Polymer-supported quenching reagents for parallel purification

Article abstract of DOI:10.1021/ja9634637

The preparation of polystyrene-divinylbenzene-supported derivatives of tris(2-aminoethyl)amine and methyl isocyanate are described. These polymeric reagents are used to quench excess reactants and remove known impurities from the crude reaction products o

Full text of DOI:10.1021/ja9634637

4882
J. Am. Chem. Soc. 1997, 119, 4882-4886
Polymer-Supported Quenching Reagents for Parallel
Purification
R. John Booth and John C. Hodges*
Contribution from Parke-DaVis Pharmaceutical Research, 2800 Plymouth Road,
Ann Arbor, Michigan 48105
ReceiVed October 3, 1996X
Abstract: The preparation of polystyrene-divinylbenzene-supported derivatives of tris(2-aminoethyl)amine and methyl
isocyanate are described. These polymeric reagents are used to quench excess reactants and remove known impurities
from the crude reaction products obtained from the solution-phase, parallel syntheses of ureas, thioureas, sulfonamides,
amides, and pyrazoles. In conjunction with the use of other polymeric reactants during the course of a reaction, the
addition of polymer-supported quench reagent(s) at the conclusion of the reaction allows isolation of the desired
product by a single filtration and evaporation of solvent. The mechanical simplicity and efficiency of this methodology
make possible the rapid, parallel purification of crude reaction products obtained via solution-phase syntheses, regardless
of whether the intended product is a single compound or a mixture of compounds, and hence offers an attractive
alternative to solid-phase organic synthesis in the practice of combinatorial chemistry.
Introduction
ability to separate a structurally diverse array of undesired
starting materials from a structurally diverse array of desired
products based upon physical properties such as solubility or
partition coefficient presents a difficult problem, especially if
one wants to maintain the equimolar ratio of desired products.
Solid-supported synthesis offers a practical solution to
purification for both parallel and batch modes of combinatorial
synthesis. The ability to remove excess reagents from solid-
supported products by filtration and rinsing of the support
dramatically reduces the time allotted to purification in parallel
synthesis and makes possible the separation of complex mixtures
associated with batch mode synthesis. The price paid for this
convenient purification is the time and effort necessary to
develop a solid-phase synthetic route to the molecules of interest.
On the other hand, solution-phase synthesis has numerous
benefits over solid-phase synthesis: A great many more
solution-phase reactions have been optimized and documented
in the literature than are currently available to solid-phase
synthesis. Reaction progress as well as identity and purity of
products may be analyzed by well established chromatographic
and spectroscopic means. A large variety of protecting group
reagents are available at reasonable cost whereas only a limited
variety of relatively expensive solid-phase synthesis resins, often
strategically employed as solid-phase protecting reagents in
anchoring of starting materials, are commercially available. Last,
but not least, there is no need for resin attachment and cleavage
steps. Recently, a number of strategies for utilizing the benefits
of solution-phase synthesis in combinatorial chemistry have been
reported.5
The recent medicinal chemistry literature has shown a
dramatic rise in the interest of utilizing combinatorial chemistry
and automation to increase the rate at which novel drug
candidates may be prepared.1 By and large, this interest has
grown out of the pioneering peptide library work of Geysen2
and Houghten.3 The efficiency of preparing combinatorial
peptide libraries via solid-phase synthesis has increased the
popularity of research upon solid-phase organic synthesis
methods that are applicable to libraries of low molecular weight,
nonoligomeric, druglike molecules.4
It is commonly argued that the major impediment to parallel,
solution-phase synthesis of large numbers of individual organic
molecules is the time and effort required for purification of the
reaction products at each synthetic step. Furthermore, if one
approaches combinatorial chemistry from the standpoint of
intentionally synthesizing compound mixtures in solution, the
alternatives for enhancing purity of that mixture are extremely
limited. The problem associated with purification of a mixture
is further exacerbated by the fact that it is desirable to produce
library mixtures consisting of structurally diverse molecules
which are present in equal molar quantities in order to increase
the likelihood of finding a biologically active compound upon
in Vitro screening of the mixture and deconvolution. Particularly
in the cases where a large number of diverse products are
involved, i.e. 100 to 1000 compounds, methods that are well
established for the purification of single compounds such as
crystallization, extraction, and flash chromatography will not
be applicable. Even if the crude reaction product is a relatively
simple mixture, consisting only of an equimolar mixture of
desired products and lesser quantities of starting materials, the
In our experience, the reasons for choosing to prepare
combinatorial libraries (with or without automation) can gener-
ally be divided by two purposes. The first is the preparation
of lead generation libraries for high volume screening. In this
case, the time investment required to develop a robust solid-
phase synthesis (several months) can be recouped if one makes
thousands of compounds via mix and split synthesis strategies.
