Crystal Engineering with Symmetrical Threefold Acceptor-Substituted Triaminobenzenes

Article abstract of DOI:10.1002/(SICI)1521-3765(19990104)5:1<29::AID-CHEM29>3.0.CO;2-5

The syntheses of threefold acceptor-substituted 1,3,5-triaminobenzene derivatives 2-6 and their crystal structure analyses are described. As acceptors, nitro, trifluoromethylsulfone, and alkylsulfone groups are employed. The combination of hydrogen bondin

Full text of DOI:10.1002/(SICI)1521-3765(19990104)5:1<29::AID-CHEM29>3.0.CO;2-5

FULL PAPER
Crystal Engineering with Symmetrical Threefold
Acceptor-Substituted Triaminobenzenes
J. Jens Wolff,*[a] Frank Gredel,[a] Thomas Oeser,[a]
Hermann Irngartinger,[a] and Hans Pritzkow[b]
Dedicated to Professor Dr. Reinhard W. Hoffmann on the occasion of his 65th birthday
Abstract: The syntheses of threefold acceptor-substituted 1,3,5-triaminobenzene
derivatives 2 ± 6 and their crystal structure analyses are described. As acceptors, nitro,
trifluoromethylsulfone, and alkylsulfone groups are employed. The combination of
hydrogen bonding, arene ´´´ arene and F ´´´ F contacts leads to remarkably similar
solid-state layer structures for sterically quite dissimilar molecules.
Keywords: arenes
neering ´ electrostatic interactions
´ crystal engi-
NO2
SO2R
Introduction
H2N
O2N
NH2
H2N
NH2
R = CF3: 2
R = Me: 3
R = Et: 4
The use of organic molecules in materials science depends
crucially on the superstructures that are formed from them.
To a zeroth-order approximation, individuality of molecules is
conserved in a material, and an observed bulk effect arises
from the vectorial summation of molecular properties. The
mutual orientation of the molecules therefore decides wheth-
er a promising molecular property can be translated into a
useful bulk property. Molecular crystals offer a highly con-
strained and precisely defined environment that guarantees
the highest degree of order observed in bulk structures. By the
same token, their structure is in many cases easily elucidated.
These advantages are countered by the difficulty of predicting
these structures. Despite the numerous crystal structures now
readily available in databases,[1] and despite great progress in
their computation,[2, 3] little is known about how to specifically
design crystal structures at will[4±6]Ðor, more specifically, how
to predict a crystal structure on the basis of a known
molecular constitution alone. It is not even possible to predict,
except in the trivial case of pure enantiomers, whether a
molecule will adopt a centrosymmetric space group or not.[7±9]
The latter question, for example, is of crucial importance to
second-order nonlinear optics.[10±14] The objective of this work
is to design layer structures. We have recently shown that
symmetrical trinitrotriaminobenzene (1, Scheme 1) exists in
NO2 RO2S
SO2R
1
NH2
H2N
NH2
SO2CF3
SO2CF3
NH2 H2N
NH2
F3CO2S
NO2 O2N
NO2
NH2
NH2
5
6
Scheme 1.
two polymorphs, one of which is noncentrosymmetric (space
group P31) and almost ideally suited for use in nonlinear
optics.[15, 16] In both polymorphs, graphitelike layer structures
of approximate trigonal symmetry are found that arise from
strong intra- and intermolecular hydrogen bonding
(Scheme 2). In the centrosymmetric structure the layers are
stacked antiparallel, whereas in the other they are stacked
parallel, but with small displacements relative to each other.
O
O
H
N
N
O
N
O
H
N
N
H
O
H
H
H
O
O
O
N
O
H
N
N
H
N
N
O
N
N
N
N
H
O
H
O
H
O
H
O
H
O
H
O
H
N
N
O
N
O
H
N
N
[a] Priv.-Doz. Dr. J. J. Wolff, Dr. F. Gredel, Dr. T. Oeser,
Prof. Dr. H. Irngartinger
H
O
H
H
H
O
O
O
N
O
H
N
N
H
N
N
O
Organisch-Chemisches Institut der Universität Heidelberg
Im Neuenheimer Feld 270, D-69120 Heidelberg (Germany)
E-mail: wolff@donar.oci.uni-heidelberg.de
H
H
[b] Dr. H. Pritzkow
Anorganisch-Chemisches Institut der Universität Heidelberg
Im Neuenheimer Feld 503, D-69120 Heidelberg (Germany)
Scheme 2.
0947-6539/99/0501-0029 $ 17.50+.50/0
Chem. Eur. J. 1999, 5, No. 1
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999
29

J. J. Wolff et al.
FULL PAPER
This situation is reminiscent of
the stacking disorder found in
graphite. We reasoned that the
intralayer interactions in 1 are
quite strong in comparison to
the interlayer interactions. How
tolerant is the formation of
layer structures with respect to
molecular asymmetry? Is it
possible to identify molecular
architectures that allow predic-
tion of crystal structures? Can
the factors that determine these
structures be identified? To
answer these questions, we
tried to keep the directional
capabilities of hydrogen bond-
ing[17±22] and studied analogues
of 1 where the nitro groups
were replaced with sulfonyl
(2 ± 4) acceptors. We also ana-
lyzed the mixed derivatives 5
and 6 (Scheme 1). While it may
be expected that the exchange
of the trigonal nitrogen present
in a nitro group with a tetrahe-
dral sulfur would lead to a
drastic change in the conforma-
tional and/or crystallographic
behavior, we show here that,
surprisingly, this is not the case.
R = Me, Et
(→ 9,10)
Cl
Cl
Cl
Cl
I
I
Cl
SO2R
Cl
CuSR
I+
NH3
Cl
I
Cl
RO2S
3, 4
MeCO3H
Cl
Cl
SO2R
7
8
11, 12
CuSCF3
Cl
SCF3
Cl
SO2CF3
Cl
CrO3
NH3
F3CS
F3CO2S
2
5
Cl
SCF3
Cl
Cl
SO2CF3
17
18
1. NO+; I3– (→ 16)
2. CuSCF3
I
I
Cl
SCF3
Cl
SO2CF3
Cl
CF3CO3H CrO3
NH3
O2N
H2N
O2N
(→ 22)
1. CuSCF3
13
Cl
SCF3
Cl
SO2CF3
2. SnCl2 (→ 14)
3. H2O2/HCl
23
15
Cl
Cl
SO2CF3
SO2CF3
1. CuSCF3; 2. CrO3
3. SnCl2
H2O2/HCl
or Cl2
SO2CF3
13
H2N
H2N
SO2CF3
20
21
NO2
NO2
CuSCF3
NH2
1. CrO3 (→ 33)
O2N
O2N
O2N
1. H2O2/HCl (→ 35)
2. CF3CO3H
2. NaHS
I
SCF3
SO2CF3
Cl
NO2
Cl
27
29
34
NH3
6
O2N
NaHS
NH2
NaHS
CrO3
NO2
Cl
SO2CF3
Cl
36
NH2
Cl
Cl
NO2
Cl
CuSCF3
CuSCF3
H2O2/HCl
O2N
O2N
O2N
Cl
O2N
Results and Discussion
(→ 31)
I
SCF3
SCF3
Cl
I
32
26
28
30
1. Syntheses: Compounds 2 ± 6
were made by nucleophilic ar-
omatic substitution of the cor-
Scheme 3. Synthesis of 2 ± 6 by nucleophilic aromatic substitution of 18, 11, 12, 23, and 36.
responding trichloro triacceptor compounds (18, 11, 12, 23, 36,
of which 23 had been prepared by others before;[23]
Scheme 3). Synthesis of the alkylsulfone precursors 11 and
12 proved straightforward: the key step is the threefold
Ullmann reaction[24] of 8 with the corresponding CuI thiolate
(modified by the addition of a,a-bipyridyl), followed by
standard oxidation with peracetic acid formed in situ.
Trichlorotriiodobenzene 8 was prepared by use of a known[25]
oxidative iodination method. Unfortunately, due to redox
processes, reaction with CuSCF3 gives a very low yield of
impure product under these conditions, so synthesis of 18 was
attempted by two routes starting from 13.[26] The route via 20
was not successful, because chlorination of this deactivated
and sterically encumbered substrate only gave the dichlori-
nated product 21. Triple chlorination, however, is easy with
the less deactivated 14.[23]
the introduction of the sulfone functionality at an early stage
and oxidation of an amino to a nitro group with concentrated
trifluoroperacetic acid. Thus, 27 is converted by a clean
reaction with CuSCF3 and subsequent CrVI oxidation to 33.
