A new procedure for preparation of carboxylic acid hydrazides

Article abstract of DOI:10.1021/jo026288n

The standard method for preparing carboxylic acid hydrazides is hydrazinolysis of esters in alcoholic solutions. However, when applied to α,β-unsaturated esters, the main product typically is the pyrazolidinone resulting from an undesired Michael-type cyclization. Other alternative methodologies reported for direct preparation of hydrazides from acids are inefficient. We developed an efficient and general process, involving preforming activated esters and/or amides followed by reaction with hydrazine, for the preparation of hydrazides including those of α,β-unsaturated acids. This process gives the desired hydrazides in excellent yield and purity under mild conditions.

Full text of DOI:10.1021/jo026288n

carbamate moieties for a general application of this
method. This paper will discuss the scope and generality
of this method.
Hydrazinolysis of esters is the conventional method for
preparing acyl hydrazides.4,5 However, when this method
was applied to the R,â-unsaturated ester 4, the predomi-
nant product was the pyrazolidinone 3, the result of
hydrazinolysis and an undesired subsequent intramo-
lecular Michael-type addition (Scheme 2).6
A New P r oced u r e for P r ep a r a tion of
Ca r boxylic Acid Hyd r a zid es
Xini Zhang,*,† Michael Breslav,‡ J effrey Grimm,†
Kailin Guan,‡ Aihua Huang,‡ Fuqiang Liu,†
Cynthia A. Maryanoff,† David Palmer,† Mitul Patel,†
Yun Qian,† Charles Shaw,‡ Kirk Sorgi,†
Stephen Stefanick,† and Dawei Xu‡
J ohnson & J ohnson Pharmaceutical Research &
Development, L.L.C., Drug Evaluation, 1000 Route 202,
Raritan, New J ersey 08869
Alternatively, acyl hydrazides may be prepared by
condensing carboxylic acids with hydrazine in the pres-
ence of coupling agents. Unfortunately, most of these
methods provide low yields and complicated product
isolations.7-10 Krysin reported good yields using bis-
(trimethylsilyl)acetamide as a condensing agent, but this
method requires strictly anhydrous conditions.11 We
evaluated the use of acid chloride12 and mixed anhy-
dride13 approaches to preparing the desired hydrazide but
observed incomplete reactions, poor yields, and a large
number of impurities.
To prepare the desired hydrazide, we turned our
attention to a commonly used carbodiimide-based cou-
pling reaction such as using 1-hydroxybenzotriazole
(HOBt) and 1-(3-dimethylaminopropyl)-3-ethylcarbodi-
imide hydrochloride (EDC) as coupling reagents.7,8 Our
initial investigation showed that the order of reagent
addition affected the outcome of the reaction, which
prompted us to examine this variable in detail (equation
1). In a typical peptide coupling procedure wherein all
reagents were added sequentially as described in method
I,7 we observed low conversion and poor yield. Modifying
the addition sequence to add hydrazine last gave slightly
better results (method II).
xzhang@prdus.jnj.com
Received August 5, 2002
Abstr a ct: The standard method for preparing carboxylic
acid hydrazides is hydrazinolysis of esters in alcoholic
solutions. However, when applied to R,â-unsaturated esters,
the main product typically is the pyrazolidinone resulting
from an undesired Michael-type cyclization. Other alterna-
tive methodologies reported for direct preparation of hy-
drazides from acids are inefficient. We developed an efficient
and general process, involving preforming activated esters
and/or amides followed by reaction with hydrazine, for the
preparation of hydrazides including those of R,â-unsaturated
acids. This process gives the desired hydrazides in excellent
yield and purity under mild conditions.
