Revision of stereochemical assignments of (2,2-dimethyl-5-phenyl-[1,3] dioxolan-4-yl)-methanol

Article abstract of DOI:10.1016/j.tetasy.2003.10.026

Opposite signs of specific rotation are reported in the literature for (4S,5S)-(2,2-dimethyl-5-phenyl-[1,3]dioxolan-4-yl)-methanol. The correct assignment was made by unambiguous synthesis of the title compound via the reaction of (S)-O-TBS-mandelic aldeh

Full text of DOI:10.1016/j.tetasy.2003.10.026

Tetrahedron:
Asymmetry
Tetrahedron: Asymmetry 15 (2004) 289–292
Revision of stereochemical assignments of (2,2-dimethyl-
5-phenyl-[1,3]dioxolan-4-yl)-methanol
Marco Lombardo,* Sebastiano Licciulli and Claudio Trombini*
Dipartimento di Chimica ‘G. Ciamician’, Universitꢀa di Bologna, via Selmi 2, I-40126 Bologna, Italy
Received 1 October 2003; accepted 27 October 2003
Abstract—Opposite signs of specific rotation are reported in the literature for (4S,5S)-(2,2-dimethyl-5-phenyl-[1,3]dioxolan-4-yl)-
methanol. The correct assignment was made by unambiguous synthesis of the title compound via the reaction of (S)-O-TBS-
mandelic aldehyde with vinyl magnesium chloride, separation of diastereomeric (1S,2S)- and (1S,2R)-1-phenyl-but-3-ene-1,2-diols,
and transformation of (1S,2S)-diol into authentic (4S,5S)-(2,2-dimethyl-5-phenyl-[1,3]dioxolan-4-yl)-methanol.
Ó 2003 Elsevier Ltd. All rights reserved.
1. Introduction
version2 of the Nozaki–Hiyama–Kishi reaction,3
a
heterosubstituted allylic chromium(III) is formed, which
adds to benzaldehyde to give enantiomerically enriched
syn-2 and anti-2 in the 71:29 ratio.1 Enantiomeric
excesses of syn-2 (64%) and anti-2 (43%) were measured,
after removal of pivaloate and trimethylsilyl groups
(LiAlH4, THF) and derivatisation of syn-3 and anti-3
diols as dioxolanes trans-4 and cis-4, respectively, by
chiral GC analysis using a cyclodextrin Megadex 5
column. The overall process is summarised in Scheme 1.
In the field of asymmetric synthesis, the assignment of
absolute configurations to newly formed stereogenic
centres is often based, besides on physical methods, on
direct chemical correlation: A chiral molecule produced
in its enantiopure or enantioenriched form is trans-
formed into one or the other of the antipodes of a ref-
erence substance by using reactions, which do not affect
the stereogenic centres. If the specific rotation of the ref-
erence molecule is known, the sign of rotation represents
an experimental criterion for the assignment of absolute
configurations. However risks are associated with the
use of a simple polarimetric analysis, particularly when
the sign of rotation shows a strong solvent, concentra-
tion, wavelength or temperature dependence, when
specific rotation is too small and a trace impurity with a
high specific rotation may overwhelm the sample rota-
tion, and, finally, when an incorrect assignment is
reported in the literature.
A representative case recently occurred to us while
developing an asymmetric catalytic entry to alk-1-ene-3,
4-diols.1
2. Results and discussion
When 3-chloro propenyl pivaloate 1 is reacted with an
€
in situ generated Salen/Cr(II) species in the Furstner
* Corresponding authors. Tel.: +39-051-2099544; fax: +39-051-2099-
456 (M.L.); tel.: +39-051-2099513; fax: +39-051-2099456 (C.T.);
e-mail addresses: marlom@ciam.unibo.it; trombini@ciam.unibo.it
Scheme 1. Reagents and conditions: (i) CrCl2 (10%)/Mn, (R,R)-Salen
(20%), Et3N (40%), TMSCl/CH3CN; (ii) LiAlH4, THF; (iii)
Me2C(OMe)2, CH2Cl2, Hþ; (iv) O3, CH2Cl2 then LiAlH4, THF.
