Conformation and stereodynamics of 1,2-diaryltetrahydropyrimidine and of its five- and seven-membered ring analogs

Article abstract of DOI:10.1016/j.tet.2011.09.091

The 1-(2-nitrophenyl)-2-(2-methylphenyl)-1,4,5,6-tetrahydropyrimidine and its five- and seven-membered ring analogs were synthesized and their conformational properties investigated by low temperature NMR spectroscopy and DFT theoretical calculations. Res

Full text of DOI:10.1016/j.tet.2011.09.091

Tetrahedron 67 (2011) 9129e9133
Contents lists available at SciVerse ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Conformation and stereodynamics of 1,2-diaryltetrahydropyrimidine and of its
five- and seven-membered ring analogs
,
b,
*
Jimena E. Díaz b, Nadia Gruber b, Lodovico Lunazzi a, Andrea Mazzanti a , Liliana R. Orelli
*
a Department of Organic Chemistry ‘A.Mangini’, University of Bologna, Viale Risorgimento 4, Bologna 40136, Italy
b
ꢀ
Departamento de Química Organica, Facultad de Farmacia y Bioquímica, Universidad de Buenos Aires, CONICET, Junin 956, Buenos Aires 1113, Argentina
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 7 June 2011
Received in revised form 29 August 2011
Accepted 20 September 2011
Available online 28 September 2011
The 1-(2-nitrophenyl)-2-(2-methylphenyl)-1,4,5,6-tetrahydropyrimidine and its five- and seven-
membered ring analogs were synthesized and their conformational properties investigated by low
temperature NMR spectroscopy and DFT theoretical calculations. Restricted rotation of the aryl sub-
stituents were observed in all cases and the corresponding barriers determined. In the case of the six-
membered ring derivative the additional conformers resulting from a ring inversion process were also
detected.
Keywords:
Ó 2011 Elsevier Ltd. All rights reserved.
Nitrogen heterocycles
Conformational analysis
Axial chirality
Dynamic NMR
DFT calculations
1. Introduction
Tetrahydropyrimidines bearing a variety of substituents display
biological and pharmacological properties1e3 in that they can be
used, for instance, as antidepressants,4 fungicides,5 and anthel-
mintics.6 In the past we had investigated, by NMR spectroscopy, the
conformational and stereodynamic properties of a number of diaryl
derivatives of these compounds7,8 that showed, at low tempera-
ture, the existence of conformational diastereoisomers due to the
restricted rotation about the aryl-pyrimidinyl rings. In the present
work we have investigated the behavior of analogous compounds
having, in addition to the six-membered ring, also the smaller and
the larger five- and seven-membered rings, respectively.
yielding two sharp lines separated by 240 Hz at ꢀ130 ꢁC with
a 51:49 intensity ratio.9 The rate constants of the process (k in sꢀ1),
obtained at various temperatures by a line shape simulation (Fig. 1),
afforded the free energy of activation
D
Gs by applying the Eyring
2. Results and discussion
equation (8.6 kcal/mol).10 As usual for conformational processes, in
this case and in all the cases investigated hereafter the activation
For this purpose compounds 1e4 were synthesized.
energy
temperature range, thus implying a very small (or negligible) ac-
tivation entropy
Ss 11
D
Gs was found to be virtually invariant in the observed
The 1H NMR spectroscopic signals (at 600 MHz in CDF2Cl/
CDFCl2) of compound 1, that are sharp at ambient temperature,
broaden on cooling and split into two groups of lines of almost
equal intensity below ꢀ100 ꢁC. This effect is most clearly visible in
the case of the methyl signal, which decoalesces below ꢀ90 ꢁC,
D
.
The observation of separate lines implies the existence of two
conformers that were interpreted as due to the restricted rotation
of both aryl groups. In fact, the conformation adopted is expected to
have the aryl planes not coplanar with the averaged dynamic plane
(due to the extremely rapid inversion) of the five-membered diaza
ring: in this non-planar situation the NO2 and Me groups can stay,
* Corresponding authors. E-mail address: andrea.mazzanti@unibo.it (A. Mazzanti).
0040-4020/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2011.09.091

9130
J.E. Díaz et al. / Tetrahedron 67 (2011) 9129e9133
identity of these two values is obviously the result of an accidental
coincidence.
