Synthesis, structure and stability of E/Z-isomers of novel conjugated enamines prepared from 9-arylmethyl- or 9-arylpropenyl-9H-carbazole with arylmethyleneanilines

Article abstract of DOI:10.1039/a604969i

Active methylene groups, substituted by 9H-carbazol-9-yl (Carb) and aryl or 2-phenylethenyl groups, condense with arylmethyleneanilines in DMF at 75°C in the presence of Bu'OK to form the corresponding enamines [(Carb)(Ar¹)C=C(Ar²)H] and dienamines [(Ar³)HC=C(Carb)CH=CHPh] in almost quantitative yield. The ¹H and ¹³C NMR spectra for the enamine 1′Z-isomers [16 (Ar¹ = Ar² = 4-fluorophenyl), 17 (Ar¹ = 4-fluorophenyl, Ar² = 4-tert-butoxyphenyl), 19 (Ar¹ = Ar² = 4-tert-butoxyphenyl)], dienamine 1′Z-isomers [14a (Ar³ = 1-naphthyl), 14b (Ar³ = 4-methoxyphenyl), 14c (Ar³ = Ph)] and 1′E-isomers [15a (Ar³ = 1-naphthyl), 15c (Ar³ = Ph)] and precursors are assigned with the aid of COSY, HMBC, and HMQC techniques. The geometrical isomerism of the different dienamines 14-15 is established by ³J_C-H NMR couplings and that of enamine 12 by a difference NOE experiment. X-Ray crystal structures for 16, 14a and 15c corroborate the isomerism results deduced by NMR studies. Dienamines 14a and 15a hydrolyse to the ketone under relatively strong acid conditions [AcOH-HCl-H₂O (18:1:1 v/v)] under reflux over 7 h. There is an equilibrium between 14c and 15c in 1,2,4-trichlorobenzene at 180 ± 1°C with K = 15c/14c = 0.77 as estimated from the kinetic rate profiles from HPLC data acquired over 4 days. However, under the same conditions, 14a and 15a undergo an equilibration concurrent with a reaction (monitored over 9 days) giving apparently a carbazolyl-substituted phenylphenanthrene. In contrast, enamine 16 is thermally stable with no detectable change after boiling for 4 days in 1,2,4-trichlorobenzene.

Full text of DOI:10.1039/a604969i

View Article Online / Journal Homepage / Table of Contents for this issue
Synthesis, structure and stability of E/Z-isomers of novel conjugated
enamines prepared from 9-arylmethyl- or 9-arylpropenyl-9H-
carbazole with arylmethyleneanilines
Martino Paventi and Allan S. Hay
Department of Chemistry, McGill University, 801 Sherbrooke St. W., Montreal, Quebec,
Canada H3A 2K6
Active methylene groups, substituted by 9H -carbazol-9-yl (Carb) and aryl or 2-phenylethenyl groups,
condense with arylmethyleneanilines in D M F at 75 ЊC in the presence of ButOK to form the
corresponding enamines [(Carb)(Ar1)C=C(Ar2)H ] and dienamines [(Ar3)H C᎐C(Carb)CH ᎐CH Ph] in
᎐
᎐
almost quantitative yield. The 1H and 13C N M R spectra for the enamine 1ЈZ -isomers [16 (Ar1 = Ar2 = 4-
fluorophenyl), 17 (Ar1 = 4-fluorophenyl, Ar2 = 4-tert-butoxyphenyl), 19 (Ar1 = Ar2 = 4-tert-butoxyphenyl)],
dienamine 1ЈZ -isomers [14a (Ar3 = 1-naphthyl), 14b (Ar3 = 4-methoxyphenyl), 14c (Ar3 = Ph)] and
1ЈE-isomers [15a (Ar3 = 1-naphthyl), 15c (Ar3 = Ph)] and precursors are assigned with the aid of COSY,
H M BC, and H M QC techniques. The geometrical isomerism of the different dienamines 14–15 is established
by 3JC-H N M R couplings and that of enamine 12 by a difference N OE experiment. X-Ray crystal
structures for 16, 14a and 15c corroborate the isomerism results deduced by N M R studies. D ienamines
14a and 15a hydrolyse to the ketone under relatively strong acid conditions [AcOH –H Cl–H 2O (18 :1 :1 v/v)]
under reflux over 7 h. There is an equilibrium between 14c and 15c in 1,2,4-trichlorobenzene at 180 ± 1 ЊC
with K = 15c/14c = 0.77 as estimated from the kinetic rate profiles from H PLC data acquired over 4 days.
H owever, under the same conditions, 14a and 15a undergo an equilibration concurrent with a reaction
(monitored over 9 days) giving apparently a carbazolyl-substituted phenylphenanthrene. In contrast,
enamine 16 is thermally stable with no detectable change after boiling for 4 days in 1,2,4-trichlorobenzene.
Introduction
To the extensive literature on enamines,1–3 we have recently con-
tributed a novel, diverse method for the preparation of the
aromatic-substituted enamines,4 illustrated in Scheme 1, whose
Scheme 1
ation when they are converted to cation radicals during the
process of hole transport.8
synthesis is discussed below. In addition to the synthesis, this
report also presents NMR, crystallographic and other physical
data on the E/Z isomerism of enamines where the more con-
ventional amino group (R1NR2–) is replaced by a 9H-carbazol-
9-yl moiety and the R group is either a substituted phenyl or
(E)-cinnamoyl group. These carbazole derivatives are more
amenable to isomerism studies because the enamines which
form readily are more crystalline, are easier to separate, dis-
criminate easily between the E- and Z-isomers, and hydrolyse
with difficulty even when subjected to harsh conditions.
When incorporated in a polymer matrix, doped aromatic
enamines 1–3 are useful electrophotographic photoconductors.5
In xerography, molecules with high mobility, which is a measure
of charge (or hole) transport capability when a potential is
applied across a matrix containing these molecules,6 are util-
ized. Compared to the analogous triaryl amines the perform-
ance measured by the mobility of the charge transport
materials7 is higher for compounds containing at least one
enamine functionality (i.e. 5 > 4, 7 > 6, 9 > 8). In conjugated
systems this property should be enhanced by the electron-rich
nature of these species (cf. mobility of 6 > 4 or 6 > 8) and
because rigid molecules may adopt a more favourable conform-
In xerographic processes,9,10 materials containing the carba-
zole moiety [e.g. poly(vinylcarbazole)] have been utilized 10,11
due to their photoconductivity, a property which was recog-
nized some time ago.12 The carbazolyl-containing enamines
illustrated in Schemes 2 and 3 show a greater resistance to
hydrolysis compared to several reported diarylamino-con-
taining enamines4 (Scheme 1) which are probably of similar
hydrolytic stability to 1–8. On the basis of this hydrolytic stabil-
ity, which is a property sought for prolonged usage,13 and their
ease of preparation and purification and good potential as hole
transport materials, enamines containing the carbazolyl group
should be attractive for use as photoreceptors and in similar
applications.
Results and discussion
Synthesis
Enamines are commonly derived from the condensation of car-
bonyl compounds with secondary amines with the removal of
water.1b,14 This type of synthesis results in enamines which are
easily hydrolysed by stirring in aqueous acid. The enamines
J. Chem. Soc., Perkin Trans. 1, 1997
1059

View Article Online
NPh
+
Ar2
H
N
Ar1
10a Ar1 = Ph
11a Ar2 = Ph
b Ar1 = 4-FC6H4
b Ar2 = 4-FC6H4
c Ar2 = 4-ButOC6H4
-PhNH2
i
100% conversion
+
N
N
1′
1′
Ar2
H
Ar1
Ar1
2′
2′
H
Ar2
(1′Z)-enamine
(1′E)-enamine
1′Z/1′E>19
Scheme
2
Reagents and conditions: i, DMF (50 cm3), ButOK
(1 equiv.), 75 ЊC, <3 min
Values read from graphical data at E = 12 V µmϪ1, T = 295 K and a
charge-transport material to polymer binder mix of 1.76 mmol gϪ1 (ref.
8)
outlined below are not accessible by a similar condensation,
probably because of the poor nucleophilicity and large steric
requirements of carbazole.
