Application of PAT tools for the safe and reliable production of a dihydro-1 H -imidazole

Article abstract of DOI:10.1021/op2001172

The application of two Process Analytical Technology (PAT) tools was studied and implemented for the safe and reliable synthesis of an advanced intermediate (4S,5R-7) of a member of the dihydro-1H-imidazole (1) class of compounds. Real time data were generated using ReactIR to track the complete breakdown of phosgene precursors (2) to phosgene (3) and confirm the absence of these hazardous materials prior to batch transfer operations. In addition, the chiral resolution by crystallization of rac7 was monitored by a Lasentec FBRM probe-based system. Implementation of the latter helped to track the crystallization process to minimize the risk of cocrystallization of undesired isomer 4R,5S-7.

Full text of DOI:10.1021/op2001172

TECHNICAL NOTE
pubs.acs.org/OPRD
Application of PAT Tools for the Safe and Reliable Production of a
Dihydro-1H-imidazole
Ana C. Barrios Sosa,* Ryan Conway, R. Thomas Williamson, James P. Suchy, William Edwards, and
Thomas Cleary
Pharmaceutical Technical Development Actives (PTDA-FL), Roche Carolina Inc., 6173 East Old Marion Highway, Florence,
South Carolina 29506-9330, United States
S Supporting Information
b
ABSTRACT: The application of two Process Analytical Technology (PAT) tools was studied and implemented for the safe and
reliable synthesis of an advanced intermediate (4S,5R-7) of a member of the dihydro-1H-imidazole (1) class of compounds. Real
time data were generated using ReactIR to track the complete breakdown of phosgene precursors (2) to phosgene (3) and confirm
the absence of these hazardous materials prior to batch transfer operations. In addition, the chiral resolution by crystallization of rac
7 was monitored by a Lasentec FBRM probe-based system. Implementation of the latter helped to track the crystallization process to
minimize the risk of cocrystallization of undesired isomer 4R,5S-7.
1. INTRODUCTION
2. DISCUSSION
p53 is a tumor suppressor, which plays a key role in the
regulation of the cell cycle. It has been found that the over-
production of the oncoprotein MDM2 in many tumors leads to
the disablement of p53 activity. Therefore, the inhibition of
MDM2 has been proposed as a novel strategy for cancer therapy.1
Several classes of MDM2 small molecule inhibitors have been
reported, including dihydro-1H-imidazoles 1 (Scheme 1).2
A representative member of this class of compounds is
dihydro-1H-imidazole 8, which can be synthesized from dihy-
dro-1H-imidazole 4 as shown in Scheme 2. In the original pro-
cedure, bis-trichloromethyl carbonate (triphosgene) (2a) was
treated with a catalytic amount of 2,6-lutidine in dichloro-
methane at <0 °C to produce carbonyl dichloride (phosgene)
(3). To the resulting phosgene (3) solution was then added com-
pound 4, followed by excess N,N-diisopropylethylamine (DIPEA)
to form intermediate 5 in situ. To this solution was then added
1-(3-methanesulfonyl-propyl)-piperazine dihydrochloride 6, and
the mixture was stirred at 25 °C overnight to provide the desired
product as a mixture of isomers (rac 8). The mixture was then
quenched with water and carried through an extractive workup.
Resolution of the product isomers was achieved by crystallization
of the desired isomer as the (1R)-(À)-10-camphorsulfonic acid
salt 4S,5R-7 from methanol in an overall yield of 38À41% and
>98% purity by chiral HPLC. Treatment of this salt with base
provides dihyrdo-1H-imidazole 8.
