The Use of Chloroformamidine Hydrochloride as a Reagent for the Synthesis of Guanidines from Electron Deficient Aromatic Amines

Article abstract of DOI:10.1002/jhet.2567

In efforts to optimize a manufacturing process for an internal development compound, a clean, efficient approach to guanidine synthesis using chloroformamidine hydrochloride was identified. To investigate the general utility of this methodology towards electron-deficient aromatic amines, a set of favorable conditions were developed from a series of screens, and the scope of the reaction was probed. The successful application of this chemistry to a variety of pyridines, anilines, and heterocyclic compounds highlights its use as an improved, alternative guanylation method for this often challenging set of aromatic amines.

Full text of DOI:10.1002/jhet.2567

Month 2015
The Use of Chloroformamidine Hydrochloride as a Reagent for the
Synthesis of Guanidines from Electron Deficient Aromatic Amines
Ian Armitage, Mingkun Fu, Frederick Hicks, Adiseshu Kattuboina, Jennifer S. N. Li, Ashley McCarron,
and Lei Zhua
b
b
a
c
a
a
*
aChemical Development Laboratories Boston, Takeda Pharmaceuticals International Company, 35 Landsdowne Street,
Cambridge, MA 02139, USA
bAnalytical Development Laboratories Boston, Takeda Pharmaceuticals International Company, 35 Landsdowne Street,
Cambridge, MA 02139, USA
cChemical Development, Albany Molecular Research Incorporated, 21 Corporate Circle, Albany, NY 12212, USA
*
E-mail: ashley.mccarron@takeda.com
Additional Supporting Information may be found in the online version of this article.
Received July 24, 2015
DOI 10.1002/jhet.2567
Published online 00 Month 2015 in Wiley Online Library (wileyonlinelibrary.com).
In efforts to optimize a manufacturing process for an internal development compound, a clean, efficient
approach to guanidine synthesis using chloroformamidine hydrochloride was identified. To investigate the
general utility of this methodology towards electron-deficient aromatic amines, a set of favorable conditions
were developed from a series of screens, and the scope of the reaction was probed. The successful applica-
tion of this chemistry to a variety of pyridines, anilines, and heterocyclic compounds highlights its use as an
improved, alternative guanylation method for this often challenging set of aromatic amines.
J. Heterocyclic Chem., 00, 00 (2015).
Guanidines are of great importance in the pharmaceuti-
species in situ followed by reaction with an amine to affect
guanidine formation (Scheme 1, Method B) [4]. The main
limitation encountered was the small number of solvent op-
tions associated with commercially available acid solutions
and water in aqueous acids. Many of those solvents hin-
dered reagent and substrate solubility, reaction tempera-
ture, and hence reaction completion. In addition,
degradation of the active formamidine reagent occurred
over time as a result of reaction with the nucleophilic sol-
vents used. When this method was attempted on starting
material 1, this degradation was significant with EtOH as
solvent and portionwise dosing of 12 equivalents of
cyanamide over 26h resulted in only 45% conversion to
2. Further addition of cyanamide did not provide notable
improvement. These inefficiencies coupled with the
difficult separation of 1 from 2 reiterated the need for an
alternate methodology for the guanidine synthesis.
A literature search to identify alternative guanylation op-
tions provided examples of pyrimidine cyclizations from
aromatic amines with an adjacent carboxylate (Scheme 2)
[5] or other functional groups such as nitriles [6], using iso-
lated chloroformamidine hydrochloride 3. In these exam-
ples, 3 presumably undergoes attack by the free amine to
form a guanidine intermediate, which immediately cyclizes
to afford a fused pyrimidine moiety. Although isolation of
the presumed guanidine intermediate was not mentioned,
without functionality at the carbon beta to the amine posi-
tion, the reaction would stop at the guanidine intermediate
and it could potentially be isolated. Further supporting the
potential of this methodology, use of preformed 3 would
cal industry due to their many synthetic applications, often
being utilized as heterocyclic precursors, strong bases and
catalysts, and their alkaloids have exhibited a broad range
of antimicrobial and antitumor activities [1]. For these rea-
sons, much effort has gone into the development of facile,
efficient guanylating reagents. Many methods for the syn-
thesis of substituted guanidines reported in the literature of-
ten perform well with electron-rich amines, but only
modestly, if at all, with electron-deficient aromatic amines
[1]. This paper reports a clean, convenient one-pot synthe-
sis of guanidines from electron-deficient aromatic amines
using chloroformamidine hydrochloride as the guanylating
reagent.
