Synthesis of 4-Substituted Phthalazin-1(2H)-ones from 2-Acylbenzoic Acids: Controlling Hydrazine in a Pharmaceutical Intermediate through PAT-Guided Process Development

Article abstract of DOI:10.1021/acs.oprd.5b00135

A simple one-pot, two-step process for the conversion of 2-acylbenzoic acids to phthalazin-1(2H)-ones was developed. A robust process was required that delivered the final isolated solid with consistently low levels of residual hydrazine, for further processing to the final drug substance. An in situ formed intermediate was critical to control reactivity and allowed for the controlled crystallization that prevented entrainment of hydrazine. Leveraging Process Analytical Technology (PAT), we investigated the reaction profile with in situ IR and Power Compensation Calorimetry (PCC) to aid development prior to a successful scale-up.

Full text of DOI:10.1021/acs.oprd.5b00135

Communication
pubs.acs.org/OPRD
Synthesis of 4‑Substituted Phthalazin-1(2H)‑ones from 2‑Acylbenzoic
Acids: Controlling Hydrazine in a Pharmaceutical Intermediate
through PAT-Guided Process Development
Steven M. Mennen,*,† Melody L. Mak-Jurkauskas,‡ Matthew M. Bio,† L. Steven Hollis,‡ Kelly A. Nadeau,‡
Andrew M. Clausen,‡ and Karl B. Hansen†
‡
†Process Development and Attribute Sciences, Amgen, Inc., 360 Binney Street, Cambridge, Massachusetts 02142, United States
S
* Supporting Information
Scheme 1. Synthesis of Phthalazin-1(2H)-one 5 from 2-
Acylbenzoic Acid
ABSTRACT: A simple one-pot, two-step process for the
conversion of 2-acylbenzoic acids to phthalazin-1(2H)-
ones was developed. A robust process was required that
delivered the final isolated solid with consistently low
levels of residual hydrazine, for further processing to the
final drug substance. An in situ formed intermediate was
critical to control reactivity and allowed for the controlled
crystallization that prevented entrainment of hydrazine.
Leveraging Process Analytical Technology (PAT), we
investigated the reaction profile with in situ IR and Power
Compensation Calorimetry (PCC) to aid development
prior to a successful scale-up.
INTRODUCTION
■
The prevalence of phthalazin-1(2H)-ones have steadily
increased in the pharmaceutical industry and have been
components of investigational therapeutic programs for
antiulcer,1 antiasthmatic,2 antimicrobial,3 anti-inflammatory,4
hypertension,5 and obesity.6 Literature examples for the
synthesis of aryl phthalazin-1(2H)-ones typically begin from
2-acylbenzoic acid derivatives and rely on the use of hydrazine
or substituted hydrazine derivatives as a key raw material.7−10
In a recent development project we required a simple, scalable
synthesis of phthalazin-1(2H)-one 5, a key intermediate in the
preparation of a clinical candidate (Scheme 1).11 Since
hydrazine is genotoxic,12 a high degree of process control was
required to ensure patient exposure would not exceed a daily
exposure as defined by the staged threshold of toxicological
concern (TTC).13
We have previously reported the preparation of 2-
acylbenzoic acid 4 by reaction of the in situ-formed Grignard
2 and phthalic anhydride (3) (Scheme 1).14 Subsequent
elaboration of 2-acylbenzoic acid 4 with 5.0 equiv of aqueous
hydrazine in 5 V (V = mL solvent/g solute) ethanol at reflux
afforded phthalazin-1(2H)-one 5 in high yield (Scheme 1). The
low solubility of 5 under the reaction conditions (1.5 mg/mL at
80 °C) led to a rapid uncontrolled crystallization and
inconsistent levels of hydrazine retained in the isolated
crystalline product 5 (529−3108 ppm). We hypothesized that
both the superstoichiometric hydrazine and an uncontrolled
reactive crystallization were the root causes of hydrazine
entrainment. We looked at improving the reactivity of 4 with in
situ formation of an activated intermediate and consequently
reduced hydrazine to a stoichiometric charge. Additionally,
identification of a solvent system with increased solubility of 5
was required to allow for a robust isolation. This improved
process achieved consistently low levels of hydrazine in isolated
5 and was characterized by in situ IR and Power Compensation
Calorimetry (PCC). This process was further demonstrated in
other 4-substituted phthalazin-1(2H)-ones.
RESULTS AND DISCUSSION
■
To control the level of hydrazine in the final drug substance,
based on its rejection in the downstream process, we set a
target of ≤110 ppm hydrazine for isolated 5. To achieve this
goal, we first investigated limiting hydrazine in the original
process to a stoichiometric charge to minimize the potential for
entrainment. Spontaneous crystallization occurred, however,
within the first 1 h of the reaction (Scheme 2). Unfortunately,
after 21 h the reaction only achieved a 66% yield of 5.
Therefore, limiting stoichiometry was not a sufficient solution
so we investigated an approach that preactivated 4 as the
acylimidazole.