Our synthesis began with an arylmethylcarbazole, easily
obtained from the alkylation of carbazole with the arylmethyl
chloride, followed by condensation with arylmethyleneanilines
in the presence of a strong base (Scheme 2). Herein we also
report some novel cross-conjugated acyclic aromatic enamines
originally intended as dienes in the Diels–Alder reaction en
route to novel polyaromatic imides (Scheme 3). These syntheses,
illustrated in Schemes 2 and 3, are a variant of the anil reaction
described in the numerous papers of Siegrist.15 Siegrist has
reported one example of enamine formation under similar
strong basic conditions16 but he did not realize the general
applicability of this reaction. High yields of enamine17,18 can
also be obtained by starting with the benzylated amines, instead
of the aryl benzyl ethers, and reacting with the imine in the
presence of ButOK (3 equiv. as also used by Siegrist 15) as the
strong base in DMF at 75 ЊC.
Scheme 3 Reagents and conditions: i, DMF (50 cm3), ButOK (1 equiv.),
0–75 ЊC, <3 min
condensation step leading to 16. We have employed this facile
displacement to prepare polymers from 16 by reaction with
bisphenols, to be reported elsewhere. Earlier we reported a
similar sequence to that shown in Scheme 4, for the preparation
of bis(4-fluorophenyl)acetylene and the displacement of fluor-
ines.3 By way of comparison, 16 is at least three times less
reactive in the fluorine displacement reaction than bis(4-
fluorophenyl)acetylene. Only the Z-isomers (vide infra) of 16–
19 are shown in Scheme 4 since these are by far the predomin-
ant isomers (Z-isomer >95% estimated by HPLC area of
peaks) and the only ones isolated. By working up the reaction
quickly by pouring the mixture into water, the amount of 16 is
maximized and may be separated by column chromatography
from minor concentrations of species 17–19. Enamine 19
can also be easily separated by letting the reaction run to
The reaction shown in Scheme 2 has been employed to
prepare several enamines in good yields when there are no
labile substituents present which would react in the strongly
basic medium.4 When fluorine atoms are present, the reaction
proceeds as illustrated in Scheme 4. The displacement of the
fluorines takes place cleanly but much more slowly than the first
1060
J. Chem. Soc., Perkin Trans. 1, 1997

View Article Online
The mass spectra of 14a and 15a both show molecular ions at
m/z 421, and very similar fragmentation patterns. UV spectro-
scopic data all show minor differences attributable to the differ-
ent degree of twisting of the butadiene moiety. Noteworthy are
the infrared bands, probably ᎐CH vibrations19b,c or C–N
᎐
stretches,19d at 1390 cmϪ1 in the spectra of the 1ЈZ-isomers and
1375 cmϪ1 in the spectra of the 1ЈE-isomers.
The fact that only two products 14 and 15 arise from the
reaction suggests that there is no isomerization of the inter-
mediate anion 20 to 21 (Scheme 6), which would result in the
Scheme 6
formation of additional isomeric products, although the
isomerization of phenylmethylamino or dimethylamino ana-
logues of 12 to their respective enamines in a strongly basic
medium is the preferred process.20
Slow hydrolysis of 14 and 15 gives the identical α,β-
unsaturated ketone 22 (NMR, MS, HRMS) (Scheme 7). The
Scheme 4 Reagents and conditions: i, DMF, ButOK (3 equiv.),
75 ЊC, <1 min
completion over a 12 h period at 75 ЊC and/or by having a
larger excess of ButOK present to expedite the completion of
the reaction. However, 17 and 18 cannot be easily separated
from the complex mixture of products from the reaction of
10b and 11b because 17 and 18 have very similar properties in
the chromatographic separation. Thus, we prepared 17 by the
reaction of 10b and 11c by the reaction sequence shown in
Scheme 5.
Scheme 7
ketone 22 could not be completely isolated from the contamin-
ating co-product carbazole, even after two attempts at column
separations. The H NMR spectrum of the mixture of 22 and
carbazole also corroborates that the conformation about the
carbon–carbon double bond in 22 is trans (J = 16 Hz) and
is thus unaffected by the sequence of steps: tertiary amine →
enamine → hydrolysis product.
1
O
O
Attempted Diels–Alder reactions of 14a and 15a with maleic
anhydride failed. Neither isomer reacted, even after 24 h in
o-xylene when heated under reflux. In this reaction some of the
more polar species were converted to the less polar compound,
presumably by acid catalysed rearrangement, thermal
rearrangement or both. This equilibration was also observed in
the NMR experiments in CDCl3 at room temperature.
1
2
Amberlyst 15
PhNH2
11c
PhH
64% conversion
3
4
OH
OBut
ButOK (3 equiv.)
10b
+
11c
17
DMF, 75 °C
>1 min
Thermal behaviour
Following these observations, we decided to study the thermal
behaviour of 14a,c and 15a,c by HPLC. N,N-Dimethyl-
acetamide (DMAc) as solvent was our first choice because it
has a relatively high boiling point and is the one which we have
used successfully in polymerizations and making derivatives of
difluoro enamines such as 16. For these synthetic scale reactions
no hydrolysis was detected, whereas at 160 ЊC the enamines 14
and 15 completely decomposed to give carbazole and other
unidentified products, presumably by hydrolysis due to the
small amount of water contained in DMAc.
However, in the absence of water the behaviour of 14a and
15a is quite different. In 1,2,4-trichlorobenzene, 14a and 15a
partly equilibrate. In two independent experiments, on a
micromolar scale, one beginning with 14a and the other with
15a in 1,2,4-trichlorobenzene at 180 ЊC, each first equilibrates
with its isomer. With time another species appears (tR = 10.2
Scheme 5
Structural evidence for 14 and 15
During the course of the synthesis of 14 and 15 we noted a
large disparity in their retention times (tR ) of up to 3 min. In
contrast, for enamines 16–19, ∆tR is ca. 0.3 min. Only the more
polar Z-isomer, 95% of the product by HPLC, was isolated and
characterized.
Alkylation of carbazole with trans-cinnamoyl chloride leads
to the trans product 12 (J2Ј,3Ј = 16.1 Hz).19a Compound 12, in
turn, condenses with 13a to produce 14 and 15. The isomeric
ratios for this reaction were 73:27, 85:15 and 90:10 for
14a :15a, 14b:15b and 14c:15c, respectively, at 75 ЊC (HPLC)
and do not vary when the reaction temperature is changed from
0 to 50 ЊC.
J. Chem. Soc., Perkin Trans. 1, 1997
1061

View Article Online
Table 1 Coupling constants (nJC,H ) for 14 and 15 in CDCl3 from
coupled/decoupled 13C NMRa spectra
min) and eventually becomes the predominant species in the
mixture over a period of nine days. This experiment was
repeated on a 0.5 g scale in order to identify the product. Thus,
14a was heated in 1,2,4-trichlorobenzene under reflux for 4
days. The HPLC chromatograms at various times indicated an
88% conversion after 4 days. The workup of the mixture led to
the isolation of the carbazolyl-substituted phenylphenanthrene
23 whose structure has been assigned pending complete charac-
terization. It presumably arises from an intramolecular Diels–
Alder reaction and subsequent air oxidation. The proposed
sequence of steps is illustrated in Scheme 8.
1JC3Ј,H 3Ј/Hz
2JC3Ј,H 4Ј/Hz
3JC3Ј,H 1Ј/Hz
14a
14b
14cb
15a
15cb
153.1
152.0
155.2
154.2
153.8
2
6.6
4.4
6.0
7.7
8.2
—
1.8
2.2
—
a
67.9 MHz. b 125.7 MHz.
3
Compound 14b shows JC3Ј,H 1Ј = 4.4 Hz, a value representa-
tive of a cis arrangement of the carbon and hydrogen atoms at
positions 3Ј and 1Ј, respectively.
3
For 14c, JC3Ј,H 1Ј = 5.9 Hz, a coupling constant which is
towards the high end of a cis arrangement of the C3Ј and H1Ј
atoms. Thus the geometry for 14c is likely to be 1ЈZ . Its corres-
ponding isomer 15c shows 3JC3Ј,H 1Ј = 8.2 Hz, which is unmistak-
ably a coupling constant resulting from a 1ЈE arrangement. For
this molecule we were able to grow a single crystal from which
an X-ray crystal structure was obtained (see below), which con-
firms the postulated 1ЈE geometry.
For species 19, a difference NOE experiment was consistent
with the structure of the Z-isomer. Thus the fluorine displace-
ment reaction by tert-butoxide takes place without a detectable
change in isomerism since the starting difluoro enamine 16 has
the 1ЈZ geometry as shown by the X-ray structure (see below).
