SmI2/H2O/amine promoted reductive cleavage of benzyl-heteroatom bonds: optimization and mechanism

Article abstract of DOI:10.1016/j.tet.2009.10.086

The SmI₂/H₂O/pyrrolidine mediated cleavage of benzylic alcohols and benzyl groups was studied and found to be a viable alternative to the Birch reduction yielding the corresponding deoxygenated product in excellent yield. The reaction has been investigated by kinetic methods, and a mechanism involving a pre-complexation of the alcohol to SmI₂ followed by an amine mediated electron transfer and subsequent bond cleavage and transfer of a second electron and proton to yield the toluene product has been proposed. The reaction is strongly inhibited at higher concentrations of water, indicating that it proceeds via an inner-sphere electron transfer from samarium(II) to the benzyl group, and excess of water prevents coordination of benzyl alcohol to samarium.

Full text of DOI:10.1016/j.tet.2009.10.086

Tetrahedron 65 (2009) 10856–10862
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
SmI2/H2O/amine promoted reductive cleavage of benzyl-heteroatom
bonds: optimization and mechanism
*
Tobias Ankner, Go¨ran Hilmersson
Department of Chemistry, University of Gothenburg, SE-412 96 Go¨teborg, Sweden
a r t i c l e i n f o
a b s t r a c t
Article history:
The SmI2/H2O/pyrrolidine mediated cleavage of benzylic alcohols and benzyl groups was studied and
found to be a viable alternative to the Birch reduction yielding the corresponding deoxygenated product
in excellent yield. The reaction has been investigated by kinetic methods, and a mechanism involving
a pre-complexation of the alcohol to SmI2 followed by an amine mediated electron transfer and
subsequent bond cleavage and transfer of a second electron and proton to yield the toluene product has
been proposed. The reaction is strongly inhibited at higher concentrations of water, indicating that it
proceeds via an inner-sphere electron transfer from samarium(II) to the benzyl group, and excess of
water prevents coordination of benzyl alcohol to samarium.
Received 4 September 2009
Received in revised form
19 October 2009
Accepted 26 October 2009
Available online 31 October 2009
Keywords:
Reduction
SmI2
Ó 2009 Elsevier Ltd. All rights reserved.
Birch
Benzyl alcohol
Benzyl amine
Benzyl thiol
1. Introduction
studying the underlying mechanism. Although it has been
demonstrated that SmI2 can mediate cleavage of activated benzyl
The Birch reduction, employing alkali metals in ammonia, is
a widely used reaction for a number of transformations in
organic synthesis.1 Although the reduction conditions are con-
sidered cumbersome and dangerous, it retains an important role
for a number of reduction processes including deoxygenation2
and deprotection3 of benzylic amines, alcohols, and thiols. Alkali
metals have a very high reduction potential that even enable
facile reduction of benzene, which results in low functional
group tolerance. In this respect it is desirable to find alternatives
that are more tolerant toward other functional groups.4 We have
discovered that the addition of an amine to a SmI2/H2O mixture
have a dramatic effect on the rate of reduction of ketones5 and
that it is capable of reducing conjugated double and triple bonds6
as well as polyaromatics7 (Fig. 1a–b). The SmI2/H2O/amine
mixture also mediates deallylation8 and detosylation9 (Fig. 1c–d).
In addition, we have found that benzylic alcohols and amines are
reduced yielding toluene. The above mentioned reactions all
share properties with the Birch reaction, and this encouraged us
to explore the scope and limitation of this reaction as well as
groups,10 there are no reports on the reduction of benzylic
alcohols and amines.
Figure 1. a) Reduction of conjugated olefins, b) Reduction of polyaromatics, c) Efficient
cleavage of allylprotected carbohydrates, d) Detosylation of primary amines.
Herein we now wish to report on the SmI2/H2O/amine mediated
reduction of the benzyl heteroatom bond along with mechanistic
studies of the reduction of benzyl alcohol.
* Corresponding author.
E-mail address: hilmers@chem.gu.se (G. Hilmersson).
