Polymorphism vs. desmotropy: The cases of 3-phenyl- and 5-phenyl-1H-pyrazoles and 3-phenyl-1H-indazole

Article abstract of DOI:10.1002/1522-2675(200209)85:9<2763::AID-HLCA2763>3.0.CO;2-R

Two desmotropes, 3-phenyl-1H-pyrazole (1a) and 5-phenyl-1H-pyrazole (1b) have been isolated and the conditions for their interconversion established. The X-ray structure of 1b has been determined (a = 10.862(1), b = 5.7620(5), c = 12.927(2) A, β = 111.435

Full text of DOI:10.1002/1522-2675(200209)85:9<2763::AID-HLCA2763>3.0.CO;2-R

Helvetica Chimica Acta Vol. 85 (2002)
2763
Polymorphism vs. Desmotropy: The Cases of 3-Phenyl- and 5-Phenyl-
1H-pyrazoles and 3-Phenyl-1H-indazole
by MarÌa Angeles GarcÌaa), Concepcio¬n Lo¬pez*a), Rosa M. Claramunt*a), Ahmed Kenzb), Marcel Pierrot*b),
and Jose¬ Elgueroc)
a
¬
) Departamento de QuÌmica Organica y BiologÌa, Facultad de Ciencias, UNED, Senda del Rey, 9, E-28040
Madrid (fax: 00-34-913988372; e-mails: clopez@ccia.uned.es & rclaramunt@ccia.uned.es)
b
¬ √
¬
) Laboratoire de Bioinorganique Structurale, Centre Scientifique de Saint Jerome, Universite d×Aix-Marseille
¬
¬
III, Avenue Escadrille Normandie-Niemen, F-13397 Marseille Cedex 20 (fax: 00.33.491.923208; e-mail:
marcel.pierrot@lbs.u-3mrs.fr)
c
¬
¬
) Instituto de QuÌmica Medica, Centro de QuÌmica Organica −Manuel Lora Tamayo×, CSIC, Juan de la Cierva 3,
E-28006 Madrid
Two desmotropes, 3-phenyl-1H-pyrazole (1a) and 5-phenyl-1H-pyrazole (1b) have been isolated and the
conditions for their interconversion established. The X-ray structure of 1b has been determined (a ˆ 10.862(1),
b ˆ 5.7620(5), c ˆ 12.927(2) ä, b ˆ 111.435(2)8, space group P21/c), and both tautomers 1a and 1b were
characterized by NMRin the solid state ( 13C- and 15N-CPMAS). In the case of 3-phenyl-1H-indazole (2a), two
concomitant polymorphs have been analyzed by X-ray crystallography, and their NMRspectral properties were
determined. The low-melting-point polymorph, at 106.78, contains three molecules in the asymmetric unit (a ˆ
41.086(1), b ˆ 7.3860(2), c ˆ 23.391(1) ä, b ˆ 117.697(1)8, space group C2/c) and the high-melting-point one,
115.38, six molecules (a ˆ 13.7818(4), b ˆ 13.7976(5), c ˆ 18.9445(5) ä, a ˆ 94.300(3), b ˆ 95.131(3), g ˆ
119.428(3)8, space group P-1). Here, too, it has been experimentally determined how to transform one form
into the other. Density-functional-theory calculations at the B3LYP/6-31G** level have been performed in both
examples to rationalize the stability of the different tautomers.
1. Introduction.
Tautomeric compounds, as other compounds, can present
polymorphism (we refer here to true polymorphism, not to solvates, also called
pseudopolymorphism), but, in addition, they can present desmotropy. The term
polymorphism is a very well-known descriptor for tautomeric heterocycles [1 5]; the
term desmotropy refers, more precisely, to the observation of crystallization of a
compound in two different tautomers (there are no known examples of a compound
that crystallize in three or more tautomeric forms). This term must not be used for
crystals containing simultaneously two tautomers that, in general, but not always, are in
a 1:1 ratio. Desmotropy is not very rare [6 9], but, in many cases, the examples are
found under the headings −tautomerism× and −polymorphism× [10 16], even when
desmotropic crystals are not polymorphs, since tautomers are two different compounds
that isomerize rapidly in solution1). The studies of the transformation of a tautomer
into another tautomer in the solid state are examples of dynamic desmotropy [13][15].
1
)
Some authors use the term −tautomerizational polymorphism× [1] or −tautomeric polymorphism× [17]
instead of desmotropy. In our opinion, the distinction between polymorphism and isomerism cannot be
based on the activation barrier separating two forms in solution, which would lead to similar problems as
those concerning the terms conformation (alkanes) vs. configuration (alkenes). We advocate using the
term polymorphism in situations where both forms are identical in terms of connectivity matrices, thus
excluding tautomerism in situations where bonds are broken and created.

