Determination of levamisole hydrochloride with HgI(2-)4 by a turbidimetric method and flow-injection analysis


Calatayud, J.M.; Falco, C.

Talanta 33(8): 685-687

1986


This paper is concerned with the use of ion-association compounds in the analysis of pharmaceutical samples by FIA. The usual extraction into an organic phase is avoided by using turbidimetric detection. Determination of levamisole with HgI(2-)(4) has been developed as a practical example: the experimental variables were optimized by the modified simplex method. The calibration graph is linear over the levamisole concentration range 7-32 microg ml . The reproducibility (rsd) and injection sample rate are 0.9% and 80/hr, respectively.

Talanta,
Vol.
33,
No.
8,
pp.
685-687,
1986
0039-9140/86
$3.00
+
0.00
Printed
in
Great
Britain.
All
rights
reserved
Copyright
(()
1986
Pergamon
Journals
Ltd
SHORT
COMMUNICATIONS
DETERMINATION
OF
LEVAMISOLE
HYDROCHLORIDE
WITH
HgIi
-
BY
A
TURBIDIMETRIC
METHOD
AND
FLOW
-INJECTION
ANALYSIS
J.
MARTINEZ
CALATAYUD*
and
CAMPINS
FALCO
Departamento
de
Quimica
AnaRica,
Facultad
de
Quimica,
Universidad
de
Valencia,
Burjasot
(Valencia),
Spain
(Received
30
October
1985.
Revised
25
March
1986.
Accepted
11
April
1986)
Summary
—This
paper
is
concerned
with
the
use
of
ion
-association
compounds
in
the
analysis
of
pharmaceutical
samples
by
FIA.
The
usual
extraction
into
an
organic
phase
is
avoided
by
using
turbidimetric
detection.
Determination
of
levamisole
with
fIgg
-
has
been
developed
as
a
practical
example:
the
experimental
variables
were
optimized
by
the
modified
simplex
method.
The
calibration
graph
is
linear
over
the
levamisole
concentration
range
7-32
pg/ml.
The
reproducibility
(rsd)
and
injection
sample
rate
are
0.9%
and
80/hr,
respectively.
Few
papers
on
fl
ow
-injection
analysis
(FIA)
are
con-
cerned
with
turbidimetric
detection,
and
so
far
only
ammonia
and
sulphate
ions
have
been
determined
in
this
way.
Krug
et
al.'
proposed
using
Nessler's
reagent
for
turbidimetric
detection
of
ammonia
(in
the
0.5-6
µg/m1
range)
in
natural
water
samples.
A
protective
agent
was
required
in
order
to
prevent
gradual
depo-
sition
of
precipitate
on
the
walls
of
the
FIA
system.
Good
precision
and
accuracy
were
achieved
with
a
sampling
rate
of
120/hr.
Sulphate
has
been
turbidimetrically
deterrnined
2--7
with
barium,
at
sampling
-rates
from
60/hr,'
to
250/
hr.
5
The
widest
application
range
was
for
sulphate
concentrations
from
5.0
to
200
µg/ml.
Hemmings
and
Macdonald'
reported
a
procedure
with
2-aminopyr-
imidine
hydrochloride
as
reagent,
using
the
merging
-
zones
technique
to
save
sample
and
reagent;
95
µI
of
both
species
are
consumed
in
each
determination,
in
the
0-10
µg/m1
sulphate
range,
and
the
sampling
-rate
is
up
to
60/hr.
Liquid
-liquid
distribution
of
ion
-association
com-
pounds
and
their
photometric
detection
is
a
broadly
used
procedure
in
drug
analysis:
it
is
a
sensitive
and
precise
technique
for
the
determination
of
basic
drugs.'"