0040-4020/$ – see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2009.10.086

T. Ankner, G. Hilmersson / Tetrahedron 65 (2009) 10856–10862
10857
2. Results and discussion
revealed that 2,3-diphenylbutane was formed via a reductive
coupling reaction in a 60% yield (Fig. 3).
2.1. Cleavage of the benzyl-heteroatom bond
2.2. Electronic and steric effects in the reduction of various
benzylic alcohols
Preliminary studies in our lab have revealed that benzyl alcohol
is reduced to toluene upon exposure to SmI2/H2O/amine in THF at
room temperature. Good conversion (85%) was achieved in 2 h at
room temperature (Table 1, entry 1), and prolonged reaction time
(24 h) resulted in marginal improvements (Table 1, entry 2). Higher
yields could be achieved when pyrrolidine was used instead of
triethylamine (Table 1, entry 3). Considerably lower yields were
The kinetic studies revealed that the relative rate was not only
dependent on the steric hindrance, but the electronic properties of
the benzylic moiety also affected the rate to a very high degree.
Electron-withdrawing groups on the aromatic ring enhanced the
rate considerably (Table 3). A Hammett plot (log k vs sꢀx ) reveals
a correlation with a slope of 4.0 (Fig. 4). This result should be taken
with some caution due to the relatively low number of substrates
(Table 3, entries 1–5). In addition, the substrates with p-F and p-H,
which has approximately the same sx, proceeds with different
rates.
obtained with an a-substituted substrate, 1-phenyl ethanol (entries
4–7). The use of more reactive amines, e.g., isopropyl amine or
pyrrolidine, resulted in a faster reaction, however the breakdown of
the reagent was also accelerated11 explaining why there was only
a modest increase in chemical yield (entries 5 and 6). Fortunately,
this unwanted reaction could be suppressed when the reaction was
carried out at ꢀ20 ꢁC (entry 7). This is in accordance with our
previously reported results where reductions of the nitro group to
the amine with SmI2/H2O/amine proceeds in much better yields
when carried out at ꢀ20 ꢁC.12
´
*
The use of Crearys
or any linear combination of the three Hammett values (
) did not result in any improvement of the correlation. We also
s
(often yielding better results for radicals)
s
,
sꢀ and
s
*
conclude that both p-F and p-OMe yields surprisingly low rates.
It should be noted that these substituents may destabilize a radical
anion intermediate via resonance thus lowering the rate.
Table 1
The introduction of a-substituents was also investigated (Table
Optimization study for the reduction of benzyl alcohols
3, entries 6–8). The rate of reduction was lowered by a factor of two
upon the introduction of a methyl group. A second methyl group
almost stopped the reduction and the rate was almost two orders of
magnitude lower (Table 3, entry 7). Surprisingly, benzhydrol (Table
3, entry 8) reacts very slowly; it is almost six times less reactive
than benzyl alcohol. These latter results indicate that steric
hindrance affects the rate of reduction significantly.
Both the benzyl amines and the thiol undergo clean reduction
by SmI2/H2O/amine (Table 2, entries 8 and 9). Thus the rates of
reduction in these cases were investigated in a similar fashion
(Table 4). Benzyl amine (entry 2) is eight times less reactive than
benzyl alcohol itself, and the highest rate of reduction is obtained
with N-methyl substitution (entry 3). However, with the
introduction of a second methyl group on the nitrogen (entry 4),
the relative rate dropped slightly. Again, further substitution at the
Entry Substrate
Time (h) Temp (ꢁC) Amine
Yielda (%)
1
2
3
4
5
6
7
Benzyl alcohol
Benzyl alcohol
Benzyl alcohol
1-Phenyl ethanol 20
1-Phenyl ethanol 24
1-Phenyl ethanol 24
1-Phenyl ethanol 24
2
24
12
RT
RT
RT
RT
RT
RT
–20
Et3N
Et3N
Pyrrolidine
Et3N
85
90
95
28
Isopropylamine 37
Pyrrolidine
Pyrrolidine
37
62
a
Yields were determined by GC/MS analysis using dodecane as internal standard
and authentic samples as references.
Several alkyl and aryl substituted benzyl alcohols were
efficiently reduced by SmI2/H2O/amine in high yields (Table 2).
