Chelation-Controlled Regioselectivity in the Synthesis of Substituted Pyrazolylpyridine Ligands. 3. Unsymmetrically Substituted Tridentates and Ditopic Bis(tridentates) en Route to Singly Stranded Helicates

Article abstract of DOI:10.1021/jo970815z

A strategy for the synthesis of helical multinuclear complexes (helicates) of any length is outlined, making use of ditopic ligands to bridge between metals and monotopic ligands to cap the chain's termini. We present the syntheses of both ditopic and monotopic ligands with tridentate binding units for octahedral metals, in liposoluble and functionalized varieties. Of special interest are ligands bearing benzoic acid or toluic acid side-chains which can ionize and serve as counteranions. Several singly N(1)-substituted 2,6-bis(4,5,6,7-tetrahydroindazol-3-yl)pyridines were prepared by regioselective, chelation-controlled mono-N-alkylation or -arylation of the tautomeric N,N'-H₂ parent compound. These were then either coupled with CH₂ linkages to provide the ditopic ligands or were further N-functionalized to give unsymmetrically substituted capping ligands. The regiochemistry of the N-substitutions and the ring-to-ring conformations were ascertained by ¹H NMR spectra with and without added complexands (H⁺, D⁺, Zn²⁺, and Na⁺). The formation of a bis-(ZnBr₂) adduct confirmed the independence of the metal binding sites while Na⁺ formed the 2:2 helicate. 1 Part 2: ref 4.

Full text of DOI:10.1021/jo970815z

J . Org. Chem. 1998, 63, 235-240
235
Ch ela tion -Con tr olled Regioselectivity in th e Syn th esis of
Su bstitu ted P yr a zolylp yr id in e Liga n d s. 3. Un sym m etr ica lly
Su bstitu ted Tr id en ta tes a n d Ditop ic Bis(tr id en ta tes) en Rou te to
Sin gly Str a n d ed Helica tes†
J erzy Zadykowicz and Pierre G. Potvin*
Department of Chemistry, York University, 4700 Keele Street, North York, Ontario, Canada M3J 1P3
Received May 6, 1997
A strategy for the synthesis of helical multinuclear complexes (helicates) of any length is outlined,
making use of ditopic ligands to bridge between metals and monotopic ligands to cap the chain’s
termini. We present the syntheses of both ditopic and monotopic ligands with tridentate binding
units for octahedral metals, in liposoluble and functionalized varieties. Of special interest are
ligands bearing benzoic acid or toluic acid side-chains which can ionize and serve as counteranions.
Several singly N(1)-substituted 2,6-bis(4,5,6,7-tetrahydroindazol-3-yl)pyridines were prepared by
regioselective, chelation-controlled mono-N-alkylation or -arylation of the tautomeric N,N′-H2 parent
compound. These were then either coupled with CH2 linkages to provide the ditopic ligands or
were further N-functionalized to give unsymmetrically substituted capping ligands. The regio-
1
chemistry of the N-substitutions and the ring-to-ring conformations were ascertained by H NMR
spectra with and without added complexands (H+, D+, Zn2+, and Na+). The formation of a bis-
(ZnBr2) adduct confirmed the independence of the metal binding sites while Na+ formed the 2:2
helicate.
In tr od u ction
Several examples of helicates, one-dimensional, oligo-
nuclear metal complexes of helical architecture, have
been reported in recent years.1 Most are formed from
polypyridines serving as oligo(bidentates) for tetrahedral
metal centers such as CuI, and a few examples of oligo-
F igu r e 1. Singly stranded helical polynuclear assembly,
where the rectangles depict the binding planes in polytetra-
hedra or polyoctahedra, and where L represents a linker group.
octahedral complexes from oligo(tridentates) are known.
The construction of long, doubly stranded helicates uses
self-assembly from long, polytopic ligands, wherein the
metals glue the intertwined ligand strands. Our ap-
proach (Figure 1) obviates the need for long polytopic
ligands. Instead, a single strand can be constructed by
linking any number of ditopic ligands via the metal
centers, with the termini being capped by monotopic
ligands. Depending on the length and flexibility of the
linker groups (L) chosen, such chains can achieve the
same helicity as double strands, especially if interplanar
stacking occurs. A greater variety of helicates are
accessible and functionality can be introduced at each
metal site. Importantly, the ditopic ligands required are
much more easily synthesized than those of higher
topicity by, as reported herein, a coupling of unsymmetri-
cally substituted monotopic precursors, while monotopic
ligands would cap the chains at the terminal metal
centers.
A suitable class of tridentate building blocks are the
2,6-bis(tetrahydroindazol-3-yl)pyridines 1. They and
their bidentate cousins comprise a class of newly devel-
oped ligands for transition metals2-4 which are readily
prepared from commercially available materials by short
routes and in good yields. They are lipophilic and readily
form stable, organo-soluble Ru(II),2,5,6 Zn(II),3 and Co(II)7
complexes. Unlike the less versatile 2,1(N)-linked 2,6-
bis(pyrazol-1-yl)pyridines 2,8-10 our 2,3(C)-linked ana-
logues can be functionalized at NH sites, and this facility
has been used to prepare pentadentate and macrocyclic
† Part 2: ref 4.
(1) Constable, E. C. Tetrahedron 1992, 48, 10013. Lehn, J .-M.;
Sauvage, J .-P.; Simon, J .; Ziesel, R. New J . Chem. 1983, 7, 413. Cathey,
C. J .; Constable, E. C.; Hannon, M. J .; Tocher, D. A.; Ward, M. D. J .
Chem. Soc., Chem. Commun. 1990, 629. Constable, E. C. Angew. Chem.
Intl. Ed. Engl. 1991, 30, 1450. Potts, K. T.; Raiford, K. G. A.;
Keshavarz, M.-K. J . Am. Chem. Soc. 1993, 115, 2793. Cathey, C. J .;
Constable, E. C.; Hannon, M. J .; Tocher, D. A.; Ward, M. D. J . Chem.
Soc. Chem. Commun. 1990, 621. Buchecker, D. O. C.; Sauvage, J .-P.
Angew. Chem., Int. Ed. Engl. 1989, 28, 189. Piguet, C.; Bernardelli,
G.; Williams, A. F. Inorg. Chem. 1989, 28, 1450. Bell, T. W.; J ousselin,
H. Nature 1994, 367, 441. Koert, U.; Harding, M. M.; Lehn, J .-M.
Nature 1990, 346, 339. Constable, E. C.; Thompson, A. M. W. Inorg.
Chim. Acta 1994, 223, 177. Crane, J . D.; Sauvage, J .-P. New J . Chem.
1992, 16, 649.
(5) Luo, Y.; Potvin, P. G.; Tse, Y.-H.; Lever, A. B. P. Inorg. Chem.
1996, 35, 5445.