The methodology presented herein evolved from the
need to identify and develop a scalable synthesis for a gua-
nidine intermediate 2 associated with an internal develop-
ment project (Scheme 1) [2]. The initial focus for this
transformation was on utilizing the two most common ap-
proaches to guanidines from electron-deficient aromatic
amines. Our first attempts involved the reaction of pyridine
amine 1 with bis-Boc-thiourea in the presence of activating
agents such as HgCl2, CuCl2, or Mukaiyama’s reagent,
followed by deprotection of the resulting Boc-protected
guanidine (Scheme 1, Method A) [3]. The use of stoichio-
metric amounts of metal salts, removal of by-products,
poor atom efficiency, column chromatography, and low
yields made this method inappropriate for further scale up.
The second guanylation strategy utilized cyanamide ac-
tivated by an acid, often HCl, to generate the formamidine
© 2015 HeteroCorporation

I. Armitage, M. Fu, F. Hicks, A. Kattuboina, J. S. N. Li, A. McCarron, and L. Zhu
Vol 000
Scheme 1. Desired transformation to guanidine intermediate 2.
with >99% purity. This new procedure for 2 was then suc-
cessfully scaled up to multiple kilograms in GMP produc-
tions. The simple translation of this method to our
previously challenging chemical transformation of 2 en-
couraged us to further investigate its utility as an effective
guanylation technique with other electron-poor aromatic
amine substrates.
Our work began with a comprehensive screen of the re-
action conditions utilizing 3-aminopyridine 4a, a com-
pound with structural simplicity and similarity to 1, as the
model substrate. The primary focus of the initial studies
was to identify the optimal solvent or solvent systems, tem-
perature(s), and chloroformamidine equivalents for this
transformation. The screen was heavily focused on polar
solvents because of the observed relative low solubility
of 3 in many common organic solvents and was carried
out over an elevated temperature range (60–120°C) for
the same reason. Attempts with acetic acid (Table 1, entry 1)
showed good solubility and conversion to the desired gua-
nidine product. Unfortunately, acetic acid also proved to
be reactive with 4a, resulting in the formation of an unde-
sired side product. The same was observed for solvents
such as DMF, DMSO, NMP, and DMAc (Table 1, entries
Scheme 2. Pyrimidine synthesis using chloroformamidine hydrochloride.
Table 1
circumvent the solvent and degradation issues observed in
previous attempts with in situ generation of 3, which al-
ready showed moderate success with substrate 1.
Screening of guanylation on model substrate 4a.
More recently, Chen et al. reported an example of direct
guanidination of a complicated aliphatic amine utilizing 3
[7]. Benefits mentioned in this reference also support our
efforts to apply the same methodology for electron-
deficient aromatic guanidine synthesis.
Applying the reported pyrimidine cyclization conditions
to compound 1 using 3, a user friendly, stable solid pre-
pared according to the literature [5,8], was highly success-
ful (Scheme 3). With dioxane or MeCN as solvent, the
reaction resulted in the exclusive formation of the desired
guanidine moiety 2 under reflux with slight excess of 3.
The only issue observed was low solubility of 3 and prod-
uct 2 in the solvents utilized, which resulted in heteroge-
neous reaction mixtures even at high temperature. The
solubility issue was readily resolved by introduction of an-
other polar solvent, with the optimal solvent combination
for 1 being MeCN/AcOH 10/1. Under these conditions,
full conversion was achieved at 80°C in less than 5 h, and
2 was isolated in 88% yield by filtration as the tris-HCl salt
Conversion (%)b
Temperature
Entrya
Solvent
(°C)
5a
4a
Imp(s).
1
2
3
4
AcOH
DMF
NMP
60
100
100
80
100
80
29
2
4
9
89
25
29
1
69
56
74
0
40
17
11
66
65
71
IPA
5
t-butanol
MeCN
MeCN/
AcOH
MeCN/
sulfolane
dioxane
dioxane/
AcOH
sulfolane
9
6
6c
7d
80
28
8c,d
80
>99
<1
0
9c
100
100
64
94
36
6
0
0
10c,d
11
100
97
2
1
aAll reactions conducted on a 2.1 mmol scale, using 1.2 eq. of 3 (Entry 5
used 1.8 eq. and Entry 10 used 1.4 eq.), in 10 vol. solvent (sampled over
24 h – table displays best conversion),
Scheme 3. Application of chloroformamidine hydrochloride conditions
to development compound 1.
bConversion was determined by LC/MS of the in-process reaction
mixture,
cHeterogeneous reaction at certain temperatures,
dSolvent ratio 10 : 1
Journal of Heterocyclic Chemistry
DOI 10.1002/jhet

Month 2015
Chloroformamidine HCl as a Reagent for the Synthesis of Guanidines
Table 2
2 and 3) [9]. However, in these instances, the reactions did
not achieve full conversion, and multiple impurities were
observed. Several higher boiling alcoholic solvents such
as Isopropyl Alcohol (IPA) (Table 1, entry 4) were
screened, but low conversion and undesired side-reactions
were encountered. Fewer impurities were observed with
the more hindered alcoholic solvents such as tert-butanol
(Table 1, entry 5). Despite preliminary favorable results,
tert-butanol was not included in the solvent list for later studies
due an observed phase separation that occurred between
the solvent and reagents at temperatures above 80°C.