0957-4166/$ - see front matter Ó 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2003.10.026

290
M. Lombardo et al. / Tetrahedron: Asymmetry 15 (2004) 289–292
Since the specific rotations for both enantiomers of anti-
3 are known,4 we were able to unambiguously assign the
(1S,2R) absolute configuration to the major isomer of
anti-2 obtained by the (R,R)-Salen/Cr(II) catalysed re-
action. Unfortunately, no data for syn-3 diols were
found in the literature.
As even specific rotations of cis-4 and trans-4 are not
known, in order to unequivocally determine the absolute
configurations and the sense of asymmetric induction of
the above reported Salen/Cr(II)-based catalytic cycle, we
subjected 4 to reductive ozonolysis (Scheme 1), since
specific rotations of the four steroisomers of 5 (Fig. 1)
are known (Table 1).5–10
Scheme 2. Reagents and conditions: (i) TBAF, THF; (ii)
Me2C(OMe)2, Hþ; (iii) O3 then LiAlH4.
chloride to give alcohols (1S,2R)-anti-6 and (1S,2S)-syn-
6 in 76% overall yield and in the 80:20 anti/syn ratio.
Protodesilylation of 6 (TBAF) afforded (1S,2R)-anti-3
and (1S,2S)-syn-3, respectively; anti-3 and syn-3 were
identified by comparison of spectroscopic data with
authentic racemic compounds synthesised in a previous
work.11 Protection of 3 as the acetonides (4S,5R)-cis-4
and (4S,5S)-trans-4 followed by reductive ozonisation,
allowed us to isolate pure (4R,5S)-cis-5 and (4S,5S)-
trans-5.
Figure 1.
Table 1. Specific rotation values of the four stereoisomers of acetonide
5
20
D
Stereoisomer
½aꢀ
Reference
(4S,5S)-trans-5
)19.5 (c 3.1, CH2Cl2)
+24 (c 0.36, CH2Cl2)
+32.7 (c 1.2, EtOH)
+19.4 (c 3.0, CH2Cl2)
+19.5 (c 2.9, CH2Cl2)
5
6
7
5
8
20
Authentic (4S,5S)-trans-5 shows an ½aꢀ +15.6 (c 0.36,
D
20
D
CH2Cl2) and ½aꢀ +14.7 (c 3.05, CH2Cl2); thus, we
(4R,5R)-trans-5
established that (i) (4S,5S)-(2,2-dimethyl-5-phenyl-
[1,3]dioxolan-4-yl)-methanol 5 does correspond to the
dextrorotatory enantiomer, and (ii) consequently, syn-2
coming from the (R,R)-Salen/Cr(II)-catalysed reaction
does correspond to the (1S,2S) isomer.
(4R,5S)-cis-5
(4S,5R)-cis-5
+80 (c 2.73, CH2Cl2)
)84 (c 3.7, MeOH)
)112.3 (c 1.3, CHCl3)
)112 (c 1.2, CHCl3)
6
9
7
10
For the sake of clarity, all the specific rotations of
products 3–6 are collected in Table 2.
Unfortunately, as apparent from data reported in Table
1, no unambiguous stereoassignment was possible for
the major isomer of trans-5 obtained by us, which shows
Table 2. Specific rotation values of products 3–6
20
a
D
20
Stereoisomer
½aꢀ
½aꢀ +20 (c 0.8, CHCl3), since opposite specific rotation
D
signs are reported in the literature.5–8
(1S,2R)-anti-6
(1S,2S)-syn-6
(1S,2R)-anti-3
(1S,2S)-syn-3
(4S,5R)-cis-4
(4S,5S)-trans-4
(4R,5S)-cis-5
(4S,5S)-trans-5
+66 (c 1.46, CHCl3)
+43 (c 1.44, CHCl3)
+73 (c 1.26, CHCl3)
+2 (c 0.71, CHCl3)
+43 (c 1.56, CHCl3)
+36 (c 1.61, CHCl3)
+96 (c 2.89, CH2Cl2)
+16 (c 0.36, CH2Cl2)
+15 (c 3.05, CH2Cl2)
+19 (c 0.81, CHCl3)
If data reported by Refs. 5 and 8 were correct, the major
syn-product in our reaction should correspond to the
(1R,2R)-diol. These results should imply a stereodiver-
gent p-face selectivity of the intermediate allyl Cr(III)
species, which preferentially should attack either the
aldehyde re face when the syn-selective reaction channel
is followed, or the si face, when anti-adducts are formed.