DFT calculations indicate the existence of four energy minima
(Scheme 1), corresponding to a situation where the two rings adopt
a twisted, thus chiral, conformation. The interconversion between
the two cis (b,d), as well as that between the two trans (a,c), occurs
via the so called p
-barrier,12 (i.e., the passage through an orthogonal
transition state), with a value, which is exceedingly low for the dy-
namic NMR technique, thus experimentally invisible. Accordingly
only a cis and a trans form, that have a sufficiently high barrier for
their reciprocal interconversion, can be experimentally detected.
According to DFT calculations the trans-a conformer, where the
dihedral angles for the ortho-nitrophenyl and for the ortho-tolyl are
ꢀ154ꢁ and ꢀ144ꢁ, respectively, represents the global minimum of
energy. The second most stable conformation corresponds to the
cis-b conformer that has dihedral angles of ꢀ160ꢁ and þ52ꢁ, re-
spectively, and a 0.7 kcal/mol higher energy value (Table 1, second
and third column).
Table 1
Dihedral angles and relative energies (kcal/mol) computed for the conformers of 1
Conformer
Dihedral anglesa
DFTb
DFT with
solventc
CCSDd
(trans-a)
(cis-b)
(trans-c)
(cis-d)
ꢀ154,ꢀ144
ꢀ160,þ52
ꢀ47,ꢀ62
0.0
0.7
2.7
1.4
0.0
0.45
1.9
1.5
2.1
0.6
0.0
ꢀ44,þ141
1.2
a
The first and the second value correspond, respectively, to the dihedral angle
Fig. 1. Left: temperature dependence of the methyl signal of 1 (600 MHz in CDF2Cl/
CDFCl2). Right: line shape simulation with the rate constants reported.
between the plane of 2-nitrophenyl and of the tolyl with that of the heterocyclic
ring.
b
B3LYP/6-311þþG(2d,p).
c
accordingly, either in a cis or in a trans relationship (Scheme 1) to
each other.
PCM-B3LYP/6-311þþG(2d,p).
d
CCSD/6-31þG(d)//B3LYP/6-311þþG(2d,p).
The trans to cis ratio theoretically predicted by this approach
(according to the Boltzmann distribution it corresponds to 92:8
at ꢀ130 ꢁC), does not match very well the almost equal proportion
experimentally observed. Such a discrepancy might be due to an
insufficient theoretical level of the calculations, or to the effects of
the solvent.
To verify the former hypothesis the four stationary points (two
trans and two cis) found by DFT were then recalculated as single
point energy at the more accurate CCSD/6-31þG(d) level. The re-
sults actually modify the energy trend, and the best structures are
now cis-d and trans-c, that are separated by 0.6 kcal/mol (fifth
column of Table 1). On the other hand, when the solvent was
considered in the DFT calculations (PCM model13) the energy dif-
ference is further reduced to 0.45 kcal/mol (fourth column of Table
1), with the trans-a, cis-b still the most stable forms.14 On the basis
the above results, the simpler DFT approach should be considered
sufficiently reliable to tackle the conformation analysis of the
present set of molecules because of its acceptable accuracy, coupled
with a reasonable computational cost.15
Scheme 1. DFT computed structures of the four ground state conformers of 1.
In principle, the barriers for the rotation of the two aryl groups
might be different, but such a difference is, apparently, too small to
be appreciated experimentally. In fact, if the rotation of only one of
the two rings had been locked at a given temperature (yielding
a non-coplanar, thus chiral conformation) with the other ring still
in rapid rotation, we should have observed diastereotopic methy-
lene hydrogens (thus yielding anisochronous signals), whereas the
methyl group should still display a single line. Only at a lower
temperature, when also the rotation of the second ring would be
frozen, the two methyl lines would separate, because only when
both rings are locked the trans and cis conformers are generated. In
contrast, the dynamic process was simultaneously detected by
monitoring the methylene as well as the methyl signals, thus in-
dicating that the rotation barriers of the two aromatic rings cannot
be distinguished, within the experimental accuracy. The near
The rotation barriers of the two rings (occurring through tran-
sition states where the aryl rings are almost coplanar with the
nearly flat five-membered ring16) were DFT computed to be very
close to each other (9.4 kcal/mol for the ortho-nitrophenyl and
8.9 kcal/mol for the ortho-tolyl), thus accounting for the impossi-
bility of distinguishing the two experimental barriers: these values
are also in reasonable agreement with observed value of 8.6 kcal/
mol. No additional dynamic features were observed when the
temperature was further lowered to ꢀ160 ꢁC.
