HPLC DETERMINATION OF ARBUTIN 37
18). Furthermore, there is no data in the literature about the isolation and quantitative
analysis of arbutin in Betula alnoides Buch. Ham., Clerodendrum petasites S. Moore, Cur-
culigo latifolia Dryand. Var. latifolia, and Hesperethusa crenulata (Roxb.) Roem. Therefore,
it is interesting to investigate the arbutin content in these medicinal plants, because
naturally occurring arbutin is very safe skin-whitening agent.
EXPERIMENT AL
APPARATUS
HPLC analyses were carried out with Hewlett Packard Model ll00 liquid chromato-
graph with autosampler, thermostatic column compartment, online degasser, and a
UV-visible detector model G 1313 A. The column used was ODS Hypersil® C
18
(125
mm x 4 mm, 5 .0 µm [Chromtech, Stockholm, Sweden}) with a Lichrosphere® 100
RP-18 (4 mm x 4 mm, 5.0 µm) guard column. The mobile phase was a mixture
containing varying ratios of methanol, water, and 0.1 M hydrochloric acid, vacuum-
filtered through 0.45-µm nylon membranes (Germany) before use. The following in-
struments were also used: a simultaneous spectrophotometer (Spekol 1200), a pH-meter
(Model pH 900, Precisa, Switzerland), a water bath and shaker (Model SB-200-10,
Thailand), an ultrasonicator (Model 889, Cole Parmer, USA), and a rotary evaporator
(EYELA N-N series).
REAGENTS
The following standard reagents were used: arbutin HPLC grade 98% (Sigma, St Louis,
MO) and resorcinol 98% (Fluka). The following reagents were used: hydrochloric acid
(AR) (Farmitalia Carlo Erba, Italy), glacial acetic acid (AR) (Farmitalia Carlo Erba),
acetonitrile (HPLC grade) (Lab-Scan Analytical Sciences, Ireland), ethyl acetate (AR)
(BDH laboratory supplies, England), methanol (HPLC grade) (Lab-Scan Analytical Sci-
ences), and ether (AR) (Lab-Scan Analytical Sciences). De-ionized distilled water was
used throughout.
SAMPLES
Skin whitening products. Arbuwhite® cream (Nature Best Health Product Co., Ltd.,
Thailand), Super Whitening® cream (Aunyamanee Herbs, Thailand), and Shiseido®
cream (Shiseido Co., Ltd., Tokyo, Japan) were used.
Plant material and location. The bark of Betula alnoides was collected at Bah Bae Village,
Mae Dtang District, Chiang Mai Province, Thailand, in October 2003 and was iden-
tified by W. Thongchai 1. Voucher specimens have been deposited at CMU Herbarium,
Chiang Mai University, Chiang Mai, Thailand. The roots of Clerodendrum petasites S.
Moore were collected from Chiang Dow, Cl1aing Mai, Thailand. The tubers of Curculigo
latifolia Dryand. Var. latifolia were collected from Papae, Maetang, Chiang Mai, Thai-
land. The trunk of HeJperethusa crenulata (Roxb.) Roem. was collected from Mae Sai,
Chiang Rai, Thailand.

38 JOURNAL OF COSMETIC SCIENCE
PROCEDURES
Sample preparation (skin-whitening cream). About 0.5 g of each whitening cream was
accurately weighed and transferred into three separate 25-ml volumetric flasks and
dissolved in methanol. To each flask 50 µg/ml- 1 of resorcinol was added as an internal
standard. The solution was sonicated vigorously for 30 min, centrifuged at 4000 rpm for
30 min, and filtered on a Millipore membrane (0.45 µm) to obtain a transparent
solution. The supernatant liquid was used for chromatographic analysis.
Extraction of medicinal plants. The dried medicinal plants were powdered. Then 6 kg of
the powder was extracted with two successive portions of 5 .0 1 of de-ionized water and
methanol. They were shaken in a wrist-action shaker for five hours and filtered. Then the
solvent of the filtrate could be removed either by using a spray-dried technique (tem-
perature 100°C and flow rate 1.0 ml/min- 1
)to give a brown powder, or by using a
rotatory evaporator to give a dark brown crude residue.
Preparation of standard solutions. A 1,000 µg/ml- 1 stock solution of arbutin standard was
prepared in methanol. A series of each standard solution containing 0.5, 1.0, 3.0, 5.0,
10.0, and 30.0 µg/ml- 1 was prepared from the stock standard solution.
Preparation of sample solutions. Three sets of medicinal crude extracts (5 g) and cosmetic
samples (0.5 g) of each set were extracted under reflux with 100 ml of 75% methanol
for 30 min and filtered. The filtrate was evaporated to about 12 ml and transferred into
a 250-ml separator followed by addition of 50 ml of water. The mixture was then
extracted with ether (2 x 30 ml). The combined aqueous layer was extracted with ethyl
acetate (3 x 50 ml). The combined ethyl acetate extract was then evaporated to dryness
and dissolved in 10 ml of methanol.
Preliminary investigation. A preliminary investigation was carried out to separate some
chemical constituents by TLC. The crude extract was extracted with 75% methanol
under reflux for 30 min and then filtered. The filtrate was evaporated to about 12 ml and
transferred to a 250-ml separating funnel together with 50 ml of water. This solution
was extracted three times with 50 ml of ethyl acetate, and the combined ethyl acetate
extracts were evaporated to dryness and the residues dissolved in 10% methanol. The
sample solution and the standard solutions were separated on a silica gel GF2
54 (20 x 20
cm) glass plate, using ethyl acetate:methanol (9: 1) as a developing solvent. The crude
extract gave five well-defined spots. The Rf value of each spot was exactly the same as
that obtained from each spot of standard.
Optimization of experimental conditions for RP-HPLC. RP-HPLC was performed under
isocratic conditions. All experimental conditions were optimized by means of a univari-
ate method as follows:
Analytical wavelength. Optimum absorbance of each standard solution was determined
by injection of the same amount of mixed standard solutions (5 .0 µg/ml- 1 )at different
wavelengths from 200 nm to 400 nm. The mobile phase was a mixture consisting of
water:methanol (80:20 v/v) with a flow rate of 1.0 ml/min- 1
.As the optimum to obtain
the best sensitivity,
"-max was chosen.
Mobile phase. Various solvent systems were tested as the mobile phase for the separation
of arbutin in the samples, e.g., water:acetonitrile:0.1 M hydrochloric acid (94:5: 1, v/v/v),
water:methanol:0.1 M hydrochloric acid (89: 10: 1, v/v/v), and methanol: 100 mM phos-
phate buffer, pH 2.1 (10:90 v/v).