π-Face diastereoselection: Stereochemistry and reactivity of reduction reactions on conformationally rigid substrates: 5-X-adamantan-2-ones and trans-10-X-decal-2-ones

Article abstract of DOI:10.1016/S0040-4020(01)00705-0

Stereochemistry and relative rates k_ax and k_eq of reduction reactions on title compounds have been measured under four different reaction conditions (NaBH₄ in i-PrOH, LiBH₄ and NaAlH₄ in THF and LiAlH₄ in Et₂O). Reaction sensitivities show that the Group III element has a prominent role in determining the structure of the transition state shape.

Full text of DOI:10.1016/S0040-4020(01)00705-0

TETRAHEDRON
Pergamon
Tetrahedron 57 *2001) 7403±7407
p-Face diastereoselection: stereochemistry and reactivity of
reduction reactions on conformationally rigid substrates:
5
-X-adamantan-2-ones and trans-10-X-decal-2-ones
p
Giorgio Di Maio, Carolina Innella and Elisabetta Vecchi
p
Dipartimento di Chimica, Universit aÁ `La Sapienza', P. le A. Moro 5, I-00185 Rome, Italy
Received 18 April 2001; revised 6June 2001; accepted 28 June 2001
AbstractÐStereochemistry and relative rates kax and keq of reduction reactions on title compounds have been measured under four different
reaction conditions *NaBH in i-PrOH, LiBH and NaAlH in THF and LiAlH in Et O). Reaction sensitivities show that the Group III
element has a prominent role in determining the structure of the transition state shape. q 2001 Elsevier Science Ltd. All rights reserved.
4
4
4
4
2
5
1
. Introduction
previously investigated in the same reaction conditions
that are: 1) NaBH in i-PrOH *208C); 2) LiBH in THF
6
,²
4
4
1
We recently discussed addition reactions on the title
compounds with particular regard to the electronic contribu-
tion of remote substituents and pointed out the importance
of having a better knowledge of the reaction TS when
exploring questions concerned with p-face diastereo-
*208C); 3) NaAlH in THF and 4) LiAlH in Et O *208C)
*Fig. 1). Experiments aimed to explore the role of the Group
I metal and Group III element on reaction sensitivities.
4
4
2
2
selection. Simple theories such as those based on dipole±
dipole interactions are unable to give a satisfactory
3
explanation, nor can theories based on ground state MO
considerations. Most of the dif®culties encountered in
4
trying to explain the relationship between e.g. inductive
effects, changes in reagent, solvent, molecularity and
p-face selection are ascribable to the fact that p-face
diastereoselection has almost always been discussed in
terms of k /k changes instead of kax and keq changes. In
ax eq
fact the stereochemical bias k /k does not indicate what is
ax eq
happening on the two sides of a stereogenic centre: the two
faces of the molecule can behave independently from one
another and suffer different effects by the remote sub-
stituent. For instance in alkylation reactions we have
found that axial reactivities behave monotonically i.e. they
always increase with increasing electronegativity of the X
Figure 1.
group. Instead k changes strictly depend on the conforma-
eq
tion of the X group and on reaction conditions.
The choice of a rigid molecular skeleton eliminates all
questions arising from conformational uncertainty.
7
±8
We describe here the stereochemical and kinetic results of
hydride reduction on a series of 10-X-adamantan-2-ones,
namely with XˆH *1), Ph *2), OH *3), CO Me *4), Cl *5)
2
and Br *6). Their reactivity has been compared to that of
trans-10-X-2-decalones: XˆH *7), CO Et *8) and Cl *9),
2. Reaction products
2
Under all the above mentioned reaction conditions we
0
obtained, besides the well known alcohol 1 , alcohols
Keywords: reduction reactions; p-face diastereoselection; kinetics;
5-X-adamantan-2-ones.
p
²
Corresponding authors. Fax: 139-6-490631;
e-mail: elisabetta.vecchi@uniromal.it
Ref. 6reports only the stereochemical ratios of reduction for this reaction
condition. Our stereochemical ratios are in good agreement with them.
0040±4020/01/$ - see front matter q 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020*01)00705-0

7404
G. Di Maio et al. / Tetrahedron 57 ꢀ2001) 7403±7407
Figure 2.