From these results we conclude that both the steric bulk around the
benzylic carbon (Table 2, entries 2–4) as well as the electronic
properties (Table 2, entries 5 and 6) of the substrate is important for
the success of the reaction.
Amazingly the introduction of one or two trifluoromethyl
groups (entries 5 and 6) facilitates the reductive cleavage consid-
erably, yielding the respective toluene derivatives seemingly
instantaneously using the less reactive triethylamine in the
reduction mixture. In contrast, the reaction almost halted upon
substitution on the phenyl ring with a 4-methoxy group, resulting
in low conversion (38%) after 48 h. The major products observed
were cyclohexadiene derivatives, which are characteristic of the
Birch reduction (Fig. 2).
In addition, benzyl substituted primary or secondary amines
and thiols also undergo smooth reduction under these conditions
(entries 8–10). Benzyl amines were efficiently reduced using SmI2/
H2O/pyrrolidine yielding toluene and the corresponding debenzy-
lated amines in 76 and 95% yields after 24 h (entries 9 and 10).
It is well known from the literature that derivatization of ben-
zylic alcohols with acetate greatly facilitates the deoxygenation.13
Thus we prepared a selection of 1-phenyl ethanol esters and
exposed them to the reagent mixture (Table 2, entries 11–14). The
reaction was instantaneous at RT and afforded the deoxygenated
product in excellent to nearly quantitative yields. The tri-
fluoroacetate derivatized alcohol (entry 14) gave a lower conver-
sion to ethyl benzene. Closer inspection of the product distribution
a-position with a methyl group lowers the rate approximately two
times (entry 5). There is no clear trend among the benzyl amines
investigated, but the difference in reactivity could possibly be due
to the presence of the electron-donating methyl groups that is
somewhat overshadowed by an increase in steric hindrance.
Converting the alcohol to an methyl ether (entry 6) also lowers
the rate considerably and it is 40 times less reactive. Benzyl thiol
(entry 7) is approximately 1.5 times more reactive than benzyl
alcohol supporting the fact that the yield for this substrate is
excellent.
2.3. Determination of the rate orders for the reduction of
benzyl alcohol
The method of initial rates can be used for precise and conve-
nient determination of the reaction orders in a complex system, as
long as the reaction follows the same kinetics throughout the
course of the reaction.14
By individually varying the component concentrations over at
least an eightfold range the reaction order in each component SmI2,
amine, water, and benzyl alcohol was obtained from the slope of
the plots of log[initial rate] vs log[component]. The rate order in
SmI2 was determined by measuring the rate constant at different
concentrations of SmI2 from 12 to 955 mM, (Fig. 5). In contrast to
our previous findings in the reduction of alkyl halides with SmI2/
H2O/amine where the reaction order for SmI2 was found to be two,
benzyl alcohol reduction yields a rate order of one. This clearly
indicates that a different mechanism is operating.

10858
T. Ankner, G. Hilmersson / Tetrahedron 65 (2009) 10856–10862
Table 2
SmI2/H2O/R3N mediated cleavage of benzyl heteroatom bondsa
Entry
1
Alcohol
Time
Temp (ꢁC)
Amine
Product
Yieldb (%)
95
22 h
ꢀ20
Pyrrolidine
2
3
4
5
24 h
ꢀ20
ꢀ20
ꢀ20
20
Pyrrolidine
Pyrrolidine
Pyrrolidine
Triethylamine
62
85c
80
5 h
22 h
>1 min
99
6
7
>1 min
20
Triethylamine
Pyrrolidine
99
88
20 h
ꢀ20
8
20 h
20 h
20 h
ꢀ20
ꢀ20
ꢀ20
Pyrrolidine
Pyrrolidine
Pyrrolidine
95
95
76
9
10
11
12
13
>1 min
>1 min
>1 min
20
20
20
Pyrrolidine
Pyrrolidine
Pyrrolidine
73
84
91
14
>1 min
20
Triethylamine
31
a
The benzylic substrate (0.04 mmol) was added to a mixture of amine (0.4 mmol) and water (0.6 mmol) in a 0.1 M solution of SmI2 (0.2 mmol) in THF at ꢀ20 ꢁC. See
experimental section for more details.
b
Yields determined with GC/MS analysis, except entry 7, which was determined with HPLC. Yields were calculated using internal standard and authentic samples as
reference.
c
Isolated yield.

