Preparation of cisplatin using microwave heating and continuous-flow processing as tools

Article abstract of DOI:10.1016/j.inoche.2011.01.005

Microwave heating has been used for the small-scale preparation of cisplatin, [cis-PtCl₂(NH₃)₂], in isomerically pure form without concomitant formation of Magnus' salt, [Pt(NH ₃)₄][PtCl₄].

Full text of DOI:10.1016/j.inoche.2011.01.005

Inorganic Chemistry Communications 14 (2011) 481–483
Contents lists available at ScienceDirect
Inorganic Chemistry Communications
journal homepage: www.elsevier.com/locate/inoche
Preparation of cisplatin using microwave heating and continuous-flow processing
as tools
Elizabeth A. Pedrick, Nicholas E. Leadbeater ⁎
Department of Chemistry, University of Connecticut, 55 N. Eagleville Road, Storrs, CT 06269-3060 USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Microwave heating has been used for the small-scale preparation of cisplatin, [cis-PtCl (NH ) ], in
2
3 2
Received 9 November 2010
Accepted 6 January 2011
Available online 14 January 2011
isomerically pure form without concomitant formation of Magnus’ salt, [Pt(NH
reaction to the gram level, continuous-flow processing was employed.
3 4 4
) ][PtCl ]. In scaling up the
©
2011 Elsevier B.V. All rights reserved.
Keywords:
Cisplatin
Cancer
Microwave heating
Flow chemistry
Platinum chemistry
Water
Cisplatin, [cis-PtCl
cancer drugs in the world due to its efficacy and surprisingly broad
applicability [1,2]. The complex was first reported by Peyrone in 1844
3] but it was not until the 1960s when its cytostatic properties were
discovered [4]. Many of the initial preparative routes to 1 were low-
yielding, time-consuming and somewhat unreliable. A significant
2
(NH
3
)
2
] 1, is one of the most widely used anti-
A current area of interest in our group is the use of microwave
heating as a tool for synthetic chemistry. The technique has seen
significant application in coordination chemistry as well as organo-
metallics [10]. In the area of organometallic chemistry, we have
found it to be very useful for the preparation of tri-ruthenium
[
3 2 4
complexes as well as Zeise's salt, KPtCl (C H ) [11,12]. Given the
issue was the formation of Magnus’ green salt, [Pt(NH
3
)
4
][PtCl
4
] 2, as a
importance of 1 and the issues associated with its synthesis, we
wanted to determine if microwave heating could be used for its
preparation in a fast, easy and selective way. Our results are pre-
sented here, together with a continuous-flow approach for scale-up
of the reaction.
by-product [5]. The fact that 2 is essentially insoluble in water and
organic solvents can make it challenging to remove from a product
mixture containing 1 [6]. An additional challenge in the synthesis of 1
is the requirement for just the cis-isomer. In an attempt to overcome
both these issues, Dhara has reported a route to 1 that proceeds via
the intermediacy of an iodo platinum complex [7]. The starting
We initiated our work by performing the synthesis on a small scale
in a 10 mL sealed glass tube using a scientific microwave unit using
material, K
2
[PtCl
4
], is converted into the tetraiodo analogue, K
2
[PtI
4
],
],
the Kukushkin procedure as our starting point. Using 20 mg of K
[PtCl ] and 4.3 eq of ammonium acetate and using 1 mL water as the
2
by addition of KI. Addition of ammonia to this yields cis-[PtI
which is then converted to cis-[Pt(H
2
(NH
by addition of
3
)
2
4
2+
2
O)
(NH )
2 3 2
]
solvent, we were able to obtain 1 in yields of 70–74% after 15 min
heating at 100 °C (Table 1, entry 1) [13]. No 2 was observed. To
confirm that the product was isomerically pure we performed the
Kurnakov test [14]. When treated with thiourea, an aqueous solution
of 1 forms a deep yellow solution while the trans isomer produces a
white insoluble precipitate. In addition, Pt NMR showed one signal
(−2104.0 ppm) corresponding to 1. We probed the effects of tem-
perature and time on the outcome of the reaction. Working at 100 °C,
reducing the reaction time had a deleterious effect on product yield
(Table 1, entries 2 and 3). Performing the synthesis at 140 °C for
aqueous silver nitrate. Treatment of this with KCl, results in the
precipitation of 1. Building on this and another procedure [8], a one-
step process has been developed and reported by Kukushkin et al.,
again yielding the cis isomer selectively [9]. However, the formation of
small quantities of metallic platinum and 2 are also observed. Their
2 4
route involves the reaction of K [PtCl ] with ammonium acetate and
KCl in water as the solvent (Scheme 1). Refluxing the reaction mixture
for 2 h followed by a hot filtration and, if necessary, a recrystallization
yields pure 1.