One nitro group has then to be reduced to activate the ring for
the chlorination. While partial reduction of sym-trinitroben-
zene with ammonium sulfide is possible in moderate yield,[27]
the formation of Meisenheimer complexes and their respec-
tive follow-up reactions compete with reduction. A trifluoro-
methylsulfone favors their formation even at lower pH with
respect to a nitro group.[28±30] The pH of the reaction mixture
had therefore to be kept as low as possible, and a solution of
NaHS[31] was employed to produce 34 in moderate yield.
Chlorination and oxidation then gave 36.
2. Crystal structures
Four routes had to be tried for the synthesis of 36. It turns
out that reaction with CuSCF3 is unsuccessful both at the
hexa-substituted stage and when an amino functionality is
present. In addition, oxidation of the sulfide to the sulfone as
the final step proved capricious. The successful route relies on
2.1 Molecular parameters: Due to the substitution with three
donors and three acceptors, p-electron density is shifted from
the benzene core to the substituents and the p system can
adopt several multipolar resonance structures (Figure 1).[32]
30
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999
0947-6539/99/0501-0030 $ 17.50+.50/0
Chem. Eur. J. 1999, 5, No. 1

Crystal Engineering
29±38
δ−
A
Structures with highly nonplanar rings result in the case of N-
substituted derivatives, but six intramolecular hydrogen
bonds planarize the ring.[33] These are also present in 2 ± 6 to
oxygen atoms of both types of acceptor groups (Table 1).
The intramolecular hydrogen bonding to a sulfonyl oxygen is
weaker than to a nitro oxygen atom: the distances H ´´´ O(S)
and (H)N ´´´ O(S) are larger than the corresponding dis-
tances H ´´´ O(N) and (H)N ´´´ O(N). This can be ascribed to
the longer C S bond and the tetrahedral environment of
sulfur.
The C C bond length alternation in 2 ± 6 is insignificant, but
the average endocyclic C C bonds are elongated and the
exocyclic C N and C S bonds shortened (Table 1) as the
comparison with the following reference values[34] reveals:
substituted Car ± Car: 1.397, Car ± Nsp2H2: 1.355, Car ± NO2: 1.468,
Car ± SO2C: 1.763 Š. The effects are less pronounced in 3 and
4; this is in accordance with the weaker electron-withdrawing
character of alkylsulfones.[35] Nevertheless, the radialene form
assumes considerable weight in the description of all of the p
δ+
δ−
δ+
δ−
D
A
D
D
A
D
A
A
D
A
D
A
D
A
D
0δ
D
δ+
A
Benzenoid
Octopolar Charge Distribution
Radialenoid
6δ−
δ+
δ+
H
δ+
δ+
H
H
H
H δ+
δ+
δ+ δ+
δ+
δ+
H
H
H
H
H
H
6δ−
6δ−
H
H
δ+
δ+
δ+
H
H
δ+
δ+
H
Quadrupolar Interactions
Figure 1. Charge distribution in symmetrical threefold donor± acceptor
substituted benzenes, and preferred stacking of quadrupolar benzene
derivatives.
Table 1. Selected structural features of 2 ± 6.[a]
5A
1B
3A
1B
5A
1B
O
O
O
O
O
O
H
S
H
H
S
H
H
S
1
H
1
1
3
1
2
1
3
1
N
N
N
N
N
N
1
4
3
6
1
4
5B H
H1A 3B H
H1A 5B H
H1A
6
5
2
3
4
5
2
1
6
5
2
3
3
S
2
S
2
S
4
N
5
N
4
N
O
O
5 O
O
O
O
O
N
2
O
O
N
3
O
O
N
2
O
2
5
6
H
3A
H
3B
H
5A
H
5B
H
3A
H
3B
3A
1B
5B
1A
O
O
O
O
H
S
H
H
S
H
1
1
2
1
3
1
N
N
N
N
1
4
1
4
3B
1A
5A
1B
H
O
H
O
H
O
H
O
6
5
2
3
6
5
2
3
1'
S
2
S
2'
S
2
S
O
N
3
O
O
N
2
O
3
4
H
H
H
H
5A
5B
3B
3A
2
3
4
5
6
bond lengths [Š]
C1 ± C2
C2 ± C3
C3 ± C4
C4 ± C5
1.443(3)
1.434(5)
1.436(5)
1.442(3)
1.431(5)
1.435(4)
1.437
1.322
±
1.727
29.2
1.426(2)
1.424(2)
1.424(2)
1.426(2)
1.422(2)
1.422(2)
1.424
1.348
±
1.757
32.2
1.427(3)
1.417(3)
1.427(3)
1.427(3)
1.417(3)
1.427(3)
1.424
1.340
±
1.763
10.0
1.447(3)
1.433(4)
1.430(4)
1.432(3)
1.435(4)
1.437(4)
1.436
1.322
1.419
1.733
30.3
1.435(7)
1.444(7)
1.444(7)
1.446(7)
1.448(7)
1.444(7)
1.444
1.322
1.413
1.746
15.3
C5 ± C6
C6 ± C1
average C(ar) ± C(ar)
average C(ar) ± NH
average C(ar) ± NO
average C(ar) ± S
S j tors.aj [8][b]
hydrogen bonds
intramolecular
range H ´´´ O
range N ´´´ O
intermolecular[c]
range H ´´´ O
N1 ´´´ O
1.90(3) ± 2.03(3)
2.648(4) ± 2.664(3)
1.95(2) ± 2.05(2)
2.672(2) ± 2.709(2)
1.88(2) ± 2.02(2)
2.621(2) ± 2.718(3)
1.81(3) ± 1.97(3)
2.503(4) ± 2.667(3)
1.80(6) ± 1.88(8)
2.503(7) ± 2.657(7)
2.54(4) ± 2.62(3)
3.199(3)
3.194(4)
3.190(4)
3.173(3)
2.58(2)
2.991(2)
±
±
2.63(2)
2.968(3)
±
±
2.48(3) ± 2.65(3)
3.017(3)
3.088(3)
±
2.32(8) ± 2.42(5)
±
±
3.020(6)
2.984(6)
2.971(6)
3.015(6)
N1 ´´´ O
N2 ´´´ O
N2 ´´´ O
N3 ´´´ O
±
±
±
3.194(3)
3.187(3)
2.991(2)
±
2.968(3)
±
3.112(3)
3.167(3)
N3 ´´´ O
[a] Numbering as defined in accompanying scheme; does not always coincide with numbering in X-ray file due to different molecular structure and site
symmetry. [b] Sum of absolute torsional angles within the six-membered ring (measure of puckering). [c] All contacts for H ´´´ O ꢀ 2.7, and O ´´´ N ꢀ 3.2 Š.
Chem. Eur. J. 1999, 5, No. 1
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999
0947-6539/99/0501-0031 $ 17.50+.50/0
31

J. J. Wolff et al.
FULL PAPER
Table 2. Structure determination summary of 2 ± 6.[a]
2
3
4
5
6
empirical formula
crystal size
habit
solvent
crystal system
space group
a [Š]
b [Š]
c [Š]
a [8]
b [8]
C9H6F9N3O6S3
0.3 Â 0.20 Â 0.15
colorless prisms
nitromethane
triclinic
P
9.453(1)
9.464(2)
11.080(1)
81.35(1)
86.76(1)
60.09(1)
849.2
C9H15N3O6S3
0.3 Â 0.25 Â 0.2
light yellow prisms
acetonitrile
orthorhombic
Pnma
10.060(1)
16.549(2)
8.434(1)
90
C12H21N3O6S3
0.45 Â 0.40 Â 0.25
yellow prisms
toluene
orthorhombic
Pnma
10.818(3)
14.177(3)
11.461(3)
90
C8H6F6N4S2O6
0.35 Â 0.3 Â 0.2
yellow prisms
toluene/acetone
monoclinic
P21/c
13.024(2)
9.368(1)
12.491(2)
90
C7H6F3N5O6S
0.35 Â 0.27 Â 0.03
yellow platelets
dichloromethane
monoclinic
C2/c
30.345(15)
9.074(5)
8.409(4)
90
90
90
90
90
108.77(1)
90
1443.0(6)
4
97.55(3)
90
2295(2)
8
g [8]
V [Š3]
Z
1404.0(5)
4
1757.7(13)
4
2
Mr
1calcd [gcm
519
2.03
357
1.69
399
1.51
432
1.99
345
2.00
3
]
absorption correction
correction factor (max/min)
absorption coeff. [cm
F(000)
empirical
1.0/0.86
5.48
empirical
1.0/0.98
5.36
numerical
0.92/0.85
4.36
empirical
1.0/0.95
4.63
empirical
0.85/1.00
3.71
1
]
516
744
840
864
696
range of V [8]
2 ± 26.5
3754
3507
2 ± 28
1730
1730
1408
2 ± 28
2186
2186
1305
2 ± 28
3605
3455
2442
2 ± 25
2014
2014
1066
collected reflns
independent reflns
observed reflns[b]
2363
hydrogen atoms refined[c]
residual electron density [eŠ
parameters refined
R/Rw
all N ± H
0.62/ 0.17
296
0.040/0.054
2.35
all H
0.29/ 0.34
134
0.032/0.042
2.02
all N ± H
0.25/ 0.39
150
0.039/0.046
1.83
all N ± H
0.32/ 0.24
259
0.039/0.055
2.21
all N ± H
0.28/ 0.40
223
0.055/0.137[d]
1.01
3
]
S value
[a] Four-circle diffractometers, room temperature, MoKa radiation. Programs for structure solution: 2: SIR92, DIRDIF; 3: SHELXS-86; 4: SIR88 (14
reflections excluded from refinement); H51, H61, H62 on calculated positions. 5: SIR88; 6: SHELXL97. Crystallographic data (excluding structure factors)
for the structures reported in this paper have been deposited with the Cambridge Crystallographic Data Center as supplementary publications nos. CCDC-
104511 ± 104515. Copies of the data can be obtained free of charge on application to CCDC, 12 Union Road, Cambridge CB21EZ, UK (fax: (‡44)1223-336-
033; e-mail: deposit@ccdc.cam.ac.uk). [b] I ꢁ 2.5Ds(I); for 6: I ꢁ 2.0Ds(I). [c] Those hydrogen atom positions that were located on the electron density
difference map and refined isotropically. [d] Weighted R factor based on F 2 instead of F for all measured reflections.
