Regio- and stereoselective hydroformylation of glucal derivatives with rhodium catalysts

Article abstract of DOI:10.1021/om971010k

The hydroformylation of the differently protected glucal derivatives 3,4,6-tri-O-acetyl-D-glucal, 3,4,6-tri-O-benzyl-D-glucal and 3,4,6-tri-O-methyl-D-glucal was carried out with rhodium catalytic systems, and 2-formyl derivatives were obtained as the main products in yields of 58%, 68%, and 55% respectively, when [Rh₂(μ-OMe)₂(COD)₂]TP(O-o- ^tBuC₆H₄)₃ was used. The bulky phosphite needs to be used as auxiliary ligand to achieve the hydroformylation of these highly hindered cyclic olefins. A mechanistic cycle is proposed which explains the regio- and stereoselectivity of the reaction. The mononuclear rhodium complex trans[RhCl(CO)(P(O-o-^tBuC₆H₄)₃) ₂] was isolated at the end of the catalytic reaction by breaking the dinuclear complexes used as catalyst precursors. To confirm the characterization of trans-[RhCl(CO)(P(O-o-^tBuC₆H₄) ₃)₂], a separate synthesis was carried out which isolated crystals suitable for an X-ray determination.

Full text of DOI:10.1021/om971010k

Organometallics 1998, 17, 2857-2864
2857
Regio- a n d Ster eoselective Hyd r ofor m yla tion of Glu ca l
Der iva tives w ith Rh od iu m Ca ta lysts
Elena Ferna´ndez, Aurora Ruiz, Carmen Claver,* and Sergio Castillo´n*
Departament de Quı´mica, Universitat Rovira i Virgili, Pc¸a. Imperial Tarraco 1,
43005 Tarragona, Spain
Alfonso Polo
Departament de Quı´mica, Universitat de Girona, Girona, Spain
J uan F. Piniella and Angel Alvarez-Larena
Unitat de Cristal.lografı´a, Universitat Auto`noma de Barcelona, 08193 Bellaterra, Spain
Received November 17, 1997
The hydroformylation of the differently protected glucal derivatives 3,4,6-tri-O-acetyl-D-
glucal, 3,4,6-tri-O-benzyl-D-glucal and 3,4,6-tri-O-methyl-D-glucal was carried out with
rhodium catalytic systems, and 2-formyl derivatives were obtained as the main products in
yields of 58%, 68%, and 55% respectively, when [Rh2(µ-OMe)2(COD)2]/P(O-o-tBuC6H4)3 was
used. The bulky phosphite needs to be used as auxiliary ligand to achieve the hydroformyl-
ation of these highly hindered cyclic olefins. A mechanistic cycle is proposed which explains
the regio- and stereoselectivity of the reaction. The mononuclear rhodium complex trans-
[RhCl(CO)(P(O-o-tBuC6H4)3)2] was isolated at the end of the catalytic reaction by breaking
the dinuclear complexes used as catalyst precursors. To confirm the characterization of
trans-[RhCl(CO)(P(O-o-tBuC6H4)3)2], a separate synthesis was carried out which isolated
crystals suitable for an X-ray determination.
In tr od u ction
conditions of pressure and temperature. When the gas
consumption was strictly controlled, aldehydes 4 and 5
were preferentially obtained,5 but always as an R/â
mixture (Scheme 1).
Hydroformylation is a well-known industrial process
which has recently attracted the attention of researchers
with a view to functionalizing complex molecules.1 The
increasing interest in C-glycosides2 and C-disaccharides3
prompted us to explore the hydroformylation of glycals
as a direct method of obtaining 2-deoxy-1-C-formyl
glycosides, which are useful intermediates in their
synthesis. Rosenthal4 has studied the hydroformylation
of 3,4,6-tri-O-acetyl-D-glucal (1) using [Co2(CO)8] as
catalyst and has obtained an R/â mixture of the 1-hy-
droxymethyl-2-deoxy carbohydrates 2 and 3 in drastic
The hydroformylation of glycals involves an interest-
ing problem of chemo-, regio-, and stereoselectivity.
Rhodium catalysts are more active than cobalt catalysts,
which means that the pressure and temperature can
be lower and that the selectivity can be controlled better
than for cobalt catalysts under milder conditions.
Since the first attempts at the hydroformylation of
3,4,6-tri-O-acetyl-D-glucal (1) using [RhH(CO)(PPh3)3]6
(6) or the dinuclear precursors [Rh2(µ-S-tBu)2(CO)2-
(PPh3)2]7 (7) and [Rh2(µ-S(CH2)3NMe2)2(COD)2]8 (8) as
catalysts failed, a systematic study was made of the
hydroformylation of 3,4-dihydro-2H-pyran (9) and 2,3-
dihydrofuran (10), which can be considered as simple
glycal models. The 2,3-dihydrofuran was easily hydro-
formylated under mild conditions with 6 and 7 and in
the presence of PPh3, while the 3,4-dihydro-2H-pyran
was hydroformylated in very low yields even in more
drastic conditions.9,10 Only when the tris-ortho-tert-
butylphenyl phosphite11 (11) was used as auxiliary
ligand could the compound 9 be hydroformylated to give
* Corresponding authors. E-mail: claver@quimica.urv.es, castillon@
quimica.urv.es.
(1) (a) Botteghi, C.; Ganzerla, R.; Lenarda, M.; Moretti, G. J . Mol.
Catal. 1987, 40, 129. (b) Kalck, Ph. Organometallics in Organic
Synthesis; de Meijere, A., Tom Dieck, H., Eds.; Springer-Verlag:
Heidelberg, 1987; p 297-320. (c) Siegel, H.; Himmele, W. Angew.
Chem., Int. Ed. Engl. 1980, 19, 178. (d) New Synthesis with Carbon
Monoxide; Falbe, J ., Ed.; Springer-Verlag: Heidelberg, 1980. (e)
Botteghi, C.; Marchetti, M.; Paganelli, S.; Sechi, B. J . Mol. Catal. 1997,
118, 173.
(2) (a) Levy, D. E.; Tang, C. The Chemistry of C-Glycosides;
Tetrahedron Organic Chemistry Series; Baldwin, J . E.; Magnus, P.
D., Eds. Pergamon-Elsevier Science Ltd: Oxford, 1995. (b) Postema,
M. H. D. Tetrahedron 1992, 48, 8545. (c) Postema, M. H. D. C-Glycoside
Synthesis; CRC Press, Boca Raton, FL, 1995. (d) J eanneret, V.;
Meerpoel, L.; Vogel, P. Tetrahedron Lett. 1997, 38, 543.
(3) See the following and references cited therein: (a) Bols, M. J .
Chem. Soc., Chem. Commun. 1993, 791. (b) Martin, O. R.; Lai, W. J .