(6) Zadykowicz, J .; Potvin, P. G. Manuscript in preparation.
(7) J airam, R., M. Sc. Thesis, York University, 1991. Adorante, J .;
Lau, M.; Potvin, P. G. Unpublished work.
(8) Saha, N.; Kar, S. K. J . Inorg. Nucl. Chem. 1977, 39, 1236.
Watson, A. A.; House, D. A.; Steel, P. J . J . Org. Chem. 1991, 56, 4072.
Chabert, N.; J acquet, L.; Marzin, C.; Tarrago, G. New J . Chem. 1995,
19, 443.
(9) J ameson, D. L.; Blaho, J . K.; Kruger, K. T.; Goldsby, K. A. Inorg.
Chem. 1989, 28, 4312.
(2) Dash, R.; Potvin, P. G. Can. J . Chem. 1992, 70, 2249.
(3) Luo, Y.; Potvin, P. G. J . Org. Chem. 1994, 59, 1761.
(4) van der Valk, P.; Potvin, P. G. J . Org. Chem. 1994, 59, 1766.
(10) Steel, P. J .; Constable, E. C. J . Chem. Soc., Dalton Trans. 1990,
1389. Downard, A. J .; Honey, G. E.; Steel, P. J . Inorg. Chem. 1991,
30, 3733.
S0022-3263(97)00815-3 CCC: $15.00 © 1998 American Chemical Society
Published on Web 01/01/1998

236 J . Org. Chem., Vol. 63, No. 2, 1998
Zadykowicz and Potvin
octadentate species2 as well as ligands bearing acetic or
benzoic acid side-chains.3,4 Ligands bearing peripheral
ionizable groups are of particular interest as they provide
access to anionic, neutral, or cationic complexes under
pH control and we wished to employ this to modulate
the interactions between photoexcited RuII electron do-
nors and cationic electron acceptors such as methylvi-
ologen.
homogeneous, containing a suspended white solid which
we tentatively identified as the monosodio derivative of
3. We speculate that the monosodio derivative of 4a was
more soluble and hence more likely to react a second
time. The simple replacement of THF by anhydrous
DMF led to homogeneous mixtures and produced a
reproducible, statistical distribution of products (1:1:2
3:1a :4a ) in near-theoretical yields. Chromatography led
to a 57% isolated yield of 4a based on converted 3, with
recycling of unreacted material. The aromatic region of
the 1H NMR spectrum of 4a showed the expected un-
symmetric pattern typical of a syn,anti conformer of an
in,out- (or 2′,1′′-) disubstituted product:3,4 One pyridine
doublet lay downfield of the pyridine H-4 triplet and was
assigned to the proton on the methylated indazole side,
where the indazole is anti to the pyridine, while the other
doublet lay upfield of the triplet and was attributed to
the pyridine proton on the nonmethylated side, which
experiences shielding from the nearby CH2 group on the
unsubstituted indazole, which itself is turned syn with
respect to the pyridine by virtue of an internal H-bond
involving the in- (or 2-)indazole tautomer and the pyri-
dine N. The occurrence of such H-bonding has recently
been confirmed in crystal structures,13 and the postulated
shielding in the syn conformation is supported by NOESY
experiments.5,14
In this paper, we report on the synthesis of unsym-
metrical, mono- and disubstituted varieties of 1 as
capping ligands and as precursors in the elaboration of
ditopic, bis(tridentate) ligands that are suitable for the
preparation of helical polynuclear complexes. This calls
forth the problem of monofunctionalization of difunctional
compounds, and our work includes the first known
instance of a monoarylation of a bis(azole).
Resu lts a n d Discu ssion
One design consideration in the binding of octahedral
metals to a necessarily unsymmetric, ditopic bis(triden-
tate) ligand is metal-centered chirality and the possible
formation of diastereomeric oligonuclear complexes. Ac-
cording to MM2-level calculations, CH2 spacers allow
complexation without undue steric hindrance and favor
the like (chiral racemic) form of any two-metal fragment
of the chain, while aliphatic spacers longer than CH2-
CH2 present no clear advantage over the shorter ones.
In fact, too much length and flexibility enables the
formation of unlike (or meso) fragments, which disrupt
helicity. Von Zalewski et al.11 showed that (CH2)6-linked
bipyridines could even wrap around a single RuII center.
We therefore chose to use CH2 linkers and targeted the
N-methyl-capped compound 5a as the first and simplest
example. Elguero et al.12 used CH2Cl2 as a source of CH2
bridges between pyrazoles under PTC conditions, albeit
without regioselectivity. In our experience, a metal ion
template in a homogeneous medium should provide the
desired selectivity2-4 in the coupling of two equivalents
of monofunctionalized precursors 4.
It has previously been shown that treatment of bis-
(pyrazolyl)pyridine 3 with NaH in THF, followed by
alkylating agents, leads exclusively to out,out- or 1′,1′′-
disubstituted bis(indazol-3-yl)pyridines.2,4 Thus, exhaus-
tive methylation of 3 (2.4 equiv of NaH/THF/room
temperature/2.3 equiv of CH3I) produced regiomerically
pure 1a in 60% isolated yield.4 Metal chelation was
apparently critical in directing the regiochemistry and
organic bases, such as DBU or Et3N, generated compli-
cated mixtures of out,out, in,out, and in,in regiomers.
With only 1 equiv of CH3I by the same procedure,
however, a nonstatistical product ratio was obtained in
which the dialkylated product 1a dominated, with the
balance consisting mostly of the starting material and
little of the desired 4a . Reactions in THF were not
The monofunctional 4a was converted to the symmetric
ditopic 5a in quantitative yield by a modification (NaH/
CH2Cl2) of the conditions of Elguero et al.9 Its structure
was determined by 1H- and 13C NMR, MS and elemental
(11) Hayoz, P.; von Zalewski, A.; Evans, S. H. J . Am. Chem. Soc.
1993, 115, 5111.
(12) Elguero, J .; Ochoa, C. J . Heterocycl. Chem. 1960, 25, 1797.
(13) Luo, Y.; Potvin, P. G. J . Coord. Chem. 1997, in press. Zadykow-
icz, J .; Potvin, P. G. Manuscript in preparation.
(14) Luo, Y., Ph.D. Thesis, York University, 1995.