2764
Helvetica Chimica Acta Vol. 85 (2002)
It is evident that, to have a reasonable probability to crystallize two tautomers, their
intrinsic difference in energy should be rather low [18][19]. This is not generally the
case in functional tautomerism (e.g., oxo/hydroxy), but it is common for the annular
tautomerism of azoles [20]. We will report in the present paper the case of two azoles,
3(5)-phenyl-1H-pyrazole (1) and 3-phenylindazole (2).
We have devoted several papers to the tautomerism of 3(5)-phenyl-1H-pyrazole 1
[20 24]. In solution (NMR) and in the gas phase (ab initio calculations), tautomer 1a
is the most stable (ca. 2 kJ molÀ1). But, in the solid state (13C-CPMAS-NMR), the
chemical shifts correspond to the 5-phenyl tautomer 1b [23]. In all publications and
general books, this compound is reported to have a melting point of 788 [25 32]
without any indication of peculiar behavior, and, so, the conclusion would be that this
compound shows neither polymorphism nor desmotropy. However, in a review on the
application of mass spectrometry to the analysis of tautomeric organic compounds,
Terent×ev and Kalandarishvili [33] have stated on p. 346: −The transformation of 3-
phenylpyrazole into more stable 5-phenylpyrazole, which has a higher melting point, was
described×, with reference to [34]. However, careful reading of Kost and Grandberg×s
review [34] reveals nothing concerning compound 1.
The case of 3-phenylindazole is different. In an authoritative review [35], it is
stated (on p. 291): −The existence of two forms of 3-phenylindazole has been taken to
provide support for the equilibrium idea. On this basis, the two forms shown in the
equilibrium 2a/2b, in which the imino hydrogen atoms are on different nitrogen atoms,
would correspond to the two forms of 3-phenylindazole. The two forms of 3-
phenylindazole are interconvertible under various conditions. Similar interconversion of
the two forms has been observed with certain 5-substituted indazoles. If these examples
actually constitute pairs of isomers differing only in the location of the imino hydrogen,
the double bond locations should differ. However, refractometric measurements indicate
that both contain the benzene ring×. And then, on p. 323: −3-phenylindazole exists in two
forms. The lower-melting (107 1088C), if maintained slightly above its melting point,
resolidifies and then melts at 115 1168C. Both compounds give the same N-
acetylderivative, hydrolysis of which affords the low melting point. It has been suggested
by von Auwers that the two compounds are desmotropes and differ in the location of the
imino hydrogen (NÀH), and that both exist in the melt×. These conclusions are based on
Von Auwers× results [36 39]. When this compound has been reported in more recent

Helvetica Chimica Acta Vol. 85 (2002)
2765
times, either the melting point is not reported [40][41], or a melting point of 106 1078
is given [42][43].
Nevertheless, the intrinsic difference in stability between 1H-indazole (3a) and 2H-
indazole (3b) is too high (17 kJ molÀ1 [44][45]) to easily accept that one of the
compounds with a melting point of 1088 or 1168 could be 2b. In the solid state, indazole
crystallizes as 3a [46], and other indazoles like 5-phenylindazole [47] and 3-phenyl-5-
methylindazole also crystallize as 1H-tautomers, 4a and 5a [48], respectively.
2. Results and Discussion. 2.1. The Case of 3(5)-Phenyl-1H-pyrazole 1. When this
pyrazole was crystallized from petroleum ether, it was observed that, depending on the
evaporation conditions and the solute/solvent ratio, a white form, or a pale yellow and
more crystalline form of 1 was obtained. An NMRstudy in the solid state (see later) of
both samples shows that the white variety was a mixture of 1a and 1b, while the yellow
one contained exclusively 1a. Therefore, the white solid is an example of concomitant
desmotropy. To obtain exclusively one tautomer, we carried out a series of experiments,
the more significant being: i) dissolving the crude compound in petroleum ether and
slowly evaporating the solvent at room temperature afforded tautomer 1a (pale
yellow); ii) dissolving tautomer 1a in EtOH and quickly removing the solvent in vacuo
or recrystallizing 1a in CH2Cl2 at 48 (freezer) afforded tautomer 1b; iii) Solid 1a on
standing (several days) or by mechanical grinding was transformed into 1b; iv) Melting
transformed 1b into 1a.
1
NMR Spectroscopy in Solution. We have recorded the H-, 13C-, and 15N-NMR
spectra of compound 1 in (D18)HMPA, [(CD3)2N]3PO, a solvent known to slow the
annular prototropic equilibria [18]. The spectra correspond to mixtures of 65% of
tautomer 1a and 35% of tautomer 1b. The chemical shifts were summarized in Table 1,
and the different signals were assigned through standard gs-HMQC and gs-HMBC
experiments and by analogy with published results [22]. Note that the NH proton of
tautomer 1b is deshielded relative to that of 1a by the nearby Ph ring; these signals were
used to determine the ratio 1a/1b.
13C- and 15N-CPMAS-NMR Spectroscopy in the Solid State. We have already noted
that 15N chemical shifts are more sensitive to environmental effects, such as H-bonds,
than to tautomerism [49]. On the other hand, 13C chemical shifts are very similar to
those in solution and constitute a precise tool to identify and determine the purity (up
to 10%) of these tautomers. The tautomer obtained by crystallization in CH2Cl2 at 48
(freezer) used to determine the X-ray structure corresponds to tautomer 1b. In Fig. 1,