1
5 min led to formation of black precipitate (Table 1, entry 4).
Decreasing reaction time while still operating at 120 °C also led to
precipitation of undesired material (Table 1, entry 5). Reducing the
temperature to 85 °C and running the reaction for 15 min resulted
in incomplete conversion (Table 1, entry 6). Thus, our optimal
⁎
Corresponding author. Tel.: +1 860 486 5076; fax: +1 860 486 2981.
E-mail address: nicholas.leadbeater@uconn.edu (N.E. Leadbeater).
1387-7003/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.inoche.2011.01.005

482
E.A. Pedrick, N.E. Leadbeater / Inorganic Chemistry Communications 14 (2011) 481–483
Cl
Cl
Cl
Cl
Cl
Cl
NH3
NH3
Δ
Pt
+ NH OAc + KCl
Pt
K2
4
1
Scheme 1. Synthesis of Cisplatin.
conditions were: 20 mg K
water, 100 °C, 15 min.
Our next objective was to scale-up the reaction. We performed a
tenfold linear scale of our optimized reaction conditions, increasing
the quantity of K [PtCl ] used from 20 mg to 0.2 g and the volume of
2 4
water from 1 mL to 10 mL. To do this we transitioned to a 35 mL glass
2
[PtCl
4
], 4.3 eq ammonium acetate, 1 mL
In order to address the issue of Magnus’ salt formation when
scaling up the reaction, our attention turned to performing the
reaction using continuous-flow processing. Flow chemistry is seeing
significant research interest within the organic chemistry community
both in academia and industry [16]. It is possible to perform reactions
effectively both at high and low temperatures by means of heating or
cooling blocks around which the reactants pass in tubing of various
diameters (meso- and micro-flows). The technology has particular
benefit when performing reactions that involve use of potentially
hazardous, or unstable reactants. Only a small quantity of material is
directly in the reaction zone at any one time. It occurred to us that a
flow approach could be used for the preparation of 1 without the
concomitant formation of Magnus’ salt. We envisaged conditions
being similar to those for our small scale trials but now in continuous
flow. With literature precedent for translation of synthetic organic
protocols from microwave to flow chemistry [17], we decided to take
our optimized conditions for the preparation of 1 onto a continuous-
flow apparatus. We had access to a unit comprising of a dual channel
flow system, each being driven by a variable high-pressure pump.
After exiting a “T”-mixer the reaction mixture could be passed into
either a coil or column reactor. The coil reactor comprised either a coil
of stainless steel or PTFE which fitted snugly around an aluminium
heating block. Different lengths of coil were available, the choice being
dictated by the flow rate used and the required reaction time, a fast
reaction requiring a shorter coil or faster flow rate while a slow
reaction may need a slower flow rate through a longer coil.
tube as our reaction vessel [15]. A 20–35% yield of 1 was obtained
(
Table 1, entry 7). This result was somewhat surprising and made us
re-assess our product isolation strategy. When working on the small
scale, product precipitated out of solution upon cooling, filtration of
which gave good recoveries. On the larger scale we found that while a
quantity of 1 precipitated out of solution on cooling, the solution was
still yellow in color. Evaporation of some of the water, followed by
cooling in an ice bath, led to a second crop of product. Using this two-
crop strategy, the yield of recovered product could be substantially
increased to around 70% (Table 1, entry 8). Again, the product was
isomerically pure and no Magnus’ salt was formed. To determine if the
quantity of 1 obtained in the first crop could be increased by
performing the reaction more concentrated, we repeated the
synthesis on the 0.2 g scale of K
used just 2.5 mL water and 5 mL on the other. We obtained 66 and
2 4
[PtCl ] twice. On one occasion we
5
2% yields respectively in the first crops (Table 1, entries 9 and 10).