Org. Chem. 1993, 58, 176. (c) Sarabia-Garc´ıa, F.; Lo´pez-Herrera, F.
J .; Pino Gonza´lez, M. S. Tetrahedron Asymmetry 1995, 51, 5491. (d)
J arreten, O.; Skrydstrup, T.; Beau, J .-M. J . Chem. Soc., Chem.
Commun. 1996, 1661. (e) Fairbanks, A. J .; Perrin, E.; Sinay¨, P. Synlett
1996, 679.
(5) Rosenthal, A.; Abson, D.; Field, T. D.; Koch, H. J .; J . Mitchel, R.
E. Can. J . Chem. 1967, 45, 1525.
(6) Brown, C. K.; Wilkinson, G. J . Chem. Soc. A 1970, 2753.
(7) Bonnet, J . J .; Kalck, Ph.; Poilblanc, R. Inorg. Chem. 1977, 16,
1514.
(8) Bayo´n, J . C.; Real, J .; Claver, C.; Polo, A.; Ruiz, A. J . Chem. Soc.,
Chem. Commun. 1989, 1056.
(9) Polo, A.; Claver, C.; Castillo´n, S.; Ruiz, A.; Bayo´n, J . C.; Real,
J .; Mealli, C.; Massi, D. Organometallics 1992, 11, 3525.
(4) Rosenthal, A. Adv. Carbohydr. Chem. 1969, 24, 59.
S0276-7333(97)01010-8 CCC: $15.00 © 1998 American Chemical Society
Publication on Web 06/05/1998

2858 Organometallics, Vol. 17, No. 13, 1998
Ferna´ndez et al.
Sch em e 1
The NMR spectroscopy of 15 (Table 1) shows (a) the
1
signal at 197.2 in 13C NMR and at 9.54 ppm in H NMR
which is coupled with H2 (J ) 2.7 Hz), (b) three acetyl
groups at 1.98, 1.99, and 2.11 ppm in 1H NMR which
show that neither elimination nor substitution has
1
taken place, and (c) the signal at 2.99 ppm in H NMR
attributed to H2 is coupled to the signal at 3.48 ppm
(H1axial, J ) 11.3 Hz) which shows that H2 must be in
an axial position.
Compound 16 could not be separated and purified by
usual chromatographic techniques, and it was deriva-
tized to the corresponding 2,4-dinitrophenylhydrazone,
which was then separated by TLC. The structure of the
2,4-dinitrophenylhydrazone derived from 16 was deter-
mined taking into account the following 1H NMR
spectroscopic data (Table 1): (a) there are only two
acetate groups at 2.00 and 2.07 ppm, (b) there are two
pairs of signals (other than H6, H6′) at 3.79/4.10 ppm
(H1e, H1a) and at 1.16/2.72 ppm (H3e,H3a) which have
geminal coupling constants, (J ) 12 and 11 Hz, respec-
tively), and (c) H1a, H3a, and H3e have small coupling
constants which show that this proton is in an equato-
rial or pseudoequatorial position.
Sch em e 2
Sch em e 3
The resulting regioselectivity is in contrast with the
regioselectivity obtained when a cobalt catalyst was
used. It is also in contrast with the one obtained in
similar conditions in the hydroformylation of 9, where
the formyl group was mainly bonded to C1. This group
is now bonded to C2, the carbon double bond with the
highest electron density.9,13 In an attempt to improve
the yield and the selectivity of the process we made a
systematic study of the reaction conditions.
First, we analyzed how selectivity varied with reac-
tion time and we observed that there was a regular
increase in the percentage of compounds 14 and 16
(entries 1 and 2, Table 2, Figure 1). On the other hand,
compound 15 reached 14% in 3 h and then decreased,
which suggests that it was converted into the elimina-
tion product 14. In fact, the percentage of 14 + 15 was
almost constant throughout the reaction. The yield and
the selectivity did not change significantly when the
temperature was increased or decreased (entries 3 and
4, respectively), but above 130 °C and below 80 °C the
catalyst was not active. When the pressure was 40 or
90 bar (entries 5 and 6), conversions were low.
When toluene was used as the solvent, only traces of
the side product 16 were obtained, although significant
quantities of the hydrogenation product remained (entry
8). On the other hand, when THF, dioxane, and glycol
dimethyl ether were used as the solvent, the selectivity
observed was similar to the one obtained when 1,2-
dichloroethane was used, but the reaction rate was
slightly slower, in all the cases.
In a previous study, we attributed the differences in
the hydroformylation of 2,3- and 2,5-dihydrofuran when
P(OPh)3 (cone angle ) 128°) and P(O-o-tBuC6H4)3 (cone
angle ) 175°) were used as auxiliary ligands to steric
rather than electronic effects.9 To confirm this, we chose
P(o-MeC6H4)3 as the auxiliary ligand because it is a
phosphine with a large cone angle (194°).14 The very
low conversions obtained in the hydroformylation of 1
a mixture of aldehydes 12 and 139 (Scheme 2). Here
we show that glucal derivatives can be hydroformylated
using different rhodium precursors in the presence of
the bulky phosphite 11.
Resu lts a n d Discu ssion
Hyd r ofor m yla tion of 3,4,6-Tr i-O-a cetyl-D-glu ca l.
The 3,4,6-tri-O-acetyl-D-glucal (1) was hydroformylated
using the catalytic system [Rh2(µ-S(CH2)3NMe2)(COD)2]/
P(O-o-tBuC6H4)3 giving a mixture of aldehydes 14, 15,
and 16, together with some amounts of the hydrogena-
tion product 1712 (Scheme 3).
Compounds 14, 15, and 17 were separated from the
reaction mixture, and their structures were determined
by spectroscopic methods. The most significant spec-
troscopic data (Table 1) for compound 14 are (a) the
1
signals at 9.50 ppm in H NMR and at 190.6 in 13C NMR
which confirm the presence of the formyl group, (b) the
presence of only two methyl groups of acetyl groups at
2.12 and 2.14 ppm in 1H NMR, (c) the signals at 6.72
1
ppm in H NMR and at 141.6 ppm (CH) and 142.5 (C)
in 13C NMR which indicate the presence of a trisubsti-
tuted double bond, which must be formed by eliminating
an OAc group, and (d) the signals at 4.58 and 4.35 ppm
in 1H NMR with a geminal coupling constant (J ) 17
Hz), which indicate that these protons are in a carbon
bonded to oxygen. Therefore this carbon was assigned
as C1.
(10) For a recent report on hydroformylation of 2,5-dihydrofuran,
see: Horiuchi, T.; Ohta, T.; Shirakawa, E.; Nozaki, K.; Takaya, H. J .
Org. Chem. 1997, 62, 4285.
(11) van Leeuwen, P. W. N. M.; Robeck, C. F. J . Organomet. Chem.
