4706 J. Agric. Food Chem., Vol. 50, No. 16, 2002
Kumar et al.
compete with any phenol-based polymer and even other types
of polymers. It has a major drawback in that it is a mixture of
monomers; to achieve a high molecular weight polymer, purity
of the monomer is of great significance. The polymerization
products of CNSL are generally dark brown in color, and
therefore the application of CNSL in the surface-coating industry
has been limited. The dark color formed during the polymeri-
zation of CNSL was attributed to the dihydric component,
cardol. It was also observed that the presence of cardol and
methylcardol results in cross-linked polymers, affecting the
uniformity of polymer. In view of the industrial applications of
cardanol, there is a need for the development of an efficient,
industrially feasible process to separate it from cardol.
Various methods of purifying technical CNSL are known.
Pure cardanol, free from cardol, has been obtained by column
chromatography, but this method is expensive, time-consuming,
and not commercially viable. Tyman et al. (10) reported an
efficient method to reduce the dihydric component in technical
CNSL, based on formation of an amine-cardol adduct (amine
should have basicity above that of diethylamine and should be
nonvolatile) and distillation of cardanol under high vacuum.
Although this method is efficient, it has a few drawbacks: (1)
distilled cardanol contains ∼6% of cardol as an impurity; (2)
the selection of base becomes difficult and expensive due to
basicity and volatility restrictions (as mentioned above); (3)
high-vacuum distillation could lead to amine contamination in
cardanol. Recently we reported a method (11) for the separation
of cardol and cardanol by solvent extraction. Although this is
the most efficient method reported to date, it is applicable only
for solvent-extracted CNSL and not for technical CNSL. As a
part of our ongoing project, a novel method for the isolation of
cardanol from technical CNSL has been developed.
Figure 2. (a) Technical CNSL; (b) cardanol isolated from technical CNSL;
(c) cardol isolated from technical CNSL.
MATERIALS AND METHODS
filtered through Celite (15 g). The filtrate was dried over anhydrous
sodium sulfate and concentrated to get pure cardanol (65 g). The
methanolic ammonia solution was extracted with ethyl acetate/hexane
(4:1) (2 × 200 mL). The resulting organic layer was washed with 5%
HCl (100 mL) followed by distilled water (100 mL), dried over
anhydrous sodium sulfate, and concentrated to yield pure cardol (20
g). The purity of cardanol and cardol was confirmed by HPLC (Figure
2) and comparison with standard samples as previously described (12).
Hydrogenation of Cardanol. Cardanol (30 g) was dissolved in
methanol (120 mL). Five percent Pd/C (palladium adsorbed on carbon)
(0.75 g) was added slowly, and this solution was transferred to a 250
mL hydrogenation flask. Hydrogenation was carried out with 2.5 kg/
cm2 pressure for 2 h, and the solution was filtered through a Celite
bed to obtain catalyst-free solution. This was evaporated under vacuum
to get crude saturated cardanol, which was recrystallized from petroleum
ether (40-60 °C). The resulting solid was found to have a melting
point of 52-53 °C (lit. mp 51-52 °C; 13).
Technical CNSL. Technical CNSL was obtained from the cashew
nut processor, Mangalore. It contained 22% cardol and 67% cardanol
as quantitated by HPLC.
Chemicals and Solvents. All organic solvents, chemicals, and TLC
plates (silica gel GF254) were obtained from Merck.
NMR. NMR (1H and 13C) spectra were taken on a Bruker DPX200
(40 MHz for 13C and 200 MHz for 1H) and FT-NMR spectra in CDCl3
using tetramethylsilane (TMS) as an internal standard. In the case of
1H NMR typically 500 scans were accumulated. All signals were
referred to TMS to within (0.01 ppm. Typically 1000-2000 scans
were accumulated for the proton noise decoupled 13C NMR spectrum.
All of the signals were referenced to TMS to within (0.1 ppm.
GC-MS. GC-MS analysis was carried out using a Hewlett-Packard
GC 5890 and a Hewlett-Packard 5970 mass selective detector, on an
HP-1 ultra-2 column. Sample (1 mg) was dissolved in 10 mL of
dichloromethane, and 1 µL of this solution was injected into the GC-
MS. The temperature was programmed from 50 to 250 °C at 10 °C/
min and maintained at 250 °C for 30 min.
RESULTS AND DISCUSSION
HPLC. HPLC analysis was done on a modular HPLC instrument
comprising two 510 reciprocating pumps, a 481 variable-wavelength
detector, and a Rheodyne injector (20 µL loop), all from Waters Corp.
A Supelcosil LC-18 (4 mm × 150 mm) column was used. The mobile
phase was acetonitrile/water/acetic acid (80:20:1) at a flow rate of 1.80
mL/min. Absorbance was monitored at 280 nm. Each analysis was
carried out by dissolving 25 mg of sample in 5 mL of acetonitrile and
filtering through a C18 Sep-Pak cartridge (Water Associates, Milford,
MA).
Separation of Cardanol from Cardol. Technical CNSL (100 g)
was dissolved in methanol (320 mL), and ammonium hydroxide (25%,
200 mL) was added and stirred for 15 min. This solution was then
extracted with hexane (4 × 200 mL). The organic layer was washed
with 5% HCl (100 mL) followed by distilled water (100 mL). Activated
charcoal (10 g) was added to the organic layer, stirred for 10 min, and
This process is an extension of the work reported for solvent-
extracted CNSL by our group (11). In this method, technical
CNSL was dissolved in a mixture of methanol and ammonium
hydroxide. It was observed that the ratio of methanol and
ammonium hydroxide was crucial. A series of experiments were
done with different ratios of methanol and ammonium hydrox-
ide, and the optimum was found to be 8:5. In the case of our
earlier reported method (11) for solvent-extracted CNSL, the
methanol and ammonium hydroxide ratio was 4:5. This solution
was extracted with hexane to obtain cardanol in the pure form.
In the case of solvent-extracted CNSL (11), ethyl acetate (2-
5%) in hexane was essential to isolate pure cardanol. This is
due to the presence of a higher percentage of cardol than
Phani Kumar