Use of MeCN and dioxane (Table 1, entries 6 and 9),
both non-reactive solvents towards 3, showed clean, con-
version to 5a, with minimal side products observed. Simi-
lar to compound 1, low solubility of 3 occurred in these
solvents at certain temperatures, which resulted in hetero-
geneous reaction mixtures, and led to slow conversion of
1. These issues were eliminated, and higher conversions
were achieved by introducing a minimal amount of acetic
acid or sulfolane (Table 1, entries 7, 8, 10).
Scope of guanidine synthesis.
The introduction of sulfolane as a co-solvent also
highlighted its potential as a primary solvent. Sulfolane
(Table 1, entry 11) demonstrated better solubility over
the MeCN and dioxane solvent systems. In addition,
reactivity of sulfolane with 4a and 3 was minimal com-
pared with the other high boiling solvents DMF, DMSO,
NMP, and DMAc, leading to higher conversion and a
cleaner purity profile.
In attempts to further improve product conversion, the
optimal amount of 3 was investigated in the preferred sol-
vents highlighted by the first screen. Studies commenced
using a minimum of 1.2 equivalents of 3 because of poten-
tial decomposition and previously observed participation in
side-reactions [8]. Results showed an improvement in
product conversion when 3 was increased from 1.2 to 1.4
equivalents, especially at lower temperatures. Additional
increases in 3 over 1.4 equivalents, however, did not offer
consistent progress. Moreover, product isolation became
increasingly difficult with the need to remove excess
amounts of 3 present, and hence, 1.4 equivalents of 3 were
chosen to be the standard amount for further
experimentation [10].
The scope of the guanylation reaction was then explored
by applying the current best conditions to a set of pyri-
dines, anilines, and heteroaromatic amines with various
electron withdrawing and donating substituents as illus-
trated in Table 2. It should be noted that for the substrate
screen, reactions were progressed until conversion of
starting material ceased or significant impurity formation
was observed. Also, column chromatography was often
used to purify the desired products.
Conversion (%)b
5a-n
Isolated
yield (%)
Entrya SM
Solvent
sulfolane
MeCN
MeCN
dioxane/ AcOH
dioxane
dioxane
MeCN/ AcOH
sulfolane
sulfolane
MeCN
MeCN/dioxane/
sulfolane
MeCN/AcOH
MeCN
1d,g
2
4a
4b
4c
4d
4e
4f
4g
4h
4i
97
79
>99
8
98
78
61
0
95
68
91
0
83
55
59
0
3
4d
5
6d
7
8e
9c
10g
11f
74
80
0
48
59
0
4j
4k
12d
13
14
4l
4m
4n
54h
>99
69
0
47
42
sulfolane
aAll screening reactions were run using 0.5 g of 4 and 1.40 eq. of 3 at 80°C,
bConversion was determined by LC/MS of the in-process reaction mixture,
c60°C,
d100°C,
e120°C,
f100% impurity formation,
gIsolated as HCl salt (all others isolated as the freebase),
hThe conversion accounts for two compounds with the desired mass cor-
responding to the two potential products
Reaction outcome with various aminopyridine sub-
strates proved heavily dependent on the position of the
amine group relative to the pyridine nitrogen. The excellent
yields of 5a and c (Table 2, entries 1 and 3) demonstrated
that guanylation is very successful when the amine is in the
3-position. The addition of an electron withdrawing group
Journal of Heterocyclic Chemistry
DOI 10.1002/jhet

I. Armitage, M. Fu, F. Hicks, A. Kattuboina, J. S. N. Li, A. McCarron, and L. Zhu
Vol 000
to 3-aminopyridine hindered reaction slightly but still
managed to afford 5b (Table 2, entry 2) in a 68% yield.