Since
the
initial
work
of
Karlberg
and
Thelander
l2
on
determination
of
caffeine
in
acetylsali-
cylic
acid
preparations,
with
sodium
laurylsulphate
interference
eliminated
by
addition
of
tetraphenyl-
ammonium
ion
(which
forms
an
ion
-pair
with
the
laurylsulphate
and
transfers
it
to
the
organic
phase,
where
it
does
not
contribute
to
the
analyte
absorb-
ance
at
275
nm),
some
applications
of
FIA
auto-
mated
solvent
extraction
and
ion
-association
com-
plex
formation
to
drug
evaluation
have
appeared
in
the
literature.
They
include
determination
of
codeine
•Author
for
correspondence.
with
picrate,''
procyclidine
hydrochloride,
also
with
picrate,"
and
anionic
15
t6
and
cationic"
surfactants
in
pharmaceutical
formulations.
Steroid
and
bile
acid
sulphates
have
been
determined
in
clinical
analysis
with
lucigenin."
This
paper
is
concerned
with
turbidimetric
detection
in
FIA
procedures
based
on
ion
-association
compounds,
and
deals
with
evaluation
of
levamisole
with
tetraiodomercurate(II)
as
precipitating
reagent.
The
procedure
was
optimized
by
the
modified
simplex
method
(MSM),
the
parameters
concerned
being
sample
volume,
reaction
coil
-length,
fl
ow
-rate,
pH
and
reagent
concentration.
Levamisole
hydrochloride
is
the
laevo-isomer
of
tetramisole
hydrochloride;
both
are
anthelmintic
drugs,
but
the
former
has
fewer
side
-effects.
Hol-
brook
and
Scales'
have
determined
tetramisole
in
animal
tissue
extract
polarographically,
and
Mourot
et
al."
have
evaluated
it
by
HPLC
in
the
routine
analysis
of
veterinary
anthelmintics,
but
both
meth-
ods
require
mg
quantities
of
the
drug.
Some
liquid
extraction-colorimetry
methods
have
been
proposed
for
evaluation
of
tetramisole
in
pg
amounts,
with
various
dyestuffs.
2
'
Sodium
nitroprusside
and
cobalt
thiocyanate
23
have
also
been
proposed
as
reagents.
No
pharmacopoeial
method
is
given
in
the
BP
or
USP
for
determination
of
the
two
drugs.
EXPERIMENTAL
Reagents
Aqueous
solutions
of
levamisole
hydrochloride.
The
solid
(donated
by
Quimica
Farmaceutica
Bayer)
was
found,
by
non
-aqueous
potentiometric
titration
with
perchloric
acid
in
acetic
acid
medium,
to
be
100.7
±
0.3%
pure
(5
replicates).
Channing's
solution.
Mercuric
iodide
(1.00
g)
and
potas-
sium
iodide
(0.80
g)
were
dissolved,
mixed
and
diluted
to
100
ml
with
demineralized
water.
Buffer
solutions.
Made
with
citric
acid
and
disodium
hydrogen
phosphate,
at
0.5M
ionic
strength
and
adjusted
to
the
desired
pH.
685
686
SHORT
COMMUNICATIONS
Carrier
stream.
A
mixture
of
10
ml
of
buffer
solution
and
V
ml
of
Channing's
solution
was
made
up
to
50
ml
with
demineralized
water.
Other
reagents
were
of
analytical
grade.
FIA
assembly
and
procedure
The
sample
solution
was
injected
into
the
carrier
stream
and
the
turbidity
measured
spectrophotometrically
with
a
Coleman
55
(Perkin-Elmer)
instrument
provided
with
an
18-µ1
fl
ow
-cell
(Hellma)
and
a
Unicam
45
A
-R
(Pye
Uni-
cam)
recorder.
The
Tecator
5020
apparatus,
sample
injector
and
pumps
were
used
and
the
reaction
-coil
was
a
0.5
-mm
bore
Teflon
tube.