The enantiomerization barrier due solely to the rotation of the
ortho-nitrophenyl ring can be observed in the case of compound 4,
where a phenyl has replaced the ortho-tolyl substituent: this is due
to the fact that the rotational barrier of the unsubstituted phenyl is
exceedingly low. As shown in Fig. S1 of Supplementary Data, the
high-field signal of the CH2 groups broadens and decoalesces below

J.E. Díaz et al. / Tetrahedron 67 (2011) 9129e9133
9131
ꢀ44 ꢁC, showing two anisochronous signals due to the diaster-
eotopicity of the hydrogens: the corresponding barrier was found to
be 10.8 kcal/mol. This value is higher than that measured in com-
pound 1 (8.6 kcal/mol), and indeed also the calculations predicted
a value of 11.1 kcal/mol, which is also higher than that computed for
the same rotational process in compound 1 (9.4 kcal/mol). The dif-
ference could be rationalized by considering that the ground state of
compound 4 is more stabilized than that of 1. In the ground state of 4
the phenyl ring can better conjugate with the five-membered un-
saturated ring (N3]C2eCipso¼Cortho dihedral angle 32ꢁ or ꢀ144ꢁ),
a conjugation, that is, lost in the transition state where the phenyl is
driven to be perpendicular by the rotation of the ortho-nitrophenyl
ring. The stabilizing effect is smaller in the case of the tolyl group of
compound 1 (dihedral angles 42ꢁ and ꢀ127ꢁ), thus the measured
rotation barrier of the ortho-nitrophenyl is lower.
The near degeneracy of the two aryl rotation barriers observed in
1 was lifted in the case of the corresponding six-membered de-
rivative 2. The 600 MHz spectrum in CDCl3 displays, in fact, sharp
signals for all the hydrogens at þ50 ꢁC, whereas on cooling below
ambient temperature, only the methylene multiplets begin to
broaden and two of them eventually split into a pair of signals with
equal intensity at ꢀ10 ꢁC. This is most evident for the signal of the
CH2 inposition 5 (at 2.19 ppm), which decoalesces below 25ꢁ toyield
two equally intense multiplets (ABX2Y2 spin system) separated by
70 Hz at ꢀ10 ꢁC (Fig. S2 of Supplementary Data): the corresponding
barrier was determined to be 14.5 kcal/mol. As mentioned above this
effect is a consequence of the restricted rotation of only one of the
two aryl rings, which adopts a non-planar (thus chiral)17 confor-
mation making, consequently, diastereotopic only the methylene
hydrogens.18 This rotation process, in contrast, cannot affect neither
the methyl nor any other signal of the molecule, as experimentally
observed. It is impossible toascertain solely on experimental ground
whether it is the rotation of the ortho-nitrophenyl or that of the
ortho-tolyl ring, which has been frozen. However the ortho-tolyl
group experiences a less hindered environment with respect to the
o-nitro phenyl, thus making the latter a more likelycandidate for the
observedprocess. DFTcalculations indeedpredict thatthebarrier for
the o-nitro phenylrotationshould be 13.4 kcal/mol, definitely higher
than that calculated for the ortho-tolyl (9.5 kcal/mol) and, in addi-
tion, much closer to the experimental value. Such a calculated dif-
ference explains why in compound 2 the two barriers can be
distinguished, contrary to the case of 1.
Fig. 2. 13C methyl signals (150.8 MHz in CDF2Cl/CDFCl2) of compound 2 at low tem-
perature showing that the major signal of the trans conformer is split into two lines
(indicated by triangles) and likewise is split into two the signal of the minor cis con-
former (indicated by squares).
Scheme 2. DFT computed structures of the four conformers of compound 2 due to the
restricted rotation of the two aryl rings (yielding cis and trans) and to the restricted
inversion of the six-membered ring (yielding syn and anti).
On furtherlowering the temperature(in CDF2Cl/CDFCl2) a second
dynamic process involving all the signals is observed: in particular
the single methyl line broadens and then splits at ꢀ80 ꢁC into two
lines, with a 70:30 ratio. This second process is due to the restriction
of the o-tolyl ring rotation, which now generates, as in 1, the trans
and cis conformers: the experimental barrier is 10.7 kcal/mol, in
reasonable agreement with the computed value (i.e., 9.5 kcal/mol).
The fact that two different aryl rotation barriers were observed in 2
proves that the unique barrier observed in 1 was actually due to an
accidental near coincidence of the barriers of the two aryl rotations.
