Rotation in biphenyls with a single ortho-substituent

Article abstract of DOI:10.1021/jo060205b

Barriers to rotation in a range up to 15.4 kcal mol^-1 were determined by dynamic NMR spectroscopy for a series of biphenyl compounds 1a-1h and 2a-2d with a single ortho-substituent. Ab initio calculations reproduce these barriers satisfactorily

Full text of DOI:10.1021/jo060205b

Rotation in Biphenyls with a Single Ortho-Substituent
Andrea Mazzanti,*,1 Lodovico Lunazzi,1 Mirko Minzoni,1 and J. Edgar Anderson*,2
Department of Organic Chemistry “A. Mangini”, UniVersity of Bologna, Viale Risorgimento 4,
20196-Bologna, Italy, and Chemistry Department, UniVersity College, Gower Street,
London WC1E 6BT, U.K.
mazzand@ms.fci.unibo.it; j.e.anderson@ucl.ac.uk
ReceiVed January 31, 2006
Barriers to rotation in a range up to 15.4 kcal mol-1 were determined by dynamic NMR spectroscopy for
a series of biphenyl compounds 1a-1h and 2a-2d with a single ortho-substituent. Ab initio calculations
reproduce these barriers satisfactorily and indicate the ground-state conformation of these molecules.
Results are discussed in terms of the contribution of individual substituents to the barrier and of the
buttressing of adjacent positions in a benzene ring by substituents.
Introduction
of such highly substituted biphenyls, based largely on measuring
rates of racemization or mutarotation.7 In practice, studies were
thus limited to enantiomers with a lifetime of a minimum of a
few minutes at room temperature, that is, with rotational barriers
greater than about 20 kcal mol-1.8 This usually meant biphenyl
derivatives with three or four ortho-substituents, although, it
was soon shown that lifetimes are sufficiently long for study
with only two suitably large ortho-substituents.9 Ideas on the
size of substituents were developed from these studies,5,10b
together with consideration of the additivity of the contribution
Hindered rotation about the central bond in ortho-substituted
biphenyls and the consequences thereof are classical fields of
study that led to an understanding of many aspects of stereo-
chemistry and molecular conformations. In 1922, Christie and
Kenner3 prepared and isolated two diastereomeric brucine salts
from the tetrasubstituted 6,6′-dinitrobiphenyl-2,2′-dicarboxylic
acid, which they associated with the two enantiomeric structures
now labeled 3M and 3P of the parent acid (see Chart 1). This
and subsequent work4,5 clarified the three-dimensional structure
of these enantiomers as in these now familiar diagrams and
demonstrated experimentally for the first time that there is a
considerable barrier to rotation about a single bond, in this case
the one joining the two rings. The term atropisomers was
subsequently suggested to describe these enantiomeric struc-
tures,4 which can be chemically separated since rotation about
the central single bond that brings about their interconversion
is slow on the chemical time scale. Following Christie and
Kenner’s first example, there were many further investigations5,6
(6) Oki, M. Top. Sterochem. 1983, 14, 1.
(7) A racemization study might involve, for example, preferential
crystallization of a salt of one enantiomer of an acidic biphenyl derivative
and a homochiral alkaloid base to give a more-or-less pure single diastereo-
atropisomer, the other one remaining in solution. Careful regeneration of
the biphenyl derivative from the salt might yield the more-or-less pure
optically active biphenyl. A solution of this material loses optical activity
at a rate that can be measured and depends on the barrier to biphenyl
rotation. The biphenyl derivative has racemized. If the rotation barrier is
rather low, the precipitated salt itself may be studied. The initial optical
rotation of a sample solution is observed to change, but does not vanish,
settling eventually at a new value. Rotation about the biphenyl bond leads
to a mixture of two diastereo-atropisomers, but there is residual optical
activity, since different amounts of two diastereomersssalts of the homo-
chiral basesare present and are eventually at equilibrium. This is mutaro-
tation.
(1) University of Bologna.
(2) University College London.
(3) Christie, G. H.; Kenner, J. J. Chem. Soc. 1922, 121, 964.
(4) Kuhn, R. In Stereochemie; Freudenberg, H., Ed.; Franz Deutike:
Leipzig-Wien, 1933; pp 803-824.
(8) In early work, results were discussed in terms of half-lives for
racemization rather than barriers to rotation about the central biphenyl bond.
(5) Adams, R.; Yuan. H. C. Chem. ReV. 1933, 12, 261.
10.1021/jo060205b CCC: $33.50 © 2006 American Chemical Society
Published on Web 06/29/2006
5474
J. Org. Chem. 2006, 71, 5474-5481

Rotation in Biphenyls with an Ortho-Substituent
CHART 1
of each substituent to the barrier measured for a given
compound. The subject was regularly reviewed10 and became
an important example in the teaching of stereochemistry by the
middle of the 20th century.10a,11 The measuring of large numbers
of barriers and thus the systematic demonstration of substituent
effects on barriers allowed one of the earliest physical-organic
discussions of steric and electronic interactions and the complex
interrelationship of these.10,11
From 1963 onward, with the advent of dynamic NMR
spectroscopy, barriers much lower than 20 kcal mol-1 could be
measured, without resolution of enantiomers, so many more
compounds with one ortho-substituent in each ring were
studied.12 This phase culminated with the systematic study by
Bott, Field, and Sternhell (BFS)13 of a large series of biphenyls
with one ortho-substituent in each ring (see 4, in Chart 1) in
which the relatively rigid five-membered ring serves as an NMR
probe. They measured barriers in 26 compounds where X )
methyl, and six compounds where X ) methoxy. They
examined the additivity of substituent effects on the size of the
rotational barrier by dissecting the experimental barrier for a
compound into contributions from each ortho-substituent, as we
discuss below. Furthermore, on the basis of the contribution of
each substituent to the experimental barrier, they proposed the
concept of the effective van der Waals radius of a substituent,
which has proved a useful steric quantifier, for such substituents
that are not symmetrical.14
Interest in the subject has continued to develop with, for
example, regular reports of naturally occurring biphenyl deriva-
tives with two or more ortho-substituents,15 where conformations
and rotational barriers are of significance.