Various analytical methods have been reported for the detec-
tion of phosgene (3). However, infrared spectroscopy has been
identified to be particularly suitable, due to this compound’s
strong absorption bands at 849 and 1827 cmÀ1. In addition,
advances in the application of FTIR have made it possible to
apply this technique as a real time monitoring tool.3,4 For these
reasons, we opted to use a Mettler-Toledo iC10 process mid-
infrared analyzer with a 2 m AgX fiber optic probe (diamond
window) for our laboratory work. Although 2,6-lutidine is a sui-
table catalyst to initiate the breakdown of phosgene precursors
(2), for an optimized synthesis of carbonyl chloride 5 we en-
visioned using HCl salt 4 as the catalyst for the reaction of 2 to
phosgene (3). The implementation of this approach would allow
the controlled generation of 3 in the presence of a scavenger (4),
which could be activated by the addition of DIPEA and lead to
the controlled formation of intermediate 5. The proposed
mechanism for the breakdown of phosgene precursors 2a and
2b by chloride ions is shown in Scheme 3.3b
We began our studies by identifying a fingerprint region for
triphosgene (2a), diphosgene (2b), phosgene (3), dihydro-1H-
imidazole 4, carbonyl chloride 5, and product 8, which could be
tracked by ReactIR. Once key wavelengths were selected, a pro-
file for the production of rac 8 from starting material 4 was gene-
rated. Figure 1 shows the changes tracked in the fingerprint
regions throughout the synthesis of rac 8 using triphosgene (2a)
as the phosgene precursor. This data showed that addition of
intermediate 4 to a solution of triphosgene (2a) in dichloro-
methane was successful in initiating the formation of phosgene
(3). Complete reaction conversion was observed in approxi-
mately 3 h. Attempts to track diphosgene (2b) during the
In order to accomplish the safe and reliable multikilogram
scale production of 8, two main areas needed to be addressed
in the synthesis of key intermediate 4S,5R-7. One of these is
the implementation of the appropriate tool to monitor
phosgene (3) levels during the synthesis of intermediate 5.
Also of importance was the development of a method to
monitor the resolution of isomers by crystallization to ensure
the isolation of product 4S,5R-7 in the desired chiral purity
and yield.
Special Issue: Safety of Chemical Processes 11
Received: April 26, 2011
Published: June 06, 2011
r
2011 American Chemical Society
1458
dx.doi.org/10.1021/op2001172 Org. Process Res. Dev. 2011, 15, 1458–1463
|

Organic Process Research & Development
TECHNICAL NOTE
Scheme 1. Dihydro-1H-imidazoles
Scheme 2. Synthesis of dihydro-1H-imidazole 8
Figure 1. (a) Trends recorded for the reaction of triphosgene (2a) to
phosgene (3), the reaction of phosgene (3) with intermediate 4, and the
reaction of carbonyl chloride 5 with piperazine 6 over time (h). (b)
Surface plot for the reactions.
phosgene provided 3 in quantitative conversion in less than
20 min. As expected, addition of DIPEA to the mixture led to the
controlled formation of carbonyl chloride 5.
From a safety point of view, the option of adding DIPEA as
phosgene (3) is produced, and the phosgene precursor (2) is still
present in the mixture, representing an attractive approach to keep
the phosgene (3) levels low throughout the synthesis of 5. How-
ever, free amines, such as 4, have been known to react directly with
2 to form carbamic acid tert-butyl esters.3 Figure 3 shows the IR
profile of a reaction, where DIPEA was added to a solution of
triphosgene (2a), phosgene (3) and intermediate 4. As it can be
seen in Figure 3, the presence of a new IR stretch in the reaction
profile indicated the formation of a new impurity. This impurity
was also observed in the synthesis of 5 under similar conditions
when diphosgene (2b) was used as the precursor. The isolation of
this compound proved to be difficult due to its instability.5d
However, structure 9 was proposed for this impurity based on IR
data and literature reports describing the reaction of substituted
amines with triphosgene (2a) and diphosgene (2b).5 Attempts to
convert 9 to product 7 were unsuccessful.6 Further processing of
reaction mixtures containing this byproduct (9) led to a reduced
overall yield in the synthesis of intermediate 7; therefore, com-
plete conversion of 2 to phosgene (3) was deemed necessary to
achieve the expected levels of product 7.
Scheme 3. Reaction of precursors 2a,b to phosgene (3)
breakdown of triphosgene (2a) to phosgene (3) were unsuccess-
ful, presumably due to the low concentration of this intermediate
and possible overlap of the fingerprint regions under these
conditions. The mixture containing intermediate 5 was held
for 2 h, and piperazine 6 was added to complete the synthesis of
rac 8.
