A simple HPLC-UV method for the determination of dimenhydrinate and related substances - Identification of an unknown impurity

Article abstract of DOI:10.1691/ph.2007.3.6156

During the revision of the dimenhydrinate monograph of the European Pharmacopoeia a HPLC-UV method was developed. The procedure described allows a qualitative and quantitative determination of both dimenhydrinate compounds and of thirteen related substances. Furthermore a hitherto unknown impurity was identified and integrated into the purity check. Also 18 samples of dimenhydrinate have been tested. Thereby the relevant impurities of dimenhydrinate could be nominated and quantified.

Full text of DOI:10.1691/ph.2007.3.6156

ORIGINAL ARTICLES
Institute for Pharmacy, Pharmaceutical Chemistry, University of Leipzig, Germany
A simple HPLC-UV method for the determination of dimenhydrinate and
related substances – identification of an unknown impurity
U. D o¨ ge, K. Eger
Received August 2, 2006, accepted August 7, 2006
Ulrica D o¨ ge, University of Leipzig, Institute for Pharmacy, Pharmaceutical Chemistry, Br u¨ derstraße 34,
04103 Leipzig, Germany
Pharmazie 62: 174–178 (2007)
doi: 10.1691/ph.2007.3.6156
During the revision of the dimenhydrinate monograph of the European Pharmacopoeia a HPLC-UV
method was developed. The procedure described allows a qualitative and quantitative determination of
both dimenhydrinate compounds and of thirteen related substances. Furthermore a hitherto unknown
impurity was identified and integrated into the purity check. Also 18 samples of dimenhydrinate have
been tested. Thereby the relevant impurities of dimenhydrinate could be nominated and quantified.
1
. Introduction
To allow the determination of all relevant impurities within
the limits of the European Pharmacopoeia the monograph
ought to be revised. In the future the test of related sub-
stances should be carried out by liquid chromatography.
For determination of dimenhydrinate by HPLC some test
methods have been described in the past, which mostly
used reversed phases (C8 or C18) and acetonitrile and
phosphate buffer as solvent. Chromatography was per-
formed both isocratic (Kvist et al. 2000) and with gradient
Dimenhydrinate, once known as dramamine, belongs to the
small group of combined medicinal substances. The patent
in terms of synthesis was presented already in 1950 in the
USA (Cusic 1950). According to this the substance is made
up by mixing equimolar amounts of 8-chlorotheophylline
(IV) and diphenhydramine (VIII) in hot ethyl alcohol. The
resulting salt is a potent drug against nausea. Various produ-
cers supply the demand of dimenhydrinate. Due to the vari-
ety of raw material, technical equipment and manufacturing
technology a wide spectrum of impurities is inevitable.
Though dimenhydrinate is applied for more than 50 years,
literature research shows only few information about this
topic. Several authors describe degradation products of di-
menhydrinate or diphenhydramine found in stress tests, but
none has been identified (Barbas et al. 2000; Donelly 2002;
Stiles et al. 1994). The European Pharmacopoeia prescribes
a specific TLC test only for theophylline (II) (PhEur. 2005).
Henderson et al. name seven potential impurities of diphen-
hydramine remaining from synthesis: benzhydrol (XII),
benzophenone (XIII), diphenylmethane (XIV), 2-(diphenyl-
methoxy)-N-methylethanamine (VII), 2-[(4-methylphenyl)-
phenylmethoxy]-N,N-dimethylethanamine (IX), 2-[(4-bro-
mophenyl)phenyl-methoxy]-N,N-dimethylethanamine (X)
and N,N,N-[(2-diphenylmethoxyethyl)-(2-dimethylamino-
ethyl)methyl]amine (VI) (Henderson et al. 2001). Already
in 1971 8-[(2-diphenylmethoxy)ethylmethyl-amino]theo-
phylline (XI) –– a dimer of IV and VII –– was isolated from
dimenhydrinate (Santoro and Warren 1971). The German
Federal Institute for Drugs and Medical Devices discusses
additionally the dibenzhydryl ether (XV). Because of their
structural relationship also caffeine (III) and theobromine
(
(
Nassr et al. 2003), partly in addition of triethylamine
Barbas et al. 2000; Donelly 2002) acting as competitive
base. A similar procedure includes methanol, triethylamine
and acetic acid (Roos and Lau-Cam 1986).
