Characterization of the complex formed between samarium diiodide and the dehydro dimer of HMPA (diHMPA)

Article abstract of DOI:10.1016/j.tetlet.2009.06.139

A new ligand that facilitates samarium diiodide-mediated reductions has been developed. Addition of a solution of samarium diiodide to the dehydro dimer of hexamethylphosphoramide results in a purple complex which is an excellent reductant for a variety of organic functionalities. The complex was characterized by the kinetics of reduction of 1-bromodecane, visible spectroscopy, and cyclic voltammetry.

Full text of DOI:10.1016/j.tetlet.2009.06.139

Tetrahedron Letters 50 (2009) 5308–5310
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Characterization of the complex formed between samarium diiodide
and the dehydro dimer of HMPA (diHMPA)
*
Chriss E. McDonald , Jeremy D. Ramsey, James A. Grant, Kelly A. Howerter
Department of Chemistry, 700 College Place, Lycoming College, Williamsport, PA 17701, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
A new ligand that facilitates samarium diiodide-mediated reductions has been developed. Addition of a
solution of samarium diiodide to the dehydro dimer of hexamethylphosphoramide results in a purple
complex which is an excellent reductant for a variety of organic functionalities. The complex was char-
acterized by the kinetics of reduction of 1-bromodecane, visible spectroscopy, and cyclic voltammetry.
Ó 2009 Elsevier Ltd. All rights reserved.
Received 2 April 2009
Revised 16 June 2009
Accepted 29 June 2009
Available online 27 July 2009
Samarium diiodide as a useful reductant was introduced to the
synthetic organic community by Kagan.1 It was subsequently
cacy in facilitating reduction by samarium diiodide.15 Recently it
has been demonstrated that 1,3-dimethyl-2-imidazolidinone
1
6,17
determined that the addition of HMPA as a co-solvent to a THF
(DMI) has shown promise as an alternative to HMPA.
None of
2
solution of SmI
2
can accelerate reductions of alkyl halides. The
these, however, have been demonstrated to be generally applicable
complex formed between SmI
into one of the most versatile reagents for performing reductions,
often with concomitant formation of carbon–carbon bonds. Due
to its importance, this complex has been the subject of much
2
and 4 equiv of HMPA has evolved
to the wide range of synthetic applications for which SmI
ven useful.
2
has pro-
3
We have examined the dehydro dimer of HMPA (diHMPA, 2)
with regard to its ability to facilitate SmI -mediated reductions.
2
elegant mechanistic scrutiny. In THF solution, substantial
We presume that diHMPA would be less mutagenic than HMPA
due simply to its lower volatility. This compound was first syn-
]2+2I as the
ꢀ
evidence points to the complex [Sm(HMPA)
4
(THF)
2
4
–6
18
reactive species.
Hou has reported the crystal structure of
thesized by Naarman via radical dimerization of HMPA. It has
7
19
SmI
2
(HMPA)
4
.
been used previously to facilitate alkylations and conjugate
2
0
Although of great synthetic value, HMPA is also known to cause
nasopharyngeal cancer in laboratory animals. Rats were shown to
develop nasal tumors when HMPA was inhaled (50 ppb for one
additions of alkali metal enolates. There are no published re-
ports of the use of diHMPA in combination with samarium diio-
dide. We chose to synthesize diHMPA by bisphosphorylation of
8
0
year) but tumor formation was not observed when HMPA was
N,N -dimethylethylenediamine (Scheme 1). Purification was
ꢀ1
ꢀ1
9
administered orally to rats (6.25 mg kg day for two years).
accomplished by silica gel column chromatography. Storage of
this compound under argon at room temperature for up to a year
There is evidence that this carcinogenicity is related to hydroxyl-
ation at N-methyl positions of HMPA by cytochrome P450 enzymes
1
resulted in no decomposition as evidenced by H NMR or
with concomitant release of H
2
CO within the nasopharyngeal
cells. The propensity of intracellular H CO to crosslink DNA ulti-
mately leads to intra-locus and multi-locus deletions within the
discoloration.
1
0
2
2
When a 0.1 M solution of SmI in THF was exposed to 2 equiv of
diHMPA, a deep purple solution resulted. Initial evidence for the
reactivity of this mixture was obtained by the addition of 0.3 equiv
of 1-bromodecane to the solution. This resulted in rapid and
1
1
genome. In spite of this, chemists use SmI
2
with HMPA as a co-
solvent on a routine basis.