The large shielding effects24,25 (1.23 and 0.84 ppm) experi-
enced by H2Љ and H3Љ, respectively, for 14a relative to the cog-
nates in 15a, strongly suggest that these hydrogens in 14a must
be sited over the carbazole ring with little rotation possible for
the naphthalene moiety. For the pair 14c and 15c, a somewhat
smaller shielding of H2Љ and H3Љ (0.85 and 0.54 ppm, respect-
ively) for 14c relative to the cognates in 15c similarly suggests
that 14c has the 1ЈZ arrangement. The latter shielding values
are smaller, consistent with an averaging effect possible with
increased rotation of the C1Ј-phenyl ring in 14c.
Scheme 8 Reagents and conditions: i, 1,2,4-trichlorobenzene, 180 ЊC,
103 h; ii, air
In 1,2,4-trichlorobenzene at 180 ЊC, 14c and 15c behave as
equilibrium species, over 4 days, with a minor amount of
decomposition (3–5%) judging from a peak at tR = 2.27 min
which may be attributed to carbazole. Again carrying out two
independent experiments on a micromolar scale, in which 14c is
added to one test tube and 15c to another at 180 ± 1 ЊC, the
approach to equilibrium may be followed by taking aliquots at
different times and analysing them by HPLC (see Experimental
for details). From the assumed first order isomerization, the
integrated rate expressions21 give estimates of the forward (kf)
and reverse (kr) rate constants, with (kr ϩ kf) = 5.8 × 10Ϫ6 sϪ1
and an equilibrium mole fraction of 14c of 0.56. From these
values the equilibrium constant K (= 14c/14c) equals 0.77 and
Crystal structures
The single crystal of 14a from which the X-ray structure was
resolved was obtained by recrystallization from ethyl acetate
1
and is represented in the CACheTM † model in Fig. 1. The H
NMR spectrum of this recrystallized sample suggests that the
ratio of the dienamine molecules to ethyl acetate is 2:1, as does
the unit cell. The plane of the heterocycle forms an acute
dihedral angle with the plane defined by N–C2Ј–C1Ј of 64Њ. The
diene moiety deviates from coplanarity by ca. 15Њ. The plane of
the naphthyl ring system is twisted from the C1Љ–C1Ј–C2Ј plane
by ca. 39Њ. The plane of the phenyl group is twisted from the
C1ٞ–C4Ј–C3Ј by ca. 6Њ. However, the larger deviations from
coplanarity of the conjugated system for 14a gives support to
the UV interpretation for the reduced absorptivity for 14a com-
pared to 14b and 14c.
the rate constants kf and kr are 2.5 × 10Ϫ6 and 3.3 × 10Ϫ6 sϪ1
,
respectively. Thus, 14c and 15c are interconvertable isomers
with the 1ЈZ-isomer 14c being the thermodynamically preferred
isomer.
Two conformations A and B are adopted by the molecule
15c, displayed in the CAChe† model in Fig. 2. The plane of
the heterocycle is twisted by ca. 60Њ from the plane containing
C1Ј–C2Ј–N for conformations A and B. The diene moiety is not
coplanar by ca. 23Њ for A and ca. 25Њ for B. The phenyl ring at
C1Ј is twisted out of the plane defined by C1Љ–C1Ј–C2Ј by 13Њ
for A and 28Њ for B. The other phenyl ring at C4Ј is twisted from
the plane defined by C1Љ–C4Ј–C3Ј by 22Њ for A and 16Њ for B.
We observe that the crystal of 15c reflects a fluorescent blue
colour when 365 nm wavelength light is shone on it, whereas the
1ЈZ-isomer is transparent. This phenomenon may be used to
manually separate crystals of the 1ЈE-isomer in a mixture of
1ЈE- and 1ЈZ-isomers once this mixture, 85% enriched with the
1ЈE-isomer, is carefully crystallized.
NMR Spectral analyses of E/Z-isomers
1
For enamines 14–17 and 21, H, 13C, H-COSY, HMQC and
HMBC experiments were necessary to identify the resonances
of carbons and hydrogens in the ethane and butadiene moieties.
Thus we were able to identify the arrangement of H1Ј and C3Ј
3
in 14 and 15, through the JC3Ј,H1Ј coupling constants22,23 listed
in Table 1.
For the coupled 13C NMR spectrum of 14a, overlap of res-
onance lines renders the determination of the coupling con-
3
stant doubtful, with JC3Ј,H 1Ј = 6.6 Hz being towards the high
end of a cis arrangement of C3Ј and H1Ј. However, for this
molecule we were able to confirm structure 14a by X-ray crys-
tallography. Its isomer 15a shows 3JC3Ј,H 1Ј = 7.7 Hz, a value def-
initely within the region of a trans arrangement of C3Ј and H1Ј.
The geometry is thus E at position 1Ј.
† CAChe, ver. 3.6, from CAChe Scientific, Inc, 1994.
1062
J. Chem. Soc., Perkin Trans. 1, 1997

View Article Online
Fig. 3 ORTEP-type drawing from CAChe of the predominant (71%)
conformer of 16. The thermal ellipsoids are drawn with a probability
value of 0.50. The atom labels shown are consistent with numbering in
text.
Fig. 1 ORTEP-type drawing from CAChe of 14a without ethyl acetate
solvent showing. The thermal ellipsoids are drawn with a probability
value of 0.50. The atom labels shown are consistent with numbering in
text.
plane of the fluorophenyl ring bonded to C1Ј is not coplanar
with the double bond C1Ј–C2Ј and atom C1Љ with a dihedral
angle of 26Њ. The fluorophenyl moiety at C2Ј is also twisted
from the plane C1Љ–C2Ј–C1Ј by 23Њ. The fluorophenyl groups
are out of plane, i.e. C1Љ is out of the plane defined by C1Ј–C2Ј–
C1ٞ by 8Њ.
Conclusions
NMR Techniques and crystal data for some of the enamines
and dienamines convincingly define their geometry. A novel
reaction leads to the predominant formation of the 1ЈZ-isomer
regardless of greater steric constraints imposed by bulkier aryl
and carbazolylgroups. Thisisprobablya result of thegreater con-
jugation energy gained by the enamines through the aromatic
groups with little, if any, cross-conjugation with the heterocycle.
Hydrolysis, equilibration and rearrangements occur very slowly
under unusually stringent conditions. In retrospect the pre-
dominant 1ЈZ-isomers, which also probably form in the syn-
theses of the analogous benzotriazoles and benzimidazole
enamines,4 are endowed with the right geometry for the anti
elimination of the nitrogen heterocycle making possible the
formation of diarylacetylenes in the presence of strong base.3,4
Experimental
X-Ray structure determination‡
The crystals used for the X-ray study were grown by routine
recrystallization from acetonitrile, ethyl acetate and MeOH for
16, 14a and 15c respectively. The crystal data, parameters of
data collection and refinement results are given in Table 2. The
unit cell dimensions were determined by least-squares using 25
(for 16 and 14a) and 15 (for 15c) centred reflections using
graphite monochromated Cu-Kα radiation. Data were cor-
rected for Lorentz and polarization effects. A correction for
secondary extinction was applied (coefficient = 0.53319E-5 for
16, 0.888854E-5 for 14a, 0.148 for 15c). No correction was
made for absorption for 15c. The structures for 16, 14a and 15c
were solved by direct methods. The non-hydrogen atoms (for 16
and 14a) were refined anisotropically.
Fig. 2 ORTEP-type drawing from CAChe of dienamine 15c crystalliz-
ing into two conformations A and B (50:50). The thermal ellipsoids are
drawn with a probability value of 0.50. These are the relative positions
of the conformers in the unit cell. The atom labels shown are consistent
with numbering in text.
The crystal structure of compound 16 confirms its structure
as that of the 1ЈZ-isomer, as shown in the CAChe† model in
Fig. 3. The carbazolyl moiety is staggered by about 63Њ from the
plane containing the double bond and the nitrogen atom. This
also appears to be the preferred conformation in solution by
analogy to the dienamines. Thus, little conjugation exists with
the substituted stilbenyl moiety and the carbazolyl system. The
‡ Atomic coordinates, thermal parameters and bond lengths and angles
have been deposited at the Cambridge Crystallographic Data Centre
(CCDC). See Instructions for Authors, J. Chem. Soc., Perkin Trans. 1,
1997, Issue 1. Any request to the CCDC for this material should quote
the full literature citation and the reference number 207/92.
