Food

Chemistry
Food Chemistry 103 (2007) 243248
www.elsevier.com/locate/foodchem

Analytical, Nutritional and Clinical Methods

The simultaneous separation and determination of ve organic acids

in food by capillary electrophoresis
a,b a,*
Yajun Tang , Mingjia Wu
a
National Analytical Research Center of Electrochemistry and Spectroscopy, Changchun Institute of Applied Chemistry, Chinese Academy of
Sciences, 5625 Renmin Street, Changchun 130022, PR China
b
Graduate School of the Chinese Academy of Sciences, Chinese Academy of Sciences, 19 Yuquan Road, Beijing 100049, PR China

Received 25 January 2005; received in revised form 31 August 2005; accepted 25 September 2005

Abstract

The simultaneous separation of quinic acid, anisic acid, salicylic acid, benzoic acid and sorbic acid was performed by capillary elec-
trophoresis with 20 mmol L 1 NaOH20 mmol L 1 H3BO3 (pH 8.8) as the background electrolyte. The choices of the background elec-
trolytes and the applied voltage were optimized. The eects of tetrabutylammonium bromide and sodium dodecylsulfate on the
separation were investigated in detail. Under the optimum condition, the linearity, reproducibility and detection limit of ve organic
acids were shown, respectively. As an application of the method proposed, 10 samples of soy sauce and vinegar were analysed to deter-
mine benzoic acid and sorbic acid. The procedure described provides the advantages of good selectivity, rapid speed and simplicity for
the separation and determination of organic acids.
 2005 Elsevier Ltd. All rights reserved.

Keywords: Capillary electrophoresis; Organic acids; Soy sauce; Vinegar

1. Introduction be sensitive to dierent preservatives, so the kinds and

the amount of the preservatives must be controlled.
Organic acids were widely distributed in fruits, cheese, Organic acids have been separated and determined by
wine, and vegetables (Yang & Choong, 2001). Quinic acid, chromatographic techniques such as high-performance li-
anisic acid, salicylic acid and benzoic acid belong to pheno- quid chromatography (HPLC) (Escobal, Gonzalez, & Iri-
lic acid. Benzoic acid and sorbic acid are widely used as ondo, 1997; Falque-Lopez & Fernandez-Gomez, 1996;
antifungal preservatives against molds and yeasts to ensure Shui & Leong, 2002; Vonach, Lendl, & Kellner, 1998;
long-term microbial stability (Mikami, Goto, Ohno, Mat- Zhao, Guo, Chang, & Wang, 2001), gas chromatography
sumoto, & Nishida, 2002; Pant & Trenerry, 1995). Benzoic (GC) (Deng, 1997), thin-layer chromatography (Thompson
acid is permitted for use in non-dairy dips where the pH is & Hedin, 1966), gasliquid chromatography (Sarkar &
4.5 or less. As a preservative, sorbic acid inhibits the Malhotra, 1979), and ion chromatography (IC) (Ding, Su-
growth of yeasts, fungi and bacteria over a much wider zuki, & Koizumi, 1995; Morales, Gonzalez, & Troncoso,
pH range than benzoic acid and is eective up to a pH of 1998). Compared with the methods above, capillary elec-
6.5. The appropriate amount of benzoic acid or sorbic acid trophoresis (CE) oers high resolution eciency, rapid sep-
appears to be safe, which has been tested by many labora- aration speed, simple sample pretreatment, small
tories throughout the world. Meanwhile, individuals may consuming sample, and short analysis time, while having
no derivatization often needed in GC and sample extrac-
tion encountered in HPLC or IC.
*
Corresponding author. Tel.: +86 431 5262403; fax: +86 431 5685653. In this paper, a simple capillary zone electrophoresis
E-mail address: mjwu@ciac.jl.cn (M. Wu). (CZE) method for the simultaneous separation and