A small number of potential HMPA surrogates have been inves-
0
tigated. Electron-rich ligands such as N,N -dimethylpropyleneurea
2
(
DMPU)1 , 1,1,3,3-tetramethyl-guanidine (TMG), and 1,8-diazaby-
13
cyclo[5.4.0]undec-7-ene (DBU) have been employed with some
success. Mixtures of samarium diiodide with water and amines,
especially N,N,N ,N ,N -pentamethyldiethylene triamine (PMDTA),
O
P
0
00
00
Me
HN
2.5 eq Cl
NMe2
Me
N
O
P
NMe2
Me2N
were shown to be effective for the reduction of a variety of func-
tionalities. Various transition metal salts have also shown effi-
Me N
NMe2
NH
Me
2
P
N
1
4
5.0 eq Et N, CH Cl
NMe2
3
2
2
O
Me
1
85%
2
*
Corresponding author. Tel.: +1 570 321 4186; fax: +1 570 321 4073.
E-mail address: mcdonald@lycoming.edu (C.E. McDonald).
Scheme 1. The synthesis of diHMPA.
0
040-4039/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2009.06.139

C. E. McDonald et al. / Tetrahedron Letters 50 (2009) 5308–5310
5309
PA:SmI
previous studies, relatively high concentrations of the electrolytes
n-Bu NPF (0.10 M) and n-Bu NI (0.020 M) were initially chosen
to increase solution conductivity for voltammetry.
chemical analysis was complicated by the low solubility of the pre-
2
ratio on the standard reduction potential. In accord with
4
6
4
2
4,25
The electro-
2
+
ꢀ
sumed Sm(diHMPA)
high ionic strength, and consequently, no oxidation or reduction
waves were observed. When the concentration of n-Bu NPF was
lowered to 0.025 M (while maintaining n-Bu NI at 0.020 M), the
SmI –diHMPA complex had increased solubility and cyclic voltam-
2 2
(THF) ] 2I complex under conditions of
4
6
4
2
metry for the complexes was observed. Under the low electrolyte
conditions, the HMPA complex, had a measured standard potential
of ꢀ2.07 V versus Ag/AgNO
tial measured at higher electrolyte concentration (ꢀ2.10 V vs Ag/
AgNO ). The effect of diHMPA on reducing strength is illustrated
by the data in Table 2. As was shown for complexation of SmI with
HMPA , the standard potential decreases with the addition of
diHMPA. When 2 equiv of diHMPA was introduced, the standard
3
, which is similar to the standard poten-
3
2
Figure 1. Visible spectrum of 5 mM THF solution of SmI with 10 mM diHMPA.
2
2
4
complete conversion of the alkyl bromide to decane as evidenced
by gas chromatography.
potential was lowered to ꢀ2.03 V versus Ag/AgNO
3
, which is pre-
A visible spectrum of 5 mM solution of SmI
2
in THF with 2 equiv
sumably due to the reducing power of the diHMPA complex,
of diHMPA afforded a peak at 543 nm (Fig. 1). A visible spectrum of
2+
ꢀ
[
(
Sm(diHMPA)
2
(THF)
2
]
2I . When smaller amounts of diHMPA
) were utilized, the standard potential
). The
2+
ꢀ
4 2
the related [Sm(HMPA) (THF) ] 2I complex has a peak at 540 nm
1 equiv relative to SmI
2
and the two spectra are qualitatively similar over the scanned
was affected to a much lesser extent (ꢀ1.43 V vs Ag/AgNO
3
ꢀ
range of 500–700 nm.4
,5
2+
change in standard potential for [Sm(diHMPA)
(ꢀ1.33 V vs Ag/AgNO
tive to SmI ) was measured to be 0.7 V,
2 2
which is similar to the change of 0.8 V for [Sm(HMPA) ] 2I ,
2 2
(THF) ] 2I rela-
Kinetic studies were performed with 1-bromodecane to deter-
mine the relative reactivity of the HMPA, diHMPA, and DMPU com-
2
3
+
ꢀ
4
(THF)
as measured by Flowers and confirmed in this study.
