Conformational studies by dynamic NMR. 83.1 Correlated enantiomerization pathways for the stereolabile propeller antipodes of dimesityl substituted ethanol and ethers

Article abstract of DOI:10.1021/jo010420m

Below -100 °c, the Nmr spectra of dimesityl derivatives of ethanol and of various ethers reveal how these molecules exist asM and P propeller-like stereolabile enantiomers, owing to the restricted rotation about the Ar-C bond. Single-crystal X-ray diffraction of one such derivative confirmed the existence of a two-blade propeller structure. Computer analysis of the NMR line shape allowed the barriers for the enantiomerization process to be determined. Theoretical modeling (Molecular Mechanics) of the interconversion circuit produced good agreement between the computed and experimental barrier for a correlated dynamic process where a disrotatory one-ring flip pathway reverses the helicity of the conformational enantiomers. Introduction of a configurationally stable chiral center allowed two distinct NMR spectra to be detected at appropriate low temperature for two stereolabile diastereoisomers.

Full text of DOI:10.1021/jo010420m

J . Org. Chem. 2001, 66, 5853-5858
5853
1
Con for m a tion a l Stu d ies by Dyn a m ic NMR. 83. Cor r ela ted
En a n tiom er iza tion P a th w a ys for th e Ster eola bile P r op eller
An tip od es of Dim esityl Su bstitu ted Eth a n ol a n d Eth er s
2
Stefano Grilli, Lodovico Lunazzi,* and Andrea Mazzanti*
Department of Organic Chemistry “A. Mangini”, University of Bologna,
Risorgimento, 4, Bologna 40136, Italy
lunazzi@ms.fci.unibo.it
Received April 23, 2001
Below -100 °C, the NMR spectra of dimesityl derivatives of ethanol and of various ethers reveal
how these molecules exist as M and P propeller-like stereolabile enantiomers, owing to the restricted
rotation about the Ar-C bond. Single-crystal X-ray diffraction of one such derivative confirmed
the existence of a two-blade propeller structure. Computer analysis of the NMR line shape allowed
the barriers for the enantiomerization process to be determined. Theoretical modeling (Molecular
Mechanics) of the interconversion circuit produced good agreement between the computed and
experimental barrier for a correlated dynamic process where a disrotatory one-ring flip pathway
reverses the helicity of the conformational enantiomers. Introduction of a configurationally stable
chiral center allowed two distinct NMR spectra to be detected at appropriate low temperature for
two stereolabile diastereoisomers.
q
In tr od u ction
having ∆G values equal to 4.5 and 5.0 kcal mol-1
,
respectively, for dimesityl sulfoxide and dimesityl sulfone.
On the basis of the latter results it seems conceivable to
Compounds comprising three ortho-substituted phenyl
groups bonded to a configurationally stable sp atom (e.g.,
carbon or silicon) display dynamic processes involving the
interconversion of stereolabile enantiomers, according to
correlated pathways (cog-wheeling circuit) described by
3
foretell that a similar effect would also occur when two
mesityl groups are bonded to a sp3 hybridized carbon
atom, since stereolabile helical enantiomers are also
expected to be available.
9,10
3
-6
the so-called n-ring flip mechanisms.
The correspond-
To obtain an experimental verification of this predic-
tion and with the purpose of assessing the possible
consequences of the related stereochemical properties, the
dimesityl derivatives of ethanol (1), dimethyl ether (2),
ethylmethyl ether (3), diethyl ether (4), and ethyl 2-me-
thylbutyl ether (5) were investigated: Mes
Mes CHOCH (2); Mes C(Me)OCH (3); Mes
CH (4); Mes C(Me)OCH CH(Me)Et (5) (Mes ) 2,4,6-
trimethyl phenyl)
7
ing barriers were reported to cover the range 9-22 kcal
-
1
mol
.
We have recently shown that even when there are only
two ortho-substituted phenyl groups bonded to a tetra-
hedral center, like a sulfur atom, an analogous dynamic
process does take place. Thus, in the case of dimesityl
sulfoxide and sulfone a cog-wheel effect, which allows the
interconversion of the two M and P propeller-like an-
tipodes (conformational enantiomers) through the one-
2
C(Me)OH (1);
2
3
2
3
2
C(Me)OCH -
2
3
2
2
8
ring flip mechanism, was detected. In these cases,
however, the enantiomerization barriers are much lower,
Resu lts a n d Discu ssion
All these derivatives were found to display dynamic
1
13
*
To whom correspondence should be sent.
effects in their H and C NMR spectra: as an example
(1) Part 82. Casarini, D.; Lunazzi, L.; Mazzanti, A. Angew. Chem.,
13
of such temperature-dependent features, the C spectra
of 3 will be illustrated.