X Abstract published in AdVance ACS Abstracts, May 15, 1997.
(1) Recently reviewed: (a) Gallop, M. A.; Barrett, R. W.; Dower, W.
J.; Fodor, S. P. A.; Gordon, E. M. J. Med. Chem. 1994, 37, 1233. (b) Gordon,
E. M.; Barrett, R. W.; Dower, W. J.; Fodor, S. P. A.; Gallop, M. A. J.
Med. Chem. 1994, 37, 1385. (c) Czarnik, A. Chemtracts: Org. Chem. 1995,
8, 13.
(2) Geysen, M. H.; Meloen, R. H.; Barteling, S. J. Proc. Natl. Acac.
Sci. U.S.A. 1984, 81, 3998.
(3) Houghten, R. A. Proc. Natl. Acac. Sci. U.S.A. 1985, 82, 5131.
(4) Recently reviewed: (a) Terrett, N. K.; Gardner, M.; Gordon, D. W.;
Kobylecki R. J.; Steele, J. Tetrahedron 1995 51, 8135. (b) Thompson, L.
A.; Ellman, J. A. Chem. ReV. 1996, 96, 555.
(5) (a) Carrell, T.; Wintner, E. A.; Bashir-Hashemi, A.; Rebek, Jr., J.
Angew. Chem., Int. Ed. Engl. 1994, 33, 2059. (b) Boger, D. L.; Tarby, D.
M.; Myers, P. L.; Caporale, L. H. J. Am. Chem. Soc. 1996, 118, 2109. (c)
Curran, D. P. Chemtracts: Organic Chemistry 1996, 9, 75. (d) Curran, D.
P.; Hoshino, M. J. Org. Chem. 1996, 61, 6480. (e) Keating, T. A.;
Armstrong, R. W. J. Am. Chem. Soc. 1996, 118, 2574.
S0002-7863(96)03463-4 CCC: $14.00 © 1997 American Chemical Society

Polymer-Supported Quenching Reagents
J. Am. Chem. Soc., Vol. 119, No. 21, 1997 4883
aminomethyl resin at rT. In this case approximately 1 mmol
of NCO/g of 2 is the highest loading that is possible if one
starts with commercially available aminomethyl resins (e1.1
mmol of N/g). A strong IR absorption at 2260 cm-1 is
observed, indicative of the resin-bound isocyanate. Loading is
verified by reaction with excess 4-bromobenzylamine and Br
analysis of the resulting polymer-supported urea. The polymeric
base 3 is prepared by treating Merrifield resin (4.3 mmol of
Cl/g) with morpholine (266 mol %) in DMF (65 °C, 6 h).8 In
this case cross-linking is not an issue, and elemental analysis is
consistent with quantitative displacement of Cl by morpholine.
Figure 2 shows selected examples of reactions wherein the
resins 1-3 are employed. The first two examples illustrate an
important advantage of PSQ methods. One can choose which
reagent to use in excess and adjust the quenching resin
accordingly. Thus reaction of excess 4 with 5, followed by
quenching with 1, affords 6 upon filtration and evaporation of
solvent. Alternatively, excess 5 may be used, followed by
quenching with 2. Completion of both the reaction and the
quenching process can be monitored by TLC. 1H-NMR and
HPLC analyses show 6, produced by either method, to be of
excellent purity. In a closely related reaction, 9 is prepared
from 7 and excess 8, with quenching by 1. These means of
preparation of ureas and thioureas have been widely applied to
compounds of proprietary interest in our laboratories with equal
results. The only complication encountered was the occasional
precipitation of the desired product, in which cases a suitable
solvent must be added to dissolve the product before filtration
of the resin.