From the rotational barriers in disubstituted biphenyls that
they measured experimentally, BFS had assigned an interference
value IX-H to each of a variety of ortho-substituents X interacting
with an ortho-hydrogen atom H of the adjacent benzene ring,
in the coplanar transition state for rotation of the biphenyl. When
added together for two substituents, these I-values satisfactorily
reproduced rotational barriers for a range of disubstituted
biphenyls that they had studied,16 plus examples from the work
of others.13 They also suggested an interference value for
hydrogen as a substituent, that is IH-H, from the rotational barrier
of 2 kcal mol-1 in biphenyl itself.17 This begs the question, is
(9) (a) Lesslie, M. S.; Turner, E. E. J. Chem. Soc. 1932, 2021. (b) Lesslie,
M. S.; Turner, E. E. J. Chem. Soc. 1932, 2394. (c) Turner, E. E. Chem.
Ind. 1932, 51, 435. (d) Corbellini, A.; Angeletti, A. Atti Accad. Lincei 1932,
15, 968. (e) Searle, N. E.; Adams, R. J. Am. Chem. Soc. 1933, 55, 1649.
(f) Lesslie, M. S.; Mayer, U. J. H. J. Chem. Soc. 1962, 1401.
(10) (a) Shriner, R. L.; Adams, R. Optical Isomerism. In Organic
Chemistry, 2nd ed.; Gilman, H., Ed.; Wiley: New York, 1943; p 343. (b)
Westheimer, F. H. Calculation of the Magnitude of Steric Effects in Steric
Effects in Organic Chemistry; Newman, M. S., Ed.; Wiley: New York,
1956; Chapter 12. (c) Eliel, E. L.; Wilen, S. H. Stereochemistry of Organic
Compounds; Wiley: New York, 1994; Section 14.5 and passim.
(11) See, for example: Finar, L. Stereochemistry of Biphenyl Com-
pounds. In Organic Chemistry, 5th ed.; Longman: New York, 1975; Chapter
5, Vol. 2, p 215. The first edition was 1956. Other mid-20th century
textbooks deal with the subject at a similar length, while more modern ones
almost invariably mention the topic, but more briefly.
(12) (a) Meyer, W. L.; Meyer, R. B. J. Am. Chem. Soc. 1963, 85, 2170.
(b) Kessler, H. Angew. Chem., Int. Ed. Engl. 1970, 9, 219. (c) Oki, M.;
Akashi, K.; Yamamoto, G.; Iwamura H. Bull. Chem. Soc. Jpn. 1971, 44,
1683 (d) House, H. O.; Campbell, W. J.; Gall, M. J. Org. Chem. 1970, 35,
1815. (e) Colebrook, L. D.; Jahnke, J. A. J. Am. Chem. Soc. 1968, 90,
4687. (f) Oki, M.; Yamamoto, G. Bull. Chem. Soc. Jpn. 1971, 44, 266.
(13) Bott, G.; Field, L. D.; Sternhell, S. J. Am. Chem. Soc. 1980, 102,
5618.
(14) Science Citation Index reports over 200 citations of ref 13 by August
2005.
(15) (a) Leroux (ChemBioChem 2004, 5, 644) gives a brief summary of
recent investigations as well as some ab initio calculations. (b) Bringmann,
G.; Mortimer, A. J. P.; Keller, P. A.; Gresser, M. J.; Garner, J.; Breuning,
M. Angew. Chem., Int. Ed. 2005, 44, 5384.
(16) These barrier values were of course the source of the I-values, but
some substituents appeared several times and there is a self-consistency
about the I-value determined.
J. Org. Chem, Vol. 71, No. 15, 2006 5475

Mazzanti et al.
SCHEME 1
rotation in mono- and bisubstituted biphenyls similar enough
that the rotational barrier in the former case can be deduced
from the sum of the BFS interference values for the substituent
and for hydrogen? There were no reports of barriers to rotation
in biphenyls with only one ortho-substituent,18,19a so we set out
to examine a series of such compounds with a range of common
substituents. We will show by dynamic NMR studies that,
experimentally, barriers are always significantly less than the
BFS sum IX-H +IH-H
.
Beyond the topic of substituent effects on rotational barriers
and additivity when there is more than one ortho-substituent,13
we will discuss the conformation of the central bond, that is to
say the relative torsion of the two rings, and various aspects of
substituent buttressing.20
The coplanar conformation for any biphenyl with a single
ortho-substituent X has maximum destabilization from the steric
interaction of X with the nearer ortho-hydrogen on the second
phenyl ring and maximum stabilization from the π-π inter-
action of the two rings. In the orthogonal conformation, these
opposing contributions to the molecular energy are both at a
minimum, so the overall lowest energy conformation for the
biphenyl will be somewhere between these extremes, more or
less near to orthogonal, depending on the size of X.21,22
That conformation can be conveniently defined by the C2-
C1-C1′-C2′ torsion angle, τ. Inspection suggests and calcula-
tions confirm that the ground state has two doubly degenerate
minima for the conformation of the biphenyl bond. These are
the enantiomeric syn structures b and d and the enantiomeric
anti structures a and c of Scheme 1. Librational motion that
interconverts a with b, and c with d, via a transition state where
the phenyl rings are mutually orthogonal, is rapid compared
with the conventional biphenyl rotation that interconverts a with
c, or b with d, via a coplanar transition state. Such libration,
with different lifetimes in the two interconverting states, is
implicit in our subsequent discussions, and biphenyl rotation is
thus represented by the interconversion of enantiomeric struc-
tures M and P.
Me). In the NMR spectrum of compounds 1, the isopropyl
methyl signals change from being isochronous to anisochronous
as interconversion of atropisomers M and P changes from being
fast to being slow on the NMR time scale as the temperature is
lowered. Variation of the spectrum with temperature thus allows
determination of the rotational barrier.