RESULTS
AND
DISCUSSION
A
qualitative
study
of
formation
of
ion
-association
compounds
of
levamisole
was
done
by
mixing
1
ml
of
4.2
x
10
-4
M
levamisole
solution,
1
ml
of
1.20
x
10
-3
M
dyestuff
or
2.64
x
10
-5
M
inorganic
anion
and
1
ml
of
buffer.
No
precipitate
was
formed
with
Bromocresol
Purple
or
Green,
Thymol
Blue,
Bromo-
phenol
Blue,
Methylthymol
Blue,
Phenol
Red
and
Arsenazo
B,
but
Hai
-
,
Cdg
-
and
BiLi
-
gave
yellow,
white
and
orange
precipitates
respectively.
Spectra
from
400
to
750
nm
for
suspensions
obtained
from
5.3
x
10
-5
M
levamisole
plus
2.64
x
10
-3
M
HgIi
-
or
BiI4
solutions
were
recorded
in
order
to
choose
the
most
suitable
counter
-ion
for
levamisole
determination
(no
precipitate
was
ob-
tained
with
levamisole
and
CdIi
-
at
these
concen-
trations).
The
influence
of
levamisole
concentration
was
also
tested.
Figure
la
shows
the
spectra
measured
against
demineralized
water
immediately
after
production
of
the
turbidities.
The
absorption
peak
at
430
nm
for
the
levamisole-HgIi
-
system
is
due
to
the
ion
-associa-
tion
compound,
since
no
absorption
is
observed
with
the
clear
solution.
For
the
levamisole-Bilz
system
no
absorption
peak
was
observed,
so
absorbance
measure-
ments
were
made
at
700
nm,
where
the
reagent
does
not
absorb.
The
calibration
graph
is
steeper
for
the
HgIi
-
system
(Fig.
lb)
so
this
is
preferred
for
practical
use,
although
the
limit
of
detection
is
lower
for
the
system.
Spectra
over
the
range
400-500
nm
recorded
at
30
-sec
intervals
show
a
decrease
in
absorbance
with
time;
the
stability
is
not
improved
by
addition
of
protective
colloids,
such
as
sucrose,
glucose
and
starch,
and
this
prevents
use
of
this
reaction
in
a
batch
procedure.
The
influence
of
pH
was
tested
by
adding
hydro-
chloric
acid
or
sodium
hydroxide
solution
to
a
5.6
x
10'M
levamisole/2.11
x
10
-3
M
Hgg
-
mixture
and
recording
the
absorbance
at
430
nm;
the
best
pH
range
was
found
to
be
3.30-7.65.
Various
buffer
solutions
were
tested
and
the
citric
acid/phosphate
system
was
selected.
The
stoichiometry
of
the
product
was
established
by
conductometric
titration
of
55
ml
of
8.1
x
10
-4
M
HgIi
-
or
3.20
x
10
-4
M
levamisole
with
2.00
x
10
-2
M
(a)
0
400
500
600
700
nm
(b)
1.0
7,
.430
nm
0.0
/.
700nm
05
•////
5
10
Levomisole
(10
-5
M)
Fig.
1.
(a)
Spectra
of
7.0
x
10
-5
M;
levamisole
solutions:
1,
with
2.6
x
10
-3
M
HgIi
-
;
2,
blank
solution
for
1;
3,
with
2.64
x
10
-3
M
Bllz;
4,
blank
solution
for
3.
(b)
Calibration
curve:
0
2.64
x
10
-3
M
Hgli
-
,
pH
5.60;
2.64
x
10
-3
M
Bilz,
pH
2.00.
levamisole
or
2.20
x
10
-3
M
HgI4
-
respectively
(see
Fig.
2)
and
found
to
be
2:1
levamisole:Hgg
-
.
It
is
a
relatively
weak
compound,
because
precipitation
is
not
clearly
observed
until
there
is
a
20
-fold
molar
ratio
of
Hgg
-
to
levamisole
and
the
maximum
E
1.2
1
1
1
0
/
/*
A r
9/
.
09
5
2
7
X
r*
10
0.4
0.3
0.2
0.1
(r)
E
Titrant
(m1)
Fig.