In Situ Pre-Activation of Acylbenzoic Acid 4. We next
investigated a process which involved the in situ preactivation
of the starting acylbenzoic acid. The objective was to identify a
Received: April 28, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.oprd.5b00135
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Communication
Scheme 2. Phthalazin-1(2H)-one Cyclization with
Stoichiometric Hydrazine
reactive intermediate which rapidly reacted with stoichiometric
hydrazine. 1,1-Carbonyldiimidazole (CDI) rapidly converted 4
to the intermediate acylimidazole 6, which upon addition of
hydrazine, efficiently converted to 5 (Scheme 3).15 The
reaction conditions investigated were limited to non-
nucleophilic, polar aprotic solvents. When either acetonitrile
(MeCN) or N,N-dimethylformamide (DMF) was employed
both 4 and 6 were fully soluble; however, the solubility of
product 5 differed greatly in MeCN (0.42 mg/mL at 20 °C) vs
DMF (25 mg/mL at 20 °C). The low solubility of 5 in MeCN
resulted in instantaneous crystallization at the beginning of the
addition of hydrazine. Agglomerated crystals of 5 were
observable by polarized light microscopy (PLM, Figure 1),
and the final isolated product had hydrazine levels that ranged
between 719 and 4275 ppm. When the reaction was performed
in DMF, nucleation of 5 was not instantaneous, and
crystallization was observed following completion of hydrazine
addition. When 5 was isolated from the DMF crystallization,
the particles were significantly less agglomerated, as observed
by PLM (Figure 2), and hydrazine entrainment was
consistently <110 ppm. As a result, DMF was chosen for
further development.
Figure 1. PLM of crystallization from MeCN (100× magnification).
Figure 2. PLM of crystallization from DMF (100× magnification).
The solubility of phthalazin-1(2H)-one 5 was studied in a
range of pure solvents to identify a solvent combination for the
crystallization conditions which would provide high product
recovery and maximize the volumetric efficiency of the process.
In combination with DMF, water was selected as the
antisolvent because of the low solubility of 5 and the high
solubility of hydrazine and imidazole in water. As shown in
Figure 3, starting in 100% DMF, the addition of water to 70%
DMF reduces the solubility of 5 from 25 mg/mL to 1.5 mg/mL
at 20 °C.
Process Design. Our process was planned based on the
hypothesis that hydrazine is rapidly consumed at an addition-
controlled rate. Formation of activated acylimidazole 6 was
carried out by charging 1.1 equiv of CDI in four approximately
equal portions over 1 h to a solution of 4 in DMF (8 V) at 20
°C (Scheme 3). Off-gassing of CO2 immediately followed each
CDI charge, and was easily controlled by the portion-wise
addition. The solution of acylimidizole was then aged for 1 h
before a linear addition of 1.1 equiv of hydrazine (35 wt %
aqueous). The equilibrium solubility of phthalazin-1(2H)-one
ranged from 20−27 mg/mL and varied depending on the water
activity of the end-of-reaction solution. The increased solubility
Figure 3. Solubility of 5 vs percent DMF/water (v/v).
compared to the original ethanol process ensured crystallization
did not immediately occur at the initiation of hydrazine
addition. Phthalazin-1(2H)-one 5 is in the metastable zone at
the end of hydrazine addition and typically holds super-
saturation until hydrazine addition is completed. Subsequent
linear addition of 2.5 V water over 1.5 h lowered the
supernatant concentration to <10 mg/mL furnishing product
5 in >80% isolated yield in high purity.
Scheme 3. Two-Step CDI-Meditated Phthalazin-1(2H)-one Formation
B
DOI: 10.1021/acs.oprd.5b00135
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Communication
Scheme 4. CDI Mediated Phthalazin-1(2H)-one Synthesis
Figure 4. In situ IR and power compensation calorimetry of acylimidazole (8) formation, Step Ac,d; a: left axis: red = 2-acylbenzoic acid (7), blue =
◆
acylimidazole (8); green = CDI; b: right axis: black = enthalpy uncorrected for CO2 off-gassing; c: the symbol (green ) indicates charges of solid
CDI; d: component trends identified using ConcIRT (Mettler-Toledo).
Figure 5. In situ IR and power compensation calorimetry of phthalazin-1(2H)-one (9) formation, Step Bc; a: left axis: blue = 8 (1789 cm−1, CO),
−1
⧫
red = 9 (790 cm , lactam N−H); b: right axis: black = enthalpy uncorrected for heat of mixing; c: the symbols ( ) indicate the duration of linear
addition of hydrazine.
C
DOI: 10.1021/acs.oprd.5b00135
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Communication
Scheme 5. Rearrangement of Acylimidazole to N,O-Acetal Reactive Intermediate
Figure 6. Rearrangement of acylimidazole to N,O-acetal by 13C NMR (Signals are labeled with compound numbers in parentheses).
To facilitate our design, we needed to ensure hydrazine
reacts at an addition-controlled rate. We leveraged in situ IR in
tandem with calorimetry to understand the kinetics and
thermodynamics of the two-step, one-pot chemical process
and to monitor consumption of hydrazine. Phthalazin-1(2H)-
one formation was studied in a stepwise fashion using readily
available commercial 2-acylbenzoic acid 7 (Scheme 4).