J. Chem. Soc., Perkin Trans. 1, 1997
1063

View Article Online
Table 2 Crystal data, data collection and refinement parameters
14aؒ1/2(CH3CO2Et)
15c
18
Formula
Mass
C34H27NO
465.59
C28H21N
371.48
C26H17F2N
381.42
Crystal system
Space group
Dimensions/mm
a/Å
b/Å
c/Å
β (Њ)
Monoclinic
P2/c (no. 13)
0.50 × 0.22 × 0.12
16.120(2)
8.928(2)
18.517(2)
106.471(9)
4
Orthorhombic
Pna21 (no. 33)
0.25 × 0.15 × 0.05
10.239(4)
23.066(3)
17.529(3)
90
8 (zЈ = 8)
4139.9(19)
1.192
Monoclinic
P21/c (no. 14)
0.37 × 0.125 × 0.45
15.808(3)
10.728(2)
11.650(2)
104.70(1)
4
Z
V/Å3
2556(1)
1.21
1911(1)
1.326
ρ (calc.)/g cmϪ3
Radiation
Cu-Kα
5.22
Cu-Kα
4.9
Cu-Kα
7.1
µ/cmϪ1
Method/Trans max
F(000)/e
psi 0.96 to 1.00
984
None
1568
psi 0.91 to 1.00
792
T/K
294
294
294
Diffractometer
AFC6S (Rigaku)
h, k, ± l
ω–2θ, 16
45–55
AFC6S (Rigaku)
h, k, l/ Ϫh, Ϫk, Ϫl
ω/2θ, 4
AFC6S (Rigaku)
h, k, ±l
ω–2θ, 16
55–60
Index ranges
Scan mode, speed/degrees minϪ1
2θ range (Њ)
40–50
Total data collected
Unique data
5154
4977
4630
2320
3093
2929
Observed data used
System used
1693 [I > 3σ(I)]
TEXSAN a
348
0.18, 0.00
Ϫ0.17 to ϩ0.24
0.06
1803 [I > 2σ(I)]
NRCVAX b
263
0.251, NA
Ϫ0.17 to ϩ0.18
0.052
1429 [I > 3σ(I)]
TEXSAN a
329
1.8, 0.08
Ϫ0.19 to ϩ0.25
0.048
No. of parameters refined
Final shift/error (maximum and average)
Max residual density/e ÅϪ3
R = Σ Fo| Ϫ |Fc /Σ|Fo|
Rw = [Σω(|Fo| Ϫ |F c|)2/Σω(|Fo|2)]0.5
GOF = [Σω(|Fo| Ϫ |Fc|)2/(No Ϫ Nv)]0.5
0.07
1.89
0.031
1.34
0.042
1.7
a
TEXRAY Structure Analysis Package, Molecular Structure Corporation 1985. b Full System Reference, see ref. 26.
The molecule 16 is disordered over two orientations with
3. The IR spectra (32 scans) were recorded on an Analect
populations of 71 and 29%.§ The disordered atoms are C1Ј and
C2Ј (labels in 16). No disorder was resolved for the phenyl rings
attached to these carbons. Disordered atoms were refined iso-
tropically and H2Ј was introduced at a calculated position and
was not refined. All other hydrogens were located on difference
Fourier maps and were refined isotropically.
The molecule 14a crystallizes with the solvent ethyl acetate in
the proportion 2:1. The ethyl acetate molecule is disordered
over two orientations about the two-fold axis. Each orientation
corresponds to 50% occupancy. No atoms are on the axis but
the methylene and carbonyl carbons are close. The carbonyl
carbon was refined isotropically because of its proximity to the
two-fold axis. Hydrogen atoms in the solvent molecule were not
included in the model.
Instruments AQS-18 FTIR spectrometer at a resolution of 2
cmϪ1 in CDCl3 solution at ca. 0.01 M concentration at room
temperature in a KBr cell of ca. 0.05 mm path length. The UV
measurements were made on
a Unicam SP800 spectro-
photometer in MeOH solution at 25 ЊC (Table 4). A Milton
Roy® CM4000 multiple solvent delivery system equipped with
spectroMonitor® 3100 variable wavelength detector and CI-10B
integrator was used to run HPLC chromatograms using
reverse-phase columns at a MeOH flow rate of 1 cm3 minϪ1
.
Materials
Carbazole (Pfaltz & Bauer), cinnamoyl chloride, 4-fluorobenzyl
chloride (Lancaster), ButOK (Aldrich), DMF (BDH) and
2-methylpropene (Matheson) were used without further purifi-
cation. All chromatographic separations were performed using
silica gel 260 (BDH) unless otherwise specified. Compounds
10a,4 11b2 and 1317 were prepared as outlined earlier. Elem-
ental analyses are listed in Table 5.
Refinement for 15c was carried out in blocks of one molecule
at a time, all non-hydrogen atoms were refined anisotropically,
hydrogen atoms were introduced at idealized geometries [C–H
= 0.95 Å and Uiso(H) = 1.1 U(eq)C ϩ 0.01]. Acentricity in this
crystal comes from packing, not from the molecule.
9-(4-Fluorobenzyl)-9H-carbazole 10b
Instruments and techniques
Anhydrous K2CO3 (22.1 g, 0.160 mol) and KOH (33.8 g, 0.500
mol) were blended, powdered and added to tetrabutyl-
ammonium hydrogen sulfate (3.39 g, 0.01 mol). To this mixture
was added carbazole (16.7 g, 0.100 mol), toluene (300 cm3) and
4-fluorobenzyl chloride (14.5 g, 0.100 mol). This mixture was
stirred at 80 ЊC for 1 h, filtered hot and the solids washed with
toluene (100 cm3). The filtrate was evaporated and the solid
residue(24.1g, 87.5%)recrystallized from MeOH (ca. 1dm3). The
second recrystallization of a small sample gave plates, mp 124–
125 ЊC (MeOH).
1H, 13C, H–C HETCOR and H-decoupling NMR experiments
were recorded at room temperature on Varian 500 Unity NMR
or JEOL-270 spectrometers. Usually 30–50 mg of sample was
dissolved in 1.0 cm3 of CDCl3 and the solvent resonance was
1
used as the internal lock. For the H NMR determinations
tetramethylsilane was used as the internal standard. Carbon
and hydrogen NMR assignments on the Varian 500 Unity were
determined with the aid of H-COSY, HMQC and HMBC
techniques as described recently for other materials.27 The
numbers refer to the position of the carbon or hydrogen as
illustrated in the respective structures (see text). HETCOR and
COLOC experiments were used on the JEOL-270 spectrometer
for the partial assignments of 14b. The data are listed in Table
(4-tert-Butoxybenzylidine)aniline 11c
4-Hydroxybenzaldehyde, recrystallized from H2O and then
benzene (10 g, 0.082 mol), was dissolved in benzene (600 cm3).
To this solution, while stirring at 45 ЊC but without any other
external heat, was added Amberlyst 15 (BDH, 2.5 g) and
§ The minor orientation has a large error value associated with it.
1064
J. Chem. Soc., Perkin Trans. 1, 1997

View Article Online
Table 3 NMR spectroscopic data for novel productsa
Product
δC(J/Hz, assignment b)
δH (multiplicity, J/Hz, assignment b)
10b
45.00 (1Ј), 109.66 (1), 115.52 (JC3Ј,F 21.1, 3Љ), 119.27 (3),
120.53 (4), 122.45 (4a), 126.05 (2), 128.92 (JC2Љ,F 8.2, 2Љ),
134.17 (JC1Љ,F 3.7, 1Љ), 140.21 (9a), 161.53 (JC4Љ,F 242.7, 4Љ).
28.81 (CH3), 79.85 (CMe3), 122.26 (3), 130.94 (1), 131.11
(2), 161.62 (4), 190.93 (CHO).
5.62 (s, 2 H, 1Ј), 7.06 (t, J 8.8, 2 H), 7.20 (t, 4 H), 7.42 (t, 2
H), 7.62 (d, 2 H, 1), 8.16 (d, J 7.8, 2 H, 4).
4-ButOC6H4CHO
1.44 (s, 9 H, CH3 × 3), 7.10 (d, J2,3 8.5, 2 H, 3), 7.80 (d, J2,3
8.5, 2 H, 2), 9.91 (s, 1 H, COH).
12
44.37 (1Ј), 109.61 (1), 119.11 (3), 120.48 (4), 122.45 (4a),
124.84 (2Ј), 125.93 (2), 126.39 (2Љ), 127.76 (4Љ), 128.68 (3Љ),
131.4 (3Ј), 136.16 (1Љ), 140.13 (9a).
