Analysis of phenolic compounds in health care products
Chemiluminescence (CL) is a widely used detection method
for different techniques such as LC, capillary electrophoresis,
thin-layer chromatography and FIA (16) due to the high sensitiv-
ity and wide linear range attainable with simple instrumentation.
(17) Different types of CL detection have been used in FIA
schemes, although without any separation steps, for the deter-
mination of phenolic compounds. For example, Haghighi and
Dadashvand (18) have proposed N-chlorosuccinimide-KI-luminol
chemistry; Costin et al. (19) used FIA with CL detection using
acidic permanganate for the estimation of the total phenolic
content in wine with CL; and Cui et al. proposed the CL coming
from a Ce(IV) and Rhodamine 6G reaction for the determination
of different phenolic compounds. (17) Therefore, the combina-
tion of LC and CL detection offers an improvement over conven-
tional LC methods for phenolic compounds. (17,20)
Although CL detection is well known in LC systems with or
without monolithic columns, the use of CL detection with a
monolithic column and FIA manifolds is less common. Thus,
Adcock et al. (6) proposed a hybrid FIA–LC system for the separa-
tion of six alkaloids and four biogenic amines using a 25 mm
length monolithic column with tris(2,2’-bipyridyl)ruthenium(III)
and permanganate CL detection. Paull et al. (21) used a 3 mm
long chelating monolithic disk for the selective retention and CL
determination of Cu(II) based on 1,10-phenanthroline and hydro-
gen peroxide chemistry. We have described the separation of
parabens in cosmetics using an ultra-short monolithic column
and UV–vis detection in a simple FIA manifold. (11)
Apparatus and software
The flow analysis set-up consisted of two Gilson Minipuls-2 four
channel peristaltic pumps (Gilson Inc., Middleton, USA) working
at a constant flow-rate and four variable volume Rheodyne 5041
Teflon rotary valves (one as injection valve and the other three
as selection valves) controlled electromechanically by a home-
made device. PTFE tubing (Omnifit, Cambridge, UK; 0.8 mm i.d.
and 1.6 mm o.d.) and various end-fittings and connectors of
different diameters were used with a maximum flow rate
attained being 2.6 mL min−1. As the separation element in the
flow system we tested a C18 monolithic mini-column (Chromolith
Guard Cartdrige RP-18 endcapped; 5 × 4.6 mm i.d.) from Merck
(Darmstadt, Germany) and an anionic precolumn CIM from Bia
Separations (Ljubljana, Slovenia).
Chemiluminescence measurements were performed with a
CL-1 Camspec luminometer (Camspec Ltd, Cambridge, UK) inter-
faced to a personal computer via an INT7–24 bit A/D integration
board. Peristaltic pumps, rotary valves and the luminometer were
connected to PC and controlled by software designed by us.
Other instruments used were a Hewlett Packard HP-8453
diode array spectrophotometer (Nortwalk, CT, USA) interfaced to
a PC and equipped with a Hellma 138-QS flow cell with a 1 mm
light path; an Agilent 1100 series liquid chromatograph with
Diode Array Detector (DAD) provided with a C8 Zorbax column
was used to validate the proposed method, and a Crison digital
pH-meter with combined glass-saturated calomel electrode
(Crison Instruments, Barcelona, Spain). For sample extraction, C18
Isolute solid-phase extraction cartridges (Biotage AB, Uppsala,
Sweden) were used.
In this paper we are interested in the use of CL detection in LC
separations using an FIA system that includes an ultra-short
monolithic column, due to its good detection limits and versatil-
ity, in order to improve the characteristics of these methods.
We selected the phenolic compounds commonly used in
formulations of health care products such as phloroglucinol,
2,4-dihydroxybenzoic acid, salicylic acid, methyl paraben and
n-propyl gallate, to verify and propose a simple and fast analyti-
cal solution for these compounds.
Software programs used for the treatment of the data were
CSW32 software package supplied by DataApex Ltd version 1.2.5
(2001) (Prague, Czech Republic), used for the acquisition and
manipulation of the luminescence data, Statgraphics software
package (Manugistics Inc. and Statistical Graphics Corporation,
USA, 1992), version 4.0 (1993) and Microsoft Office 2003.
Procedure
Experimental
Standards and samples. A sample solution or standard con-
taining between 10−7 and 10−4 M each of PH, DHBA, SA, MP and
PG with the same composition as the carrier A was inserted into
the flow analysis system through a loop valve injector (150 μL).
For the mobile phase we used pH 3.0 acetate buffer 0.1 M (carrier
A) and 0.1 M pH 3.0 acetate buffer with 5% ACN:H2O v/v (carrier
B) working in both cases at a flow rate of 2.6 mL min−1. The CL
detection was based on Ce(IV) and Rhodamine 6G chemistry and
two different configurations were used: (1) 6 × 10−2 M Ce(IV) in
0.5 M H2SO4 (solution C) and 5 × 10−5 M Rho 6G (solution E); and
(2) 2 × 10−2 M Ce(IV) in 1.5 M H2SO4 (solution D) and 5 × 10−4 M Rho
6G (solution F). Theses solutions were moved by a second peri-
staltic pump at a flow rate of 2.6 mL min−1.
Initially, the carrier A flowed through the manifold and was
mixed with solutions C and E using a four-way connector. At 150 s
after the injection valve was opened, the carrier changed to
carrier B and at the same time solutions C and E changed to solu-
tions D and F using a programmed sequence for the three
needed selection valves. The CL analytical signal corresponded
to the difference between the maximum CL intensity (I1) and
CL background (I0). The relationship between the concentration
and peak height for phenolic compounds was established by
conventional calibration with external standards.
Reagents and chemicals
All chemicals used were of analytical reagent grade. Aqueous
solutions were prepared using reverse-osmosis type quality water
produced by a Milli-RO 12 plus Milli-Q purification system (Mil-
lipore, Bedford, MA, USA). Phloroglucinol (PH), 2,4-dihydroxy-
benzoic acid (DHBA), salicylic acid (SA), methyl paraben (MP) and
n-propyl gallate (PG) qwre supplied by Sigma (Sigma-Aldrich
Química S.A., Madrid, Spain). We prepared five 100 mg L−1 stock
solutions with each one containing the corresponding phenolic
compound and a mixed solution containing all the phenolic
compounds at 100 mg L−1 each in 5% (v/v) acetonitrile (ACN)
(Panreac Química S.A, Barcelona, Spain, 99.9% v/v HPLC grade).
These solutions were spectrophotometrically stable when
protected from light and kept in the refrigerator at 4°C for at least
2 months. Other solutions were prepared by appropriate dilu-
tions with water while maintaining 5% (v/v) ACN in all instances.
The pH 3.0 buffer acetate 2 M was prepared from acetic acid 96%
(p/v) (d = 1.052 g mL−1) by Panreac and NaOH 0.5 M. The chemi-
luminescent reagents used were 10−2 M ammonium cerium (IV)
sulfate dihydrate (99%) and 1 M sulfuric acid, both by Merck
(Madrid, Spain), and 10−3 M of Rhodamine 6G chloride (Rho 6G)
95% by Sigma.
Luminescence 2011; 26: 44–53
Copyright © 2009 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/luminescence