Experiments were monitored using a Mettler Toledo IC10
ReactIR equipped with a DiComp probe collecting 256 scans/
min. Power Compensation Calorimetry (PCC)16 experiments
were performed using a Syrris Atlas Calorimeter equipped with
a 250 mL vacuum jacketed reactor and a 25W heating probe;
data were collected at a frequency of 1 Hz.
Acylimidiazole Formation, Step A. The ConcIRT
function in the Mettler Toledo iCIR software was used to
track consumption of the 2-acylbenzoic acid and CDI, and
formation of acylimidazole (Figure 4). The profiles of the
individual components are consistent with rapid reaction of 2-
acylbenzoic acid (red) and CDI (green) to form acylimidizole
(blue). The in situ IR trends are in agreement with the
calorimetry data (black) collected simultaneously using PCC.
Rapid heat generation was observed after each charge of solid
D
DOI: 10.1021/acs.oprd.5b00135
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Communication
Scheme 6. Pilot-Plant Campaign
Figure 7. Substrate scope of CDI mediated synthesis of 4-substituted phthalazin-1(2H)-ones. (a) Isolated yield after crystallization; (b) assay yield
1
by quantitative H NMR.
CDI, in concert with rapid consumption of CDI. The overall
enthalpy of acylimidizolide formation was 11.65 kJ/mol without
correction for CO2 liberation. Additionally, the downward
sloping trend for 8 observed by in situ IR indicated that the
reactive intermediate might have a stability liability that
required an additional investigation (see below).
Phthalazin-1(2H)-one Formation, Step B. Conversion of
8 to 9 was followed by peak trending in the solvent-subtracted
in situ IR (Figure 5) trace. Conversion of acylimidazole (1789
cm−1) to phthalazin-1(2H)-one (790 cm−1) was consistent with
rapid consumption of hydrazine. Comparison of the calorim-
etry data with in situ IR trends indicated that the reaction was
nearly complete at the end of the approximately 1 h hydrazine
addition. The overall enthalpy of phthalazin-1(2H)-one
formation was measured to be 112.20 kJ/mol, which was
uncorrected for the heat of mixing of aqueous hydrazine and
DMF.
While formation of acylimidazole 8 was facile, our study
indicated potential instability of the reactive intermediate (see
Figure 4). While monitoring the stability of our desired system
for extended hold times, we observed the appearance of N,O-
acetal 10 (Scheme 5).17 We then studied the reaction of 2-
acylbenzoic acid 4 and CDI in DMF-d7 using NMR techniques
to gain an understanding of N,O-acetal 10 formation. The
extent of N,O-acetal formation was followed over several days,
and the reaction products were completely characterized by 1-
D and 2-D NMR studies (Figure 6; see Supporting
Information). Data are consistent with the kinetic formation
of 6 followed by rearrangement to N,O-acetal 10. Examination
of the downfield region of the 13C spectrum allowed conversion
of 6 to 10 to be tracked and required greater than 6 days at 25
°C to achieve full conversion to 10. We then demonstrated that
isolated 10 rapidly reacts with hydrazine to form phthalazin-
1(2H)-one 5 with comparable rate and yield as 6, which
minimized the risk of a long hold time negatively impacting the
batch (see Experimental Section).
Pilot Plant Campaign. With an improved understanding of
the kinetic and thermodynamic profile of the phthalazin-1(2H)-
one process, we initiated a demonstration campaign to produce
the desired phthalazin-1(2H)-one 5 (Scheme 6). Formation of
acylimidazole 6 proceeded as expected based on our develop-
ment studies. Crystallization initiated at the predicted time,
following complete addition of aqueous hydrazine. Phthalazin-
1(2H)-one 5 was isolated in 82% yield (3.47 kg) as a crystalline
solid, the purity was 99.7% AUC containing <110 ppm
hydrazine which fell within the target for downstream
processing.
Substrate Scope. The two-step CDI-mediated process was
developed specifically to maximize rejection of hydrazine in the
synthesis of 5; however, we wanted to understand the synthetic
limitations of the process for related phthalazin-1(2H)-ones. As
shown in Figure 7, the process performed well to afford a
variety of aryl-substituted phthalazin-1(2H)-ones (5, 9−9d)
tolerating a range of functional groups. Simple alkyl (9e) or
C7−H (9f) phthalazin-1(2H)-ones would require further
optimization for preparative scale; however, our process allows
E
DOI: 10.1021/acs.oprd.5b00135
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Communication
the synthesis in modest yield. The presence of a nucleophilic
phenol was well-tolerated, affording 9d in 86% isolated yield.
Interestingly, the strained tetracyclic fluorenone 9c was easily
constructed and isolated in 84% yield. In all cases hydrazine was
controlled to <200 ppm in the final isolated solid when a 1.5 h
addition of hydrazine was employed. Compounds 9b and 9c
both exhibit low solubility and crystallized during the 1.5 h
hydrazine addition. Prolonged hydrazine addition times of 4
and 6 h were studied for compounds 9 and 9b along with
seeding and it was shown that hydrazine can be controlled to
<10 ppm by minimizing the level of supersaturation for
compounds with low solubility.
cutaneous absorption. Consult the current MSDS prior to use and
ensure appropriate personal protective equipment). The slurry was
aged for 12 h at 20 °C and after a 12 h hold, water (3.0 V) was
added over 2 h by pump. Following a 3 h hold, the slurry was
filtered by vacuum using a medium porosity sintered glass filter.