2766
Helvetica Chimica Acta Vol. 85 (2002)
Table 1. NMR Chemical Shifts of 3(5)-Phenylpyrazole 1 in (D18)HMPA at 300 K and in the Solid State
Nucleus
1
2
3
4
5
Ph Group
3-Phenyl-1H-pyrazole (1a) 65%
1Ha)
13Ca)
15Na)
13Cb)
15Nb)
12.85
150.44 101.54 135.35 135.35 (ipso), 125.63 (ortho), 128.73 (meta), 127.14 (para)
150.4 101.8 133.8 133.9 (ipso), 127.5 (ortho), 127.5 (meta), 127.5 (para)
À 171.9 À 86.7
À 173.9 À 84.4
5-Phenyl-1H-pyrazole (1b) 35%
1Ha)
13Ca)
15Na)
13Cb)
15Nb)
13.28
139.69 101.54 142.30 130.89 (ipso), 125.63 (ortho), 128.73 (meta), 129.33 (para)
143.0 101.6 142.9 129.7 (ipso), 125.8 (ortho), 129.4 (meta), 129.4 (para)
À 176.6 À 76.3
À 176.8 À 93.0
a) HMPA. b) CPMAS.
Fig. 1. Quaternary C-atoms obtained by means of the NQS sequence in 13C-NMR CPMAS
for desmotropes 1a and 1b
we present the quaternary C-atoms C(3) and Cipso for 1a, and C(5) and C(ipso) for 1b,
in the solid state (see Exper. Part).
X-Ray-Crystal and Molecular Structure of 5-Phenyl-1H-pyrazole (1b). The crystals
of 1a obtained by slow evaporation in petroleum ether were not suitable for crystal-
structure determination. Moreover, on standing, crystals of 1a evolved to 1b. On the

Helvetica Chimica Acta Vol. 85 (2002)
2767
Fig. 2. Molecular structure of 5-phenyl-1H-pyrazole (1b)
Fig. 3. Molecular packing for 5-phenyl-1H-pyrazole (1b)
other hand, those of 1b proved suitable. Their main characteristics are reported in the
Exper. Part and are depicted in Figs. 2 and 3.
The compound crystallizes as a chain of NÀH ¥¥¥ N H-bonded molecules (catemer),
this being one of the typical motifs of NH-pyrazole crystals [50]. Phenylpyrazoles,

2768
Helvetica Chimica Acta Vol. 85 (2002)
unless the two rings are in the same plane (q ˆ 08) or are orthogonal (q ˆ 908), present
axial chirality, the rings adopting (M) (Minus) or (P) (Plus) conformations. These
conformations, being separated by a very low barrier, are in rapid equilibrium in
solution. The helices formed by these structures could crystallize in three structures: i)
the helix contains both conformations in alternation (M,P)n; ii) in the same crystal,
there are two adjacent helices, one (M)n and the other (P)n; iii) a spontaneous
resolution has occurred (conglomerate), and in the batch are found pure (M)n and pure
(P)n crystals. Compound 1b (q ˆ 18.43(1)8) belongs to the P21/c space group with a ˆ
10.862(1), b ˆ 5.7620(5), c ˆ 12.927(2) ä and b ˆ 111.435(2)8. The helix of this racemic
crystal corresponds to class ii; to describe the catemer, the connectivity of pyrazole
centroids would be used. Compound 1b forms a chain of order 2 with the properties,
reported in Fig. 4, very similar to those found for another catemer, the 4-phenyl-5-
azidopyrazole [50].
Fig. 4. Geometries of catemers of order 2 as defined by the centroids
DSC Experiments. When we recorded the differential scanning calorimetry (DSC)
plots of compounds 1a and 1b at a rate of 108 minÀ1, tautomer 1b melted at 75.18 and
resolidified to yield tautomer 1a with a melting point at 60.58.
DFT Calculations (B3LYP/6-31G**). The optimized geometries correspond to the
following values: 1a (E ˆ À457.27423 hartrees, m ˆ 2.42 D, q ˆ 18), 1b (E ˆ À457.27389
hartrees, m ˆ 2.55 D, q ˆ 288), 1a being most stable by 0.91 kJ molÀ1. By applying the
zero-point energy (ZPE) correction, the stability difference increases to 1.26 kJ molÀ1,
that is, at 300 K, KT ˆ 1.66, 62% 1a/38% 1b. Since the dipole moments are very similar,
no important solvent effects are expected in the composition mixture. We have now all
the information to summarize in Scheme 1 the behavior of 3(5)-phenylpyrazole 1.
Note that, in the gas phase (DFT calculations) and in solution (NMR), tautomer 1a
is the more stable, while, in the solid state, 1a corresponds to a metastable structure.
2.2. The Case of 3-Phenylindazole (2). This compound, as will be shown in the
following discussion, is an example of polymorphism involving tautomer 2a. It
crystallized in two forms: the low-melting point (LMP, needles, 106.78 by DSC at a rate
of 108 minÀ1) and the high-melting-point (HMP, cube-like shape, 115.38 by DSC at a
rate of 108 minÀ1) form. The following experiments have been carried out: i)
crystallization in petroleum ether (b.p. 35 608) yielded both forms; ii) crystallization