It is noted that Donelly and Kvist et al. sign only one
peak for dimenhydrinate in HPLC chromatograms. Doubt-
less it is 8-chlorotheophylline. Iterating the published
methods diphenhydramine appears in the chromatograms
late with plane and wide peaks. Therefore it has probably
been overlooked. So diphenylalkylamines such as diphen-
hydramine are not quantifiable in this way.
Moreover the asymmetry (intense tailing) of the diphenhy-
dramine peaks is the most important handicap of all men-
tioned techniques. This comes along with elution times of
several minutes. As structural related substances cause the
same effect no sufficient peak resolution could be
achieved.
Instructions for determination of diphenhydramine –– with-
out 8-chlorotheophylline –– as the European Pharmacopoeia
test of related substances (PhEur. 2005) and the method
from Stiles et al. (Stiles et al. 1994) have the same limits.
Utilisation of ion pair formation can improve the peak
shape considerably. Employed are octanesulfonic acid
(Paciolla et al. 2001), heptanesulfonic acid (Qi et al. 2003)
and laurylsulfonic acid (Henderson et al. 2001). Two dis-
advantages restrain their applicability. First the reagents
must be of high purity to avoid ghost peaks, especially in
gradient HPLC. Secondly dilution series of diphenhydra-
mine ion pairs have only limited linearity.
(I) must be taken into account.
H C
O
3
+
N
CH3
H C
3
N
N
O
N
H
Cl
O
N
CH3
The intention of this study was to name relevant impuri-
ties of dimenhydrinate and to develop and validate a
Dimenhydrinate
174
Pharmazie 62 (2007) 3

ORIGINAL ARTICLES
HPLC method for qualitative and quantitative determina-
tion of the drug and related substances. For these purposes
8 samples of dimenhydrinate from several producers
were available.
A gradient in eluent composition and flow rate (Table 1)
caused the best possible resolution of fifteen compounds.
The high percentage of triethylamine and an acetonitrile
minimum of 40% was essential to obtain a sufficient peak
shape for diphenylalkylamines, like diphenhydramine. The
working temperature was 30 C. Always 10 ml were in-
jected. The UV detector was set at 225 nm. Separation of
15 substances occured within 30 min. Equilibration took
1
ꢀ
2. Investigations, results and discussions
2.1. Identification of an impurity in dimenhydrinate
1
5 min. Fig. 1 shows a spectrum of all regarded sub-
Within the scope of the development of the HPLC method
in chromatograms of two dimenhydrinate samples a peak
was detected, which could not be associated with any
known impurity. All substances related to diphenhydra-
mine and 8-chlorotheophylline should be involved into the
purity test. That is why identification was necessary. A
content of more than one percent in two samples allowed
the isolation by column chromatography. The structure
elucidation qualified the substance as 8-chlorocaffeine (V).
This special xanthine was not commercially available. By
methylation of 8-chlorotheophylline a reference substance
of high purity could be synthesized.
stances. Table 2 summarizes chromatogram data of dimen-
hydrinate and its impurities. Beside retention times/relative
retention times and peak resolution the UV-factor was de-
termined, because of the varying UV-activity at 225 nm.
The standard for comparison was diphenhydramine.
Table 1: Gradient for dimenhydrinate determination
Time (min)
Mobile phase
A (% V/V)
Mobile phase
B (% V/V)
Flow rate
(ml ꢁ min
ꢂ1
)
0
1
–15
82 ! 50
50 ! 20
20
18 ! 50
50 ! 80
80
1.2
5–20
1.2 ! 2.0
O
CH3
20–30
30–32
2.0
H C
20 ! 82
80 ! 18
2.0 ! 1.2
1.2
3
N
N
3
2–45
82
18
Cl
N
O
N
CH3
1
.50
.25
.00
.75
.50
.25
.00
8
-Chlorocaffeine
1
1
0
0
0
0
2
.2. HPLC of dimenhydrinate and related substances
The Scheme shows all substances analysed. As also VI,
IX, X, XI and XV have not been commercially available,
reference substances were synthesized. Extensive tests
considering the described analytical problems resulted in
the following HPLC parameters:
As matrix material a high-purity, metal free, pH-stable
pH 1.5–10) silica gel with complete endcapping was cho-
(
sen. Perfect spherical particles with smooth surface guar-
anteed symmetric peaks, short elution times and as a re-
sult good resolution of structural related substances. The
eluent consisted of a 1% triethylamine solution (eluent A;
pH 2.5 with phosphoric acid) and acetonitrile (eluent B).