Synthesis of Substituted Pyrazolylpyridine Ligands
J . Org. Chem., Vol. 63, No. 2, 1998 237
analysis. The presence of a CH2 bridge was indicated
by the appearance of new singlets in the 1H NMR
insoluble even in polar solvents (D2O, CD3CN, CD3-
SOCD3, CD3COCD3). We presume this to be highly
charged, polynuclear material of the type depicted in
Figure 1. The 1H NMR spectrum of the supernatant
revealed a number of species, including free ligand, and
the signals were not in exchange. The FAB-MS spec-
trum, however, revealed the presence of the 2:2 complex,
1
spectrum at 6.26 ppm and in the H-decoupled 13C NMR
spectrum at 61.07 ppm, both entirely in line with data
from similar formaminals.12 A DEPT experiment con-
firmed that the one new 13C peak was indeed owing to a
CH2 group. Since the indazoles’ CH2 and CH3 substitu-
ents are similar, the spectra show pseudo-symmetry. The
+
as the highest mass peaks corresponded to [(5a )Zn]2Br2
1
+
two partially overlapping pyridine H doublets (H-3 and
and [(5a )Zn]2
.
H-5) were clearly downfield of the H-4 doublet of dou-
blets, signaling anti pyrazole-pyridine bonds. Not sur-
prisingly, 5a fragmented easily in the mass spectrometer,
with two fragment peaks attributable to N-CH2 bond
cleavage.
Treatment of 5a in CDCl3 with an excess of anhydrous
sodium picrate (NaPic) apparently led to the 2:2 product
in a two-stage process. There was an initial, instanta-
neous uptake of the otherwise insoluble solid, with a
strong yellow coloration of the solution and the appear-
ance of a sharp singlet due to Pic- at 8.7 ppm.4 Signal
integration indicated an initial excess of 5a over Na+ but,
over the course of about 1 h with vigorous shaking, there
was a slower, second stage of uptake that eventually led
to a stable ratio of 5a to bound Na+ of approximately 1:1.
We interpret this two-stage process as an initial forma-
tion of [(5a )2Na]+ while 5a was in large excess over the
dissolved Na+, followed by further uptake and rearrange-
ment to [(5a )Na]22+. At all times, there was only one set
of pyridine signals, indicating fast ligand exchange, and,
as had occurred with ZnBr2, the (partially overlapped)
pyridine doublets shifted to positions upfield of the
triplet, again indicating rotation to syn,syn conformations
at the binding sites. The CH2 signal was broad but there
was no clear evidence of diastereotopicity, although the
rate of ligand exchange may have disallowed its observa-
tion. The FAB-MS spectrum revealed much fragmenta-
tion, but a peak corresponding to [(5a )Na]nn+ at m/z 701
supported our formulation.
N-Arylations of 3 are more difficult. The original
reaction4 with excess ethyl 4-fluorobenzoate and Na2CO3
in DMSO (150 °C/2 d) had produced the monoester 4b
as a side-product (20% isolated yield) in the synthesis of
1b (31% isolated yield). Several alternative procedures
were explored in model reactions of 3,5-dimethylpyrazole
with ethyl 4-fluorobenzoate, ethyl 4-nitrobenzoate, and
ethyl 2,4-dinitrobenzoate. We found that the 4-fluo-
robenzoate gave the cleanest reactions but the conditions
employed by J ameson et al.9 (K/diglyme/110-130 °C)
afforded only a very slow reaction with this reagent.
However, replacement of the solvent with DMSO after
deprotonation in diglyme gave access to higher temper-
atures without interference by dimsyl ion and afforded
the best yields. By this process, 3 and 2 equiv of ethyl
4-fluorobenzoate produced the di- and monoesters in 66%
and 20% isolated yields, respectively, after 4 d at 130-
150 °C. By the same procedure with 1 equiv of ethyl
4-fluorobenzoate over 2 d at a more modest temperature
(70 °C), we obtained 20% of 1b and 40% of 4b after
purification. This 2:1 product ratio signals equal rates
of reaction at each step, in this homogeneous environ-
ment at least.
To help resolve the partial overlaps, a CDCl3 solution
of 5a was treated with an excess of anhydrous ZnBr2.
This produced a symmetric spectrum with no overlap of
the pyridine doublets. These were upfield of the triplet,
the pattern typical of a syn,syn conformer, and thus
served to prove that the desired out,out substitution
pattern had been achieved. Although a FAB-MS spec-
trum of the product was uninformative, we conclude that
it was the dinuclear 1:2 species, 5a (ZnBr2)2 by elimina-
tion of possible alternatives. With a symmetric spectrum,
unsymmetric 1:1, 2:3, and higher stoichiometries can be
eliminated. Second, the presence of more than one unit
of 5a per metal, such as in the helical 2:2 complex, [(5a )-
Zn]2Br4, can be discounted for two reasons. First, ZnII
readily forms neutral pentacoordinate species, for in-
stance with terpyridine,15 whereas the formation of
4+
dicationic octahedral centers, such as in [(5a )Zn]2
,
would be disfavored in the solvent used. Second, one
would expect the N-CH3 and N-CH2 groups in [(5a )Zn]2-
Br4 to suffer interligand shielding by the ring current of
the neighboring ligand on the same metal, and for their
signals to therefore migrate upfield relative to the free
ligand positions. That phenomenon occurs in complexes
with suitable, orthogonally disposed ligands as in, for
instance, Ru(bipyridine)2L2+ species10,16 but not in com-
plexes lacking them such as Ru(terpyridine)(PMe3)2Cl+17
and Zn(bidentate)Cl2 species,3 where metal induction
causing downfield migrations is the dominant effect. This
is also true for the N-CH3 signals in Ru(1a )22+ and related
complexes.5,6 Instead, both the N-CH3 and N-CH2 signals
in the present case migrated downfield, consistent with
5a (ZnBr2)2. Finally, one might expect diastereotopic CH2
bridges in the helical [(5a )Zn]2Br4 due to chirality at the
metal centers, but none was evident in the present case.
The hydrolysis of 1b in TFA/H2O produced diacid 1c
as the TFA salt in 74% yield, after recrystallization. This
was a high melting solid that was difficult to dissolve
but which remained soluble upon concentration and so
was difficult to recrystallize. The similar TFA/H2O
hydrolysis of monoester 4b proceeded to give 4c in 82%
isolated yield without salt formation. Hydrolysis of 1b
or neutralization of 1c with Me4NOH gave a quantitative
crude yield of the H2O-soluble bis(tetramethylammo-
nium) salt 1d but all attempts to recrystallize this
Treatment of 5a with exactly 1.0 equiv of ZnBr2 led to
a mixture of products, including a precipitate that was
(15) Einstein, F. W. B.; Penfold, B. R. Acta Crystallogr. 1966, 20,
924.
(16) Steel, P. J .; Lahousse, F.; Lerner, D.; Marzin, C. Inorg. Chem.
1983, 22, 1488. Marzin, C.; Budde, F.; Steel, P. J .; Lerner, D. New J .
Chem. 1987, 11, 33.
(17) Leising, R. A.; Kubow, S. A.; Churchill, M. R.; Buttrey, L. A.;
Ziller, J . W.; Takeuchi, K. J . Inorg. Chem. 1990, 29, 1306.