The 3-isopropyl group in 1 adopts two rapidly interconverting
conformations with the methine hydrogen in the plane of the
benzene ring. According to MM3 calculations, the conformation
with that hydrogen pointing toward the 4-position is slightly
more stable than that where it points to the 2-position, although
ab initio calculations suggest that these two conformers have
the same energy. Since the barrier to rotation of an isolated
isopropyl group on a benzene ring is about 1.5 kcal mol-1 23
,
interconversion of these conformations is rapid on the NMR
time scale at all accessible temperatures. The possibility of these
two isopropyl group conformations and of syn-anti libration
will not be mentioned again, and both are implicit in our
discussions henceforth.
In contrast, each of the terphenyls 1h and 2a-2d exists as a
mixture of stereoisomeric cis and trans-conformations,24 rather
different25 from the syn- and anti-conformations of Scheme 1,
as shown in Figure 1. When interconversion of these, i.e.,
rotation about both phenyl-phenyl bonds,19b is slow on the
NMR time scale, two sets of signals of different intensities are
seen in the NMR spectrum, reflecting the relative population
of these two conformations, and the rotation barrier can be
derived from changes in the spectrum with temperature.19a
Buttressing, now widely recognized in many intramolecular
situations, was first postulated as the indirect effect of substit-
uents in the meta- and para-positions on the biphenyl rotation
Results
We now report results for rotation in compounds with a single
ortho-substituent, i.e., the two series 1a-1h and 2a-2d,19a and
for compound 5 with two ortho-substituents (Chart 1). The latter
was prepared to make a comparison with the equivalent
disubstituted compound 6 in the BFS series (i.e. 4, X ) Y )
(17) Katon, J. E.; Lippincott, E. R. Spectrochim. Acta 1959, 19, 627,
with a temperature correction as described.13
(18) Mutarotation7 of a camphorsulfonate salt of a biphenyl with a single
ortho-substituent, trimethylarsonium, was reported at a surprisingly early
date (Lesslie, M. S.; Turner, E. E. J. Chem. Soc. 1933, 1588) but were not
quantified.
(19) (a) A preliminary account of some of this work (compounds 2a-
2d) is given: Lunazzi, L.; Mazzanti, A.; Minzoni, M.; Anderson, J. E. Org.
Lett. 2005, 7, 1291. (b) The ∆Gq values reported in ref 19a are obtained
from the interconversion rates between the cis and trans atropisomers, an
interconversion that can take place by rotation of either of the two terminal
aryl groups. Comparison with compounds of type 1 requires the rate constant
for rotation of only one of the two aryl groups, which is obviously half of
the measured atropisomerization value. Thus the values to be used here for
the comparison of 2a-2d with 1a-1d are RT ln 2 ) 0.2-0.4 kcal mol-1
higher (i.e., 7.0, 8.1, 10.2, and 15.0 kcal mol-1 for 2a, 2b, 2c, and 2d,
respectively) than those reported in ref 19a.
(23) Schaefer, T.; Sebastian, R.; Penner, G. H. Canad. J. Chem. 1988,
66, 1495.
(24) The central ring may be ortho-substituted (1h), para-substituted (2a-
2d), or meta-substituted. There are two cis-structures (both-X-up and both-
X-down) and two anti-structures (up, down and down, up)25 but that need
not be taken further account of in our discussions.
(20) Rieger, M.; Westheimer, F. J. Am. Chem. Soc. 1950, 72, 19.
(21) When there is an ortho-substituent in each ring, the ground state
conformation is expected to be close to orthogonal (τ ) 90°), except when
both substituents are fluorine.15a,22
(25) In our preliminary communication19a we used syn and anti as labels
for 2a-2d stereoisomeric conformations, but with the use of these labels
in Scheme 1, we have changed these labels as cis and trans (Figure 1) in
the hope of avoiding confusion.
(22) Grein, F. J. Phys. Chem. A 2002, 106, 3823.
5476 J. Org. Chem., Vol. 71, No. 15, 2006

Rotation in Biphenyls with an Ortho-Substituent
TABLE 1. Computed (ab initio) and Experimental Barriers ((0.2 kcal mol-1) for the Interconversion of the Biphenyl Atropisomers 1a-1h
and 5h
barrier
∆ν (in Hz at
150.8 MHz)
compound
experimentala
computed
interpolatedb
τc
1a (X ) Me)
1b (X ) Et)
1c (X ) i-Pr)
1d (X ) t-Bu)
1e (X ) Cl)
1f (X ) Br)
1g (X ) I)
1h (X) 3-i-PrPh)
2af
41 at -146°
7.4 [-130°, -110°]
8.7 [-100°, -90°]
11.1 [-54°, -36°]
15.4 [+20°, +55°]
7.7 [-120°, -115°]
8.75 [-100°, -90°]
9.9 [-73°, -67°]
7.7d [-120°, -100°]
6.8 [-145°, -135°]
7.9 [-120°, -104°]
9.9 [-86°, -60°]
14.6 [-2°, +25°]
18.7 [+81°, +86°]
19.3g [+64°, +87°]
7.1
8.6
11.1
15.3
7.3
7.7
not feasible
7.4
6.7
8.1
10.7
14.8
16.7
17.1
(9.7 + 1) ) 10.7
127
118
122
95
126
127
NA
129, -53
127
117
123
94
32 at -107°
28 at -75°
(12.6 + 1) ) 13.6
(18.3 + 1) ) 19.3
(9.1 + 1) ) 10.1
(10.2 + 1) ) 11.2
(10.9 + 1) ) 11.9
(7.9 + 1)e ) 8.9
39 at -7°
22 at -131°
57 at -120°
67 at -96°
96 and 212 at -135°
2bf
2cf
2df
5
6
18 at +50°
90
91
a The values in square brackets represent the temperature range where line shape simulations yield the rate constants used for the determination of the
barrier. b The sum (IX-H + IH-H) of the interference values13 for substituent X and for hydrogen; see the text. c Computed dihedral angle. d Barrier for the
transition major isomer to minor. e The interference value for the phenyl rather than for the m-isopropylphenyl is used. f Data from ref 19a (see also ref 19b).
g Value from ref 13. h The chemical shift separations are those of the 13C methyl signals of the isopropyl substituent in the meta position. Our earlier results
for 2a-2d and those of 6 discussed in the text are also reported.