Int. Ed. 2001, 40, 2536. Part 81. Grilli, S.; Lunazzi, L.; Mazzanti, A.
J . Org. Chem. 2001, 66, 4444.
(
2) In partial fulfilment of the requirements for the Ph.D. in
Chemical Sciences, University of Bologna.
3) (a) Sabacky, M. J .; J ohnson, S. M.; Martin, J . C. Paul, I. C. J .
At ambient temperature down to -30 °C (Figure 1)
three signals due, respectively, to the methyl bonded to
the quaternary carbon (24.9 ppm), to the four methyls
in the ortho position (23.0 ppm) and to the two methyls
in the para position (19.0 ppm) are observed in the 19-
26 ppm spectral region. Owing to the different relaxation
(
Am. Chem. Soc. 1969, 91, 7542. (b) Rieker, A.; Kessler, K. Tetrahedron
Lett. 1969, 1227. (c) Kessler, K.; Moosmayer, A.; Rieker, A. Tetrahedron
1
969, 25, 287.
4) (a) Gust, D.; Mislow, K. J . Am. Chem. Soc. 1973, 95, 1535. (b)
(
Finocchiaro, P.; Gust, D.; Mislow, K. J . Am. Chem. Soc. 1974, 96, 2165
and 2176. (c) Mislow, K. Chemtracts Org. Chem. 1989, 2, 151.
13
times the intensity of these C lines is not proportional
(5) (a) Boettcher, R. J .; Gust, D.; Mislow, K. J . Am. Chem. Soc. 1973,
9
5, 7157. (b) Hutchings, M. G.; Maryanoff, C. A.; Mislow, K. J . Am.
Chem. Soc. 1973, 95, 7158. (c) Kates, M. R.; Andose, J . P.; Finocchiaro,
P.; Gust. D.; Mislow, K. J . Am. Chem. Soc. 1975, 97, 1772. (d) Mislow,
K. Acc. Chem. Res. 1976, 9, 26.
(9) The two enantiomeric forms of the highly hindered derivatives
Ar
2
CH-COOH (Ar ) 2,4,2′,4′-tetra-tert-butyl-6,6′-dimethylphenyl and
cognates) could be separated at ambient temperature having racemi-
-1
(
6) Sedo, J .; Ventosa, N.; Molinos. Ma. A.; Pons, M.; Rovira, C.;
Veciana, C. J . Org. Chem. 2001, 66, 1579.
7) Oki, M. Applications of Dynamic NMR Spectroscopy to Organic
Chemistry; VCH: Deerfield Beach, 1985; Chapter 5, p 226.
8) Casarini, D.; Grilli, S.; Lunazzi, L.; Mazzanti, A. J . Org. Chem.
001, 66, 2757.
sation barriers in the range 21.5-22.9 kcal mol (Akkerman, O. S.;
Coops, J . Rec. Trav. Chim. 1967, 86, 755. See also: Akkerman, O. S.
Rec. Trav. Chim. 1970, 89, 673.).
(
(10) Dynamic processes were detected in the ArAr′CHMe and
ArAr′CHOH derivatives (where Ar * Ar′), see: Finocchiaro, P. Gazz.
Chim. Ital. 1975, 105, 149.
(
2
1
0.1021/jo010420m CCC: $20.00 © 2001 American Chemical Society
Published on Web 07/26/2001

5
854 J . Org. Chem., Vol. 66, No. 17, 2001
Grilli et al.
F igu r e 2. The aromatic region of the 13C NMR spectra (100.6
MHz in CHF
2
Cl) of 3 displays four lines at -10 °C (top)
whereas twelve lines are detected at -146 °C (bottom).
q
Ta ble 1. Exp er im en ta l F r ee En er gies of Activa tion (∆G )
-
1
a n d Com p u ted Ba r r ier s (∆E), Both in k ca l m ol , for th e
a
Dyn a m ic P r ocesses of Com p ou n d s 1-5
cmpd
1
5.8
2
4.4
3
7.45
4
8.3
8
8.6
5
q
∆G (exp.)