1983, 258, 343.
(12) For a preliminary communication, see: Polo, A.; Ferna´ndez,
E.; Claver, C.; Castillo´n, S. J . Chem. Soc., Chem. Commun. 1992, 639.
(13) Ojima, I.; Kato, K.; Okabe, M.; Fuchikami, T. J . Am. Chem.
Soc. 1987, 109, 7714.
(14) Tolman, C. A. Chem. Rev. 1977, 77, 313.

Hydroformylation of Glucal Derivatives with Rh Catalysts
Organometallics, Vol. 17, No. 13, 1998 2859
Ta ble 1. Selected 1H a n d 13C NMR δ Da ta (in p p m ) of Com p ou n d s 14, 15, 16, a n d 22a a
1H NMR δ/(J )
13C NMR
CHO
compd
CHO
H1e
H1a
H2e
H2a
H3e
H3a
H4
14b
9.50 (s) 4.58 (ddd)
4.35 (ddd)
-
-
6.72 (q)
5.50 (dq)
(8.4, 2.7, 2.7, 2.2)
4.95 (dd)
(14, 11)
5.12 (ddd)
190.6
197.2
-
(17, 2.2, 2.2) (17, 2.7, 2.7)
(2.7, 2.7, 2.2)
- 5.22 (dd)
15
9.54 (d) 4.00-4.25
3.40-3.60
(m)
3.79 (dd)
(12, 3.3)
3.83 (dd)
(12.5, 5.2)
-
2.99 (ddt)
(17, 17, 8, 2.7)
-
(2.7)
(m)
(17, 14)
16c
22a
-
4.10 (d)
(12)
-
2.91 (td)
(5.4, 3.3, 3.3)
2.42 (td)
2.72 (dd) 1.16 (ddd)
(11, 4.4) (11, 10.3, 5.4) (10.3, 10, 4.4)
9.70 (s)
1.54 (dt)
-
201.2
(12.5, 5, 2.5) (12.5, 12.5, 11.5)
a
b
Sugar numbering has been used for C and H. See Scheme 3. Double-bond carbons appear at 141.6 and 142.5 ppm. c As
2,4-dinitrophenylhydrazone derivative.
Ta ble 2. Hyd r ofor m yla tion of 3,4,6-Tr i-O-a cetyl-D-glu ca l (1) w ith th e Ca ta lytic System
e
[Rh 2(µ-S(CH2)3NMe2)(COD)2]/P (O-o-tBu C6H4)3
entry no.
solvent
pressure (bar)
temp. (°C)
time (h)
conversiona (%)
14b
15b
16b
17b
1
2
3
4
5
6
7
8
9
Cl2C2H4
Cl2C2H4
Cl2C2H4
Cl2C2H4
Cl2C2H4
Cl2C2H4
Cl2C2H4
toluene
Cl2C2H4
75
75
60
60
40
90
60
60
60
120
120
130
110
120
120
120
120
120
3
24
24
24
24
24
24
24
24
16
84
94
93
38
68
72
80
15
17
34
34
30
16
33
31
40
4
14
6
8
-
12
15
12
13
6
27
31
26
27
32
30
trace
3
6
20
13
13
3
7
c
10
17
17
d
20
a
b
Percentage of transformed product. Percentage of the products identified, related to the products of the reaction detected by GC.
c PCO/PH
)
3/2. Auxiliary ligand P(o-MeC6H4)3. e Standard conditions: substrate (5 mmol), auxiliary ligand (0.5 mmol), precursor of
d
catalyst 2(0.05 mmol), solvent: 15 mL, PCO/PH ) 1/1.
2
a basic amino group in the catalyst. So, to increase the
selectivity we decided to use other nonbasic catalytic
precursors and to change the protecting groups.
Initially, we focused on the fact that all the experi-
ments performed with the catalytic system [Rh2(µ-
S(CH2)3NMe2)(COD)2]/P(O-o-tBuC6H4)3 in chlorinated
solvents afforded the mononuclear rhodium complex
trans-[RhCl(CO)(P(O-o-tBuC6H4)3)2] (18) at the end of
the reaction. This suggests that the active species is
mononuclear and probably similar to the isolated spe-
cies. On the other hand, it is known that the dinuclear
rhodium complex [Rh2(µ-OMe)2(COD)2] (19) easily gives
monuclear species under catalytic conditions. Moreover,
this catalytic precursor has no basic groups in its
structure, which makes it an attractive catalytic precur-
sor for the hydroformylation of glucal derivatives.
The catalytic precursor 19 was, in general, slightly
less active than 8 and required longer reaction times
(entries 1 and 2, Table 3). However, as far as selectivity
is concerned, the 2-formyl derivative 15 was obtained
in higher percentages than when catalytic system 8/P(O-
o-tBuC6H4)3 was used. The yield and selectivity were
not significantly affected by pressure changes (entries
3 and 4), but the percentage of elimination and hydro-
genation products were affected by changes in the
temperature and in the solvent (entries 5 and 6). One
important difference is that when precursor 19 was
used, compound 16 was not obtained.
The complex trans-[RhCl(CO)(P(O-o-tBuC6H4)3)2]9 (18)
was not a good precursor of the catalyst in the hydro-
formylation of 3,4,6-tri-O-acetyl-D-glucal, and the main
product observed after 48 h was the hydrogenated
product 17 (entry 8).
F igu r e 1. Evolution of the hydroformylation of 3,4,6-tri-
O-acetyl-D-glucal. (2) Conversion, (b) selectivity on 14, (9)
selectivity on 15, (O) selectivity on 16, (4) selectivity on
17 (in %).
(entry 9) showed that electronic factors of the auxiliary
ligand also affect the activity of the catalytic systems
in the hydroformylation of hindered olefins.
The very narrow range of conditions in which the
catalytic systems were active and the slight variations
observed in the yield and selectivity prevented us from
making a more extensive study. On the other hand, the
elimination product 14 and the aldehyde 16 may have
been obtained because of the presence of the acetates
as protecting group, which are good leaving groups, and
The presence of halogens in the reaction mixture is
known to inhibit the catalytic reaction.15 Et3N was
added to scavenge hydrogen chlorine (entry 9), but the

2860 Organometallics, Vol. 17, No. 13, 1998
Ferna´ndez et al.