5.15 (d, J 6.37, 2 H, 1Ј), 5.61 (d, J2Ј,3Ј 16.11, 1 H, 3Ј), 6.38
(tt, J 5.86, J2Ј,3Ј 16.11, 1 H, 2Ј), 7.14–7.27 (m, 7 H, 3 and
2Љ–4Љ), 7.46 (t, J1,2 = J2,3 = 7.5, 2 H, 2), 7.63 (d, J1,2 7.81, 2
H, 1), 8.18 (d, J3,4 7.81, 2 H, 4).
14a
110.62 (1), 119.6 (3), 120.12 (4), 123.3 (4a), 123.41 (8Љ),
125.17 (2Љ), 125.35 (3Љ), 125.65 (6Љ), 125.88 (2), 126.33 (7Љ),
126.52 (3Ј), 126.81 (2Љ), 128.04 (4ٞ), 128.25(1Ј), 128.44
(4Љ), 128.58 (3ٞ), 128.77 (5Љ), 130.82 (1Љ), 131.25 (4Ј),
131.35 (8aЉ), 133.35 (4aЉ), 135.28 (2Ј), 136.28 (1ٞ), 139.91
(9a).
6.05 (d, J3Ј,4Ј16.1, 1 H, 4Ј), 6.60 (d, J2Љ,3Љ 7.3, 1 H, 2Љ), 6.75
(t, J2Љ,3Љ = J3Љ,4Љ = 7.8, 1 H, 3Љ), 7.16–7.28 (m, 11 H, 1–3 and
2ٞ–4ٞ), 7.31 (d, J3Ј,4Ј 15.7, 1 H, 3Ј), 7.45 (br d, 2 H, 4Љ and
6Љ), 7.56 (t, J6Љ,7Љ 7.5, 1 H, 7Љ), 7.70 (d, J5Љ,6Љ 7.8, 1 H, 5Љ),
7.75 (s, 1 H, 1Ј), 8.07 (d, J3,4 7.3, 2 H, 4), 8.27 (d, J7Љ,8Љ 8.3, 1
H, 8Љ).
14b
14cc
15a
55.02 (CH3), 110.51 (1), 113.94 (3Љ), 119.69 (3), 120.25 (4),
123.29 (9a), 126.05 (2), 126.58 (2ٞ or 3ٞ), 126.85 (1Љ),
127.09 (3Ј), 128.48 (2ٞ or 3ٞ), 129.29 (4Ј), 130.23 (2Љ),
131.34 (2Ј), 131.59 (1Ј), 136.51 (1ٞ), 139.27 (4a), 159.38
(4Љ).
110.52 (1), 119.78 (3), 120.26 (4), 123.36 (4a), 126.72 (2),
126.72 (2ٞ), 126.8 (3Ј), 127.9 (4ٞ), 128.14 (4Љ), 128.42 (3Љ),
128.5 (3ٞ), 128.62 (2Љ), 130.57 (4Ј), 131.84 (1Ј), 133.48 (2Ј),
134.11 (1Љ), 136.28 (1ٞ), 139.31 (9a).
3.62 (s, 3 H, CH3), 5.72 (d, J3Ј,4Ј 15.8, 1 H, 4Ј), 6.44 (d,
J2Љ,3Љ 8.6, 2 H, 3Љ), 6.58 (d, J2Љ,3Љ 9.0, 2 H, 2Љ), 7.01 (s, 1 H, 1Ј),
7.10–7.30 (m, 11 H), 7.13 (d, 1 H, 3Ј), 8.14 (d, J3,4 7.9, 2 H,
4).
5.85 (d, J3Ј,4Ј 16.11, 1 H, 4Ј), 6.68 (d, J2Љ,3Љ 7.3, 2 H, 2Љ),
6.92 (t, J2Љ,3Љ = J3Љ,4Љ = 7.3, 2 H, 3Љ), 6.97 (d, 1 H, 4Љ), 7.07 (s, 1
H, 1Ј), 7.16 (d, J3Ј,4Ј 16.11, 1 H, 3Ј), 7.16–7.26 (m, 9 H, 1
and 3 and 2ٞ–4ٞ), 7.31 (t, J2,3 = J1,2 = 7.8, 2 H, 2), 8.16 (d,
J3,4 7.8, 2 H, 4).
6.21 (d, J3Ј,4Ј 16.1, 1 H, 4Ј), 7.09 (m, 5 H, 2ٞ–4ٞ), 7.30 (t,
J2,3 = J3,4 = 7.3, 2 H, 3), 7.37 (d,d 1 H, 3Ј), 7.39 (s,d 1 H, 1Ј),
7.44 (t, J1,2 = J2,3 = 7.3, 2 H, 2), 7.47–7.54 (m, 4 H, 1 and
6Љ–7Љ), 7.59 (t, J2Љ,3Љ = J3Љ,4Љ = 7.5, 1 H, 3Љ), 7.83 (d, J2Љ, 3Љ 7, 1
H, 2Љ), 7.90 (d, J3Љ,4Љ 7,e 2 H, 4Љ–5Љ), 7.99 (d, J7Љ,8Љ 8.8, 1 H,
8Љ), 8.18 (d, J3,3 7.3, 2 H, 4).
110.45 (1), 119.72 (3), 120.3 (4), 121.6 (3Ј), 123.26 (4a),
124.71 (8Љ), 125.41 (3Љ), 126.02 (2), 126.3 (6Љ), 126.59 (7Љ),
126.94 (2ٞ), 128.13 (4ٞ), 128.28 (2Љ), 128.5 (3ٞ), 128.61
(4Љ), 128.84 (5Љ), 130.33 (1Ј), 131.88 (8aЉ), 132.52 (1Љ), 133
(4Ј), 133.68 (4aЉ), 136.28 (1ٞ), 136.61 (2Ј), 141.3 (9a).
15c
16
110.51 (1), 119.65 (3), 120.21 (4), 121.28 (3Ј), 123.21 (4a),
125.89 (2), 126.97 (2ٞ), 128.04 (4Љ), 128.2 (4Љ), 128.58 (3ٞ),
128.71 (3Љ), 129.56 (2ٞ), 131.96 (1Ј), 133.33 (4Ј), 135.04
(2Ј), 135.69 (1Љ), 136.36 (1ٞ), 141.18 (9a).
6.19 (d, J3Ј,4Ј 16.1, 1 H, 4Ј), 6.85 (s, 1 H, 1Ј), 7.15–7.20 (m,
5 H, 2ٞ–4ٞ), 7.26 (dd, J2,3 = J3,4 = 7.8, 2 H, 3), 7.28–7.35
(t, 1 H, 4Љ), 7.38 (t, J1,2 = J2,3 = 7.5, 2 H, 2), 7.42 (d, J1,2 7.5,
2 H, 1), 7.46 (t, J2Љ,3Љ = J3Љ,4Љ = 7.5, 2 H, 3Љ), 7.52 (d, J3Ј,4Ј 16.1,
1 H, 3Ј), 7.53–7.56 (br d, 2 H, 2Љ), 8.14 (d, J3,4 8.3, 2 H, 4).
6.66 (t, J2ٞ,3ٞ = JH 3ٞ,Fٞ = 8.8, 2 H, 3ٞ), 6.80 (dd, J2ٞ,3ٞ 8.8,
JH 2ٞ,F4ٞ 5.4, 2 H, 2ٞ), 6.93 (t, J2Љ,3Љ = JH 3Љ,F4Љ = 8.8, 2 H, 3Љ),
7.02 (d, J1,2 7.8, 2 H, 1), 7.16 (dd, J2Љ,3Љ 9, JH 2Љ,F4Љ 5, 2 H, 2Љ),
7.21 (s, 1 H, 2Ј), 7.22–7.27 (m, 4 H, 2 and 3), 8.12 (d, J3,4
8.3, 2 H, 4).
110.91 (1), 115.44 (JC3ٞ,F4ٞ 21.1, 3ٞ), 115.8 (JC3Љ,F4Љ 22, 3Љ),
120.19 (3), 120.31 (4), 123.72 (4a), 125.78 (2Ј), 126.07 (2),
127.71 (JC2Љ,F4Љ 8.2, 2Љ), 130.17 (JC2ٞ,F4ٞ 8.2, 2ٞ), 130.8
(JC3ٞ,F4ٞ 2.7, 1ٞ), 132.57 (JC1Ј,F4Љ 1.8, 1Ј), 133.7 (JC1Љ,F4Љ 3.7,
1Љ), 139.2 (9a), 162.03 (JC4ٞ,F4ٞ 249.1, 4ٞ), 163.03 (JC4Љ,F4Љ
249.1, 4Љ).