T. Ankner, G. Hilmersson / Tetrahedron 65 (2009) 10856–10862
10859
3
2
1
0
y = 4.04x - 0.74
R2 = 0.90
-1
Figure 2. Product distribution in the reduction of anisyl alcohol.
-2
-3
-0.3
-0.1
0.1
0.3
σx
0.5
0.7
0.9
-
Figure 4. Hammett plot (log k vs sꢀx ) for the SmI2/amine/water mediated reduction of
benzyl alcohol in THF.
Figure 3. Reductive coupling of benzylic trifluoroacetate.
Table 4
Relative rates in the reduction of various benzylic heteroatoms, determined by the
initial rate method
Table 3
Relative rates in the reduction of benzyl alcohol derivatives to the respective tolu-
enes determined by the initial rate method
Entry
Substrate
krel
Entry
Alcohol
krel
1
1
1
1
2
3
4
0.125
0.25
0.18
2
3
4
0.012
0.012
93
5
0.05
6
7
0.025
1.5
5
>500
6
7
0.5
-2
-1.5
-1
log [SmI2]
-2
0.012
0.18
-2.5
-3
y = 1.09x - 2.02
R2 = 0.92
8
-3.5
-4
Similarly the rate order in benzyl alcohol was determined to be
unity at concentrations below 19 mM, while at higher concentra-
tions it gradually increased toward 0.5 (Fig. 6). There can be several
explanations for this behavior. It can indicate that benzyl alcohol is
a monomer at lower concentrations and that it dimerizes at higher
concentrations in THF at room temperature. Another explanation
can be that at higher concentrations the benzyl alcohol begins to
effectively compete with water at the metal center thus altering the
mechanism.
The rate order in triethylamine was found to be unity between
in the range 18 and 388 mM (Figs. 7 and 8).
To our surprise the rate dependence for water in the reaction
was found to be non-linear. At lower concentrations (24 mM to
-4.5
Figure 5. Initial log(rate) vs log (concentration) of SmI2 (12–955 mM). Reaction con-
ditions; 200 mM Et3N, 300 mM H2O, 14 mM benzyl alcohol studied at 20 ꢁC in THF.
approx 300 mM) the rate appears to increase linearly with a slope
corresponding to rate order 1. It then reaches a plateau (300 mM),
and at yet higher concentrations the rate dropped dramatically,
with a slope of ꢀ4 in the log–log plot.
Although there is no reaction at this point the solution remains
dark, indicating that SmI2 is still intact. Notably, in control experi-
ments the reaction mixture containing large excess of H2O is still

10860
T. Ankner, G. Hilmersson / Tetrahedron 65 (2009) 10856–10862
0
0.5
1
1.5
2
2.5
3
the rate order is unity in each component,16 and with the
assumption that the concentration of each component is approxi-
mately equal to their initial concentration (pseudo-first order rate
condition). An Eyring plot was obtained by plotting ln(kobsh/kBT) vs
1/T (Figs. 9–11).
-1
log [benzyl alcohol]
-1.5
-2
-2.5
-3
y = 0.50x - 3.21
0.003
-31
0.0032
0.0034
0.0036
0.0038
1/T
R
2 = 0.97
-3.5
-4
y = 1.04x - 3.90
2 = 0.97
R
-32
-33
Figure 6. Initial rate vs concentration of benzyl alcohol (2.4–654 mM). Reaction con-
ditions; 100 mM SmI2, 200 mM H2O, 300 mM H2O studied at 20 ꢁC in THF.
y = -2269x - 24.4
R2 = 0.97
1
1.5
2
2.5
3
-2
-2.5
-3
log [amine]
Figure 9. The Eyring plot (ln[kobsh/kBT] vs 1/T, T¼273–323) for the reduction of benzyl
y = 0.99x - 4.99
R2 = 0.98
alcohol in THF.
