminal work of Hoz et al. demonstrated that
the interplay between proton-donor coordination and acidity
is likely to play an important mechanistic role in reductions.[7]
Studies by DahlØn and Hilmersson have shown that coordi-
nating alcohols enhance the rate of ketone reduction sub-
stantially and that the rate increase is proportional to the
number of ethereal oxygen atoms in the proton donor
source.[8] More recent studies in our group have shown that
the high affinity of water for SmI2 in THF leads to substrate
reduction through a unique mechanistic pathway involving a
SmII–water complex.[9] Taken together, these studies show
that proton-donor coordination enhances the reactivity of
SmI2. Unfortunately, the paucity of well-characterized sys-
tems for which the structures of the ground-state reductants
are known limits our understanding of the mechanistic
rationale for this phenomenon. Presented herein are crystal-
lographic, spectroscopic, and kinetic studies on substrate
reduction by SmI2 in solutions containing diethylene glycol
(dg), diethylene glycol monomethyl ether (dgme), and
diglyme (dgde). These studies show1) that initial coordina-
tion of the additives liberates THF or iodide from SmI2, thus
providing open coordination sites for substrate, 2) that
saturation of SmI2 with the additive decreases its reactivity,
and 3) that replacement of a hydroxyl proton with a methyl
group significantly decreases the affinity of the additive for
SmI2.
In searching for appropriate systems to explore, we sought
proton donors and related systems that would have a high
affinity for SmII and simultaneously provide stable complexes
that could be characterized. Many complexes of SmI2 with tri-
and tetraglyme provide complex aggregates which would
complicate mechanistic analysis.[10] We were drawn to the
structure of the [Sm(dgde)3]2+ cation.[11] This species is unique
among SmII–glyme structures because it is monomeric, thus
indicating that an opportunity existed for isolation and
characterization of complexes of dg, dgme, and dgde with
SmI2. Furthermore, this system offers an opportunity to
examine the impact of subtle changes to the system through
substitution of the proton of the hydroxyl group with methyl
substituents.
To initially examine the system, the UV/Vis spectrum of
SmI2 (1 mmol in THF) was monitored in the presence of
increasing amounts of dg, dgme, and dgde. A subset of these
experiments is displayed in Figure 1. Upon addition of two
[*] J. A. Teprovich, Jr., Prof. R. A. Flowers, II
Department of Chemistry
Figure 1. UV/Vis spectra of SmI2 (1 mmol) in THF(a) with increasing
amounts of dg (b=2, c=6, and d=8 equivalents). Inset shows
UV/Vis spectra of SmI2 (1 mmol) in THF(e) with increasing amounts
of dgme (f=30, g=80, h=500 equivalents).
Lehigh University
Bethlehem, PA 18015 (USA)
Fax: (+1)610-758-6536
E-mail: rof2@lehigh.edu
Homepage: http://www.lehigh.edu/~rof2/
M. N. Balili, Prof. T. Pintauer
Department of Chemistry & Biochemistry
Duquesne University
equivalents of dg, the long-wavelength absorption peaks of
SmI2 (lmax = 555 and 618 nm) begin to merge, indicating
coordination. Upon further addition of dg, the long-wave-
length absorption peak decreases in intensity, and a new
absorption appears at 476 nm. During this process the
solution turned from blue to red, consistent with the spectral
changes shown in Figure 1. Addition of more than eight
equivalents of dg produces no further changes in the UV/Vis
Pittsburgh, PA 15282 (USA)
[**] R.A.F. thanks the National Science Foundation (CHE-0413845) for
support of this work. T.P. acknowledges a Duquesne University
Start-up Grant.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
8160
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 8160 –8163

Angewandte
Chemie
À
spectrum. The isosbestic point at 522 nm is consistent with the
conversion of the SmI2–thf complex to a SmII–dg complex.