2.
Conductometric
titrations:
1,
titrant
levamisole;
2,
titrant
HgIi-
SHORT
COMMUNICATIONS
687
Table
1.
Range
of
variables
examined
Reaction
coil
length
55-315
cm
Flow
-rate
1.27-3.33
ml/min
pH
5.00-7.00
HgI
concentration
2.2
x
10
-5
-2.20
x
10
-2
M
Sample
size
30-200
pl
absorbance
is
obtained
at
>
50
-fold
molar
concen-
tration
ratio.
MSM
of
optimization
The
range
of
FIA
variables
to
be
studied
was
pro-
vided
by
the
results
of
the
preliminary
spectrophoto-
metric
work.
The
peak
-height
was
the
parameter
to
be
optimized.
The
range
of
variables
is
shown
in
Table
1.
Once
a
stable
chart
-recorder
base
-line
had
been
obtained,
a
sample
was
injected,
the
reaction
took
place
and
the
resulting
peak
was
collected
and
the
absorbance
value
at
430
nm
read.
This
was
repeated
until
an
rsd
4
1%
was
obtained
for
the
peak
-height
(four
or
fi
ve
repetitions
usually
sufficed).
The
program
of
the
MSM
for
this
work
was
operated
with
six
vertices,
and
was
written
on
the
basis
of
references
24-26.
The
initial
simplex
was
chosen
according
to
Yarbro
and
Deming'"
with
a
side
-length
of
1
and
the
vertex
at
the
origin
of
the
co-ordinates.
The
region
of
the
variables
was
normalized
by
the
modification
proposed
by
Morgan
and
Deming!'
The
optimal
conditions
obtained
for
1.0
x
10
-4
M
levamisole
are:
pH
5.00,
fl
ow
-rate
3.14
ml/min,
reaction
-coil
length
77
cm,
sample
size
200
µ1
and
reagent
concentration
5.8
x
10
-3
M.
The
influence
of
ionic
strength
on
the
peak
-height
is
constant
over
the
range
0.07-0.12M
(Fig.
3.).
The
calibration
graph
was
linear
for
7-32
µg/ml
levamisole,
with
8.55
x
10
-3
M
Hai
-
as
carrier
(57
-
fold
molar
ratio
to
the
maximum
concentration
of
levamisole).
The
linear
range
is
wider
than
those
for
the
methods'
using
Solochrome
Dark
Blue
or
Black
T,
Bromothymol
Blue,
Bromophenol
Blue
and
Bromo-
cresol
Green
(which
lie
between
2.0-2.5
µg/ml
as
the
lower
limit
and
10-20
µg/ml
as
the
upper),
but
nar-
rower
than
that
obtained
with
Bromocresol
Purple
(5-50
p
g/m1).
6
E
°
4
2
01
0.2
1
-
(M)
Fig.
3.
Influence
of
ionic
strength:
levamisole
concentration
7.1
x
10
-5
M,
Hgg
-
concentration
4.03
x
10
-3
M
in
carrier
stream
(pH
5.00,
buffer
citric
acid/Na
2
HPO
4
).
The
reproducibility
of
the
analysis
was
tested
by
injecting
into
the
reagent
stream
40
samples
of
9.1
x
10
-5
M
levamisole;
the
rsd
was
0.9%
(similar
to
that
of
the
method
of
Sane
et
al.
3
).
The
sample
injection
rate
obtainable
was
80/hr.
No
cleaning
solution
was
used
for
the
manifold
tubing,
but
of
those
tested
6M
hydrochloric
acid
gave
good
results.
The
levamisole
content
of
Nemanthel
was
deter-
mined
(veterinary
drug,
declared
formulation:
10
g
of
levamisole,
0.1
g
of
clorferinamine
maleate,
and
ex-
cipient
to
give
100
g
total
weight
of
sample).