The peak separation between the oxidation and reduction
waves ( ) provides a measure of the electron transfer rate con-
stant (k°). Daasbjerg et al. observed that a significant decrease in k°
accompanied complexation of SmI by HMPA, with the retardation
attributed to a large increase in the reorganization energy of
plexes of SmI
analysis because it is the most common alternative ligand to HMPA
for SmI -mediated reductions. In each case 1-bromodecane was
added to a stirred 0.020 M solution of SmI that contained the indi-
cated amount of each ligand, 1-butanol as the proton source, and
tetradecane as an internal standard. Sufficient quantities of SmI
2
(Table 1). It was decided to include DMPU in this
26
2
DE
p
2
2
2
and 1-butanol were included to ensure pseudo-first-order condi-
tions for the analyses. Aliquots were removed and immediately
+
ꢀ 28
[
Sm(HMPA)
greater decrease in k° upon introduction of diHMPA. In our study,
for SmI was measured to be 0.45 V and increased to 1.2 V for
the presumed complex, [Sm(diHMPA)
cates very sluggish electron transfer kinetics.
The ability of the SmI /diHMPA complex to effect the formation
4 2 2
(THF) ] 2I . The results in Table 2 indicate an even
2
1
quenched with I
2
.
Gas chromatographic analysis of the resultant
D
E
p
2
mixtures was performed. Linear plots of time versus ln[decane]
afforded the results shown in Table 1. The complex formed from
2+
ꢀ
2
2
(THF) ] 2I , which indi-
2
equiv of diHMPA and SmI
to that of [Sm(HMPA) (THF)
was used, the rate constant was significantly smaller as expected.
Because DMPU has a lesser affinity for SmI , 8 equiv of this ligand
was used in these studies. The reactivity of the DMPU complex of
SmI was approximately the same as that of the 1:1 complex of
diHMPA and SmI . In the absence of exogenous ligands, but under
2
afforded a rate constant fairly similar
2
] 2I . When only 1 equiv of diHMPA
2+
ꢀ
2
4
of carbon–carbon bonds was also examined. The first example in-
volves the addition of the ketyl radical anion derived from 4-phen-
2
29
ylbutanone to styrene (Table 3). Efficient conversion to adduct 4
was observed using either HMPA or diHMPA as a ligand. The use of
DMPU as a ligand afforded the product in significantly lower yield.
A second example that illustrates the effect of the ligand on
2
2
otherwise analogous conditions, no product was observed in a 4-h
period (10 times the length of the other kinetic analyses).
SmI
2
-induced formation of a carbon–carbon bond is shown in
30
Table 4. Pummerer rearrangement of sulfoxide 5 affords a very
It is known that the addition of 10 equiv of HMPA to SmI
affords the complex [Sm(HMPA) ] 2I .
6
2
in THF
Kinetic experiments
with this complex have shown it to be of comparable reactivity
31
reactive
tion components under vacuum, the substrate was transferred to a
mixture of SmI , ligand, and t-BuOH. All three ligands examined
a-trifluoroacetoxy sulfide. After removal of volatile reac-
2+
ꢀ 5,22
2+
ꢀ
23
2
to [Sm(HMPA)
4
2
(THF) ]
2I with alkyl iodide substrates. In anal-
32
provided similar yields of cyclized products.
ogous fashion, SmI
2
was added to 5 equiv of diHMPA. The resultant
purple complex immediately precipitated from the THF solution.
Addition of 1-bromodecane to the stirred mixture did not afford
detectable quantities of decane over a 5-h period of analysis.
Table 2
a
Effect of ligands on the standard potentials of SmI
2
2
Cyclic voltammetry (CV) of SmI /diHMPA complexes in THF was
also undertaken in an effort to characterize the effect of the diHM-
Reductant Equiv ligand/
Standard potentialb
D
to SmI
E relative
DE
p
(V)
equiv SmI
2
(V vs Ag/AgNO
3
)
2
SmI
2
—
4
4
1
2
ꢀ1.35 ± 0.03
ꢀ2.07 ± 0.01
—
0.45
0.83
0.93
1.41
1.09
Table 1
SmI :HMPA
0.72
0.78
0.08
0.68
2
2
2
c
Pseudo-first-order kinetic study on the effect of ligand choice on the reduction of 1-
bromodecane (21 °C)
SmI :HMPA
ꢀ2.10 ± 0.01
SmI :diHMPA
ꢀ1.43 ± 0.02
ꢀ2.03 ± 0.06
2
SmI :diHMPA
a
ꢀ1
Ligand
Equiv ligand/equiv SmI
2
k
obs (s
)