Helvetica Chimica Acta Vol. 85 (2002)
2769
Scheme 1. The Case of 3(5)-Phenylpyrazole 1 a) on Standing at Room Temperature or by Mechanical Grinding,
b) by Melting
in benzene/cyclohexane yielded the high-melting-point form; iii) careful melting of the
LMP form afforded the HMP polymorph.
NMR Spectroscopy in Solution. To exclude desmotropy, it is necessary to determine
the tautomeric structures of both forms by solid-state NMR. To assign the signals of
both possible tautomers of this compound, 2a (1H) and 2b (2H), we have prepared the
corresponding N-Me derivatives 6 and 7. The results are reported in Table 2.
A simple examination of the 13C chemical shifts of Table 2 shows that only tautomer
1
2a is present in solution (we have carried out other NMRexperiments, both H and 13C,
and never observed signals of the other tautomer, nor even broadening, for instance in
(D18)HMPA at 255 K). The differences in 15N chemical shifts are due to the effect of
the Me group and to modifications of the H-bonds involving the N-atoms.
13C- and 15N-CPMAS-NMR Spectroscopy in the Solid State. Both polymorphs, LMP
and HMP, show very similar 13C-CPMAS spectra, which are also very close to that of 2a
in solution. Therefore, we are in the presence of a case of polymorphism and not of
desmotropy. Since, in the same crystallization dish, both are found, this is a new
example of concomitant polymorphs.
X-Ray-Crystal and Molecular Structures of 3-Phenylindazole Polymorphs (2a
(LMP) and 2a (HMP)). 3-Phenyl-1H-indazole crystallizes in two different habits: long
needles for 2a (LMP) and cube-shaped crystals for 2a (HMP). The two crystal types
belong to different crystal systems: monoclinic (2a (LMP): a ˆ 41.086(1), b ˆ 7.3860(2),
c ˆ 23.391(1) ä, b ˆ 117.697(1)8, space group C2/c) and triclinic (2a (HMP): a ˆ

2770
Helvetica Chimica Acta Vol. 85 (2002)
Table 2. NMR Chemical Shifts of 1-Methyl-3-phenylindazole (6), 2-Methyl-3-phenylindazole (7) and 3-
Phenylindazole (2) in (D6)DMSO at 300 K, 2 in (D8)THF at 205 K, and 2 in the Solid State
Nucleus
1
2
3
3a
4
5
6
7
7a
Ph Group
1-Methyl-3-phenyl-1H-indazole (6) (N-Me at 35.42 ppm)
13Ca)
133.36 120.61 120.75 121.13 126.12 110.07 141.17 141.96 (ipso), 128.86 (ortho),
126.67 (meta), 127.66 (para)
15Na)
À 204.6 À 63.1
2-Methyl-3-phenyl-2H-indazole (7) (N-Me at 38.62 ppm)
13Ca)
134.99 120.51 119.87 121.57 125.74 116.78 147.14 143.17 (ipso), 129.35 (ortho),
129.13 (meta), 128.64 (para)
15Na)
À 92.3 À 166.2
3-Phenylindazole (2)
13Ca)
133.77 120.03 120.99 120.57 126.06 110.55 141.54 143.15 (ipso), 128.84 (ortho),
126.72 (meta), 127.61 (para)
15Na)
13Cb)
À 197.0 À 86.8
135.31 121.24 121.66 121.61 126.69 111.14 142.81 144.34 (ipso), 129.51 (ortho),
127.65 (meta), 128.27 (para)
133.2 121.1 121.1 121.1 126.2 110.9 141.2 145.3 (ipso), 128.9 (ortho),
126.2 (meta), 128.9 (para)
13Cc)
15Nc)
13Cd)
À 197.0 À 86.8
134.0 119.8 119.8 119.8 127.2 109.2 141.4 145.0 (ipso), 127.2 (ortho),
127.2 (meta), 127.2 (para)
15Nd)
a) DMSO. b) THF. c) CPMAS of LMP polymorph. d) CPMAS of HMP polymorph.
13.7818(4), b ˆ 13.7976(5), c ˆ 18.9445(3) ä, a ˆ 94.300(3), b ˆ 95.131(3), g ˆ
119.428(3)8, space group P-1). The asymmetric unit cell is composed of three 3-
phenyl-1H-indazole molecules for 2a (LMP) and of six molecules for 2a (HMP). The
structure of the asymmetric unit is shown in Figs. 5 (2a (LMP)) and 7 (2a (HMP)). In
both structures, 3-phenyl-1H-indazole molecules are linked by H-bonds to form
trimers, as can be observed in Figs. 5 and 6 for 2a (LMP) and in Fig. 7 for 2a (HMP).
DSC Experiments. When the DSC of the LMP form was recorded at 108 minÀ1, it
melted at 106.78. When heating is maintained above the melting point for 20 min and
then the temperature first decreased (À 508) and then increased, the compound melted
at 115.38 (HMP). When the temperature is increased to 1708, the compound reaches a
glassy state needing several days to resolidify, and its DSC record shows both
polymorphs. When the DSC were recorded at 18 minÀ1 an exothermic transition of
LMP towards the HMP was observed.
DFT Calculations (B3LYP/6-31G**). Both tautomers of 3-phenylindazole have
been fully optimized (no imaginary frequencies). They have the following intrinsic
properties: 2a: E ˆ À610.92081 hartrees, m ˆ 1.73 D, q ˆ 29.58, and 2b: E ˆ À610.91378
hartrees, m ˆ 2.70 D, q ˆ 33.78. As expected, the 1H-tautomer 2a is calculated to be by
18.44 kJ molÀ1 (19.25 kJ molÀ1 with the ZPE correction) more stable than 2b, and the
increase in the dipole moment should not modify the equilibrium in solution. Since 1a
and 1b are of similar stability, the observation that 2b is significantly less stable than 2a
is not due to the Ph substituent but to 1H-indazoles being more aromatic than 2H-