17
19
23
28.71 (CH3), 78.72 (CMe3), 111.09 (1), 115.72 (JC3Љ,F 21.1,
3Љ), 119.98 (3), 120.18 (4), 123.68 (3ٞ), 125.9 (2), 126.54f
(4a), 126.56f (2Ј), 127.57 (JC2Љ,F 8.2, 2Љ), 129.19 (3ٞ), 129.69
(1ٞ), 131.57 (1Ј), 134.05 (JC1Љ,F 3.7, 1Љ), 139.33 (9a), 155.31
(4ٞ), 162.85 (JC4Љ,F 249.1, 4Љ).
28.72 and 28.84 (CH3), 78.62 and 78.77 (CMe3), 111.24
(1), 119.77 (3), 120.05 (4), 123.58 (4a), 123.68 (3ٞ), 123.82
(3Љ), 125.52 (2Ј), 125.78 (2), 126.51 (2Љ), 129.04 (2ٞ),
130.06 (1ٞ), 132.34 (1Ј), 132.63 (1Љ), 139.5 (9a), 155.02
(4ٞ), 156.04 (4Љ).
109.82 (1), 120.12 (3), 120.23 (4Ј), 120.42 (4), 123.06 (5Ј),
123.53 (4a), 124.31, 126.10 (2), 126.60 (2Ј) 126.99, 127.30,
127.42 (8Ј), 127.70 (4Љ), 128.46 (3Љ), 128.62, 129.02, (8aЈ or
10aЈ), 130.07, (4bЈ), 130.15 (2Љ), 132.00 (8aЈ or 10aЈ),
132.04 (4aЈ), 135.36 (3Ј), 140.09 (1Љ), 141.12 (9a), 143.00
(1Ј).
1.18 (s, 9 H, CH3 × 3), 6.59 (d, J2ٞ,3ٞ 8.3, 2 H, 3ٞ), 6.72 (d,
J2ٞ,3ٞ 8.5, 2 H, 2ٞ), 6.93 (dd, J2Љ,3Љ 8.8, J3Љ,F 8.5, 2 H, 3Љ), 7.02
(d, J1,2 8.3, 2 H, 1), 7.17 (dd, J2Љ,3Љ 8.5, J2Љ,F 5.4, 2 H, 2Љ),
7.20–7.25 (m, 5 H, 2, 3 and 2Ј),g 8.12 (d, J3,4 8.8, 2 H, 4).
1.22 (s, 9 H, CH3 × 3), 1.35 (s, 9 H, CH3 × 3), 6.62 (d, J2ٞ,3ٞ
8.5, 2 H, 3ٞ), 6.73 (d, J2ٞ,3ٞ 8.8, 2 H, 2ٞ), 6.90 (d, J2Љ,3Љ 8.8, 2
H, 3Љ), 7.08 (d, J1,2 7.8, 2 H, 1), 7.15 (d, J2Љ,3Љ 8.8, 2 H, 2Љ),
7.22–7.28 (m, 5 H, 2, 3 and 2Ј), 8.12 (d, J3,4 8, 2 H, 4).
7.32 (t, J 7.8, 2 H, 3), 7.44 (t, J 8.3, 2 H, 2), 7.47 (t, J 7, 1
H, 4Љ), 7.53–7.62 (m, 6 H), 7.63–7.68 (m, 2 H), 7.75–7.78
(m, 2 H), 7.91–7.95 (m, 2 H), 8.20 (d, J 7.8, 2 H, 1), 8.64–
8.67 (m, 1 H, 5Ј), 8.93 (d, J 1.5, 1 H, 4Ј).
a
1
The H NMR spectra at room temperature were referenced using SiMe4 as the internal standard. The 13C NMR spectra were referenced to the
middle line of CDCl3 at δ 77.00 as per ref. 23(c). Assignments were aided by H-COSY, HMQC (null = 0.65 s) and HMBC (taumb = 0.05 s)
b
experiments similar to those recently described in ref. 27. The atom numbering is indicated on the respective structures. c HMQC (null = 0.7 s).
g
d Doublet and singlet are superimposed. e Two doublets with the same chemical shift. f Values may be interchanged. From difference NOE
experiment, H2Ј resonates at δ 7.22.
2-methylpropene gas was bubbled slowly into the reaction for
17 h. An aliquot indicated a 64% conversion (HPLC). After this
period the solution was filtered, the filtrate extracted with
10% NaOH (2 × 100 cm3), washed with H2O (2 × 100 cm3) and
dried over MgSO4. The mixture was filtered and the filtrate
evaporated. The residue weighed 11 g. This residue was
adsorbed on alumina (basic) and chromatographed using light
petroleum (bp 35–60 ЊC)–EtOAc (97:3) as eluent. Evaporation
of the solvent from the later fractions gave 4-tert-butoxy-
benzaldehyde28 (2.0 g) as an oil.
aniline (1.04 g, 0.0110 mol) and benzene (30 cm3), and the water
was removed by azeotropic distillation in a Dean–Stark trap for
1 h. Unreacted aldehyde remained (HPLC) and aniline was
added in increments of 0.2 cm3 to the boiling benzene and dis-
tillation continued for 15 min after each addition. The 1H NMR
spectrum of the mixture, where the equilibrium is completely
over to the imine side, shows ca. 75% mole excess of aniline.
Attempted purification by chromatography using either silica
gel or aluminum oxide (basic), with light petroleum–EtOAc
(97:3) as eluent, gave a mixture containing 10–15% of the alde-
hyde by hydrolysis. Since an excess of aniline is not detrimental
To 4-tert-butoxybenzaldehyde (2.0 g, 0.011 mol) was added
J. Chem. Soc., Perkin Trans. 1, 1997
1065

View Article Online
Table 4 Spectroscopic data for novel products
Compound
m/z (%)
λmax(MeOH)/nm (log E)
12
14a
260 (4.41), 288 (4.02), 294 (4.16), 329 (3.49), 343 (3.55)
259 (4.31), 289 (4.35), 336 (4.45)
421 (Mϩ, 100), 255 (26.6), 239 (16.9), 172 (22.8), 129 (45.9),
112 (15.8), 83 (40.3), 58 (38.9), 43 (56.4)
15a
14b
14c
15c
23
421 (Mϩ, 100), 255 (28.1), 239 (17.6), 172 (24.3), 129 (17.6)
259 (4.34), 289 (4.49), 326 (4.33), 336 (4.32)
260 (4.28), 293 (4.33), 337.5 (4.64)
259 (sh, 4.23), 297.5 (4.42), 326 (4.62)
255 (sh, 4.44), 291 (4.37), 318 (4.36)
HRMS: calc. for C32H21N 419.167 39. Found, 419.167 20.
214 (4.99), 252 (4.95), 290 (4.55), 321 (4.35), 323 (4.33)
Table 5 Elemental microanalyses for the new compounds
Compound
Formula
Found (%)
Calculated (%)
10b
12
C19H14FN: M, 275.33
C21H17N: M, 283.38
C32H23N: M, 421.55
C32H23N: M, 421.55
C29H23NO: M, 401.51
C28H21N: M, 371.49
C28H21N: M, 371.49
C26H17F2N: M, 381.42
C30H26FNO: M, 435.54
C34H35NO2: M, 489.66
C, 82.42: H, 5.26; N, 4.84
C, 88.73; H, 6.17; N, 4.94
C, 91.18; H, 5.60; N, 3.67
C, 91.13; H, 5.71; N, 3.40
C, 86.71; H, 5.61; N, 3.45
C, 90.26; H, 5.62; N, 3.68
C, 89.98; H, 5.83; N, 3.74
C, 81.77; H, 4.71; N, 3.65
C, 83.02; H, 6.04; N, 3.16
C, 83.31; H, 7.37; N, 2.82
C, 82.89; H, 5.13; N, 5.09
C, 89.01; H, 6.05; N, 4.94
C, 91.18; H, 5.50; N, 3.32
C, 91.18; H, 5.50; N, 3.32
C, 86.75; H, 5.77; N, 3.49
C, 90.53; H, 5.70; N, 3.77
C, 90.53; H, 5.70; N, 3.77
C, 81.87; H, 4.49; N, 3.67
C, 83.73; H, 6.02; N, 3.22
C, 83.40; H, 7.20; N, 2.86
14a
15a
14b
14c
15c
16
17
19
for the enamine formation this mixture was used below, as pre-
pared, with excess aniline.
Similar workup as for 14a and chromatography of the precipi-
tate on silica gel 60 using light petroleum–CHCl3 (9:1) as eluent
gave 15c (tR = 5.60 min) followed by 14c (tR = 3.75 min).