Table 1. Stereochemical product ratios *kax/keq) for 10-X-adamantan-2-ones *1±6)
Reaction conditions Stereochemical product ratios *kax/keq
)
0
0
00
0
00
0
00
0
00
0
00
1
, s
I
ˆ0.00
2 /2 , s
I
ˆ0.12
3 /3 , s
I
ˆ0.25
4 /4 , s
I
ˆ0.32
5 /5 , s
I
ˆ0.47
6 /6 , s
I
ˆ0.47
1
2
3
4
) NaBH
) LiBH
) NaAlH
4
, i-PrOH, 208C
, THF, 208C
, THF, 208C
, Et O, 208C
1
1
1
1
1.28
1.22
1.09
1.27
1.05
1.37
1.38
1.23
4
1.061.12
1.18
1.10
1.34
1.19
4
1.30
1.25
1.41
1.12
1.38
1.23
) LiAlH
4
2
0
and *2±6) derived from equatorial attack *Fig. 2).
*
2±6) that derived from axial attack of the reduction agents,
the two sides of the ketonic group only from kinetic data and
therefore performed competitive reaction sets on equi-
molecular mixtures of compounds 1 and 2±6, respectively.
00
0
0
and 3 , respectively, were unsuccessful: we characterised
All our attempts to separate alcohols 2 and 3 from alcohols
2
00
00
them in the epimeric mixture via H and C NMR by
Competitive kinetic experiments provided highly repro-
ducible results largely independent of the concentration of
the reactants. The relative reaction rates were obtained by
1
13
6
comparison with literature data. In particular the axial
,9
1
3
C
³
hydroxyl diastereoisomers exhibit the carbinol
quaternary carbon signal at higher ®elds with respect to
GLC determination of the reaction yields. Relative rates k
and k were computed taking the overall rate of compound
ax
eq
1
0
the equatorial diastereomers, and have higher propensities
towards loss of water from parent ion in mass spec-
1 *kax ˆ k ˆ 1) as two, and by assuming that all reactions
eq
are ®rst order in ketone and the same order in reducing agent
for all ketones and are reported in Table 2 as a mean of at
least ®ve separate experiments.
1
1
§
trometry. Alcohols derived from 5-X-adamantan-2-ones
4±6) were separated from each other by HPLC and
individually characterised by comparison with literature
*
6
,12
data.
Reactions of Table 2 give fairly good LFER collected in
Table 3 in decreasing r value order.
3
. Results
Competitive reactions were also performed on equimolecu-
lar mixtures of compounds 1 and 7; 4 and 8; 5 and 9 *that is
admantanones and decalones bearing the same substituent).
The experimental data are collected in Table 4.
3.1. Stereochemistry: relative axial and equatorial
reactivity
For each set of reaction conditions we determined the
stereochemistry of reduction reactions by GLC. Table 1
collects the stereochemical outcome of several reactions
*®ve experiments at least for each substrate under all
reaction conditions).
³
We measured the areas of starting materials and products; each area was
divided by the corresponding molecular weight. Preliminary experiments
showed that GLC responses of compounds *1±6) on the one hand and the
alcohols *1±6) and *2±6) on the other were very close to each other.
Thus no correction was introduced. Although yields varied from run to
run, the material balance *i.e. the sum of starting and ®nal products) was
always greater than 90% of the starting material.
Yields varied in competition experiments depending on quenching times.
In order to minimize errors in computing the relative rates, we used only
data from reactions with yields ranging from 15 to 85%: outside this
range, either the errors in reading the GLC peaks of products originated
from less reactive competitors, or those in reading areas of unreacted
substances increase up to unacceptable limits.
0
00
The stereochemical results do not show a great difference
between either substrates or reaction condition: we have a
rather small constant increase of the k /k ratio with the
§
ax eq
1
3
increasing electronegativity of substituents which means
either that there are no large rates changes on the two
diastereotopic faces of the molecule, or that we are looking
at parallel rates variations. We can infer what happens on

G. Di Maio et al. / Tetrahedron 57 ꢀ2001) 7403±7407
Table 2. Overall ratio of rates and relative rates for 5-X-adamantan-2-ones
7405
Reaction conditions
Overall ratio of rates
Relative rates
k
ax
k
eq
a
k
1
/k
2
/k
3
/k
4
/k
5
/k
6
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
) NaBH
) LiBH
) NaAlH
) LiAlH
4
, i-PrOH
, THF
, THF
, Et
1/2.32/2.63/4.55/12.5/12.5
1/2.27/1.78/2.94/10.2/6.67
1/1.64/2.16/1.83/2.80/4.12
1/1.28/1.31/1.39/2.10/1.73
1
1
1
1
2.6
2.5
1.0
1.2
2.7
1.8
1.0
0.9
5.3
3.1
1.4
1.4
14.5
11.7
1.9
13.8
7.2
2.9
2.0
1
1
1
1
2.1
2.6
3.8
10.5
8.7
1.4
1.7
11.2
6.1
2.1
1.7
4
2.0
0.9
1.0
1.7
0.9
0.9
2.8
1.1
1.1
4
4
2
O
1.9
a
Mean standard deviation: 0.02.