Additional information on the phosgene formation process
was generated by using diphosgene (2b) as the precursor. As
shown in Figure 2, the conversion of diphosgene (2b) to
1459
dx.doi.org/10.1021/op2001172 |Org. Process Res. Dev. 2011, 15, 1458–1463

Organic Process Research & Development
TECHNICAL NOTE
Figure 2. Trends recorded over time (h) for the reaction of diphosgene
(2b) to phosgene (3), followed by the reaction of phosgene (3) with
intermediate 4 upon addition of DIPEA.
Figure 4. (a) Product mixture obtained from the reaction of DIPEA
with phosgene (3). (b) Product mixture obtained from the reaction of
N-ethylisopropylamine with phosgene (3).
Figure 3. Surface plot of the IR profile from a reaction of triphosgene
(2a) and phosgene (3) with intermediate 4 in the presence of DIPEA.
Figure 5. Setup for the addition of diphosgene (2b) to the reactor.
Diphosgene (2b) is transferred from a mobile unit (a) to a reactor (c).
The line is cleared using low pressure nitrogen (b). Diphosgene (2b)
and phosgene (3) are monitored in real time using IR (d). After
complete consumption of phosgene (3), the reaction mixture is
transferred to a second reactor (e).
In order to ensure full conversion of 4 to 5 and maximize the
yield for product 7, we opted to use a slight excess of phosgene
(3) in the process. Due to the hazardous nature of this reagent, it
became important to determine the fate of the excess phosgene
remaining in the system. An experiment was devised to deter-
mine the levels of phosgene (3) in the headspace. The data
collected showed that no significant amounts of this reagent were
lost from the solution.7 This information combined with the data
collected by IR during the synthesis of 5 indicated that any excess
phosgene (3) remaining in the system was being scavenged
under the reaction conditions. Additional experiments were then
run to study the reaction of phosgene (3) with DIPEA. For this
purpose, a solution of phosgene (3) in dichloromethane was
treated with DIPEA using similar reaction conditions to the ones
used in the synthesis of 5. The resulting product mixture was
analyzed by nuclear magnetic resonance (NMR) spectroscopy.
Previous studies have reported that in the reaction of tertiary
amines with phosgene (3) one of the alkyl groups in the amine
is transferred and released in the form of an alkyl chloride.3a
Suspecting that an isopropyl group would be more likely
to migrate away from the amine under the reaction conditions,
the reaction of N-ethylisopropylamine with phosgene (3) was
investigated in parallel to confirm the identity of the products
formed. Analysis of the data collected showed that both reactions
generated a similar product mixture, which was composed of
amidoyl chloride 10 and urea 11 (Figure 4). These results con-
firmed that excess DIPEA acts a suitable scavenger for excess
phosgene (3).3a,8
From the data collected up to this point it was concluded that
both triphosgene (2a) and diphosgene (2b) could be used as
phosgene precursors for the synthesis of 5. However, for the first
multikilogram production runs, we opted to use diphosgene
(2b). Diphosgene (2b) could be effectively contained and trans-
ferred in our plant setting by configuring the equipment as shown
in Figure 5.9
For large scale runs, real time monitoring of diphosgene (2b)
and phosgene (3) in the reactor (c) was done with a Mettler-
Toledo MonARC process mid-infrared analyzer. A 3 m AgX fiber
1460
dx.doi.org/10.1021/op2001172 |Org. Process Res. Dev. 2011, 15, 1458–1463

Organic Process Research & Development
TECHNICAL NOTE
Figure 6. Trends recorded over time (h) during a production run for
the reaction of triphosgene (2a) to phosgene (3), followed by the
reaction of phosgene (3) with intermediate 4.
Figure 8. Profile of the particle changes tracked (counts/sec, No Wt,
10À50) in the crystallization of product 4S,5R-7 from a methanol
solution over time (0.055 °C/min). (a) Mixture diluted by a factor of
1.13 before crystallization. (b) Mixture diluted by a factor of 1.23 before
crystallization. Values in % indicate the purity by chiral HPLC.