1.2
y = 0.0088x + 0.26
R
2 = 0.95
1
0.8
0.6
0.4
0.2
0
-3.5
-4
Figure 7. Initial rate vs concentration of amine (18–388 mM). Reaction conditions;
100 mM SmI2, 300 mM H2O, 14 mM benzyl alcohol studied at 20 ꢁC in THF.
y = 0.0045x + 0.15
2 = 0.95
R
1.5
-1.5
2
2.5
3
3.5
0
20
40
60
80
100
log [H2O]
-2
-2.5
-3
time (s)
Figure 10. Initial rates of the reduction of benzyl alcohol with SmI2 with H2O and D2O,
the kinetic isotope effect was calculated to be 1.9. Reaction conditions; 100 mM SmI2,
200 mM Et3N, 130 mM L2O and 14 mM benzyl alcohol at 20 ꢁC in THF.
-3.5
-4
1.4
1.2
-4.5
0 M BnsOH
1 M BnsOH
Figure 8. Initial log (rate) vs log (concentration) of water (67–1316 mM). Reaction
conditions; 100 mM SmI2, 200 mM Et3N, 14 mM benzyl alcohol studied at 20 ꢁC in THF.
1.0
4 M BnsOH
8 M BnsOH
0.8
0.6
a
powerful reductant that instantaneously reduces ketones
quantitatively to alcohols. Such dependence indicates that the
coordination of the substrate, i.e., benzyl alcohol to the metal center
is of paramount importance for the electron transfer step of the
reaction, which would be inhibited at higher water concentrations
since water is a better ligand for SmI2 than benzyl alcohol. The
reductive cleavage of the benzyl group thus appears to be an inner-
sphere electron transfer process and if the water concentration is
high it will occupy all available coordinating sites around samarium.
Ketones on the other hand are much stronger Lewis bases and they
might compete effectively with water at the coordination sites on
samarium. Flowers has previously observed that large excess of the
very powerful Lewis base HMPA may indeed inhibit reduction
reactions with SmI2.15 Alternatively, it is also possible that reduction
of a ketone, which can proceed with a much weaker reducing agent,
may proceed via an alternative outer-sphere electron transfer.
0.4
0.2
0.0
400
450
500
550
600
650
700
750
800
nm
Figure 11. Absorption spectra of SmI2 (2 mM) in the presence of increasing amount of
benzyl alcohol (0, 0.5, 1, 2, 4, 8 M). Several lines are omitted for clarity.
Activation parameters for the reduction reaction were obtained
from the linear correlation according to the Eyring equation
[ln(kobsh/kBT)¼ꢀ
D
Hz/RTþ
D
Sz/R],
i.e.
D
Hz¼18.9ꢂ1.0 kJ molꢀ1
,
D
Sz¼ꢀ260ꢂ30 J Kꢀ1 molꢀ1 and
D
Gz
¼97ꢂ10 kJ molꢀ1
.
298K
The large negative entropy of activation is obviously a result of
the observed high order in the reaction components.
2.4. Activation parameters
2.5. The amine influences the rate of reaction
The temperature dependence of the reduction of benzyl alcohol
with SmI2/H2O/Et3N was studied at temperatures from 0 ꢁC to 50 ꢁC
at temperature increments of 10 ꢁC. The values for kobs was
obtained by determining the initial rates at concentrations where
The rate of reduction was also found to be dependent on the
amine, and there is a clear Brønsted correlation between pKBH and
rate (Table 5).

T. Ankner, G. Hilmersson / Tetrahedron 65 (2009) 10856–10862
10861
Table 5
Relative rates
presumably destabilized by substituents capable of donating
electrons (–OMe and –F) thus lowering the reaction rate. On the
contrary, electron withdrawing substituents will enhance the rate,
and this is also the case as the difference in rate between
4-methoxy benzyl alcohol and 4-trifluoromethyl benzyl alcohol is
approximately three orders of magnitude. We can also conclude
that the rate of reduction is highest when three equivalents of
water are present. Higher amounts of water will effectively displace
coordinated benzyl alcohol leading to a lower rate. This proves that
the alcohol must coordinate to the metal center in order to receive
an electron and this strongly indicates that it occurs via an inner-
sphere mechanism.