When even lower temperatures are reached, a third dynamic
process becomes visible. This was best followed by monitoring the
methyl lines at the 13C frequency since they are clearly isolated from
the CH2 carbon lines. Both the lines of the major trans and of the
minor cis conformer broaden and then split into two at ꢀ130 ꢁC:
these four methyl lines are shown in Fig. 2. The barrier for this third
process was estimated to be about 7 kcal/mol. The analogous pro-
cess, observed in similar six-membered heterocyclic derivatives
containinga double bond, hadbeen assigned7,8,19 totherestrictionof
the ring inversionprocess that have barriers of about 7 kcal/mol. DFT
computations indicate that the two conformers (invertomers) aris-
ing from this motion (Scheme 2) actually correspond to theoretical
minima: they were labeled anti or syn, depending as to whether the
CH2 in position 5 is anti or syn to the NO2 group. As conceivable, the
experimental ratio of the two major is not equal to that of the two
minor signals (Fig. 2), because they reflect the anti to syn ratio oc-
curring in two different conformers (trans and cis).
The computed energy differences between the anti and syn
invertomers are however quite high (1.3 and 2.2 kcal/mol, in the
case of the conformer trans and cis, respectively) and do not agree
with the nearly equal proportion experimentally observed in Fig. 2.
Again this feature is likely to be a consequence of the insufficient
accuracy of the level of computation or to the solvent effects, as
mentioned above: being 2 an even larger molecule than 1, more
accurate computations were, unfortunately, outside the capability
of our computation facilities.13
In the case of the seven-membered derivative 3 the dynamic
effect is observed simultaneously on the methyl as well as on the
methylene lines, indicating, that the rotation of the ortho-nitro-
phenyl and ortho-tolyl rings take place almost simultaneously and
cannot be distinguished, as already observed in compound 1. The
ratio of the two rotational conformers, as detected by 13C at
ꢀ140 ꢁC, is about 70:30. Only two lines were visible, indicating that
only the two conformers trans and cis are present, in that the
conformational effects of ring inversion were not observed. This is
quite reasonable since the rate of the ring inversion process is

9132
J.E. Díaz et al. / Tetrahedron 67 (2011) 9129e9133
expected to be very fast in a seven-membered ring and to escape
NMR observation, even at quite low temperature.20 The complexity
of the spectrum of compound 3 also prevented the determination
of the experimental barrier for the cis/trans interconversion. Since
such a process occurs, however, in the same temperature range as
that of 1, it is conceivable to expect a similar value for the in-
terconversion barrier.
50.0, 44.7, 21.7, 19.4. HRMS (ESI) MHþ, found 296.1403. C17H18N3O2
requires 296.1399.
4.2.3. 1-(2-Nitrophenyl)-2-(2-methylphenyl)-1,4,5,6-tetrahydrodiaz-
epine 3. This compound was obtained as a yellow solid (278.1 mg,
90%), mp 98e99 ꢁC (chloroform/hexane). nmax (KBr) 2859, 2099,
1636, 1522, 1482, 1347, 1295, 1154, 1049, 852, 768, 707 cmꢀ1
.
1H
NMR (500 MHz, CDCl3, 25 ꢁC, TMS):
d
¼7.69 (1H, dd, J¼8.2, 1.6 Hz),
7.40e7.45 (1H, m), 7.33 (1H, dd, J¼8.2, 1.3 Hz), 7.15 (1H, dd, J¼7.7,
1.3 Hz), 7.02e7.10 (2H, m), 6.91e6.99 (2H, m), 4.00e4.05 (2H, m),
3.93e3.98 (2H, m), 2.37 (3H, s), 2.08e2.16 (2H, m), 2.02 (2H, p,
3. Conclusion
The restricted rotation of aryl rings linked to five-, six-, and
seven-membered heterocyclic rings were detected by Dynamic
NMR and computed by DFT calculations. In the case of the six-
membered tetrahydropyrimidine ring the additional conformers
resulting from a ring inversion process were also detected.
J¼6.1 Hz). 13C NMR (125 MHz, CDCl3, 25 ꢁC):
¼157.6, 144.9, 140.3,
d
136.7, 136.4, 133.2, 130.5, 129.7, 128.9, 126.9, 125.8, 125.4, 124.8,
52.5, 49.0, 26.1, 24.8, 19.8. HRMS (ESI) MHþ, found 310.1559.
C18H20N3O2 requires 310.1556.
N-(2-Methylbenzoyl)-N0-(2-nitrophenyl)-1,n-diamines
were
synthesized by selective monoacylation of the corresponding N-(2-
nitrophenyl)-1,n-diamines. N-benzoyl-N0-(2-nitrophenyl)-1,2-etha-
nediamine was described in the literature.23 Reactions were per-
formed ina 1 mmolscale. Yields andphysicaldata ofnewcompounds
are as follows.