7 using diphosgene (2b), the continued use of this reagent was
hindered by its low commercial availability and the restrictions
associated with the transportation of a liquid phosgene precursor.
For this reason, the use of a DoverPac containment unit for solid
charges was investigated and ultimately implemented to transfer
solid triphosgene (2a) to the reactor (c) (Figure 5).11 Figure 6
shows the ReactIR profile acquired during a production run for
the reaction of 4 to 5 using triphosgene (2a).
Figure 7. Profile of the particle changes tracked (counts/sec, No Wt,
10À50) in the crystallization of product 4S,5R-7 from a methanol
solution over time (0.055 °C/min). Analysis of the solids by chiral
HPLC after 10 h showed a decrease in the product purity. Values in %
indicate the purity by chiral HPLC.
As an optimized process for the synthesis of 5 was being deve-
loped, the resolution of the mixture of isomers by crystallization
of salt 4S,5R-7 was also investigated. For this purpose we opted
for the use of a Lasentec FBRM model D600VL,12 which is a
probe-based system that allows for the real time tracking of the
change of particles in the process. In a standard crystallization
procedure, the reaction mixture containing the product isomers
in methanol is treated with (1R)-(À)-10-camphorsulfonic acid
and seeded at 50 °C.13 A linear ramp of 0.055 °C/min is applied,
and the slurry is filtered at 23 °C.14 Figure 7 shows the trend for
particle changes tracked during the crystallization process.
As shown in Figure 7, the rate of crystal growth was fastest
in the first 2 h of the cooling process. Once the temperature
reached 23 °C and the crystallization appeared to have reached a
state of equilibrium, a second nucleation stage was observed. A
sample of the slurry was collected after the second nucleation
point, and the solids were filtered and analyzed by chiral HPLC.
The results showed that this sudden crystal growth was indicative
of the crystallization of the undesired isomer 4R,5S-7 from the
optic probe with a diamond window was inserted into the reactor
via a dip tube fitted to the solids charging port. As it had been
observed in the laboratory runs, addition of DIPEA after con-
version of diphosgene (2b) to phosgene (3) initiated the reaction
of dihydro-1H-imidazole 4 with phosgene (3) to generate
intermediate 5. Once a solution of 5 was generated and phosgene
(3) was no longer detected by IR in the reactor (c), the reaction
mixture was transferred to a second reactor (e) containing
piperazine 6, where the reaction of 5 with 6 was completed. Com-
partmentalization of the reaction from compound 4 to inter-
mediate 5 in a single reactor prevented the accumulation of
residual amounts of reagents or solvents used in other operations
in the synthesis of 7. This approach minimized the risk of an
uncontrolled generation of phosgene (3) gas in subsequent
batches, which could be triggered by the incompatible exposure
of diphosgene (2b) with these solvents and reagents.10 Despite
the successful completion of the first multikilogram campaign of
1461
dx.doi.org/10.1021/op2001172 |Org. Process Res. Dev. 2011, 15, 1458–1463

Organic Process Research & Development
TECHNICAL NOTE
showing the cocrystallization of compound 4R,5S-7. Once
analysis of a sample of the material confirmed the presence of
isomer 4R,5S-7 in the solids of the slurry, the batch was heated to
dissolve the material and the solution was diluted by 10% with
methanol. The cooling ramp was repeated, and the desired
product 4S,5R-7 was isolated with the expected chiral purity.
These results confirmed that dilution of the system allows for
extended processing times to achieve the desired product quality.
Figure 9b shows a representative Lasentec profile for subsequent
batches, for which a 10% dilution was implemented in the crys-
tallization process.
3. CONCLUSIONS
The synthesis of dihydro-1H-imidazole 5 is a key step in the
production of dihydro-1H-imidazole 8. Formation of 5 requires
the use of a phosgene precursor (2) to generate phosgene (3)
in situ. Key to the safe and reliable formation of 5 was the
implementation of IR as a real time monitoring tool to ensure the
complete breakdown of the phosgene precursor (2) to phosgene
(3) and confirm the absence of these hazardous materials prior to
transfer operations. In addition, particle changes in the crystal-
lization of intermediate 4S,5R-7 were tracked using a Lasentec
FBRM probe-based system. Implementation of the latter helped
to understand the overall crystallization process and to identify
conditions which could minimize the risk of crystallization of
undesired isomer 4R,5S-7.
