Characterization of a Stable 2,2′-Azobis(2-Methylpropanenitrile) Degradant and Its Use to Monitor the Oxidative Environment During Forced Degradation Studies by Liquid Chromatography/Mass Spectrometry

Article abstract of DOI:10.1016/j.xphs.2019.05.012

Azo compounds are commonly used to study radical-mediated degradation of pharmaceutical compounds. The favorable chemical and physical properties of 2,2′-azobis(2-methylpropanenitrile) (AIBN) have made it one of the most widely used compound for these type of studies. This article describes the characterization of a stable product, N-(1-cyano-1-methylethyl)-2-methylpropanamide, formed during the decomposition of AIBN. This product is easily detected by liquid chromatography/mass spectrometry and can serve as a marker to confirm the AIBN is working as intended and to monitor the kinetic formation of free radical species.

Full text of DOI:10.1016/j.xphs.2019.05.012

Journal of Pharmaceutical Sciences xxx (2019) 1-4
Contents lists available at ScienceDirect
Journal of Pharmaceutical Sciences
journal homepage: www.jpharmsci.org
Note
0
Characterization of a Stable 2,2 -Azobis(2-Methylpropanenitrile)
Degradant and Its Use to Monitor the Oxidative Environment During
Forced Degradation Studies by Liquid Chromatography/Mass
Spectrometry
*
Kevin J. Wells-Knecht , Derek Dunn
Teva Pharmaceutical, West Chester, Pennsylvania 19380
a r t i c l e i n f o
a b s t r a c t
Article history:
Azo compounds are commonly used to study radical-mediated degradation of pharmaceutical com-
pounds. The favorable chemical and physical properties of 2,2 -azobis(2-methylpropanenitrile) (AIBN)
Received 7 September 2018
Revised 2 May 2019
Accepted 14 May 2019
0
have made it one of the most widely used compound for these type of studies. This article describes the
characterization of a stable product, N-(1-cyano-1-methylethyl)-2-methylpropanamide, formed during
the decomposition of AIBN. This product is easily detected by liquid chromatography/mass spectrometry
and can serve as a marker to confirm the AIBN is working as intended and to monitor the kinetic for-
mation of free radical species.
Keywords:
degradation product
forced conditions
free radical
© 2019 Published by Elsevier Inc. on behalf of the American Pharmacists Association.
mass spectrometry
LC-MS
oxidation
stability
Introduction
on their solubility and stability properties. One of the most
commonly used is azobisisobutyronitrile (AIBN), which ultimately
4
The development of pharmaceutical compounds of interest
routinely involves forced degradation studies to characterize
products that may form during manufacture and storage. These
studies subject the active pharmaceutical ingredient (API) to
various pH, temperature, oxidation and light conditions to accel-
erate the kinetics of degradant formation and concentrations
forms carbon centered 2-cyanopropyl radicals that react with dis-
solved oxygen to form the desired 2-cyano-2-propyl peroxy radical
intermediates for oxidation of test articles (Fig. 1).
These types of studies rarely incorporate a positive control, so it
is rational to wonder if the AIBN is working as intended when no
API degradation is detected. In one of our studies utilizing AIBN, a
compound was detected by liquid chromatography/mass spec-
trometry (LC-MS) that increased with time but was unrelated to the
API. This work describes the characterization of this product that
could be used as a marker to confirm an API has indeed been
exposed to an oxidative free radical environment.
1,2
suitable for detection and characterization.
Free radicalemediated oxidation of pharmaceutical compounds
typically employs azo-compounds that thermally degrade to per-
oxy radical species.3 While the number of azo-compounds to
choose from is quite large, only a handful are routinely used based
Experimental Section
0
Abbreviations used: AIBN, 2,2 -azobis(2-methylpropanenitrile); CMMP, N-(1-
General Experimental Information
cyano-1-methylethyl)-2-methylpropanamide; CID, collision induced dissociation;
DKI, dimethyl-N-(2-cyano-2-propyl)ketenimine; IBN, isobutyronitrile; LC/MS,
liquid chromatography/mass spectrometry; MAN, methacrylonitrile; TMSN, tetra-
methylsuccinodinitrile; ToF, time-of-flight; UPLC, ultra performance liquid
chromatography.
All buffers and solvents used in this study were of HPLC grade or
of the highest available purity and used without further purifica-
1
tion. AIBN was obtained from Sigma Aldrich (St. Louis, MO). H NMR
Conflicts of interest: The authors declare no competing financial interest.
13
and C NMR were recorded on a Bruker 400 MHz using d
as a solvent.
6
-acetone
*
Correspondence to: Kevin J. Wells-Knecht (Telephone: þ1-610-883-5849).
E-mail address: kevin.wells-knecht@tevapharm.com (K.J. Wells-Knecht).
https://doi.org/10.1016/j.xphs.2019.05.012
022-3549/© 2019 Published by Elsevier Inc. on behalf of the American Pharmacists Association.
0