FIGURE 1. Conformation of terphenyls 2a-2d. Each biphenyl bond
undergoes libration as in Scheme 1.
barrier.20,26 Such additional substituents bolster the substituent,
or even the hydrogen atom, in the adjacent ortho-position,
destabilizing a coplanar state and thereby raising the biphenyl
rotational barrier.13,20 We will comment on buttressing in our
simply substituted biphenyls.
Dynamic NMR Studies. Barrier measurements for com-
pounds studied are shown in Table 1. A typical example of the
temperature-dependence of the 13C NMR spectra of compounds
1a-1g is shown, for the case of 1a, in Figure 2 (left). The single
FIGURE 2. Left: temperature dependence of the 13C NMR (150.8
line of the isopropyl methyl signal broadens below -100 °C
MHz) isopropyl methyl signal of 1a in CHF2Cl/CHFCl2. Right: selected
examples of line shape simulation obtained with the rate constants
indicated.
and eventually splits into a pair of equally intense peaks at -146
°C. On the right are shown line shape simulations obtained with
the rate constants indicated, from which the ∆Gq value (7.4 kcal
mol-1 as in Table 1) was obtained.
The terphenyl compounds 2a-2d of ref 19a show spectral
The spectra of compound 1h shows an additional feature to
all other compounds 1, since the cis- and trans-rotational
conformers yield separate sets of signals of different intensity
at low temperatures. Thus at -135 °C the 13C signal of the
isopropyl group appears as two doublets of relative intensity
60:40 with doublet separation of 96 and 212 Hz, respectively
(see Figure 3 and Table 1).
On the basis of ab initio computations, wherein the trans-
conformation is more stable than the cis- (albeit by only 0.02
kcal mol-1, which is well below the uncertainty of the computed
energy values), we tentatively assign the more intense signals
to the trans-conformer of 1h.
changes similar to those for 1h and these give rise to the barriers
reported in Table 1. The barrier in compound 5 is 18.7 kcal
mol-1, whereas in the equivalent BFS compound 6 it is 19.3
kcal mol-1
.
Molecular mechanics27 and ab initio28 calculations were
carried out for ground state conformations and for rotation of
the biphenyl bond in the compounds studied experimentally and
some others. Table 1 shows relevant values of the barriers to
rotation (the computed dihedral angles that define each ground
state conformation are reported in Table 1).
Discussion
Ground-State Conformation. For the same ortho-substituent,
(26) Chien, S. L.; Adams, R. J. Am. Chem. Soc. 1934, 56, 1787.
the calculated torsion angles (Table 1) are not consistently
J. Org. Chem, Vol. 71, No. 15, 2006 5477

Mazzanti et al.
SCHEME 2
represents the rotation needed to reach the coplanar transition
state. The smaller this torsion angle, the greater the relative steric
strain already present in the ground state, so the smaller the
steric contribution to the observed barrier. On the other hand,
a small torsion angle represents enhanced π-π stabilization in
the ground state, so there is less additional electronic stabilization
offsetting the steric interaction in the coplanar transition state.
This would lead to a somewhat higher barrier, but since in all
compounds studied, the steric strain of the planar conformation
is greater than the electronic stabilization, the expectation is
that the smaller the torsion angle in the ground state, the smaller
the barrier; Table 1 shows that this is in fact the case (see also
Supporting Information).
While all monosubstituted biphenyls, except 1d and 2d, are
calculated to have ground-state torsion angles less than 90°, in
the 50°-80° range, biphenyls with two ortho-substituents22
usually have torsion angles close to 90°. These two tendencies
mean that rotation has significantly different features in the
mono- and disubstituted series.
FIGURE 3. Left: temperature dependence of the 13C NMR (150.8
MHz) isopropyl methyl signal of 1h in CHF2Cl/CHFCl2 showing two
conformers in a 60:40 ratio at -135 °C. Right: selected examples of
line shape simulation obtained with the rate constants indicated.
Barriers in Monosubstituted Biphenyls. The steric inter-
actions during rotation in a substituted biphenyl are not as simple
as a transition-state diagram with coplanar rings might imply.
It is well-known29 that if structural constraints are great, the
benzene ring is relatively easily distorted and also that exocyclic
bond angles can be significantly altered by atom displacement
in or out of the plane. Thus during rotation, the maximum
interaction of the 2-position with the adjacent 2′-position need
not coincide with the maximum interaction of the 6- and 6′-
positions. Nonetheless, the size of the barriers along the series
1 clearly fits with widely accepted ideas of the steric size of
substituents.
The variation of barriers is particularly markedsmore than
doublingswith the alkyl groups of 1a-1d and 2a-2d. Alkyl-
phenyl bond rotation takes place relatively easily, and this helps
to reduce congestion when the two phenyl groups are more or
less coplanar. Further relief of strain comes from splaying apart
of the C1-phenyl and C2-X bonds. As biphenyl rotation
continues, reversal of the alkyl group rotation (or perhaps in
the case of a methyl or tert-butyl group, completion of 120° of
rotation) and relaxation of the bond splaying deliver the
molecule to the enantiomeric ground state conformation.