0308-8146/$ - see front matter  2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.foodchem.2005.09.022
244 Y. Tang, M. Wu / Food Chemistry 103 (2007) 243248

determination of quinic acid, anisic acid, salicylic acid, ben- was adjusted with H3BO3 or NaOH. The samples and
zoic acid and sorbic acid was performed with 20 mmol L 1 BGEs were ltered through a 0.45 lm cellulose acetate l-
NaOH20 mmo L 1 H3BO3 (pH 8.8) as the background ter membrane (Jiangsu Qilin Medical Instrument Factory,
electrolyte (BGE). The eects of tetrabutylammonium bro- China).
mide (TBAB) and sodium dodecylsulfate (SDS) on the sep- The capillary was conditioned for 30 min with 0.1 mol
aration were investigated in detail. As an application of this L 1 NaOH and 10 min with water. Additionally, the capil-
method, 10 samples of soy sauce and vinegar were analysed lary washed for 2 min with 0.1 mol L 1 NaOH, 2 min with
to determine benzoic acid and sorbic acid, which were often water and 5 min with the running buer before each run.
added as preservative. Between runs, the capillary was washed with running buf-
fer. The experiment was done under constant humidity
2. Materials and methods (40%).

2.1. Chemicals 2.3. Samples and sample preparation

Quinic acid was purchased from Takeda Chemical The real samples were Taoda soy sauce (Shanghai Tao-
Industries Ltd. (Japan), sorbic acid from the Pharmacy da Food Co. Ltd.), Hubao seasoning sauce (Tianjin Hubao
Faculty of Second Military Medical University (China), Flavour-Making Co. Ltd.), Donghu mature vinegar (Shan-
anisic acid from International Dispensary Co., Ltd. (Chi- xi Meiheju Mature Vinegar Co. Ltd.), Zilin mature vinegar
na), benzoic acid from Shanghai Laize Fine Chemicals (Shanxi Qingxu Laifu Mature Vinegar Co. Ltd.), Haitian
Factory (China), salicylic acid from Xinzhong Chemical light soy sauce (Foshan Haitian Flavour-Making Co.
Factory Co., Ltd. (China), cinnamic acid from Beijing Ltd.), Beikang dark soy sauce (Jilin Beikang Fermentation
Chemical Reagent Research Institute (China), TBAB from Food Co. Ltd.), Haitian seafood soy sauce (Foshan Hai-
E.MERCK (West Germany), SDS from Zibo Jujin Chem- tian Flavour-Making Co. Ltd.), Beikang vinegar (Jilin
ical Co., Ltd. (China). Sodium hydroxide was from Beijing Beikang Fermentation Food Co. Ltd.), Beikang rice vine-
Chemical Plant (China), (hydroxymethyl) aminomethane gar (Jilin Beikang Fermentation Food Co. Ltd.) and Beik-
(Tris) from Changshu Fushan Biology and Chemistry Re- ang avor vinegar (Jilin Beikang Fermentation Food Co.
agents Factory (China), boric acid from Beijing Chemical Ltd.). All the samples were purchased from a supermarket
Reagent Factory (China), borax from Beijing Xinguang in Changchun. Soy sauce and vinegar with the labeling of