9.5
9.3
8.9
b
c
.1
∆E (calcd)
7.0
2.3
7.6
F igu r e 1. 13C NMR signals (100.6 MHz in CHF
Cl) due to
2
the carbon-bonded methyl groups of 3 as a function of
temperature (the OMe line lies outside the displayed region).
a
Unless otherwise specified, the experimental values were
obtained by monitoring the signals of the ortho methyl groups.
b
Value obtained by monitoring the signals of the methylene
hydrogens. c Value obtained by monitoring the C signals of the
13
aliphatic methine group.
to the number of carbons. On lowering the temperature,
the ortho methyl signal broadens considerably and
reaches the coalescence point at -108 °C, splitting
subsequently into a pair of widely separated lines at
The asymmetry of the adopted conformation is also
reflected in the spectrum of the aromatic region. Whereas
at -10 °C the aromatic carbons display only four 13
C lines
-
121 °C. On further cooling to -146 °C each of these
corresponding to the ipso, ortho, meta, and para positions
within each enantiotopic mesityl ring, twelve lines are
observed at -146 °C (Figure 2). This is because the two
rings are now diastereotopic, as are diastereotopic the
six carbons within each ring.
lines, and the para methyl line as well, display an
additional splitting so that, as a whole, six anisochronous
lines are eventually observed for the six phenyl bonded
methyl groups (the seventh line appearing in the picture
at 24.9 ppm cannot be obviously influenced by this
dynamic process in that belongs to a single methyl group).
This indicates that the six methyl groups bonded to
phenyl have been rendered diastereotopic, thus implying
that compound 3 adopts a conformation which lacks any
Analogous results were obtained in the case of 1 and
2
and the corresponding interconversion barriers are
collected in Table 1.
11
Theoretical (Molecular Mechanics ) calculations agree
with the NMR results in predicting asymmetric confor-
mations for 1-3: in the computed propeller-like struc-
tures the pairs of dihedral angles between each mesityl
ring and the C1-C-C1′ plane were found 46° and 47°
in 1, 50° and 53° in 2, and 43° and 50° in 3.
element of symmetry (C
1
point group). Line shape
simulations carried out at various temperatures provided
the set of rate constants that yielded the free energy of
q
-1
activation (∆G ) 7.45 ( 0.15 kcal mol ) for the observed
q
dynamic process. The same ∆G value was also obtained
by simulating the shape of the H methyl signals, where
A single-crystal X-ray diffraction structure (see Ex-
perimental Section) of 1 confirmed the existence of the
1
five lines were observed at -146 °C for these phenyl
bonded methyl groups, due to the accidental degeneracy
of two such lines.
(11) MMX force field as implemented in the computer package PC
Model, Serena Software, Bloomington, IN.

Conformational Studies by Dynamic NMR
J . Org. Chem., Vol. 66, No. 17, 2001 5855
Sch em e 1. Sch em a tic Rep r esen ta tion of th e Rin g
F lip Cir cu it of 3
F igu r e 3. Computed structures (Molecular Mechanics11) of
the P and M enantiomers of 1 (top). Underneath is reported
the experimental structure (X-ray diffraction) of the four
molecules within the crystal cell. In both the computed and
experimental structures the carbons of the enantiomer P are
shaded dark whereas those of the enantiomer M are repre-
sented as colorless.
the one- and two-ring flip mechanism and, to decide
which one is expected to have the transition state with
the lower energy, the dynamic process was modeled
according to a Molecular Mechanics11 approach.
helical shape, the corresponding dihedral angles mea-
sured in the crystalline state (46° and 55°) being close to
those computed for the isolated molecule. The crystal cell
of 1 contains in fact four molecules, two of them having
a helicity opposite to that of the other two (heterochiral
crystals). In Figure 3 the M and P conformational
enantiomers predicted by calculations are displayed
together with the experimental structures of 1.
In Scheme 2 is displayed the two-dimensional contour
plot, derived from the three-dimensional energy surface
computed as function of the two rotation angles between
1
5
the mesityl rings and the C1-C(OMe)-C1′ plane of 3.
The dotted line between the energy minima correspond-
ing to the M and P enantiomers represents the two-ring
flip mechanism, which is a conrotatory motion leading
to a transition state where both the mesityl rings are
orthogonal to the C1-C(OMe)-C1′ plane (gear clashing).