When such an o-alkyl group is methyl, ethyl, or isopropyl in
1a-1c and 2a-2c, it no doubt points a hydrogen atom toward
the biphenyl bond, as in Scheme 2.
different in the terphenyl series 2 and in the m-isopropyl series
1. We determined the crystal structure conformation for
compounds 2a and 2d, and the dihedral angles τ measured (55°
and 99.5°) are in close agreement with the values calculated,
55° and 93°, respectively.19a
Except with the biggest tert-butyl substituent, i.e., 1d or 2d,
the rings are quite far removed from orthogonal, showing that,
in the monosubstituted biphenyls, there is a significant conflict
between steric repulsion, favoring an orthogonal arrangement,
and conjugative interaction of the two phenyl rings, favoring a
coplanar system. It has been calculated22 that in biphenyl itself
(at the B3LYP/6-311+G(d) level) the ground-state torsion angle
is 43°, and the coplanar and orthogonal states are 2.17 and 1.79
kcal mol-1 less stable, suggesting that these two effects are of
a comparable size when the ortho-substituents are hydrogen
atoms.
The barrier to rotation in a monosubstituted biphenyl should
reflect somewhat the ground-state torsion angle τ, which
(27) (a) Allinger, N. L.; Yuh, Y. H.; Lii, J.-H. J. Am. Chem. Soc. 1989,
111, 8551, 8566, 8576. (b) Allinger, N. L.; Li, F.; Yan, L.; Tai, J. C. J.
Comput. Chem. 1990, 11, 868.
(28) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K.
N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.;
Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.;
Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.;
Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li,
X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.; Adamo, C.;
Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.;
Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.;
Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich,
S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A.
D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A.
G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.;
Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham,
M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.;
Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian
03, Revision C.02, Gaussian, Inc., Wallingford CT, 2004.
This apparently similar arrangement leads however to dif-
ferent interconversion barriers of 7.4, 8.7, and 11.1 kcal mol-1
for 1a, 1b, and 1c and 6.8, 7.9, and 9.9 kcal mol-1 for 2a, 2b,
and 2c, respectively.19b The increase in barrier in going along
each series reflects the increasing difficulty of the postulated
(29) Qiao, X.; Padula, M. A.; Ho, D. M.; Vogelaar, N. J.; Schutt, C. E.;
Pascal, R. A. J. Am. Chem. Soc. 1996, 118, 741. This paper reports two
extreme examples of distorted biphenyl-like structures, octaphenylnaph-
thalene and decaphenylanthracene, and discusses the distortions in their
structures and earlier work.
5478 J. Org. Chem., Vol. 71, No. 15, 2006

Rotation in Biphenyls with an Ortho-Substituent
TABLE 2. Rotational Barriers (kcal mol-1) in Biphenyls with One and with Two Ortho-Substituentsa
monosubs tituted compd
disubstituted compounds
ortho-
substituent
exp barrier U
ortho-
substituent
exp barrier Q
barrier for disubstituted
from monosubstituted
(U + Q - 2.0)
exp barrier (ref 13) at
the mean temp of
columns 2,4
sum of interference
values (ref 13)
IU,H + IQ,H
1a
1a
1a
1a
1a
1a
1c
1c
Me 7.4 [153]
Me 7.4 [153]
Me 7.4 [153]
Me 7.4 [153]
Me 7.4 [153]
Me 7.4 [153]
i-Pr 11.1 [228]
i-Pr 11.1 [228]
1g
1f
I 9.9 [203]
15.3
14.2
13.1
12.8
16.5
13.1
20.2
24.5
17.0 [178]
17.7 [165]
14.9 [155]
15.7 [153]
17.4 [190]
13.4 [158]
25.5 [228]b
30.8 [270]b
20.5
19.8
17.8
19.2
22.1
17.5
25.0
30.8
Br 8.8 [178]
Cl 7.7 [156]
Me 7.4 [153]
i-Pr 11.1 [228]
Ar 7.7 [163]
i-Pr 11.1 [228]
t-Bu 15.4 [311]
1e
1a
1c
1h
1c
1d
a The values in square brackets are the absolute temperatures. b These experimental barriers are from ref 32 corrected to the temperature (T) shown using
an entropy of activation of -11.0 eu.32
changes in the alkyl group conformation during biphenyl
rotation, either by alkyl group rotation or by splaying. The
hydrogen atom in the adjacent C3-position buttresses the alkyl
group against both these changes in shape.
Rotation barriers also increase, although rather less markedly
with the size of the halogen atom (see 1e-1g). It is not
uncommon30 that the relative effect of chlorine, bromine, and
iodine in rotations and in conformational equilibria is less
marked than their covalent radii might suggest. Examples of
alkyl group rotation in neopentylbenzenes with two halogen
atoms as ortho-substituents give barriers31 in exactly the same
ratio as in the present biphenyls.
Activation entropies for rotation in biphenyls determined by
classical methods shed light on this problem. Hall and Harris35
have reviewed the literature36 and showed that classically
determined activation entropies for four disubstituted biphenyls
averaged -6.6 eu while for eight trisubstituted and four
tetrasubstituted biphenyls, the average values are -11.6 and
-13.8 eu, respectively. These averages are striking and do
confirm the need for caution in dynamic NMR derivations of
activation entropies. Furthermore, the progression in the aver-
ages with substitution indicates that an activation entropy of
near to zero in monosubstituted biphenyls is plausible.
If rotational barriers in monosubstituted biphenyls are derived
from the sum of the BFS interference values for that substituent
and for hydrogen, the predictions (see the interpolated values
in the last column of Table 1) are invariably higher than
Barriers in Ortho-Disubstituted Biphenyls. It is interesting
to compare the barriers for biphenyls with two ortho-substituents
with those for the two corresponding monosubstituted com-
pounds, and relevant results are shown in Table 2.
experimental values, usually by much more than 1 kcal mol-1
.
Since the BFS interference values do serve well for predicting
results for further examples of disubstituted biphenyls, there
must be some factor that differs between the disubstituted and
monosubstituted series. This is buttressing, which is progres-
sively greater with increasing substitution, but whose effects
can be observed even in simply substituted compounds.