The cake was washed twice with 60:40 (v/v) water: N,N-
dimethylformamide (1.0 V) followed by a water (4.0 V) wash.
The filter cake was dried on the filter by pulling dry nitrogen
through the cake for 12 h whereupon phthalazin-1(2H)-one
was isolated and characterized.
4-Phenylphthalazin-1(2H)-one (9). 9 was prepared follow-
ing the general procedure employing 2-benzoylbenzoic acid
(22.0 g, 97.0 mmol) as the acylbenzoic acid. 4-Phenyl-
phthalazin-1(2H)-one (9) (20.29 g, 94% yield) was isolated
as a colorless crystalline solid: GC-MS <10 ppm hydrazine; mp
CONCLUSION
■
In summary, a safe and simple process was developed for the
conversion of 2-acylbenzoic acids to phthalazin-1(2H)-ones
using an acylimidazole as an activated intermediate. The mild
conditions leveraged a fully soluble in situ-generated activated
acylimidazole. The acyl imidazole was found to react with
aqueous hydrazine at an addition-controlled rate which allowed
hydrazine to be reduced from >5.0 equiv to 1.1 equiv. A robust
crystallization was developed to control residual hydrazine
levels in the isolated solid, providing a critical control point to
ensure product quality and the process was successfully
demonstrated on multikilogram scale.
1
(DSC) 241.92 °C; H NMR (400 MHz, DMSO-d6, δ): 12.83
(s, 1H), 8.34 (t, J = 4.5 Hz, 1H), 7.88 (app t, J = 4.2 Hz, 2H),
7.66 (m, 1H), 7.59−7.52 (m, 5H); 13C NMR (100 MHz,
DMSO-d6, δ): 159.1, 146.3, 135.0, 133.5, 131.5, 129.2, 128.9,
128.8, 128.4, 127.9, 126.4, 126.0; HRMS-TOF (m/z): [M +
H]+ calcd for C14H11N2O, 223.0866; found, 223.0873.
4-(4-Chlorophenyl)phthalazin-1(2H)-one (9b). 9b was
prepared following the general procedure employing 2-(4-
chlorobenzoyl)benzoic acid (46.0 g, 176 mmol) as the
acylbenzoic acid. Crystallization occurred approximately half
way through the 1.5 h addition of hydrazine. 4-(4-
Chlorophenyl)phthalazin-1(2H)-one (9b) (40.96 g, 90%
yield) was isolated as a colorless crystalline solid: GC-MS =
EXPERIMENTAL SECTION
■
1
197 ppm hydrazine; mp (DSC) 273.90 °C; H NMR (400
General. Reactions were conducted under an atmosphere of
nitrogen with a suitable outlet to accommodate modest
pressure changes. Reaction temperatures were monitored by
an internal thermocouple. Reaction progress was determined by
LCMS analysis of derivatized18 reaction mixture and analysis by
HPLC-MS using an XBridge C18, 3.0 mm × 50 mm, 2.5 μm
column, with a 5−100% gradient method using 0.1% (v/v)
formic acid/water and 0.1% (v/v) formic acid/acetonitrile as
mobile phases. HRMS (ESI-TOF) spectra were obtained using
Agilent 1100 systems. Melting points were determined using a
TA Instruments Q200 DSC at 10 °C/min ramp. Hydrazine was
monitored using a headspace GC-MS method.19 Power
compensation calorimetry (PCC) was executed in a glass-
lined jacked reactor insulated with vacuum jacket. PCC
temperature was maintained at a batch temperature of 20.00
0.05 °C and employed a 25 W heating probe; data were
collected at a frequency of 1 Hz. In situ IR experiments were
conducted using a Mettler-Toledo iC10 ReactIR affixed with a
DiComp probe collecting 256 scans/min.
MHz, DMSO-d6, δ): 12.87 (s, 1H), 8.34 (m, 1H), 7.91−7.88
(m, 2H), 7.67−7.59 (m, 5H); 13C NMR (100 MHz, DMSO-d6,
δ): 159.1, 145.2, 133.9, 133.7, 133.6, 131.6, 131.1, 128.7, 128.5,
127.8, 126.3, 126.0; HRMS-TOF (m/z): [M + H]+ calcd for
C14H10ClN2O, 257.0476; found, 257.0483.