PSQ is also applicable to the synthesis of sulfonamide 12
and amide 15. In these cases, 3 is employed as a base during
the reaction. It is not necessary to remove 3 before adding the
quenching resin. Worth noting is the synthesis of 15, wherein
excess 14 is reacted with 13 in the presence of 3. TLC
monitoring of this reaction shows a small amount of 4-meth-
ylbenzoic acid, presumably an impurity in the starting acid
chloride. Therefore, both 1 and 2 were added in the quench
step. The excess amine is removed by covalent reaction with
2, and the acid impurity is removed by salt formation with 1 (3
is insufficiently basic to completely remove the benzoic acid
in addition to the HCl produced). Alternatively, one could
consider using a mixture of acidic and basic resins to remove
the excess amine and benzoic acid impurities, respectively.
Experience has taught us, however, that removal of impurities
by covalent binding to a resin is typically more reliable than an
acid-base equilibrium and thus requires less experimentation
to choose the amount of resin that ensures a quantitative removal
of impurities. Hence we employ the isocyanate resin 2 as a
first resort to remove amine starting materials so long as there
are no other groups in the products that would also react
inappropriately. The purification of 15 thus illustrates another
important advantage of the PSQ method. A multiplicity of
quenching reagents may be added concurrently to remove a
multiplicity of impurities and excess reagents since reactive
groups on separate polymer-supports are known not to interact.9
In principle, so long as one understands what impurities are in
the crude product, one can devise the necessary PSQ reagents
to selectively remove them and still maintain a single filtration
workup.
Figure 1. Polymeric reagents.
The second purpose is the optimization of an existing lead. If
the existing lead came from a library that was prepared by solid-
phase synthesis, there is good reason to try and optimize its
properties by making analogous compounds with the same
synthesis. On the other hand, many leads are found by screening
historical collections of compounds. In such circumstances,
parallel solution-phase synthesis becomes the method of choice
for two reasons: (1) A solid-phase synthetic route is not
available and the time necessary to devise one is inconsistent
with the relatively small number of compounds (hundreds) one
generally needs to transform a mass screening lead to a drug
candidate. (2) The purpose for any individual synthetic array
is typically unidimensional, i.e. a single starting material is
modified with a collection of synthons to derive a structure-
activity relationship at one position of the molecule. In such a
unidimensional array, mix and split synthesis strategies cannot
be used to reduce the number of reactions.
Results and Discussion
Recognizing two major advantages of solid-phase synthesis,
we set out to devise a method whereby (1) excess reagent could
be similarly employed to drive solution-phase reactions to
completion and (2) the reaction workup would be simplified to
a single filtration and evaporation of solvent at each synthetic
step. We have found that polystyrene beads bearing reactive
groups which mimic the limiting reagent of a reaction can be
used to remove the remainder of excess reagent(s) from crude
product solutions and are thereby convenient and effective tools
for performing rapid purifications.6 The resulting products may
or may not meet conventional purity standards by elemental
analysis. However their purity, as determined by HPLC and
1H-NMR, is sufficient for routine in Vitro biological testing.
Described below are polymeric reagents and reactions which
illustrate the polymer-supported quench (PSQ) method of
purification.
Three polymeric reagents which are readily made in one step
from commercially available polymers are shown in Figure 1.
Contrary to polymer-supported synthesis, it is advantageous to
have as high a loading of functionality as possible in PSQ.
Hence the polyamine resin 1 is prepared by heating Merrifield
resin with tris(2-aminoethyl)amine (400 mol %) in DMF (65
°C, 6 h).7 The starting loading of chloromethyl groups affects
the amount of amine cross-linking that occurs in this reaction.
We have found by comparing a number of commercially
available resins with loadings in the range of 0.5-4.3 mmol of
Cl/g, that a starting loading of 1.7 mmol of Cl/g gives a maximal
loading of approximately 3.2 mmol of NH and NH2/g of 1. The
NH/NH2 loading is deduced from N analysis and confirmed by
treatment of the resin with excess 3,4-dichlorophenyl isocyanate,
followed by Cl analysis of thoroughly washed and dried product.