Ta ble 3. Hyd r ofor m yla tion of 3,4,6-Tr i-O-a cetyl-D-glu ca l (1) w ith th e Ca ta lytic System [Rh 2(µ-OMe)(COD)2]
f
(19)/P (O-o-tBu C6H4)3
entry no.
solvent
pressure (bar)
temp. (°C)
time (h)
conversiona (%)
14b
15b
16b
17b
1
2
3
4
5
6
7
8
9
Cl2C2H4
Cl2C2H4
Cl2C2H4
Cl2C2H4
Cl2C2H4
toluene
Cl2C2H4
Cl2C2H4
Cl2C2H4
55
55
70
45
55
55
55
55
55
100
100
100
100
130
100
100
100
100
24
48
48
48
48
48
48
48
48
54
90
91
94
92
94
14
44
37
20
22
19
21
37
37
11
1
58
54
58
54
5
-
7
8
8
-
-
-
8
trace
trace
-
23
37
5
76
19
5
c
d
e
14
23
4
-
-
7
a
b
Percentage of transformed product. Percentage of the products identified, related to the products of the reaction detected by GC.
c Auxiliary ligand P(o-MeC6H4)3. Precursor of catalyst: [RhCl(CO)(P(O-o-tBuC6H4)3)2]. e Precursor of catalyst: [RhCl(CO)(P(O-o-
d
tBuC6H4)3)2] + 0.5 mmol of NEt3. f Standard conditions: substrate (5 mmol), auxiliary ligand (0.5 mmol), precursor of catalyst (0.05
mmol), solvent: 15 mL, PCO/PH ) 1/1.
2
Ta ble 4. Hyd r ofor m yla tion of
3,4,6-Tr i-O-ben zyl-D-glu ca l (20a ) a n d
compound 23a decreased to 39% (entry 3, Table 4). A
small amount of reaction mixture was purified by TLC,
and pure compound 22a was isolated. The most sig-
nificant spectroscopic NMR data (Table 1) which sup-
ported this structure were (a) an aldehydic proton at
3,4,6-Tr i-O-m eth yl-D-glu ca l (20b) w ith th e Ca ta lytic
c
System 19/P (O-o-tBu C6H4)3
entry no.
sub.
solvent
conv.a (%)
21b
22b
23b
1
9.70 ppm in the H NMR and at 201.2 ppm in the 13C
1
2
3
20a
20a
20b
Cl2C2H4
toluene
Cl2C2H4
99
99
99
9
9
11
10
10
16
68
39
55
NMR, (b) the presence of three benzyl groups, and (c)
the appearance of two signals at 2.42 and 1.54 ppm with
a
b
a
2J ) 12.5 Hz. If the δ value is taken into account,
Percentage of transformed product. Percentage of the prod-
ucts identified, related to the products of the reaction detected by
GC. c Standard conditions: substrate (5 mmol), auxiliary ligand
these signals can be assigned to a CH2 group which is
not bonded to oxygen (δ ) 31.3 ppm in 13C NMR), so
they are bonded to C2, and (d) the signal at 1.54 ppm
has two coupling constants of 11 Hz, which show that
the neighboring protons H1 and H3 are axial; therefore,
the formyl group must be equatorial in C1.
(0.5 mmol), precursor of catalyst (0.05 mmol), solvent 15 mL, PCO
PH ) 1/1, T ) 100 °C, P ) 50 bar, t ) 9 h.
/
2
Sch em e 4
The 2,4-dinitrophenylhydrazone derivatives of the
reaction mixure products were prepared and the hy-
drazone of the main product 23a was isolated. In the
1H NMR spectrum the hydrazone derivative of 23a has
a signal at 2.85 ppm with four coupling constants.
Double resonance experiments showed that two of these
were trans-diaxial, that one was an axial-equatorial
coupling, and that the fourth coupled with the imino
proton. This proton was assigned as H2, and so the
formyl group at position 2 is equatorial.
The hydroformylation of 3,4,6-tri-O-methyl-D-glucal
20b with the catalytic system 19/P(O-o-tBuC6H4)3 gave
a quantitative conversion in aldehydes after 9 h under
the conditions shown in Table 4. In this case also, the
main product was the 2-formyl derivative 23b of the
gluco configuration. Neither the aldehydes nor the 2,4-
dinitrophenylhydrazone derivatives of the reaction prod-
ucts could be purified by usual chromatographic tech-
niques.
hydrogenated product was still the main product ob-
tained. This proves that the complex 18 is not the active
species and that it is probably formed when the catalytic
reaction has finished.
Hyd r ofor m yla tion of 3,4,6-Tr i-O-ben zyl-D-glu ca l
(20a ) a n d 3,4,6-Tr i-O-m eth yl-D-glu ca l (20b) w ith
th e Ca ta lytic System [Rh 2(µ-OMe)(COD)2]/P (O-o-
tBu C6H4)3. The use of ether type protecting groups in
glucal prevents the secondary products 14 and 16 from
being formed. Thus, when 3,4,6-tri-O-benzyl-D-glucal
(20a ) was hydroformylated in 1,2-dichloroethane, using
the catalytic system 19/P(O-o-tBuC6H4)3 under the
conditions shown in Table 4, the conversion was com-
pleted in 9 h, and only aldehydes were obtained.
Compounds 21a , 22a , and 23a were identified (Scheme
4), but five aldehyde signals were detected. However,
the three products identified make up 87% of the total
products. Compound 23a , the 2-formyl derivative, was
obtained in the highest percentage, 68% (entry 1, Table
4).
Nevertheless, using GC-MS analysis, compounds with
the formyl group at C1 and C2 were differentiated.
While the mass spectra of two compounds show the loss
•
of the CHO and CO fragments from their molecular
ions, the mass spectrum of the major compound does
not. It has been described16 that substituents at C1 in
sugars are more easily removed under electron impact
conditions because it generates a stabilized carbocation.
This indicates that the two minor products are 21b and
22b, while the major one is 23b (Scheme 5).
When toluene was used as the solvent, the selectivity
after 3 h was similar to that obtained in 1,2-dichloro-
ethane, with a conversion of 65%. After 9 h the
conversion was complete. However, the percentage of
P r oblem of Regio- a n d Ster eoselectivity. In all
substrates and with all rhodium catalysts tested,
(16) DeJ ongh, D. C. In The Carbohydrates: Chemistry and Bio-
chemistry, 2nd ed.; Horton, D., Ed.; Academic Press: San Diego, 1970;
p 1327.
(15) Bayo´n, J . C.; Esteban, P.; Real, J .; Polo, A.; Ruiz, A.; Claver,
C.; Castillo´n, S. J . Organomet. Chem. 1991, 403, 393.

Hydroformylation of Glucal Derivatives with Rh Catalysts
Organometallics, Vol. 17, No. 13, 1998 2861
Sch em e 5
Sch em e 7
Sch em e 8
Sch em e 6
be because of the fast isomerization process from II2 to
II1 analogues and the faster formation of III1 species.
This isomerization process depends on two different
factors: the high temperature required to perform the
hydroformylation reaction (∼100 °C) and the auxiliary
ligand used. The auxiliary ligand can increase or
decrease the rate of â-elimination. This may be due to
the size of the phosphite which reduces the number of
ligands coordinated to the metal, thus creating the
vacant site necessary for the â-elimination.
