Rhodium(II)-catalyzed decomposition of 3-O-(2-diazo-2-phenylacetyl)-1,2;5,6-di-O-isopropylidene-α-d-allofuranose: diastereoselective ether formation

Article abstract of DOI:10.1016/j.carres.2008.04.005

Standard diazo transfer to 3-O-(2-phenylacetyl)-1,2;5,6-di-O-isopropylidene-α-d-allofuranose (2), using p-acetamidobenzenesulfonyl azide (p-ABSA, 3) and DBU as base, provides the expected 3-O-(2-diazo-2-phenylacetyl)-1,2;5,6-di-O-isopropylidene-α-d-allofuranose (4) as an orange syrup in 49% isolated yield. Subsequent decomposition of 4 using Rh₂(OAc)₄ yields ether 5 in a highly diastereoselective manner and in 58% isolated yield. The X-ray crystal structure of 5 proves that both newly produced stereocenters have the (S) configuration; the conformation of the ester group at O-3 of the furanose ring of 5 is used to discuss the possible cause of the observed stereoselectivity.

Full text of DOI:10.1016/j.carres.2008.04.005

Available online at www.sciencedirect.com
Carbohydrate Research 343 (2008) 1819–1823
Note
Rhodium(II)-catalyzed decomposition of
3-O-(2-diazo-2-phenylacetyl)-1,2;5,6-di-O-isopropylidene-
a-D-allofuranose: diastereoselective ether formation
Iulia A. Sacui, Matthias Zeller and Peter Norris*
Department of Chemistry, Youngstown State University, 1 University Plaza, Youngstown, OH 44555, USA
Received 24 January 2008; received in revised form 29 March 2008; accepted 2 April 2008
Available online 7 April 2008
Abstract—Standard diazo transfer to 3-O-(2-phenylacetyl)-1,2;5,6-di-O-isopropylidene-a-D -allofuranose (2), using p-acet-
amidobenzenesulfonyl azide (p-ABSA, 3) and DBU as base, provides the expected 3-O-(2-diazo-2-phenylacetyl)-1,2;5,6-di-O-isopro-
pylidene-a-D-allofuranose (4) as an orange syrup in 49% isolated yield. Subsequent decomposition of 4 using Rh2(OAc)4 yields
ether 5 in a highly diastereoselective manner and in 58% isolated yield. The X-ray crystal structure of 5 proves that both newly pro-
duced stereocenters have the (S) configuration; the conformation of the ester group at O-3 of the furanose ring of 5 is used to discuss
the possible cause of the observed stereoselectivity.
Ó 2008 Elsevier Ltd. All rights reserved.
Keywords: Allofuranose; Diazoester; Rhodium(II)-catalyzed decomposition; Diastereoselective insertion
Metal-stabilized carbenes have proven to be reliable and
versatile reagents in organic chemistry, particularly for
the formation of new carbon–carbon bonds by insertion
into C–H bonds or addition to alkenes.1 These species
offer other interesting modes of reactivity that have scar-
cely been studied in the carbohydrate field. As such, we
are exploring the possibilities of intramolecular carbe-
noid insertion chemistry for the synthesis of branched-
chain sugars and C-glycosides and have been focusing
on Rh(II)-catalyzed decomposition of stabilized diazo
compounds derived from furanose platforms.2 Others
have found that decomposition of benzyl-protected
pyranose-derived diazoesters does not lead to intra-
molecular insertion into the carbohydrate framework
but to products formed by insertion into the protecting
groups, into the solvent (benzene), or from dimerization
of the carbenoid intermediate.3 The choice of furanose-
derived substrates, as well as careful selection of protect-
ing groups, is seen as a way of increasing the chances of
intramolecular insertion, however we are also finding
that intermolecular processes are competing. Here we
describe the diastereoselective intermolecular insertion
of two sugar-derived carbenoids into the O–H bonds
of a molecule of water to give a symmetrical ether as
the major product.