238 J . Org. Chem., Vol. 63, No. 2, 1998
Zadykowicz and Potvin
material provided the H2O-insoluble free diacid 1c con-
taminated with variable amounts of Me4N+ salts.
EtOH,18 to produce the pure monoethyl ester 6a and pure
diethyl ester 6c in 61% and 71% overall yields, respec-
tively. In terms of flexibility, the mixed diester 6c
presents a compromise between the more rigid 1b and
the looser 1e.
With monoacid 4c in DMSO-d6, there was a severe
overlap of peaks in the aromatic region but the 2:4:2:8
distribution of aliphatic signals was consistent with the
syn,anti conformation expected of the nonzwitterionic
form of 4c, and was similar to the pattern with the parent
monoester 4b, also in the syn,anti form. Addition of
ZnBr2 produced the aromatic pattern typical of the syn,
syn conformer, just as it had with the diacid 1c. Treat-
One possible route to these side-products is the attack
of the benzoate ester’s ethyl group, in either the starting
material or in the desired product, by the X- (Br- or I-)
liberated during the alkylations to generate carboxylate
ions that then become otherwise O-alkylated. This would
also generate C2H5X species and one would therefore also
expect N-ethyl groupings in the product mixtures. Such
1
ment with TFA had the same result. H NMR analysis
of the ammonium salt 1d indicated the syn,syn orienta-
tion, just as with the bis(tetramethylammonium) salt of
the acetic acid analogue,4 which we had ascribed to either
the enforcement of this conformation by ionic interactions
or to the operation of new shielding effects. The 1H NMR
spectrum of salt 1c‚TFA in DMSO-d6 was deceptive as
the aromatic pattern indicated an anti,anti conformation
which is rather expected of the free diacid in nonzwitte-
rionic form. Treatment with CH3COOD or CF3COOD
caused no change, as expected, but an excess of ZnBr2
did produce the anticipated pattern indicative of com-
plexation of the syn,syn conformer.
1
groupings were indeed found in the H NMR spectra of
the crude reaction products in support of this explana-
tion, but we cannot exclude the intervention of adventi-
tious H2O, despite the care taken to avoid it, in also
producing carboxylate intermediates. To overcome this
problem and the necessity for a second operational step
in forming an N-alkyl monoester such as 6a , ethyl ester
4b was also alkylated with C2H5Br, leading directly to
the N,O-diethyl compound 6d in 92% isolated yield.
Despite the steric bulk of substituents in some cases,
many of these ligands have been successfully incorpo-
rated into homoleptic and heteroleptic, mononuclear and
dinuclear RuII and CoII complexes, and these will be
reported elsewhere.
Compound 4b was converted in quantitative yield to
5b using the same conditions that produced 5a . Unlike
with 5a , there was no overlap of pyridine 1H NMR signals
1
with 5b. Its H NMR, 13C NMR, and mass spectra clearly
supported the structural assignment and the melting
point was sharp. However, combustion analyses were
consistently low for all elements, though the C/N and H/N
ratios matched those expected.
Exp er im en ta l Section
THF was distilled over K and benzophenone, and DMSO
was dried over CaO, filtered, and distilled over molecular
sieves (5 Å) and then was frozen and stored in sealed vials
under Ar. All other solvents used were reagent grade and used
as such. The petroleum ether (PE) used was the light fraction
(bp 30-60 °C). The melting points are not corrected. NMR
spectra were obtained on a 400-MHz Bruker instument in
CDCl3, unless otherwise indicated. Acetone was used as an
internal reference for spectra in D2O. Mass spectroscopy was
carried out by Dr. B. Khouw on a Kratos Profile machine. The
microanalyses were perfomed by Guelph Chemical Laborato-
ries Ltd. of Guelph, Ontario, or National Chemical Consulting
Inc. of Tenafly, NJ .
2,6-Bis[1-[4-(et h oxyca r b on yl)p h en yl]-4,5,6,7-t et r a h y-
d r oin d a zol-3-yl]p yr id in e (1b). A mixture of bis(pyrazolyl)-
pyridine 34 (0.319 g, 1 mmol) and K metal (0.078 g, 2 mmol)
was stirred at 60 °C under Ar for 3 h in anhydrous diglyme.
After removal of the solvent, the yellow solid residue was
dissolved in dry DMSO, and diethyl 4-fluorobenzoate (0.673
g, 4 mmol) was added. The resulting mixture was stirred for
4 d at 150 °C. After removal of the solvent, the crude product
was taken up in CHCl3 and washed with H2O. The oily residue
was purified by column chromatography on alumina, using as
eluent AcOEt-PE (50:50). The first fraction (0.406 g, 66%)
contained diester 1b. Subsequent fractions, eluted by EtOAc,
contained monoester 4b (0.093 g, 20%). The spectra and
melting points of these materials were identical with those
previously reported.4
In view of the ease of monoalkylation of 3, compared
to its monoarylation, an alternative carboxyl-bearing
substituent was sought. Although more mobile than a
benzoic acid substituent, which provides a preorganized
and entropically advantaged site in diacid 1c for meth-
ylviologen binding, the more flexible p-toluic acid group
was chosen to possibly adjust and better accommodate
the guest in enthalpically favored binding. Further,
because we have found in other work6 that the nature of
the side-chain in RuII complexes of 1 has little effect on
the MLCT band positions and the CV oxidation waves,
the interposition of a CH2 group between the tridentate
core and the benzoic acid group was not expected to much
affect the ligand’s electronic properties. Thus, in analogy
to the preparation of monobenzoate 4a , a reaction of 3
with 1 equiv of methyl 4-(bromomethyl)benzoate pro-
duced the monotoluate 4d and the ditoluate 1e in 47%
and 25% isolated yields, respectively. The coproduct 1e
is interesting in its own right as a carboxyl-functionalized
ligand.
Because free NH sites are at times undesirable in
complexes (for instance, they are slower to complex and
strongly decrease the Ru3+/2+ oxidation potential5,6), the
monosubstituted compounds 4 to be used as monotopic
ligands were N-capped. Thus, monoester 4b was analo-
gously alkylated with CH3I and methyl 4-(bromometh-
yl)benzoate to produce the monoester 6a and mixed
diester 6b, respectively. However, analysis of the reac-
tion mixtures revealed new ester groups: ethyl ester 6a
was accompanied by a small amount of the corresponding
methyl ester while the preparation of the mixed ester 6b
produced a complex mixture presenting some 4-(ethoxy-
carbonyl)benzyl ester groupings. The crude reaction
products were therefore separately subjected to a sub-
sequent transesterification step, using DBU/LiCl in
2,6-Bis[1-(4-ca r boxyp h en yl)-4,5,6,7-tetr a h yd r in d a zol-
3-yl]p yr id in e (1c). A solution of 0.128 g (0.208 mmol) of
diester 1b in a mixture of CF3COOH (10 mL) and H2O (3 mL)
was heated to reflux for 24 h. Removal of solvent gave a yellow
solid residue, which was washed with water and CHCl3 and
then vacuum-dried. Recrystallization from THF gave 0.096 g
(66%) of the white solid diacid 1c as its TFA salt: mp > 300
1
°C; H NMR (DMSO-d6): δ 1.80 (m, 8H), 2.89 (m, 4H), 2.97
(m, 4H), 7.81 (d, 4H, J ) 8.5 Hz), 7.91 (t, 1H, J ) 7.9 Hz),
7.97 (d, 2H, J ) 7.95 Hz), 8.08 (d, 4H, J ) 8.5 Hz); 13C NMR:
δ 22.15, 22.35, 22.71, 23.60, 117.64, 120.01, 122.14, 128.66,
(18) Seebach, D.; Thaler, A.; Blaser, D.; Ko, S. Y. Helv. Chim. Acta
1991, 74, 1102.