Chemical Reagent Factory (China), disodium hydrogen food preservatives were shown in Table 1. Vinegar was di-
phosphate from Hongxing Chemical Reagent Factory (Bei- rectly diluted with double-distilled water to 1:10 (v/v) and
jing, China) and carbon tetrachloride from Beijing Chemi- ltered through a 0.45 lm lter membrane. Two millilitres
cal Reagent Factory (China). All reagents were of of soy sauce and 1 ml CCl4 were mixed and the solution
analytical purity and used without further purication. was vortexed for 10 min. Then, the lipids extracted into
the organic phase were discarded. The aqueous phase solu-
2.2. Apparatus and method tion obtained was transferred into a dry vial. The aqueous
phase solution was diluted with double-distilled water to
All experiments were performed on a laboratory-built 1:10 (v/v) and ltered through a 0.45 lm lter membrane.
CE unit, which comprised a 3 kV high voltage supply
(Shanghai Institute of Nuclear Research), a CV4 variable- 2.4. Linearity, reproducibility and detection limit
wavelength detector (ISCO, Lincoln, NE, USA), and a
75-lm-I.D., 360-lm-O.D. fused-silica capillary tube Using the optimized instrumental and operational
(Yongnian Optical Conductive Fiber Plant, Hebei, China). parameters for the standard solution, the linearity of the
The total length of the capillary was 55 cm, and the detec- method was calculated and expressed as the correlation
tion was performed at 40 cm downstream. The samples coecient, and the reproducibility of the method was mea-
were introduced into the anode end of the capillary by
gravity injection at 6.5 cm height for 8 s. The pH meter Table 1
was from Shanghai Analytical Instrument Factory of Description of samples analysed
China. Sample name Labeling of food preservatives
Boric acid, borax, disodium hydrogen phosphate, Taoda soy sauce Benzoic acid Sorbic acid
TBAB, SDS and Tris were dissolved in double-distilled Beikang dark soy sauce Benzoic acid
water, respectively, to prepare the concentration of Haitian light soy sauce Benzoic acid Sorbic acid
Hubao seasoning sauce Sorbic acid
0.1 mol L 1 as the stock solutions. Stock standard solu-
Haitian seafood soy sauce Sorbic acid
tions of organic acids were prepared at a concentration Beikang vinegar Benzoic acid
of 1 10 3 mol L 1 using double-distilled water, respec- Beikang rice vinegar Benzoic acid
tively. In all cases, the pH value was measured in a PHS- Donghu mature vinegar Benzoic acid Sorbic acid
3CT pH meter (Shanghai Analytical Instrument Factory, Zilin mature vinegar Benzoic acid
Beikang avor vinegar Benzoic acid
China) with a combined electrode. The required pH value
 Y. Tang, M. Wu / Food Chemistry 103 (2007) 243248 245