The inset of Figure 1 shows addition of successive portions of
dgme. Similar behavior was observed, but much higher
concentrations of dgme were required to produce a SmII–
dgme complex. Addition of dgde showed no perceptible
changes to the UV/Vis spectrum of SmI2 even at high
concentrations (500 equivalents, see the Supporting
Information). These experiments showthat the affinity of
the additives for SmI2 in THF is in the order dg > dgme >
dgde and indicate that replacement of a hydroxyl proton on
dg with a methyl substituent significantly alters the affinity of
the additive for SmII.
ligand, however, all three Sm O bond lengths are very
À
similar. In [Sm(dgme)3]I2, d(Sm O) = 2.585(6), 2.679(5), and
2.653(6) Š for the first ligand; 2.638(5), 2.702(5), and
2.610(5) Š for the second ligand; and 2.554(6), 2.711(5), and
2.717(5) Š for the third ligand. For two chelating ligands,
elongation of the bond to the internal O atom is observed,
À
and in the third ligand, two Sm O bond lengths are quite
long. Furthermore, in each dgme ligand, the hydroxyl oxygen
À
atom has a shorter Sm O bond than the methoxy oxygen
atom. Analysis of the [SmI2(dgde)2] structure shows that the
two dgde ligands have the following Sm O bond distances:
for the first ligand d(Sm O) = 2.656(3), 2.688(3), and
2.696(3) Š; for the second ligand d(Sm O) = 2.650(3),
2.642(3), and 2.702(3) Š. In each ligand, one terminal Sm
À
À
À
À
We next set out to crystallize the series of complexes to
gain an understanding of the structures formed in solution.
Addition of an eight-equivalent excess of dg to 0.1m SmI2 in
THF under an argon atmosphere provided ruby-red crystals
of [Sm(dg)3]I2 (Figure 2) upon standing overnight. Inspection
À
OCH3 bond is elongated. The Sm I bond lengths of 3.3081(4)
and 3.3185(4) Š are consistent with other eight-coordinate
SmII complexes, and all data are similar to those previously
reported for this structure.[12]
The crystallographic data showthat in both [Sm(dg) ]I2
3
and [Sm(dgme)3]I2, the first two ligands have shorter contacts
than the third ligand, suggesting that the latter has a lower
affinity. Replacement of a hydroxyl hydrogen atom with a
À
methyl group leads to a longer Sm O bond, and the
replacement of both hydroxyl hydrogen atoms with methyl
groups decreases the affinity to such an extent that only two
ligands coordinate to SmII, leaving the iodides bound to the
inner sphere. The crystallographic data is consistent with the
UV/Vis spectrum and indicates that the emergence of the
absorption at 476 nm is due to formation of the complexes
[Sm(dg)3]I2 and [Sm(dgme)3]I2.
Because coordinating additives are known to alter the
reducing power of SmII, cyclic voltammetry was employed to
characterize the impact of addition of dg, dgme, and dgde to
SmI2 in THF. In the absence of additive, a quasi-reversible
voltammogram was obtained with an estimated E1/2 value of
À1.57 Æ 0.05 V versus a saturated Ag/AgNO3 electrode.[13]
Addition of a large excess of dgme or dgde had no measurable
impact on the potential compared to SmI2 alone. Addition of
dg led to moderate changes in the redox properties. Samples
Figure 2. Molecular structure of [Sm(dg)3]I2 shown with 20% proba-
bility displacement ellipsoids. H atoms have been omitted for clarity.
Samarium–oxygen bond lengths [Š]: Sm1–O6 2.587(11), Sm1–O1
2.590(10), Sm1–O4 2.603(10), Sm1–O9 2.618(11), Sm1–O3 2.624(11),
Sm1–O7 2.635(11), Sm1–O5 2.657(10), Sm1–O8 2.666(10), Sm1–O2
2.728(10).
containing four equivalents of dg provided an estimated E1/2
=
À1.63 Æ 0.05 V, and samples containing ten equivalents of dg
gave an estimated E1/2 = À1.70 Æ 0.05 V. Further addition of
dg had no impact on the redox potential. While addition of dg
to SmI2 produces a more powerful reductant in comparison to
dgme and dgde, its effect is modest in comparison to other
additives.[14]
To further study the mechanistic impact of coordination
on the reactivity of SmII, the rates of reactions of SmI2 with
benzyl bromide were monitored in the presence of increasing
amounts of dg, dgme, and dgde to determine the relationship
between the affinity of each ligand for SmII and the impact on
the reactivity of the resulting complex. Benzyl bromide was
chosen as a model substrate to simplify the kinetic analy-
sis.[9b,15] Initial experiments were designed to determine the
rate orders of additive, SmI2, and benzyl bromide. All rate
studies were carried out at 258C under pseudo-first-order
conditions with the concentration of benzyl bromide kept
high relative to that of SmI2 using a stopped-flowspectro-
photometer. Analysis of reaction products showed that the
of the structure clearly shows that the iodide ions are
displaced to the outer sphere. The UV/Vis spectrum of the
crystals in THF was identical to that of a solution of SmI2
containing three equivalents of dg (Figure 1). Isolation of
crystals of [Sm(dgme)3]I2 (see the Supporting Information)
was accomplished through a similar procedure but required
100 equivalents of dgme to produce crystals. Finally, blue
crystals of cis-[SmI2(dgde)2] were obtained by addition of a
100-equivalent excess of dgde to a 0.1m solution of SmI2 in
THF (see the Supporting Information).