Indeno[1,2,3-de]phthalazin-3(2H)-one (9c). 9c was pre-
pared following the general procedure employing 9-oxo-9H-
fluorene-1-carboxylic acid (1.01 g, 4.49 mmol) as the
acylbenzoic acid. Crystallization quickly occurred at the
beginning of hydrazine addition. Indeno[1,2,3-de]phthalazin-
3(2H)-one (9c) (0.826 g, 84% yield) was isolated as a yellow
crystalline solid: GC-MS = 43 ppm hydrazine; mp (DSC)
268.13 °C; 1H NMR (400 MHz, DMSO-d6, δ): 12.73 (s, 1H),
8.18 (d, J = 7.1 Hz, 1H), 7.95 (d, J = 7.3 Hz, 1H), 7.91 (d, J =
7.8 Hz, 1H), 7.81 (app t, J = 8.2, 7.3 Hz, 1H), 7.49 (t, J = 7.2
Hz, 1H), 7.43 (d, J = 7.3 Hz, 1H); 13C NMR (100 MHz,
DMSO-d6, δ): 159.7, 146.8, 139.8, 136.0, 134.2, 133.5, 130.4,
130.2, 128.9, 125.8, 125.0, 123.9, 122.8, 121.7; HRMS-TOF
(m/z): [M + H]+ calcd for C14H9N2O, 221.0709; found,
221.0717.
4-(4-(Dibutylamino)-2-hydroxyphenyl)phthalazin-1(2H)-
one (9d). 9d was prepared following the general procedure
employing 2-(4-(dibutylamino)-2-hydroxybenzoyl) benzoic
acid (20.0 g, 54.1 mmol) as the acylbenzoic acid. 4-(4-
(Dibutylamino)-2-hydroxyphenyl)phthalazin-1(2H)-one (9d)
containing 0.84 equiv N,N-dimethylformamide (19.84 g, 86%
purity adjusted yield) was isolated as a yellow solid: GC-MS
<10 ppm hydrazine; mp (DSC) 85.03 °C, 99.20 °C, 167.38 °C;
1H NMR (400 MHz, DMSO-d6, δ): 12.56 (s, 1H), 9.17 (s,
1H), 8.26 (m, 1H), 7.80 (ddd, J = 7.6, 5.2, 1.8 Hz, 2H), 7.50
(m, 1H), 7.01 (d, J = 8.4 Hz, 1H), 6.26−6.22 (m, 2H), 3.25 (t,
J = 7.8 Hz, 4H), 1.58−1.51 (m, 4H), 1.34 (dd, J = 14.9, 7.5 Hz,
4H), 0.94 (t, J = 7.3 Hz, 6H); 13C NMR (100 MHz, DMSO-d6,
δ): 159.4, 156.1, 149.7, 145.9, 132.8, 131.4, 130.9, 130.4, 127.8,
General Procedure. Acylbenzoic acid (1.00 equiv) was
charged to a reactor affixed with overhead agitation and a
condenser. The reactor was inerted with nitrogen followed by
the addition of 8.0 V (V = mL solvent/g acylbenzoic acid) N,N-
dimethylformamide. The solution was agitated at 20 °C until
the acylbenzoic acid was fully dissolved. Solid 1,1′-carbon-
yldiimidazole (1.10 equiv) was charged in 4 approximately
equal portions over 20 min (CAUTION: evolution of carbon
dioxide, ensure reactor is adequately vented to avoid pressure
buildup). The solution was aged for approximately 0.5 h when
an analytical sample was quenched into amyl amine and
analyzed by LCMS to show acylbenzoic acid consumption.
A 35 wt % aqueous solution of hydrazine (1.10 equiv) was
added by pump over approximately 1.5 h while maintaining the
batch temperature was maintained between 20−25 °C
(CAUTION: hydrazine is acutely toxic and has a danger for
F
DOI: 10.1021/acs.oprd.5b00135
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Communication
127.4, 125.2, 109.5, 103.0, 98.1, 50.0, 29.1, 19.7, 13.8; HRMS-
TOF (m/z): [M + H]+ calcd for C22H28N3O2, 366.2176; found,
366.2185.
aged for 2 h at 20 °C, and HPLC analysis showed a supernatant
concentration of 8 mg/mL of 4-(4-methylthiophen-2-yl)-
phthalazin-1(2H)-one (target ≤10 mg/mL).
4-Methylphthalazin-1(2H)-one (9e). 9e was prepared
following the general procedure employing 2-acetylbenzoic
acid (25.0 g, 152 mmol) as the acylbenzoic acid. 4-
Methylphthalazin-1(2H)-one (9e) (12.3 g, 50% yield) was
isolated as a colorless crystalline solid: GC-MS <10 ppm
hydrazine; mp (DSC) 226.17 °C; 1H NMR (400 MHz,
DMSO-d6, δ): 12.39 (s, 1H), 8.25 (d, J = 7.7 Hz, 1H), 7.94−
7.81 (m, 3H), 2.50 (s, 3H); 13C NMR (100 MHz, DMSO-d6,
δ): 159.4, 143.1, 133.3, 131.3, 129.7, 127.3, 125.7, 125.4, 18.3;
HRMS-TOF (m/z): [M + H]+ calcd for C9H9N2O, 161.0709;
found, 161.0713.