The isocyanate resin 2 is prepared by reacting triphosgene (200
mol % equiv of phosgene) and Et3N (500 mol %) with
Based upon the success of these simple examples, it seems
rational that the ability to do traditional multistep syntheses with
(6) During the course of review of this manuscript a similar strategy
was published: Kaldor, S. W.; Siegel, M. G.; Fritz, J. E.; Dressman, B. A.;
Hahn, P. J. Tetrahedron Lett. 1996, 37, 7193. Kaldor, S. W.; Fritz, J. E.;
Tang, J.; McKinney, E. R. Bioorg. Med. Chem. Lett. 1996, 6, 3041.
(7) Other polyamine resins have been reported: Yamskov, I. A.;
Budanov, M. V.; Davankov, V. A. Bioorg. Khim. 1979, 5, 757.
(8) Previously prepared at 110 °C. Metelko, M.; Zupan, M. Synth.
Commun. 1988, 18, 1821.
(9) For selected references on compatibility of polymer-supported
reactants, see Cohen, B. J.; Kraus, M. A.; Patchornik, A. J. Am. Chem.
Soc. 1981, 103, 7620. Parlow, J. J. Tetrahedron Lett. 1995, 36, 1395.

4884 J. Am. Chem. Soc., Vol. 119, No. 21, 1997
Booth and Hodges
Figure 2. Polymer-supported quench-examples.
18 which is of excellent purity as determined by 1H-NMR and
HPLC. The moderate yield of this reaction is poorly understood.
By TLC the reaction appears to be very clean. Literature
precedent10 would appear to rule out the possibility that the
alternate pyrazole regioisomer is also formed, crystallizes from
the reaction mixture, and is being removed in the filtration. It
is plausible that some of the desired product is retained by the
weakly basic morpholino resin; however, attempts to more
vigorously wash the resin gave substantially less pure product
so we chose to accept the moderate yield in favor of greater
purity. In the second step, 18 is converted to a mixed anhydride
by treatment with isobutyl chloroformate in the presence of 3
which, in the third step, is treated in situ with 7 to afford 19,
also of excellent purity by 1H-NMR and HPLC, following PSQ
purification with a mixture of 1 and 2.
An example of the application of PSQ in mixed synthesis is
shown in Figure 4. Equimolar quantities of three unusual
carboxylic acids (2011, 2112, and 2213), which were obtained
from the Parke-Davis historical collection, are combined and
converted to the corresponding mixed anhydrides as above.
Treatment with an excess of an unusual amine from the P-D
collection (2314), followed by quenching with a mixture of 2
and 1 (added for good measure to ensure that any traces of
Figure 3. Pyrazole synthesis with PSQ-purification.
a quick PSQ purification after each step would make possible
the application of combinatorial synthesis paradigms to many
known solutionphase syntheses. As such, PSQ methods would
apply to preparation of both lead optimization and lead
generation libraries. An example of the application of PSQ
purification to a traditional synthesis of pyrazoles10 is illustrated
by the conversion of 16 to 19 (Figure 3). In the first step a
1,3-diketone (16) is condensed with excess phenylhydrazine-
4-carboxylic acid hydrochloride (17) in the presence of 3.
Removal of excess 17 by quenching the reaction with 2 affords
(11) Short, F. W.; Hoefle, M. L. United States Patent 3,657,270.
(12) Prepared according to the method of Gates, M.; and Dickinson, C.
L. J. Org. Chem. 1957, 22, 1398. For compound 21: mp 129-130 °C
(benzene/hexane). Anal. Calcd for C14 H16O3S2: C, 56.75; H, 5.44; S, 21.6.
Found: C, 56.97; H, 5.37; S, 21.52.
(13) Prepared according to the method of Creger P. L.; Neuklis W. A.
United States Patent 3,707,566. For compound 22: mp 79-80 °C (hexane).
Anal. Calcd for C17 H18O3: C, 75.54; H, 6.71. Found: C, 75.76; H, 6.71.
(10) Murray, W.; Wachter, M.; Barton, D.; Forero-Kelly, Y. Synthesis
1991, 18.

Polymer-Supported Quenching Reagents
J. Am. Chem. Soc., Vol. 119, No. 21, 1997 4885
complete removal of the starting amine(s), but complete removal
of all of the starting carboxylic acids would be much more
difficult to establish conclusively by NMR. LC-MS would
likely be a preferable analytical tool for this purpose. This
simple example illustrates the utility of PSQ for removing
multiple starting materials from a crude reaction mixture when
multiple products are intentional. In such batch mode combi-
natorial syntheses, PSQ has an inherent advantage over tradi-
tional purification methods since it focuses on the chemical,
not physical properties of the contaminants. In common
practice, one would likely identify contaminants and validate
PSQ purification protocols with a representative variety of
starting materials in small mixture and/or single compound
syntheses and assume similar success with the same PSQ
protocols on a larger mixture.