4. Experimental
4.1. General
€
Melting points were determined with a Buchi capillary appa-
ratus and are uncorrected. IR spectra were recorded on a Per-
kineElmer Spectrum One FT-IR spectrometer. 1H NMR spectra were
recorded on a Bruker Avance II 500 MHz spectrometer, using
deuteriochloroform as the solvent. Chemical shifts are reported in
4.2.4. N-(2-Methylbenzoyl)-N0-(2-nitrophenyl)-1,2-ethanediamine.
This compound was obtained as a yellow solid (218.3 mg, 73%), mp
82e83 ꢁC (chloroform/hexane). nmax (KBr) 3435, 2103, 1641, 1573,
1511, 1418, 1353, 1320, 1262, 1231, 1157, 1037, 741 cmꢀ1
.
1H NMR
parts per million (d) relative to TMS as an internal standard. D2O
(500 MHz, CDCl3, 25 ꢁC, TMS):
d
¼8.23 (1H, br s ex), 8.18 (1H, dd,
was employed to confirm exchangeable protons (ex). Splitting
multiplicities are reported as singlet (s), broad signal (br s), doublet
(d), double doublet (dd), doublet of double doublets (ddd), triplet
(t), double triplet (dt), quartet (q), pentet (p), and multiplet (m).
HRMS (ESI) were performed with a Bruker MicroTOF-Q II spec-
trometer. Reagents, solvents, and starting materials were pur-
chased from standard sources and purified according to literature
procedures.
J¼8.5, 1.5 Hz), 7.48 (1H, ddd, J¼8.5, 7.0, 1.1 Hz), 7.29e7.37 (2H, m),
7.16e7.25 (2H, m), 7.03 (1H, dd, J¼8.5, 0.8 Hz), 6.70 (1H, ddd, J¼8.5,
7.1, 1.1 Hz), 6.29 (1H, br s ex), 3.74 (2H, q, J¼6.0 Hz), 3.64 (2H, q,
J¼6.0 Hz), 2.45 (3H, s). 13C NMR (125 MHz, CDCl3, 25 ꢁC):
¼170.7,
d
145.3, 136.5, 136.2, 135.7, 131.2, 131.1, 130.2, 127.0, 126.7, 125.8, 115.9,
113.7, 42.3, 39.1, 19.8. HRMS (ESI) MNaþ, found 322.1171.
C16H17N3NaO3 requires 322.1168.
4.2.5. N-(2-Methylbenzoyl)-N0-(2-nitrophenyl)-1,3-propanediamine.
This compound was obtained as a yellow solid (256.7 mg, 82%), mp
105e106 ꢁC (chloroform/hexane). nmax (KBr) 3435, 2102, 1641, 1573,
4.2. Synthesis of compounds 1e4
Amidines 1e4 were synthesized by microwave-assisted ring
closure of the corresponding N-acyl-N0-aryl-1,n-diamines.21,22 Re-
actions were performed in a 1 mmol scale. Compound 4 was al-
ready described in the literature.23 Yields and physical data of new
compounds are as follows.
1512, 1440, 1418, 1354, 1309, 1268, 1232, 1154, 1037, 741 cmꢀ1
.
1H
NMR (500 MHz, CDCl3, 25 ꢁC, TMS):
d
¼8.19 (1H, dd, J¼8.5, 1.5 Hz),
8.13 (1H, br s ex), 7.44e7.49 (1H, m), 7.30e7.37 (2H, m), 7.18e7.26
(2H, m), 6.89 (1H, dd, J¼8.70, 0.9 Hz), 6.68 (1H, ddd, J¼8.5, 7.0,
1.1 Hz), 6.04 (1H, br s ex), 3.61 (2H, q, J¼6.9 Hz), 3.47 (2H, td, J¼6.9,
5.6 Hz), 2.46 (3H, s), 2.07 (2H, p, 6.9). 13C NMR (125 MHz, CDCl3,
4.2.1. 1-(2-Nitrophenyl)-2-(2-methylphenyl)-imidazoline
1. This
25 ꢁC):
d
¼170.4, 145.3, 136.3, 136.2, 136.0, 132.1, 131.1, 130.0, 127.0,
compound was obtained as a yellow solid (222.0 mg, 79%), mp
126.58, 125.8, 115.5, 113.6, 40.7, 37.5, 29.4, 19.8. HRMS (ESI) MNaþ,
found 336.1330. C17H19N3NaO3 requires 336.1324.