2
Table 3. LFER data *r in decreasing order and corresponding r ) in differ-
ent reaction conditions
the identity of the third group metal being higher with
boron reactants than with aluminium and with changes of
the ®rst group metal being of less importance. It also
means that the boron TSs are more polar than the
aluminium ones. Some data in the literature indicate
that the ®rst group metal has a role in determining the
2
r
Reaction conditions
Attack
r
1
) NaBH
) LiBH
4
, i-PrOH, 208C
, THF, 208C
, THF, 208C
Ax.
Eq.
Ax.
Eq.
Ax.
Ax.
Eq.
Eq.
2.35
2.15
1.92
1.75
0.860.76
0.63
0.54
0.96
0.97
0.86
0.89
2
4
overall reaction rate with reactions using LiBH being
4
1
4,15
3) NaAlH
4) LiAlH
3) NaAlH
4) LiAlH
4
much faster than those with NaBH4.
holds for literature data concerning LiAlH vs NaAlH
The same
4
, Et
, THF, 208C
2
O, 208C
0.72
0.52
0.66
4
4
4
1
6
rates. In conjunction with experiments we are describ-
ing here, those data depict for the ®rst time, as far as we
4
, Et
2
O, 208C
0.54
Table 4. Relative rates *kax and keq) of 5-X-adamantan-2-ones and trans-10-X-decal-2-ones
Reaction conditions Relative rates
k
ax *adam.)
k
ax *dec.)
k
eq *adam.)
k
eq *dec.)
1
4
5
7
8
9
1
4
5
7
8
9
1
2
3
4
) NaBH
) LiBH
) NaAlH
) LiAlH
4
, i-PrOH
, THF
, THF
, Et
1
1
1
1
5.3
3.1
1.4
1.4
14.5
11.7
1.9
3.7
2.2
1.61.3
2.2
10.1
4.1
1.9
3.0
38.2
20.8
1
1
1
1.1
1
3.8
2.8
1.4
10.5
8.7
0.2
1.7
0.5
0.3
0.1
0.3
0.7
0.2
0.0
0.2
0.0
0.0
4
4
4
2
O
1.9
4.7
1.1
0.0
4. Discussion
know, the two different roles of I and III Group metal in
determining, respectively, reaction rates and sensitivity.
The kinetic data in Tables 2 and 3 show that:
For what concerns the decalones series *axial substituent),
Table 4, one can observe:
1
2
3
. both k and k slightly increase on increasing substituent
ax eq
electronegativity;
Axial reactivity:
. the small positive stereochemical bias *Table 1) is due to
a greater increase for axial reactivity *Table 2);
. the fastest reaction *k ) is also the most sensitive one
1. as in the adamantanone series, reactions rates increase
with increasing the substituent electronegativity;
ax
towards the effect of substituent; the attack is nucleo-
philic on both sides of the molecule;
k
2. as in the adamantanone series, the greatest effect of
changing substituent electronegativity is again observed
in reaction conditions 1) and 2);
4
. Table 3 clearly shows that r values group in two sets: the
boron and the aluminium set. This sharp division
suggests that the solvent change *i.e. i-PrOH!THF)
does not produce kinetic effects larger than those
produced by the B!Al change. Instead the solvent effect
3
. these effects have the same magnitude of those observed
in the adamantanone series *< ten times in passing from
H to Cl in reaction conditions 1) and 2) and two times in
reaction conditions 3) and 4);
*
i.e. THF!Et O) may be in the range of those produced
2
4. the change from B to Al has a larger effect on the reaction
sensitivity when Group I metal is Na.
by the Na!Li change as possibly suggested by the
mixing of values in the aluminium set;
. this means that the reaction sensitivity mainly depends on
5
Point 1 strictly parallels what we already described for
addition reactions and has the same rationale: the LUMO
carbonyl orbital is more extended on the axial face both
under the in¯uence of b-C±C bond hyperconjugation
k
5
We found that substrates bearing an axial substituent show, in analogous
reducing conditions, that keq decreases on increasing the substituents
electronegativity *see keq*dec) in Table 4): this is peculiar for a reaction
`electrophilic' in nature.
and C ±X bonds irrespective of their axial or equatorial
4

7406
G. Di Maio et al. / Tetrahedron 57 ꢀ2001) 7403±7407
0
*170); 9, 9 , 9 *170) and *in parentheses Toven,
00
0
00
0
00
0
00
conformation: this more rigidly preserves the nucleophilic
character of axial attack with changing reactant.