The eluted fractions which contained more than 93% of 15c
were combined, the solvent evaporated and the residue
recrystallized three times from acetonitrile into clear prisms.
The yield of this pure material was ca. 1%. The sample submit-
ted for X-ray crystal structure analysis was recrystallized from
MeOH. An analytical sample was recrystallized from acetone,
mp 158–160 ЊC.
(2ЈE)-9-(3Ј-Phenylprop-2Ј-enyl)-9H-carbazole 12
Following the procedure for the preparation of 10b, but using
cinnamoyl chloride (15.26 g, 0.10 mol) instead of 4-fluoro-
benzyl chloride, gave 12 (23.5 g, 83%), mp 147–150 ЊC (acetone).
(1ЈZ,3ЈE)- and (1ЈE, 3ЈE)-9-[1Ј-(1Љ-Naphthyl)-4Ј-phenylbuta-
1Ј,3Ј-dien-2Ј-yl]-9H-carbazole 14a and 15a
A solution of 12 (2.83 g, 0.01 mol) and 13a (2.60 g, 0.01 mol) in
DMF (10 cm3) was added to ButOK (1.50 g, 0.0134 mol) in
DMF (40 cm3) at 75 ЊC. After 12 min the conversion was 100%
by HPLC: the isomers had tR = 5.44 (73%) and 9.07 min (27%).
The solution was poured into H2O (150 cm3) and the solvent
was decanted from the oily precipitate which was washed with
H2O. Chromatography using light petroleum–CHCl3 (9:1) as
eluent gave 15a (tR = 9.07 min) followed by 14a (tR = 5.44 min).
The eluted fractions which contained more than 90% of
15a were combined, the solvent evaporated and the residue
recrystallized twice from acetone to give 15a as needles (23%),
mp 189–190 ЊC.
Combination of the fractions containing 14a and evapor-
ation of the solvent left a residue. This was recrystallized from
acetone to give clear prisms (2.65 g, 63%). When heated to
135 ЊC the compound effervesced and the prisms became
opaque; upon further heating the prisms melted: mp 175–
177 ЊC (acetone).
Eluted fractions which contained more than 85% of 14c were
combined, the solvent evaporated and the residue recrystallized
three times from acetonitrile into clear prisms. The yield of this
pure material was 45%. An analytical sample was recrystallized,
mp 143–145 ЊC (MeOH).
(1ЈZ)-9-[1Ј,2Ј-Bis(4-fluorophenyl)ethenyl]-9H-carbazole 16
To a solution of ButOK (6.72 g, 0.06 mol) in DMF (40 cm3)
at 75 ЊC was added a solution of 10b (5.5 g, 0.02 mol) and
11b (4.0 g, 0.02 mol) in DMF (10 cm3), and the mixture was
stirred at 75 ЊC for 1 min. After this period it was immediately
poured into H2O (200 cm3). This mixture was stirred and the
precipitate filtered off. The oily solid was dissolved in CHCl3
(200 cm3) and the solution dried over MgSO4. The solution was
filtered and the solvent evaporated. The residue was taken up
in CCl4 and chromatographed (CCl4). The first fractions con-
tained exclusively 16. These were combined and the solvent
evaporated. The residue (4.73 g, 62%) was recrystallized, mp
199 ЊC (acetonitrile).
(1ЈZ,3ЈE)-9-[1Ј-(4Љ-Methoxyphenyl)-4Ј-phenylbuta-1Ј,3Ј-dien-
2Ј-yl]-9H-carbazole 14b
(1ЈZ)-9-[2Ј-(4-tert-Butoxyphenyl)-1Ј-(4-fluorophenyl)ethenyl]-
9H-carbazole 17
Following the preparation for 14a on a 0.01 mol scale, using 13b
instead of 13a, gave after ca. 5 min a conversion of 100% by
HPLC: the isomers had tR = 4.69 (85%) and 6.46 min (15%).
Similar workup as for 14a and three recrystallizations from
EtOAc gave 14b (45%). An analytical sample was recrystallized
from methanol, mp 185–187 ЊC.
To a solution of ButOK (3.3 g, 0.029 mol) in DMF (40 cm3) at
75 ЊC was added a solution of 10b (2.75 g, 0.01 mol) and 11c (3 g
of a mixture containing a ca. 75% mole excess of aniline, 0.0093
mol) in DMF (10 cm3) and the mixture was stirred at 75 ЊC for 1
min. After this period it was immediately poured into water (150
cm3). This mixture was extracted with diethyl ether (3 × 50 cm3)
and the solvent evaporated over 5 g of silica gel. The adsorbed
material was chromatographed on 40 g of silica gel using light
petroleum–EtOAc (97:3) as eluent. Unreacted 10b eluted first
followed by 17. The fractions which contained greater than 85%
of 17 (HPLC) were combined and the eluent evaporated.
(1ЈZ,3ЈE)- and (1ЈE,3ЈE)-9- [1Ј,4Ј-Diphenylbuta-1Ј,3Ј-dien-2Ј-
yl]-9H-carbazole 14c and 15c
Following the preparation for 14a on a 0.05 mol scale, using 13c
instead of 13a, gave after ca. 2 min a conversion of 100% by
HPLC: the isomers had tR = 3.75 (90%) and 5.60 min (10%).
1066
J. Chem. Soc., Perkin Trans. 1, 1997

View Article Online
Recrystallization from propan-2-ol gave 1.84 g (45%). An ana-
lytical sample was recrystallized three times from MeOH, mp
147–150 ЊC.
X14c(PA) = PA14c/(PA14c ϩ PA15c
)
(1)
of 14c and 15c, each 0.0102 g in 1,2,4-trichlorobenzene (2.107
g) (0.013 03 M), a calibration graph was obtained by plotting
X14c obtained from the weighed aliquots of 14c and 15c diluted
with MeOH (ca. 5 cm3) versus X14c(PA) [eqn. (1)]. These calibra-
tion data are represented by eqn. (2). Using eqn. (2) we evaluate
(1ЈZ )-9-[1Ј,2Ј-bis(4-tert-butoxyphenyl)ethenyl]-9H-carbazole 19
To a solution of ButOK (5.60 g, 0.05 mol) in DMF (40 cm3) at
75 ЊC was added a solution of 10b (0.01 mol) and 11b (0.01
mol) in DMF (10 cm3) and the mixture was stirred at this temp-
erature for 18 h. The solution was poured into water (150 cm3)
and the precipitate was filtered. The oily precipitate was chrom-
atographed using light petroleum–EtOAc (9:1) as eluent to give
19. The fractions which contained not less than 90% of 19 were
combined, the solvent evaporated and the residue recrystallized
twice from MeOH (20% yield of purified material), mp 151–
153 ЊC.
X14c = 0.28149[X14c(PA)]2 ϩ 0.6924[X14c(PA)] ϩ 0.0101 (2)
X
14c(t) by substituting the quantity X14c(PA) which was cal-
culated [using eqn. (1)] and from peak areas from HPLC
chromatograms of the reaction aliquots taken at different
times.¶
Acknowledgements
Hydrolysis of dienamines 14a and 15a: 4-(1-naphthyl)-1-phenyl-
but-1-en-3-one 22
We thank Anne Marie Lebuis, McGill University, for the dif-
ficult task of determining the crystal structures and for verify-
ing the experimental report of all crystallographic data. This
work was realized through the financial support of the General
Electric Company and the Natural Sciences and Engineering
Research Council of Canada.
The dienamine 15a (0.5 g) was hydrolysed in AcOH–H2O–HCl
(18:1:1 cm3) by heating under reflux for 7 h. The mixture was
poured into H2O (50cm3)and extracted with CHCl3 (3 × 20cm3).
The CHCl3 layer was washed with saturated aqueous K2CO3
(30 cm3) and H2O (50 cm3), dried over MgSO4 and chromato-
graphed twice using light petroleum–EtOAc (94:6). Only
repeated extractions with light petroleum removed much of the
carbazole. The residual oil contained 22 and carbazole (84:16
¶ Further detailed spectroscopic data, including HPLC peak areas, are
available as supplementary data (Suppl. No. 57217) from the British
Library. For details of the Supplementary Publications Scheme, see
Instructions for Authors, J. Chem. Soc., Perkin Trans. 1, 1997, Issue 1.
1
estimated by HPLC), H NMR (300 MHz, CDCl3): δ 4.3 (s, 2
H, CH2), 6.77 (d, J 16 Hz, 1 H, H1), 7.18–7.53 (m, 10.6 H,
aromatic), 7.62 (d, J 16 Hz, 1 H, H2), 7.74–8.04 (m, 4 H,
aromatic).