116e118 ꢁC (chloroform/hexane). nmax (KBr) 3006, 2989, 1621, 1524,
1487, 1347, 1275, 1138, 1090, 851, 706 cmꢀ1 1H NMR (500 MHz,
.
CDCl3, 25 ꢁC, TMS):
d
¼7.75 (1H, dd, J¼8.1,1.4 Hz), 7.36e7.42 (1H, m),
4.2.6. N -(2-Methylbenzoyl)-N0-(2-nitrophenyl)-1,4-butanediamine.
This compound was obtained as a yellow solid (209.3 mg, 64%),
mp 94e95 ꢁC (chloroform/hexane). nmax (KBr) 3383, 2096, 1637,
1573, 1473, 1440, 1418, 1379, 1354, 1308, 1232, 1153, 1037,
7.19e7.26 (2H, m), 7.15e7.19 (1H, m), 7.12 (1H, dt, J¼7.6, 0.7 Hz),
7.03e7.09 (2H, m), 4.18e4.23 (2H, m), 3.95e4.01 (2H, m), 2.37 (3H,
s). 13C NMR (125 MHz, CDCl3, 25 ꢁC):
d
¼162.1, 144.7, 137.4, 136.6,
133.2, 130.8, 129.6, 129.6, 129.4, 127.8, 125.6, 125.4, 125.3, 54.6, 52.7,
19.9. HRMS (ESI) MHþ, found 282.1248. C16H16N3O2 requires
282.1242.
777 cmꢀ1
.
1H NMR (500 MHz, CDCl3, 25 ꢁC):
d
¼8.18 (1H, dd,
J¼8.7, 1.6 Hz), 8.08 (1H, br s ex), 7.42e7.48 (1H, m), 7.30e7.37
(2H, m), 7.17e7.25 (2H, m), 6.88 (1H, dd, J¼8.5, 0.8 Hz), 6.67 (1H,
ddd, J¼8.5, 7.0, 1.1 Hz), 5.94 (1H, br s ex), 3.52 (2H, q, J¼6.6 Hz),
3.36e3.44 (2H, m), 1.74e1.90 (4H, m), 2.45 (3H, s). 13C NMR
4.2.2. 1-(2-Nitrophenyl)-2-(2-methylphenyl)-1,4,5,6-tetrahydropyri-
midine 2. This compound was obtained as a yellow solid (268.4 mg,
91%), mp 115e116 ꢁC (chloroform/hexane). nmax (KBr) 2853, 2099,
(125 MHz, CDCl3, 25 ꢁC, TMS):
d
¼170.3, 145.4, 136.4, 136.3, 136.0,
131.9, 131.0, 129.88, 126.9, 126.6, 125.7, 115.3, 113.7, 42.5, 39.2,
27.4, 26.4, 19.8. HRMS (ESI) MNaþ, found 350.1483. C18H21N3NaO3
requires 350.1481.
1634, 1523, 1483, 1349, 1301, 1217, 1090, 852, 766, 705 cmꢀ1
.
1H
NMR (500 MHz, CDCl3, 25 ꢁC, TMS):
d
¼7.75 (1H, dd, J¼8.1, 1.5 Hz),
7.41e7.45 (1H, m), 7.25 (1H, d, J¼7.8 Hz), 7.17e7.21 (1H, m),
7.05e7.08 (1H, m), 7.03 (1H, dd, J¼7.6, 1.4 Hz), 6.96e6.99 (1H, m),
6.90e6.94 (1H, m), 3.79 (2H, br s), 3.74 (2H, t, J¼5.5 Hz), 2.35 (3H, s),
4.3. Variable temperature NMR spectroscopy
2.19 (2H, br s). 13C NMR (125 MHz, CDCl3, 25 ꢁC):
d
¼154.7, 139.8,
NMR spectra were recorded using a Varian INOVA spectrometer
135.7 (br s), 133.4, 130.3, 128.7, 128.5, 126.8 (br s), 125.4 (br s), 125.4,
operating at a field of 14.4 T (600 MHz for 1H and 150.8 for 13C). The

J.E. Díaz et al. / Tetrahedron 67 (2011) 9129e9133
9133
NMR tubes containing the compounds were linked to a vacuum
Supplementary data
line, immersed in liquid nitrogen and evacuated in order to con-
dense about 0.6 ml of a mixture of gaseous CDFCl2/CDF2Cl obtained
by reacting CDCl3 with SbF3 and SbF5 as described in the litera-
ture.24 The tubes were subsequently sealed under reduced pressure
(0.01 mbar) using a methane/oxygen torch. The samples were
cautiously warmed to þ25 ꢁC, where the solvent develops a pres-
sure of about 8 atm. After a few hours at ambient temperature, the
samples could be safely introduced into the probe head of the
spectrometer, already cooled to e30 ꢁC. Low temperature 600 MHz
1H spectra were acquired without spinning using a 5 mm dual di-
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.tet.2011.09.091.