Using 1,2;5,6-di-O-isopropylidene-a-D-allofuranose
(1) as the starting platform, esterification with phenyl-
acetic acid was accomplished using standard conditions
(DCC, 4-DMAP) to provide the 2-phenylacetyl ester 2
in 70% isolated yield (Scheme 1) as a colorless syrup.
1
The H NMR spectrum of 2 showed the expected 2H
singlet at 3.70 ppm for the a-hydrogens of the phenyl-
acetyl group and a 5H multiplet at 7.25–7.34 ppm for
the protons on the phenyl ring. The 13C NMR spectrum
of this material featured a signal at 171.5 ppm, which
corresponds to the ester carbonyl of 2.
Diazo transfer to compound 2 was accomplished
using commercially available p-acetamidobenzenesulfo-
nyl azide (p-ABSA, 3) and DBU as base. After aqueous
workup and purification by silica gel chromatography,
diazoester 4 was isolated as an orange syrup in 49%
yield, and proved to be stable for months upon storing
in the freezer. The 1H NMR spectrum of 4, when
*
Corresponding author. Tel.: +1 330 941 1553; fax: +01 330 941
1579; e-mail: pnorris@ysu.edu
0008-6215/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carres.2008.04.005

1820
I. A. Sacui et al. / Carbohydrate Research 343 (2008) 1819–1823
O
S
O
O
O
O
O
O
O
AcHN
N3
PhCH2CO2H
3
O
O
O
N2
O
O
DCC, 4-DMAP
CH2Cl2, CH3CN
rt, 12 h
O
DBU
O
O
OH O
O
O
CH2Cl2, CH3CN
rt, 12 h
O
O
70%
49%
1
4
2
Scheme 1.
compared to that of precursor 2, showed the loss of the
2H singlet at 3.70 ppm and the IR spectrum of 4
revealed the diagnostic diazo band at 2100 cmꢀ1. The
the major process occurring under these conditions
was the insertion of 2 equiv of a carbenoid into one mol-
ecule of H2O.
O
O
O
O
O
O
O O
Rh2(OAc)4
O
O
O
N2
O
O
CH2Cl2, rt
ð1Þ
O O
O
O
O
O
O
O
O
4
5
diazoester was freely soluble in acetone, THF, CHCl3,
and CH2Cl2, and was homogenous by TLC (silica gel,
Rf = 0.45, 2:1 hexanes–ethyl acetate).
Insertion of metal-stabilized carbenoids into O–H
bonds is a well-known process,4 the intramolecular var-
iant of which has been applied in several situations to
the synthesis of carbohydrate analogs.5 Related to the
current work, the use of chiral auxiliaries to influence
the stereochemical course of intermolecular O–H inser-
tion processes in non-sugar examples has been studied
by Moody and colleagues who observed differing levels
of diastereoselectivity in insertions, using simple
alcohols and water, with diazoesters based on enantio-
merically pure alcohols such as (ꢀ)-borneol.6 Although
the mechanistic details of such insertions are still not
fully understood, a model for diastereoselection was
proposed that depends upon the preferred geometry of
the carbenoid intermediate and involves the water (or
alcohol) molecule attacking the highly electrophilic
carbenoid carbon from the more accessible face.6 The
stereoselection observed upon the decomposition of
diazoester 4 may be explained by a similar model.
The fact that we isolate an ether and not an alcohol
product here is most likely due to only a limited amount
of water being available in the reaction mixture thus
allowing for sequential O–H insertions. With the known
stereochemical identity of the allofuranose rings as refer-
ence, it can clearly be seen that both new stereocenters in
5 (which are identical) are of the (S) configuration (Fig.