The dienamine 14a (0.5 g) was hydrolysed similarly to give
the same mixture.
References
1 (a) The Chemistry of Enamines Part
1 and Part 2, ed.
Z. Rappoport, Wiley, New York, 1994; (b) P. W. Hickmott,
Tetrahedron, 1982, 38, 1975; (c) P. W. Hickmott, Tetrahedron, 1982,
28, 3363, (d) P. W. Hickmott, Tetrahedron, 1984, 40, 2989.
2 M. Paventi, E. Elce, R. J. Jackman and A. S. Hay, Tetrahedron
Lett., 1992, 33, 6405.
3 M. Paventi and A. S. Hay, Tetrahedron Lett., 1993, 34, 999.
4 M. Paventi and A. S. Hay, Tetrahedron Lett., 1991, 32, 1617.
5 M. Ogata, K. Tsukamoto, T. Kamisaka, T. Watanuki and
N. Saruwatari, Jpn. Kokai Tokkyo Koho, JP 02 13,963 [90 13,963], 18
Jan 1990; Chem. Abstr., 1990, 113, 31973a.
6 H. Naarmann and P. Strohriegel, Conducting and Photo-
conducting Polymers, in Handbook of Polymer Synthesis. Part B,
ed. H. R. Kricheldorf, Marcel Dekker, New York, 1992, ch. 21,
p. 1353.
7 H. Sugimura, Y. Kojima, S. Nishigaki and K. Emoto, IS&T’s
Int. Congr. Adv. Non-Impact Print. Technol., Final Program Proc.,
9th, 1993, 627.
9-(1-Phenylphenanthren-3-yl)-9H-carbazole 23
The dienamine 14a (0.50 g, 1.2 mmol) was heated in 1,2,4-
trichlorobenzene (7.0 g) under reflux for 103 h. After this period
the conversion was 88% by HPLC. The solvent was removed by
blowing N2 over the heated solution. The residue was cooled
and taken up with CHCl3. The solution was adsorbed on ca. 5 g
of silica gel and chromatographed over silica gel (30 g) using
light petroleum–EtOAc (95:5) as eluent. The fractions which
contained over 90% 23 (tR = 10.2 min) were combined and the
solvent evaporated. The residue was boiled in MeOH (30 cm3)
and the volume of solvent reduced to ca. 3 cm3 before filtering.
The process of boiling with MeOH, reducing the solvent vol-
ume and filtering was repeated twice more. Total material
amassed was 0.20 g. This was recrystallized from EtOAc as
small needles, mp 227–229 ЊC.
8 A. Itoh, M. Aida, N. Hirota, T. Kai and M. Okaji, IS&T’s Int.
Congr. Adv. Non-Impact Print. Techno., Final Program Proc., 9th,
1993, 631.
Technique of thermal equilibration: 14c to 15c and 15c to 14c
In two different test tubes, each containing 1,2,4-
trichlorobenzene (2.0 g) and a small stirrer bar, was placed 14c
(0.025 g, 67 µmol) and 15c (0.010 g, 27 µmol). The two test tubes
were each wired to a thermometer so that the bulb was at the
level of the liquid in the tubes and this setup was lowered into a
silicone oil bath stirring at 180 ЊC. This temperature was moni-
tored periodically so that any variation did not exceed ±2 ЊC.
The effect of any changes in temperature were largely damp-
ened by the very slow rate of equilibration. Evaporation of
solvent is not critical because at 180 ЊC only a slow evaporation
of solvent takes place and, more importantly, the mole frac-
tions at different times [X14c(t)], weight fractions (X14c) or peak
area (HPLC) fractions [X14c(PA)] being measured are volume-
independent. The aliquots were taken by dipping an open-ended
pasteur pipette into the solution. The small volume that was
withdrawn was diluted with methanol (ca. 5 cm3) and the time
elapsed was recorded. The methanolic aliquot was injected into
the HPLC system and the ratio of peak areas obtained at 300
nm at two different retention times of the same injected sample
gives the quantity X14c(PA) from eqn. (1). From stock solutions
9 J. W. Weigl, Angew. Chem., Int. Ed. Engl., 1977, 16, 374.
10 K.-Y. Law, Chem. Rev., 1993, 93, 449.
11 K. Meerholz, B. L. Volodin, Sandalphon, B. Kippelen and
N. Peyghambarian, Nature, 1994, 371, 497.
12 H. Hoegl, O. Süs and W. Neugebauer, Ger. Pat., 1,068,115 to
Kalle AG (1961); Chem Abstr., 1961, 55, 20742a; H. Hoegl, J. Phys.
Chem., 1965, 69, 755.
13 T. Watanuki, T. Kamisaka and N. Saruwatari, Jpn. Kokai
Tokkyo Koho, JP 02 47, 666 [90 47,666], 16 Feb 1990; Chem. Abstr.,
1990, 113, 88221h.
14 O. Cervinka, Preparation of enamines, in The Chemistry of
Enamines. Part 1, ed. Z. Rappoport, Wiley, New York, 1994, ch. 9,
p. 467.
15 I. J. Fletcher and A. E. Siegrist, Adv. Heterocycl. Chem., 1978,
23, 171.
16 B. Weickhardt and A. E. Siegrist, Helv. Chim. Acta, 1972, 55,
173.
17 M. Paventi and A. S. Hay, J. Org. Chem., 1991, 56, 5875.
18 A. S. Hay and M. Paventi, US Pat. 5,011,998, 30 Apr 1991;
Chem. Abstr., 1991, 115, 182798q.
19 (a) E. Pretsch, J. Seibl and W. Simon, in Tables of Spectral Data
for Structure Determination of Organic Compounds: 13C-NMR, IR,
MS, UV/VIS, 2nd edn., ed. W. Frenius, J. F. K. Huber, E. Pungor,
G. A. Rechnitz, W. Simon and T. S. West, Springer-Verlag, Berlin
J. Chem. Soc., Perkin Trans. 1, 1997
1067

View Article Online
1989, p. H205; (b) p. I36; p. I37; p. I152; (c) p. I20; (d) p. I105;
(e) p. C255.
26 E. J. Gabe, Y. Le Page, Y.-P. Charland, F. L. Lee and P. S. White,
J. Appl. Crystallogr., 1989, 22, 384.
20 R. M. Coates and E. F. Johnson, J. Am. Chem. Soc., 1971, 93,
4016; J. Hine, S.-M. Linden, A. Wang and V. Thiagarajan, J. Org.
Chem., 1980, 45, 2821; J. Hine and M. J. Skoglund, J. Org. Chem.,
1982, 47, 4785; J. Hine and M. J. Skoglund, J. Org. Chem., 1982, 47,
4766.
27 M. Paventi, K. P. Chan and A. S. Hay, J. Macromol. Sci., Pure
Appl. Chem., 1996, 33, 133.
28 M. Barl, D. Degner, H. Siegel and W. Hoffmann (BASF A.-G.),
Ger. Offen., 2,935,398 1981; Chem. Abstr., 1981, 95, P24556y;
D. Degner, H. J. Pander and H. Siegel (BASF A.-G.), Ger. Offen.,
3,002,543, 1981; Chem. Abstr., 1981, 95, P187672z; A. Dutta and
J. S. Morley, J. Chem. Soc., Perkin Trans. 1, 1975, 1712.
21 F. Daniels and R. A. Alberty, Physical Chemistry, 3rd edn.,
Wiley, New York, 1966, ch. 10, pp. 338–339.
22 J. E. Anderson, Tetrahedron Lett., 1975, 4079.
23 R. Stradi, P. Trimarco and A. Vigevani, J. Chem. Soc., Perkin
Trans. 1, 1978, 1.
24 K. Müllen, T. Meul, P. Schade, H. Schmickler and E. Vogel,
J. Am. Chem. Soc., 1987, 109, 4992.
25 B. J. Whitlock and H. W. Whitlock, J. Am. Chem. Soc., 1990,
112, 3910; J. E. Cochran, T. J. Parrot, B. J. Whitlock and H. W.
Whitlock, J. Am. Chem. Soc., 1992, 114, 2269; B. J. Whitlock and
H. W. Whitlock, J. Am. Chem. Soc., 1994, 116, 2301.
Paper 6/04969I
Received 16th July 1996
Accepted 4th November 1996
1068
J. Chem. Soc., Perkin Trans. 1, 1997