Phthalazin-1(2H)-one (9f). 9f was prepared following the
general procedure employing 2-formylbenzoic acid (25.0 g, 167
mmol) as the acylbenzoic acid. The final solution was
partitioned with saturated aqueous ammonium chloride (200
mL, 8.0 V) and extracted with 3 × 200 mL (8.0 V) ethyl
acetate. The combined organics were dried over 30 g of
anhydrous magnesium sulfate and filtered. The combined
organics were concentrated at 75 mmHg/40 °C to 150 g of
solution whereupon 100 g heptane was charged. After 5 min,
the thin slurry was again concentrated at 75 mmHg/40 °C to
150 g of slurry. After 6 h at 20 °C, the slurry was filtered by
vacuum using a medium porosity sintered glass filter. The cake
was washed with heptane (100 mL, 4.0 V), and the filter cake
was dried in a 40 °C vacuum oven with a nitrogen bleed for 12
h.
Phthalazin-1(2H)-one (9f) (3.49 g, 14% yield) was isolated
as a colorless crystalline solid along with 10.41 g (43% yield) of
phthalazin-1(2H)-one (9f) lost to the mother liquor giving a
combined chemical yield of 13.9 g, 57% yield. GC-MS <10 ppm
hydrazine; mp (DSC) 180.30 °C; 1H NMR (400 MHz,
DMSO-d6, δ): 12.63 (s, 1H), 8.35 (d, J = 0.4 Hz, 1H), 8.22
(ddd, J = 7.8, 1.0, 0.9 Hz, 1H), 7.92−7.90 (m, 2H), 7.83 (dt, J =
7.92, 4.3 Hz, 1H); 13C NMR (100 MHz, DMSO-d6, δ): 159.5,
138.1, 133.5, 131.6, 129.9, 127.5, 126.6, 125.3; HRMS-TOF
(m/z): [M + H]+ calcd for C8H7N2O, 147.0553; found,
147.0553.
Pilot Plant Campaign: 4-(4-Methylthiophen-2-yl)-
phthalazin-1(2H)-one (5). 2-(4-Methylthiophene-2-carbonyl)-
benzoic acid (4.35 kg, 17.66 mol) was charged to a 100 L
Hastelloy reactor followed by N,N-dimethylformamide (25.67
kg, 27.02 L). A homogeneous solution resulted in less than 25
min after initiating agitation. Solid 1,1′-carbonyldiimidazole
(3.18 kg, 19.61 mol) was charged through the man-way in four
approximately equal portions over 1.25 h followed by charging
a N,N-dimethylformamide (7.18 kg, 7.56 L) rinse. CAUTION:
carbon dioxide evolution! Ensure adequate venting is available and
a rupture disk is used. The batch was held for 1 h whereupon 2-
(4-methylthiophene-2-carbonyl)benzoic acid was not detected
by HPLC analysis. The batch was transferred to a 160L glass-
lined steel reactor through a 10 μm inline filter followed by
rinsing of the line with N,N-dimethylformamide (2.0 kg).
A 35 wt % aqueous solution of hydrazine (1.78 kg, 1.76 L,
19.44 mmol) was transferred over 1.5 h to the batch through a
charge tank while maintaining the temperature at 16−18 °C
using jacket cooling (CAUTION: hydrazine is acutely toxic and
has a danger for cutaneous absorption. Consult the current MSDS
prior to use and ensure appropriate personal protective equipment).
After a 4 h hold at 20 °C, water (10.88 kg, 10.88 L) was added
to the slurry through a charge tank over 1 h. The slurry was
The slurry was transferred to a 80 L Hastelloy reactor,
equipped with an agitator and a filter dryer fitted with a 25 μm
polypropylene filter cloth. The cake was washed with a
premade mixture of N,N-dimethylformamide (7.18 kg, 7.56
L) and water (1.74 kg, 1.74 L), followed by two water washes
(4.35 kg, 4.35 L). The filter cake was dried under vacuum with
a nitrogen sweep at ambient temperature for 13 h whereupon
TGA analysis indicated LOD = 0.2% (target ≤1.0%).
4-(4-Methylthiophen-2-yl)phthalazin-1(2H)-one (5) was
isolated as a colorless crystalline solid (3.47 kg, 81.5% yield).
GC-MS analysis indicated the final isolated solid contained
<110 ppm hydrazine. Characterization data were identical to
that previously reported.14
N,O-Acetal (10). 2-(4-Methylthiophene-2-carbonyl)benzoic
acid (4) (500 mg, 2.03 mol) was charged to a 4 dram vial with a
magnetic stir bar. Acetonitrile (1.0 mL, 2.0 V) was added
resulting in a thick paste. Solid 1,1′-carbonyldiimidazole (325
mg, 2.03 mol) was charged in one portion immediately
resulting in an orange solution. CAUTION: carbon dioxide
evolution! After 1 h, an additional solid 1,1′-carbonyldiimidazole
(100 mg, 0.617 mol) was charged, and the vial was sonicated
for 5 min resulting in primary nucleation. The slurry was cooled
to 0 °C for 2 h and filtered by vacuum. The cake was washed
with 5 mL of dry acetonitrile and dried at 40 °C in a vacuum
oven with a nitrogen sweep.