Contrary to the 3-aminopyridine substrates, reactions with
pyridines containing the amine moiety at the 2- or 4-
position were poor presumably because of conjugation ef-
fects. Some formation of the desired guanidine product
was observed with 4-aminopyridine 4d (Table 2, entry 4),
while 2-aminopyridine showed virtually none. Prolonged
heating to force the reaction only resulted in impurity
formation.
specifically for the often difficult electron-deficient aro-
matic amines, and also expands upon the number of sol-
vent options benefitting reaction temperature and
substrate solubility.
EXPERIMENTAL
General methods. All starting materials and reagents
were purchased commercially and used without further
purification. NMR spectra were taken on either a 300 or
400 MHz NMR spectrometer. All HPLC analyses were
performed using an LC/MSD system equipped with a
reverse-phase column and a flow rate of 1.0mL/min.
Purification of compounds was performed on an
automated liquid chromatography purification system
using pre-packed silica columns or a preparative HPLC
system. Because of the unknown stability of the new
The aniline substrates all proved to efficiently undergo
guanylation using the current system. Most successful
was the electron-deficient aniline 4e (Table 2, entry 5),
with 83% yield of 5e. The more challenging 4f (Table 2,
entry 6), an aniline bearing multiple electron withdrawing
groups, exhibited slower reactivity than 4e, which was ex-
pected because of its highly electron-deficient nature, but
did offer reasonable conversion and isolated yield of 55%
desired product 5f after chromatography. Substrate 4g with
an electron donating group (Table 2, entry 7) resulted in a
slightly lower, but still moderate yield compared with 4e
because of impurity formation, however, no functional
group intolerance with the hydroxyl group was observed.
Some commercial heteroaromatic amines were also
explored as part of substrate scope investigation. The less
reactive aminopyrimidine 4h (Table 2, entry 8) showed
essentially no conversion to desired product 5h under the
optimal conditions, potentially because of conjugation ef-
fects, and resulted in only impurity formation under forced
conditions. Substrates 4i and 4j (Table 2, entries 9 and 10)
with the amine group on the benzene portion of the
compound proved successful. Moderate yields (48% and
59%, respectively) were obtained because of difficult
isolation. Further investigation into the effect of amino
group position on a heterocycle with more than one
heteroatom showed this to have a significant influence on
the reaction. Heterocycles with multiple heteroatoms were
prone to significant impurity formation, especially when
the amine moiety is located between the two heteroatoms
as can be seen with 4k (Table 2, entry 11). Heterocyclic
substrates with multiple reactions sites such as 4l (Table 2,
entry 12) underwent reaction at both sites. However, when
the additional sites were blocked, as in 4m (Table 2, entry
13), the reaction showed complete conversion. Unfortu-
nately, only a 55% yield was obtained because of loss of
material during the difficult isolation and the free basing pro-
cess. Guanylation was also possible with electron-rich het-
erocycles such as 4n (Table 2, entry 14), albeit with slightly
lower conversion and yield due to impurity formation.
In summary, carrying out this methodology on a variety
of different aromatic amines successfully proved the use of
chloroformamidine hydrochloride as a facile reagent for
the guanylation. This approach provides a clean and effi-
cient alternative to the current guanylation methods, more
compounds
synthesized,
high-resolution
mass
spectrometry (HRMS) was performed in-house instead of
elemental analysis, which would have required shipping
the compounds off-site. All HRMS data were acquired
using a mass spectrometer coupled with ultra-high
performance liquid chromatography (UHPLC) and
charged aerosol detector (CAD).
Chloroformamidine hydrochloride(3).
1,4-Dioxane
(4.16mol, 0.325L) and cyanamide (0.714mol, 30.0 g)
were added to a jacketed glass reactor and cooled to 0°C
while stirring at 450rpm.
A
solution of 4.0M
hydrochloric acid in 1,4-dioxane (2.34mol, 0.584L) was
added dropwise maintaining reaction temperature below
20°C. Upon addition completion, the mixture was stirred
at room temperature for 2h. The reaction was filtered and
solids washed with another portion of 1,4-dioxane
(2.08mol, 0.162L) . The cake was then dried in vacuo at
40°C resulting in 78.3 g (95.4%) of a white solid. Anal.
Calcd. for CH4Cl2N2: C, 10.45, H, 3.51, N, 24.37, Cl,
61.68. Found: Batch 1 (20.0 g scale) C, 10.93, H, 3.28,
N, 24.39, Cl, 61.63, Batch 2 (30.0 g scale) C, 10.27, H,
3.30, N, 24.54, Cl, 61.81.
General procedure for the synthesis of mono-substituted
guanidines. To a scintillation vial was added amine 4
(1.00 equivalent), solvent (5–25 volumes), and
chloroformamidine HCl 3 (1.40 equivalent). The vial was
flushed with nitrogen and heated with stirring at 80–100°C
until liquid chromatography–mass spectrometry (LC/MS)
indicated the reaction was complete or conversion of 4
ceased. Each substrate followed individual work-up
procedures to isolate guanidine 5 as the HCl salt or freebase.