Summary
In summary, PSQ purification offers an efficient and readily
automated means for obtaining organic molecules of suitable
purity for routine in Vitro biological testing. Relative to
traditional chromatography, PSQ uses less solvent, requires less
solid support, and eliminates the need to collect multiple
fractions. Thus it is amenable to parallel application and
automation. Furthermore, when applied to batch mode syn-
thesis, mixtures of desired products can be separated from
mixtures of starting materials, a process that would be cumber-
some, if not impossible, by traditional methods of purification.
Although the PSQ resins described are expensive relative to
silica gel on a per gram basis, the small quantity of resin and
solvent employed, combined with labor savings, actually make
PSQ purification significantly cheaper than flash chromatog-
raphy in most cases. Additionally, the PSQ reagents described
are easily made in large quantity from some of the least
expensive resin starting materials, and each PSQ reagent has
many possible applications in the selective removal of excess
reagents and identified impurities from crude product mixtures.
The combination of solution synthesis and PSQ purification
provides a convenient alternative to solid-phase synthesis in the
practice of combinatorial chemistry.
Experimental Section
General. Unless otherwise indicated, all reactions were run in
capped glass vials and were shaken on an orbital shaker. Reagents
and solvent were commercially available and used without further
purification. In order to represent the purity of compounds as they
would be found following a solution-phase synthesis of a combinatorial
library, all characterization listed before the term “Recrystallized mp”
is derived from compounds as they are obtained following filtration of
PSQ resins and evaporation of solvent. HPLC analysis was achieved
using a Perkin Elmer, reduced activity, C18 column (4.6 mm ID, 8 cm
length). The mobile phase (acetonitrile/25 mM phosphate buffer pH
3, 1.0 mL/min) was used as a linear gradient of 20-90% acetonitrile
over 7 min; detection was at 210 nm. Unless otherwise stated, NMR
spectra were obtained at a field strength of 400 MHz with samples
dissolved in CDCl3. Coupling constants (J) are reported in hertz. IR
Figure 4. Amide mixture synthesis.
starting carboxylic acids are removed) affords an approximately
equimolar mixture of 24, 25, and 26, as confirmed by 1H-NMR
analysis, HPLC (Figure 4), and CI-MS. Complete removal of
the excess amine is further verified by the use of a ninhydrin
test. Complete removal of unreacted carboxylic acid starting
spectra were taken as KBr pellets, and absorptions are reported in cm-1
.
High resolution mass spectra were obtained by chemical ionization with
methane or by electrospray ionization from a soluton in MeOH/H2O
(2:1) with 2.5% HOAc. Melting points are uncorrected.
1
materials is further verified by comparison of the H-NMR
spectrum for each of the individual starting acids with that of
the amide product mixture. By inference, one could similarly
use a much larger number of carboxylic acids to prepare a larger
mixture of desired amides with comparable results. The
ninhydrin test would still be applicable for demonstrating
Polymer-Supported Tris(2-aminoethyl)amine (1). A suspension
of Merrifield resin (Fluka, 50 g, 1.7 mmol of Cl/g resin, 85 mmol) in
DMF (500 mL) was treated with tris(2-aminoethyl)amine (50 mL, 342
mmol). The resulting mixture was shaken at 65 °C for 6 h under N2
atmosphere. After cooling to room temperature, the resin was filtered
and washed successively with MeOH, DMF, Et3N, MeOH, DCM, Et3N,
MeOH, DCM, MeOH, DCM, and MeOH. The resulting amine resin
was dried at 45-50 °C, 20 mmHg for 24 h, and stored in tightly sealed
(14) Trivedi, B. K.; Holmes, A.; Stoeber, T. L.; Blankley, C. J.; Roark,
W. H.; Picard, J. A.; Shaw, M. K.; Essenburg, A. D.; Stanfield, R. L.;
Krause, B. R. J. Med Chem. 1993, 36, 3300.