References and notes
1. Dunbar, P. G.; Durant, G. J.; Fang, Z. J. Med. Chem. 1993, 36, 842e847.
2. Dunbar, P. G.; Durant, G. J.; Rho, T. J. Med. Chem. 1994, 37, 2774e2782.
3. Bernard, T.; Jebbar, M.; Rassouli, Y.; Himdi-Kabbab, S.; Hamelin, J.; Blanco, C. J.
Gen. Microbiol. 1993, 139, 129e136.
4. Weinhardt, K.; Wallach, M. B.; March, M. J. Med. Chem. 1985, 28, 694e698.
5. Carter, P.A.; Pfrendle, W.F. UK Patent Appl. GB 2,277,090, 1994; Chem. Abstr.
1994, 122, 31544b.
rect probe with a 9000 Hz sweep width, 2.0 m
s (20ꢁ tip angle) pulse
width, 3 s acquisition time, and 1 s delay time. A shifted sine bell
weighting function25 equal to the acquisition time (i.e., 3 s) was
applied before the Fourier transformation. Low temperature
150.8 MHz 13C spectra were acquired without spinning and under
6. McFarland, J. W.; Howes, H. L. J. Med. Chem. 1972, 15, 365e368.
7. Garcia, M. B.; Grilli, S.; Lunazzi, L.; Mazzanti, A.; Orelli, L. R. J. Org. Chem. 2001,
66, 6679e6684.
8. Garcia, M. B.; Grilli, S.; Lunazzi, L.; Mazzanti, A.; Orelli, L. R. Eur. J. Org. Chem.
2002, 4018e4023.
proton decoupling conditions with a 38,000 Hz sweep width, 4.2
m
s
9. The fact that the ratio of these two lines is different from 1:1 is proved by the
observation that the two lines have a 58:42 intensity in a Me2O solution at the
same temperature.
(60ꢁ tip angle) pulse width, 1 s acquisition time, and 1 s delay time.
A line broadening function of 1e2 Hz was applied before the
Fourier transformation. To reach the required low temperature
a flow of dry nitrogen was first passed through a pre-cooling unit
adjusted to e50 ꢁC. The gas entered into an inox steel heat-
exchanger immersed in liquid nitrogen and connected to the
NMR probe head by a vacuum-insulated transfer line. Gas flows of
10e40 L minꢀ1 were required to descend to the desired tempera-
ture. Temperature calibrations were performed before the experi-
ments, using a digital thermometer and a Cu/Ni thermocouple
placed in an NMR tube filled with isopentane. The conditions were
kept as equal as possible with all subsequent work. In particular, the
sample was not spun and the gas flow was the same as that used
during the acquisition of the spectra. The uncertainty in tempera-
ture measurements can be estimated as ꢂ2 ꢁC.
10. Eyring, H. Chem. Rev. 1935, 17, 65e77.
11. (a) Lunazzi, L.; Mancinelli, M.; Mazzanti, A. J. Org. Chem. 2007, 72, 5391e5394;
(b) Casarini, D.; Lunazzi, L.; Mazzanti, A. Eur. J. Org. Chem. 2010, 75, 2035e2056.
12. (a) Anderson, J. E.; Casarini, D.; Lunazzi, L. Tetrahedron Lett. 1988, 29,
3141e3144; (b) Lunazzi, L.; Mancinelli, M.; Mazzanti, A. J. Org. Chem. 2007, 72,
10045e10050.
13. For a review see: Tomasi, J.; Mennucci, B.; Cammi, R. Chem. Rev. 2005, 105,
2999e3093 (the calculations were performed by setting CHCl3 as the solvent,
because its parameters are the most similar to those of CDFCl2).
14. CCSD calculations including the solvent are behind our computing facilities.
15. A single point CCSD/6-31þG(d) calculation on compound 1 takes about one
week on an 8-cores job running on Xeon X7355 processors @ 2.93 GHz, with
a scratch file of about 120 Gbytes. CCSD calculations scale roughly with N6 (N
being the number of the basis functions, i.e., of the orbitals), so this approach
would require computational times close to a month when dealing with larger
molecules like compounds 2 and 3. On the other hand, an optimization with
DFT and the subsequent frequency calculation usually takes 8e16 h.