1-(5-(3-(Dimethylamino)propyl)-2-methylpyridin-3-yl)guanidine
tris-hydrochloride (2). To a 50mL round-bottom flask was
added 5-(3-(dimethylamino)propyl)-2-methylpyridin-3-amine
2HCl
1 (4.35 mmol, 1.00 g), chloroformamidine HCl
(5.22mmol, 0.600g), and acetonitrile (12.0 mL). The flask
was then flushed with nitrogen. Acetic acid (1.20mL) was
Journal of Heterocyclic Chemistry
DOI 10.1002/jhet

Month 2015
Chloroformamidine HCl as a Reagent for the Synthesis of Guanidines
added, and the mixture was left to stir at room temperature for
an addition 15min. The reaction was heated to 80°C and
monitored by LC/MS until complete. The reaction was then
cooled to room temperature, and the solids isolated by
filtration under nitrogen. The solids were washed with
acetonitrile (2.00 mL) and dried in vacuo to give the tris-HCl
a 2-dram vial. The vial was flushed with nitrogen and
heated to 80°C for 70 h. The reaction was then cooled
to room temperature. Unreacted chloroformamidine
was removed by filtration and washed with minimal
acetonitrile. The filtrate was concentrated, dissolved in
minimal water, and then freebased with saturated
sodium bicarbonate solution. The mixture was left to
stir at room temperature for 30 min, and a fine white
precipitate formed. The solids were then filtered,
washed with another portion of water, and dried in
vacuo to give the freebase as a white solid (0.236 g,
1
salt as an off-white solid (1.32g, 88.0% yield). H NMR
(300MHz, DMSO-d6) δ 10.88 (s, 1H), 10.67 (s, 1H), 8.70
(d, J=1.7Hz, 1H), 8.37 (d, J=1.7Hz, 1H), 7.97 (s, 3H),
3.03 (dt, J=9.9, 5.7Hz, 2H), 2.84 (t, J=7.5Hz, 2H), 2.68 (d,
J= 4.7Hz, 6H), 2.65 (s, 3H), 2.05 (dq, J= 14.5, 7.2Hz, 2H).
13C NMR (75MHz, DMSO) δ 157.0, 149.9, 143.8, 139.7,
139.2, 134.1, 55.6, 42.3, 28.3, 24.5, 16.4. HRMS: m/z for
C12H24Cl3N5 (M+ +H) 236.1870, found 236.1865.
1
91% yield). H NMR (300 MHz, DMSO-d6) δ 8.43 (d,
J = 2.5 Hz, 1H), 7.88–7.80 (m, 1H), 7.75–7.68 (m, 1H),
7.54–7.51 (m, 1H), 7.49–7.39 (m, 2H), 5.69 (s, 4H).
13C NMR (75 MHz, DMSO) δ 154.6, 150.8, 144.3,
143.4, 129.6, 128.8, 127.2, 126.5, 126.2, 124.2. HRMS:
m/z for C10H10N4 (M+ + H) 187.0978, found 187.0973.
1-(4-(Trifluoromethyl)phenyl)guanidine (5e). 4-(Trifluoro
methyl)aniline (3.10mmol, 1.00g), 1,4-dioxane (15.0mL),
and chloroformamidine HCl (4.34mmol, 1.00g) were
added to a 20mL scintillation vial. The vial was flushed
with nitrogen and then heated to 80°C for 24h. Once
complete, the reaction was cooled to room temperature
and concentrated. The mixture was purified via column
chromatography with a 40g silica column using a 0–
10% MeOH/0.05% triethylamine in EtOAc system. The
product fractions were concentrated and dried in vacuo
to yield the freebase as an off-white solid (1.05g, 83.3%
yield). 1H NMR (300MHz, Methanol-d4) δ 7.78–7.67
(m, 2H), 7.47–7.36 (m, 2H). 13C NMR (75 MHz,
CD3OD) δ 156.2, 140.8, 127.8, 127.4, 125.9, 122.3.
HRMS: m/z for C8H8F3N3 (M+ + H) 204.0743, found
204.0742.
1-(Pyridin-3-yl)guanidine dihydrochloride (5a).
3-Amino-
pyridine (26.6mmol, 2.50g), sulfolane (25.0mL), and
chloroformamidine HCl (31.9mmol, 3.66 g) were added to a
40mL scintillation vial. The vial was flushed with nitrogen
and heated to 100°C until LC/MS showed full conversion of
starting material. Once complete, the mixture was transferred
to a 100 mL round-bottom flask with acetonitrile (25mL).