Conversely, if data reported by Refs. 6 and 7 were
correct, the major syn-diol in our reaction should cor-
respond to the (1S,2S)-isomer.
a The relative errors were in the range 3–5% (n ¼ 10).
To definitely ascertain the right assignments, we carried
out a synthesis of authentic (4R,5S)-cis-5 and (4S,5S-
trans)-5, as outlined in Scheme 2.
We believe that correcting literature stereoassignments
has the major importance of avoiding later misunder-
standings of the stereochemical outcome of an asym-
metric transformation,12 as would happen if a wrong
specific rotation value were used to identify stereo-
products.
(S)-Mandelic aldehyde, protected as tert-butyldimeth-
ylsilyl (TBS) ether, is reacted with vinyl magnesium

M. Lombardo et al. / Tetrahedron: Asymmetry 15 (2004) 289–292
291
3. Experimental
3.3. (1S,2S)-1-Phenyl-but-3-ene-1,2-diol syn-3
The purity of all title compounds was established to
1
The analogous reaction of TBAF with (1S,2S)-syn-6
(0.153 g, 0.55 mmol), afforded (1S,2S)-syn-3 in 95% yield
(0.087 g, 0.52 mmol). 1H NMR (200 MHz, CDCl3) d
4.27 (ddt, J ¼ 1:5=5:5=6:9 Hz, 1H), 4.52 (d, J ¼ 6:9 Hz,
1H), 5.16 (dt, J ¼ 1:4=10:6 Hz, 1H), 5.25 (dt,
J ¼ 1:4=17:3 Hz, 1H), 5.75 (ddd, J ¼ 5.3/10.4/17.3 Hz,
1H), 7.26–7.40 (m, 5H); 13C NMR (75 MHz, CDCl3) d
76.9, 77.5, 116.9, 127.0, 128.0, 128.3, 136.3, 140.2.
be >97% by inspection of H and 13C NMR spectra,
GC–MS analyses and elemental analysis ( 0.4% for C
and H). Optical rotations were measured on a Perkin–
Elmer 241 polarimeter, using a 10 mm cell path
length.
3.1. 1-Phenyl-1-(tert-butyldimethylsilyl)-but-3-ene-2-ol 6
Vinylmagnesium chloride (1.0 M in THF, 6 mL, 6 mmol)
was added at 0 °C to a solution of (S)-O-tert-butyldi-
methylsilyl mandelic aldehyde (1.0 g, 4 mmol) in THF
(8 mL), and the reaction mixture stirred at 0 °C for 3 h.
The reaction was quenched by the addition of aq
NaHCO3 and the organic layer extracted with ether.
The combined organic layers were dried over Na2SO4
and evaporated at reduced pressure. Purification by
flash chromatography on SiO2 (cyclohexane/ethyl ace-
tate 95:5) afforded 0.167 g of syn-6 (0.60 mmol, 15%) and
0.675 g of anti-6 (2.43 mmol, 61%).
3.4. (4S,5R)-2,2-Dimethyl-4-phenyl-5-ethenyl[1,3]dioxo-
lane cis-4
2,2-Dimethoxy-propane (0.17 mL, 1.4 mmol) was added
at rt to a solution of (1S,2R)-anti-3 (0.113 g, 0.69 mmol)
in CH2Cl2 (3 mL), in the presence of a catalytic amount
of Amberlystâ-15H. The reaction mixture was stirred at
rt overnight, filtered through a small pad of Celiteâ and
the organic layer evaporated at reduced pressure. Puri-
fication by flash chromatography on SiO2 (cyclohexane/
ethyl acetate 95:5) afforded 0.138 g of (4S,5R)-cis-4
1
(0.68 mmol, 99%). H NMR (200 MHz, CDCl3) d 1.52
(s, 3H), 1.69 (s, 3H), 4.82 (br t, J ¼ 6:7 Hz, 1H), 4.94–
5.04 (m, 1H), 5.22 (dd, J ¼ 3:3=17:4 Hz, 1H), 5.30 (d,
J ¼ 7:2 Hz, 1H), 5.26–5.41 (m, 1H), 7.26–7.42 (m, 5H);
13C NMR (50 MHz, CDCl3) d 24.9, 27.3, 80.1, 80.6,
108.8, 117.8, 126.8, 127.6, 128.0, 134.9, 137.4.