N,O-Acetal (10) (343 mg, 57%) was isolated as a white
1
crystalline solid: mp(DSC) 128.00 °C; H NMR (400 MHz,
CDCl3, δ): 7.99 (d, J = 8.4 Hz, 1H), 7.83 (t, J = 7.5 Hz, 1H),
7.73−7.69 (m, 2H), 7.53 (s, 1H), 7.08 (m, 2H), 7.00 (s, 1H),
6.89 (t, J = 1.5 Hz, 1H), 2.23 (s, 3H); 13C NMR (100 MHz,
CDCl3, δ): 166.5, 147.1, 138.5, 138.2, 136.5, 135.3, 131.5,
129.9, 129.8, 126.5, 125.0, 123.5, 123.5, 118.3, 92.0, 15.6;
HRMS-TOF (m/z): [M + Na]+ calcd for C16H12N2O2SNa,
319.0517; found, 319.0533.
Conversion of N,O-Acetal (10) to Phthalazin-1(2H)-one
(5). N,O-Acetal (10): (200 mg, 0.682 mmol) was charged to a 1
dram vial containing a magnetic stir bar. N,N-Dimethylforma-
mide (0.80 mL, 4.0 V) was added resulting in a colorless
solution followed by addition of 35 wt % aqueous hydrazine
(0.067 mL, 0.75 mmol) (CAUTION: hydrazine is acutely toxic
and has a danger for cutaneous absorption. Consult the current
MSDS prior to use and ensure appropriate personal protective
equipment). Within 5 min, primary nucleation was observed.
After an 8 h hold, water (1.4 mL, 7.0 V) was added to the
colorless slurry, and the slurry was aged for 12 h. It was then
filtered by vacuum using a medium porosity sintered glass filter
and washed with water (5 mL). The filter cake was dried in a 40
°C vacuum oven with a nitrogen bleed for 12 h.
4-(4-Methylthiophen-2-yl)phthalazin-1(2H)-one was iso-
lated as a colorless crystalline solid (150 mg, 91% yield).
Characterization data were identical to that previously
reported.14
ASSOCIATED CONTENT
■
S
* Supporting Information
1H and 13C NMR spectra for compounds 9, 9b−9f and 10, 2D-
HSQC NMR spectra of compounds 6 and 10. The Supporting
Information is available free of charge on the ACS Publications
website at DOI: 10.1021/acs.oprd.5b00135.
G
DOI: 10.1021/acs.oprd.5b00135
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Communication
Stanforth, S. P. Tetrahedron 2000, 56, 5523−5534. (c) Raposo, M. M.
M.; Sampaio, A. M. B. A.; Kirsch, G. J. Heterocycl. Chem. 2005, 42,
AUTHOR INFORMATION
Corresponding Author
*E-mail: smennen@amgen.com.
Present Addresses
M.L.M-J.: Analytical Development, Alkermes, 852 Winter
Street, Waltham, Massachusetts 02451.
M.M.B.: Process Development, Amgen, Inc., One Amgen
Center Drive, Thousand Oaks, California 91320, United States.
■
1245−1251. (d) Vina, D.; del Olmo, E.; Lopez-Per
́
ez, J. L.; San
̃
Feliciano, A. Tetrahedron 2009, 65, 1574−1580.
(11) (a) Payton, M.; Bush, T. L.; Chung, G.; Ziegler, B.; Eden, P.;
McElroy, P.; Ross, S.; Cee, V. J.; Deak, H. L.; Hodous, B. L.; Nguyen,
H. N.; Olivieri, P. R.; Romero, K.; Schenkel, L. B.; Bak, A.; Stanton,
M.; Dussault, I.; Patel, V. F.; Geuns-Meyer, S.; Radinsky, R.; Kendall,
R. L. Cancer Res. 2010, 70, 9846−9854. (b) Cee, V. J.; Deak, H. L.;
Du, B.; Geuns-Meyer, S. D.; Hodous, B. L.; Nguyen, H. N.; Olivieri, P.
R.; Patel, V. F.; Romero, K.; Schenkel, L. Chem. Abstr. 2007, 147,
235186. Preparation of Substituted Phthalazinamines as Aurora Kinase
Modulators. PCT Int. Appl. WO/2007/087276, 2007.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
(12) (a) Robinson, D. I. Org. Process Res. Dev. 2010, 14, 946−959.
and references therein. (b) Snodin, D. J. Org. Process Res. Dev. 2011,
15, 1243−1246 and references therein.
(13) (a) Fiori, J. M.; Meyerhoff, R. D. Reg. Toxicol. Pharmacol. 2002,
35, 209−216. (b) FDA Draft Guidance, December 2008: “Guidance
for Industry Genotoxic and Carcinogenic Impurities in Drug
Substances and Products: Recommended Approaches”.
■
We thank Eric A. Bercot, Judy R. Ostovic and Tiffany L. Thiel
for helpful discussions in the preparation of this manuscript.
Funding Sources: Amgen Inc.
ABBREVIATIONS
■
CDI = 1,1-carbonyldiimidazole; DMF = N,N-dimethylforma-
mide; PAT = Process Analytical Technology; PCC = Power
Compensation Calorimetry; PLM = polarized light microscopy;
TTC = threshold of toxicological concern; V = mL solvent/g
solute
(14) Asselin, S. M.; Bio, M. M.; Langille, N. F.; Ngai, K. Y. Org.
Process Res. Dev. 2010, 14, 1427−1431.
(15) Kolkmann, R.; Leistner, E. Tetrahedron Lett. 1985, 26, 1703−
1704.