The resulting solids were isolated by filtration under nitrogen
and washed with a portion of acetonitrile (20 mL). To
remove the residual chloroformamidine, the material was
then taken up in MeOH (35 mL), heated to 60°C for 2 h and
then cooled to room temperature for an additional 2h. The
suspension was filtered and dried in vacuo to give the di-HCl
1
salt as an off-white solid (5.30g, 95.4% yield). H NMR
(400MHz, DMSO-d6) δ 10.98 (s, 1H), 8.88 (d, J=2.4Hz,
1H), 8.74 (dd, J=5.5, 1.2Hz, 1H), 8.34 (ddd, J=8.4, 2.5,
1.3Hz, 1H), 8.15 (s, 4H), 8.00 (dd, J=8.4, 5.4Hz, 1H). 13C
NMR (101 MHz, DMSO) δ 156.1, 139.7, 139.1, 138.8,
135.5, 127.3. HRMS: m/z for C6H8N4 (M+ +H) 137.0822,
found 137.0818.
1-(2-Nitro-4-(trifluoromethyl)phenyl)guanidine (5f). 4-Amino
-3-nitrobenzotrifluoride (2.42 mmol, 0.500 g), 1,4-dioxane
(5.00mL), and chloroformamidine HCl (3.40 mmol, 0.390 g)
was added to a 20 mL scintillation vial. The vial was flushed
with nitrogen and then heated to 100°C until all starting
material was consumed as indicated by LC/MS. A second
portion of chloroformamidine HCl (3.40 mmol, 0.390 g) was
then added to push the reaction to completion. Once
complete, unreacted chloroformamidine was removed by
filtration and washed with minimal dioxane. The crude
solution was concentrated, dissolved in minimal water, and
freebased with saturated sodium bicarbonate solution. The
mixture was concentrated and run down a 24 g silica column
using 0.05% triethylamine in EtOAc to remove salts and then
another 24g column using 50–80% EtOAc in hexanes to
purify. The tubes containing product were concentrated and
dried in vacuo to produce the freebase as an orange solid
1-(6-(Trifluoromethyl)pyridin-3-yl)guanidine (5b). 3-Amino-
6-(trifluoromethyl)pyridine (3.08mmol, 0.500 g), acetonitrile
(12.5 mL), and chloroformamidine HCl (4.32mmol, 0.496 g)
were added to a 20mL scintillation vial and flushed with
nitrogen. The reaction mixture was heated to 80°C until all
starting material was consumed as indicated by LC/MS. The
reaction mixture was cooled to ambient, the unreacted
chloroformamidine filtered, and the filtrate concentrated. The
filtrate was freebased using saturated sodium bicarbonate
solution and water (1 mL) added to dissolve any salts. The
slurry was filtered and clean freebase product was isolated as
an off-white solid (0.429 g, 68.1% yield). 1H NMR
(400MHz, DMSO-d6) δ 8.11 (d, J=2.5Hz, 1H), 7.69 (d,
J= 8.3 Hz, 0H)? int of 0.22, 7.58 (dd, J=8.5, 0.7Hz, 1H),
7.28 (dd, J=8.5, 2.4Hz, 1H), 5.65 (s, 3H). 13C NMR
(75MHz, DMSO) δ 155.0, 151.4, 145.6, 129.6, 124.8,
121.3, 110.0. HRMS: m/z for C7H7F3N4 (M+ +H) 205.0696,
found 205.0690.
1
(0.330g, 54.8% yield). H NMR (300MHz, DMSO-d6) δ
1-(Quinolin-3-yl)guanidine (5c).
(1.39 mmol, 0.200 g), acetonitrile (5.00 mL), and chloro
formamidine HCl (1.94 mmol, 0.233 g) were added to
3-Aminoquinoline
7.97 (dd, J= 2.3, 1.0 Hz, 1H), 7.61 (dd, J= 8.8, 2.3 Hz, 1H),
7.15 (dd, J=8.7, 1.0 Hz, 1H), 5.90 (s, 4H). 13C NMR
(75 MHz, DMSO) δ 155.8, 149.8, 143.2, 129.1 (d, J=3.8Hz),
Journal of Heterocyclic Chemistry
DOI 10.1002/jhet

I. Armitage, M. Fu, F. Hicks, A. Kattuboina, J. S. N. Li, A. McCarron, and L. Zhu
Vol 000
126.8, 122.1 (d, J =4.5 Hz), 118.3, 117.9. HRMS: m/z for
C8H7F3N4O2 (M+ + H) 249.0594, found 249.0586.