We wanted to transition to the gram scale. Working in a 35 mL glass
vessel, we performed the reaction using 1 g of K [PtCl ] and 10 mL
2 4
water. After 10 min at 100 °C, significant Magnus’ salt formation was
observed, as evidenced by the green colour of the product mixture
(
Table 1, entry 11). We thought this could be mitigated by increasing the
Although flow chemistry has a number of advantages, a significant
issue arises when performing reactions in which a precipitate is
formed. Since transformations are often performed at temperatures
above the boiling point of the solvent, the product mixture exits the
flow unit via a back-pressure regulator and into a collection vessel.
There is the potential for clogging either the back-pressure regulator
or the exit tube. In the case of 1, solubility in water is limited, even at
elevated temperatures. We therefore were concerned that precipita-
tion of 1 could be a problem. Our initial trial confirmed this. We
volume of water used in the reaction. However, when using 15 mL
water, although we obtained a reasonable quantity of 1, Magnus salt
formation meant that product isolation was challenging (Table 1,
entry 12).
Table 1
2 3 2
Probing reaction conditions for the synthesis of [cis-PtCl (NH ) ].
2 4
performed the reaction using 2 g of K [PtCl ] and 25 mL water and
flowed at a rate of 1 mL/min through a coil of 14 mL capacity in the
heated zone, this equating to a time at temperature of approximately
Entry
Reaction scale and conditionsa
Yield (%)
1
2
3
4
5
6
7
8
9
20 mg K
20 mg K
20 mg K
20 mg K
20 mg K
20 mg K
2
[PtCl
2
[PtCl
2
[PtCl
2
[PtCl
2
[PtCl
2
[PtCl
4
], 1 mL H
4
], 1 mL H
4
], 1 mL H
4
], 1 mL H
4
], 1 mL H
4
], 1 mL H
2
0, 100 °C, 15 min, MW
2
0, 100 °C, 5 min, MW
2
0, 100 °C, 10 min, MW
2
0, 140 °C, 15 min, MW
2
0, 140 °C, 5 min, MW
2
0, 85 °C, 15 min, MW
70–74b
28
1
4 min. We were able to process approximately half the mixture
34
c
before a clog occurred in the back-pressure regulator (Table 1, entry
13). However, the trial did show promise; no Magnus’ salt formation
was observed. In its original configuration, the back-pressure
regulator was not located in the heated zone. It occurred to us that
clogging could be somewhat ameliorated by moving the regulator to
inside the heated area. We made this modification and also slightly
reduced the flow rate to 0.8 mL/min (corresponding to around 17 min
in the heated zone) to account for the fact that the reaction using
microwave heating was ramped to 100 °C over the period of 1–2 min
and then held at that temperature for 15 min. Unfortunately clogging
still occurred, this time in the exit line between the back-pressure
regulator and the product collection vessel (Table 1, entry 14). The
solution to this problem that we envisaged was to perform the
reaction using the same methodology but at half the concentration
(i.e. using 50 mL water). We decided to trial this using microwave
0
c
0
d
16
b,e
0.2 g K
0.2 g K
0.2 g K
0.2 g K
2
[PtCl
2
[PtCl
2
[PtCl
2
[PtCl
4
4
4
4
], 10 mL H
], 10 mL H
], 2.5 mL H
2
0, 100 °C, 15 min, MW
0, 100 °C, 15 min, MW
20–25
~70
f
2
e
2
0, 100 °C, 15 min, MW
66
e
1
1
1
1
1
1
0
1
2
3
4
5
], 5 mL H 0, 100 °C, 15 min, MW
2
52
g
1 g K
1 g K
2 g K
2 g K
2 g K
2
2
2
2
2
[PtCl
[PtCl
[PtCl
[PtCl
[PtCl
4
4
4
4
4
], 10 mL H
], 15 mL H
], 25 mL H
], 25 mL H
], 50 mL H
2
2
2
2
2
0, 100 °C, 15 min, MW
0, 100 °C, 15 min, MW
0, 100 °C, 1 mL/min, F
0, 100 °C, 0.8 mL/min, F
0, 100 °C, 0.8 mL/min, F
_
g
38
40
_
72
h
h,i
i, j
a
b
c
d
e
f
MW = microwave heating; F = flow processing.
Average of a number of trials.
Black precipitate formed.
Unreacted starting material remaining.
Collecting just the initial crop of product.
Adopting a two crop strategy.