Synthesis of Substituted Pyrazolylpyridine Ligands
J . Org. Chem., Vol. 63, No. 2, 1998 239
130.56, 137.16, 140.35, 142.83, 148. 86, 152. 56, 166.72 ppm;
MS m/z (%) 559 (67, M•+), 515 (15, M•+ - CO2), 467 (18), 439
(5, M•+ + H - C6H4COOH), 368 (5), 319 (5, M•+ - 2C6H4COO‚),
280 (11), 236 (15), 187 (10), 137 (37), 97 (54), 81 (81), 69 (100,
C3H7CN), 57 (74), 43 (65). Anal. Calcd for C33H29N5O4‚CF3
COOH‚H2O: C, 60.69; H, 4.80; N, 10.11. Found: C, 60.56; H,
4.92; N, 10.10.
Bis(tetr a m eth yla m m on iu m ) 2,6-Bis[1-(4-ca r boxyla to-
ph en yl)-4,5,6,7-tetr ah ydr oin dazol-3-yl]pyr idin e (1d). Fr om
1b. A mixture of 0.308 g (0.5 mmol) of diester 1b and Me4-
NOH‚5H2O (0.181 g, 1 mmol) in EtOH was heated to reflux
overnight. The volatiles were removed, and the crude yellow
solid was washed with CHCl3 and then vacuum-dried to leave
a quantitative yield of salt 1d in nearly pure form: mp 276-8
°C; 1H NMR (D2O): δ 1.33 (m, 8H), 2.34 (m, 4H), 2.47 (m, 4H),
3.13 (s, 24H), 7.28 (bd, 4H, J ) 8.2 Hz), 7.49 (bd, 2H), 7.52
(bt, 1H), 7.88 (d, 4H) ppm; 13C NMR (D2O): δ 23.17, 23.40,
23.76, 24.27, 56.32, 117.98, 121.34, 123.60, 131.21, 136.57,
138.32, 141.56, 141.82, 149.73, 152.90, 174.79 ppm; MS m/z
(%) 705 (0.1, M•+), 602 (1), 560 (3), 481 (6), 368 (42), 334 (6),
313 (8), 257 (25), 236 (46), 110 (82), 71 (100), 56 (85).
Calcd for C20H23N5‚0.5 H2O: C, 70.15; H, 7.06; N, 20.45.
Found: C, 70.78; H, 7.39; N, 20.66.
2-[1-[4-(Eth oxyca r bon yl)p h en yl]-4,5,6,7-tetr a h yd r oin -
d a zol-3-yl]-6-(2H -4,5,6,7-t e t r a h yd r oin d a zol-3-yl)p yr i-
d in e (4b). In the same fashion as for diester 1b, a 2 d
arylation in DMSO at 70 °C, using bis(pyrazolyl)pyridine 3
(0.106 g, 0.33 mmol), K metal (0.028 g, 0.716 mmol), and ethyl
4-fluorobenzoate (0.061 g, 0.36 mmol), gave two products.
Isolation as before provided diester 1b (0.040 g, 20%) and the
desired monoester 4b (0.062 g, 40%). The spectra and melting
points of these materials were identical with those previously
reported.4
2-[1-(4-Ca r b oxyp h en yl)-4,5,6,7-t et r a h yd r oin d a zol-3-
yl]-6-(2H-4,5,6,7-tetr a h yd r oin d a zol-3-yl)p yr id in e (4c). As
for diacid 1c, 0.234 g (0.5 mmol) of monoester 4b gave 0.160
g (73%) of white solid acid 4c, after recrystallization from
THF: mp > 300 °C; 1H NMR (DMSO-d6); δ 1.78 (m, 8H), 2.65
(m, 2H), 2.89 (m, 4H), 3.10 (m, 2H), 7.80 (m, 3H), 7.90 (m,
2H), 8.09 (d, 2H, J ) 8.5 Hz); 13C NMR: δ 20.79, 21.60, 22.38,
22.54, 23.21, 23.63, 113.67, 114.76, 117.81, 119.36, 122.02,
122.29, 124.55, 128.62, 130.472, 130.71, 140.33, 143.82, 148.90,
152.52, 158.59, 166.76 ppm; MS m/z (%) 439 (100, M•+), 410
(20), 395 (10, M•+ - CO2), 303 (4), 256 (4), 241 (9), 220 (19),
170 (13), 149 (16), 129 (19), 95 (16), 81 (37), 69 (70, C3H7CN),
55 (27, C2H5CN), 41 (24). Anal. Calcd for C26H25N5O2: C,
71.05; H, 5.73; N, 15.93. Found: C, 70.65; H, 6.10; N, 15.52.
2-[1-[[4-(Meth oxyca r bon yl)p h en yl]m eth yl]- 4,5,6,7-tet-
r a h yd r oin d a zol-3-yl]-6-(1H-4,5,6,7-tetr a h yd r oin d a zol-3-
yl)p yr id in e (4d ). A mixture of bis(pyrazolyl)pyridine 3 (0.160
g, 0.5 mmol) and NaH (0.014 g, 0.6 mmol) was stirred under
Ar in anhydrous DMF at room temperature for 2 h until the
solution became clear and the evolution of H2 had stopped.
The mixture was treated with methyl 4-(bromomethyl)ben-
zoate (0.115 g, 0.5 mmol) and allowed to stir under Ar at room
temperature for 24 h. The solvent was removed, and the
resulting oil was washed with H2O and extracted into CHCl3.
Further purification was carried out by column chromatogra-
phy on alumina, using MeOH:EtOAc (2:98) as eluent. After
liberation from solvents, the first isolated fraction yielded the
disubstituted diester 1e as a white solid (0.080 g, 26%), and
the second fraction, a clear oil, was the desired monoester 4d .