2
K.J. Wells-Knecht, D. Dunn / Journal of Pharmaceutical Sciences xxx (2019) 1-4
Synthesis of N-(1-cyano-1-methylethyl)-2-methylpropanamide
CMMP) [84,213-57-0]
(
The route used to synthesize CMMP followed a procedure and
5
used a similar compound. All chemicals used were obtained from
Sigma Aldrich. Briefly, 340
mL (3.57 mmol) of 2-amino-2-propane
nitrile [19,355-69-2], 374
mL (3.57 mmol) of isobutryl chloride
[
79-30-1] and 498 mg (3.57 mmol) of potassium carbonate were
combined in 5 mL of ethyl acetate and stirred for 72 h at room
temperature, The reaction was quenched by adding 300 L of water
m
followed by several washes with saturated brine to remove
unreacted compounds, salts, and water from the ethyl acetate layer.
The ethyl acetate was evaporated to dryness to yield 463 mg of
1
CMMP (84% yield). H NMR (400 MHz, Acetone-d
2
6
)
d
7.45 (s, 1H),
Figure 1. Kinetics for appearance of the AIBN degradation product at pH 3, 6, and 9.
Peak areas are derived from selected ion chromatograms (m/z 155.1 þ 177.1).
1
3
.51-2.41 (sep, 1H), 1.66 (s, 6H), 1.10-1.08 (d, 6H); C NMR (400
6
MHz, Acetone-d ) d 177.1, 122.3, 46.9, 35.4, 27.1, 19.6. HRMS (ion
electrospray) m/z calc for C
177.0979.
8
H
14
N
2
ONa [M þ Na]þ 177.1004, found:
AIBN Incubations
The final concentration of AIBN for all incubations was 2.5 mM
in acetonitrile/methanol (40/10, v/v) containing 50 mM sodium
phosphate buffer (pH 3 or 6) or 30 mM sodium pyrophosphate (pH
LC/MS Conditions
Samples were injected (5
formance liquid chromatography system using a Waters HSS T3
m). The mobile phases were 0.1% formic
mL) onto a Waters Acquity ultra per-
9
). The buffer concentration at pH 9 was lower due to the reduced
solubility of pyrophosphate buffer with organic solvents. Methanol
was present to quench undesired 2-cyano-2-propoxy radical for-
column (1.0 ꢁ 100 mm,1.7
m
acid inwater (A) and 0.1% formic acid in acetonitrile (B). The gradient
was 0% B ramped to 7.5% B at 6.0 min, then 20% B at 9.0 min, then 70%
B at 10 min. The column was held at 70% B for 1.0 min, then returned
to 0% B in 1.0 min and held for 2.0 min to re-equilibrate the ultra-
3
mation. To initiate the formation of oxidative 2-cyano-2-peroxy
radicals, the vial was immediately placed in the LC/MS autosam-
ꢀ
pler compartment kept at 40 C. Repeated injections were made at
approximately 45 min intervals to evaluate the kinetic formation of
AIBN degradant.
performance liquid chromatography column. The flow rate was
ꢀ
0.15 mL/min, and the column temperature was maintained at 30 C.
A divert valve was used to direct flow to waste during the first
0.85 min to minimize contamination of the mass spectrometer
source. The kinetic studies were carried out with an autosampler
Forced Degradation of Solid AIBN
ꢀ
Small amounts of AIBN (<10 mg) were placed in 1.5 mL plastic
compartment kept at 40 C to promote the degradation of AIBN.
screw cap microcentrifuge tubes and then incubated for 6 days and
1
The mass spectrometer was a Waters Synapt G2 HDMS Q-ToF
operated in positive ion electrospray mode scanning from 85 to 600
Da using leucine enkephalin (m/z ¼ 556.2771) as a lock mass
standard. The parameters used were capillary voltage ¼ 3.25 kV,
ꢀ ꢀ
6 days at 45 C (below the 50 C self-accelerating decomposition
temperature listed for AIBN). The degraded AIBN samples were
ꢀ
placed at 5 C until used for analysis.
Figure 2. Thermal degradation of AIBN to radical products and various secondary reaction products. MAN, methacrylonitrile; TMSN, tetramethylsuccinodinitrile; IBN,
isobutyronitrile.