Significant quantities of Magnus’ salt observed.
Process stopped due to clogging.
2 4
heating first. We ran the reaction on the 0.2 g scale of K [PtCl ] using
g
h
i
5 mL water. After heating at 100 °C for 15 min, we cooled the reaction
vessel and collected the first crop. Performing some additional trials,
we discovered that the highest yield of pure 1 could be obtained if,
Back-pressure regulator located inside the heated zone.
Reaction mixture exposed to two heating cycles.
j

E.A. Pedrick, N.E. Leadbeater / Inorganic Chemistry Communications 14 (2011) 481–483
483
[
10] For a review, see:. G.L. Powell, in: N.E. Leadbeater (Ed.), Microwave Heating as a
Tool for Sustainable Chemistry, Taylor and Francis, Boca Raton FL, 2010,
pp. 175–205.
after collecting the first crop, the filtrate was placed back into the
microwave tube and heated again to 100 °C for 15 min. Moving to the
flow apparatus and passing the material through the heated zone at a
rate of 0.8 mL/min, the entire reaction mixture could be processed
without any clogging issues and again without the formation of 2 [18].
We obtained a 45% yield of 1 after the first crop. Passing the filtrate
through the flow reactor a second time allowed us to obtain a further
[11] N.E. Leadbeater, K.A. Shoemaker, Organometallics 27 (2008) 1254.
[
[
12] N.E. Leadbeater, K.A. Shoemaker, Inorg. Chem. Commun. 12 (2009) 341.
13] Synthesis of cisplatin on the 20 mg scale of K PtCl using microwave heating: To a
2
4
dry 10 mL microwave tube equipped with a magnetic stirbar was added a solution
of K [PtCl ] (0.0482 mmol, 20 mg), NH CO CH (0.208 mmol, 16 mg), KCl (0.268
2
4
4
2
3
mmol, 20 mg) in water (1 mL). This tube was then sealed in a microwave
apparatus (CEM Discover). The reaction mixture, being stirred continuously, was
heated using a maximum microwave power of 150 W to the target temperature of
100 °C and then held at this temperature for 15 min. The product mixture was
then cooled to 60 °C, the tube was taken out of the microwave unit and allowed
cool to room temperature before completing precipitation by placing in an ice
2
6% giving an overall product yield of 72% for the entire process
(
Table 1, entry 15). We thereby had a set of conditions for the larger
scale, continuous-flow preparation of 1.
In summary, we show that microwave heating can be used for the
small-scale preparation of cisplatin in isomerically pure form without
concomitant formation of Magnus’ salt. In scaling up the reaction to
the gram level, continuous-flow processing was employed.
bath. The precipitate was filtered and then washed with ice water. Drying gave
195
cis-[PtCl
ppm.
2
(NH
3
2
) ], 1, as a yellow solid in 71% yield.
2
Pt NMR (H O): δ = -2104.0
[
[
14] J.D. Woollins, A. Woollins, B. Rosenberg, Polyhedron 2 (1983) 175.
15] Synthesis of cisplatin on the 0.2 g scale of K PtCl using microwave heating: To a
dry 35 mL microwave tube equipped with a magnetic stirbar was added a solution
of [PtCl (0.482 mmol, 0.2 g), NH CO CH (2.08 mmol, 160 mg), KCl
2.68 mmol, 0.2 g) in water (2.5 mL). This tube was then sealed in the microwave
unit. The reaction mixture, being stirred continuously, was heated using
2
4
Acknowledgements
K
2
4
]
4
2
3
(
a
We thank the University of Connecticut and the National Science
Foundation (CAREER award CHE-0847262) for funding. CEM Corp. is
thanked for microwave equipment support and Uniqsis Inc. for access
to a FlowSyn continuous-flow unit.
maximum microwave power of 150 W to the target temperature of 100 °C and
then held at this temperature for 15 min. The product mixture was then cooled to
60 °C, the tube was taken out of the microwave unit and allowed cool to room
temperature before completing precipitation by placing in an ice bath. The
precipitate was filtered and then washed with ice water. The mother liquor was
concentrated by gently boiling off the excess water. It was washed with 0.1 M HCl.
A second crop of 1 was obtained, the combined crops equating to an 82% yield.