The results obtained with the glucal derivatives in
the same conditions than for compound 9 confirm that
the II2 alkyl-metal species are principally formed and
that there is no â-elimination. This may be due to the
demands of the â-elimination mechanism, which needs
a “synperiplanar” arrangement of the metal and the
â-hydride. Therefore, there must be a conformational
change in the molecule, which is more difficult in
substituted rings such as glucals than in nonsubstituted
rings such as dihydropyran (Scheme 7).
The results of hydroformylating glucals using cobalt
catalysts show that the isomerization process is also fast
in this case. This may be because of the very high
reaction temperature (∼200 °C) needed and the shorter
cobalt-carbon bond which makes cobalt derivatives
more sensitive to steric hindrance.
2-formylcarbohydrates were the main derivatives ob-
tained when hydroformylating glucal derivatives, which
is in contrast to the regioselectivity observed when
cobalt catalysts were used5 and to the results obtained
when hydroformylating the closely related 3,4-dihydro-
2H-pyran and 2,3-dihydrofuran, which gave mainly
1-formyl9 derivatives.
In a previous paper,9 we showed that the regioselec-
tivity in the hydroformylation of dihydrofuran and
dihydropyran derivatives depended on the polarization
of the double bond, the relative stability of the alkyl-
metal complexes, the tendency to isomerize (i.e., the rate
of â-elimination), and the ratio between the rates of
formation of the acyl-metal intermediates.
On the other hand, glucals are chiral compounds and
the coordination of the double bond to the metal can
give two different adducts, IR and Iâ (Scheme 8). Glucal
4
derivatives mainly have a H5 conformation17 although
4
5
the conformers H5 and H4 may undergo a process of
equilibrium at the reaction temperature, which would
mean that the setereoselectivity is poor. However, the
results show that the hydroformylation reaction of
glucals is stereoselective and that the main product
obtained is the result of the attack from the R face
(adduct IR) of the glucal. Moreover, the stereoselectivity
is much better when the formyl group is in C2 than
when it is in C1.
We believe that for compound 9 (Scheme 6, R ) R′)
H), the alkyl-metal complex II2 was principally ob-
tained because the new metal-carbon bond was formed
at the carbon with the highest electron density.9,13
However, because the C1-metal bond in II1 is more
polarized than the C2-metal bond in II2, III1 forms from
CO
II1 faster than III2 from II2 (K1
> K2CO). In this
context, the reason the hydroformylation of 3,4-dihydro-
2H-pyran principally gives the 1-formyl derivative must
(17) Rico, M.; Santoro, J . J . Magn. Reson. 1976, 8, 49.

2862 Organometallics, Vol. 17, No. 13, 1998
Ferna´ndez et al.
The 1,2 addition of palladium complex and nucleo-
philes to glycals has been extensively studied by Doyle-
Daves and colleagues.18,19 When the starting material
was 3,4,6-tri-O-acetyl-D-glucal 1 the main product ob-
tained was also the result of the attack on the R face at
C2. In general, the stereoselectivity observed means
that the palladium attacks the face which is opposite
to the substituent at position 3. The fact that the
stereoselectivity is high is due to stereoelectronic effects
with a πox---π*CdC---σ*3-OR (vinylogous effect) interac-
tion, which force the substituent at position 3 to adopt
a pseudoaxial arrangement.18
The stereoselectivity in the hydroformylation of glu-
cals with a rhodium catalyst, although smaller than in
the reaction with palladium, can be explained in a
similar way. There are practically no products with the
2-formyl axial group (manno configuration) because of
the instability of the axial σ-alkyl-metal intermediate,
which strongly interacts with the substituent at position
3 and with other substituents in the ring.
F igu r e 2. Molecular structure of the complex trans-[RhCl-
(CO)(P(O-o-tBuC6H4)3)2] (18).
In conclusion, the catalytic system [Rh2(µ-OMe)2-
(COD)2]/P(O-o-tBuC6H4)3 hydroformylates glucal deriva-
tives in good yields and selectivities. The stereochem-
istry of the substituent at position C3 of the glucal seems
to be responsible for controlling the stereoselectivity of
the reaction. The regioselectivity depends mainly on
the polarization of the olefin and on the difficulty of
producing the â-elimination process in conformationally
rigid substrates.
tion conditions. In our work the complex 18 was
isolated at the end of the hydroformylation reaction. The
presence of only one phosphite coordinated to the
rhodium center in RhH(CO)3P(O-o-tBuC6H4)3
is at-
21,22
tributed to the large cone angle of the tris-ortho-tert-
butylphenyl phosphite (175°).14 This behavior is differ-
ent from triphenylphosphine rhodium systems in which
RhH(CO)2(PPh3)2 is considered to be the main species
during the hydroformylation reaction, in rapid equilib-
rium via CO and PPh3 dissociation steps with other
related species.23,24
The isolation of trans-[RhCl(CO)(P(O-o-tBuC6H4)3)2]
shows the cleavage of the dinuclear [Rh2(µ-S(CH2)3-
NMe2)(COD)2] and [Rh(µ-OMe)(COD)]2 complexes, and
the exchange of the hydride in the [RhH(CO)(PR3)2]
species with chloride in the solvent.
X-r a y Str u ctu r e of tr a n s-[Rh Cl(CO)(P (O-o-tBu -
C6H4)3)2] (18). To confirm the characterization of trans-
[RhCl(CO)(P(O-o-tBuC6H4)3)2], we carried out a separate
synthesis using [Rh(µ-Cl)(COD)]2 as starting material,
using a methodology previously described for related
compounds,25 and we obtained a crystal suitable for an
X-ray structure determination. It should be noted that
although several complexes with a trans-[RhCl(CO)-
(P(OR)3)2] molecular structure have previously been
synthesized (R ) Me,20 Ph,20,26,27 p-MePh,27 p-ClPh27),
none of these compounds has yet been structurally
determined by X-ray diffraction, although X-ray struc-
tures of related phosphine complexes have been exten-
sively reported.
The structure of trans-[RhCl(CO)(P(O-o-tBuC6H4)3)2]
consists of discrete mononuclear units in which the
geometry at rhodium is square planar (Figure 2). The
rhodium is placed on a crystallographic inversion center,
and a CO/Cl structural disorder is present. The pres-
ence of CO was unequivocally determined by IR spec-
Isola tion a n d Ch a r a cter iza tion of Rh od iu m Sp e-
cies a t th e En d of th e Ca ta lytic Rea ction . When
the catalytic systems 8/P(O-o-tBuC6H4)3 and 19/P(O-o-
tBuC6H4)3 are used as the hydroformylation catalyst,
the mononuclear species trans-[RhCl(CO)(P(O-o-t-
BuC6H4)3)2]9 (18) can be isolated at the end of the
catalytic reaction. The evaporation of the solvent and
the extraction with methanol of the reaction products
lead to the isolation of this species as a pale yellow
crystalline solid which was characterized by elemental
1
microanalysis, IR spectroscopy, and 31P and H NMR.