À
Comparison of the Sm O bond lengths for the coordi-
nated ligands in each structure is informative. In [Sm(dg)3]I2,
À
d(Sm O) = 2.590(10), 2.728(10), and 2.624(11) Š for the first
ligand; 2.587(11), 2.657(10), and 2.603(10) Š for the second
ligand; and 2.635(11), 2.666(10), and 2.618(11) Š for the third
ligand. From this analysis, it is apparent that in the case of two
À
coordinating ligands, the central Sm O bond is elongated
with respect to the terminal oxygen atoms. In the third dg
Angew. Chem. Int. Ed. 2007, 46, 8160 –8163ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
8161

Communications
presence of dg and dgme provided toluene exclusively as
Further experiments using deuterated dg provided a deute-
rium isotope effect of 0.92 Æ 0.04, thus demonstrating that
proton transfer does not occur in the rate-limiting step of the
reduction. The rate orders of benzyl bromide and SmI2 were
approximately first order over the entire concentration range
of additives examined in these studies.
product, whereas the presence of dgde gave bibenzyl as
product. To determine the rate order of dg, dgme, and dgde, a
series of kinetic experiments were initiated in which the rate
of reduction of benzyl bromide was monitored over a range of
additive concentrations as summarized in Figure 3 and
Taken together, these studies provide the following
experimental observations: 1) The affinity of the additives
for SmI2 in THF is in the order dg > dgme > dgde. 2) SmII can
accommodate the coordination of two diglyme and two iodide
ligands. 3) The impact of additive on the rate of benzyl
bromide reduction is proportional to the affinity of the
additive for SmI2. 4) Additives capable of displacing iodide
from SmI2 increase the rate of substrate reduction. 5) The
highest rate of substrate reduction by SmI2 is achieved at an
additive concentration lower than that required for formation
of [Sm(dg)3]I2 or [Sm(dgme)3]I2 in solution. Concentrations of
dg or dgme that drive the equilibrium to the coordinatively
saturated complexes decrease the rate of substrate reduction.
6) Only dg provided a modest increase in the reducing power
of SmI2. On the basis of these observations, we propose that
the data support the equilibrium process shown in Scheme 1.
Upon coordination of one equivalent of dg or dgme, solvent is
displaced from the inner sphere of SmI2. Addition of a second
equivalent of additive displaces remaining solvent, and
further addition displaces the iodide ligands.
~
*
Figure 3. Plot of kobs vs. [dg] (&), [dgme] ( ), and [dgde] ( ) for the
reduction of benzyl bromide (100 mm) by SmI2 (2 mm) at 258C.
Table 1. In the case of dg, the reaction is approximately
first-order in additive at concentrations up to 6 mm (3 equiv-
alents), whereas at higher concentrations, the reaction is
Table 1: Summary of rate studies for the reduction of benzyl bromide
(BnBr) by SmI2 in THFcontaining dg, dgme, or dgde.
Rate order
[dg]
dg[a]
SmI2
BnBr[c]
[b]
Scheme 1. Possible equilibrium for the coordination of dg and dgme
to SmI2 in THF.
2–6 mm
6–60 mm
1.1Æ0.1
0.96Æ0.05
0.9Æ0.1
À0.87Æ0.03
[dgme]
dgme[a]
SmI2
BnBr[c]
[b]
The enhanced reactivity and positive reaction order for dg
or dgme belowconcentrations necessary for coordinative
saturation of samarium are consistent with displacement of
iodide (or solvent) by the additive to open up coordination
sites for substrate in the activated complex.[16] Higher
concentrations of dg or dgme inhibit interaction of substrate
with samarium, and additive displacement is necessary for
reduction. Although dg does increase the reducing power of
SmI2, it does so at concentrations that saturate the coordina-
tion sphere of the Sm center. As a consequence, the driving
force for increased reactivity of SmI2 upon addition of dg (or
other coordinating proton donors) is due predominantly to
solvent or iodide displacement. With this information in hand,
we are currently examining reactions in which proton transfer
to substrate plays an important mechanistic role to ascertain
the importance of proton transfer from Sm–donor complexes
versus bulk solution.