systems investigated. Likewise, the charge distribution has a
strong octopole moment, but most likely a low quadrupole
moment. For example, both benzene and hexafluorobenzene
have strong quadrupole moments of opposite signs, but the
small moment of 1,3,5-trifluorobenzene closely corresponds
to the arithmetic mean of the two former moments.[36] The
significance for crystal packing lies in the fact that packing an
arrangement of planar p systems with like quadrupole mo-
ments is optimized in T-shape geometries (Figure 1). Thus the
familiar herringbone patterns, but not layer structures, are
observed as the characteristic motifs for the packing of
aromatic hydrocarbons (although many exceptions are
known) that are also favorable in solution.[37, 38] Such a
counteracting force is not present in molecules with a low
quadrupolar moment. Layer structures with p ± p stacking
only result from intermolecular complexes of planar p
systems with opposite quadrupolar moments, like in the
well-known benzene ± hexafluorobenzene complex and sim-
ilar structures.[36, 39±43] It should be noted that minimal
intermolecular charge transfer occurs in these complexes.[36]
The presence of an octopolar charge distribution (Figure 1)
should also favor a layer structure formed from like mole-
cules, as the absence of a quadrupolar moment does. As the
interaction energy of octopoles is small, however, the purely
electrostatic influence is not likely to be structure-determin-
ing.
2.2 Crystal packing: The structures of fluorinated sulfones
will be discussed first. Sulfone 2, in which all three nitro
groups have been replaced, isÐin contrast to the general
observations with fluorinated sulfonesÐa poorly soluble
compound, and can be induced to form crystals suitable for
X-ray analysis only with difficulty (the results of the X-ray
structure determination are given in Table 2). The crystals are
easily cleaved. Only one crystal form suitable for X-ray
analysis could be found. The structure in the solid state is
strikingly similar to the one of 1: intermolecular hydrogen
bonds are present which guarantee the formation of layers
(Figure 2). The latter are of approximate trigonal symmetry
and stacked in parallel. As a consequence, the CF3 groups of
one molecule all point in one direction, a conformation not
likely to be favored in solution simply on the basis of statistics.
Two basic stacking arrangements of layers are compatible
with this arrangement of layers in 2, namely, one where the
layers expose like sides to each other, and one where unlike
sides are in contact. Sulfone 2 forms a structure reminiscent of
a micelle: the layers are arranged in a centrosymmetric
fashion, hence, like sides are in contact. Apparently, the
contact F ´´´ p system is not favorable in 2. A particular
attractive interaction between the fluorine groups that point
to each other cannot be rigorously excluded, but we note that
the closest distance between them, 3.087(5) Š, is slightly
larger than the sum of the van der Waals radii (a radius of
32
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999
0947-6539/99/0501-0032 $ 17.50+.50/0
Chem. Eur. J. 1999, 5, No. 1

Crystal Engineering
29±38
Figure 4. Intermolecular hydrogen bonding in layers of 5.
between the least squares plane through the fluorine layer and
the plane though the six-membered ring amounts to 278. The
ribbonlike packing pattern with its hydrogen-bonded inner
side and its outer fluorine layer creates the impression of base-
pairing in DNA (although, of course, no helical twist is
present here), or a micellar structure where the hydrophobic
fluorine layers are in contact and enclose a hydrophilic
hydrogen-bonded arene layer. The overall features of this
structure are reproduced in solid 6, which shows a surprisingly
similar arrangement of layers (Figures 5 and 6; closest F ´´´ F
contact: 2.967 Š). At 538, the inclination angle of the layers is
Figure 2. Packing diagram of 2.
1.47 Š is generally assumed for Ar± F[44]). A recent statistical
survey[45] of the 1995 release of CSD[1] has revealed 385
structures with a range of F ´´´ F contacts of 2.40 ± 2.95 Š, and
1908 structures with 2.95 ± 3.50 Š. Hence, the closest distance
found here is rather on the long side.
If the structural motifs (layer structures through hydrogen-
bonding and exposition of like sides resulting in p ± p
stacking) persisted, planar layer structures would no longer
be expected with 5 and 6 because there would be a void in the
F ´´´ F layer, especially in 6. Space filling can be effected,
however, if the layers are tilted with respect to each other.
Accordingly, a modified layer structure is present in crystal-
line 5 (Figures 3 and 4; structure determination details in
Figure 5. Packing diagram of 6.
Figure 3. Packing diagram of 5.
Table 2). Again, intermolecular hydrogen bonding is notice-
able (Figure 4), but only the nitro groups and two of the amino
groups are involved in it. The amino group in the position
between the sulfonyl groups does not engage in intermolec-
ular hydrogen bonding, nor does one of the sulfonyl groups.
No indication of any F ´´´ H hydrogen bonding is detectable in
accordance with recent statistical surveys;[46, 47] the closest
distance amounts to 2.71(3) Š. Tilted layers, for the arrange-
ment of arenes better described as ribbons, are formed where
like sides are exposed as in the structure of 2. The tilt angle
Figure 6. Intermolecular hydrogen bonding in layers of 6.
close to double the value for 5 for reasons connected with
space-filling: the fluorine layer is formed only by one instead
of two CF3 groups. Further hypothetical increase of the angle
on exchange of the last sulfone with a nitro group to give 1
again yields a value of close to 908Ðwhich corresponds to a
parallel stacking of the benzene rings and thus to the planar
layer structure which is indeed found for 1 (Figure 7).
Chem. Eur. J. 1999, 5, No. 1
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999
0947-6539/99/0501-0033 $ 17.50+.50/0
33

J. J. Wolff et al.
FULL PAPER
VT-NMR (see Experimental Section). If the methyl groups
are replaced with bulkier ethyl groups, as in 4, the conforma-
tion changes to the one favored in solution, where the alkyl
chains point in both directions (Figure 9). VT-NMR shows
0°
3 SO2CF3 (2)
27°
0 SO2CF3 (1)
2 SO2CF3 (5)
57°
1 SO2CF3 (6)
Figure 7. Inclination of planes through fluorine layers with respect to
planes through six-membered rings in 1, 2, 5, and 6.
If the structures of 2, 5, and 6 are determined both by the
attractive hydrogen bonding and the tendency for like sides of
the molecule to be exposed, then exchange of CF3 with CH3
groups should lead to a drastic change in the observed crystal
structure, although, on a purely geometric basis, this change is
minor in comparison to the structural differences between 2,
5, and 6. As shown in Figure 8, 3 does not form a layer
structure (see also Table 2). The packing can be described as
Figure 9. Packing diagram of 4.
1
that this conformation is favored by at least 7.8 kJmol at
240 K. However, the structural features are preserved to some
extent; it may therefore be concluded that this arrangement is
particularly favorable. It should be noted, however, that the
dimers slip apart and the hydrogen bonding, again with quite
an acute angle O-H-N of 1078 and with the distances on the
verge of the accepted boundaries, now no longer occurs within
the dimeric units, but between them.