This was crystallized from EtOAc:PE to yield a white solid
(0.110 g, 47%), mp: 105-107 °C; 1H NMR: δ 1.80 (m, 4H),
1.90 (m, 4H), 2.50 (t, 2H, J ) 5.6 Hz), 2.78 (m, 2H), 2.87 (m,
2H), 3.05 (t, 2H, J ) 5.4 Hz), 3.92 (s, 3H), 5.38 (s, 2H), 7.22 (d,
2H, J ) 7.96 Hz), 7.41 (d, 1H, J ) 7.70 Hz), 7.76 (dd, 1H),
7.92 (d, 1H, J ) 7.90 Hz), 8.00 (d, 2H, J ) 8.05 Hz) ppm; 13C
NMR: δ 21.73, 22.36, 22.41, 23.01, 23.10, 23.31, 23.52, 23.69,
52.13, 52.76, 113.61, 116.80, 117.74, 119.27, 126.80, 129.61,
130.10, 137.05, 138.02, 140.38, 142.05, 146.96, 147.96, 150.60,
153.74, 166.72 ppm; MS m/z (%) 467 (21, M•+), 448 (4), 411
(3), 368 (5), 318 (13, M•+ - CH2C6H4COOCH3), 301 (3), 284
(5), 249 (23), 2218 (17), 181 (18), 167 (12), 137 (48), 111 (57),
91 (96), 81 (84), 69 (96, C3H7CN), 55 (85), 43 (100). Anal. Calcd
for C28H29N5O2‚0.5 H2O: C, 70.57; H, 6.34; N, 14.70. Found:
C, 70.45; H, 6.15; N, 14.25.
Fr om 1c. An EtOH solution of diacid salt 1c‚CF3COOH‚H2O
(0.063 g, 0.091 mmol) was treated with Me4NOH‚5H2O (0.055
g, 0.303 mmol). Workup as above provided nearly pure 1d
(0.060 g, 93%).
2,6-Bis[1-[[4-(m eth oxyca r bon yl)p h en yl]m eth yl]-4,5,6,7-
tetr a h yd r oin d a zol-3-yl]p yr id in e (1e). Solid NaH (0.053 g,
2.2 mmol) was added to a solution of 0.319 g (1 mmol) of bis-
(pyrazolyl)pyridine 3 in dry THF. H2 evolution was immediate,
and the solution became milky-white. The mixture was stirred
under Ar for 2 h and then treated with methyl 4-(bromo-
methyl)benzoate (0.504 g, 2.2 mmol). After heating to reflux
overnight, THF was removed, and the light yellow solid residue
was dissolved in CHCl3 and washed with water. After the
removal of CHCl3, the resulting oil was purified by column
chromatography on alumina, using EtOAc as eluent, to yield
the solid diester 1e (0.536 g, 87%), mp: 147-149 °C; 1H
NMR: δ 1.79 (m, 8H), 2.47 (t, 4H, J ) 5.90 Hz), 2.99 (t, 4H, J
) 5.75 Hz), 3.90 (s, 6H), 5.35 (s, 4H), 7.20 (d, 4H, J ) 8.01
Hz), 7.70 (t, 1H, J ) 7.97 Hz), 7.86 (d, 2H, J ) 7.85 Hz), 7.98
(d, 4H, J ) 8.50 Hz) ppm; 13C NMR: δ 21.73, 22.43, 22.97,
23.23, 52.10, 52.67, 116.47, 119.44, 126.83, 129.49, 130.03,
136.48, 139.83, 142.32, 148.00, 153.33, 166.78 ppm; MS m/z
(%) 616 (10, M•+ + H), 524 (5), 466 (5, M•+ - CH2C6H4-
COOCH3), 336 (44), 350 (20), 313 (13), 257 (16), 236 (28), 218
(59), 129 (34), 97 (82), 83 (83), 43 (100). Anal. Calcd for
C
37H37N5O4: C, 72.18; H, 6.06; N, 11.37. Found: C, 72.19; H,
6.00; N, 11.29.
2-(1-Meth yl-4,5,6,7-tetr ah ydr oin dazol-3-yl)-6-(2H-4,5,6,7-
tetr a h yd r oin d a zol-3-yl)p yr id in e (4a ). Solid NaH (0.030 g,
1.26 mmol) was added to a solution of 0.190 g (0.6 mmol) of
bis(pyrazolyl)pyridine 3 in anhydrous DMF. H2 evolution was
immediate, and the solution became yellow. The mixture was
stirred under Ar for 2 h and then treated with CH3I (0.085 g,
0.6 mmol). After stirring overnight at room temperature, DMF
was removed, and the light yellow solid residue was dissolved
in CHCl3 and washed with water. After the removal of CHCl3,
the resulting oil was purified by column chromatogrsphy on
alumina, using EtOAc as eluent. The first fraction was the
N,N′-dimethyl compound 1a (0.050 g, 24%), whose spectra and
melting point were identical with those previously reported.4
Subsequent fractions, eluted with MeOH:EtOAc (1:99), con-
tained the desired N-monomethyl compound 4a (0.096 g, 48%)
and unreacted bis(pyrazolyl)pyridine 3 (0.031 g, 16%). Re-
crystallization of crude 4a from EtOAc:PE gave a white solid,
mp: 130-132 °C; 1H NMR: δ 1.79 (m, 4H), 1.89 (m, 4H), 2.62
(t, 2H, J ) 6.2 Hz), 2.76 (m, 2H), 2.84 (m, 2H), 2.99 (t, 2H, J
) 5.75 Hz), 3.81 (s, 3H), 7.38 (d, 1H, J ) 7.7 Hz), 7.73 (dd,
1H), 7.83 (d, 1H, J ) 7.86 Hz) ppm; 13C NMR: δ 21.69, 22.43,
23.02, 23.16, 23.28, 23.52, 23.70, 35.20, 108.20, 113.53, 116.00,
117.53, 118.96, 136.99, 137.60, 140.28, 147.20, 148.80, 153.80
ppm; MS m/z (%) 333 (62, M•+), 318 (7, M•+ - CH3), 304 (13),
279 (16), 256 (14), 241 (14), 213 (14), 185 (14), 149 (42), 129
(84), 111 (30), 81 (63), 69 (100, C3H7CN), 57 (39), 43 (23). Anal.