As mentioned, the interconversion of these M and P
stereolabile enantiomers is expected to occur through a
correlated rotation process (cog-wheeling circuit) of the
two mesityl groups. In principle three possible pathways
are available: they can be classified according to the
n-ring flip mechanisms reported in Scheme 1, where n
-
1
The computed barrier for this process (18.2 kcal mol
)
is much too high to be compatible with the experimental
value. The full line is that for the one-ring flip mecha-
nism, which is a disrotatory motion where the enanti-
omerization takes place via a transition state having one
ring orthogonal to and the other coplanar with the
mentioned plane (gear meshing). This process obviously
entails a degenerate transition state where the two rings
1
2-14
can be 0, 1, or 2.
The zero-ring flip pathway is a conrotatory motion that
not only leads to a very crowded transition state (coplanar
mesityl rings) but, in addition, does not allow for the
1
6
exchange of the ortho and meta edge positions, so that,
switch their relative position. The corresponding com-
being “NMR silent”,1
2,13b,14
cannot account for the ob-
-1
puted barrier (7.6 kcal mol ) is much lower than the
served dynamic process. The choice is thus restricted to
previous one and essentially coincides with that (7.45 kcal
(
12) Nugiel, D. A.; Biali, S. E.; Rappoport, Z. J . Am. Chem. Soc. 1984,
06, 3357.
13) (a) Rappoport, Z.; Biali, S. E. Acc. Chem. Res. 1988, 21, 442 (b)
Rappoport, Z.; Biali, S. E. Acc. Chem. Res. 1997, 30, 307.
14) (a) Grilli, S.; Lunazzi, L.; Mazzanti, A.; Casarini, D.; Femoni,
(15) The rotation barrier about the C-OMe bond was computed to
be too low (2 kcal mol- ) to be detectable by NMR spectroscopy. Thus
the observed dynamic spectral features only depend on the restricted
rotation involving the mesityl rings.
(16) Kaftory, M.; Nugiel, D. A.; Biali, S. E.; Rappoport, Z. J . Am.
Chem. Soc. 1989, 111, 8181. Rappoport, Z.; Biali, S. E.; Kaftory, M. J .
Am. Chem. Soc. 1990, 112, 7742.
1
1
(
(
C. J . Org. Chem. 2001, 66, 488. (b) Grilli, S.; Lunazzi, L.; Mazzanti,
A.; Mazzanti, G. J . Org. Chem. 2001, 66, 748.

5
856 J . Org. Chem., Vol. 66, No. 17, 2001
Grilli et al.
Sch em e 2. Tw o Dim en sion a l Con tou r P lot of th e
En er gy Su r fa ce of 3, Com p u ted a s F u n ction of th e
Dih ed r a l An gles Ma d e by th e Mesityl Su bstitu en ts
a
w ith th e C1-C(OMe)-C1′ P la n e
a
The full and the dotted line represent the one-ring and the
two-ring flip pathway, respectively.
F igu r e 4. 1H NMR spectrum (300 MHz in CHF
Cl) of 4 at
2
two different temperatures. Not only the para methyl signal
splits into two and the ortho signal into four lines at -130 °C,
2
but also the quartet signal of the hydrogens of the CH group
splits into two multiplets. (At this temperature the arrowed
single line of the methyl group bonded to the quaternary
carbon broadens considerably).
mol-1) measured by the dynamic NMR experiment. As a
consequence the latter mechanism is the most plausible
pathway for accomplishing the M, P enantiomerization
in derivative 3.
The computations of the energy surfaces for derivatives
1
and 2 gave similar results: the corresponding data have
and 8.3 kcal mol-1, respectively, as in Table 1) were found
been collected in Table 1.
These observations entail two stereochemical conse-
quences that we were able to verify experimentally.
coincident within the experimental errors (( 0.15 kcal
-
1
mol ).
i) If a prochiral probe,17 such as an ethyl group, is
(ii) If a configurationally stable chiral center is intro-
(
duced in the molecule, it is expected that two stereolabile
diastereoisomers become observable at a temperature
where the rate for the propeller interconversion is
rendered slow: this is because two sources of chirality
would be simultaneously present. Derivative 5 fulfils
these conditions since the aliphatic methine carbon is a
configurationally stable chiral center. Actually the low
temperature 13C spectrum of 5 displays a number of lines
much greater than those expected for a single species.