Buttressing. The structural phenomenon of buttressing was
first demonstrated with polysubstituted biphenyls26 and is
particularly associated with benzene rings where the relative
position of adjacent substituents on a rigid framework is well-
If rotational barriers in the disubstituted series are predicted
from the sum of barriers in the appropriate monosubstituted
cases, reduced by 2.0 kcal mol-1 (i.e., the rotational barrier in
biphenyl itself due to the interaction of two pairs of coplanar
hydrogen atoms), the outcomes are always markedly lower than
the experimental values.13 We believe that our monosubstituted
biphenyl barriers serve for predicting only a lower limit for those
in the disubstituted series.
There is however a problem associated with a comparison
based on such experimental rotational barriers. From the nature
of the dynamic NMR experiment, disubstituted biphenyl barriers
are determined in a markedly higher temperature range ∆T than
the two compared monosubstituted biphenyls. If there is a
significant entropy of activation ∆Sq for rotation, one of the
two barriers compared has to be adjusted to the temperature of
the other by ∆T∆Sq. We have found that free energies of
activation for series 1 and 3, inevitably determined over narrow
temperature ranges, are essentially constant within the limits
of experimental error,33 so for lack of better information we
take ∆Sq to be 0 in these series.
(33) As often observed in conformational processes, the ∆Gq value was
found to be independent of temperature within the experimental uncertainty
of the NMR measurements. See: Hoogosian, S.; Bushweller, C. H.;
Anderson, W. G.; Kigsley, G. J. Phys. Chem. 1976, 80, 643. Lunazzi, L.;
Cerioni, G.; Ingold, K. U. J. Am. Chem. Soc. 1976, 98, 7484. Forlani, L.;
Lunazzi L.; Medici, A. Tetrahedron Lett. 1977, 18, 4525. Bernardi, F.;
Lunazzi, L.; Zanirato, P.; Cerioni, G. Tetrahedron, 1977, 33, 1337. Lunazzi,
L.; Magagnoli, C.; Guerra, M.; Macciantelli, D. Tetrahedron Lett. 1979,
3031. Cremonini, M. A.; Lunazzi, L.; Placucci, G.; Okazaki, R.; Yamamoto,
G. J. Am. Chem. Soc. 1990, 112, 2915. Anderson, J. E.; Tocher, D. A.;
Casarini, D.; Lunazzi, L. J. Org. Chem. 1991, 56, 1731. Borghi, R.; Lunazzi,
L.; Placucci, G.; Cerioni, G.; Foresti, E.; Plumitallo, A. J. Org. Chem. 1997,
62, 4924. Garcia, M. B.; Grilli, S.; Lunazzi, L.; Mazzanti, A.; Orelli, L. R.
J. Org. Chem. 2001, 66, 6679. Garcia, M. B.; Grilli, S.; Lunazzi, L.;
Mazzanti, A.; Orelli, L. R. Eur. J. Org. Chem. 2002, 4018. Casarini, D.;
Rosini, C.; Grilli, S.; Lunazzi, L.; Mazzanti, A. J. Org. Chem. 2003, 68,
1815. Casarini, D.; Grilli, S.; Lunazzi, L.; Mazzanti, A. J. Org. Chem. 2004,
69, 345. Bartoli, G.; Lunazzi, L.; Massacesi, M.; Mazzanti, A. J. Org. Chem.
2004, 69, 821. Casarini, D.; Coluccini, C.; Lunazzi, L.; Mazzanti, A.;
Rompietti, R. J. Org. Chem. 2004, 69, 5746.
BFS on the other hand reported a mean ∆Sq value of -18.8
( 6.7 eu for the series of disubstituted biphenyls that they
measured.13 The difficulties in determining ∆Sq from dynamic
NMR experiments are well-known and lead to an excessively
negative value.34
(30) Anderson, J. E.; Doecke, C. W.; Pearson, H. J. Chem. Soc., Perkin
Trans. 2 1976, 336.
(31) Nilsson, B.; Martinson, P.; Olsson, K.; Carter, R. E. J. Am. Chem.
Soc. 1974, 96, 3190.
(34) Allerhand, A.; Gutowsky, H. S.; Jonas, J.; Meinzer, R. J. Am. Chem.
Soc. 1966, 88, 3185.
(35) Hall, D. M.; Harris, M. M. J. Chem. Soc. 1960, 490. See also later
classical measurements.36
(32) Wolf, C.; Hochmuth, D. H.; Konig, W. A.; Roussel, C. Liebigs Ann.
Chem. 1996, 357.
(36) (a) Lesslie, M. S.; Mayer, U. J. H. J. Chem. Soc. 1961, 611. (b)
Cheung King Ling, C.; Harris, M. M. J. Chem. Soc. 1964, 1825.
J. Org. Chem, Vol. 71, No. 15, 2006 5479

Mazzanti et al.
ortho-group as the only substituent.19b Similar basic barriers for
SCHEME 3
o-halogen substituents can be obtained by reducing the series 1
barrier by 0.6 kcal mol-1
.
Another kind of buttressing occurs in a system like that in
Scheme 3 (right) when the process is represented as B-B bond
rotation. Groups A and C will not only affect the process at B
by direct steric interaction, but each will also restrict distortion
at B to accommodate the other. The A-B and B-C interactions
will buttress each other to increase rotational barriers. A
comparison with our results for monosubstituted compounds
suggests that this is what happens in biphenyls with an ortho-
substituent in each ring.
The experimental barrier values for such disubstituted com-
pounds contain a substantial contribution from mutual buttress-
ing that is inevitably much diminished in the monosubstituted
case when one substituent is hydrogen. It is thus unlikely that
there is a simple link between the rotational barriers of a series
of monosubstituted biphenyls and of any member of the
corresponding set of disubstituted biphenyls.
defined. It is essentially an effect of steric interaction of these
substituents, although any one substituent may have an electronic
effect as well.
In any system of three groups, A, B, and C, as in Scheme 3
(left), when a process at A is studied, the process being a
chemical reaction or a movement (here the process is rotation
about the bond to A), B has a direct steric effect on this process.