Conclusion
We have shown that structures with planar or tilted layers are
formed from 1,3,5-triaminobenzenes substituted with any
combination of nitro or trifluoromethylsulfonyl groups. The
arrangement is always centrosymmetric. These structures are
satisfactorily explained as arising from a combination of
intermolecular hydrogen bonding and the tendency of the
fluorine substituents to segregate and to form layers (or
ribbons). We suggest that the combination of these two effects
should be a powerful control element in crystal engineering.
This view is corroborated by the completely different
structures observed when alkylsulfones are used as electron-
attracting substituents.
Figure 8. Packing diagram of 3.
an arrangement of sandwich-type dimers that are oriented in
an almost perpendicular fashion with respect to each other.
There are two symmetry-related hydrogen bonding contacts
between the two molecules in the dimeric unit. The total of
four such interactions probably assists in its formation, even
though with 1138 the angle O-H-N is quite acute. In order to
form these dimers effectively, the methyl groups have to point
in one direction, a conformation not observed in solution by
Experimental Section
Melting points: hot-stage microscope or Büchi B-540. 1H NMR recorded at
300.135 MHz and 13C NMR at 75.469 MHz at RT, unless noted otherwise.
1,3,5-Triiodo-2,4,6-trichlorobenzene (8): A known iodination method was
used.[25] Arene 8 has been claimed to form in 36% yield from reaction of
‡
trichlorobenzene (7) with Py2I BF4 , but no characterization was given.[48]
Powdered iodine (19.2 g, 75.1 mmol) was added to a vigorously stirred
34
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999
0947-6539/99/0501-0034 $ 17.50+.50/0
Chem. Eur. J. 1999, 5, No. 1

Crystal Engineering
29±38
solution of periodic acid (5.8 g, 25 mmol) in conc sulfuric acid (180 mL).
After 15 min, 7 (2.722 g, 15.00 mmol) was added to the dark solution, and
the mixture stirred for 18 h at RT. It was poured onto ice, the precipitate
was filtered, and washed with water, with a solution of bisulfite until the
color disappeared, with methanol, and finally with ether. The residue
(8.266 g, 98.6%), nearly pure material, could be used directly in the next
step. It could be recrystallized from dioxane to give colorless needles, m.p.
280.5 ± 281.58C (6.615 g, 11.83 mmol, 78.9%). 1H NMR (CDCl3): no signal;
13C NMR (dioxane/CDCl3, 50.3 MHz, 350 K): d ˆ 97.13, 144.95; anal. calcd
for C6Cl3I3: (559.13) C 12.89, Cl ‡ I 87.11, I 68.09; found: C 12.96, Cl ‡ I
87.08, I 67.96.
precipitate washed with water. 12 was obtained as small colorless prisms
(983 mg, 2.15 mmol, 90.0%), m.p. ca. 260 ± 2758C (decomp) which could be
recrystallized from dioxane without improvement in purity or alteration of
the decomposition point. 1H NMR ([D6]DMSO; decomposes slowly in this
solvent): d ˆ 1.12 (t, J ˆ 7.3 Hz, 6H), 1.26 (t, J ˆ 7.4 Hz, 3H), 3.40 (q, J ˆ
7.4 Hz, 2H), 3.49 (q, J ˆ 7.3 Hz, 4H); 13C NMR ([D6]DMSO): d ˆ 6.77, 7.20
(double int), 50.16 (double int), 50.99, 114.17, 126.38 (double int), 140.07
‡
(double int), 168.95; MS (EI): 456 (M for 35Cl3, 6.3%; isotope cluster in
accordance with proposed formula); anal. calcd for C12H15Cl3O6S3 (457.80):
C 31.48, H 3.30, Cl 23.23; S 21.01; found: C 31.59, H 3.42, Cl 23.07, S 20.80.
1,3,5-Tris(methylsulfonyl)-2,4,6-triaminobenzene (3): Under slight pres-
sure, dry ammonia was bubbled through a solution of the trichloride 11
(0.800 g, 1.92 mmol) in dry DMSO (10 mL). A yellow color developed, and
the mixture was kept tightly closed overnight. It was diluted with water and
neutralized with dil HCl, then extracted with dichloromethane. The organic
layers are washed with water and satd brine, and dried to give a faintly
yellow powder (673 mg, 1.88 mmol, 97.9%) which is further purified by
chromatography (silica, CH2Cl2/2% MeOH); from toluene: large, light
1,3,5-Tris(methylthio)-2,4,6-trichlorobenzene (9): A mixture of 8 (5.048 g,
9.028 mmol), methylthiocopper (3.919 g, 35.42 mmol; prepared from
methanethiol and copper(ii) acetate) and a,a-dipyridyl (5.532 g,
35.42 mmol) in DMF (50 mL) was stirred at RT for 2.5 d. The mixture is
then diluted with diethyl ether (300 mL) and is filtered. The precipitate was
washed with the same solvent, and the combined extracts were washed
three times with water, then satd brine, and were dried. Chromatography
(silica, CCl4) gave colorless prisms (2.042 g, 6.387 mmol, 70.7%). An
analytical sample was recrystallized from n-heptane, m.p. 114 ± 114.58C.
1H NMR (CDCl3): d ˆ 2.43 (s); 13C NMR (CDCl3): d ˆ 18.44, 135.28,
146.73; anal. calcd for C9H9Cl3S3 (319.73): C 33.81, H 2.84, Cl 33.26, S 30.09;
found: C 33.85, H 2.83, Cl 33.42, S 30.04.
1
yellow plates, decomp >2808C. H NMR ([D6]DMSO): d ˆ 3.23 (s, 9H),
7.73 (brs, 6H); 13C NMR ([D6]DMSO): d ˆ 42.86, 93.41, 152.35. The
molecule has Cs symmetry on the NMR time scale at 220 K and below (one
methyl group up, two down). The 1H NMR shows the intramolecular
hydrogen bonds to be frozen because the three expected singlets of equal
intensities in the N ± H region are found: 1H NMR (240 K, [D6]acetone):
d ˆ 3.25 (s, 6H), 3.29 (s, 3H), 7.90 (s, 2H), 7.94 (s, 2H), 8.03 (s, 2H); anal.
calcd for C9H15N3O6S3 (357.43): C 30.24, H 4.23, N 11.76, S 26.91; found: C
30.27, H 4.27, N 11.88, S 26.68.
1,3,5-Tris(ethylthio)-2,4,6-trichlorobenzene (10):
A suspension of 8
(5.048 g, 9.028 mmol), copper ethylthiolate (4.416 g, 35.42 mmol; from
freshly precipitated cuprous oxide and ethanethiol), and a,a'-bipyridyl
(5.532 g, 35.42 mmol) in dry DMF (50 mL) was vigorously stirred for 4 d at
RT. The mixture was diluted with diethyl ether (200 mL), and the red-
brown copper complex filtered and washed with this solvent (200 mL). The
filtrate was washed with water (4 Â 300 mL) and brine, and dried (Na2SO4).
The solvent was removed and the residue (4.007 g) chromatographed on
silica gel eluting with CCl4/light petroleum (1/1). Compound 10 was
obtained as a colorless solid (2.160 g, 2.52 mmol, 66.1%), m.p. 51 ± 578C.
An analytical sample was recrystallized from ethanol to give colorless
prisms, m.p. 57.5 ± 588C. 1H NMR (CDCl3): d ˆ 1.19 (t, J ˆ 7.4 Hz, 9H), 2.93
(q, J ˆ 7.4 Hz, 6H); 13C NMR (CDCl3): d ˆ 14.52, 29.66, 133.54, 147.93;
anal. calcd for C12H15Cl3S3 (361.80): C 39.84, H 4.18, Cl 29.40, S 26.59;
found: C 39.75, H 4.18, Cl 29.15, S 26.70. Reduction is a side reaction; 1,3-
bis(ethylthio)-2,4,6-trichlorobenzene was isolated as a colorless oil in 12%
yield: 1H NMR (CDCl3): d ˆ 1.19 (t, J ˆ 7.4 Hz, 6H), 2.91 (q, J ˆ 7.4 Hz,
4H), 7.52 (s, 1H); 13C NMR (CDCl3): d ˆ 14.52, 29.45, 128.78, 132.86,
141.35, 148.26.