K.J. Wells-Knecht, D. Dunn / Journal of Pharmaceutical Sciences xxx (2019) 1-4
3
Figure 3. Selected ion chromatograms (m/z 155.1 þ 177.1) of (a) AIBN incubation at pH
after 18 h and (b) authentic CMMP standard.
6
ꢀ
sampling cone ¼ 25 V, desolvation temperature ¼ 475 C, des-
Figure 4. Kinetics for appearance of CMMP from fresh AIBN and degraded AIBN (solid
ꢀ
olvation gas flow ¼ 800 L/h, source temperature ¼ 120 C, and cone
ꢀ
ꢀ
AIBN incubated at 45 C for 6 days and 16 days) when incubated at 40 C in solution at
ꢀ
gas flow ¼ 20 L/h. Product ion spectra were obtained with a colli-
sion energy ¼ 21 V.
pH 6. Fresh AIBN was also incubated at 10 C in solution at pH 6 as a control. (a) CMMP
peak areas derived from selected ion chromatograms (m/z 155.1 þ 177.1). (b) CMMP
peak areas relative to the initial peak area for each sample.
Results and Discussion
peroxy radical concentrations are not high enough to completely
ameliorate DKI so it is able to form CMMP. McCarthy and Hegarty
reported the hydration of DKI was acid catalyzed; however, the
similar kinetics observed in Figure 1 may be explained by the
The unknown compound formed in the AIBN incubations eluted
þ
at 7.8 min with a low intensity [M þ H] ion of 155.1 and a much
þ
larger (~50-fold) relative intensity [M þ Na] ion of 177.1. Inter-
þ
þ
estingly, the [M þ Na] /[M þ H] signal ratio decreased to ~5 by
changing the mobile phase additive to 0.1% difluoroacetic acid
instead of formic acid (results not shown). High resolution mass
spectrometry results suggested an elemental composition of
11
presence of phosphate buffers which are also catalytic. This work
shows that CMMP is easily detected with LC/MS conditions
commonly used for analysis of pharmaceutical APIs exposed to
forced degradation conditions with AIBN. For APIs that exhibit
excellent stability, the detection of CMMP could serve as a positive
control to confirm the AIBN is behaving as intended, and the API
was indeed exposed to a suitable free radical oxidative environ-
ment. Indeed, Figure 4a illustrates that when AIBN is compromised,
there is a significant amount of CMMP present at t ¼ 0, and the
slope is decreased which suggests less radical formation from AIBN.
When the CMMP peak area is plotted relative to the area at t ¼ 0
C
8
H
14
N
2
O for the unionized molecule. The CID-MS/MS spectrum of
þ
the [M þ H] ion produced 2 major fragments, m/z 128 (neutral loss
of HCN) and m/z 86 (neutral loss of isobutyronitrile). Incubations
performed at pH 3, 6, and 9 displayed similar kinetics for appear-
ance of this compound out to 5 hours (Fig. 1).
The initial solvent caged radicals formed after release of nitro-
gen from AIBN are known to produce several products via dispro-
6
portionation and recombination reactions outlined in Figure 2. In
(Fig. 4b), there was an approximately 15-fold increase for the fresh
an inert, dry solvent devoid of oxygen, ~55% of AIBN forms
dimethyl-N-(2-cyano-2-propyl)ketenimine (DKI). When oxygen is
present, no DKI is detected owing to the formation of peroxy rad-
AIBN compared with only a 1.5- to 2-fold increase for the degraded
AIBN. No increase in CMMP was observed when fresh AIBN was
incubated at 10 C which confirms CMMP is derived from AIBN
radical products. The detection of CMMP is also amenable to ex-
periments performed over a wide range pH range. While the pre-
sent work was specific for AIBN, it may also extend to alternative
azonitrile compounds used for forced degradation studies and their
respective amide products.
ꢀ
7,8
icals that degrade the DKI. However, the presence of water allows
an additional reaction pathway for DKI that forms CMMP. Early
work with DKI did not detect CMMP with the introduction of water
because liquid-liquid extractions were performed with dilute hy-
9
10
drochloric acid that hydrolyzed the nitrile group. Rodríguez et al.
later detected CMMP in a problematic batch of solid AIBN that was
underperforming as a catalyst for radical-mediated synthetic re-
actions owing to water contamination.
References
The identity of the AIBN degradation product noted in this
article was confirmed to be CMMP after synthesis of the authentic
compound which had a similar retention time (Fig. 3), accurate
mass and MS/MS spectrum. Identical retention times were main-
tained when formic acid was substituted with difluoroacetic acid as
a mobile phase modifier. To our knowledge, this is the first time
that CMMP has been identified in a system that is typical for AIBN-
mediated oxidation of pharmaceutical compounds. This suggests
1. Baertschi SW, Alsante KM, Santafiano D. Stress testing: the chemistry of drug
degradation. In: Baertschi SW, Alsante KM, Reed RA, eds. Pharmaceutical Stress
Testing: Predicting Drug Degradation. 2nd ed. New York: Informa Healthcare;
2011:49-141.
2. Blessy M, Patel RD, Prajapati PN, Agrawal YK. Development of forced degra-
dation and stability indicating studies of drugs - a review. J Pharm Anal.
2014;4(3):159-165.
3
. Harmon P, Boccardi G. Oxidative susceptibility testing. In: Baertschi SW,
Alsante KM, Reed RA, eds. Pharmaceutical Stress Testing: Predicting Drug
Degradation. 2nd ed. New York: Informa Healthcare; 2011:168-191.