(16) Zogg, A.; Stoessel, F.; Fischer, U.; Hungerbuhler, K.
̈
Thermochim. Acta 2004, 419, 1−17.
(17) Sloan, K. B.; Koch, S. A. M. J. Org. Chem. 1983, 48, 635−640.
(18) Derivatization: 10 mL of reaction mixture was digested in 100
mL of amylamine for 10 min at 20 °C followed by dilution with 1 mL
1:1 acetonitrile/water.
(19) Sun, M.; Bai, L.; Liu, D. Q. J. Pharm. Biomed. Anal. 2009, 49,
529−533.
REFERENCES
■
(1) Yamada, T.; Nobuhara, Y.; Yamaguchi, A.; Ohki, M. J. Med.
Chem. 1982, 25, 975−982.
(2) (a) Yamaguchi, M.; Kamei, K.; Koga, T.; Akima, M.; Kuroki, T.;
Ohi, N. J. Med. Chem. 1993, 36, 4052−4060. (b) Van der Mey, M.;
Hatzelmann, A.; Van der Laan, I. J.; Sterk, G. J.; Thibaut, U.;
Timmerman, H. J. Med. Chem. 2001, 44, 2511−2522.
(3) Hemdan, M. M.; Taha, S. M.; Gabr, A. M.; Elkady, M. Y. J. Chem.
Res. 2010, 34, 102−105.
(4) Kaila, N.; Huang, A.; Moretto, A.; Follows, B.; Janz, K.; Lowe, M.;
Thomason, J.; Mansour, T. S.; Hubeau, C.; Page, K.; Morgan, P.; Fish,
S.; Xu, X.; Williams, C.; Saiah, E. J. Med. Chem. 2012, 55, 5088−5109.
(5) Demirayak, S.; Karaburun, A. C.; Beis, R. Eur. J. Med. Chem. 2004,
39, 1089−1095.
(6) Lim, C. J.; Kim, S. H.; Lee, B. H.; Oh, K.-S.; Yi, K. Y. Bioorg. Med.
Chem. Lett. 2012, 22, 427−430.
(7) (a) Bernard, A. M.; Cocco, M. T.; Congiu, C.; Onnis, V.; Piras, P.
P. Synthesis 1998, 317−324. (b) Ishikawa, M.; Eguchi, Y. Heterocycles
1981, 16, 25−30. (c) Pal, M.; Batchu, V. R.; Parasuraman, K.;
Yeleswarapu, K. R. J. Org. Chem. 2003, 68, 6806−6809.
(8) For related basic conditions see: (a) Knott, E. B. J. Chem. Soc.
1963, 402. (b) Yamada, T.; Nobuhara, Y.; Yamaguchi, A.; Ohki, M. J.
Med. Chem. 1982, 25, 975−982. (c) Scheffler, G.; Engel, J.; Kutscher,
B.; Sheldrick; William, S.; Bell, P. Arch. Pharm. 1988, 321, 205−208.
(d) Yamaguchi, M.; Kamei, K.; Koga, T.; Akima, M.; Kuroki, T.; Ohi,
N. J. Med. Chem. 1993, 36, 4052−4060. (e) Yamaguchi, M.; Kamei, K.;
Koga, T.; Akima, M.; Maruyama, A.; Kuroki, T.; Ohi, N. J. Med. Chem.
1993, 36, 4061−4068. (f) Epsztajn, J.; Malinowski, Z.; Brzezinki, J. Z.;
Karzatka, M. Synthesis 2001, 2085−2090. (g) Botero Cid, H. M.;
Trankle, C.; Baumann, K.; Pick, R.; Mies-Klomfass, E.; Kostenis, E.;
̈
Mohr, K.; Holzgrabe, U. J. Med. Chem. 2000, 43, 2155−2164. (h) Van
der Mey, M.; Hatzelmann, A.; Van der Laan, I. J.; Sterk, G. J.; Thibaut,
U.; Timmerman, H. J. Med. Chem. 2001, 44, 2511−2522. (i) Hemdan,
M. M.; Taha, S. M.; Gabr, A. M.; Elkady, M. Y. J. Chem. Res. 2010, 34,
102−105.
(9) For related acidic conditions see: (a) Desai, A. C.; Desai, C. M. J.
Indian Chem. Soc. 1980, 57, 757−758. (b) Shubin, K. M.; Kuznetsov,
V. A.; Galishev, V. A. Tetrahedron Lett. 2004, 45, 1407−1408.
(c) Kuznetsov, V. A.; Shubin, K. M.; Schipalkin, A. A.; Petrov, M. L.
Tetrahedron 2006, 62, 10018−10030.
(10) For cyclization from ketoesters see: (a) Bethune, R. J.; Gabbutt,
C. D.; Guinot, S. G. R.; Hepworth, J. D.; Heron, B. M. J. Chem. Soc,
Perkin Trans. 1 1994, 14, 1925−1934. (b) Morgan, D. O.; Ollis, W. D.;
H
DOI: 10.1021/acs.oprd.5b00135
Org. Process Res. Dev. XXXX, XXX, XXX−XXX