Bis[3-[2-(1-m eth yl-4,5,6,7-tetr ah ydr oin dazol-3-yl)pyr id-
6-yl]-4,5,6,7-tetr a h yd r oin d a zol-1-yl]m eth a n e (5a ). A mix-
ture of mono-N-methyl derivative 4a (0.040 g, 0.12 mmol) and
NaH (0.009 g, 0.38 mmol) was stirred under Ar in CH2Cl2 at
room temperature for 2 h until evolution of H2 stopped. The
mixture was heated to reflux overnight and then washed with
H2O. After separation of the organic phase and removal of
solvent, the clear oily residue was purified by column chro-
matography on alumina using EtOAc as the eluent. The white
solid residue was subsequently recrystallized from MeOH,
yielding 0.038 g (93%) of ditopic ligand 5a as white solid, mp
1
238-240 °C; H NMR: δ 1.78 (m, 8H), 1.89 (m, 8H), 2.63 (m,
6H), 2.95 (m, 10H), 3.81 (s, 6H), 6.26 (s, 2H), 7.70 (dd, 2H),
7.83 (d, 4H) ppm; 13C NMR: δ 21.72, 21.88, 22.43, 22.52, 22.87,
22.93, 23.11, 23.28, 35.75, 61.07, 115.67, 116.73, 118.93,
119.22, 136.41, 139.78, 140.83, 148.49, 153.00, 153.28, 153.44
ppm; MS m/z (%) 679 (6, M•+), 604 (6), 578 (3), 538 (4), 524
(6), 510 (4), 403 (5), 346 (28, M•+ - C20H22N5 (4a - H)), 333
(46, C20H22N5 (4a - H)), 264 (16), 236 (22), 211 (16), 135(23),

240 J . Org. Chem., Vol. 63, No. 2, 1998
Zadykowicz and Potvin
109 (38), 97 (61), 83 (73), 69 (86, C3H7CN), 55 (100 C2H5CN),
43 (89, C3H7). Anal. Calcd for C41H46N10: C, 72.54; H, 6.83;
N, 20.63. Found: C, 72.06; H, 6.84; N, 20.54.
mmol) was stirred under Ar in dry THF for 2 h until the
solution became milky white from formation of the Na+ disalt.
This mixture was treated with methyl 4-(bromomethyl)-
benzoate (0.0.041 g, 0.18 mmol) and stirred overnight. The
solvent was removed under reduced pressure, and the oily
residue was washed with H2O and extracted into CHCl3. The
1H NMR spectrum of the crude product showed that undesir-
able transesterification had occurred. Thus, retro-transes-
terification was performed in refluxing EtOH with DBU and
LiCl.18 After 16 h, the solvent was removed, and the crude
oily residue was washed with dilute HCl solution and extracted
into CHCl3. After removal of solvent, the resulting oil was
purified by column chromatography on alumina with EtOAc:
PE (50:50) as eluent. Recrystallization from EtOAc yielded
white crystals of diethyl ester 6c (0.080 g, 71%), mp 199-201
°C; 1H NMR: δ 1.40 (t, 3H, J ) 6.94 Hz), 1.45 (t, 3H, J ) 6.96
Hz), 1.78 (m, 3H), 1.90 (m, 5H), 2.51 (m, 2H), 2.81 (m, 2H),
3.02 (m, 2H), 3.10 (m, 2H), 4.39 (q, 2H, J ) 6.92 Hz), 4.43 (q,
2H, J ) 6.96 Hz), 5.38 (s, 2H), 7.24 (d, 2H, J ) 8.09 Hz), 7.74
(d, 2H, J ) 8.58 Hz), 7.77 (d, 1H, J ) 7.86 Hz), 7.94 (d, 1H, J
) 7.71 Hz), 8.00 (m, 3H), 8.17 (d, 2H J ) 8.72 Hz) ppm; 13C
NMR: δ 14.38, 14.74, 21.74, 22.42, 22.93, 22.98, 23.05, 23.22,
24.17, 24.55, 52.73, 61.02, 61.15, 116.51, 118.52, 119.84,
119.98, 122.21, 126.84, 128.21, 129.87, 130.03, 130.66, 136.65,
139.95, 140.03, 142.10, 143.65, 147.81, 150.08, 152.80, 153.28,
166.06, 166.32 ppm; MS m/z (%): 629 (31, M•+), 509 (48), 481
(100, M•+ - C6H4COOCH2CH3), 452 (31, M•+ + H - C6H4-
COOCH2CH3 - CH2CH3), 408 (13, M•+ + H - C6H4COOCH2-
CH3 - COOCH2CH3), 353 (33), 313 (34), 239 (54), 213 (48),
185 (37), 135 (61), 96 (81), 69 (89, C3H7CN), 56 (76, C4H8). Anal.
Calcd for C38H39N5O4:C, 72.48; H, 6.24; N, 11.12. Found: C,
72.25; H, 6.20; N, 10.75.
Bis[3-[2-[1-[4-(eth oxyca r bon yl)p h en yl]-4,5,6,7-tetr a h y-
d r oin d a zol-3-yl]p yr id -6-yl]-4,5,6,7-tetr a h yd r oin d a zol-1-
yl]m eth a n e (5b). Using the same procedure as for 5a ,
monoester 4b (0.040 g, 0.086 mmol) and NaH (0.006 g, 0.26
mmol) in CH2Cl2 provided ditopic ligand 5b in a quantitative
yield as a white powder, after column chromatography on
alumina with 1:1 EtOAc:PE: mp 267-269 °C; 1H NMR: δ 1.44
(t, 6H, J ) 7.2 Hz), 1.78 (m, 4H), 2.00 (m, 12H), 2.86 (m, 4H),
2.96 (m, 4H), 3.02 (m, 4H), 3.08 (m, 4H), 4.43 (q, 4H, J ) 7.1
Hz), 6.29 (s, 2H), 7.73 (d, 4H, J ) 8.6 Hz), 7.78 (t, 2H, J ) 7.8
Hz), 7.89 (d, 2H, J ) 7.7 Hz), 8.02 (d, 2H, J ) 7.7 Hz), 8.17 (d,
4H, J ) 8.7 Hz) ppm; 13C NMR: δ 14.36, 21.88, 22.40, 23.03,
23.11, 24.52, 61.00, 61.12, 116.70, 118.55, 119.91, 119.96, 122.
19, 128.20, 130.63, 136.50, 140.02, 140.91, 143.63, 148.42,
149.96, 152.91, 153.10, 166.03 ppm; MS m/z (%) 798 (0.02, M•+
- C6H4COOEt), 647 (0.05, M•+ - 2C6H4COOEt), 480 (0.4, M•+
- C20H22N5 (4b - H)), 466 (0.4, C20H22N5 (4b - H)), 386 (4),
368 (7), 149 (49, C6H4COOEt), 43 (100). Anal. Calcd for
C
57H58N10O4: C, 72.28; H, 6.17; N, 14.79. Found: C, 60.78;
H, 5.09; N, 12.27.