3.1.1. (1S,2R)-anti-6. 1H NMR (300 MHz, CDCl3) d
)0.18 (s, 3H), 0.02 (s, 3H), 0.86 (s, 9H), 4.14 (ddt,
J ¼ 1:2=4:8=5:6 Hz, 1H), 4.61 (d, J ¼ 4:8 Hz, 1H), 5.11
(dt, J ¼ 1:2=10:6 Hz, 1H), 5.18 (dt, J ¼ 1:2=17:2 Hz,
1H), 5.76 (ddd, J ¼ 5:6=10:6=17:2 Hz, 1H), 7.20–7.31
(m, 5H); 13C NMR (75 MHz, CDCl3) d )5.0, )4.6,
18.1, 25.7, 77.3, 78.2, 116.6, 127.0, 127.6, 127.9, 136.6,
140.8.
3.5. (4S,5S)-2,2-Dimethyl-4-phenyl-5-ethenyl[1,3]dioxo-
lane trans-4
By applying the same procedure to (1S,2S)-syn-3
(0.062 g, 0.38 mmol), we obtained (4S,5S)-trans-4 in 97%
yield (0.075 g, 0.37 mmol). H NMR (200 MHz, CDCl3)
1
3.1.2. (1S,2S)-syn-6. 1H NMR (300 MHz, CDCl3) d
)0.19 (s, 3H), 0.05 (s, 3H), 0.90 (s, 9H), 4.12 (ddt,
J ¼ 1:3=5:5=6:8 Hz, 1H), 4.44 (d, J ¼ 6:8 Hz, 1H), 5.10
(dt, J ¼ 1:4=10:6 Hz, 1H), 5.21 (dt, J ¼ 1:4=17:3 Hz,
1H), 5.68 (ddd, J ¼ 5:5=10:6=17:3 Hz, 1H), 7.26–7.34
(m, 5H); 13C NMR (75 MHz, CDCl3) d )5.1, )4.6,
18.1, 25.8, 77.3, 79.1, 116.5, 127.2, 127.8, 128.0, 136.2,
140.9.
d
1.56 (s, 3H), 1.60 (s, 3H), 4.20 (ddt, J ¼
1:1=7:0=8:4 Hz, 1H), 4.67 (d, J ¼ 8:4 Hz, 1H), 5.26 (dt,
J ¼ 1:0=10:0 Hz, 1H), 5.27 (dt, J ¼ 1:0=17:5 Hz, 1H),
5.90 (ddd, J ¼ 6:9=9:9=17:5 Hz, 1H), 7.26–7.42 (m, 5H);
13C NMR (50 MHz, CDCl3) d 27.00, 27.07, 82.9, 84.6,
109.2, 119.2, 126.4, 128.1, 128.4, 133.8, 137.1.
3.6. (4R,5S)-(2,2-Dimethyl-5-phenyl-[1,3]dioxolan-4-yl)-
methanol cis-5
3.2. (1S,2R)-1-Phenyl-but-3-ene-1,2-diol anti-3
Tetrabutylammonium
fluoride
(TBAF,
0.347 g,
A solution of (4S,5R)-cis-4 (0.138 g, 0.68 mmol) in
CH2Cl2 (15 mL) was allowed to react with ozone for
30 min at )78 °C. The reaction mixture was allowed to
reach rt while argon was bubbled into the solution to
remove excess ozone. The solvent was evaporated at
reduced pressure and the residue redissolved in anhy-
drous THF (5 mL) after which LiAlH4 (1.0 M in THF,
0.14 mL, 1.44 mmol) was added at 0 °C. The reaction
mixture was stirred at rt for 2 h, quenched with aq
NaHCO3, filtered on Celiteâ and extracted with ether.