1). It is reasonable to expect that 5 is formed by initial
insertion of a molecule of carbenoid (6, Scheme 2) into
H2O to give alcohol 7, which then serves as the substrate
Decomposition of diazoester 4 was achieved by addi-
tion of a dilute (0.04 g/mL) CH2Cl2 solution of the sub-
strate by syringe pump over 1 h to a stirred suspension
of Rh2(OAc)4 in CH2Cl2. After stirring for a further
12 h, TLC showed that 4 had been consumed and that
several very minor products had been formed along with
a major component with Rf = 0.08 (2:1 hexanes–ethyl
1
acetate). Analysis of the crude H NMR spectrum con-
firmed that the bulk of the crude mixture was indeed one
compound and careful column chromatography on sil-
ica gel yielded this material as a colorless syrup that
could be crystallized from ethanol. The mass spectrum of
the product showed a molecular ion at 793.6 [M+Na]+
which suggested an intermolecular process involving the
loss of N2 from two molecules of the precursor diazoest-
er and the introduction of the elements of a molecule of
H2O. That a symmetrical species had been formed was
supported by the 1H NMR spectrum; eight signals
appear between 3.42 and 5.82 ppm corresponding to
the seven protons on the sugar backbone plus a
deshielded (5.13 ppm) singlet. The X-ray structure of 5
(Fig. 1, Tables 1 and 2) gave conclusive evidence that
The catalyst and solvents employed in the decomposition experi-
ments were used as received from the supplier without further drying.

I. A. Sacui et al. / Carbohydrate Research 343 (2008) 1819–1823
1821
Figure 1. X-ray crystal structure with atom labeling scheme of ether 5.
˚
Table 1. Crystallographic data for ether 5
Table 2. Selected bond lengths (A) and bond angles (°) for ether 5
Empirical formula
Formula weight
Crystal size (mm)
Crystal system
Z
C
770.80
40H50O15
Bond lengths
C-1–C-2
C-2–C-3
C-3–C-4
C-1–O-4
C-4–O-4
1.530(5)
1.566(5)
1.540(5)
1.396(4)
1.444(4)
C-3–O-3
1.433(4)
1.349(5)
1.200(5)
1.525(5)
1.422(4)
0.60 ꢁ 0.09 ꢁ 0.06
C-13–O-3
C-13–O-2
C-13–C-14
C-14–O-1
Monoclinic
2
C2
Space group
˚
a (A)
28.638(3), a = 90°
5.4002(5), b = 907.249(2)°
12.3306(12), c = 90°
1891.7(3)
1.353
820
0.103
90(2)
0.71073
Needle, colorless
Bond angles
O-4–C-1–C-2
C-1–C-2–C-3
C-2–C-3–C-4
C-3–C-4–O-4
˚
b (A)
108.1(3)
102.5(3)
104.8(3)
103.9(3)
C-2–C-3–O-3
111.4(3)
115.6(3)
108.7(3)
113.0(4)
˚
c (A)
C-3–O-3–C-13
C-14–C-13–O-3
C-14–O-1–C-140
3
˚
V (A )
Density (calcd) (Mg/m3)
F(000)
Absorption coefficient (mmꢀ1
Temperature (K)
)
ter. Once alcohol 7 is formed, insertion of a second
equivalent of carbenoid (6) should occur in a similar
manner to produce the same stereochemical outcome
which is seen in 5.
˚
Wavelength (A)
Crystal shape, color
h range for data collection
Limiting indices
1.43–26.37°
ꢀ35 6 h 6 35, ꢀ6 6k 6 6,
ꢀ15 6l 6 15
8530
2160 (R(int) = 0.0453)
100.0%
Isolation of ether 5 as the major product here suggests
that the bulky acetal group at O-5–O-6 plays an impor-
tant role not only in the stereochemical course of the
intermolecular process, but that this large group might
also preclude the possibility of intramolecular insertion.
Once carbenoid 6 is formed (Scheme 2), the only intra-
molecular C–H insertion process that seems feasible is
into the cis-aligned C-4–H-4 bond of the furanose ring.
Unfavorable steric interactions between the bulky ‘Rh
2Ln’ group and the O-5–O-6 acetal in 6 likely prevents
the intramolecular process from occurring.