4886 J. Am. Chem. Soc., Vol. 119, No. 21, 1997
Booth and Hodges
bottles. Calcd: N, 8.02; Cl, 0.00. Found: N, 5.96; Cl, 0.42 (indicates
approximately 25% cross-linking). A small sample reacted with excess
3,4-dichlorophenyl isocyanate in DCM indicates a quenching capacity
of 3.18 mmol/g resin, consistent with 3/4 of the N content in the amine
resin. Anal. Calcd: N, 6.51; Cl, 14.15. Found: N, 6.25; Cl, 13.99.
butanedione (16, 81.5 mg, 0.5 mmol) and (4-carboxyphenyl)hydrazine
hydrochloride (17, 113 mg, 0.6 mmol) in MeOH (2 mL) was shaken
for 2.5 h. The methanol was blown off under a stream of N2. DCM (4
mL) and polymer-supported isocyanate 2 (350 mg) were added, and
the reaction mixture was shaken for 16 h. An additional portion of
polymer-supported isocyanate 2 (120 mg) was added. After 4 h the
resin was filtered and washed with DCM (2 × 1.5 mL). The combined
organic phases, when concentrated to dryness, gave the title compound
(67 mg, 48%) as an orange solid: mp 159-162 °C; tR ) 6.12; IR
Polymer-Supported Isocyanate 2. A suspension of aminomethyl
resin (Fluka, 1.1 mmol of N/g resin, 15 g, 16.5 mmol) in DCM (150
mL) was treated with Et3N (11.5 mL, 83 mmol) and triphosgene (3.25
g, 2 mmol equiv of phosgene) and shaken 5 h at room temperature.
The resulting isocyanate resin was filtered and washed with DCM (2
× 200 mL), CHCl3 (2 × 200 mL), Et2O (1 × 200 mL), THF (1 × 200
mL), Et2O (1 × 200 mL), THF (1 × 200 mL), and Et2O (1 × 200
mL). The resin was then dried at 35-40 °C, 25 mmHg, for 24 h. IR
(KBr) 2260. Anal. Calcd: N, 1.50. Found: N, 1.45.
1
(KBr) 1699; H NMR δ 2.36 (3H, s), 6.3 (1H, s), 7.2 (2H, m), 7.33
(3H, m), 7.35 (2H, d, J ) 7), 8.0 (2H, d, J ) 7); predicted mass for
(C17H14N2O2 + H)+, 278.1055; found by HRMS (CI), 278.1055.
Recrystallization from EtOAc/hexane gives the hydrate:15 recrystallized
mp 169-170 °C. Anal. Calcd for C17H14N2O2‚H2O: C, 68.90; H,
5.45; N, 9.45. Found: C, 68.75; H, 5.46; N, 9.38.
Polymer-Supported Morpholine 3. A suspension of Merrifield
resin (Fluka, 20 g, 4.3 mmol of Cl/g resin, 86 mmol) in DMF (100
mL) was treated with morpholine (20 mL, 229 mmol). The resulting
mixture was shaken at 65 °C for 6 h under N2 atmosphere and then
allowed to stand at room temperature 24 h. After cooling to room
temperature, the resin was filtered and washed successively with MeOH,
DMF, MeOH, Et3N, DCM, MeOH, Et3N, DCM, MeOH, EtOAc, and
hexanes. The resulting polymer-supported morpholine, 3, was dried
at 45-50 °C, 20 mmHg, for 48 h and stored in tightly sealed bottles.
Anal. Calcd: N, 4.83; Cl, 0.00. Found: N, 4.98; Cl, 0.21.
N-(3-Isopropoxypropyl)-4-(3-methyl-5-phenylpyrazol-1-yl)benza-
mide (19). A solution of 4-(3-methyl-5-phenylpyrazol-1-yl)benzoic
acid (18, 20 mg, 70 µmol) in DCM (0.7 mL) was treated with polymer-
supported morpholine (3, 100 mg) and 0.1 M isobutyl chloroformate
in DCM (0.75 mL, 75 µmol). The resulting slurry was shaken under
N2, at rT, for 35 min and then treated with a solution of (3-
isopropoxypropyl)amine (7, 100 mg, 85 µmol) in DCM (0.5 mL). The
reaction was shaken at rT for 2.5 h. Polymer-supported isocyanate (2,
75 mg) and polymer-supported tris(2-aminoethyl)amine (1, 100 mg)
were added, and the mixture was shaken an additional 2 h. Resins
were removed by filtration and rinsed with DCM (2 × 2.5 mL).