2-(1-Me t h yl-4,5,6,7-t e t r a h yd r oin d a zol-3-yl)-6-[1-[4-
(eth oxyca r bon yl)p h en yl]-4,5,6,7-tetr a h yd r oin d a zol-3-yl]-
p yr id in e (6a ). F r om 4b. Solid NaH (0.003 g, 0.128 mmol)
was added to a solution of 0.050 g (0.11 mmol) of monoester
4b in dry THF. H2 evolution was immediate, and a white
precipitate of the Na+ salt was formed. The mixture was kept
under Ar for 2 h and then treated with CH3I (0.016 g, 0.11
mmol). After stirring overnight at room temperature, THF
was removed, and the white solid residue was dissolved in
CHCl3 and washed with water. After the removal of CHCl3,
the resulting oil was purified by column chromatography on
1
alumina using EtOAc:PE (50:50), but, according to its H NMR
spectrum, this was an inseparable mixture of methyl and ethyl
esters. Retro-transesterification in EtOH with DBU and LiCl18
was performed. After 24 h reflux, EtOH was removed and
the remaining yellow oil was washed with acidified (HCl) H2O
and extracted into CHCl3. Removal of solvent afforded the
white solid N-methyl monoester 6a (0.032 g, 60%), mp 166-
168 °C; 1H NMR: δ 1.42 (q, 3H, J ) 7 Hz), 1.77 (m, 3H), 1.88
(m, 5H), 2.63 (m, 2H), 2.85 (m, 2H), 2.96 (m, 2H), 3.06 (m,
2H), 3.81 (s, 3H), 4.4 (t, 2H, J ) 7 Hz), 7.7 (d, 2H, J ) 8.7 Hz),
7.74 (d, 1H, J ) 7.8 Hz), 7.85 (d, 1H, J ) 7.7 Hz), 7.95 (d, 1H
J ) 7.7 Hz), 8.14 (d, 2H, J ) 8.7 Hz) ppm; 13C NMR: δ 14.39,
21.72, 22.50, 22.94, 22.97, 23.02, 23.29, 24.55, 35.77, 61.14,
115.67, 118.54, 119.62, 119.68, 122.18, 128.15, 130.65, 136.54,
139.88, 139.98, 143.66, 147.15, 150.11, 152.77, 153.41, 166.08
ppm; MS m/z (%) 481 (100, M•+), 466 (7, M•+ - CH3), 452 (23,
M•+ - CH2CH3), 436 (23), 408 (16), 332 (4, M•+ - C6H4-
COOCH2CH3), 318 (7, M•+ + H - C6H4COOCH2CH3 - CH3),
269 (13), 237 (16), 223 (13), 204 (10), 185 (16), 170 (21), 155
(11), 142 (14), 135 (32), 103 (13), 77 (18), 65 (12). Anal. Calcd
for C29H31N5O2: C, 72.31; H, 6.49; N, 14.55. Found: C, 72.37;
H, 6.69; N, 14.26.
F r om 4a . A mixture of mono-N-methyl compound 4a (0.029
g, 0.087 mmol) and K metal (0.004 g, 0.102 mmol) was stirred
at 70 °C under Ar for 12 h in anhydrous diglyme. A slow
evolution of H2 was observed, and the white solid K+ salt was
formed. After removal of the solvent, the light yellow, solid
residue was dissolved in dry DMSO, and diethyl 4-fluoroben-
zoate (0.023 g, 0.137 mmol) was added. The resulting mixture
was stirred for 4 d at 140 °C. After removal of the solvent,
the crude product was taken up in CHCl3 and washed with
H2O/NaCl. The 1H NMR analysis of the crude oil indicated
some hydrolysis. Thus, esterification was carried out in
refluxing triethyl orthoformate for 24 h. After removal of the
volatiles under reduced pressure, the resulting oil was purified
as above, and a white powder was obtained (0.013 g, 31%)
whose spectra and melting point matched those described.
2-[1-[4-(Eth oxyca r bon yl)p h en yl]-4,5,6,7-tetr a h yd r oin -
d a zol-3-yl]-6-[1-[[4-(e t h oxyca r b on yl)p h e n yl]m e t h yl]-
4,5,6,7-tetr a h yd r oin d a zol-3-yl]p yr id in e (6c). A mixture of
monoester 4b (0.085 g, 0.18 mmol) and NaH (0.008 g, 0.33
2-(1-E t h y l-4,5,6,7-t e t r a h y d r o in d a zo l-3-y l)-6-[1-[4--
(et h oxyca r b on yl)p h en yl]-4,5,6,7-t et r a h yd r oin d a zol-3-
yl]p yr id in e (6d ). A 0.034 g sample of monoester 4b (0.073
mmol) was dissolved in 10 mL of C2H5Br. Then 0.004 g (0.17
mmol) of NaH was added, and the mixture was stirred at room
temperature until the H2 evolution ceased. Then the solution
was refluxed under Ar for 1 h. The solvent was removed, and
the resulting oil was washed with H2O and extracted into
CHCl3. The crude oily residue was purified by column chro-
matography on alumina using EtOAc:PE (50:50) to afford a
white solid 6d (0.033 g, 92%). Further purification was carried
out by the diffusion of Et2O vapor into a CHCl3 solution; this
1
yielded a white powder, mp 152-4 °C; H NMR: δ 1.46 (m,
6H), 1.80 (m, 2H), 1.90 (m, 6H), 2.66 (m, 2H), 2.87 (m, 2H),
2.99 (m, 2H), 3.08 (m, 2H), 4.13 (q, 2H, J ) 7.4 Hz), 4.43 (q,
2H, J ) 7.5 Hz), 7.74 (m, 3H), 7.89 (d, 1H, J ) 7.7 Hz), 7.97
(d, 1H, J ) 7.7 Hz), 8.16 (d, 2H, J ) 8.6 Hz) ppm; 13C NMR:
δ 14.38, 15.62, 20.14, 20.67, 21.41, 21.70, 22.57, 22.92, 23.04,
23.40, 24.55, 43.85, 61.14, 115.58, 118.56, 119.54, 119.81,
122.19, 128.15, 130.65, 130.90, 136.51, 138.87, 139.98, 143.68,
147.16, 150.16, 152.74, 153.60, 166.10 ppm; MS m/z (%) 495
(100, M•+), 480 (52, M•+ - CH3), 465 (50, M•+ - 2CH3), 450
(22, M•+ - H - CH2CH3 - CH3), 420 (44), 405 (87), 375 (90),
346 (86, M•+ - C6H4COOCH2CH3), 332 (35, M•+ + H - C6H4-
COOCH2CH3 - CH3), 318 (7, M•+ + H - C6H4COOCH2CH3
-
CH2CH3), 248 (85), 188 (52), 149 (64, C6H4COOCH2CH3), 124
(68), 111 (69), 98 (59), 85 (68), 71 (73), 57 (87), 44 (51, CO2).
Anal. Calcd for C30H33N5O2‚0.4CHCl3: C, 67.20; H, 6.20; N,
12.89. Found: C, 67.16; H, 6.38; N, 12.69.
Ack n ow led gm en t. The authors thank the Natural
Sciences and Engineering Research Council of Canada
for funding.
J O970815Z