1,3,5-Tris(ethylsulfonyl)-2,4,6-triaminobenzene (4): Dry ammonia was
bubbled through a suspension of the sulfone 12 (326 mg, 0.712 mmol) in
dry DMSO (15 mL); the vessel was held at an overpressure of 0.5 bar. The
mixture turned yellow, and the sulfone dissolved. The mixture was allowed
to stand overnight; dil HCl was then added and the mixture extracted with
dichloromethane (3 Â 15 mL). The combined organic layers were washed
with water (3 Â ) and brine, and dried (Na2SO4). The solvent was
evaporated and the residue (284 mg) chromatographed over silica gel
(200 g) eluting with diethyl ether. Compound 4 was isolated as colorless
prisms (201 mg, 0.503 mmol, 70.7%), m.p. 176.5 ± 1788C (from n-heptane/
toluene). 1H NMR (CDCl3): d ˆ 1.32 (t, J ˆ 7.4 Hz, 9H), 3.23 (q, J ˆ 7.3 Hz,
6H), 7.89 (brs, 6H); 13C NMR (CDCl3): d ˆ 6.94, 49.31, 91.18, 154.04. The
molecule has Cs symmetry on the NMR time scale at 240 K and below (one
ethyl group up, two down). The 1H NMR shows the intramolecular
hydrogen bonds to be frozen because the three expected singlets of equal
intensities are found in the N ± H region. A small additional singlet in this
region integrating for 0.1H may be ascribed to the presence of the C3
symmetric conformer where all N ± H are isochronous. This conformer
would then be present in roughly 2%: 1H NMR (500 MHz, 240 K, CD2Cl2):
d ˆ 1.24 (t, J ˆ 7.2 Hz, 9H), 3.19 (m, 8H), 7.76 (s, 2H), 7.82 (s, 0.1H), 7.88 (s,
2H), 7.93 (s, 2H); 13C NMR (125 MHz, 240 K, CD2Cl2): d ˆ 7.17, 48.98 and
49.06 (int 1:2), 90.11 and 90.33 (int 1:2), 153.53 and 154.55 (int 2:1); anal.
calcd for C12H21N3O6S3 (399.51): C 36.08, H 5.30, N 10.52, S 24.08; found: C
36.23, H 5.27, N 10.54, S 24.11.
The standard conditions to effect this Ullmann reaction (hot quinoline[49]
)
failed to produce any desired compound from 8; the addition of bipyridyl
(we would like to thank Prof. H.-J. Grützmacher for this suggestion) has
been found to drastically increase the yield because the reaction can
proceed at RT; otherwise, reductive removal of one iodine atom is
observed to a considerable extent.
Reaction of 8 with CuSCF3 failed to give the desired substitution product
17 in acceptable yields; about 5% of an inseparable mixture of 17 and the
corresponding reduction product bis(trifluoromethylthio)trichlorobenzene
was isolated.
3,5-Bis(trifluoromethylsulfonyl)-2,6-dichloroaniline (21): A solution of the
aniline 20[26, 50] (418 mg, 1.17 mmol) in acetic acid (10 mL) and 15% HCl
(2 mL) was saturated with chlorine gas at RT. After 30 min, colorless
needles separate; the solution is saturated with chlorine gas again after 3 h.
After standing overnight, the solution is extracted with dichloromethane to
give almost pure (TLC) 21 (502 mg, quant) containing only small amounts
of the trichlorination product as judged by MS. Conducting the reaction at
reflux temperature with continuous saturation with chlorine only decreased
the yield of 21; only small amounts of trichlorinated product were formed.
Faintly yellow, large prisms, m.p. 163.5 ± 1648C (slow evaporation from
CH2Cl2), ref. [23] m.p. 157 ± 1588C. 1H NMR ([D6]acetone, 500 MHz): d ˆ
6.82 (brs, 2H), 8.11 (s, 1H); 13C NMR ([D6]DMSO, 125 MHz): d ˆ 120.63
(q, J ˆ 326.4 Hz), 124.71 (CH), 126.87, 130.13 (q, J ˆ 2.0 Hz), 148.62; MS
1,3,5-Tris(methylsulfonyl)-2,4,6-trichlorobenzene (11): To a stirred solution
of the tristhioether 9 (1.610 g, 5.035 mmol) in glacial acetic acid (20 mL) at
1008C, 30% hydrogen peroxide (20 mL) was added dropwise. After being
stirred for a further 2 h, the mixture was cooled, and the precipitate formed
filtered and washed with water (1.608 g, 3.868 mmol, 76.8%). An analytical
sample was crystallized from nitromethane, colorless needles, decomp
>2858C. The trisulfone is poorly soluble in most solvents; solutions of
analytically pure samples in CD3CN or DMSO decomposed appreciably.
Impurities are not quoted in the following NMR data: 1H NMR (CD3CN):
d ˆ 3.42 (s); 13C NMR ([D6]DMSO): 45.34, 116.61, 128.91 (double
intensity), 138.73 (double int), 167.99; anal. calcd for C9H9Cl3O6S3
(415.72): C 26.00, H 2.18, Cl 25.58, S 23.14; found: C 26.28, H 2.24, Cl
25.78, S 23.31.
‡
(EI): M : 431 (0.8), 430 (1.2), 429 (11.5), 428 (5.5), 427 (45.7), 426 (8.2), 425
(53.5); intensities in accordance with composition C8H3Cl2F6NO4S2.
1,3,5-Tris(ethylsulfonyl)-2,4,6-trichlorobenzene (12):
A
solution of 10
1,3,5-Tris(trifluoromethylsulfonyl)-2,4,6-triaminobenzene (2): From 18
(260 mg, 0.450 mmol) and ammonia in dry dichloromethane. The precip-
itate was filtered after 3 d and washed with water. The residue was
crystallized from nitromethane to give colorless small platelets with a
pinkish tinge, m.p. 261.5 ± 262.58C (55 mg, 0.11 mmol, 24%). The analyti-
(863 mg, 2.39 mmol) in glacial acetic acid (10 mL) was heated to 1008C
and treated dropwise with 30% hydrogen peroxide (10 mL, 88 mmol).
After 15 min, a colorless precipitate began to form, and heating was
continued for 2 h. After cooling, the suspension was filtered and the
Chem. Eur. J. 1999, 5, No. 1
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999
0947-6539/99/0501-0035 $ 17.50+.50/0
35

J. J. Wolff et al.
FULL PAPER
cally pure compound shows two signals in the 1H NMR and also two signals
for C ± S, possibly indicating freezing in the conformation with Cs symmetry.
1H NMR ([D6]DMSO): d ˆ 8.33 and 8.53 (brs; int 2.7:1); 13C NMR
([D6]DMSO): d ˆ 80.72 (br), 80.94 (br), 120.28 (q, J ˆ 328.1 Hz), 157.07
(br); anal. calcd for C9H6F9N3O6S3 (519.35): C 20.81, H 1.16, N 8.09, S 18.52;
found: C 20.78, H 1.13, N 8.06, S 18.39.
CH2Cl2/PE ˆ 1/1; 9.553 g). Recrystallizion from n-heptane gives yellow
needles (9.142 g, 38.38 mmol, 57.6%), m.p. 117 ± 1218C. For analysis, a
sample is recrystallized again from heptane and then sublimed (1108C/0.1
Torr), m.p. ca. 908C, recrystallizing and then melting at 118.5 ± 1208C.
1H NMR (CD2Cl2): 7.24 (prob. ddq, J ˆ 1.5, 2.5 and 0.5 Hz), 7.58 (pseudo-t,
J ˆ 2.2 Hz), 7.79 (ddq, J ˆ 1.5, 2.5 and 0.5 Hz); 13C NMR (CD2Cl2): d ˆ
111.38, 119.61 (q, J ˆ 1.0 Hz), 126.95 (q, J ˆ 2.5 Hz), 127.12 (q, J ˆ 0.8 Hz),
129.78 (q, J ˆ 308.4 Hz), 148.95, 149.86 (br); anal. calcd for C7H5F3N2O2S
(238.19): C 35.30, H 2.12, N 11.76, S 13.46; found: C 35.28, H 2.15, N 11.80, S
13.58.
1,3-Bis(trifluoromethylsulfonyl)-5-nitro-2,4,6-trichlorobenzene (23): From
the nitroso compound 22 by oxidation with CrO3.[23] Compound 22 in turn
was obtained from 15 by reaction with trifluoroperacetic acid. We found the
reaction 15 !22 to proceed at 10 to 58C for 2 h, but only slowly at the
reported temperature of 208C.[23] Prolonged reaction times at RT led to
overoxidation. Data for 23: m.p. 136 ± 1388C (lit. m.p.); 1H NMR (CD3CN)
no signal; 13C NMR (CD3CN): d ˆ 120.78 (q, J ˆ 327.5 Hz), 127.29, 134.45
(q, J ˆ 2.9 Hz), 139.77, 147.95, 152.88 (br).