4
K.J. Wells-Knecht, D. Dunn / Journal of Pharmaceutical Sciences xxx (2019) 1-4
4
. Alsante KM, Martin L, Baertschi SW. A stress testing benchmarking study.
Pharm Technol. 2003;27(2):60-72.
8. Hammond GS, Wu CHS, Trapp OD, Warkentin J, Keys RT. The mechanism of
decomposition of azo compounds. II. Cage effects in the decomposition of a,a-
5
6
7
. Coussanes G, Gaus K, O'Sullivan AC. The synthesis of ketone-derived enamides
by elimination of HCN from cyanoamides. Eur J Org Chem. 2016;2016:4176-
azoisobutyronitrile and related compounds. J Am Chem Soc. 1960;82(20):5394-
5399.
^
0
4188.
9. TalAt-Erben M, Bywater S. The thermal decomposition of 2,2 -azo-bis-iso-
butyronitrile. Part I. Products of the reaction. J Am Chem Soc. 1955;77(14):3710-
3711.
. Jaffe AB, Skinner KJ, McBride JM. Solvent steric effects II. The free-radical
chemistry of azobisisobutyronitrile and azobis-3-cyano-3-pentane in viscous
and crystalline media. J Am Chem Soc. 1972;94(24):8510-8515.
. Hammond GS, Trapp OD, Keys RT, Neff DL. Isolation and study of the inter-
mediate [dimethyl-N-(2-cyano-2-propyl)-ketenimine] formed in the decom-
10. Rodríguez V, Barba V, Bern eꢀ s S, Quintero L, Sartillo-Piscil F. N-(2-Cyanopropan-
2-yl)isobutyramide. Acta Crystallogr E. 2015;E61:o2733-o2735.
11. McCarthy DG, Hegarty AF. Rapid acid-catalysed and uncatalysed hydration of
ketenimines. J Chem Soc Perkin Trans. 1980;2:579-591.
position of a,a'-azoisobutyronitrile. J Am Chem Soc. 1959;81(18):4878-4882.