The combined organic layers were dried over Na2SO4
and concentrated at reduced pressure. Purification by
flash chromatography on SiO2 (cyclohexane/ethyl ace-
tate 80:20) afforded 0.095 g of (4R,5S)-cis-5 (0.46 mmol,
1.1 mmol) was added at 0 °C to a solution of (1S,2R)-
anti-6 (0.257 g, 0.93 mmol) in THF (2 mL). After being
stirred for 2 h at rt, the reaction was quenched with
water and the organic layer extracted with ether. The
combined organic layers were dried over Na2SO4 and
evaporated at reduced pressure. Purification by flash
chromatography on SiO2 (cyclohexane/ethyl acetate
70:30) afforded 0.133 g of (1S,2R)-anti-3 (0.81 mmol,
88%). 1H NMR (300 MHz, CDCl3) d 4.34 (ddt,
J ¼ 1:2=4:9=6:1 Hz, 1H), 4.78 (d, J ¼ 4:9 Hz, 1H), 5.24
(dt, J ¼ 1:2=10:4 Hz, 1H), 5.30 (dt, J ¼ 1:3=17:3 Hz,
1H), 5.83 (ddd, J ¼ 6:1=10:4=17:3 Hz, 1H), 7.26–7.42
(m, 5H); 13C NMR (75 MHz, CDCl3) d 76.3, 76.5, 117.5,
126.6, 127.6, 128.1, 135.5, 139.8.
1
68%). H NMR (300 MHz, CDCl3) d 1.51 (s, 3H), 1.65

292
M. Lombardo et al. / Tetrahedron: Asymmetry 15 (2004) 289–292
(s, 3H), 3.12 (dd, J ¼ 4:6=11:6 Hz, 1H), 3.25 (dd,
J ¼ 8:0=11:6 Hz, 1H), 4.47 (ddd, J ¼ 4:6=6:9=8:0 Hz,
1H), 5.33 (d, J ¼ 6:9 Hz, 1H), 7.26–7.42 (m, 5H); 13C
NMR (75 MHz, CDCl3) d 24.9, 27.4, 62.7, 78.2, 78.8,
108.8, 126.2, 128.0, 128.3, 136.2.
References and notes
1. Lombardo, M.; Licciulli, S.; Morganti, S.; Trombini, C.
Chem. Commun. 2003, 1762.
2. Furstner, A.; Shi, N. J. Am. Chem. Soc. 1996, 118,
12349.
€
€
3. Furstner, A. Chem. Rev. 1999, 99, 991.
4. Barrett, A. G. M.; Malecha, J. W. J. Chem. Soc., Perkin
20
3.7. (4S,5S)-(2,2-Dimethyl-5-phenyl-[1,3]dioxolan-4-yl)-
methanol trans-5
Trans. 1 1994, 1901, (1S,2R)-anti-3: ½aꢀ +74.1 (c 1.2,
D
20
D
CHCl3); (1R,2S)-anti-3: ½aꢀ )75.7 (c 1.1, CHCl3).
5. Kazmaier, U.; Schneider, C. Synthesis 1998, 1314.
6. Untersteller, E.; Xin, Y. C.; Sinay, P. Tetrahedron Lett.
1994, 35, 2537.
The same reductive ozonisation reported above, was
applied to (4S,5S)-trans-4 (0.074 g, 0.36 mmol), to afford
(4S,5S)-trans-5 in 61% yield (0.045 g, 0.22 mmol). 1H
NMR (300 MHz, CDCl3) d 1.55 (s, 3H), 1.60 (s, 3H),
3.66 (dd, J ¼ 4:6=13:1 Hz, 1H), 3.84–3.92 (m, 2H), 4.90
(d, J ¼ 8:2 Hz, 1H), 7.29–7.45 (m, 5H); 13C NMR
(75 MHz, CDCl3) d 27.06, 27.11, 60.3, 78.6, 83.6, 109.3,
126.5, 128.3, 128.6, 137.6.
7. Zhou, W.-S.; Yang, Z.-C. Tetrahedron Lett. 1993, 34,
7075.
8. Effenberger, F.; Hopf, M.; Ziegler, T.; Hudelmayer J.
Chem. Ber. 1991, 124, 1651.
9. Banwell, M. G.; Coster, M. J.; Karunaratne, O. P.;
Smith, J. A. J. Chem. Soc., Perkin Trans. 1 2002,
1622.
10. Yang, Z.-C.; Zhou, W.-S. J. Chem. Soc., Chem. Commun.
1995, 743.
11. Lombardo, M.; Morganti, S.; Trombini, C. J. Org. Chem.
2003, 68, 997.
Acknowledgement
This work was supported by MIUR-Rome (FIRB
Funds to C.T., 2002).
12. Trost, B. M.; Crawley, M. L.; Lee, C. B. J. Am. Chem.
Soc. 2000, 122, 6120.