5 , 5 *170); 6, 6 , 6 *160); 7, 7 , 7 *150); 8, 8 , 8
0
initial isotherm, T
00
0
00
, T ) for 2, 2 , 2 *170, 5 min,
®nal
increase
0
00
Equatorial reactivity:
208C/min, 220); 3, 3 , 3 *80, 1 min, 108C/min, 220);
T ˆT ˆ2308C.
inj
det
The in¯uence of a C ±X axial substituent is far less predict-
4
able. In the present case we observe a decrease in reactivity
with increasing substituent electronegativity in all reduction
conditions: an electrophilic behaviour is mimicked and
therefore, one must imagine the TS with the O±metal
bond more developed than the C±H bond. In the alkylation
6.1. Starting materials
2-Adamantanone is commercially available *Aldrich) and
was used as such. Published procedures were used for the
preparation of 5-phenyl-adamantan-2-one 2 and 5-hydroxy-
1
1
7
reaction the equatorial attack was instead `nucleophilic' or
`
adamantan-2-one 3, 5-carbomethoxy-adamantan-2-one
4, 5-chloro-adamantan-2-one 5 and 5-bromo-adamantan-
1
8
electrophilic' depending on the used reactant. This is not
1
9
20
2-one 6, trans-decal-2-one 7, trans-10-carbethoxy-
schizophrenic behaviour: it depends on that the HOMO
carbonyl orbital suffer opposite distortion effects from the
b-C±C and the axial C ±X bond hyperconjugation. The
2
decal-2-one 8 and trans-10-chloro-decal-2-one 9.
1
5
4
balance is determined by the identity of the X group and
the reaction outcome, nucleophilic or electrophilic, could
depend on the reactant in a way till now unexplored. In
similar cases, the need of precise TS knowledge is more
evident. In our opinion many of the uncertainties and
debates in the literature concerning p-face diastereo-
selection on less rigid substrates are due to the fact that
products ascribed to axial attack partly originate from
6
.2. Preparation of reagents
1 M solns of NaBH , LiBH and LiAlH , respectively, in
anhydrous i-PrOH, THF and Et O were prepared, kept
4
4
4
2
2
under dry N and titrated before use. Commercial solns
2
2
*1 M) of NaAlH in THF *Aldrich) were titrated and diluted
to the desired concentration just before use.
4
the more equivocal equatorial attack on the C ±X axial
4
conformation.
6.3. Reactions
7
All the reactions were carried out under a pure dry nitrogen
atmosphere and the glassware was carefully ¯amed and
5. Conclusions
¯
ushed with dry nitrogen before use. Typically: a solution
of the reducing reagent *0.1 M) was added via a syringe into
a ¯ask containing a solution 0.1 M of each substrate in the
suitable anhydrous solvent, with n-hexadecane as internal
standard. Reactions lasted a few minutes. After this time, the
reaction mixtures were slowly hydrolysed with satd. aq.
NH Cl and extracted three times with Et O. The ethereal
We determined the axial and equatorial rates of attack on
-X-adamantan-2-ones and trans-10-X-decal-2-ones in
5
reduction reactions. Changes of Group I metals *Li and
Na) only have a small in¯uence on the reaction sensitivity:
this mainly depends on the Group III element, being larger
for reducing agents containing boron with respect to those
containing aluminum. This means that the transition state
shape is mainly determined by the Group III element. The
TS is more polar, trapezoidal with boron reactants and
has the O´´´B bond less developed than the C´´´H bond.
Aluminum hydrides generate a more square, pericyclic
TS. Solvent changes do not produce kinetic effects larger
than those produced by the Group III element change. A
comparison of axial and equatorial conformation of sub-
stituents *adamantanones and decalones series) shows for
axial reactivity similar behaviour in the two series; on the
other side, equatorial reactivity shows a different and less
predictable behaviour when the substituent is axial.
4
2
solns washed with water were combined, dried over
Na SO , ®ltered and evaporated. Analyses of reaction
mixtures by GLC were carried out as described.
2
4
6.4. Competition experiments
Three ¯asks *10 ml) were equipped with magnetic stirrer
and connected by means of a three-point star-rotating
receiver to a graduated burette, gas inlet and CaCl tube.
2
The apparatus was carefully dried by ¯aming it under a
nitrogen ¯ow. Each ¯ask contained an equimolecular
mixture of 1 and 3 or, respectively, depending on the chosen
partner for that particular experiment, 1, 4, 5, or 1, 2, 6, or 1,
7
and dissolved in 2 or 3 ml of anhydrous solvent *i-PrOH,
¶
or 4, 8, or 5, 9); 0.1 mmol of each substance were used
6. Experimental
Et O or THF) depending on whether the competition
2
reaction was with two or three competitors. The graduated
burette was ®lled via a syringe with the suitable, con-
veniently diluted, reactant, and the equimolar amount of it
was added to the substrates mixtures under vigorous stirring.