a
Data were recorded at 5 mM SmI
2
with n-Bu
4
6 4
NPF (0.025 M) and n-Bu NI
HMPA
4
2
1
8
0.0094 ± 0.0009
0.0069 ± 0.0007
0.0022 ± 0.0003
0.0020 ± 0.0003
(
0.020 M) at 100 mV/s.
diHMPA
diHMPA
DMPU
b
Standard potentials were estimated from cyclic voltammograms and confirmed
by semi-integration of the cyclic voltammetry data, Ref. 27.
c
Data were recorded at 5 mM SmI
2 4 6 4
with n-Bu NPF (0.10 M) and n-Bu NI
a
[
2
1-bromodecane] = 0.0030 M, [SmI ] = 0.020 M, [1-butanol] = 0.040 M.
(0.020 M).

5
310
C. E. McDonald et al. / Tetrahedron Letters 50 (2009) 5308–5310
Table 3
Joseph Keane, Jeffrey Newman, and Jason Stamm for helpful
discussions.
2
Effect of ligand choice on the SmI -mediated addition of 4-phenylbutanone to styrene
O
2
2
.0 eq styrene, 2.0 eq t-BuOH
.4 SmI2, THF, Ligand, RT, 2 h
OH
References and notes
Ph
Ph
Ph
3
4
1
2
3
.
.
.
Girard, P.; Namy, L.; Kagan, H. B. J. Am. Chem. Soc. 1980, 102, 2693–2698.
Inanaga, J.; Ishikawa, M.; Yamaguchi, M. Chem. Lett. 1987, 1485–1486.
Kagan, H. B. Tetrahedron 2003, 59, 10351–10372.
Ligand
HMPA
diHMPA
DMPU
Equiv ligand/equiv SmI
2
Yielda (%)
4
2
4
97b
82
52
4. Shotwell, J.; Sealy, J.; Flowers, R. A., II J. Org. Chem. 1999, 64, 5251–5255.
5
.
Enem
754.
ꢀ
´rke, R.; Hertz, T.; Skrydstrup, T.; Daasbjberg, K. Chem. Eur. J. 2000, 3747–
3
6
.
For an excellent review of mechanistic work done on various samarium
a
Isolated yield of chromatographically pure product.
Ref. 29.
diiodide reductions, see: Flowers, R. A., II Synlett 2008, 1427–1439.
7. Hou, Z.; Wakatsuki, Y. J. Chem. Soc., Chem. Commun. 1994, 1205–1206.
b
8
9
.
.
Kimbrough, R.; Gaines, T. Bull. Environ. Contam. Toxicol. 1973, 10, 225–226.
Lee, K.; Trochimowicz, H. J. Natl. Cancer Inst. 1982, 68, 157–171.
1
1
0. Jones, A.; Jackson, H. Biochem. Pharmacol. 1970, 19, 603–604.
1. Aguirrezabalaga, I.; Nivard, M.; Comendador, M.; Vogel, E. Genetics 1995, 139,
Table 4
649–658.
Effect of ligand choice on the Pummerer rearrangement and SmI
tion of sulfoxide 5
2
-mediated cycliza-
1
1
2. Hasegawa, E.; Curran, D. J. Org. Chem. 1993, 58, 5008–5010.
3. Cabri, W.; Candiani, I.; Colombo, M.; Franzoi, L.; Bedeschi, A. Tetrahedron Lett.
1
995, 36, 949–952.
4. Dahlén, A.; Hilmersson, G.; Knettle, B.; Flowers, R. A., II J. Org. Chem. 2003, 68,
870–4875.