16] For an overview see:. Chemical Reactions and Processes under Flow Conditions,
in: S.V. Luis, E. Garcia-Verdugo (Eds.), Royal Society of Chemistry, Cambridge UK,
Appendix A. Supplementary data
[
A description of the apparatus used is given as supplementary
material. Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.inoche.2011.01.005.
2
010.
[17] M. Damm, T.N. Glasnov, C.O. Kappe, Org. Proc. Res. Dev. 14 (2010) 215.
18] Synthesis of cisplatin on the 2 g scale of PtCl using continuous-flow
processing: To a 100 mL capacity glass jar equipped with a top that allows tube
access was added solution of [PtCl (4.8 mmol, 2.0 g), NH CO CH
(20.8 mmol, 1.6 g), KCl (26.8 mmol, 2.0 g) in water (50 mL). Using the Uniqsis
FlowSyn, 14 mL capacity PTFE coil reactor was put in place around the
[
K
2
4
a
K
2
4
]
4
2
3
References
a
[
[
1] For an introduction to the chemistry and activity of cisplatin, see:. R.A. Alderden,
M.D. Hall, T.W. Hambley, J. Chem. Educ. 83 (2006) 728.
2] For discussion of the anti-cancer activity of cisplatin, the reader is directed to one
of the many books and review articles on the area. See for example
aluminium heating block. At the exit of the coil reactor, a 100 psi rate back-
pressure regulator was attached and kept in close proximity to the heater block. A
small length of PTFE tubing was attached to the exit of the back-pressure
regulator, this being used to take product mixture from the heated zone to a
100 mL capacity collection vessel. The block was heated from room temperature
to 100 °C, passing water through the coil reactor, from bottom to top, at a rate of
0.8 ml/mL. The flow was then changed from solvent (water) to reaction mixture
by means of a switch on the control unit. The reaction mixture was then passed
through the coil reactor at a rate of 0.8 mL/min. As the mixture neared the end of
the coil reactor, an empty, clean collection vessel was put in place. After all the
reaction mixture has entered the coil reactor, the flow was changed back to
solvent and water flowed through the reactor at a rate of 0.8 mL/min to push the
remaining reaction mixture through and out into the collection vessel. As soon as
this was achieved, the flow was stopped so as to avoid dilution of the product
mixture with water. The product mixture was then placed in ice to facilitate
precipitation of cisplatin. This solution was then filtered giving a first crop of
cisplatin in approximately 45% yield (648 mg). The mother liquor was then re-run
through the flow reactor using the same protocol. Again the product solution was
cooled and the precipitated cisplatin collected by filtration. The mother liquor was
concentrated by gently boiling off most of the water. A small quantity of 0.1 M HCl
was added and a second crop of 1 was obtained, the combined crops equating to a
72% yield.
(a) B. Lipp (Ed.), Cisplatin: Chemistry and Biochemistry of a Leading Anticancer
Drug, Wiley-VCH, Weinheim, 1999;
(
(
(
b) A.S. Abu-Surrah, M. Kettunen, Curr. Med. Chem. 13 (2006) 1337;
c) T. Boulikas, M. Vougiouka, Oncol. Rep. 11 (2004) 559;
d) E. Wong, C.M. Giandomenico, Chem. Rev. 99 (1999) 2451.
[
[
[
3] M. Peyrone, Ann. 51 (1844) 1.
4] B. Rosenberg, L. Van Camp, T. Krigas, Nature 205 (1965) 698.
5] Magnus’ salt has interesting chemistry itself, being a quasi-one-dimensional
compound with semiconductor properties
(a) E.G. Kim, K. Schmidt, W.R. Caseri, T. Kreouzis, N. Stingelin-Stutzmann, J.L.
Bredas, Adv. Mater. 18 (2006) 2039;
(b) W.R. Caseri, Platinum Met. Rev. 48 (2004) 91.
[6] A procedure has been published for conversion of Magnus’ salt into cis- and
trans-platin:. A.K. Starkov, G.A. Kozhukhovskaya, Russ. J. Inorg. Chem. 53 (2008)
1426.
[
[
[
7] S.C. Dhara, Ind. J. Chem. 8 (1970) 193.
8] V.V. Lebedinsky, V.A. Golovnya, Izv. Sektora Platiny SSSR 20 (1946) 95.
9] V.Y. Kukushkin, A. Oskarsson, L.I. Elding, N. Farrell, Inorg. Synth. 32 (1998) 141.