Reaction mixtures were then hydrolysed and worked up
under standard methods, and ®nally examined by GLC in
order to measure the relative amounts of products and
starting materials.
1
13
H and C NMR spectra were recorded on a GEMINI 200.
GC±MS analyses were performed with a GC±MS HP 5970
Chemstation Mass Selective Detector connected with a HP
5
gas chromatograph. GLC analyses were carried out on a
Carlo Erba HRGC Mega Series 5300 apparatus using a
3
phase O.V.1), He ¯owˆ0.5 ml/min. We report, in sequence,
the elution order of compounds from each mixture and the
most suitable temperature conditions *in parentheses
890 gas chromatograph and on a HP G1800A GCD System
0 m, 0.25 mm i.d. fused silica capillary column *stationary
¶
As a consequence of peaks overlapping in the GLC analysis it was not
possible to perform competition experiments in which all products were
present at the same time.
0
0
00
Toven8C) for, respectively,: 1, 1 *160); 4, 4 , 4 *170); 5,

G. Di Maio et al. / Tetrahedron 57 ꢀ2001) 7403±7407
7407
Acknowledgements
9. Srivastava, S.; Cheung, C. K.; le Noble, W. J. Magn. Reson.
Chem. 1985, 23, 232±235.
We are indebted to the Italian MURST and to the University
of Rome `La Sapienza' *National Project `Stereoselezione
in Sintesi Organica. Metodologie ed Applicazioni') for
10. Senda, Y.; Ishiyana, J.; Imaizumi, S. J. Tetrahedron 1975, 31,
1601±1605.
11. Green, M. M. Mass Spectrom. Stereochem. 1975, 35±38.
12. Adcock, W.; Cotton, J.; Trout, N. A. J. Org. Chem. 1994, 59,
®
nancial support.
1
867±1876.
1
3. Charton, M. Prog. Phys. Org. Chem. 1981, 13, 119±251.
4. Nystrom, R. F.; Chaikin, S. W.; Brown, W. G. J. Am. Chem.
Soc. 1949, 71, 3245±3246.
References
1
1
. Di Maio, G.; Solito, G.; Var Áõ , M. R.; Vecchi, E. Tetrahedron
000, 56, 7237±7243.
. See e.g. the recent review by Cieplak, A. S. Chem. Rev., 1999,
9, 1265±1336 and references cited therein.
. Wu, Y.-D.; Tucker, J. A.; Houk, K. N. J. Am. Chem. Soc.
991, 113, 5818±5027.
1
5. Brown, H. C.; Ichikawa, K. J. Am. Chem. Soc. 1961, 83,
4
2
372±4374.
2
3
16. Cha, J. S.; Brown, H. C. J. Org. Chem. 1993, 58, 4727±4731.
17. Geluk, H. W. Synthesis 1972, 374±375.
18. Cheung, C. K.; le Noble, W. J.; Srivastava, S. J. Org. Chem.
9
1
1
983, 48, 1099±1101.
9. Geluk, H. W.; Schlatlman, J. L. M. Tetrahedron 1968, 24,
369±5376.
4
. Klein, J. Tetrahedron Lett. 1973, 44, 4307±4310.
. Di Maio, G.; Migneco, L. M.; Vecchi, E. Tetrahedron 1990,
1
2
2
2
5
5
4
6, 6053±6060 for stereochemical ratios kax/keq of compounds
0. Monson, R. S. Advanced Organic Synthesis; Academic Press:
New York, 1971; pp 27±28 and 80±81.
1. Dreiding, A. S.; Tomasewsky, A. J. J. Am. Chem. Soc. 1955,
7, 8 and 9.
6
7
8
. Cheung, C. K.; Tseng, L. T.; Lin, M.-H.; Srivastava, S.; le
Noble, W. J. J. Am. Chem. Soc. 1986, 108, 1598±1605.
. Cianetti, C.; Di Maio, G.; Pignatelli, V.; Tagliatesta, P.;
Vecchi, E.; Zeuli, E. Tetrahedron 1983, 39, 657±666.
77, 411±414.
2. Orchin, M.; Wender, I. Anal. Chem. 1949, 21, 875±876.
. Migneco, L. M.; Vecchi, E. Gazz. Chim. Ital. 1997, 127, 19±
2
4.