4. EXPERIMENTAL SECTION
4.1. General Methods. All solvents and reagents were ac-
quired from commercial sources and used without further puri-
fication unless otherwise stated.
4.1.1. Chiral HPLC Analysis. Chiral HPLC analysis was per-
formed on an Agilent 1200 or equivalent equipped with a UV de-
tector at 254 nm and a Chiralcel OD column (250 mm  4.6 mm,
10 μm). Injection volume: 10 μL. Flow rate: 1.0 mL/min. Iso-
cratic Gradient: 80:20 hexanes/ethanol. Run time: approximately
25 min.
Figure 9. Lasentec profiles for the crystallization of product 4S,5R-7
from methanol over time in a plant setting. (a) Plant profile showing the
cocrystallization 4R,5S-7 and subsequent recrystallization profile to isolate
4S,5R-7 with the desired chiral purity. (b) Representative crystallization
profile of subsequent batches, for which a 10% dilution was implemented.
4.1.2. General Laboratory Reaction Setup for the Synthesis of
4S,5R-7. In a typical procedure, a 1 L half jacketed cylindrical
vessel was equipped with a 4-neck PTFE lid, overhead stirrer,
thermocouple, and ReactIR probe. The reaction was carried out
under an inert atmosphere. Upon completion of the extractive
workup, the ReactIR probe was replaced with a Lasentec probe
and the vessel was equipped with a jacketed distillation head.
Cooling and heating were achieved using a Huber Unistat 340w
chiller.
4.2. Representative Procedure for the Synthesis of [(4S,5R)-
2-(4-tert-Butyl-2-ethoxyphenyl)-4,5-bis-(4-chlorophenyl)-4,5-
dimethyl-4,5-dihydroimidazol-1-yl]-[4-(3-methanesulfonyl-
propyl)-piperazin-1-yl]-methanone (1R)-(À)-10-Camphor-
sulfonic Acid Salt (4S,5R-7). Triphosgene (2a) (17 g, 0.057 mol)
is dissolved in 204 g of methylene chloride. The temperature
is adjusted to À10 °C. A solution of dihydro-1H-imidazole 4
(71.0 g, 0.133 mol) in 439 g of methylene chloride is charged to
the mixture while keeping the temperature below 0 °C. The
reaction is monitored by ReactIR. N,N-Diisopropylethylamine
(DIPEA) (127 g, 0.983 mol) is then added at 13 g/min while
maintaining the temperature below 0 °C. The mixture is held
for approximately 2 h. The reaction is monitored by ReactIR.
To the solution is added 1-(3-methanesulfonylpropyl)-piperazine
mixture. The concentration of the reaction mixture containing
the product isomers was then varied to study the effect of changes
in supersaturation on the rate of crystal growth. The profiles
shown in Figure 8 show the particle changes recorded at two
different dilution levels over time. As expected, dilution of the
process by a factor of 1.13 (Figure 8a) and 1.23 (Figure 8b) led to
a time expansion in the overall crystallization profile.
The data in Figure 8 suggested that unexpected delays could
result in the crystallization of the undesired isomer 4R,5S-7.