0
5
10
15
Minutes
20
25
30
Fig. 1: Dimenhydrinate and related substances
O
O
O
O
O
CH3
CH3
CH3
H
N
H
H
C
H
C
H
C
H C
3
N
3
3
N
N
3
N
N
HN
N
N
N
N
Cl
Cl
N
N
N
N
O
N
O
N
O
N
O
N
O
N
CH3
CH3
CH3
CH3
CH3
I
II
III
IV
V
H
CH3
CH3
N
CH3
CH3
N
O
CH3
N
CH3
N
N
O
N
O
CH3
O
CH3
O
CH
3
CH3
VII
CH3
Br
VI
VIII
IX
X
O
N
H
N
H C
CH
3
3
OH
O
N
N
O
N
O
O
CH3
XIV
XII
XIII
XV
XI
I: theobromine; II: theophylline; III: caffeine; IV: 8-chlorotheophylline; V: 8-chlorocaffeine; VI: N,N,N-[(2-diphenylmethoxyethyl)-(2-dimethylamino-
ethyl)methyl]amine; VII: 2-(diphenylmethoxy)-N-methylethanamine; VIII: diphenhydramine; IX: 2-[(4-methylphenyl)-phenylmethoxy]-N,N-dimethylethan-
amine; X: 2-[(4-bromophenyl)phenyl-methoxy]-N,N-dimethylethanamine; XI: 8-[(2-Diphenylmethoxy)ethylmethylamino]theophylline; XII: benzhydrol;
XIII: benzophenone; XIV: diphenylmethane; XV: dibenzhydryl ether
Pharmazie 62 (2007) 3
175

ORIGINAL ARTICLES
Table 2: Chromatogram data of dimenhydrinate and related substances
Substance
I
II
III
IV
VI
V
VII
VIII
IX
X
XII
XI
XIII
XIV
XV
tret. (min)
Rel. tret.
Resolution
UV-factor
2.96 3.83 5.65 6.38 9.08 9.79 12.65 13.15 15.13 16.16 19.06 20.25 20.98 22.60 26.00
0.23 0.29 0.43 0.49 0.69 0.74
4.72 7.84 2.70 8.92 3.61 9.69
1.06 1.20 1.08 1.09 1.24 0.93
0.96
1.90
1.08
1.00
6.75
1.00
1.15
4.03 13.63
0.72 0.84
1.23
1.45
6.58
1.15
1.54
1.60
1.72
1.98
n.a.
1.19
5.19 10.83 16.46
0.94
2.10
1.87
2.3. Validation
VI
To test the suitability of the method a validation was carried
out. Table 3 shows the parameters, the requirements and the
results. A high value was set on the robustness. It could be
verified that variations in temperature, triethylamine percen-
tage and pH value had no influence on retention times of
the analysed substances, with exception of VI. In this case a
raising pH value resulted in increasing retention time
7
,6
,5
7
CH
3
N
CH3
7,4
7,3
O
N
CH
3
(
Fig. 2). The effect was caused by the two-step protonation
7
7
,2
,1
of the diamine structure. At pH 2 there was a hydrophilic
cation with low retention on the lipophilic matrix. Above
pH 2.8 (increasing deprotonation) the lipophilic character
became more important and the retention time enhanced.
This was essential to consider because above pH 3 resolu-
tion between VI und V was not possible any more.
Also the application of columns from various producers
demonstrated good robustness. There was no deviation in
retention times above 10%. Furthermore excellent repeat-
ability, accuracy, linearity in a wide range and good recov-
ery could be verified.
7
6
,9
1,8
2,3
2,8
pH-value
Fig. 2: pH-sensitivity of VI
importance. Amounts up to 1.3% were determined. IX, X,
XII und XV could be detected only in traces. I, III, XI,
XIII and XIV have not been found. Nevertheless III, XI
and XIII should be taken into account because of a higher
probability of appearance in dimenhydrinate samples. I and
XIV need not be included into succeeding purity tests.