The 31P{1H} NMR spectrum shows only one doublet
centered at 112.0 ppm, attributed to the two equivalent
phosphorus atoms, with a relative trans conformation.
The high coupling constant (J P-Rh ) 213 Hz) is in the
range reported for analogous rhodium complexes trans-
[RhCl(CO)(P(OR)3)2] (R ) Me, J P-Rh ) 195.0 Hz; R )
Ph, J P-Rh ) 217.4 Hz).20
In the carbonyl region, the infrared spectrum shows
one ν(CO) frequency at 2012.5 cm-1 (s) in dichlo-
romethane solution and at 2002.5 cm-1 in the solid state
of the isolated product, according to the trans geometry
of complex 18. These values for ν(CO) frequences are
in the range for other analogous rhodium complexes
reported.20
Previous mechanistic studies by van Leeuwen et
al.21,22 on the use of [Rh]/P(O-o-tBuC6H4)3 systems have
shown the presence of RhH(CO)3P species, P ) tris-
ortho-tert-butylphenyl phosphite, under hydroformyla-
(23) Pruett, R. Adv. Organomet. Chem. 1979, 17.
(24) (a) Brown J . M.; Kent, A. G. J . Chem. Soc., Perkins Trans. 2
1987, 1597. (b) Moser, W. R.; Papile, C. J .; Brannon, D. A.; Duwell, R.
A. J . Mol. Catal. 1987,41, 271.
(25) (a) Vallarino, L. J . Chem. Soc. 1957, 2287. (b) Hieber, W.;
Hensinger, H.; Vohler, O. Chem. Ber. 1957, 90, 2425.
(26) Gracheva, L. S.; Yurchenko, E. N.; Triotskaya, A. D. Koord.
Khim. 1977, 3, 1718.
(18) Doyle-Daves, G.; Hallber, A. Chem. Rev. 1989, 89, 1433.
(19) Arai, I.; Doyle-Daves, G. J . Am. Chem. Soc. 1981, 103, 7683.
(20) Liwu, M.; Desmond, M. J .; Drago, R. S. Inorg. Chem. 1979, 18,
679.
(21) J ongsma, T.; Challa, G.; van Leeuwen, P. W. N. M. J . Orga-
nomet. Chem. 1991, 421, 121.
(22) van Rooy, A.; Orij, E. N.; Kamer, P. C. J .; van Leeuven, P. W.
N. M. Organometallics 1995, 14, 34.
(27) Vallarino, L. J . Chem. Soc. 1957, 2473.

Hydroformylation of Glucal Derivatives with Rh Catalysts
Organometallics, Vol. 17, No. 13, 1998 2863
Ta ble 5. Selected Bon d Len gth s (Å) a n d An gles
(d eg) w ith Estim a ted Sta n d a r d Der iva tion s in
P a r en th eses for 18
After each run, the system was allowed to cool and the gas
vented. The reaction mixture was removed from the autoclave
and immediately analyzed by FT-IR spectroscopy to determine
the metal carbonyl species and by gas chromatography and
1H NMR spectroscopy to determine the conversion and the
selectivity of the reaction.
Hyd r ofor m yla tion of 3,4,6-Tr i-O-a cetyl-D-glu ca l. 3,4,6-
Tri-O-acetyl-D-glucal was hydroformylated following the gen-
eral procedure. The resulting reaction mixture was evaporated
to dryness and purified by flash chromatography (ethyl
acetate/hexane 2:1) so obtaining the pure compounds 14, 15,
and 17. Pure compound 16 was recovered as its 2,4-dinitro-
phenylhydrazone derivative.
Rh-Cl
2.370 (3)
1.797 (6)
2.2856 (7)
1.145(7)
1.592 (2)
P-O(20)
1.600(2)
1.603(2)
1.384(2)
1.378(2)
1.369(3)
Rh-C(1)
Rh-P
P-O(30)
O(10)-C(11)
O(20)-C(21)
O(30)-C(31)
C(1)-O(1)
P-O(10)
C(1)-Rh-P
Cl-Rh-P
O(10)-P-Rh
O(20)-P-Rh
89.7 (4)
90.70 (8)
114.49 (7)
118.65 (8)
O(30)-P-Rh
117.76 (8)
104.11 (11)
101.54 (11)
97.51 (8)
O(10)-P-O(20)
O(10)-P-O(30)
O(20)-P-O(30)
troscopy. Table 5 shows the most significant intramo-
lecular distances and bond angles with their standard
deviations.
The bond lengths and angles are similar to those
reported previously for trans-[Rh(Cl)(CO)](PPh3)2]28 and
trans-[Rh(Cl)(CO)](PtBu3)2].29 The Rh-P distance
(2.287(2) Å) for 18 is slightly shorter than the related
Rh-P distance in the case of trans-[Rh(Cl)(CO)](P-
Ph3)2],28 average (2.323(6) Å).
4,6-Di-O-a cetyl-1,5-a n h yd r o-2,3-d id eoxy-2-C-for m yl-D-
2-en er yth r itol (14). 1H NMR (CDCl3): δ 2.12 (s, 3H,
OCOCH3), 2.14 (s, 3H, OCOCH3), 3.70 (ddd, J ) 8.4, 5.4, 2.7,
1H, H5), 4.21 (dd, J ) 12, 5.4, 1H, H6), 4.29 (dd, J ) 12, 2.7,
1H, H6′), 4.35 (ddd, J ) 17, 2.7, 2.7 1H, H1a), 4.58 (ddd, J )
17, 2.2, 2.2, 1H, H1e), 5.50 (dq, J ) 8.4, 2.7, 2.7, 2.2, 1H, H4),
6.72 (ddd, J ) 2.7, 2.7, 2.2, 1H, H3), 9.50 (s, 1H, CHO). 13C
NMR (CDCl3): δ 20.6, 20.7, 62.8, 63.4, 65.2, 73.9, 141.6, 142.5,
169.9, 170.6, 190.6. Anal. Found: C, 53.52; H, 5.06. Calcd:
C, 53.77; H, 5.39.