(dd, J= 8.3, 1.9 Hz, 1H). 13C NMR (75MHz, DMSO) δ
156.8, 156.4, 156.2, 135.1, 128.1, 123.8, 123.5, 122.4, 122.3,
121.6, 120.4, 112.1, 108.7. HRMS: m/z for C13H11N3O
(M+ + H) 226.0975, found 226.0939.
1-(2-Hydroxyphenyl)guanidine (5g). To a 20 mL scintillation
vial was added 2-aminophenol (9.16 mmol, 1.00 g), chloro
formamidine HCl (12.3 mmol, 1.47g), and a co-solvent
system of acetonitrile (10.0 mL) and acetic acid (1.00mL).
The vial was flushed with nitrogen and then heated to 80°C
for 24h. Once complete, unreacted chloroformamidine was
removed by filtration and washed with minimal acetonitrile.
The filtrate was concentrated and freebased using saturated
sodium bicarbonate solution. The mixture was concentrated,
dissolved in methanol, and passed through a silica plug
(Whatman C18 ODS-3 Partisil, Clifton, NJ) to remove salts.
The material was concentrated and further purified by prep
HPLC to yield the freebase product as an off-white solid
1-(1-Methyl-1H-indazol-3-yl)guanidine (5m).
1-Methyl-
1H-indazol-3-amine (0.652mmol, 0.100 g), acetonitrile
(2.50 mL), and chloroformamidine HCl (0.754mmol,
0.0867g) were added to a reaction vial. The vial was flushed
with nitrogen and heated to 80°C for 24 h. The reaction was
then cooled to room temperature. Solids were collected by
filtration and washed with minimal acetonitrile. The material
was dissolved in minimal water and then freebased with
saturated sodium bicarbonate solution. The mixture was left
to stir at room temperature overnight, filtered, and washed
with a small amount of water. The solids were dried in
vacuo to give the freebase as a pale yellow solid (0.0583 g,
1
(1.02 g, 58.5% yield). H NMR (300 MHz, Methanol-d4) δ
1
7.26–7.14 (m, 2H), 7.01–6.83 (m, 2H). 13C NMR (75MHz,
CD3OD) δ 157.1, 152.8, 129.2, 127.6, 121.2, 119.8, 116.3.
HRMS: m/z for C7H9N3O (M++ H) 152.0818, found 152.0813.
47.2% yield). H NMR (300MHz, DMSO-d6) δ 7.58 (dt,
J=8.0, 1.0Hz, 1H), 7.35 (dt, J=8.4, 1.0Hz, 1H), 7.32–7.23
(m, 1H), 6.92 (ddd, J=7.8, 6.7, 1.0Hz, 1H), 6.29 (s, 2H),
3.80 (s, 3H). 13C NMR (75 MHz, DMSO) δ 155.5, 150.5,
140.2, 126.4, 120.6, 119.0, 118.0, 108.5, 34.8. HRMS: m/z
for C9H11N5 (M+ +H) 190.1087, found 190.1084.
1-(Benzo[d]oxazol-4-yl)guanidine (5i).
1,3-Benzoxazol-
4-amine (3.73 mmol, 0.500 g), sulfolane (2.50mL), and
chloroformamidine HCl (5.22mmol, 0.600 g) were added to
a 20 mL scintillation vial. The vial was flushed with nitrogen
and heated to 60°C for 2 h. The reaction was then cooled
to room temperature. A small portion of water was added
to the vial to dissolve solids, and the solution was
transferred to a separatory funnel. The aqueous layer was
washed three times with DCM and then concentrated.
Saturated sodium bicarbonate solution was added to
freebase the material, and the mixture is concentrated once
again. The product was isolated via prep HPLC and dried
in vacuo to give the freebase as an off-white solid (0.314 g,
1-(Benzo[b]thiophen-3-yl)guanidine (5n). 1-Benzothiophen-
3-amine hydrochloride (0.517mmol, 0.100g), sulfolane
(0.50 mL), and chloroformamidine HCl (0.724mmol,
0.0832 g) were added to a reaction vial. The vial was flushed
with nitrogen and heated to 80°C for 36 h. The reaction was
then cooled to room temperature. A small portion of water
was added to the vial to dissolve solids, and then the solution
was transferred to a separatory funnel. The aqueous layer was
washed three times with DCM and then concentrated.