Compounds A and B are potential fibrinogen receptor
antagonists under development (Scheme 1).1 A key
intermediate common to both compounds is E-3-(N-t-
butoxycarbonyl-4-piperidinyl)propenoic acid hydrazide
(1). To prepare multikilogram drug quantities for further
study, we were in need of an efficient and high-yielding
method for preparing this R,â-unsaturated hydrazide
from the corresponding acid 2 (Scheme 1). Though this
seemed to be straightforward initially, the transformation
turned out to be quite complex. While acyl hydrazides
are important building blocks in many syntheses,2,3 there
are no efficient and general procedures for preparing
hydrazides, especially R,â-unsaturated hydrazides, from
the corresponding acids. Most reported procedures are
low yielding and require chromatographic purification
that is not suitable for large-scale preparations. Several
literature approaches were investigated without satisfac-
tory results. The most promising lead came from a
traditional carbodiimide-based coupling of acids and
hydrazine. Our careful investigation resulted in an
efficient, high-yielding procedure, involving preforming
activated esters and/or amides followed by reaction with
hydrazine, for preparing acyl hydrazides from the cor-
responding acids. This coupling procedure is rapid,
convenient, proceeds under mild conditions, and is suit-
able for large-scale preparations. We have studied a
variety of conjugated and unconjugated unsaturated
acids as well as saturated acids bearing ester and
1
f
2
(1)
Method I: (1) 4-methylmorpholine; (2) HOBt;
(3) N2H4; (4) EDC <30%
Method II: (1) 4-methylmorpholine; (2) HOBt;
(3) EDC; (4) N2H4 30-70%, inconsistent
Method III: (1) HOBt;
(2) DCC, 5 min, filter into N2H4 60%
Attempts to drive the reaction to completion in either
method I or II by using excess coupling agent, different
(4) Carter, D.; Vranken, D. J . Org. Chem. 1999, 64, 8537.
(5) Mazaleyrat, J .; Wakselman, M.; Formaggio, F.; Crisma, M.;
Toniolo, C. Tetrahedron Lett. 1999, 40, 6245.
(6) Harada, R.; Kondo, H. J . Org. Chem. 1968, 41, 2521.
(7) Patterson, J .; Ollmann, I.; Cravatt, B.; Boger, D.; Wong, C.;
Lerner, R. J . Am. Chem. Soc., 1996, 118, 5938.
(8) J ansen, R.; Schummer, D.; Irschik, H.; Hoefle, G. Liebigs Ann.
Chem. 1990, 10, 975-88.
(9) Pavlov, A.; Berdnikova, T.; Olsufyeva, E.; Miroshnikova, O.;
Filipposyanz S.; Preeobrazzhenskaya, M. J . Antibiot. 1996, 49, 194.
(10) Miyasaka, T.; Hibino, S. J . Chem. Soc., Perkin Trans. 1 1986,
479.
(11) Krysin, E.; Karel’skii, V.; Antonov, A.; Rostovskaya, G. Chem.
Nat. Compd. 1979, 15, 601.
† Chemical Development.
‡ Analytical Development.
(1) Hoekstra, W.; Lawson, E.; Maryanoff, B. PCT Int. Appl. WO
0006570, 2000.
(2) Khodairy, A. Synth. Commun. 2001, 31, 2697-2712.
(3) Wasfy, A.; El-Shenawy, A.; Nassar, S. Heterocycl. Commun. 2001,
7, 493-500.
(12) Plehnert, R.; Schroeter, J .; Tschierske, C. J . Mater. Chem. 1998,
8, 2611.
(13) Monge, A.; Palop, J .; Goni, T.; Fernandez-Alvarez, E. J . Het-
erocycl. Chem. 1986, 23, 141.
10.1021/jo026288n CCC: $22.00 © 2002 American Chemical Society
Published on Web 12/04/2002
J . Org. Chem. 2002, 67, 9471-9474
9471

SCHEME 1
SCHEME 2
13 was easily isolated. Compounds 12 and 13 were
identified by NMR and MS as the ester and the amide,
respectively.19,20 The structure of amide 13 was further
confirmed by the observation of a long-range (three
bonds) heteronuclear correlation between the R-proton
of the acyl group and one nitrogen of the benzotriazole
moiety in a gradient-enhanced HMBC NMR experiment.
The activated ester 12 forms initially and then slowly
rearranges to the amide 13. The intermediate formation
is complete within 2 h, resulting in a 1:2 mixture of 12
and 13.21 Both 12 and 13 are equally reactive toward
hydrazine, and their conversion to the desired hydrazide
is instantaneous.
bases, and longer reaction times did not improve the yield
and/or purity of the product. Instead, we found, in
addition to the desired product, a multitude of undesired
byproducts (Scheme 3). Under the reaction conditions,
acid, EDC, hydrazine, and hydrazide react with each
other to form 5-9.14,15 We also detected the saturated
hydrazide 10 and acid 11, the products of double-bond
reductions by diimide, a contaminant in the commercial
hydrazine.16,17 These impurities were each formed in the
range of 3-15% and were very difficult to remove from
the desired product. Folkers reported that Cbz-L-Pro-
NHNH2 could be prepared in 89% yield by treating Cbz-
L-ProOH with HOBt and 1,3-dicyclohexylcarbodiimide
(DCC) and waiting for 5 min before filtering the mixture
into a hydrazine solution.18 However, when we applied
this method to our R,â-unsaturated acid 2, we observed
only 60% conversion (method III, eq 1). It is clear from
these observations that the seemingly “simple” conver-
sion of R,â-unsaturated acids to hydrazides is a rather
complex and a poorly understood reaction.