3,4,6-Tr i-O-a cet yl-1,5-a n h yd r o-2-d eoxy-2-C-for m yl-D-
glu citol (15). 1H NMR (CDCl3): δ 1.98 (s, 3H, OCOCH3), 1.99
(s, 3H, OCOCH3), 2.11 (s, 3H, OCOCH3), 2.99 (ddt, J ) 17,
17, 8, 2.7, 1H, H2), 3.40-3.60 (m, 1H, H1a), 4.00-4.25 (m, 4H,
Exp er im en ta l Section
Gen er a l Com m en ts. All syntheses of rhodium complexes
were performed using standard Schlenk techniques under a
nitrogen atmosphere. Solvents were dried by standard meth-
ods and distilled under nitrogen immediately prior to use. The
starting materials [{Rh(µ-Cl)(COD)}2],30 [{Rh(µ-OMe)(COD)}2],31
H
1e, H5, H6, H6′), 4.95 (dd, J ) 14, 11, 1H, H4), 5.22 (dd, J )
17, 14, 1H, H3), 9.54 (d, J ) 2.7, 1H, CHO). 13C NMR
(CDCl3): δ 20.6, 20.7, 53.8, 62.3, 64.8, 68.3, 70.8, 78.9, 169.7,
170.1, 171.0, 197.2.
11
and P(O-o-tBuC6H4)3 were prepared according to reported
4,6-Di-O-a cetyl-1,5-a n h yd r o-2,3-d id eoxy-2-C-for m yl-D-
a r a bin itol (16) (a s 2,4-Din itr op h en ylh yd r a zon e Der iva -
tive). To a solution of the mixture of hydroformylated
compounds and acetic acid (three drops) in ethanol was added
dropwise a saturated solution of 2,4-dinitrophenylhydrazine
in ethanol until no color change was observed. When the
reaction had finished, water was added until it became turbid
and the solution was first left at room temperature and then
cooled to 5 °C. The solid formed was filtered off and vacuum-
dried. The mixture of hydrazone obtained was purified by TLC
(ethyl acetate/hexane 1:1) and the pure compound 16 was
recovered as its 2,4-dinitrophenylhydrazone derivative. 1H
NMR (CDCl3): δ 1.16 (ddd, J ) 11, 10.3, 5.4, 1H, H3a), 2.00 (s,
3H, OCOCH3), 2.07 (s, 3H, OCOCH3), 2.72 (dd, J ) 11, 4.4,
1H, H3e), 2.91 (td, J ) 5.4, 3.3, 3.3, 1H, H2), 3.59 (dt, J ) 10,
5.3, 5.3, 1H, H5), 3.79 (dd, J ) 12, 3.3, 1H, H1a), 4.10 (d, J )
12, 1H, H1e), 4.10 (d, J ) 5.3, 2H, H6, H6′), 5.12 (ddd, J ) 10.3,
10, 4.4, 1H, H4), 7.57 (d, J ) 3.3, 1H, CH)), 8.20-9.10 (m,
5H, Ar), 11.12 (bs, 1H, NH).
3,4,6-Tr i-O-a cetyl-1,5-a n h yd r o-2-d eoxy-D-h exoa r a bin i-
tol (17). 1H NMR (CDCl3), δ 1.82 (dq, J ) 12.5, 12.5, 12.5, 5,
1H, H2a), 2.03 (s, 3H, OCOCH3), 2.04 (s, 3H, OCOCH3), 2.07
(d, J ) 12.5, 1H, H2e), 2.09 (s, 3H, OCOCH3), 3.44-3.48 (m,
2H, H1a, H1e), 4.05 (ddd, J ) 11.5, 5, 2, 1H, H5), 4.10 (dd, J )
12.5, 2, 1H, H6′), 4.23 (dd, J ) 12.5, 5, 1H, H6), 4.94-5.00 (m,
2H, H3, H4). 13C NMR (CDCl3): δ 20.9, 21.0, 21.1, 31.2, 63.6,
65.7, 69.5, 72.6, 76.8, 170.5, 171.1, 171.5.
methods. RhCl3‚xH2O and phosphorus reactants were com-
mercial samples and were used without further purification.
Infrared spectra (range 4000-400 cm-1) were recorded on a
Nicolet 5ZDX-FT spectrophotometer in CH2Cl2 solutions or in
KBr pellets. Elemental analyses were carried out on a Carlo-
Erba microanalyzer, and fast atom bombardment mass spectra
were obtained on a VG Autospect in a 3-nitrobenzyl alcohol
matrix. 1H and 13C NMR spectra were recorded on a Varian
Gemini 300 spectrophotometer and chemical shifts are quoted
in ppm downfield from internal TMS. 31P NMR spectra were
obtained on the same instrument at 120 MHz, using external
85% H3PO4 as reference. Mass spectrometry was performed
on a Hewlett-Packard CG/MS 5988A spectrometer, using an
Ultra-2 (diphenylsilicone 5%, dimethylsilicone 95%) 25 m ×
0.2 mm column. Gas chromatography was performed on a
Hewlett-Packard 5890 II chromatograph equipped with the
column mentioned above. Flash chromatography was per-
formed on silica gel 60 A CC. Solvents for chromatography
were distilled at atmospheric pressure prior to use.
Ca t a lysis. High-pressure hydroformylation experiments
(50-75 bar) were carried out in a Berghof autoclave, and were
magnetically stirred and electrically heated. These experi-
ments were not performed at constant pressure, but for the
amount of substrate used the pressure drop was never more
than 3 bar.
Gen er a l P r oced u r e. In a standard experiment, a solution
of the substrate (5 mmol), the catalyst (0.05 mmol), and the
phophorous cocatalyst (0.5 mmol) in 15 mL of the solvent was
placed in the evacuated autoclave and heated while stirring.
Once the system reached thermal equilibrium the gas mixture
was introduced to reach the working pressure. Small samples
of the catalytic solution were taken at intervals for analysis.
Hyd r ofor m yla tion of Tr i-O-ben zyl-D-glu ca l. 4,5,7-Tr i-
O-ben zyl-2,6-a n h yd r o-D-glu coh ep top yr a n ose (22a ). 1H
NMR (CDCl3): δ 1.54 (dt, J ) 12.5, 12.5, 11.5, 1H, H2a), 2.42
(td, J ) 12.5, 5, 2.5, 1H, H2e), 3.83 (dd, J ) 12.5, 2.5, 1H, H1),
3.60-4.00 (m, 3H, H5, H6, H6′), 4.60-4.80 (m, 8H, CH2Ph, H3,
H4), 7.10-7.60 (m, 15H, Ph), 9.70 (s, 1H, CHO). 13C NMR
(CDCl3), δ 31.4, 69.2, 71.4, 73.4, 73.5, 75.1, 79.1, 79.4, 80.2,
127.0-128.7, 137.0-139.0, 201.2.
(28) Cerotti, A.; Ciani, G.; Sironi, A. J . Organomet. Chem. 1983, 247,
345.
(29) Schumann, H.; Heisler, M.; Pickardt, J . Chem. Ber. 1977, 110,
1020.
(30) Chatt, J .; Venanzi, L. M. Nature 1956, 177, 852.
(31) (a) Vizi-Orosz, A.; Palyi, G.; Marko, L. J . Organomet. Chem.