The strong dependency on electronic effects indicate that the
reaction proceeds likely through a radical anion intermediate that
rearranges to a benzyl radical intermediate (Scheme 1) followed by
transfer of another electron and proton to yield the reduced
product, toluene. The role of the amine is similar to that reported
for the SmI2/H2O/R3N mediated reduction of alkyl halides, i.e. the
amine deprotonates the SmI2(H2O)x complex, yielding a much
more powerful reducing agent.19
a
Entry
Amine
Rate
Log rate
pKBH
1
2
3
4
Pyrrolidine
Isopropylamine
Et3N
0.029
0.029
0.0050
0.0005
ꢀ1.53
ꢀ1.54
ꢀ2.31
ꢀ3.30
11.26
10.68
10.62
9.99
Tributylamine
a
The values were predicted using ACD Lab software for THF.
2.6. Kinetic isotope effect studies
The rate of reduction of benzyl alcohol with SmI2/amine/water
shows a substantial kinetic isotope effect kH O=kD O ¼ 1:9. The
isotope effect were found to be independent of the water concen-
trations, [L2O]¼130 or 667 mM.
2
2
The result is expected as the water is of first order in the rate
expression (Eq.1) and the proton transfer therefore is part of the rate
determining step. The incorporation of deuterium was complete as
only a-deuterium toluene could be detected in the reaction mixture.
2.7. Spectroscopic investigation UV
The UV–Vis spectrum of dilute SmI2 solution with added benzyl
alcohol reveals a 50 nm hypochromic shift. The UV–Vis band at
617 nm from the THF complex of SmI2 shifts to 567 nm upon
addition of MeOH. ([SmI2]¼2 mM, [Bn–OH]¼0.5–8 M).
3. Conclusion
The reactivity pattern displayed by SmI2/H2O/R3N is in close
resemblance with that of the Birch reaction, but in contrast to this
reaction the reduction is shut down once the heteroatom is cleaved
off and the aromaticity is preserved. The reactions proceeds
smoothly and the reaction do not involve any hazardous reagents or
conditions and no precautions is required for its use. We have
shown that SmI2/H2O/R3N has potential use as deoxygenation and
debenzylation reagent. Furthermore, as a result of the detailed
mechanistic studies we have revealed that the amount of
coordinating water can have a dramatic effect on the reactivity of
SmI2/H2O/R3N, possibly by inhibiting an inner-sphere electron
transfer pathway. This behavior was not observed during our
previous studies on the reduction of benzyl halides, which most
likely proceeds via an outer-sphere mechanism. It is thus possible
that excess of water can inhibit other inner-sphere reactions
mediated by SmI2/H2O/R3N, i.e., the amount of water can possibly
From this data it is possible to obtain the equilibrium constant
between SmI2 and benzyl alcohol. Regression analysis of the
UV–Vis data for benzyl alcohol gave a complex equilibrium
constant of 0.095 Mꢀ117,18
.
This proves that there is an equilibrium between SmI2 and
benzyl alcohol in THF. The UV–Vis spectra of SmI2 and aliphatic
amines were also studied; Et3N, isopropyl amine, pyrrolidine. Et3N
gave no effect on the UV spectra while isopropylamine and pyrro-
lidine additions gave rise to a strong bathochromic shift.
2.8. Mechanism proposal
The evidence collected indicates that the first electron is
transferred to the aromatic ring forming a radical anion. This is
Scheme 1. A suggested mechanism for the reduction of benzyl alcohol derivatives by SmI2/H2O/R3N.

10862
T. Ankner, G. Hilmersson / Tetrahedron 65 (2009) 10856–10862
control the functional group selectivity of this reagent. We are
currently exploring similar reactions in our laboratory with the aim
of developing selective reactions mediated by SmI2/H2O/R3N.
4.4. UV–Vis experiments
The benzyl alcohol was distilled under nitrogen using a column
packed with small glass helices and a variable take off distillation
head and subsequently dried with 4 Å molecular sieves. Use of
dried benzyl alcohol from the bottle was not pure enough and
immediately quenched the dilute SmI2 solution used. Measure-
ments were recorded between 400 and 800 nm.