2
.4. Sample tests
To evaluate the relevance of the suspected impurities, all
available samples of dimenhydrinate have been analysed.
Table 4 shows the results. II, V, VI and VII are of high
Table 3: Validation data
Test item
Test range
Requirement
Results
*
*
ꢂ1
Peak symmetry
pH robustness
0–1.2 mg ꢁ ml
symmetry factor 0.8–1.5
rel. retention time 0.95–1.05
rel. retention time 0.95–1.05
rel. retention time 0.95–1.05
content 99.0–101.0%
rel. standard deviation of retention time max. 10%
resolution at least 1.5
<1.45
pH 2.0–pH 2.8
0.5%–1.5%
0.98–1.02
0.98–1.02
0.98–1.01
100.1–100.8
<10%
>1.9
<0.25%
<1.5%
0.9997
0.9987–1.0
100.2–101.8%
<0.001 mg ꢁ ml
*
TEA robustness
Temperature robustness
*
ꢀ
ꢀ
25 C–35 C
134 h
**
Stability of solution
Column material
*
3 columns
*
Selectivity
Repeatability
15 substances
6 injections
6 weighted samples
0–1.3 mg/ml
0–0.1 mg/ml
6 samples
*
*
rel. standard deviation of peak areas max. 0.85%
rel. standard deviation of peak areas max. 2.0%
coefficient of correlation at least 0.990
coefficient of correlation at least 0.990
content 98–102%
**
Accuracy
**
Linearity
Linearity
Recovery
*
**
*
ꢂ1
ꢂ1
Limit of determination
15 substances
<0.001 mg ꢁ ml (10 ng)
*
dimenhydrinate and impurities, ** dimenhydrinate only
Table 4: Contents of impurities in 18 dimenhydrinate samples (%)
Impurity
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
II
V
0.44 1.50 0.17 0.12 0.06 0.05 0.02 0.04 0.06 0.07 0.05 0.02
1.37 0.01
0.02 0.05 0.20 0.24 0.13
0.86 1.00
VI
VII
0.02 0.04 0.01 0.01 0.01 0.01 0.01 0.02 0.01 0.01 0.02 0.01 0.01
0.02 0.02 0.09 0.13 0.26 0.14 0.17 0.18 0.15 0.16 0.23 0.13 0.18 0.13 0.13 0.18 0.04 0.04
0.06
IX
X
0.01 0.03
0.01
0.01
0.01
0.01
XII
XV
other
total
0.06 0.07
0.05 0.05
0.44 0.47 0.16 0.13
0.02
0.02 0.02 0.02
0.02 0.01
0.17 0.29 0.04
0.01
0.04
0.01
2.38 2.12 0.46 0.47 0.33 0.20 0.22 0.23 0.23 0.26 0.34 0.17 0.21 0.18 0.19 1.36 1.61 0.30
176
Pharmazie 62 (2007) 3

ORIGINAL ARTICLES
For purity checks it is recommended to accept at most
.2% of II and VII. It seems to be difficult to avoid the
formation of these by-products in syntheses. Also there is
no information about toxic effects. According to the Euro-
pean Pharmacopoeia every other impurity should be re-
stricted to 0.1%.
3.4.3. 2-[(4-Methylphenyl)phenylmethoxy]-N,N-dimethylethanamine ––
hydrochloride (IX)
0
Sodium hydride (2 g) was suspended in 200 ml of ice cooled absolute to-
luene. To the suspension, 4 g of methylbenzhydrol in 20 ml of toluene
ꢀ
were added dropwise. Then the mixture was heated up to 95 C. After
3
0 min 4 g of 1-chloro-2-dimethylaminoethane hydrochloride were added
in portions. Then the temperature in the reaction vessel was kept for four
ꢀ
additional hours at 95 C. Next the suspension was cooled down, and
5
0 ml of cold water were dropped in carefully to inactivate remained so-
dium hydride. Subsequently the organic layer was washed with water
50 ml) three times and then extracted with hydrochloric acid (1 M,
50 ml). The acid extract was alkalified with sodium hydroxide solution
1 M, about 200 ml) to pH 12–13. The white product precipitated and was
3. Experimental
(
1
(
3
.1. Materials
Acetonitrile (gradient grade for liquid chromatography) was obtained from
Merck (Germany), triethylamine, phosphoric acid and methylamine solu-
tion (33% in ethyl alcohol) from Fluka (Germany). Theobromine (I), theo-
phylline (II), caffeine (III) and 15 samples of dimenhydrinate were pro-
vided by the German Federal Institute for Drugs and Medical Devices.