Group C, as a meta-substituent in a biphenyl for example, is
too distant from the rotating phenyl ring A to exert any direct
steric compression (van der Waals repulsion), but the very
presence of C, or introducing a larger substituent as C, may
increase the effect that B has on the process at A, and this is
known as the buttressing of B by C: the effect of the interaction
between B and C on a process at A.
Although hydrogen as a substituent B or C is often taken as
the starting point in discussing buttressing by larger substituents,
it may itself be involved in buttressing. Thus, the experimental
rotational barriers for the meta-substituted compounds 1a-1d
are each higher than in the para-substituted terphenyl compounds
2, by 0.6 kcal mol-1 on average.19b Ab initio calculations for
the same compounds suggest that the barriers are 0.45 kcal
mol-1 higher on average in the meta-substituted series 1.
Molecular mechanics calculations for 2-chlorobiphenyl suggest
that the barrier is 9.6 kcal mol-1, whereas in meta-substituted
analogues, 1e and 7, which also have a chlorine as a single
ortho-substituent, the barrier is somewhat higher at 9.9 kcal
mol-1. In 5, where there is an isolated isopropyl group as a
meta-substituent, free to rotate so that its methyl groups need
not be near the o-hydrogen atom, we measured a barrier of 18.7
kcal mol-1. In compound 6, whose structure around the biphenyl
bond is superficially similar to 5 but has the meta-substituent
held rigid with methyl groups toward the o-hydrogen atom, the
barrier is 19.3 kcal mol-1. This varied evidence suggests that
the buttressing effect of an isolated meta-substituent (such as
is used as an NMR probe for the biphenyl rotation) is small
but real and is transmitted by the o-hydrogen atom.37
The extent to which the BFS interference value for a
substituent, derived from disubstituted biphenyls, is greater than
the monosubstituted biphenyl rotation barrier (see columns 3
and 5 of Table 1) can thus be taken as a measure of the
buttressing effect of a substituent. On this basis, the phenyl group
has a small effect (0.2 kcal mol-1), halogen atoms have a
similarly sized intermediate effect (1.0-1.4 kcal mol-1), and
alkyl groups have larger effects (1.5-2.9 kcal mol-1).
Calculations. Computations of barriers to rotation in mono-
substituted biphenyls have an unusual outcome. Molecular
mechanics calculations (MM3) driving the inter-ring dihedral
angle invariably lead to a rotational barrier that is significantly
higher than experimental values. While it may be that driving
the torsion angle needs to be accompanied by some rotation of
the ortho-substituent to better reproduce what happens experi-
mentally during rotation, this cannot be the case for simple
ortho-substituents such as halogen, where the calculated barriers
are also too high. In the chloro compound 1e, for example, the
calculated barrier is 9.9 kcal mol-1, more than 25% higher than
the experimental value of 7.7 kcal mol-1
.
If, however, the structure for the ground state and transition
state are calculated by ab initio methods (B3LYP/6-31G(d)
level28), optimization of these latter calculations leads to a
calculated barrier that agrees well with the experimental values
for the monoalkylbiphenyls and reasonably well for the mono-
halobiphenyls.39
Given the size of this buttressing by an isolated meta-
substituent, any intrinsic buttressing effect by an isolated para-
substituent in series 2, transmitted by intervening m- and
o-hydrogens, should be negligible. An electronic effect on the
rotation barrier might be expected,38 but the p-phenyl substituent
in series 2 is far from coplanar, which reduces any electronic
interaction, and this does not necessarily change during the
biphenyl rotation being observed. We thus suggest that the
barriers measured in the series 2 are close approximations to
the basic barriers, those to be expected in a biphenyl with an
Conclusions
A series of biphenyl compounds with a single ortho-
substituent adopts a ground state conformation where the
biphenyl dihedral angle varies between about 40° and 90°,
(39) A reviewer has suggested that for the series of chiral compounds
1a-1g, the experimental ∆Gq include a ∆Sq contribution of -R ln 2 from
the loss of the entropy of mixing of the enantiomeric ground-state structures
with a planar, achiral rotational trensition state. However, the two almost
equivalent coplanar transition states should balance the ground-state
degeneracy in such a way that the whole contribution becomes almost
negligible. The reviewer suggests that the loss of entropy might be allowed
for by correcting the experimental barrier by RT ln 2. This would affect
the comparison of the barriers and would reduce the small discrepancy
between the calculated activation enthalpies and the experimental activation
energies for the series 1. It would also reduce the difference between the
experimental barriers for chiral series 1 with a meta-substituent and achiral
series 2, a difference that we have attributed to a buttressing effect of the
meta-substituent.
(37) Stoughton and Adams (J. Am. Chem. Soc. 1932, 54, 4426)
considered the effect of an isolated m-methyl or methoxy substituent on
rotation, but could detect no significant difference in the half-lives for
racemization of appropriate chiral biphenyls. Confusingly, Chien and
Adams26 cited this paper shortly afterwards and stated that Stoughton and
Adams “demonstrated that a group in the 5′-position increased the antipodal
stability of the molecule.”
(38) Gallo, R.; Roussel, C.; Berg, U. AdV. Heterocycl. Chem. 1988, 43,
173.
5480 J. Org. Chem., Vol. 71, No. 15, 2006

Rotation in Biphenyls with an Ortho-Substituent
General Procedure for Compounds 1a-1h and 5. To a
solution of the appropriate ortho-substituted bromobenzene (1 mmol
in 6 mL of benzene) were added K2CO3 (2 M solution, 1.25 mL),
3-isopropylphenylboronic acid (2.5 mmol, suspension in 4 mL of
ethanol), and Pd(PPh3)4 (0.2 mmol) at room temperature (in the
case of 1h, 7.5 mmol of 3-isopropylphenylboronic acid was added).