1,3-Bis(trifluoromethylsulfonyl)-5-nitro-2,4,6-triaminobenzene (5): Am-
monia was bubbled through a solution of 23 (258 mg, 0.526 mmol) in dry
CH2Cl2 (30 mL). After 1 d the precipitate was filtered over silica (acetone,
CHCl3) to give a yellow powder (232 mg, quant) which was recrystallized
from toluene to give yellow prisms (122 mg), 243.5 ± 244.58C. 1H NMR
([D6]DMSO): d ˆ ca. 8.4 and 9.0 (vbrs; int ca. 1:2); 13C NMR
([D6]DMSO): d ˆ 79.73 (br), 114.62, 120.33 (q, J ˆ 328.4 Hz), 152.98,
156.10; anal. calcd for C8H6F6N4O6S2 (432.28): C 22.23, H 1.40, N 12.96;
found: C 22.32, H 1.45, N 13.04.
2,4,6-Trichloro-3,5-dinitroaniline (25):[51] A solution of 3,5-dinitroaniline[52]
(5.494 g, 30.60 mmol) in glacial acetic acid (130 mL) and conc HCl (85 mL)
was treated at 10 ± 168C with 30% H2O2 (43 mL, 0.42 mol); the cooling
bath was removed, and the temperature was allowed to rise to 328C. Yellow
crystals separated and were collected once the exothermic reaction had
subsided (7.539 g, 26.32 mmol, 87.7%). Crystallization from toluene/n-
heptane gave yellow prisms, m.p. 229.5 ± 231.58C (lit. m.p. 2308C[51] from
EtOH). 1H NMR ([D6]DMSO): 7.24 (s); 13C NMR ([D6]DMSO): d ˆ
100.61, 110.48, 144.04, 146.64.
2,4,6-Trichloro-3-nitro-5-trifluoromethylthioaniline (31): The aniline 30
(0.800 g, 3.36 mmol) is chlorinated (9.5 mL HCl, 15 mL AcOH, 4.8 mL
30% H2O2) as described for 25. Crude yield, 1.084 g (3.17 mmol, 94%),
yellow prisms from heptane (766 mg, 2.24 mmol, 66%) that contain a small
amount of dichlorinated product, m.p. 92.5 ± 998C. The mother liquor is
chromatographed (PE/CH2Cl2 ˆ 7/3) and crystallized again from heptane;
yellow prisms, m.p. 99 ± 1018C. 1H NMR (CDCl3): 5.02 (brs); 13C NMR
(CDCl3): d ˆ 113.79, 120.47 (q, J ˆ 1.2 Hz), 123.92 (q, J ˆ 2.3 Hz), 128.38 (q,
J ˆ 0.8 Hz), 128.56 (q, J ˆ 312.2 Hz), 141.38, 143.22 (?, br); anal. calcd for
C7H2Cl3F3N2O2S (341.53): C 24.62, H 0.59, N 8.20, Cl 31.14, S 9.39; found: C
24.61H, 0.70, N 8.24, Cl 31.37, S 9.13.
2,4,6-Trichloro-3,5-dinitrotrifluoromethylthiobenzene (32): CAUTION!
Concentrated hydrogen peroxide and trifluoroperacetic acid are powerful
oxidants and may explode in contact with organic matter (dust particles,
residues in sintered glass funnels) or metal salts! For the handling of these
compounds, only glassware thoroughly cleaned with conc sulfuric acid and
rinsed with distilled water should be used. The aniline 31 was oxidized as
described for 35 (see below) and chromatographed (PE/CH2Cl2 ˆ 9/1).
M.p. 86.5 ± 87.58C (heptane); 1H NMR (CDCl3): no signal; 13C NMR
(CDCl3): d ˆ 123.66, 127.52 (q, J ˆ 2.4 Hz), 128.02 (q, J ˆ 313.4 Hz), 137.06
(q, J ˆ 0.8 Hz), 147.83 (br); anal. calcd for C7Cl3F3N2O4S (371.51): C 22.63,
H 0.00, N 7.54, S 8.63, Cl 28.63; found: C 22.60, H 0.00, N 7.63, S 8.84, Cl
28.41; MS (EI): 376 (3.1), 375 (2.2), 374 (35.0), 373 (7.9), 372 (100), 371 (8.0),
2,4,6-Trichloro-3,5-dinitroiodobenzene (26): A solution of the aniline 25
(3.000 g, 10.47 mmol) in hot glacial acetic acid (20 mL) was rapidly cooled
while being stirred. The resulting fine suspension was added to nitro-
sylsulfuric acid [from 787 mg (11.4 mmol) NaNO2 and conc sulfuric acid
(25 mL)] at 58C. the mixture was stirred for 30 min at RT, and was then
added to KI/I2 (4 g each) in water (50 mL). The mixture was heated to
boiling, then cooled, the excess iodine removed by bisulfite, the precipitate
filtered and washed with water, satd bicarbonate, then water. It was then
washed through with diethyl ether, and the solution dried (Na2SO4).
Filtration over silica gel (CH2Cl2/PE ˆ 1/1) gave colorless crystals (3.592 g,
9.040 mmol, 86.3%). An analytical sample had m.p. 192.5 ± 194.58C (from
‡
370 (85.8): M , intensities in accordance with proposed composition. The
oxidation can be stopped at the nitroso stage, if the colorless precipitate
that forms at 108C is collected after 150 min (yield, 62%); pure by TLC,
m.p. 139 ± 1408C. 2,4,6-Trichloro-3-nitro-5-nitrosotrifluoromethylthioben-
zene gives no signal in 1H NMR, and a complex 13C NMR (probably due to
association). However, the MS (EI) is unequivocal: 360 (0.5), 359 (0.3), 358
(5.3), 357 (1.1), 356 (14.1), 355 (1.1), 354 (13.7). Both the nitroso compound
and 32 only give water-soluble material on attempted oxidation with CrO3
in TFAA/H2SO4 mixtures.
3,5-Dinitrotrifluoromethylsulfonylbenzene (33): By CrO3 (2.5 g, 25 mmol)
oxidation of the above thioether (2.326 g, 8.674 mmol) in TFAA/sulfuric
acid mixture at RT, and purified by chromatography (silica gel, CH2Cl2).
Colorless powder, m.p. 93 ± 958C (2.216 g, 7.382 mmol, 85.1%). 1H NMR
([D6]acetone): d ˆ 9.20 (dq, J ˆ 2.0 and 0.5 Hz, 2H), 9.49 (t, J ˆ 2.0 Hz,
1H); 13C NMR ([D6]acetone): d ˆ 120.44 (q, J ˆ 325.4 Hz), 127.38, 131.49
(q, J ˆ 0.8 Hz), 134.99 (q, J ꢂ 1.4 Hz), 150.75. Compound 33 has been
described in the literature, but no characterization was given.[57]
1
n-heptane/toluene). H NMR (CDCl3): no signal; 13C NMR (CDCl3): d ˆ
106.53, 119.53, 134.31, 146.10 (br); anal. calcd for C6Cl3IN2O4 (399.51): C
18.14, H 0.00, N 7.05, Cl 26.77, I 31.94; found: C 18.27, H 0.00, N 7.10, Cl
26.96, I 32.20.
Reaction of 26 with CuSCF3 in DMF did not give the substitution product
32.
3,5-Dinitroiodobenzene (27):[53] M.p. 100.5 ± 1038C, then recrystallizing and
melting at 104.5 ± 105.58C (first from EtOH, then from n-heptane/toluene),
lit. m.p.[53] 99 ± 1008C (EtOH). 1H NMR (CDCl3): d ˆ 8.86 (d, J ˆ 2.0 Hz,
2H), 8.98 (t, J ˆ 2.0 Hz, 1H); 13C NMR (CDCl3): d ˆ 93.33, 118.26, 137.59,
148.43.
3-Nitro-5-trifluoromethylsulfonylaniline (34): To a boiling solution of the
dinitrosulfone 33 (2.091 g, 6.966 mmol) in methanol (20 mL), a solution of
NaHS (16 mL of a 0.70m solution,[31] 11 mmol; pH 10 ± 11) was added in
one portion. The deep red color (Meisenheimer complex) changes within
30 min further reflux to a deep brown. The methanol is removed on a rotary
evaporator, the mixture filtered, and the residue washed thoroughly with
diethyl ether. The filtrate is washed with water, satd brine, and dried.