4. Experimental
4.1. General
All chemicals were purchased from Aldrich except samarium
(Alfa) and used without further purification unless stated other-
wise. All handling of the samarium(II) iodide solutions were done
in a nitrogen glovebox.
Acknowledgements
Financial support from the Swedish research council is grate-
fully acknowledged.
Column chromatography: SiO2 Kiselgel 60 (Merck flash chro-
matography grade). TLC: Merck pre-coated SiO2 plates 60F254.
1H
NMR and 13C NMR were recorded on a Varian Unity 400, chemical
shifts is given in ppm relative to CDCl3. GC/MS analysis used for
identification and kinetics were recorded on a Varian 3400 GC with
a Varian Saturn 2000 MS, column: SB-5 MS, and Helium was used
as carrier gas. Temperature program: 40–330 ꢁC (12 ꢁC/min), hold
time 4 min. UV–Vis was recorded on a Varian Cary Bio 100.
References and notes
1. For recent examples see: (a) Kartika, R.; Taylor, R. E. Angew. Chem., Int. Ed. 2007,
46, 6874–6877; (b) Reddy, M. S.; Narender, M.; Rao, R. Tetrahedron 2007, 63,
11011; (c) Rodriquez, V.; Quintero, L.; Sartillo-Priscil, F. Tetrahedron Lett. 2007,
48, 4305; (d) Hutt, O. E.; Mander, L. N. J. Org. Chem. 2007, 72, 10130.
2. Birch, A. J. J. Chem. Soc. 1945, 809–813; Goering, B. K.; Lee, K. J.; An, B. J.; Cha, J. K.
J. Org. Chem. 1993, 58, 1100–1103; Liu, H. J.; Yip, J.; Shia, K. S. Tetrahedron Lett.
1997, 38, 2253–2256; Srikrishna, A.; Beeraiah, B. Synth. Commun. 2007, 37,
2855–2860.
4.2. General procedure for the reductive cleavage of the
benzyl heteroatom bond (Table 2)
3. Greene, T. W.; Wuts, P. G. M. Protective Groups in Organic Synthesis, 4th ed.; John
Wiley & Sons: New York: NJ, 2007; and references therein.
SmI2 (0.52 mmol, 4 mL, 0.13 M in THF containing decane as
4. Other methods for deoxygenations include catalytic hydrogenation: Hartung,
W. H.; Simonoff, R. Org. React. 1953, 7, 263; Lewis acid and a silane: Mayr, H.;
Dogan, B. Tetrahedron Lett. 1997, 38, 1013; LAH and TiCl4: Sato, F.; Tomuro, Y.;
Ishikawa, H.; Oikawa, T.; Sato, M. Chem. Lett. 1980, 103; Indirect methods such
as Barton–McCombie: Barton, D. H. R.; McCombie, S. W. J. Chem. Soc., Perkin
Trans. 1 1975, 16, 1574.
5. (a) Dahlen, A.; Hilmersson, G. Tetrahedron Lett. 2002, 43, 7197–7200; (b) Dahlen, A.;
Hilmersson, G. Chem.d Eur. J. 2003, 9, 1123–1128.
6. Dahlen, A.; Hilmersson, G. Tetrahedron Lett. 2003, 44, 2661–2664.
7. Dahlen, A.; Nilsson, A.; Hilmersson, G. J. Org. Chem. 2006, 71, 1576–1580.
8. Dahlen, A.; Sundgren, A.; Lahmann, M.; Oscarson, S.; Hilmersson, G. Org. Lett.
2003, 5, 4085–4088.
internal standard), pyrrolidine (1.04 mmol, 90 mL) and the substrate
(0.104 mmol) was cooled to ꢀ20 ꢁC (if specified) and water was
added (1.56 mmol, 28 mL). The reaction was left to stir for the
specified time. The reaction mixture was quenched with the
addition of diethyl ether (4 mL) followed by dilute hydrochloric
acid (4 mL, 0.5 M) and one drop saturated sodium thiosulfate
solution. The organic phase was analyzed on GC/MS comparing
with authentic samples and using the internal standard for
quantification.