To minimize these side reactions, it is important to
avoid the coexistence of acid 2 and hydrazide 1 or
hydrazine and EDC in the reaction mixture. A stepwise
reaction in which the acid is first converted to an
activated intermediate followed by reaction with hydra-
zine would satisfy these requirements and preclude most
of the side reactions. To that effect, treatment of 2 with
HOBt and EDC in acetonitrile resulted in a heavy
suspension in 2 h. Analysis of the reaction mixture by
HPLC revealed the formation of two isomeric intermedi-
ates 12 and 13. These two compounds showed different
solubilities in acetonitrile, and the less soluble compound
The order of the reagent addition is critical to the
success of the reaction as shown in Scheme 4. To suppress
the reduction of the double bond, cyclohexene17 was added
to the hydrazine solution to consume any formed diimide.
Among the solvents investigated, we found that CH2Cl2,
DMF, and CH3CN are the most suitable for this reaction.
The preformation of the intermediates and a reverse
mode of addition effectively drives the reaction to comple-
tion, suppresses the formation of side products, and
affords the desired hydrazide in a nearly quantitative
yield with 98% purity. This procedure enabled us to
successfully prepare multikilogram quantities of hy-
drazide 1 without chromatographic purification.
To explore the generality of this procedure, we applied
it to a variety of functionalized carboxylic acids as shown
in Table 1. The stepwise hydrazide formation procedure
was compared to a standard coupling condition that
represents the best scenario of methods II and III. While
DCC and EDC give comparable results, EDC is safer and
the byproducts are easier to remove from the product;
thus, EDC was used in all entries. In addition to
expanding the study to other alkenoic acids, we investi-
gated alkynoic acids, since hydrazinolysis of alkynoic
esters is also complicated by formation of pyrazolones 14.
(14) Side products were identified by LC/MS and mostly confirmed
by comparison to authenic samples.
(15) Pottorf, R.; Szeto, P. Encyclopedia of Reagents for Organic
Synthesis; Wiley and Sons: New York, 1995; Vol. 4, p 2430.
(16) Corey, E. J .; Mock, W.; Pasto, D. Tetrahedron Lett. 1961, 347
(17) Maryanoff, B.; Greco, M.; Zhang, H.; Andrade-Gordon, P.;
Kauffman, J .; Nicolaou, K.; Liu, A.; Brung, P. J . Am. Chem. Soc. 1995,
117, 1225.
(19) Okawara, T.; Ikeda, N.; Yamasaki, T.; Forukawa. M. Chem.
Pharm. Bull. 1988, 36, 36628.
(18) Folkers, K.; Bowers, C.; Lutz, W.; Friebel, K.; Kubiak, T.;
Schircks, B.; Rampold, G. Anorg. Chem. Org. Chem. 1982, 37, 1075.
(20) Meldal, M. Acta Chem. Scand. 1986, B 40, 242.
(21) Prolonged stirring resulted in complete conversion to the amide.
9472 J . Org. Chem., Vol. 67, No. 26, 2002

SCHEME 3
TABLE 1. P r ep a r a tion of Hyd r a zid es: Com p a r ison
betw een th e New P r oced u r e a n d Sta n d a r d Cou p lin g
P r oced u r e
SCHEME 4
We included acids bearing ester and carbamate moieties,
since entry into this class of hydrazides may not be easily
accomplished via hydrazinolysis. Finally, the method was
applied to enantiomeric carboxylic acids with stereo-
centers at the potentially racemizable R-positions. We
also extended the methodology to include phenylhydra-
zine.
The results in Table 1 demonstrate the superiority of
the stepwise procedure. The mild conditions tolerate the
presence of a variety of functional groups. The rate of
the activated intermediate formation in all entries was
determined by HPLC analysis and found to be substrate
dependent. For most alkenoic acids (entries 1-10), the
formation of activated intermediates requires about 2 h
to reach completion. One exception is entry 5, which
requires only 15 min for the intermediate formation. The
slow intermediate formation suggests that for this group
of acids, it is necessary to preform the activated ester/
amide before reaction with hydrazine for the success of
hydrazide preparation. Most saturated and nonconju-
gated acids (entries 11-16) require only 10-20 min for
the intermediate formation. The activated esters/amides
in this group are too labile to be analyzed by HPLC
directly. One exception, however, is 2-phenylpropanoic
acid (entries 14 and 15) where the intermediates of this
acid form in 2 h and can be analyzed by HPLC directly.