Helvetica Chimica Acta Vol. 85 (2002)
2771
Fig. 5. Molecular structure of 3-phenyl-1H-indazole polymorph 2a (LMP)
Fig. 6. Molecular packing for 3-phenyl-1H-indazole polymorph 2a (LMP)

2772
Helvetica Chimica Acta Vol. 85 (2002)
Fig. 7. Molecular structure of 3-phenyl-1H-indazole polymorph 2a (HMP). Trimer 1: N(1)ÀN(2)/
(N(21)ÀN(22)/N(41)ÀN(42) and Trimer 2: N(61)ÀN(62)/N(101) À N(102)/N(81)ÀN(82)
indazoles. This is generally the case for all benzazoles: the benzenoid tautomer is more
stable than the quinonoid one in the absence of other factors, such as lone pair/lone pair
repulsions [44][45]. At the B3LYP/6-31G** level, the three C(H)ÀC(H) bonds show
less bond alternation in 2a (1.387, 1.415, and 1.386 ä) than in 2b (1.377, 1.428, and
1.376 ä). Scheme 2 summarizes the information we have gathered on compound 2.
Scheme 2. The Case of 3-Phenylindazole (2)

Helvetica Chimica Acta Vol. 85 (2002)
2773
Conclusions.
In two closely related compounds, we have found a case of
desmotropy and a case of concomitant polymorphism, thus clarifying a long-standing
problem. In the case of 3-phenylindazole, the difference in stability between the two
tautomers is too great for the phenomenon of desmotropy to appear. On the other
hand, 3(5)-phenyl-1H-pyrazole has precisely the thermodynamic properties to allow
desmotropy, the two tautomers being of similar energy. This condition is not sufficient,
however; for instance, a closely related compound, 3(5)-phenyl-5(3)-methylpyrazole
crystallized, forming tetramers in which both tautomers are present [50]. Finally, we
feel that the present research work is a posthumous homage to the memory of the great
German chemist, Karl von Auwers, who was the most prominent figure in pyrazole and
indazole chemistry in the last century.
Financial support was provided by the Spanish DGICYT (Project No. BQU-2000-0252). One of us
(M. A. G.) thanks the UNED for an FPI grant.
Experimental Part
Preparation of 3(5)-phenyl-1H-pyrazole 1 [51]. Acetophenone (2.00 g, 0.0166 mol) and HCOOEt (1.85 g,
0.0249 mol) were added, in one portion, with rapid stirring, to dry MeONa (0.93 g, 0.0166 mol in 15 ml of
toluene) in a 50-ml round-bottomed flask provided with a magnetic stirrer and a reflux condenser topped with a
drying tube. The mixture was stirred, after 1 h a solid appeared, the solid was filtered and washed with hot
toluene and then with hexane. Finally, the product was air-dried and then made into a slurry with 10 ml of
MeOH. To this MeOH slurry, hydrazine monohydrochloride (1.14 g, 0.0166 mol) and 10 ml of H2O were added.
After 30 min of stirring, MeOH was removed under reduced pressure. The remaining aq. soln. was extracted
three times with CH2Cl2 (3 Â 15 ml). The soln. was dried (Na2SO4), and CH2Cl2 was removed in vacuo. The
product was crystallized from petroleum ether, and 1.11 g of solid was obtained (yield 50%, isolated product).
Preparation of 3-Phenyl-1H(2H)-indazole 2 [52]. In a 50-ml, double-necked, round-bottomed flask
provided with a magnetic stirrer and a reflux condenser was placed o-aminobenzophenone (2.00 g, 0.0101 mol),
and HCl (1.70 ml, 0.0203 mol) was added carefully. The soln. was cooled in ice, and the temp. was maintained at
0 to À58, and a soln. of NaNO2 (0.813 g, 0.0118 mol) in 7 ml of H2O was added dropwise. The diazotized
compound was neutralized with a soln. of Na2SO3. The mixture was extracted with AcOEt, dried, and
evaporated to yield the 2-oxo-3-phenylindazole. This crude product was placed in a 250-ml round-bottomed
flask provided with a reflux condenser and containing 50 ml of EtOH and was boiled gently until most
compound had dissolved.
Meanwhile 10 g of a good grade of SnCl2 in 30 ml of conc. HCl was dissolved. This soln. was added to the
contents of the flask and boiled under reflux for a further 30 min; after cooling, EtOH was removed and then
extracted with AcOEt (3 Â 15 ml). The combined org. layers were washed with H2O and dried (Na2SO4).
Evaporation of the solvent gave an oily material, which was purified by column chromatography (silica gel;
hexane/AcOE) 7:3) (Rf 0.24): yellow solid (0.808 g, 41%).
Methylation of 2 [53]. In a 50-ml round-bottomed flask provided with a magnetic stirrer and a reflux
condenser, 3-phenylindazole (0.179 g, 0.000923 mol), KOH (0.175 g, 0.00312 mol), MeI (0.525 g, 0.00370 mol),
and MeOH (5 ml) were boiled for 4 h, cooled, and diluted with H2O (20 ml). The mixture was extracted with
CHCl3 (3 Â 10 ml) and dried (NaSO4). A portion of the CHCl3 soln. was evaporated and directly investigated by
1H-NMRspectroscopy (a ratio of 70 :30 1-methyl/2-methyl was found, which agrees with previous results, 74 :26
[53]). Evaporation of the bulk of the CHCl3 extract gave the crude mixture of N-methyl-3-phenylindazoles. This
was separated by chromatography (silica gel; CHCl3); the 1-Me isomer 6 (m.p. 798; [53]: 808) was eluted first,
followed by the 2-Me isomer 7 (m.p. 488; [53]: oil).
NMR Experiments. 13C- (100.62 MHz) and 15N-NMR(40.56 MHz) spectra in soln. were obtained with a
Bruker DRX-400 instrument (corresponding to 9.4 T). Variable-temp. 1H-, 13C-, and 15N-NMRexperiments
were conducted on the same spectrometer with (D8)THF and (D18)HMPA as solvents. 2D-Inverse-proton-
detected heteronuclear-shift-correlation spectra, gs-HMBC, gs-HMQC, were obtained with the standard pulse
sequence [54]. Solid-state 13C- (50.32 MHz) and 15N- (20.28 MHz) CPMAS-NMRspectra have been obtained
with a Bruker AC-200 spectrometer at 298 K and a 7-mm Bruker DAB-7 probehead, which achieves rotational