These data also confirmed that a 10 to 25% dilution of the system
would allow for extended holding times, while keeping the
reaction yields within the expected range. To monitor the particle
changes in the crystallization process in a plant setting we chose a
Lasentec FBRM model D600R with an Alloy C22 400 mm
probe. For the scale-up of the crystallization process, measures
were taken to ensure careful review of the Lasentec profile before
proceeding with subsequent operations. Review of the real time
data could help to prevent the isolation of 4S,5R-7 containing
high levels of 4R,5S-7 in the event a second nucleation point was
observed. As shown in Figure 9a, implementation of the original
crystallization process at multikilogram scale generated a profile
1462
dx.doi.org/10.1021/op2001172 |Org. Process Res. Dev. 2011, 15, 1458–1463

Organic Process Research & Development
TECHNICAL NOTE
dihydrochloride (6) (45.0 g, 0.161 mol), the temperature is
adjusted to 25 °C, and the mixture is agitated overnight. The
reaction mixture is washed twice with 284 g of water and
concentrated to an oil. The solution is diluted with 236 g of
methanol and concentrated two times to remove residual DIPEA
and residual methylene chloride. To the resulting oil is then added
400 g of methanol. The temperature is adjusted to 50 °C, and a
solution of (1R)-(À)-10-camphorsulfonic acid(34.2g, 0.147 mol)
in 95 g of methanol is added. The mixture is seeded with product
4S,5R-7 (0.24 g, 0.2 mmol), and a cooling ramp of 0.055 °C/min
is initiated with a target temperature of 23 °C. The slurry is filtered
and washed with 44 g of methanol. The solids are dried under
vacuum at 50 °C for 12 h to give product 4S,5R-7 as a white solid
(48.7 g, 38%) in >98% purity by chiral HPLC.
Technologies (PAT) In Organic Process R&D, Clearwater, FL, 2010.
(b) Sholl, P. Enhanced Development of Batch and Continuous Pro-
cesses Using Real-Time in Situ FTIR Analytics as a PAT Tool. Presented
at The 5th International Conference and Exhibition on Process Analy-
tical Technologies (PAT) In Organic Process R&D, Clearwater, FL,
2010.
(5) (a) Bermudez, J.; Dabbs, S.; Joiner, K. A.; King, F. D. J. Med.
Chem. 1990, 33, 1929.(b) Barsanti, P. A.; Xia, Y.; Wang, W.; Mendenhall,
K. G.; Lagniton, L. M.; Ramurthy, S.; Phillips, M. C.; Subramanian, S.;
Boyce, R.; Brammeier, N. M.; Constantine, R.; Duhl, D.; Walter, A. O.;
Abrams, T. J.; Renhowe, P. A. Substituted imidazole compounds as KSP
inhibitors and their preparation, pharmaceutical compositions and use in
the treatment of cancers. U.S. Pat. Appl. US 20070037853, 2007; CAN
146:251846. (c) Gao, H.; Belvedere, S.; Webb, Y.; Landry, D.; Deng, S.;
Cheng, Z.; Yan, J. Preparation of benzoxazepines, benzodiazepines, and
benzazepines derivatives as agents for treating disorders involving
modulation of ryanodine receptors. PCT Int. Appl. WO 2008144483,
2008; CAN 150:5777. (d) This byproduct was produced in 10 to 20 area %
based on HPLC data. Characterization by LCMS was hindered by the
instability of this compound. Attempts to react this byproduct with
piperazine 6 to produce rac 8 under the reaction conditions were
unsuccessful.
(6) Product 9 was formed in approximately 7% by HPLC. This
product decomposes during the workup of the reaction. Data obtained
from the analysis of the mixture after workup by LCMS suggest that
product 9 decomposes to give primarily the corresponding carbamic acid.
(7) Phosgene (3) can be trapped by sweeping the headspace with
nitrogen into a midget impinger containing a solution of aniline.3
Quantification of the urea product can help to determine the amounts
of phosgene (3) present in the headspace. Analysis of the aniline solu-
tion produced in our study showed that no significant amounts of phos-
gene were present in the headspace of the vessel during the synthesis of
intermediate 5.
(8) For information regarding analysis of carbamoyl chloride con-
formers by 1H NMR, refer to: Koyano, K.; Suzuki, H.; McArthur, C. R.
Bull. Chem. Soc. Jpn. 1977, 50, 1872.
(9) Diphosgene is a liquid with a bp of 128 °C and a density of
1.65 g/mL.
(10) Phosgene is a gas with a bp of 8.3 °C and a density of 1.43 g/mL
at 0 °C.
(11) DoverPac units are commercially available containment sys-
tems provided by ILC Dover.