In particular it can be observed (Figure 5) that the single
line of the mentioned methine carbon broadens on cooling
and decoalesces, eventually yielding two lines (separated
by 0.265 ppm at -114 °C) with a 53:47 intensity ratio.
These lines are due to the pair of conformational dia-
stereoisomers RM and R P (Scheme 3): each of them also
comprises an enantiomeric form (respectively SP and
introduced in an appropriate position, the existence of
the molecular asymmetry should be also revealed by the
anisochronicity of the signals of the two methylene
hydrogens. Thus in 1,1-dimesityl diethyl ether, (4) the
2
quartet of the CH group (coupled to the Me triplet with
a vicinal J ) 7.1 Hz) broadens on cooling and eventually
splits into a pair of separate multiplets (displaying an
additional J geminal coupling of -7.9 Hz) at -130 °C
(Figure 4): the two geminal hydrogens become in fact
diastereotopic when the interconversion of the M and P
enantiomeric conformers is blocked. Since the source of
the diastereotopicity for the hydrogens within the me-
thylene moiety corresponds to the same restricted rota-
tion process which makes the conformational antipodes
sufficiently long living in the NMR time-scale, the barrier
obtained by monitoring the methylene signals must be
the same as that derived by monitoring the ortho methyl
signals of the mesityl groups. Actually the two values (8.1
11
SM). Molecular Mechanics calculations suggest that the
pair RM ≡ SP should be slightly more stable than the
pair RP ≡ SM, the computed energy difference being 0.03
-
1
kcal mol . Although such a difference corresponds
exactly to the relative conformer population observed at
(17) J ennings, W. B. Chem. Rev. 1975, 75, 307.

Conformational Studies by Dynamic NMR
J . Org. Chem., Vol. 66, No. 17, 2001 5857
Sch em e 3. Com p u ted Str u ctu r es (Molecu la r
Mech a n ics11) of th e R, M (m or e sta ble) a n d R, P
(
less sta ble) Dia ster eoisom er s of 5a
a
The relative values of the computed energies (E) are in kcal
.
mol-
1
mol- higher. Also, replacement of the OH group of 1 with
1
the OMe group in 3 increases the barrier by 1.6 kcal
-
1
mol (Table 1).
Exp er im en ta l Section
Ma ter ia ls. 1,1-Dimesitylethanol, (1) was prepared as re-
ported in the literature.18 Products 2-5 were prepared accord-
ing to the following procedure. To a suspension of KH (3 mmol
in 5 mL of anhydrous THF), a solution of 1,1-Dimesitylethanol
(
°
5
1 mmol in 2 mL of anhydrous THF) was slowly added at 0
C. After 10 min a solution of dibenzo-18-crown-6 (1 mmol in
mL of THF) was added and the mixture warmed to room
temperature. After stirring for 30 min, the appropriate elec-
trofile (MeI, EtI, 2-methyl-1-butyl-tosilate,19 5 mmol in 2 mL
of anhydrous THF) was added, and the mixture stirred for 3-6
h until the starting product disappeared (TLC). After quench-
F igu r e 5. Left: experimental 13C signal (100.6 MHz in CHF
-
2
Cl) of the aliphatic CH carbon of 5 as a function of tempera-
ture, eventually showing two lines (ratio 53:47) corresponding
to the two diastereoisomers of Scheme 3. Right: line shape
simulation obtained with the rate constants reported.
ing with water, the organic layers were extracted (Et
2
O), dried
(
Na SO ), and evaporated at reduced pressure. The crude
2
4
products were purified on preparative TLC (eluent: pentane)
to give the products in 20-50% yields. Note: the products are
extremely sensitive to acidity to give 1,1-dimesityl-etylene.
-114 °C, the smallness of this computed value does not
1
Dim esitylm eth yl Meth yl Eth er (2). H NMR (200 MHz,
allow one to exclude that the good agreement is the
consequence of an accidental coincidence.
CDCl
3
, 22 °C, TMS): δ ) 2.08 (s, 12H, o-CH
), 3.29 (3H, s, OCH ), 5.60 (1H, s CH), 6.72 (s, 4H, CH).