Two dimenhydrinate samples were purchased from Synopharm. Also di-
menhydrinate-CRS was available. 8-Chlorotheophylline (IV) was bought
from Aldrich, diphenhydramine hydrochloride (VIII) likewise from Syno-
pharm, benzhydrol (XII) from Lancaster, benzophenone (XIII) from Jena-
pharm and diphenylmethane (XIV) also from Fluka. 2-Chloroethyldiphe-
nylmethyl ether was offered by Frinton Laboratories. Water was obtained
from a purification system (Millipore). Eluent A was filtered through a
extracted with diethyl ether (Wolf and Schunack 1996). The diethyl ether
was concentrated to a small volume. By gassing with hydrogen chloride
gas from heated hydrochloric acid the salt sedimented (M u¨ ller et al. 2001).
After recrystallization from ethyl acetate white crystals were obtained:
yield 10%; m.p. 148 C; H NMR (300 MHz, CDCl3) d (ppm): 1.81 (1 H,
s), 2.36 (3 H, s), 2.89–2.92 (6 H, m), 3.30–3.33 (2 H, m), 3.95–3.98 (2 H,
m), 5.44 (1 H, s), 7.16–7.39 (9 H, m).
ꢀ
1
3.4.4. 2-[(4-Bromophenyl)-phenylmethoxy]-N,N-dimethylethanamine
maleate (X)
Bromobenzene (8 ml) was heated with 10.5 g of benzoyl chloride and
0
.45 mm cellulose nitrate membrane.
ꢀ
1
1.3 g of aluminium chloride at 80–90 C. After 10 h the light brown sub-
stance hardened during cooling. The residue was dissolved in 100 ml acet-
one and filtered through a frit with 10 g of aluminium oxide. Recrystalliza-
tion in 40 ml petroleum ether yielded 6.9 g of bromobenzophenone.
Bromobenzophenone (4 g) was mixed with 5.6 g of NaBH4 in a mortar
and stored for five days, stirred once a day (TLC reaction control: CHCl3).
The powder was extracted with diethyl ether. Then the organic layer was
evaporated and the oily product solidified during 12 h in the fridge; 3 g of
bromobenzhydrol were obtained (Toda et al. 1989).
3
.2. Apparatus
Liquid chromatography was performed with a Dionex system. The equip-
ment consisted of a pump P 680, an autosampler Gina 50, an UV detector
UVD 170 U and a column oven STH 585 (software: Chromeleon 6.50).
The column was a Luna C18(2) 250 ꢃ 4.6 from Phenomenex. For valida-
tion a second Luna column and additional a EC 250 ꢃ 4.6 Nucleodur 100-
5
C18 ec (Machery-Nagel) had been used. NMR-spectra were recorded
The alcohol was etherified according to 3.4.3., precipitated in diethyl ether
with maleic acid and recrystallized in acetone. The separated powder was
flocculent and white. The ratio of base and acid was 1/1: yield 10 %; m.p.
with a Gemini 300 spectrometer (Varian).
ꢀ
1
1
52 C; H NMR (300 MHz, CDCl3) d (ppm): 2.93 (6 H, s), 3.33 (2 H, s),
3
.3. Isolation of 8-chlorocaffeine (V)
3
.96 (2 H, s), 5.45 (1 H, s), 6.50 (maleic acid), 7.09–7.52 (Ph, m).