Chromatography (silica, CH2Cl2) and crystallization from toluene/n-
heptane gives orange-brown needles, m.p. 165 ± 1728C (395 mg, 1.46 mmol,
21.0%). For analysis, a sample was rechromatographed (silica, Et2O) and
crystallized again from n-heptane/toluene to give yellow needles, m.p. 171 ±
1728C. 1H NMR ([D6]acetone): 7.68 (ddq, J ˆ 1.7, 2.3 and 0.6 Hz), 7.87 (ddq,
J ˆ 1.6, 2.1 and 0.5 Hz), 8.00 (pseudo-t, J ˆ 2.2 Hz); 13C NMR ([D6]ace-
tone): d ˆ 111.88 (q, J ˆ 0.8 Hz), 115.95, 120.12 (q, J ˆ 0.7 Hz), 120.61 (q,
J ˆ 325.3 Hz), 133.74 (q, J ˆ 1.5 Hz), 150.87, 152.69; anal. calcd for
C7H5F3N2O4S (270.19): C 31.12, H 1.87, N 10.37, S 11.87; found: C 31.22,
H 1.87, N 10.39, S 11.97.
3-iodo-5-nitroaniline (28): Obtained from 27 according to ref. [54], but with
NaHS solution as described for 34, 50% yield after recrystallization from
EtOH, still impure, m.p. 131 ± 1378C (ref. [54]: yield 35% with m.p. 135.5 ±
1398C and 140 ± 1418C when pure). 1H NMR ([D6]acetone]: d ˆ 5.53 (brs,
2H), 7.41 (dd, J ˆ 2.1 and 3.2 Hz, 1H), 7.48 (t, J ˆ 3.1 Hz, 1H), 7.65 (dd, J ˆ
2.0 and 3.1 Hz, 1H). It did not give 3-nitro-5-trifluoromethylthioaniline on
reaction with CuSCF3.
3,5-Dinitrotrifluoromethylthiobenzene (29): From 27[53] by reaction with
[55]
CuSCF3 in DMF (cf. ref. [56]). Purified by chromatography (silica gel,
CCl4/Et2O ˆ 98/2); yellowish oil, yield, 94%.
3-Nitro-5-trifluoromethylthioaniline (30): A solution of NaHS in water/
MeOH (237 mL of a 0.464m solution, 110 mmol) was added in a thin
stream to
a refluxing solution of the dinitrosulfide 29 (17.855 g,
66.581 mmol) in MeOH (200 mL). The solution turns orange, then brown.
After the addition, reflux is continued for 30 min. After cooling, the
solution is extracted with CH2Cl2 (13.97 g) and chromatographed (silica,
2,4,6-Trichloro-3-nitro-5-trifluoromethylsulfonylaniline (35): To a stirred
and cooled solution of the aniline 34 (500 mg, 1.85 mmol) in acetic acid
(8 mL) and conc HCl (5 mL), 30% H2O2 (3 mL) was added. A precipitate
36
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999
0947-6539/99/0501-0036 $ 17.50+.50/0
Chem. Eur. J. 1999, 5, No. 1

Crystal Engineering
29±38
formed. After 3.5 h, the mixture was heated briefly to reflux. After standing
overnight, it was diluted with water and extracted with dichloromethane
(3 Â ). The extracts were washed with bicarbonate solution and dried.
Chromatography (silica, Et2O/CH2Cl2/PE ˆ 1/1/2) and crystallization from
n-heptane gave faintly yellow prisms; m.p. 137 ± 138.58C (525 mg,
1.41 mmol, 76.0%). 1H NMR ([D6]acetone): 6.73 (brs); 13C NMR
([D6]acetone): d ˆ 114.44, 117.36, 120.71 (q, J ˆ 327.4 Hz), 123.90, 128.10
(q, J ˆ 2.3 Hz), 145.94, 150.36 (br); anal. calcd for C7H2Cl3F3N2O4S
(373.52): C 22.51, H 0.54, N 7.50, S 8.58, Cl 28.47; found: C 22.59, H 0.67,
N 7.53, S 8.34, Cl 28.40; MS (EI): 378 (2.3), 377 (1.4), 376 (16.2), 375 (4.6),
[10] D. J. Williams, Angew. Chem. 1984, 96, 637 ± 651; Angew. Chem. Int.
Ed. Engl. 1984, 23, 690 ± 703.
[11] D. M. Burland, R. D. Miller, C. A. Walsh, Chem. Rev. 1994, 94, 31 ± 75.
Ã
[12] C. Bosshard, K. Sutter, P. Pretre, J. Hulliger, M. Flörsheimer, P. Kaatz,
P. Günther, Organic Nonlinear Optical Materials, Gordon and Breach,
Basel, 1995.
[13] Nonlinear Optics of Organic Molecules and Polymers (Eds.: H. S.
Nalwa, S. Miyata), CRC, Boca Raton, 1997.
[14] J. J. Wolff, R. Wortmann, J. Prakt. Chem./Chem. Ztg. 1998, 340, 99 ±
111.
[15] I. G. Voigt-Martin, G. Li, A. Yakimanski, G. Schulz, J. J. Wolff, J. Am.
Chem. Soc. 1996, 118, 12830 ± 12381.
‡
374 (48.3), 373 (4.7), 372 (49.4): M , intensities in accordance with proposed
formula.
[16] I. G. Voigt-Martin, L. Gao, A. Yakimanski, H. Gross, J. J. Wolff, J.
Phys. Chem. A 1997, 101, 7265 ± 7276.
2,4,6-Trichloro-3,5-dinitro-5-trifluoromethylsulfonylbenzene (36): Cau-
tion! See warning above for the handling of conc peroxide! ꢂ85%
hydrogen peroxide[58] (3 mL) was added in portions to trifluoroacetic
anhydride (20 mL) with stirring at 08C. After several min, the biphasic
mixture became homogeneous in an exothermic (Caution!) reaction. This
reaction has to be awaited before the next portion is added. The aniline 35
(113 mg, 0.303 mmol) was added at 08C and slowly dissolved. A blue-
greenish color was observed (probably nitroso compound) and a precip-
itate formed. The ice bath was removed after 4 h, and the mixture was
stirred at RT overnight, whereupon the precipitate slowly dissolved to give
a colorless solution. This was cooled again, and ice was added, whereupon a
colorless precipitate was formed. After 30 min stirring, the almost pure 36
was filtered, washed with water, and dried (130 mg, 0.322 mmol, quant);
m.p. 142 ± 144.58C. For analysis, it was purified by sublimation (1208C,
0.1 Torr); almost no residue was observed, m.p. 146 ± 1478C. 1H NMR
(CD3CN): no signal; 13C NMR (CD3CN): d ˆ 120.56 (q, J ˆ 326.9 Hz),
128.34, 123.90, 131.27 (q, J ˆ 2.1 Hz), 134.62, 150.56 (br); anal. calcd for
C7Cl3F3N2O6 (403.51): C 20.84, H 0.00, N 6.94, S 7.95, Cl 26.36; found: C
20.93, H 0.00, N 7.04, S 8.23, Cl 26.53; MS (EI): 408 (0.5), 407 (0.3), 406
[17] C. B. Aakeröy, K. R. Seddon, Chem. Soc. Rev. 1993, 22, 397 ± 407.
[18] R. Taylor, O. Kennard, Acc. Chem. Res. 1984, 17, 320 ± 326.
[19] A. Vedani, J. D. Dunitz, J. Am. Chem. Soc. 1985, 107, 7653 ± 7658.
[20] J. C. McDonald, G. M. Whitesides, Chem. Rev. 1994, 94, 2383 ± 2420.
[21] J. Bernstein, R. E. Davis, L. Shimoni, N.-L. Chang, Angew. Chem.
1995, 107, 1689 ± 1708; Angew. Chem. Int. Ed. Engl. 1995, 34, 1545 ±
1564.
[22] P. Brunet, M. Simard, J. D. Wuest, J. Am. Chem. Soc. 1997, 119, 2737 ±
2738.
[23] V. N. Boiko, I. V. Gogoman, G. M. Shchupak, L. M. Yagupolꢁskii, J.
Org. Chem. USSR Engl. Transl. 1987, 23, 544 ± 548 and 2282 ±
2286.
[24] J. Lindley, Tetrahedron 1984, 40, 1433 ± 1456.
[25] D. L. Mattern, J. Org. Chem. 1984, 49, 3051 ± 3053.
[26] J. J. Wolff, D. Längle, D. Hillenbrand, R. Wortmann, R. Matschiner, C.
Glania, P. Krämer, Adv. Mater. 1997, 9, 138 ± 143.
[27] B. Flürscheim, J. Prakt. Chem. [2] 1905, 71, 497 ± 539.
[28] G. A. Artamkina, M. P. Egorov, I. P. Beletskaya, Chem. Rev. 1982, 82,
427 ± 459.