2774
Helvetica Chimica Acta Vol. 85 (2002)
frequencies of ca. 3.5 4.5 kHz. Samples were carefully packed in ZrO2 rotors, and the standard CPMAS pulse
sequence was employed. To observe only the quaternary C-atoms, we run the NQS (Non-Quaternary
Suppression) experiments by conventional cross-polarization (CP) at different contact times and with the
dipolar dephased technique [55]. Chemical shifts (d) in ppm are relative to Me4Si and 15NH4Cl (these were
converted to MeNO2 according to the relationship: d (15N (MeNO2)) ˆ d (15N(NH4Cl)) À338.1 ppm).
Crystallographic Data Collection and Structure Determination of 1b, 2a (LMP), and 2a (HMP). The results
are reported in Table 3. Intensities were measured in the f-scan mode on a Nonius CCD diffractometer
(graphite monochromator, MoKa, l ˆ 0.71073 ä), and the Shelx 97 program for refinement of crystal structures
was used [56].
Crystallographic data (excluding structure factors) for the structures reported in this paper have been
deposited with the Cambridge Crystallographic Data Centre as deposition No. CCDC-184315, 184316, and
184317 for 1b, 2a (LMP), and 2a (HMP), respectively. Copies of the data can be obtained, free of charge, on
application to the CCDC, 12 Union Road, Cambridge CB21EZ, UK (fax: ‡ 44(1223)336033; e-mail:
deposit@ccdc.cam.ac.uk).
Table 3. Crystallographic Data Collection and Structure Determination of 1b, 2a (LMP) and 2a (HMP)
1b
2a (LMP)
2a (HMP)
Crystallized from
Empirical formula
Formula weight
CH2Cl2 at 48
C9H8N2
144.17
Petroleum ether
C13H10N2
194.23
Petroleum ether
C13H10N2
194.23
Crystal color, habit
Crystal dimensions [mm]
Temp. [K]
Colorless, cube
0.35 Â 0.30 Â 0.30
293
Pale yellow, needles
0.5 Â 0.3 Â 0.1
293
yellow, prismatic
0.35 Â 0.30 Â 0.20
178
Crystal system
Space group
Monoclinic
P21/c
Monoclinic
C2/c
Triclinic
P-1
Reflections for cell determination
5627
21322
43711
2q Range [8]
a [ä]
b [ä]
c [ä]
a [8]
4
52
4
52
4 52
10.862(1)
5.7620(5)
12.927(2)
90.000
111.435(2)
90.000
753.1(1)
4
41.086(1)
7.3860(2)
23.391(1)
90.000
117.697(1)
90.000
6284.9(4)
24
1.232
13.7818(4)
13.7976(5)
18.9445(3)
94.300(3)
95.131(3)
119.428(3)
3095.5(2)
12
b [8]
g [8]
V [ä3]
Z
Dx [g cmÀ3
]
1.272
0.078
1.259
0.076
Absorption coefficient (MoKa) [cmÀ1
]
0.074
2qmax [8]
52.6
52.7
50.8
Total reflections measured
5627
21322
43711
Unique reflections
1350
5994
11588
Absorption correction
none
none
none
Rint
0.036
0.047
0.051
Refined parameters
100
406
811
Reflections observed (I > 2s(I))
1266
4840
9735
H-Atoms
R
constrained
0.0622
0.1380
k1 ˆ 0.0508;
k2 ˆ 0.2248
0.005
constrained
0.0470
0.0998
k1 ˆ 0.0333;
k2 ˆ 3.2696
0.001
constrained
0.0801
0.1127
k1 ˆ 0.1089;
k2 ˆ 3.3126
0.000
Rw
Weightsa)
Final (shift/e.s.d.)max
Goodness-of-fit
D1fin (max/min) [eäÀ3
1.164
0.121/ À 0.131
1.090
0.13/ À 0.15
1.097
0.700/ À 0.387
]
a) w ˆ [s2Fo2 ‡ k12 P2 ‡ k2 P]À1, with P ˆ (Fo2 ‡ 2 Fc2)/3