16. There are actually two possible transition states for the rotation of each aryl
ring, the threshold transition states correspond to the passage of the nitro
group across the CH2 in position 5, and to the passage of the ortho-methyl
group across the nitrogen in position 3. The alternative transition states have
much higher energies and therefore are not feasible.
17. As discussed in the case of 1 the frozen aryl ring adopts a conformation where
its plane is not coplanar with respect to the dynamic plane of the six-
membered ring, which undergoes a rapid ring inversion at this temperature.
18. (a) Mislow, K.; Raban, M. Top. Stereochem. 1967, 1, 1e38; (b) Jennings, W. B.
Chem. Rev. 1975, 75, 307e322; (c) Eliel, E. L. J. Chem. Educ. 1980, 57, 52e55; (d)
Casarini, D.; Lunazzi, L.; Macciantelli, D. J. Chem. Soc., Perkin Trans. 2 1992,
1363e1370.
19. (a) Oki, M. Applications of Dynamic NMR Spectroscopy to Organic Chemistry;
VCH: Deerfield Beach, Florida, 1985; 306; (b) Kleinpeter, E. In Advances in
Heterocyclic Chemistry; Katritzky, A. R., Ed.; Elsevier: 2004; Vol. 86, 109.
20. In principle it cannot be excluded that the ring inversion process had been
frozen but only one of the resulting conformers (invertomers) is populated.
This occurrence would also be compatible with the absence of observable NMR
dynamic effects.
21. García, M. B.; Torres, R. A.; Orelli, L. R. Tetrahedron Lett. 2006, 47, 4857e4859.
22. Díaz, J. E.; Bisceglia, J. A.; Mollo, M. C.; Orelli, L. R. Tetrahedron Lett. 2011, 52,
1895e1897.
Line shape simulations were performed using a PC version of the
QCPE DNMR6 program.26 Electronic superimposition of the original
spectrum and of the simulated one enabled the determination of
the most reliable rate constant.
4.4. Calculations
Geometry optimization were carried out at the B3LYP/6-31G(d)
level by means of the Gaussian 09 series of programs27 The
standard Berny algorithm in redundant internal coordinates and
default criteria of convergence were employed. The harmonic vi-
brational frequencies were calculated for all the stationary points.
For each optimized ground state the frequency analysis showed
the absence of imaginary frequencies, whereas each transition
state showed a single imaginary frequency. Visual inspection of
the corresponding normal mode was used to confirm that the
correct transition state had been found. If not explicitly indicated,
the reported energy values represent the total electronic energies.
In general, these give the best fit with experimental Dynamic-
NMR data.26b,28 This approach avoids artifacts that might result
from the ambiguous choice of the adequate reference tempera-
ture, and from the idealization of low-frequency vibrators as
harmonic oscillators (very important in the present cases, where
about one third of the calculated frequencies fall below the
500e600 cmꢀ1 range).
23. Perillo, I. A.; Lamdan, S. J. Heterocycl. Chem. 1970, 7, 791e798.
24. Siegel, J. S.; Anet, F. A. L. J. Org. Chem. 1988, 53, 2629e2630.
25. Claridge, T. D. W. High-resolution NMR Techniques in Organic Chemistry; Per-
gamon: Oxford, 1999; 71.
26. (a) Brown, J. H.; Bushweller, C. H. DNMR6: (program 633) QCPE Bulletin; Bloo-
mington: Indiana, 1983; Vol. 3; 103e103; (b) A copy of the program is available
on request from the authors (L. L. and A. M.).
27. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;
Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Na-
katsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng,
G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.;
Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery,
J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K.
N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.;
Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, N. J.; Klene, M.;
Knox, J. E.; 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.;
Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dan-
Acknowledgements
Financial contribution (to L.L and A.M.) was received from the
University of Bologna (RFO) and from MIUR, Rome (PRIN national
project ‘Stereoselection in Organic Synthesis, Methodologies, and
Applications’). Support (to L.R.O.) was also obtained from Uni-
versidad de Buenos Aires and CONICET
€
nenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O; Foresman, J. B.; Ortiz, J. V.;
Cioslowski, J.; Fox, D. J. Gaussian 09, Revision A.02; Gaussian: Wallingford CT,
2009.
28. Ayala, P. Y.; Schlegel, H. B. J. Chem. Phys. 1998, 108, 2314e2325.