Combined filtrate and washings were evaporated to a solid and dried
at 0.25 mmHg, rT, overnight to afford 19 as a pale orange solid (19.7
1-Butyl-3-(2-thiophen-2-yl-ethyl)urea (6). Method 1: To a solu-
tion of 2-(thieny-2-yl)ethyl isocyanate (4, 47 mg, 0.3 mmol) in DCM
(2 mL) was added n-butylamine (5, 25 µl, 0.25 mmol). The reaction
mixture was shaken for 1 h, and then polymer-supported tris(2-
aminoethyl)amine (1, 50 mg) was added. After 2 h the resin was
filtered and washed with DCM (2 × 1.5 mL). The combined organic
phases, when concentrated to dryness, gave the title compound (53
mg, 94%) as an oil that crystallizes upon prolonged standing: mp 46-
1
mg, 75%): mp 109-111 °C; tR ) 6.90; IR (KBr) 1629; H NMR δ
1.15 (6H, d, J ) 6), 1.87 (2H, m), 2.39 (3H, s), 3.55-3.64 (5H,
complex), 6.32 (1H, s), 7.2 (2H, m), 7.3 (6H, m), 7.71 (2H, d, J ) 9);
predicted mass for (C23H27N3O2 + H)+, 378.2181; found by HRMS
(CI), 378.2166. Recrystallization from EtOAc/hexanes gives a nearly
colorless solid: recrystallized mp 126-127 °C. Anal. Calcd for
C23H27N3O2‚0.3H2O: C, 72.14; H, 7.27; N, 10.97. Found: C, 72.15;
H, 6.97; N, 10.94. Found by HRMS (CI) 378.2170.
1
50 °C; tR ) 5.92; IR (KBr) 1626; H NMR δ 0.9 (3H, t, J ) 8), 1.33
(2H, m), 1.42 (2H, m), 3.0 (2H, t, J ) 7), 3.11 (2H, m), 3.44 (2H, m),
4.61 (1H, brs), 4.77 (1H, brs), 6.81 (1H, dd, J ) 3, 1), 6.92 (1H, dd,
J ) 5, 3), 7.13 (1H, dd, J ) 5, 1); 13C NMR δ 13.8, 20.0, 30.7, 32.3,
40.2, 41.8, 123.7, 125.2, 126.9, 141.7, 158.22; predicted mass for
(C11H18N2OS + H)+, 227.1218; found by HRMS (CI), 227.1210. Anal.
Calcd for C11H18N2OS: C, 58.40; H, 8.02; N, 12.38. Found: C, 58.69;
H, 8.22; N, 12.06. Attempted recrystallization from a variety of solvents
gave at best an amorphous solid from hexanes: mp 42-46 °C.
Supporting Information Available: Complete experimental
procedures and data for compounds 9, 12, 15, 24, 25, and 26
1
plus photoreductions of actual H NMR spectra and HPLC
chromatograms, for 6, 9, 12, 15, 18, 19, and the mixture of
24-26 (18 pages). This material is contained in many libraries
on microfiche, immediately follows this article in the microfilm
version of the journal, can be ordered online from the ACS,
and can be downloaded from the Internet; see any current
masthead page for ordering information and Internet access
instructions.
Method 2: To a solution of 4 (30 mg, 0.2 mmol) in DCM (2 mL)
was added 5 (25 µl, 0.25 mmol). The reaction mixture was shaken
for 1 h, and then polymer-supported isocyanate (2, 50 mg) was added.
After 2 h the resin was filtered and washed with DCM (2 × 1.5 mL).
The combined organic phases, when concentrated to dryness, gave the
title compound (44 mg, 99%) as an oil that solidifies upon prolonged
standing: mp 40-42 °C. Spectral and HPLC data are identical to those
from method 1 above.
JA9634637
4-(3-Methyl-5-phenylpyrazol-1-yl)benzoic Acid (18).10,15 A sus-
pension of polymer-supported morpholine (3, 170 mg), 1-phenyl-1,3-
(15) The hydrate of 18 has previously been described, mp 159-160 °C
(EtOH).