Helvetica Chimica Acta Vol. 85 (2002)
2775
DSC Experiments. The melting temp. of the samples were determined with a Seiko DSC 220C instrument.
The heating rates were normally 108 minÀ1, but other heating and cooling rates were used to study the
transitions, with N2 as the purge gas. As known, slight differences in the melting temp. appear depending on the
heating rates. Melting points were also measured on a Microscope Axiolab −Zeiss× with a TMS 92 LINKAN
heating stage.
Computational Details. Calculations were carried out at the B3LYP/6-31G** level with basis sets of
Gaussian-type functions with Windows Titan 1.0.5 from Wavefunction Inc.
REFERENCES
[1] S. R. Byrn, −Solid-State Chemistry of Drugs×, Academic Press, New York, 1982.
[2] J. Bernstein, −Conformational Polymorphism× in −Organic Solid State Chemistry×, Ed. G. R. Desiraju,
Elsevier, Amsterdam, 1987, Chapt. 13.
[3] J. D. Dunitz, J. Bernstein, Acc. Chem. Res. 1995, 28, 193.
[4] J. Bernstein, R. J. Davey, J.-O. Henck, Angew. Chem., Int. Ed. 1999, 38, 3440.
[5] −Polymorphism in Pharmaceutical Solids×, Ed. H. G. Brittain, Marcel Dekker, New York, 1999.
¬
[6] C. Foces-Foces, A. L. Llamas-Saiz, R. M. Claramunt, C. Lopez, J. Elguero, Chem. Commun. 1994, 1143.
¬
[7] J. S. Casas, E. E. Castellano, A. Macias, N. Pleya, A. Sanchez, J. Sordo, J. M. Varela, J. Zukerman-
Schpector, Inorg. Chim. Acta 1995, 238, 129.
[8] P. Boldt, S. Zippel, M. Ratzkowsky, J. Chem. Soc., Perkin Trans. 1 1996, 2615.
[9] M. Bauer, R. K. Harris, R. C. Rao, D. C. Apperley, C. A. Rodger, J. Chem. Soc., Perkin Trans. 2 1998, 475.
[10] P. G. Takla, C. J. Dakas, J. Pharm. Pharmacol. 1989, 41, 227.
[11] M. Annese, A. B. Corradi, L. Forlani, S. C. Rizzoli, P. Sgarabotto, J. Chem. Soc., Perkin Trans. 2 1994, 615.
[12] J. Elguero, N. Jagerovic, C. Foces-Foces, F. H. Cano, M. V. Roux, F. Aguilar-Parrilla, H.-H. Limbach, J.
Heterocycl. Chem. 1995, 32, 451.
[13] Y. Eichen, M. Botoshansky, U. Peskin, M. Scherl, D. Haarer, S. Khatib, J. Am. Chem. Soc. 1997, 119, 7167.
[14] I. Alkorta, I. Rozas, J. Elguero, J. Chem. Soc., Perkin Trans. 2 1998, 2671.
[15] S. Khatib, S. Tal, O. Godsi, U. Peskin, Y. Eichen, Tetrahedron 2000, 56, 6753.
[16] A. Schmidt, S. Kababya, M. Appel, S. Khatib, M. Botoshansky, Y. Eichen, J. Am. Chem. Soc. 1999, 121,
11291.
[17] T. L. Threlfall, Analyst 1995, 120, 2435.
[18] J. Elguero, A. R. Katritzky, C. Marzin, P. Linda, −The Tautomerism of Heterocycles×, Academic Press, New
York, 1976.
[19] J. Elguero, A. R. Katritzky, O. V. Denisko, Adv. Heterocycl. Chem. 2000, 76, 1.
¬
ƒ
[20] J.-L. M. Abboud, P. Cabildo, T. Canada, J. Catalan, R. M. Claramunt, J. L. G. De Paz, J. Elguero, H.
Homan, R. Notario, C. Toiron, G. I. Yranzo, J. Org. Chem. 1992, 57, 3938.
[21] C. Cativiela, J. I. GarcÌa, J. Elguero, E. Elguero, Gazz. Chim. Ital. 1991, 121, 477.
[22] F. Aguilar-Parrilla, C. Cativiela, M. D. DÌaz de Villegas, J. Elguero, C. Foces-Foces, J. I. GarcÌa Laureiro,
F. H. Cano, H.-H. Limbach, J. A. S. Smith, C. Toiron, J. Chem. Soc., Perkin Trans. 2 1992, 1737.
¬
[23] C. Lopez, R. M. Claramunt, S. Trofimenko, J. Elguero, Can. J. Chem. 1993, 71, 678.
[24] C. Cativiela, J. I. GarcÌa, J.-P. Fayet, J. Elguero, Bull. Soc. Chim. Belg. 1995, 104, 383.
[25] L. Knorr, Liebigs Ann. Chem. 1894, 279, 254.
[26] K. von Auwers, Liebigs Ann. Chem. 1934, 508, 51.
[27] H. Brederek, Chem. Ber. 1963, 96, 1505.
[28] M. S. Gibson, A. W. Murray, J. Chem. Soc. 1964, 6250.
[29] J. C. Parham, J. Org. Chem. 1969, 34, 1474.
[30] R. Fusco, −Pyrazoles×, in −Pyrazoles, Pyrazolines, Pyrazolidines, Indazoles and Condensed Rings×, Eds. L. C.
Behr, R. Fusco, C. H. Jarboe, Interscience, John Wiley & Sons, 1967, p. 403.
[31] K. Schofield, M. R. Grimmett, B. R. T. Keene, −The Azoles×, Cambridge University Press, Cambridge, 1976,
p. 257.
[32] K. Kirsche, in −Methoden der Organischen Chemie× (Houben-Weyl), 1994, − Heterarene III×, pp. 415, 419.
[33] P. B. Terent×ev, A. G. Kalandarishvili, Mass Spectrom. Rev. 1996, 15, 339.
[34] A. N. Kost, I. K. Grandberg, Adv. Heterocycl. Chem. 1966, 6, 347.