The isolation of 8-chlorocaffeine (V) from dimenhydrinate was carried out
by column chromatography. A glass column (length 60 cm, diameter
3
cm, downward reducing) was packed with silica gel. The solvent con-
3.4.5. N,N,N-[(2-diphenylmethoxyethyl)-(2-dimethylaminoethyl)methyl]-
amine maleate (VI)
sisted of dichloromethane/methanol/ammonia solution (90/9/1). From 1.3 g
dimenhydrinate 10 mg of 8-chlorocaffeine were obtained in about 80%
To 105 g of iced 80% formic acid 16 g of 2-(2-aminoethylamino)ethanol
were added carefully. After the yellow liquid was heated up to 105 C
ꢀ
1
purity; m.p. 178 C; H NMR (300 MHz, CDCl3) d (ppm): 3.44 (3 H, s),
3
6
ꢀ
þ
.58 (3 H, s), 3.99 (3 H, s); MS m/z: 228 (M ), 207, 193, 171, 143, 82,
7.
5
3 g of formalin (30%) were dropped into the mixture (vigorous evolution
of carbon dioxide). Then the reaction mixture was boiled under reflux for
h (Nakajima 1961). The light brown fluid was cooled down to room
4
temperature and 45 ml of hydrochloric acid (6 N) were added. Afterwards
the solution was evaporated under reduced pressure. To the resulting vis-
cous substance 100 ml of sodium hydroxide solution (30%) were added.
The pH value had to be at least 12. The alkaline liquid was extracted
exhaustively with diethyl ether (10 ꢃ 70 ml!). After the removal of the
diethyl ether the residue (yellow oil) was distilled under vacuum. 2-[(2-
Dimethylaminoethyl)methylamino]ethanol was obtained as a colourless oil,
3
.4. Syntheses
The syntheses based on known procedures which have partly been adapted
to similar reactants. Also there were some modifications in techniques to
improve the purity and the yield of the required products. New or differing
analytical data have been added.
ꢀ
yield: 70%; refraction index 1,4545 (25 C). 16,5 g of 2-[(2-Dimethylami-
3
.4.1. 8-Chlorocaffeine (V)
noethyl)methylamino]ethanol were etherified with 23 g of chlorodiphenyl-
methane according to 3.4.3. (TLC reaction control: acetonitrile/chloroform/
triethylamine 1/1/0.1). The base was precipitated from diethyl ether with
maleic acid. The mole ratio of base to acid was 1 : 2. After recrystallization
1
.93 mmol (414 mg) of 8-chlorotheophylline were suspended in 20 ml of
DMF. After addition of 3.86 mmol (533 mg) of K2CO3 and 2.51 mmol
156 ml) of methyl iodide the mixture was stored for 6 h at room tempera-
(
ꢀ
ture. Afterwards 40 ml of water were added. Thereby a clear solution was
obtained. From this solution 8-chlorocaffeine precipitated in colourless
needles after 2 h at 4 C (Vollmann and M u¨ ller 2002): yield 100%; m.p.
1
3
white crystals were obtained: yield 10%; m.p. 152–156 C (lit. 155–
ꢀ
1
156 C (Stelt and Tersteege 1964)); H NMR (300 MHz, D2O) d (ppm):
2.72 (6 H, s), 2.83 (3 H, s), 3.39 (2 H, s), 3.48 (4 H, s), 3.77 (2 H, s), 5.52
(1 H, s), 6.20 (maleic acid), 7.25–7.39 (10 H, m).
ꢀ
ꢀ
1
88 C; H NMR (300 MHz, CDCl3) d (ppm): 3.43 (3 H, s), 3.58 (3 H, s),
13
.99 (3 H, s); C NMR (300 MHz, CDCl3) d (ppm): 28.2 (CH3), 30.0
(
CH3), 32.9 (CH3), 108.5 (C), 139.2 (C), 147.3 (C), 151.5 (C), 154.8 (C);
ꢂ1
3.4.6. 8-[(2-Diphenylmethoxy)ethylmethylamino]theophylline (XI)
IR cm : 3446, 3130, 1713, 1641, 1550, 1449, 1375, 1216, 984; MS m/z:
2
þ
28 (M ), 207, 193, 171, 143, 128, 82, 67 (Sono et al. 1996).