Table 2 Faster separations are obtained in CE when the electropho-

Summary of calibration and precision data for the method retic vector and the electroosmotic ow are in the same
Acids The regression r RSD Linearity LOD direction. The separations of ve organic acids in four dif-
equation (%) (mg L 1) (mg L 1) ferent BGEs including 20 mmol L 1 borax20 mmol L 1
Quinic acid y = 0.11x + 0.02a 0.9978 2.59 5.41288.67 1.80 H3BO3, 20 mmol L 1 Tris20 mmol L 1 H3BO3, 20
Anisic acid y = 0.12x + 0.03 0.9969 2.70 3.19106.54 1.21 mmol L 1Na2HPO420 mmol L 1 H3BO3 and 20
Sorbic acid y = 0.24x + 0.03 0.9979 2.58 1.12112.13 0.44
Benzoic acid y = 1.31x + 0.05 0.9965 2.80 2.9373.27 2.19
mmol L 1 NaOH20 mmol L 1 H3BO3, were tried and
Salicylic acid y = 1.23x + 0.05 0.9966 2.78 3.1796.68 2.07 the migration times were shown in Table 3. Benzoic acid
Mean values of seven determinations (n = 7).
and salicylic acid had the same migration time in
BGE: 20 mmol L 1 NaOH20 mmol L 1 H3BO3 (pH 8.8); applied volt- 20 mmol L 1 borax20 mmol L 1 H3BO3 (pH 8.8). Asinic
age: 20 kV; injection: 6.5 cm 8 s. acid and sorbic acid could not be separated completely
a
y, the peak area ratio (y) of the respective organic acid to cinnamic and synchronously in 20 mmol L 1 Tris20 mmol L 1
acid; x, concentration (mg L 1). H3BO3 (pH 8.8). In 20 mmol L 1Na2HPO420 mmol L 1
H3BO3 (pH 8.8), ve organic acids were separated, but
sured by replicate injections of the standard solution. Cin- benzoic acid and salicylic acid had leading peaks. In
namic acid was used as the internal standard. The regres- 20 mmol L 1 NaOH20 mmol L 1 H3BO3 (pH 8.8) ve or-
sion equations of ve organic acids, the correlation ganic acids were separated and detected absolutely with
coecient, the relative standard deviation (RSD) (n = 7), sharp and symmetrical peaks, compared with the other
the linearity range and the limits of detection (LOD) (sig- BGEs.
nal-to-noise ratio of three) were shown in Table 2. Dilute electrolyte solutions are desirable to shorten
migration time. However, it is important that electrolyte
2.5. Validation of the proposed methods solutions have sucient buer capacity to maintain a xed
pH and guarantee the strong mutual action in the presence
Typical electropherograms of the separation of ve or- of the analytes. Inadequate buer capacity may result in
ganic acids by UV detection were shown in Figs. 4(a) the poor resolution. So, we alternated the BGE from 5
and 5(a). The reproducibility of peak area and migration mmol L 1 NaOH5 mmol L 1 H3BO3 (pH 8.8) to 25
time in this experiment was determined by injecting a solu- mmol L 1 NaOH25 mmol L 1 H3BO3 (pH 8.8). The ef-
tion of the mixture of ve organic acids (5 10 5 mol L 1 fects of dierent electrolyte concentrations on the separa-
in each) into the system for seven times, with 20 mmol L 1 tion of ve organic acids showed that 20 mmol L 1
NaOH20 mmol L 1 H3BO3 (pH 8.8) as BGE under the NaOH20 mmol L 1 H3BO3 (pH 8.8) gave complete sepa-
optimum conditions. RSDs of peak area and migration ration while the higher electrolyte concentrations brought
time were 2.0% and 0.9%, respectively, for quinic acid, about longer migration time and broader peak. The elec-
2.1% and 1.1%, respectively, for anisic acid, 2.1% and tropherograms of the above BGEs revealed that 20
1.2%, respectively, for sorbic acid, 2.3% and 1.3%, respec- mmol L 1 NaOH20 mmol L 1 H3BO3 (pH 8.8) was
tively, for benzoic acid and 2.6% and 1.4%, respectively, for suitable.
salicylic acid. The high reproducibility indicates that this
method is suitable for the analysis of real samples. 3.2. The eect of TBAB concentration

3. Results and discussion Cationic surfactants can aect the EOF in micellar elec-
trokinetic chromatography (MEKC) owing to the electro-
3.1. Optimization of the background electrolyte static interactions between the positively charged
surfactant monomers and the negatively charged fused-sil-
BGE aects the migration time and the separation be- ica wall. The feasibility of making use of the interactions
tween compounds, directly. It is necessary to provide high between charged analytes and TBAB in the capillary was
electroosmotic ow (EOF) in basic conditions for high- studied. The resolution variation of ve organic acids as
speed and high-resolution separations of organic acids. the function of TBAB concentration was shown in Fig. 1.

Table 3
Migration times of organic acids obtained in dierent BGEs
BGE Migration time
Quinic acid Cinnamic acid Anisic acid Sorbic acid Benzoic acid Salicylic acid
1 1
20 mmol L borax20 mmol L H3BO3 2.44 4.88 4.97 5.08 5.61 5.61
1
20 mmol L Tris20 mmol L 1 H3BO3, 2.23 4.04 4.11 4.16 4.45 4.60
1
20 mmol L Na2HPO420 mmol L 1 H3BO3 2.36 4.10 4.16 4.29 4.72 5.05
1
20 mmol L NaOH20 mmol L 1 H3BO3 2.19 3.93 4.01 4.08 4.43 4.65
Mean values of ve determinations (n = 5).
Applied voltage: 20 kV; injection: 6.5 cm 8 s.
246 Y. Tang, M. Wu / Food Chemistry 103 (2007) 243248