C NMR (50.3 MHz, CDCl , 22 °C, TMS): δ ) 19.8 (p-CH ),
19.9 (o-CH ), 56.0 (OCH ), 81.0 (CHO), 129.8 (CH), 135.9 (q),
3
), 2.19 (s, 6H,
p-CH
3
3
The line shape simulation of the two 13C methine lines
of 5 yields the rate constants, hence the free energy of
activation, required to intercovert the two diastereoiso-
13
3
3
3
3
136.9 (q), 143.8 (q). Anal. Calcd. For C20
9.28. Found: C, 85.01; H, 9.26.
H
26O: C, 85.06; H,
-
1
mers (9.3 kcal mol ). The barrier must be the same as
that obtained by monitoring, for instance, the lines of the
ortho methyl signals, since the pathways for intercon-
verting the diastereoisomers is the same ring-flip mech-
anism involving the mesityl groups. Within the experi-
1
1
,1-Dim esityleth yl Meth yl Eth er (3). H NMR (200 MHz,
C
2
6
D
6
, 22 °C, TMS) δ ) 1.87 (s, 3H, CH
.22 (s, 12H, o-CH ), 2.86 (3H, s, OCH
NMR (50.3 MHz, C , 22 °C, TMS): δ ) 21.0 (CH
), 48.8 (OCH ), 87.4 (q), 132.7 (CH), 136.0
q), 137.3 (q), 142.3 (q). Anal. Calcd. For C21 28O: C, 85.08;
3 3
), 2.10 (s, 6H, p-CH ),
1
3
3
3
), 6.69 (s, 4H, CH).
C
6
D
6
3
), 24.7
(o-CH
3
), 26.6 (p-CH
3
3
-1
mental uncertainty, the latter value (9.5 kcal mol ) was
found indeed equal to the previous one (Table 1).
The interconversion barriers measured in these com-
pounds are significantly dependent upon the steric
hindrance. Since derivative 5 bears an oxygen-bonded
(
H
H, 9.52. Found: C, 85.05; H, 9.56.
1,1-Dim esityleth yl Eth yl Eth er (4). H NMR (300 MHz,
, 22 °C, TMS) δ ) 1.01 (t, J ) 6.9 Hz, 3H, CH ), 1.91 (s,
H, CH ), 2.10 (s, 6H, p-CH ), 2.24 (s, 12H, o-CH ), 3.09 (q, J
6.9 Hz, 2H, CH
), 6.70 (s, 4H, CH). 13C NMR (75.5 MHz,
, 22 °C, TMS): δ ) 16.0 (CH ), 20.5 (CH ), 24.4 (o-CH ),
6.8 (p-CH ), 56.9 (CH ), 86.5 (q), 132.2 (CH), 135.4 (q), 136.7
1
C
6
D
6
3
3
3
3
3
)
6
2
2
substituent (-CH CHMeEt) quite bulkier than the meth-
yl group of 3 or the ethyl group of 4, the corresponding
C
D
6
3
3
3
2
3
2
free energy of activation turns out to be appreciably
-1
higher (respectively by 2.0 and 1.2 kcal mol ). Likewise,
substitution of the methine hydrogen atom of 2 with a
methyl group in 3 makes the barrier of the latter 3.0 kcal
(18) Timberlake, J . W.; Pan, D.; Murray, J .; J ursic, B. S.; Chen, T.
J . Org. Chem. 1995, 60, 5295.
(19) Nakamura, Y.; Mori, K. Eur. J . Org. Chem. 1999, 2175

5
858 J . Org. Chem., Vol. 66, No. 17, 2001
Grilli et al.
(
q), 142.3 (q). Anal. Calcd. For C22 30O: C, 85.11; H, 9.74.
H
and condensing therein the gaseous solvents by means of liquid
nitrogen. The tubes were subsequently sealed in vacuo and
introduced into the precooled probe of the 300 MHz spectrom-
Found: C, 85.08; H, 9.77.
1
,1-Dim esityleth yl-2-m eth ylbu tyl Eth er (5). 1H NMR
1
3
(
300 MHz, C
CH ), 0.88 (d, J ) 6.8 Hz, 3H, CH
.29-1.56 (m, 2H, CH ), 1.96 (s, 3H, CH
.25 (s, 12H, o-CH ), 3.02 (m, 2H, CH
), 6.70 (s, 4H, CH). 13
, 22 °C, TMS): δ ) 11.6 (CH
), 24.5 (o-CH ), 26.8 (CH ), 26.9 (p-CH
), 86.8 (q), 132.1 (CH), 135.4 (q), 136.7 (q), 142.3
q). Anal. Calcd. For C25 36O: C, 85.17; H, 10.29. Found: C,
5.13; H, 10.32.