2-Chloroethyldiphenylmethyl ether (300 ml, 340 mg) was dissolved in
1
0 ml of ethyl alcohol and added dropwise into 50 ml of methylamine
solution (33% in ethyl alcohol). The mixture was stirred for 14 days at
room temperature. Then the colourless liquid was evaporated. The residue
was dissolved in ethyl ether. After the addition of maleic acid in ethyl
ether white crystals of 2-(diphenylmethoxy)-N-methylethanamine maleate
were formed: yield 75 %; m.p. 159 C; H NMR (300 MHz, DMSO) d
(ppm): 2.58 (3 H, s), 3.17 (2 H, t), 3.57 (2 H, t), 5.51 (1 H, s), 6.02 (maleic
acid, s), 7.24–7.41 (10 H, m), 8.45 (1 H, s).
3
.4.2. Dibenzhydryl ether (XV)
Benzhydrol (1.84 g) was pulverized in
a
mortar with 3.08 g of
ꢀ
Fe(NO3)3 ꢁ 9 H2O. Then the mixture was heated gently for 30 min at 55 C
in a round-bottom flask (Namboodiri and Varma 2002). Afterwards the
resulted auburn mass was suspended in 20 ml of water and extracted with
ꢀ
1
8
0 ml of diethyl ether. After drying with anhydrous sodium sulphate the
organic layer was evaporated on a rotary evaporator. The residue contained
dibenzhydryl ether, benzhydrol and benzophenone (TLC reaction control:
hexane/ethyl acetate 9/1). Purification was carried out by column chroma-
For the final reaction the maleic acid was removed from 175 mg XI-mal-
eate by addition of 20 ml of 1 M NaOH and extraction with 50 ml of di-
chloromethane. The organic layer was evaporated. Afterwards the basic
residue was diluted in ethyl alcohol together with 57 mg of 8-chlorotheo-
phylline. The mixture was stirred under reflux for 14 days. Then it was
evaporated again. For the separation of the product from the starting sub-
stances column chromatography was used (silica gel; dichloromethane/
ꢀ
tography (silica gel, hexane/ethyl acetate 9/1): yield 50 %; m.p. 106 C;
1
H NMR (300 MHz, CDCl3) d (ppm): 5.62 (2 H, s), 7.44-7.58 (20 H, m);
C NMR (300 MHz, CDCl3) d (ppm): 80.4, 127.69, 127.88, 128.84,
13
1
24.64; MS m/z: 168, 167, 105, 77.
Pharmazie 62 (2007) 3
177

ORIGINAL ARTICLES
methanol/ammonia solution, 98/2/1). After recrystallization from ethyl al-
cohol white crystals were obtained: yield 10%; m.p. 205 C; H NMR
Nakajima K (1961) A new method for the preparation of piperazines. Pre-
paration of N,N -disubstituted piperazines. Bull Chem Soc Jap 34: 655–
ꢀ
1
0
(
(
2
300 MHz, DMSO) d (ppm): 3.07 (3 H, s), 3.18 (3 H, s), 3.32 (3 H, s), 3.55
2 H, t), 3.68 (2 H, t), 5.44 (1 H, s), 7.25–7.4 (10 H, m). MS m/z: 419,
52, 167, 152.
659.
Namboodiri V, Varma R (2002) Iron-catalyzed solvent-free conversion of
alcohols and phenols into diphenylmethyl (DPM) ethers. Tetrahedron
Lett 43: 4593–4595.
Nassr S, Dubuc MC, Lavoie P, Brazier JL (2003) HPLC-DAD methods for
studying the stability of solutions containing hydromorphone, ketorolac,
haloperidol, midazolam, famotidine, metoclopramide, dimenhydrinate,
and scopolamine. J Liq Chromatogr Rel Technol 26: 2909–2929.
Paciolla MD, Jansen SA, Martellucci SA, Osei AA (2001) A fast and effi-
cient determination of amines and preservatives in cough and cold liquid
and suspension formulations using a single isocratic ion-pairing high
power liquid chromatography method. J Pharm Biomed Anal 26: 143–
3
.5. Determination of mole ratio in maleates
The mole ratio of base to acid in the salts was determined indirectly
measuring the content of maleic acid by the HPLC-method which was
already described. 10 ml of 2-[(4-bromophenyl)-phenylmethoxy]-N,N-di-
ꢂ1
methylethanamine maleate (0.605 mg ꢁ ml ; PAmaleic acid ¼ 42.8), N,N,N-
[
(
(
(2-diphenylmethoxyethyl)-(2-dimethylaminoethyl)methyl]amine
maleate
ꢂ1
ꢂ1
0.419 mg ꢁ ml ; PAmaleic acid ¼ 48.6) and a dilution series of maleic acid
ꢂ1
:
0.169 mg ꢁ ml ꢂ 0.844 mg ꢁ ml
y
(PA) ¼ 254.99 ꢃ þ4.4362) were
1
49.