O
S
SPh
O
1
O
i - iii
4
Ph
1
1
5. Machrouhi, F.; Hamann, B.; Namy, J.; Kagan, H. B. Synlett 1996, 633–634.
6. Inokuchi, T. J. Org. Chem. 2005, 70, 1497–1500.
OEt
OEt
5
6
17. Kumaran, R.; Bruedgam, I.; Reissig, H. Synlett 2008, 991–994.
1
8. Naarmann, H.; Beaujean, M.; Merényi, R.; Viehe, H. Polymer Bull. 1980, 2, 417–
25.
i excess (CF3CO)2O, 2 eq lutidine, RT; ii remove volatiles; iii 6 eq SmI2, Ligand, THF, 2 eq t-BuOH, 0 o
C
4
19. Nee, G.; Bottin-Strzalko, T.; Seyden-Penne, J.; Beaujean, M.; Viehe, H. J. Org.
Chem. 1983, 48, 1111–1114.
Ligand
Equiv ligand/equiv SmI
2
Yielda (%)
2
2
2
2
0. Cossentini, M.; Strzalko, T.; Seyden-Penne, J. Bull. Chim. Soc. Fr. 1987, 531–534.
1. Fuh, M.; Lin, T.; Chang, S. Talanta 1998, 46, 861–866.
2. Hou, Z.; Zhang, Y.; Wakatsuki, Y. Bull. Chem. Soc. Jpn. 1997, 70, 149–153.
3. Prasad, E.; Flowers, R. A., II J. Am. Chem. Soc. 2002, 124, 6895–6899.
HMPA
diHMPA
DMPU
4
2
4
59
62
50
a
24. Shabangi, M.; Flowers, R. A., II Tetrahedron Lett. 1997, 38, 1137–1140.
Isolated yield of chromatographically pure product.
2
2
2
5. Enemꢀ
´rke, R.; Daasbjberg, K.; Skrydstrup, T. Chem. Commun. 1999, 343–344.
6. Shabangi, M.; Kuhlman, M. L.; Flowers, R. A. Org. Lett. 1999, 1, 2133–2135.
7. Oldham, K. B.; Myland, J. C. Fundamentals of Electrochemical Science; Academic
Press: San Diego, 1994. p 426.
In conclusion, diHMPA is a reasonable synthetic alternative to
HMPA for SmI -mediated reductions. Spectroscopic, electrochemi-
cal, and kinetic studies indicate that the complex formed between
SmI and 2 equiv of diHMPA is structurally analogous to the com-
plex formed between SmI and 4 equiv of HMPA. Work continues
2
8. Enem
747–3754.
9. Ujikawa, O.; Inananga, J.; Yamaguchi, M. Tetrahedron Lett. 1989, 30, 2837–2840.
ꢀ
´rke, R. J.; Hertz, T.; Skrydstrup, T.; Daasbjerg, K. Chem. Eur. J. 2000, 6,
2
3
2
ꢀ
1
m
3055, 2980, 2935, 1712, 1652, 1443; 1H NMR
30. Sulfoxide 5 IR spectrum (cm ):
spectrum (300 MHz, CDCl ) d 7.60 (m, 2H), 7.56 (m, 3H), 6.90 (dt, J = 15.6,
.0 Hz, 1H), 5.81 (d, J = 15.6 Hz, 1H), 4.20 (t, J = 7.2 Hz, 2H), 2.80 (t, J = 8.2 Hz,
2
3
2
6
2
in this laboratory to find alternative ligands for SmI
less mutagenic and more reactive than HMPA.
2
that will be
13
H), 2.22 (m, 2H), 1.81 (m, 1H), 1.64 (m, 3H), 1.30 (t, J = 7.2 Hz, 3H); C NMR
) d 166.4, 147.8, 143.9, 131.0, 129.3, 124.0, 122.0,
0.2, 56.9, 31.7, 27.1, 21.8, 14.3; HRMS (ES): calcd for C15
03.1025, found: 303.1022.
spectrum (75 MHz, CDCl
6
3
3
+
3
O SH20Na (MNa ):
Acknowledgments
3
3
1. Tanikaga, R.; Yabuki, Y.; Ono, N.; Kaji, A. Tetrahedron Lett. 1976, 17, 2257–2258.
2. Adduct 6 has previously been constructed by Bu SnH—mediated cyclization of
the analogous -chloro sulfide, see: Tsai, Y.; Chang, F.; Huang, J.; Shiu, C.; Kao,
C.; Liu, J. Tetrahedron 1997, 53, 4291–4308.
3
a
K.A.H. would like to thank the Merck-SURF program for provid-
ing a research stipend. The authors are indebted to Holly Bendorf,