High yields and purities were achieved with our new
procedure, while the standard coupling procedure gave
poor results. For some acids in this group, method III
(eq 1) will result in fare yields if enough time is allowed
before reaction with hydrazine. This observation explains
the good yield employing coupling method III (eq 1) on a
a
Yields calculated by HPLC analysis with internal and external
standards unless otherwise indicated. b As described in the general
procedures. c Isolated yield. Determined by enantiomeric HPLC
d
analysis.
J . Org. Chem, Vol. 67, No. 26, 2002 9473

remove HOBt. Removal of the solvents under reduced pressure
yielded the hydrazide. Water-insoluble hydrazides were easily
isolated by filtration after diluting the reaction mixture with
water. Highly water-soluble hydrazides were isolated as p-
toluenesulfonic acid or oxalic acid salts without extraction. DCC
worked equally well. DMF and methylene chloride can also be
used as solvents for this reaction.
Rep r esen ta tive Sp ectr oscop ic Da ta for Com p ou n d s 1
a n d 13-31. E-3-(N-t-Bu toxyca r bon yl-4-p ip er id in yl)p r op e-
n oic Acid Hyd r a zid e (1, a s Tosyla te): white solid; mp 161-
163 °C; 1H NMR (DMSO-d6) δ 1.19 (m, 2H), 1.39 (s, 9H), 1.67
(m, 2H), 2.30 (s, 3H), 2.35 (m, 1H), 2.76 (m, 2H), 3.94 (m, 2H),
5.93 (d, J ) 15.6 Hz, 1H), 6.83 (dd, J ) 15.7, 6.4 Hz, 1H), 7.10
(d, J ) 7.6 Hz, 2H), 7.50 (d, J ) 7.6 Hz, 2H); 13C NMR (DMSO-
d6) δ 21.13, 28.43, 32.25, 37.96, 78.97, 118.81, 125.84, 128.43,
138.05, 145.91, 150.41, 154.18, 164.74. Anal. Calcd for
C20H31N3O6S: C, 54.40; H, 7.08; N, 9.52; S, 7.26. Found: C,
54.23; H, 7.09; N, 9.47; S, 7.47.
saturated acid as reported by Folkers and the poor yield
when it was applied to R,â-unsaturated acid 2 in our
initial evaluation. Interestingly, alkynoic acids (entries
17 and 18) form the activated intermediates rapidly
within 5 min, and the intermediates are stable to analysis
by HPLC. Both the standard coupling procedure and our
new procedure work equally well for alkynoic acids.
Esters and carbamates are unaffected under our reaction
conditions (entries 1, 5, 11-13, and 16). The reaction
conditions do not cause any measurable racemization as
demonstrated by entries 15 and 16. The procedure is not
limited to hydrazine: reactions with phenylhydrazine
under our conditions are clearly superior to those per-
formed using standard coupling procedure (entries 9, 15,
and 16).
In conclusion, our investigation led to a clear under-
standing of the poorly studied reaction of acid and
hydrazine. We isolated the intermediates, studied a
variety of acids, and observed the structure-reaction rate
relationship in different classes of substrates. An ef-
ficient, convenient, and practical procedure was devel-
oped for preparing carboxylic acid hydrazides from the
corresponding acids in excellent yield and purity. This
procedure is particularly effective in the preparation of
R,â-unsaturated carboxylic acid hydrazides. The mild
reaction conditions tolerate the presence of such groups
as ester and carbamate, cause no racemization, and are
applicable to substituted hydrazines such as phenyl
hydrazine.
ter t-Bu tyl 4-[3-Oxo-3-(3-oxyben zotr ia zol-1-yl)p r op en yl]-
p ip er id in e-1-ca r boxyla te (13): white solid; mp 144-146 °C;
1H NMR (acetic acid-d4) δ 1.48 (m, 2H), 1.48 (s, 9H), 1.88 (dd, J
) 13.3, 2.5 Hz, 2H), 2.60 (m, 1H), 2.91 (m, 2H), 4.20 (d, J ) 11.3
Hz, 2H), 7.08 (dd, J ) 15.5, 1.4 Hz, 1H), 7.42 (dd, J ) 15.6, 6.7
Hz, 1H), 7.68 (ddd, J ) 8.4, 7.2, 0.8 Hz, 1H), 7.88 (ddd, J ) 8.4,
7.2, 1.0 Hz, 1H), 8.10 (d, J ) 8.4 Hz, 1H), 8.47 (d, J ) 8.4 Hz,
1H); 13C NMR (acetic acid-d4) δ 28.5, 31.0, 39.8, 44.2, 81.3, 116.2,
117.0, 118.4, 128.4, 133.6, 133.8, 134.2, 158.5, 162.6. Anal. Calcd
for C19H24N4O4: C, 61.28; H, 6.50; N, 15.04. Found: C, 61.11;
H, 6.62; N, 15.11.