Fig. 1. The eect of TBAB concentration on the separation of organic acids. (1) quinic acid; (2) cinnamic acid; (3) anisic acid; (4) sorbic acid; (5) benzoic
acid; (6) salicylic acid. (a) 0 mmol L 1 TBAB; (b) 5 mmol L 1 TBAB; (c) 10 mmol L 1 TBAB; (d) 15 mmol L 1 TBAB; (e) 20 mmol L 1 TBAB; (f) 25
mmol L 1 TBAB. BGE: 20 mmol L 1 NaOH20 mmol L 1 H3BO3 (pH 8.8); applied voltage: 20 kV; injection: 6.5 cm 8 s; UV wavelength: 240 nm.

The migration order of the organic acids was quinic acid, 3.3. The eect of SDS concentration
followed by cinnamic acid, anisic acid, sorbic acid, benzoic
acid and salicylic acid. When 5 mmol L 1 TBAB was SDS is the most regularly used anionic surfactant in
added to the electrolyte, the separation window between MEKC separations. SDS of 10, 20, 30, 40 and 50
benzoic acid and salicylic acid was diminished. Benzoic mmol L 1 were added sequentially to the electrolyte solu-
acid and salicylic acid had the same migration time when tion of 20 mmol L 1 NaOH20 mmol L 1 H3BO3 (pH
TBAB concentration increased to 10 mmol L 1. The sepa- 8.8) to investigate whether the presence of SDS could inu-
ration between benzoic acid and salicylic acid became bet- ence the resolution. From these experiments, it was found
ter gradually along with the TBAB concentration changed that the resolution between cinnamic acid and anisic acid
from 15 to 20 mmol L 1. However, when the concentration reduced from 1.97 to 1.20 after the addition of SDS, and
of TBAB reached 25 mmol L 1, benzoic acid overlapped that the migration times of all the organic acids increased.
with sorbic acid, partly. During all the processes, the sepa- Fig. 2 showed the electropherograms of the separation of
ration window between anisic acid and sorbic acid got ve organic acids by MEKC with SDS as the anionic
enlarged. surfactant.
 Y. Tang, M. Wu / Food Chemistry 103 (2007) 243248 247

3.4. The choice of applied voltage 3.5. Application

The eciency of the separation improves and the migra- As an application, soy sauce and vinegar were prepared
tion time shortens when the applied voltage is increased. using the procedure described in Section 2.3. Cinnamic acid
But the higher applied voltage will result in the eciency was used as the internal standard and the electrophero-
of the separation falling because of the Joule heating eect. grams were depicted (Figs. 4 and 5). By standard addition
On the other hand, the reproducibility becomes worse and method and comparing migration times of analytes with
the migration time prolongs if the applied voltage gets low- those of the standard mixture solution, benzoic acid and
er. In this experiment, the ve organic acids were separated sorbic acid in soy sauce and vinegar were determined.
as the applied voltage increased from 12.5 to 22.5 kV The recovery and reproducibility experiments under the
(Fig. 3). The current increased slightly and was always be- optimum conditions were conducted to evaluate the preci-
low 10 lA. Under the applied voltage above, ve organic sion and accuracy of the method. Recovery was determined
acids were well resolved except for 22.5 kV. A 20 kV ap- by standard addition method and the determination results
plied voltage was found to be an acceptable compromise of the real samples were listed in Tables 4 and 5. Nowa-
between adequate resolution and short migration time. days, HPLC (Garcya, Ortiz, Sarabia, Vilches, & Gredilla,
2003; Saad, Bari, Saleh, Ahmad, & Talib, 2005) is the most
common analytical procedure for the determination of
8 Quinic acid
Cinnamic acid
benzoic acid and sorbic acid. Compared with HPLC, the
Anisic acid method proposed in this paper simplies considerably the
Sorbic acid analysis, reducing its cost and shortening migration time.
7 Benzoic acid
Salicylic acid
The procedure described provides the advantages of good
Migration time (min)

10 20 30 40 50 60
-1
SDS concentration (mmolL )

Fig. 2. The eect of SDS concentration on the separation of ve organic

acids (the values expressed as average of ve injections). Other conditions
were the same as in Fig. 1.