Reflections collected
Independent reflections
Completeness to h = 26.37°
Absorption correction
Refinement method
Data/restraints/parameters
Goodness-of-fit on F2
Final R indices [I > 2r(I)]
R indices (all data)
Multi-scan
Full-matrix least-squares on F2
2160/1/253
1.287
R1 = 0.0577, wR2 = 0.1151
R1 = 0.0621, wR2 = 0.1167
0.378 and ꢀ0.219
Largest difference in peak and
ꢀ3
˚
hole (e A
)
for a second insertion to give 5. If carbenoid 6 adopts a
conformation similar to that shown in Scheme 2 (cf.
Ref. 6) this would favor attack at the ‘front’ face of the
carbenoid by the incoming H2O nucleophile with ap-
proach to the ‘back’ of the carbenoid being hindered
by the presence of the O-5–O-6 isopropylidene acetal.
This leads to the preferential creation of an (S) stereocen-
1. Experimental
1.1. General methods
1H and 13C NMR spectra were recorded on a Varian
Gemini 2000 system, at 400 and 100 MHz, respectively,
as solutions in CDCl3; IR spectra were collected on a

1822
I. A. Sacui et al. / Carbohydrate Research 343 (2008) 1819–1823
O
O
O
O
O
O
O
Rh2+
O
H2O
6
4
5
O
O
O
O
O
O
Rh2Ln
H
HO
6
7
Scheme 2.
Thermo Electron Corporation IR 200 spectrophoto-
meter; low resolution mass spectra were obtained using
a Bruker Esquire LC–MS instrument; high resolution
mass spectra were carried out at the Campus Chemical
Instrumentation Center at The Ohio State University;
optical rotations were recorded on a Perkin Elmer model
343 automatic polarimeter as solutions in CH2Cl2.
Diffraction data for compound 5 was collected on a
Bruker AXS SMART APEX CCD diffractometer at
90(2) K using monochromatic MoKa radiation with
omega scan technique using the SMART software.7 The
unit cell was refined and the data were integrated using
SAINT+. 8 The structure was solved by direct methods
and refined by full matrix least squares against F2 with
all reflections using SHELXTL.9 All non-hydrogen atoms
were refined anisotropically. Hydrogen atoms were
placed in calculated positions and refined with an isotro-
pic displacement parameter 1.5 (CH3) or 1.2 times (all
others) that of the adjacent carbon atom. Friedel equiva-
lents were merged prior to refinement and the absolute
structure of the compounds was assigned based on the
unchanged configuration of carbon atoms also present
in the starting material. CCDC 675209 contains the
supplementary crystallographic data for this paper.These
data can be obtained free of charge via www.ccdc.cam.
ac.uk/data_request/cif, by e-mailing data_request@
ccdc.cam.ac.uk, or by contacting the Cambridge Crystal-
lographic Data Centre, 12 Union Road, Cambridge CB2
1EZ, UK; fax: +44(0)1223-336033.
purifiedon a silicagelcolumn (4:1 hexanes–EtOAc) to give
1.025 g (70% yield) of 2 as a colorless syrup: [a]D +63.4 (c
5, CH2Cl2); 1H NMR: d 1.32 (s, 6H, 2 ꢁ –CH 3), 1.34 (s,
3H, –CH3), 1.50 (s, 3H, –CH3), 3.70 (s, 2H, –CH2–), 3.75
(dd, 1H, H-4, J = 6.0, 8.6 Hz), 4.00 (dd, 1H, H-3,
J = 6.8, 8.6 Hz), 4.13 (dd, 1H, H-2, J = 4.4, 8.1 Hz), 4.30
(ddd, 1H, H-5, J = 4.2, 6.0, 10.4 Hz), 4.80 (m, 2H, H-6
and H-60), 5.85 (d, 1H, H-1, J = 3.7 Hz), 7.20–7.30 (m,
5H, Ar-H); 13C NMR: d 26.3, 27.4, 27.8, 28.0, 42.1, 66.7,
74.0, 76.0, 78.6, 78.7, 105.2, 110.9, 114.1, 128.2, 129.5
(2 ꢁ C), 130.4 (2 ꢁ C), 134.4, 171.5. HRMS m/z calcd
for C20H26O7Na: 401.1576. Found: 401.1551.