A
0.5-g
portion
of
previously
powdered
sample
was
dissolved
and
diluted
accurately
to
250
ml
with
demineralized
water,
and
further
diluted
ten
-fold
(5
ml
to
50
ml)
and
200µI
were
directly
injected
into
the
carrier;
the
result
obtained
was
11.8
±
0.2%
(average
value
of
three
replicates).
This
sample
was
checked
by
the
method
of
Sane
et
al.
3
with
Bromocresol
Purple,
the
average
of
3
replicates
being
12.0
±
0.2%.
REFERENCES
1.
F.
J.
Krug,
J.
Riiiieka
and
E.
H.
Hansen,
Analyst,
1979,
104,
47.
2.
H.
Bergamin
F:,
B.
F.
Reis
and
E.
A.
G.
Zagatto,
Anal.
Chim.
Acta,
1978,
97,
427.
3.
F.
J.
Krug,
H.
Bergamin
F:,
E.
A.
G.
Zagatto
and
S.
S.
Jorgensen,
Analyst,
1977,
102,
503.
4.
S.
Baban,
D.
Beetlestone,
D.
Betteridge
and
P.
Sweet,
Anal.
Chim.
Acta,
1980,
114,
319.
5.
W.
D.
Basson
and
J.
F.
van
Staden,
Water
Research,
1981,
15,
333.
6.
.1.
F.
van
Staden,
Z.
Anal.
Chem.,
1982,
310,
239.
7.
F.
J.
Krug,
E.
A.
G.
Zagatto,
B.
F.
Reis,
F.
D.
Bahia,
A.
0.
Jacintho
and
S.
S.
Jorgensen,
Anal.
Chim.
Acta,
1983,
145,
179.
8.
P.
Hemmings
and
A.
M.
G.
Macdonald,
Flow
Analysis
II,
Lund,
Sweden,
June
1982.
9.
H.
Mohammed
and
F.
F.
Cantwell,
Anal.
Chem.,
1979,
51,
1006.
10.
Idem,
ibid.,
1980,
52,
553.
11.
F.
F.
Cantwell
and
M.
Carmichael,
ibid.,
1982,
54,
697.
12.
B.
Karlberg
and
S.
Thelander,
Anal.
Chim.
Acta,
1978,
98,1.
13.
B.
Karlberg,
P.
A.
Johansson
and
S.
Thelander,
ibid.,
1979,
104,
21.
14.
L.
Fossey
and
F.
F.
Cantwell,
Anal.
Chem.,
1982,
54,
1693.
15.
J.
Kawase,
A.
Nakae
and
M.
Yamanaka,
ibid.,
1979,
51,
1640.
16.
K.
Kine,
Dojin,
1979,
14,
9.
17.
J.
Kawase,
Anal.
Chem.,
1980,
52,
2124.
18.
M.
Maeda
and
A.
Tsuji,
Analyst,
1985,
110,
665.
19.
A.
Holbrook
and
B.
Scales,
Anal.
Biochem.,
1967,
18,
46.
20.
D.
Mourot,
B.
Delepine,
J.
Boisseau
and
G.
Gayot,
J.
Pharm.
Sci.,
1979,
68,
796.
21.
R.
T.
Sane,
D.
S.
Sapre
and
V.
G.
Nayak,
Talanta,
1985,
32,
148.
22.
D.
M.
Shingbal
and
S.
V.
Joshi,
Indian
Drugs,
1984,
21,
396.
23.
R.
T.
Sane,
B.
R.
Shinde,
A.
K.
Parikh
and
S.
P.
Tikekar,
ibid.,
1984,
21,
257.
24.
L.
A.
Yarbro
and
S.
N.
Denting,
Anal.
Chim.
Acta,
1973,
73,
391.
25.
S.
L.
Morgan
and
S.
M.
Deming,
Anal.
Chem.,
1974,
46,
1170.
26.
J.
A.
Nelder
and
R.
Mead,
Computer
J.,
1965,
7,
308.