5-P en tyl-1,2-d ih yd r o-p yr a zol-3-on e (14): yellow solid; mp
194-195 °C; 1H NMR (DMSO-d6) δ 0.86 (t, J ) 6.8 Hz, 3H),
1.27 (m, 4H), 1.52 (m, 2H), 2.42 (t, J ) 7.5 Hz, 2H), 5.22 (s, 1H),
10.27 (br s, 2H); 13C NMR (DMSO-d6) δ 14.23, 22.15, 25.93,
28.66, 31.14, 68.24, 149.00, 161.21. Anal. Calcd for C8H14N2O:
C, 62.31; H, 9.15; N, 18.17. Found: C, 62.07; H, 8.93; N, 18.18.
3-P h en yla cr ylic Acid N′-P h en ylh yd r a zid e (22). The ac-
tivated ester/amide mixture was added slowly to phenylhydra-
zine (2 equiv) in CH3CN at room temperature. The resulting
suspension was stirred at room temperature for 30 min and then
diluted with water. The product precipitated and was collected
by filtration and dried. Data were identical with those of
commercial compounds.
Exp er im en ta l Section
1H and 13C NMR spectra were recorded at 300 and 75.45 MHz,
respectively. Accurate mass measurements were obtained under
electrospray ionization conditions. Most reagents were com-
mercially available reagent-grade chemicals and used without
further purification.
Gen er a l P r oced u r es: Sta n d a r d Cou p lin g P r oced u r e.
The acid (40 mmol) was dissolved or suspended in CH3CN (80
mL) at room temperature. NMM (48 mmol), HOBt (48 mmol),
and EDC (48 mmol) were added sequentially followed by
hydrazine (80 mmol) 5 min later. The mixture was stirred at
room temperature overnight. The yield was determined by HPLC
analysis compared to product standard. The yields obtained with
this procedure were similar to those obtained with method III.
Method III was not used here due to the insolubility of the
activated intermediates in some entries.
New P r oced u r e. The acid (40 mmol) was dissolved or
suspended in CH3CN (80 mL) at room temperature. HOBt (48
mmol) was added in one portion followed by EDC (48 mmol).
The mixture was stirred at room temperature, and the reaction
progress was monitored by HPLC until all of the acid was
converted to the activated ester/amide mixture. The resulting
mixture was then slowly added to a solution of hydrazine (80
mmol) and cyclohexene (1 mL) in CH3CN (40 mL) while the
temperature was maintained at 0-10 °C. The reaction was
usually complete upon the completion of addition. Water (40 mL)
was added. The aqueous CH3CN mixture was extracted with
EtOAc followed by a carbonate wash of the organic layer to
Ben zyl Hyd r a zin oca r bon yl Aceta te (24, a s Oxa la te).
Upon reaction completion, the mixture was filtered and the
filtrate was evaporated to dryness under reduced pressure. The
residue was suspended in ethyl acetate and filtered to remove
any insoluble solid. Oxalic acid (1 molar equiv) in ethyl acetate
was added to the clear filtrate, and the salt of the product
precipitated and was isolated by filtration. White solid; mp 106-
1
107 °C; purity, 98.5%; H NMR (DMSO-d6) δ 3.25 (s, 2H), 5.10
(s, 2H), 7.40 (m, 5H), 8.05-8.50 (br s, 5H); 13C NMR (DMSO-d6)
δ 66.33, 127.27, 128.20, 128.39, 128.76, 136.19, 163.05, 164.84,
167.61; HRMS calcd for C10H12N2O3, 209.0926; found, 209.0931.
Ackn owledgm en t. The authors thank Richard Dun-
phy for high-resolution mass spectral analysis, Yong
Guo for helpful discussions, and Michael J . Humora for
critical reading and review.
Su p p or tin g In for m a tion Ava ila ble: Experimental de-
tails. This material is available free of charge via the Internet
at http://pubs.acs.org.
J O026288N
9474 J . Org. Chem., Vol. 67, No. 26, 2002