‡
(3.9), 405 (1.0), 404 (11.0), 403 (1.0), 402 (10.8): M , intensities in
accordance with proposed formula.
[29] F. Terrier, Chem. Rev. 1982, 82, 77 ± 152.
1,3,5-Triamino-2,4-dinitro-6-trifluoromethylsulfonylaniline (6): Ammonia
was bubbled through a solution of the trichloro compound 36 (205 mg,
0.508 mmol) in dry CH2Cl2. The mixture soon turned yellow, and a cloudy
precipitate was formed. The mixture was kept in a tightly closed container
for 1.5 d, during which time the triamine 6 crystallized as yellow platelets,
which were removed by filtration and washed with water to remove
ammonium chloride (170 mg, 0.492 mmol, 96.9%), m.p. 273 ± 2748C.
Recrystallization from either acetonitrile or nitromethane did not alter
the observed m.p.. 1H NMR ([D6]DMSO): 9.19 (brs); 13C NMR
([D6]DMSO): d ˆ 78.38 (q, J ˆ 2.1 Hz), 113.77, 120.40 (q, J ˆ 328.5 Hz),
149.71, 151.86; anal. calcd for C7H6F3N5O6S (345.22): C 24.35, H 1.75, N
20.29, S 9.29; found: C 24.33, H 1.78, N 20.22, S 9.28.
[30] F. Terrier, Nucleophilic Aromatic Displacement: The Influence of the
Nitro Group, VCH, New York, 1991.
[31] H. H. Hodgson, E. R. Ward, J. Chem. Soc. 1949, 1316 ± 1317.
[32] J. J. Wolff, H. Irngartinger, F. Gredel, I. Bolocan, Chem. Ber. 1993, 126,
2127 ± 2131.
[33] J. J. Wolff, F. Gredel, D. Hillenbrand, H. Irngartinger, Liebigs Ann.
1996, 1175 ± 1182; K. K. Baldridge, J. S. Siegel, J. Am. Chem. Soc.
1993, 115, 10782 ± 10785.
[34] F. H. Allen, O. Kennard, D. G. Watson, J. Chem. Soc. Perkin Trans 2
1987, 1 ± 19.
[35] J. Shorter, in The Chemistry of Sulphones and Sulphoxides (Eds.: S.
Patai, Z. Rappoport, C. Stirling), Wiley, London, 1988, p. 483 ± 539.
[36] J. H. Williams, Acc. Chem. Res. 1993, 26, 593 ± 598.
[37] G. R. Desiraju, A. Gavezzotti, Acta Crystallogr. Sect. B 1989, 45, 473 ±
482.
[38] C. A. Hunter, Chem. Soc. Rev. 1994, 23, 101 ± 109; H. Adams, F. J.
Carver, C. A. Hunter, J. C. Morales, E. M. Seward, Angew. Chem.
1996, 108, 1542 ± 1544; Angew. Chem. Int. Ed. Engl. 1996, 35, 1628 ±
1631.
[39] T. Dahl, Acta Chem. Scand. 1994, 48, 95 ± 106.
[40] T. Dahl, Acta Crystallogr. Sect. B 1990, 46, 283 ± 288.
[41] R. Laatikainen, J. Ratilainen, R. Sebastian, H. Santa, J. Am. Chem.
Soc. 1995, 117, 11006 ± 11010.
Acknowledgements
We would like to thank the Deutsche Forschungsgemeinschaft (Wo495/1-5
and Heisenberg Fellowship), the Fonds der Chemischen Industrie, and
Solvay GmbH (gift of trifluoroacetic anhydride) for support of this work.
[42] F. Cozzi, F. Ponzini, R. Annunziata, M. Cinquini, J. S. Siegel, Angew.
Chem. 1995, 107, 1092 ± 1094; Angew. Chem. Int. Ed. Engl. 1995, 34,
1019 ± 1020.
[43] G. W. Coates, A. R. Dunn, L. M. Henling, D. A. Dougherty, R. H.
Grubbs, Angew. Chem. 1997, 109, 290 ± 293; Angew. Chem. Int. Ed.
Engl. 1997, 36; G. W. Coates, A. R. Dunn, L. M. Henling, J. W. Ziller,
E. B. Lobkovsky, R. H. Grubbs, J. Am. Chem. Soc. 1998, 120, 3641 ±
3649; S. M. Ngola, D. A. Dougherty, J. Org. Chem. 1998, 63, 4566 ±
4567 and references therein.
[1] F. H. Allen, O. Kennard, D. G. Watson, in Structure Correlation, Vol. 1
(Eds.: H.-B. Bürgi, J. D. Dunitz), VCH, Weinheim, 1994, p. 71 ± 110.
[2] A. Gavezzotti, G. Filippini, J. Am. Chem. Soc. 1996, 118, 7153 ± 7157.
[3] J. J. Wolff, Angew. Chem. 1996, 108, 2339 ± 2441; Angew. Chem. Int.
Ed. Engl. 1996, 35, 2195 ± 2197.
[4] G. R. Desiraju, Angew. Chem. 1995, 107, 2541 ± 2558; Angew. Chem.
Int. Ed. Engl. 1995, 34, 2328 ± 2361.
[5] G. R. Desiraju, Crystal EngineeringÐThe Design of Organic Solids;
Materials Science Monographs, Vol. 54, Elsevier, Amsterdam, 1989.
[6] M. L. Greer, B. J. McGee, R. D. Rogers, S. C. Blackstock, Angew.
Chem. 1997, 109, 1973 ± 1976; Angew. Chem. Int. Ed. Engl. 1997, 36,
1864 ± 1866.
[44] A. Bondi, J. Phys. Chem. 1964, 68, 441 ± 451.
Â
Ä
[45] C. Fernandez-Castano, C. Foces-Foces, F. H. Cano, R. M. Claramunt,
Â
C. Escolatico, A. Fruchier, J. Elguero, New J. Chem. 1997, 21, 195 ± 213.
[46] J. D. Dunitz, R. Taylor, Chem. Eur. J. 1997, 3, 89 ± 98.
[47] J. A. K. Howard, V. J. Hoy, D. OꢁHagan, G. T. Smith, Tetrahedron
1996, 52, 12613 ± 12622.
[7] A. J. C. Wilson, Acta Crystallogr. Sect. A 1990, 46, 742 ± 754.
[8] A. J. C. Wilson, Acta Crystallogr. Sect. A 1993, 49, 210 ± 212.
[9] C. P. Brock, J. D. Dunitz, Chem. Mater. 1994, 6, 1118 ± 1127.
Chem. Eur. J. 1999, 5, No. 1
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999
0947-6539/99/0501-0037 $ 17.50+.50/0
37

J. J. Wolff et al.
FULL PAPER
Â
[48] J. Barluenga, J. M. Gonzalez, M. A. García-Martín, P. J. Campos,
[55] J. H. Clark, C. W. Jones, A. P. Kybett, M. A. McClinton, J. M. Miller,
D. Bishop, R. J. Blade, J. Fluorine Chem. 1990, 49, 249 ± 253.
[56] N. V. Kondratenko, A. A. Kolomeytsev, V. I. Popov, L. M. Yagupol-
skii, Synthesis 1985, 667 ± 669.
[57] V. N. Boiko, G. M. Shchupak, J. Org. Chem. USSR Engl. Transl. 1977,
13, 958 ± 961.
Tetrahedron Lett. 1993, 34, 3893 ± 3896.
[49] R. A. Adams, A. Ferretti, J. Am. Chem. Soc. 1959, 81, 4927 ± 4931.
[50] L. M. Yagupolskii, N. V. Kondratenko, V. P. Sambur, Synthesis 1975,
721 ± 723.
[51] J. J. Blanksma, Rec. Trav. Chim. Pays-Bas 1902, 21, 254 ± 268.
[52] W. C. Lothrop, G. R. Handrick, R. M. Hainer, J. Am. Chem. Soc. 1951,
73, 3581 ± 3584.
[58] M. Schmeisser, F. Huber, in Handbuch der Präparativen Anorgani-
schen Chemie (Ed.: G. Brauer), Enke, Stuttgart 1975, p. 156.
[53] J. Arotsky, R. Butler, A. C. Darby, J. Chem. Soc. C 1970, 1480 ± 1485.
[54] T. L. Fletcher, M. J. Namkung, W. H. Wetzel, H.-L. Pan, J. Org. Chem.
1960, 25, 1342 ± 1348.
Received: April 22, 1998 [F1112]
38
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999
0947-6539/99/0501-0038 $ 17.50+.50/0
Chem. Eur. J. 1999, 5, No. 1