chromatographed. For 2-[(4-bromophenyl)-phenylmethoxy]-N,N-dimethyl-
ethanamine maleate a content of 24.9% was calculated which corre-
sponded to a mole ratio of 1 : 1 (theoretical percentage: 25.7%). N,N,N-[(2-
Diphenyl-methoxyethyl)-(2-Dimethylaminoethyl)methyl]amine maleate con-
tained 41.3% of maleic acid which corresponded to a mole ratio of 1 : 2
PhEur. (2005) deutsche Ausgabe, Dimenhydrinat. 1964–1965.
PhEur. (2005) deutsche Ausgabe, Diphenhydramin-Hydrochlorid. 1976–
1
978.
Qi M, Wang P, Zhou L, Sun Y (2003) Simultaneous determination of four
active components in a compound formulation by liquid chromatogra-
phy. Chromatographia 58: 183–186.
Roos RW, Lau-Cam CA (1986) General reversed-phase high-performance
liquid chromatographic method for the seperation of drugs using triethyl-
amine as a competing base. J Chromatogr A 370: 403–418.
Santoro RS, Warren RJ (1971) Isolation and identification of 8-(2-diphenyl-
methoxy-N-methylethylamine)-1,3-dimethylxanthine from dimenhydri-
nate. J Pharm Sci 60: 791–792.
(
theoretical percentage 42.6%). (PA ¼ peak area).
Acknowledgements: This work was supported by the German Federal In-
stitute for Drugs and Medical Devices.
References
Barbas C, Garcia A, Saavedra L, Castro M (2000) Optimization and vali-
dation of a method for the determination of caffeine, 8-chlorotheophyl-
line and diphenhydramine by isocratic high-performance liquid chroma-
tography –– stress test for stability evaluation. J Chromatogr A 870: 97–
Sono M, Toyoda N, Shimizu K (1996) Functionalization including fluori-
nation of nitrogen-containing compounds using electrochemical oxida-
tion. Chem Pharm Bull 44: 1141–1145.
Stelt C, Tersteege HM (1964) The effect of alkyl substitution in drugs.
Arzneim.-Forsch./Drug Res 14: 1053–1055.
1
03.
Cusic JW (1950) Haloxanthine salts. US Pat. 2499058 (1950), C.A. 4926.
Donelly RR (2002) Chemical stability of dimenhydrinate in minibags and
polypropylene syringes. Can J Hosp Pharm 55: 307–312.
Henderson L, Miller JHM, Skellern GG (2001) Control of impurities in
diphenhydramine hydrochloride by an ion-pair, reversed-phase liquid
chromatographic method. J Pharm Pharmacol 53: 323–331.
Kvist LC, Andersson S-B, Berglund J, Wennergren B, Fors SM (2000)
Equipment for drug release testing of medicated chewing gums. J Pharm
Biom Anal 22: 405–411.
Stiles ML, Allen L, Prince SJ (1994) Stability of dexamethasone sodium
phosphate, diphenhydramine hydrochloride, lorazepam and metoclopr-
amide in portable infusion-pump reservoirs. Am J Hosp Pharm 51:
514–517.
Toda F, Kiyoshige K, Yagi M (1989) Festk o¨ rperreduktion von Ketonen mit
NaBH4. Angew Chem 101: 329–330.
Vollmann K, M u¨ ller CE (2002) Synthesis of 8-substituted xanthine deri-
vates by suzuki cross-coupling reaction. Heterocycles 57: 871–879.
Wolf C, Schunack W (1996) Synthesis and pharmacology of combined
histamine H1-/H2-receptor antagonists containing diphenhydramine and
cyproheptadine derivates. Arch Phar Pharm Med Chem 329: 87–94.
M u¨ ller P, Nury P, Bernardinelli G (2001) Asymmetric nucleophilic substi-
tution of acetales. Eur J Org Chem 4137–4147.
178
Pharmazie 62 (2007) 3