X-r a y Diffr a ction . Crystal Data for 1,1-dimesityl-1-ethanol
1): C20 26O (282.41), monoclinic, space group P2 /c, Z ) 4, a
8.6804(5), b ) 15.9597(9), c ) 11.8673(6) Å, â ) 95.858(2),
6
D
6
, 22 °C, TMS) δ ) 0.75 (t, J ) 7.4 Hz, 3H,
eter operating at 75.45 MHz for C or 400 MHz operating at
1
3
3
3
), 1.00-1.14 (m, 1H, CH),
100.6 MHz for C. The temperatures were calibrated by
substituting the sample with a precision Cu/Ni thermocouple
before the measurements. Total line shape simulations were
1
2
2
3
), 2.10 (s, 6H, p-CH
3
),
3
2
C
2
0
NMR (75.5 MHz, C
6
D
6
3
), 17.6
), 36.0
achieved by using a PC version of the DNMR-6 program.
(
(
(
CH
3
), 20.5 (CH
3
3
2
3
Since at the low temperatures required to observe the ex-
change process the intrinsic line width of the compounds was
significantly temperature dependent, the corresponding de-
pendence of the solvent signal was measured. The appropriate
ratio with the line width of the compound was then taken into
account. We also checked that errors as large as 50% on this
value affected the activation energy by less than 0.05 kcal
CH), 67.2 (CH
2
H
8
(
)
H
1
3
-3
-1
21
V ) 1635.47(16) Å , D
c
) 1.147 g cm , F(000) ) 616, µMo
)
mol in the temperature range investigated.
Ca lcu la tion s. In the Molecular Mechanics computations
-
1
11
0
.068 cm , T ) 293 K. Data were collected using a graphite
monochromated Mo KR X-radiation (λ ) 0.71073 Å) in the
range 2.15° < θ < 30.08°. Of 21 769 reflections measured, 4791
were found to be independent (Rint ) 0.1053), 1743 of which
were considered as observed [I > 2σ(I)] and were used in the
the dihedral drivers were obtained using a 10° rotational step
over the bonds connecting the quaternary carbon to the
mesitylene rings. In the case of the two diastereoisomers of
derivative 5, a search of the preferred conformation was
performed by making use of the GMMX computer package
(Serena Software, Bloomington, IN). The best structures were
minimized again with the PC Model and cross-checked over
the four possible stereoisomers in order to reduce the chances
of slightly different final energy.
1
refinement of 198 parameters leading to a final R of 0.0405
and a Rall of 0.1469. The structure was solved by direct method
2
and refined by full-matrix least squares on F , using SHELXTL
9
7 program packages. In refinements were used weights
2
2
2
-1
according to the scheme w ) [σ (F
o
) + (0.0589P) + 0.0000P]
)/3. The hydrogen atoms were located
by geometrical calculations and refined using a “riding”
method. wR was equal to 0.1328. The goodness of fit param-
eters S was 0.767. Largest difference density between peak
2
2
where P ) (F
o
) + 2 F
c
Ack n ow led gm en t. Thanks are due to the CNR
I.Co.C.E.A. Institute (Bologna) for access to the 400
MHz spectrometer. Financial support has been received
from MURST (national project “Stereoselection in Or-
ganic Synthesis”) and from the University of Bologna
2
-
3
and hole was 0.157 and -0.152 eÅ . Crystallographic data
(excluding structure factors and including selected torsion
angles) have been deposited with the Cambridge Crystal-
lographic Data Center, CCDC 164872.
(Funds for selected research topics 1999-2001).
NMR Mea su r em en ts. The assignment of the 13C signals
was carried out by DEPT sequence. The samples for the low-
temperature measurements were prepared by connecting to
a vacuum line the NMR tubes containing the desired com-
J O010420M
(
(
20) QCPE program No. 633, Indiana University, Bloomington, IN.
21) Grilli, S.; Lunazzi, L.; Mazzanti, A. J . Org. Chem. 2000, 65,
pounds dissolved in some C
6
D
6
or CD
2
Cl
2
for locking purpose
3653.