The stirred solution was refluxed for 2-3 h, the reaction being
monitored by GC-MS. Then CHCl3 and H2O were added, and the
extracted organic layer was dried (Na2SO4) and evaporated. The
crude was prepurified by chromatography on silica gel (hexane).
The analytically pure compounds were obtained from preparative
HPLC on a Kromasil C18 column (250 mm × 10 mm, 5 µm, eluent
CH3CN/H2O). The spectroscopic and analytical data for all the new
compounds are reported in the Supporting Information.
NMR Measurements. NMR spectra were recorded at 600 MHz
for 1H and 150.8 MHz for 13C. The assignments of the 13C signals
were obtained by DEPT and 2D experiments (gHSQC sequence42).
The samples for the 13C NMR low-temperature measurements were
prepared by connecting to a vacuum line the NMR tubes containing
the compound and some C6D6 for locking purpose and condensing
therein the gaseous CHF2Cl and CHFCl2 under cooling with liquid
nitrogen. The tubes were subsequently sealed in vacuo and
introduced into the precooled probe of the spectrometer. The
temperatures were calibrated by substituting the sample with a
precision Cu/Ni thermocouple before the measurements. Complete
fitting of dynamic NMR line shapes was carried out using a PC
version of the DNMR-6 program.43 At least five different temper-
ature spectra were used for the simulations.
reflecting the size of the substituent. Barriers to rotation between
enantiomeric conformations reflect steric size. These barriers
are somewhat smaller than expected from literature values for
biphenyls with an ortho-substituent in each ring. Remote
substituents in the meta-position affect the barrier somewhat.
These last two observations are evidence of differing kinds of
buttressing effects that are present even when three of the four
ortho-positions carry hydrogen atoms.
Experimental Section
Materials. 3-Isopropylboronic Acid. To a solution of 1-bromo-
3-isopropylbenzene (15 mmol in 100 mL of THF) cooled at -78
°C was added 10 mL (16 mmol) of BuLi (1.6 M in hexane). After
2 h the resulting 3-isopropylphenyllithium was transferred dropwise
into a solution of triisopropylborate (45 mmol in 20 mL of THF)
kept at -78 °C. (Note: the transferring line has to be cooled to
avoid decomposition of the organometallic reagent.) When the
addition was terminated, the reaction was maintained for 1 h at
this temperature and subsequently warmed to room temperature
for about 1 h. Finally, the reaction was quenched with aqueous
HCl and extracted with Et2O. The crude product was crystallized
from hexane to yield 1.36 g (10 mmol) of 3-isopropylphenylboronic
acid.
1-Bromo-2-tert-butylbenzene.40 A solution of 2-tert-butylphen-
ylamine (0.1 mol 15.6 mL) in 30 mL of 40% w/w hydrobromic
acid was prepared in a 250-mL round-bottomed flask. After cooling
to 5 °C by immersion in an ice/salt bath, diazotization was
performed by gradual addition of a solution of 15.1 g (0.22 mol)
of sodium nitrite in 20 mL of water, with stoppering of the flask
after each addition and shaking until all red fumes were adsorbed.
The temperature was kept between 5 and 10 °C for 2 h. When the
diazotization was completed, 0.4 g of copper powder was added.
WARNING: the solution was refluxed very cautiously because
of evolution of gas. When the vigorous evolution of nitrogen
moderates, the system was kept at 50 °C for 30 min and was then
diluted with 80 mL of water and extracted with Et2O. The organic
layer was washed with 10% solution of KOH, and the crude purified
by chromatography on silica gel (hexane) and distilled at 85 °C (3
mmHg) to obtain 6.0 g (28.3 mmol) of 1-bromo-2-tert-butylben-
zene.
1-Bromo-p-cymene and 2-Bromo-p-cymene. To a 250-mL
round-bottomed flask containing 63.9 g (0.475 mol) of p-cymene
and 0.2 g of iron powder was added 64 g (0.40 mol) of Br2 slowly
under a nitrogen flux. During the addition the reaction can be
monitored by the vigorous evolution of HBr, and the temperature
should be kept between 50 and 60 °C. When the evolution of gas
ends, the mixture was warmed to 70 °C for about 4 h and then
quenched by addition of 100 mL of a 6 M KOH solution, and the
resulting mixture was extracted with chloroform. The dried solution
(Na2SO4) was evaporated at reduced pressure. The unreacted
p-cymene was removed by distillation (bp 75 °C at 25 mmHg).
The brown crude product was distilled at reduced pressure to obtain
a mixture of the two isomers (bp 116 °C at 19 mmHg).41
Computational Details. Ab initio computations were carried out
at the B3LYP/ 6-31G(d) level by means of the Gaussian 03 series
of programs28 (the standard Berny algorithm in redundant internal
coordinates and default criteria of convergence was employed).
Harmonic vibrational frequency was calculated in order to ascertain
the nature of all the stationary points. For each optimized ground
state the frequency analysis showed the absence of imaginary
frequencies, whereas for each transition state the frequency analysis
showed a single imaginary frequency. The corresponding optimized
structures are reported in the Supporting Information.
Acknowledgment. Financial support was received from the
University of Bologna (Funds for selected research topics and
RFO) and from MIUR-COFIN 2003, Rome (national project
“Stereoselection in Organic Synthesis”).
Supporting Information Available: Analytical and spectro-
scopic data for compounds 1a-1h and 5; Ab initio calculated
ground state dihedral angles and computational details of com-
pounds 1a-1h and 5. This material is available free of charge via
the Internet at http://pubs.acs.org.
JO060205B
(41) Strubell, W. Chem. Techn. 1957, 9, 597.
(42) Wilker, W.; Leibfritz, D.; Kerssebaum, R.; Bermel, W. Magn. Reson.
Chem. 1993, 31, 287.
(43) PC version of the QCPE program no. 633, Indiana University,
Bloomington, IN.
(40) Hou, Y.; Meyers, C. Y. J. Org. Chem 2004, 69, 1186.
J. Org. Chem, Vol. 71, No. 15, 2006 5481