1.3. 3-O-(2-Diazo-2-phenylacetyl)-1,2;5,6-di-O-isopropyl-
idene-a-D-allofuranose (4)
2-Phenylacetyl derivative 2 (2.5 g, 6.6 mmol) and p-acet-
amidobenzenesulfonyl azide (3, 1.590 g, 6.6 mmol) were
added to a flame-dried flask and dissolved in dry CH2Cl2
(25 mL) and dry CH3CN (25 mL). While stirring at
room temperature, 1,8-diazabicyclo [5,4,0]undec-7-ene
(1.10 mL, 7.2 mmol) was added dropwise producing an
orange solution. TLC showed the product at an Rf of
0.44 (3:1 hexanes–EtOAc) and, after stirring for 12 h,
the reaction solvent was evaporated under reduced pres-
sure. The residue was dissolved in CH2Cl2 (30 mL) and
washed with 5% H2SO4 (3 ꢁ 15 mL) and then H2O
(2 ꢁ 15 mL). After drying with MgSO4 the solvent was
evaporated to give the crude product which was then
purified on silica gel (6:1 hexanes–EtOAc) to afford
1.31 g (49%) of 4 as an orange syrup: [a] D +127.4 (c
1.2. 3-O-(2-Phenylacetyl)-1,2;5,6-di-O-isopropylidene-a-
D-allofuranose (2)
1
0.5, CH2Cl2); IR (thin film): 2100 cmꢀ1; H NMR: d
1.35 (s, 6H, 2 ꢁ –CH3), 1.43 (s, 3H, –CH3), 1.56 (s, 3H,
–CH3), 3.90 (dd, 1H, H-4, J = 5.3, 8.6 Hz), 4.10 (dd,
1H, H-6, J = 6.6, 8.6 Hz), 4.20 (dd, 1H, H-60, J = 5.9,
8.6 Hz), 4.30 (ddd, 1H, H-5, J = 5.1, 6.6, 10.3 Hz), 4.90
(t, 1H, H-3, J = 4.8 Hz), 5.00 (dd, 1H, H-2, J = 4.9,
8.7 Hz), 5.86 (d, 1H, H-1, J = 4.0 Hz), 7.20–7.50 (m,
5H, Ar-H); 13C NMR: d 26.2, 27.6, 27.9, 28.0, 67.1,
74.6, 76.4, 78.8, 79.0, 105.3, 111.1, 114.3, 125.0 (2 ꢁ C),
126.0, 127.1, 129.9 (2 ꢁ C), 165.0.
1,2;5,6-Di-O-isopropylidene-a-D-allofuranose (1) (1.0 g,
3.9 mmol), phenylacetic acid (0.582 g, 4.3 mmol), and 4-
dimethylaminopyridine (0.085 g, 0.69 mmol) were added
to a flame-dried round-bottom flask under N2 atmosphere
and dissolved in anhydrous CH2Cl2 (10 mL) and anhy-
drous CH3CN (10 mL). 1,3-Dicyclohexylcarbodiimide
solution (4.3 mL, 1.0 M in CH2Cl2) was then added drop-
wise resulting in a white precipitate. The reaction mixture
was stirred overnight at room temperature, gravity fil-
tered, and the solvent was removed under reduced pres-
sure. The residue was dissolved in CH2Cl2 (20 mL) and
the solution washed with 5% H2SO4 (3 ꢁ 10 mL) and
H2O (2 ꢁ10 mL). After drying with MgSO4, the filtrate
was evaporated to give the crude product, which was then
1.4. (2S,20S)-Di-(3-O-1,2;5,6-di-O-isopropylidene-a-D-
allofuranose)-2,20-oxybis(2-phenylacetate) (5)
Rh2(OAc)4 (0.025 g, 0.05 mmol) was suspended in anhy-
drous CH2Cl2 (10 mL) under N2 atmosphere and a solu-

I. A. Sacui et al. / Carbohydrate Research 343 (2008) 1819–1823
1823
tion of diazoester 4 (0.400 g, 0.989 mmol) dissolved in
anhydrous CH2Cl2(10 mL) was added dropwise over
1 h using a syringe pump. TLC showed a major product
forming with an Rf value of 0.08 (3:1 hexanes–EtOAc)
and, after stirring for 12 h at room temperature, the sol-
vent was removed under reduced pressure and the resi-
due was purified using a flash column eluted with 3:1
hexanes–EtOAc. Ether 5 was isolated as a colorless syr-
for the CCD X-ray diffractometer at YSU were pro-
vided by the National Science Foundation (CCLI-
0087210) and the Ohio Board of Regents Action Fund
(Grant CAP-491).