2776
Helvetica Chimica Acta Vol. 85 (2002)
[35] L. C. Behr, −Indazoles and Condensed Types×, in −Pyrazoles, Pyrazolines, Pyrazolidines, Indazoles and
Condensed Rings×, Eds. L. C. Behr, R. Fusco, C. H. Jarboe, Interscience, John Wiley & Sons, 1967, p. 291,
323.
[36] K. von Auwers, Liebigs Ann. Chem. 1937, 527, 291.
[37] K. von Auwers, Chem. Ber. 1896, 29, 1255.
[38] K. von Auwers, Chem. Ber. 1922, 55B, 1112.
[39] K. von Auwers, Chem. Ber. 1929, 62B, 1671.
[40] J. I. Kim, B. C. Kim, S. W. Moon, Y. Jahng, Heterocycles 1995, 41, 1471.
[41] Z. Q. Zhong, Y. Z. Qui, Z. Zhang, H. W. Hu, Chin. Chem. Lett. 1995, 6, 13.
[42] D. H. R. Barton, G. Lukacs, D. Wagle, J. Chem. Soc., Chem. Commun. 1982, 451.
[43] P. Dalla Croze, C. La Rosa, Synthesis 1984, 982.
¬
¬
¬
[44] J. Catalan, J. C. Del Valle, R. M. Claramunt, G. Boyer, J. Laynez, J. Gomez, P. Jimenez, F. Tomas, J.
Elguero, J. Phys. Chem. 1994, 98, 10606.
¬
[45] J. Catalan, J. L. G. De Paz, J. Elguero, J. Chem. Soc., Perkin Trans. 2 1996, 57.
[46] A. Escande, J. Lapasset, R. Faure, E. J. Vincent, J. Elguero, Tetrahedron 1974, 30, 2903.
[47] O. Hager, C. Foces-Foces, N. Jagerovic, J. Elguero, S. Trofimenko, Acta Crystallogr., Sect. C 1996, 52, 2894.
[48] A. A. Dvorkin, A. S. Yavorskii, Yu. A. Simonov, G. E. Kolodeev, S. A. Andronati, T. I. Malinovskii, Dokl.
Akad. Nauk SSSR, 1989, 305, 1378.
[49] R. M. Claramunt, M. D. Santa MarÌa, L. Infantes, F. H. Cano, J. Elguero, J. Chem. Soc., Perkin Trans. 2
2002, 564.
[50] C. Foces-Foces, I. Alkorta, J. Elguero, Acta Crystallogr., Sect. B 2000, 56, 1018.
[51] S. Trofimenko, J. C. Calabrese, J. S. Thompson, Inorg. Chem. 1987, 26, 1507.
[52] K. v. Auwers, K. H¸ttenes, Chem. Ber. 1922, 55B, 1112.
[53] M. H. Palmer, R. H. Findlay, S. M. F. Kennedy, P. S. McIntyre, J. Chem. Soc., Perkin Trans. 2 1975, 1695.
[54] S. Braun, H.-O. Kalinowski, S. Berger, −150 and More Basic NMRExperiments×, Wiley-VCH, 1998.
[55] S. J. Opella, D. M. H. Frey, J. Am. Chem. Soc. 1979, 101, 5855.
[56] G. M. Sheldrick, −SHELX97 Refinement Program×, University of Gˆttingen, Germany, 1997.
Received April 23, 2002