(12) (a) Wheatcroft, H. Use of PAT in Monitoring and Under-
standing Crystallization and Polymorphic Transformations. Presented
at The 5th International Conference and Exhibition on Process Analy-
tical Technologies (PAT) In Organic Process R&D, Clearwater, FL,
2010. (b) Black, S. N.; Quigley, K.; Parker, A. Org. Process Res. Dev. 2006,
10, 241.(c) The unit was equipped with a D600L Probe (diameter: 19
mm, length: 400 mm) constructed from Alloy C-276 with Kalrez seals.
(13) The mixtures used for crystallization studies were mixtures
produced starting from compound 4.
(14) Different cooling ramps and filtration temperatures were
studied for the crystallization of 7. A linear cooling ramp of approxi-
mately 0.055 to 0.065 °C/min provided the highest product yield and
purity.
’ ASSOCIATED CONTENT
S
Supporting Information. Key wavelengths for com-
b
pounds 2, 3, 4, and 5. Solubility data for compounds 4S,5R-7
and 4R,5S-7 as well as LCMS data for decomposition products
arising from mixtures containing 9. This material is available free
of charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: ana.barrios_sosa@roche.com. Telephone: 843-629-
4000. Fax: (843)-629-4128.
’ ACKNOWLEDGMENT
We thank Geremia Jennings for her technical support with the
Lasentec unit and Jennifer Andrews from Mettler Toledo for her
technical support with the ReactIR units. We also thank Jason
Greer for his support with the DoverPac containment unit and
Ram Narayan for project support.
’ REFERENCES
(1) Xia, M.; Knezevic, D.; Tovar, C.; Huang, B.; Heimbrook, D. C.;
Vassilev, L. Cell Cycle 2008, 7, 1604.
(2) (a) Fry, D. C. Curr. Protein Pept. Sci. 2008, 9, 240. (b) Graves, B.;
Fotouhi, N. Curr. Top. Med. Chem. 2005, 5, 159. (c) Vassilev, L. T.; Vu,
B. T.; Graves, B.; Cavajal, D.; Podlaski, F.; Filipovic, Z.; Kong, N.;
Kammlott, U.; Lukacs, C.; Klein, C.; Fotouhi, N.; Liu, E. A. Science 2004,
303, 844.(g) Ding, Q.; Graves, B. J.; Kong, N.; Liu, J.; Lovey, A. J.;
Pizzolato, G.; Roberts, J. L.; So, S.; Vu, B. T.; Wovkulich, P. M. 2,4,5-
Triphenyl imidazoline derivatives as inhibitors of the interaction be-
tween p53 and MDM2 proteins for use as anticancer agents. PCT Int.
Appl. WO 2007063013, 2007; CAN 147:52890. (d) Markey, M. P.
Frontiers Biosci. 2011, 16, 1093. (e) Azmi, A. S.; Philip, P. A.; Almhanna,
K.; Beck, F. W.; Kafri, Z. K.; Sarkar, F. H.; Mohammad, R. M. Mini-Rev.
Med. Chem. 2010, 10, 518.(f) Vassilev, L.; Fry, D. In Small-Molecule
Inhibitors of Protein-Protein Interactions; Aktories, K., Compas, R. W.,
Cooper, M. D., Gleba, Y. Y., Honjo, T., Koprowski, H., Malissen, B.,
Melchers, F., Oldstone, M. B. A., Olsnes, S., Vogt, P. K., Eds.; Current
Topics in Microbiology and Immunology; Springer-Verlag: Berlin,
Germany, 2011; Vol. 348, pp 151À172.
(3) (a) Cotarca, L, Eckert, H. Phosgenations À A Handbook; WILEY-
VCH Verlag GmbH & Co. KGaA; Weinheim, Germany, 2003. (b)
Pasquato, L.; Modena, G.; Cotarca, L.; Delogu, P.; Mantovani, S. J. Org.
Chem. 2000, 65, 8224.
(4) (a) Williamson, R. T. Recent PAT Applications at Roche
Carolina, Inc.; Adventures with Phosgene and Beyond. Presented at
The 5th International Conference and Exhibition on Process Analytical
1463
dx.doi.org/10.1021/op2001172 |Org. Process Res. Dev. 2011, 15, 1458–1463