2–30 mm
30–240 mm
0.29Æ0.02
1.03Æ0.05
0.99Æ0.03
À0.46Æ0.03
[b]
[dgde]
50–800 mm
dgde[a]
0
SmI2
BnBr[c]
0.93Æ0.07
1.0Æ0.1
[a] The concentration of SmI2 in all experiments is 2 mm and the
concentration of BnBr is 100 mm. [b] Rate order was extracted using the
fractional rates method. [c] Rate order determined from concentrations
of BnBr ranging from 25–100 mm.
nearly inverse first-order. These findings are consistent with
the involvement of one equivalent of dg in the reduction of
benzyl bromide at lower concentrations of additive. The
inverse first-order behavior of dg at higher concentrations is
consistent with dg displacement in the transition state of the
reduction.
Examination of dgme reveals similar but less pronounced
behavior. At dgme concentrations up to 60 mm (30 equiv-
alents), the rate order is 0.29 Æ 0.02. At concentrations above
60 mm, the rate order of dgme is À0.46 Æ 0.03. Addition of
dgde over a large concentration range had no measurable
impact on the rate of reduction of benzyl bromide by SmI2.
Experimental Section
THF was dried using a solvent purification system produced by
Innovative Technology, Inc. (Newburyport, MA). SmI2 was prepared
8162
www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 8160 –8163

Angewandte
Chemie
by stirring Sm metal and iodine in THF until the characteristic color
of SmII appeared. The concentration of SmI2 was determined by
iodometric titration or by the method developed by Hilmersson.[17]
Benzyl bromide, dg, dgme, and dgde were dried over activated
molecular sieves. Dried solvents and reagents were stored in an
Innovative Technology, Inc. drybox containing an argon atmosphere
and a platinum catalyst for drying.
UV/Vis experiments were performed on a Shimadzu UV-1601
spectrophotometer controlled by UV Probe software (version 1.11).
All rates were determined using an SX-18 MV stopped-flow
spectrophotometer (Applied Photophysics Ltd., Surrey, UK) at
258C. Cyclic voltammograms for SmI2 and the SmI2–additive com-
plexes were measured as described in reference [9b], except the
concentration of SmI2 in each experiment was 15 mm.
Steel, J. Chem. Soc. Perkin Trans. 1 2001, 2727 – 2751; d) D. J.
Edmonds, D. Johnston, D. J. Procter, Chem. Rev. 2004, 104,
3371 – 3404; e) A. DahlØn, G. Hilmersson, Eur. J. Inorg. Chem.
2004, 3393 – 3404.
[2] R. A. Flowers II, E. Prasad in Handbook on the Physics and
Chemistryof Rare Earths , Vol. 36 (Eds.: K. A. Gschneidner, Jr.,
J.-C. G. Bünzli, V. K. Pecharsky), Elsevier, 2006, chap. 230,
pp. 393 – 473.
[3] A. Yoshida, T. Manamoto, J. Inanaga, K. Mikani, Tetrahedron
Lett. 1998, 39, 1777 – 1780.
[4] T. K. Hutton, K. W. Muir, D. J. Procter, Org. Lett. 2003, 5, 4811 –
4814.
[5] D. J. Edmonds, K. W. Muir, D. J. Procter, J. Org. Chem. 2003, 68,
3190 – 3198.
Crystals of [Sm(dg)3]I2 were prepared by diluting dg (76 mL) in
THF (5 mL) and adding SmI2 (1.0 mL, 0.10m in THF) to the flask
with subsequent final dilution to 10 mL. The solution was shaken
thoroughly and allowed to stand open in the drybox to permit slow
evaporation. Red crystals began to appear after a fewhours, and the
solution was allowed to stand overnight. The bulk solution was
discarded, and the crystals were collected and dried under slight
vacuum. Crystals of [Sm(dgme)3]I2 and [SmI2(dgde)2] were prepared
using analogous procedures employing dgme (1.2 mL) or dgde
(1.4 mL). Crystals of [Sm(dgme)3]I2 were ruby-red, whereas crystals
of [SmI2(dgde)2] were blue.
[6] a) H. Lodberg Pedersen, T. B. Christensen, R. J. Enemærke, K.