Supplementary data
up (0.22 g, 58% yield), which was crystallized from eth-
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.carres.
2008.04.005.
1
anol: mp 168–170 °C; [a]D +111.3 (c 1.3, CH Cl2); H
2
NMR: 1.19 (s, 3H, –CH3), 1.26 (s, 3H, –CH3), 1.32 (s,
3H, –CH3), 1.48 (s, 3H, –CH3), 3.40 (dd, 1H, H-6,
J = 6.6, 8.8 Hz), 3.70 (dd, 1H, H-60, J = 6.7, 8.6 Hz),
4.10 (dd, 1H, H-2, J = 3.7, 8.4 Hz), 4.20 (ddd, 1H, H-
5, J = 3.8, 6.8, 10.4 Hz), 4.80 (dd, 1H, H-3, J = 5.1,
8.4 Hz), 4.90 (dd, 1H, H-4, J = 4.0, 5.1 Hz), 5.13 (s,
1H, –CH–), 5.80 (d, 1H, H-1, J = 3.7 Hz), 7.30–7.50
(m, 5H, Ar-H); 13C NMR: d 26.3, 27.2, 27.9, 28.1,
66.2, 74.1, 75.6, 78.6, 78.7, 80.0, 105.2, 110.8, 114.1,
128.8 (2 ꢁ C), 129.6 (2 ꢁ C), 130.0, 136.1, 170.3. HRMS
References
1. Doyle, M. P.; Forbes, D. C. Chem. Rev. 1998, 98, 911–935.
2. Berndt, D. F.; Norris, P. Tetrahedron Lett. 2002, 43, 3961–
3962.
3. Branderhorst, H. M.; Kemmink, J.; Liskamp, R. M. J.;
Pieters, R. J. Tetrahedron Lett. 2002, 43, 9601–9603.
4. Im, C. Y.; Okuyama, T.; Sugimura, T. Eur. J. Org. Chem.
2008, 285–294, and references cited therein.
m/z
calcd for C40H50O15Na: 793.3047. Found:
´
5. Sarabia, F.; Chammaa, S.; Lopez Herrera, F. J. Tetra-
793.2997.
hedron 2001, 57, 10271–10279.
6. Aller, E.; Brown, D. S.; Cox, G. G.; Miller, D. J.; Moody,
C. J. J. Org. Chem 1995, 60, 4449–4460.
Acknowledgments
7. SMART, Software for the CCD Detector System, version
5.630; Bruker AXS: Madison, WI, 2002.
8. SAINT+, Data Processing Software for the CCD Detector
System, version 6.45; Bruker AXS: Madison, WI, 2003.
9. SHELXTL, Software Package for the Refinement of Crystal
Structures, version 6.10; Bruker AXS: Madison, WI, 2000.
This work was supported by the Petroleum Research
Fund of the American Chemical Society (Type B Award
to P.N.) and the School of Graduate Studies and
Research at Youngstown State University. The funds