Saturated sodium bicarbonate solution was added slowly
resulting in a precipitate. The slurry was stirred at room
temperature for 2h and then filtered under nitrogen. The
solids were washed with minimal water and dried in vacuo to
give the freebase as a white solid (0.0413g, 41.8% yield). 1H
NMR (300 MHz, Methanol-d4) δ 7.89–7.81 (m, 1H), 7.73–7.66
(m, 1H), 7.44–7.34 (m, 2H), 7.22 (s, 1H). 13C NMR (75MHz,
CD3OD) δ 155.3, 139.1, 135.4, 124.6, 123.8, 122.6, 120.8,
117.3, 110.0. HRMS: m/z for C9H9N3S (M++H) 192.0590,
found 192.0586.
1
48% yield). H NMR (300MHz, Methanol-d4) δ 8.03 (s,
1H), 7.12–6.99 (m, 2H), 6.63 (dd, J=6.3, 2.3Hz, 1H). 13C
NMR (75MHz, CD3OD) δ 163.1, 147.9, 141.1, 140.2,
128.9, 124.7, 108.2, 107.0. HRMS: m/z for C8H8N4O
(M+ +H) 177.0771, found 135.0549 (consistent with 4i).
HRMS data for the product could not be obtained, likely
because of the instability of this compound. All other data
support the formation of the desired guanidine 5i.
1-(Dibenzo[b,d]furan-3-yl)guanidine hydrochloride (5j). Di-
benzo[b,d]furan-3-amine (0.546mmol, 0.100g), acetonitrile
(2.50 mL), and chloroformamidine HCl (0.764 mmol,
0.0878g) were added to a 20 mL scintillation vial. The vial
was flushed with nitrogen and heated to 80°C for 36 h. The
reaction was then cooled to room temperature. The mixture
was filtered and washed with minimal acetonitrile to obtain
crude product. The solids were slurried in water (10 mL) at
0°C for 30min. The slurry was filtered; the solids washed
with a small portion of cold water and dried in vacuo to
obtain the mono-HCl salt as an off-white solid (0.418g, 59%
Acknowledgment. A special thank you to Ashok Patil and Matthew
Jones from the Oncology Chemistry Department at Takeda
Pharmaceuticals for assistance in compound separation and isolation.
REFERENCES AND NOTES
[1] (a) Katritzky, A. R.; Rogovoy, B. Arkivoc 2005, Part iv 49; (b)
Katritzky, A. R.; Khashab, N.; Bobrov, S. Helv Chim Acta 2005, 88,
1664.
[2] (a) Bharathan, I. T. et al. U.S. Patent 7,998,952 B2, August 16,
2011; (b) Bharathan, I. T. et al. U.S. Patent 8,268,992 B2, September 18,
2012; (c) Bharathan, I. T. et al. U.S. Patent 8,507,667 B2, August 13, 2013.
[3] (a) Kim, K. S.; Qian, L. Tetahedron Lett 1993, 34, 7677; (b)
Yong, Y. F.; Kowalski, J.; Lipton, M. J Org Chem 1997, 62, 1540.
1
yield). H NMR (300MHz, DMSO-d6) δ 10.40 (s, 1H),
8.22–8.10 (m, 2H), 7.74–7.65 (m, 4H), 7.61 (d, J=1.8Hz,
1H), 7.55–7.48 (m, 1H), 7.40 (td, J= 7.5, 1.0 Hz, 1H), 7.25
Journal of Heterocyclic Chemistry
DOI 10.1002/jhet

Month 2015
Chloroformamidine HCl as a Reagent for the Synthesis of Guanidines
[4] (a) Braun, C. E. J Am Chem Soc 1933, 55, 1280; (b) Andrews, D.;
Finlay, M. R.; Green, C.; Jones, C.; Oza, V. PCT Int Appl WO 2006/095159
A1, September 14, 2006.
[5] Jones, A.; Wilson, F.; Dyke, H.; Price, S.; Cramp, S.; Szabo, A.;
Repasi, J. European Patent 1 925 619 A1, May 28, 2008.
[6] Colbry, N.; Elslager, E.; Werbel, L. J Heterocycl Chem 1984,
21, 1521.
[7] Jia, F.; Hong, J.; Sun, P.; Chen, J. Chen W Synth Commun
2013, 43, 2641.
[8] Before any work could be performed, an analysis method for
the determination of chloroformamidine quality was necessary. To date,
the only assessment of quality was elemental analysis and this was not
considered sufficient for our purposes. See supporting information for full
chloroformamidine hydrochloride analysis method development, synthe-
sis, and stability details.
[9] See Supporting Information for data from full condition screen.
[10] See Supporting Information for data from the full
chloroformamidine hydrochloride equivalent screen.
Journal of Heterocyclic Chemistry
DOI 10.1002/jhet