Daasbjerg, T. Skrydstrup, Eur. J. Org. Chem. 1999, 565 – 572;
b) T. B. Christensen, D. Riber, K. Daasbjerg, T. Skrydstrup,
Chem. Commun. 1999, 2051 – 2052; c) H. C. Aspinall, N.
Greeves, C. Valla, Org. Lett. 2005, 7, 1919 – 1922.
[7] S. Hoz, A. Yacovan, I. Bilkis, J. Am. Chem. Soc. 1996, 118, 261 –
262.
[8] A. DahlØn, G. Hilmersson, Tetrahedron Lett. 2001, 42, 5565 –
5569.
[9] a) P. R. Chopade, E. Prasad, R. A. Flowers II, J. Am. Chem. Soc.
2004, 126, 44 – 45; b) E. Prasad, R. A. Flowers II, J. Am. Chem.
Soc. 2005, 127, 18093 – 18099.
[10] M. Vestergren, B. Gustafsson, A. Johansson, M. Hꢀkansson, J.
Organomet. Chem. 2004, 689, 1723 – 1733.
[11] J. P. White III, H. Deng, E. P. Boyd, J. Gallucci, S. G. Shore,
Inorg. Chem. 1994, 33, 1685 – 1694.
[12] A. Sen, V. Chebolu, A. L. Rheingold, Inorg. Chem. 1987, 26,
1821 – 1823.
[13] This value is within experimental error of previous reports.
a) R. J. Enemærke, K. Daasbjerg, T. Skrydstrup, Chem.
Commun. 1999, 343 – 344; b) M. Shabangi, M. L. Kuhlman,
R. A. Flowers II, Org. Lett. 1999, 1, 2133 – 2135.
[14] a) M. Shabangi, R. A. Flowers II, Tetrahedron Lett. 1997, 38,
1137 – 1140; b) M. Shabangi, J. M. Sealy, J. R. Fuchs, R. A.
Flowers II, Tetrahedron Lett. 1998, 39, 4429 – 4432; c) R. J.
Enemærke, T. Hertz, T. Skrydstrup, K. Daasbjerg, Chem. Eur.
J. 2000, 6, 3747 – 3754.
Crystal structure analysis data for [Sm(dg)3]I2: C12H30I2O9Sm,
Mr = 722.51, monoclinic, space group P2(1)/c, a = 16.121(4), b =
14.581(4), c = 10.438(3) Š, b = 106.911(4)8, V= 2347.7(10) Š3, Z = 4,
1calcd = 2.044 MgmÀ3, m = 5.164 mmÀ1, F(000) = 1368, 12252 reflec-
tions collected, 2478 unique, GOF = 0.825, R1 = 0.0525, wR2 = 0.1281.
Crystal structure analysis data for [Sm(dgme)3]I2: C15H36I2O9Sm,
Mr = 764.90, orthorhombic, space group Pna2(1), a = 17.2297(12), b =
9.9738(7), c = 15.2855(10) Š, V= 2626.7(3) Š3, Z = 4, 1calcd
=
1.933 MgmÀ3, m = 4.621 mmÀ1, F(000) = 1464, 31086 reflections col-
lected, 8200 unique, GOF = 0.988, R1 = 0.0469, wR2 = 0.1120.
CCDC-649549 ([Sm(dg)3]I2) and CCDC-649550 ([Sm(dgme)3]I2)
contain the supplementary crystallographic data for this paper. These
data can be obtained free of charge from The Cambridge Crystallo-
graphic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
Received: June 13, 2007
[15] a) C. P. Andrieux, I. Gallardo, J.-M. Saveant, K.-B. Su, J. Am.
Chem. Soc. 1986, 108, 638 – 647; b) J.-M. Saveant, J. Am. Chem.
Soc. 1987, 109, 6788 – 6795.
[16] E. Prasad, B. W. Knettle, R. A. Flowers II, J. Am. Chem. Soc.
2004, 126, 6891 – 6894.
Published online: September 7, 2007
Keywords: chelates · proton donors · reaction mechanisms ·
.
samarium
[17] A. DahlØn, G. Hilmersson, Eur. J. Inorg. Chem. 2004, 3020 –
3024.
[1] a) G. A. Molander, Chem. Rev. 1992, 92, 29 – 68; b) G. A.
Molander, C. R. Harris, Chem. Rev. 1996, 96, 307 – 338; c) P. G.
Angew. Chem. Int. Ed. 2007, 46, 8160 –8163ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
8163