1973, 57, 379. (b) Giordano, G.; Martinengo, S.; Strumdo, D.; Chini,
P. Gazz. Chim. Ital. 1975, 105, 613.
3,4,6-Tr i-O-b en zyl-1,5-a n h yd r o-2-d eoxy-2-C-for m yl-D-
glu citol (23a ) (a s 2,4-Din itr op h en ylh yd r a zon e Der iva -
tive, see Meth od ology for Der iva tiza tion of 16). 1H NMR
(CDCl3), δ 2.85 (ddd, J ) 10.5, 10.3, 5.3, 1H, H2), 3.44 (dd, J
) 11, 10.5, 1H, H1a), 3.45 (m, 2H, H4, H5), 3.66 (t, J ) 10.3,

2864 Organometallics, Vol. 17, No. 13, 1998
Ferna´ndez et al.
Ta ble 6. Cr ysta l Da ta a n d X-r a y Exp er im en ta l
Deta ils for 18
1H, H3), 3.72 (m, 2H, H6, H6′), 4.08 (dd, J ) 11, 5.3, 1H, H1e),
4.50-4.90 (m, 6H, CH2Ph), 6.85 (d, J ) 5.3, 1H, CHdN), 7.10-
9.10 (m, 15H, Ph), 10.60 (s, 1H, NH). 13C NMR (CDCl3): δ
45.9, 67.7, 68.7, 75.5, 74.8, 75.1, 79.3, 79.7, 82.1, 116.5, 123.5,
127.7, 128.7, 129.0, 130.1, 137.8, 137.9, 138.0, 138.1, 144.7,
148.7.
empirical formula
fw
C61H78ClO7P2Rh
1123.53
wavelength
0.71069 Å
monoclinic
C2/c
crystal system
space group
unit cell dimensions
P r ep a r a tion of [Rh Cl(CO)(P (O-o-tBu C6H4)3)2] (18). A
stream of CO was bubbled through a solution of [Rh2(µ-Cl)2-
(cod)2] (40 mg, 0.1 mmol) in dichloromethane (5 mL) and then
tris-(o-tert-buylphenyl)phosphite (95.6 mg, 0.2 mmol) was
added. A brisk gas evolution and a change in color to pale
yellow were observed. After 30 min, the solution was concen-
trated and ethanol was added and a yellow crystalline solid
precipitated, which was filtered off, washed with cold ethanol,
and vacuum-dried to give a yellow crystalline compound. Yield
85% (133.4 mg). Elemental analysis: Found: C, 65.2; H, 6.9.
a ) 12.150(3) Å
b ) 24.457(2) Å
c ) 20.045(5) Å
R ) 90°
â ) 96.80(2)°
γ ) 90°
volume
Z
density (calcd)
abs coeff
F(000)
5914(2) Å3
4
1.262 Mg/m3
0.437 mm-1
2368
Calcd for C61H78O7P2ClRh: C, 65.6; H, 7.1. IR (ν(CO) cm-1
)
crystal size
θ range for data collection
0.5 × 0.4 × 0.2 mm
in CH2Cl2 solution, 2012.5; in KBr, 2002.5. 1H NMR (CDCl3):
δ 1.40(s, 54H, CH3), 6.89 (t, J ) 8.2, 8.1 Hz, 6H, Ar), 6.96 (t,
J ) 8.2, 8.1 Hz, 6H, Ar), 7.31 (d, J ) 8.1 Hz, 6H, Ar), 7.70 (d,
J ) 8.1 Hz, 6H, Ar). 13C NMR (CDCl3): δ 30.3, 35.0, 120.6,
123.9, 126.6, 127.3, 139.2, 150.4. 31P NMR: δ 112.0 (d, J P-Rh
) 213 Hz).
1.67-24.97°
-14 e h e 14
0 e k e 28
index ranges
0 e l e 23
no. of indep reflcns
5202
no. of obsd reflcns
refinement method
data/restraints/params
goodness-of-fit on F2
final R indicesa (F > 4σ(F))
3936
Cr ysta l Str u ctu r e Deter m in a tion of 18. A suitable
yellow single crystal of 18 was obtained by slow evaporation
from a CH2Cl2/EtOH solution. The crystal was mounted in
an Enraf-Nonius CAD4 automatic diffractometer with graphite-
monochromated Mo KR radiation (λ ) 0.710 69). Crystal-
lographic and experimental details are summarized in Table
6. Data were collected at room temperature. Intensities were
corrected for Lorentz and polarization effects. Empirical
absorption correction32 from ψ-scans was applied. The struc-
ture was solved by direct methods (SHELXS86)33 and refined
by least squares on F2 for all reflections (SHELXL9334). The
asymmetric unit is half a molecule, the rhodium is placed in
a crystallographic inversion center. On the basis of chemical
and spectroscopic grounds, the rhodium is bonded to a carbonyl
group and a chlorine atom, so the complex is not centrosym-
metric and the chlorine and the CO group show positional
disorder. The refinement was carried out using a restrained
geometry affecting the Cl-Rh-CO moiety (dRh-C1 ) 1.85(1),
dRh-O1 ) 3.00(1), dC1-O1 ) 1.15(1)). All non-H atoms were
refined anisotropically, and the H-atoms were refined using a
riding model with thermal parameters fixed at 1.5 (methyl H)
or 1.2 (the rest) times the Uiso of the corresponding carbon
atoms.
full-matrix least-squares on F2
5202/39/304
1.107
R(F) ) 0.039
Rw(F2) ) 0.115
R(F) ) 0.057
Rw(F2) ) 0.118
w ) 1/[σ2(Fo2) + (0.075P)2,
final R indices (all data)
weighting scheme
where P ) (Fo + 2Fc2)/3
2
largest diff peak and hole
0.39 and -0.43 e/Å3
R(F) ) ∑||Fo| - |Fc||∑/Fo|; Rw(F2) ) [∑[w(Fo - Fc2)2]/
a
2
∑[w(Fo2)2]]0.5
.
Ack n ow led gm en t. This research was supported by
DGR (Direccio´ General de Recerca, de la Generalitat de
Catalunya), Grants QFN95-4725-C03-2 and 1995DGR
00528.
Su p p or tin g In for m a tion Ava ila ble: ORTEP diagram of
the molecular structure with full numbering scheme and tables
of the atomic coordinates of heavy atoms and bond distances
and angles for [RhCl(CO)(P(O-o-tBuC6H4)3)2] are available (13
pages). Ordering information is given on any current mast-
head page.
(32) North, A. C. T.; Phillips, D. C.; Mathews, F. S. Acta Crystallogr.
1968 A24, 351-359.
(33) Sheldrick, G. M. Acta Crystallogr. 1990 A46, 467-473.
(34) Sheldrick, G. M. SHELXL93. Program for the Refinement of
Crystal Structures; University of Go¨ttingen: Germany, 1993.
OM971010K