9. Ankner, T.; Hilmersson, G. Org. Lett. 2009, 11, 503–506.
On large scale exemplified for benzhydrol: Benzhydrol
(0.42 mmol, 77 mg) was added to SmI2 (2.08 mmol, 16 mL 0.13 M in
THF) followed by pyrrolidine (4.16 mmol, 360 mL) and the mixture
was cooled to ꢀ20 ꢁC. Water was added and the reaction was
allowed to proceed for 21 h. The reaction was quenched with
diethyl ether (20 mL) and shaken with hydrochloric acid (30 mL
0.5 M). The organic phase was dried over MgSO4 and evaporated to
yield crude diphenylmethane (70 mg). This was purified with
column chromatography to yield 59 mg (85%) product identical to
an authentic sample diphenylmethane.
10. (a) Kato, Y.; Mase, T. Tetrahedron Lett. 1999, 40, 8823–8826; (b) Kusuda, K.;
Inanaga, J.; Yamaguchi, M. Tetrahedron Lett. 1989, 30, 2945–2948; (c) Kamochi,
Y.; Kudo, T.; Masuda, T.; Takadate, A. Chem. Pharm. Bull. 2005, 53, 1017.
11. Amine (2 equiv) and water (3 equiv) was added to a SmI2 solution in THF.
Aliquots of the mixture were titrated with our previously published method
with regular intervals.
12. Ankner, T.; Hilmersson, G. Tetrahedron Lett. 2007, 48, 5707.
13. He, Y.; Pan, X.; Zhao, H.; Wang, S. Synth. Commun. 1989, 19, 3051–3054; Tabuchi,
T.; Inanaga, J.; Yamaguchi, M. Tetrahedron Lett. 1986, 27, 601–602; Milne, J. E.;
Murry, J. A.; King, A.; Larsen, R. D. J. Org. Chem. 2009, 74, 445–447; Markgraf, J. H.;
Basta, S. J.; Wege, P. M. J. Org. Chem. 1972, 37, 2361; Rashkin, M. J.; Hughes, R. M.;
Calloway, N. T.; Waters, M. L. J. Am. Chem. Soc. 2004, 126, 13320–13325.
14. (a) Espenson, J. H. In Chemical Kinetics and Reaction Mechanisms, 2nd ed.; Speer,
J. B., Morris, J. M., Eds.; McGraw-Hill: 1995; p 8, 32; (b) Cox, B. G. In Modern Liquid
Phase Kinetics; Compton, R. G., Davies, S. G., Evans, J., Eds.; Oxford University:
New York, NY, 1994; pp 12–13; (c) Casado, J.; Lopez-Quintela, M. A.; Lorenzo-
Barral, F. M. J. Chem. Educ. 1986, 63, 450; (d) Hall, K. J.; Quickenden, T. I.; Watts, D.
W. J. Chem. Educ. 1976, 53, 493.
4.3. General procedure for the kinetic experiments
SmI2 (0.4 mmol, 4 mL containing decane as internal standard)
was added to a Schlenck tube and equilibrated in a water bath at
20.0 ꢁC. Water and amine was added and after a short period of
15. Prasad, E.; Knettle, B. W.; Flowers, R. A. J. Am. Chem. Soc. 2004, 126, 6891.
16. kobs is obtained by dividing the initial rate with [SmI2][BnOH][amine][H2O].
17. K defined as K¼[SmI2(BnOH)]/[SmI2][BnOH].
time benzyl alcohol was added. Five aliquots (200 mL) was with-
18. This is an expected value, considering that Hoz reported an equilibrium
constant of 0.56 Mꢀ1 for the SmI2–MeOH complex formation. See Yacovan, A.;
Hoz, S. J. Am. Chem. Soc. 1996, 118, 261.
drawn and quenched in hexane saturated with iodine (0.5 mL).
Dilute hydrochloric acid (0.5 mL 0.1 M) and diethyl ether (0.5 mL)
was added and the organic phase was analyzed using GC/MS.
19. Dahlen, A.; Hilmersson, G. J. Am. Chem. Soc. 2005, 127, 8340–8347.