Quinic acid
7 Cinnamic acid
Anisic acid
Sorbic acid
Migration time (min)

6 Benzoic acid
Salicylic acid

12 14 16 18 20 22
Applied voltage(kV) Fig. 4. The electropherograms of determining benzoic acid in the samples.
(a) The standard solution; (b) Taoda soy sauce + 5 10 5 mol L 1
Fig. 3. Migration times of organic acids under dierent applied voltage cinnamic acid; (c) Beikang vinegar + 5 10 5 mol L 1 cinnamic acid.
(the values expressed as average of ve injections). Other conditions were BGE: 20 mmol L 1 NaOH20 mmol L 1 H3BO3 (pH 8.8); applied
the same as in Fig. 1. voltage: 20 kV; injection: 6.5 cm 8 s; UV wavelength: 225 nm.
248 Y. Tang, M. Wu / Food Chemistry 103 (2007) 243248

selectivity, rapid speed and simplicity for the separation

and determination of organic acids and these results indi-
cates that the proposed method can be useful for research
and routine analysis of organic acids in many application
areas.

References

Deng, C. R. (1997). Determination of total organic acids in wine by

interfacial deviatization gas chromatographic methods. Sepu, 15,
505507.
Ding, M. Y., Suzuki, Y., & Koizumi, H. (1995). Simultaneous determi-
nation of organic acids, inorganic acids and cations in beverages by ion
chromatography with a mixed-bed stating phase of anion and cation
exchanges. The Analyst, 120, 17731777.
Escobal, A., Gonzalez, J., & Iriondo, C. (1997). Laborra. Liquid
chromatographic determination of organic acids in txakoli from
Bizkaia. Food Chemistry, 58, 381384.
Falque-Lopez, E., & Fernandez-Gomez, E. (1996). Simultaneous deter-
mination of the major organic acids, sugars, glycecol, and ethanol by
HPLC in grape musts and white wine. Journal of Chromatographic
Science, 35, 254257.
Garcya, I., Ortiz, M. C., Sarabia, L., Vilches, C., & Gredilla, E. (2003).
Advances in methodology for the validation of methods according to
the International Organization for Standardization Application to the
determination of benzoic and sorbic acids in soft drinks by high
performance liquid chromatography. Journal of Chromatography A,
992, 1127.
Mikami, E., Goto, T., Ohno, T., Matsumoto, H., & Nishida, M. (2002).
Simultaneous analysis of dehydroacetic acid, benzoic acid, sorbic acid
and salicylic acid in cosmetic products by solid-phase extraction and
high-performance liquid chromatography. Journal of Pharmaceutical
and Biomedical Analysis, 28, 261267.
Morales, M. L., Gonzalez, A. G., & Troncoso, A. M. (1998). Ion-
Fig. 5. The electropherograms of determining sorbic acid in the samples. exclusion chromatographic determination of organic acids in vinegars.
UV wavelength: 255 nm.Other conditions were the same as in Fig. 4. Journal of Chromatography A, 822, 4551.
Pant, I., & Trenerry, V. C. (1995). The determination of sorbic acid
and benzoic acid in a variety of beverages and foods by micellar
Table 4 electrokinetic capillary chromatography. Food Chemistry, 53,
The results for CE analysis of benzoic acid in the samples 219226.
Sample The average value RSD Recovery Saad, B., Bari, M. F., Saleh, M. I., Ahmad, K., & Talib, M. K. M. (2005).
found (mg L 1) (%) (%) Simultaneous determination of preservatives (benzoic acid, sorbic acid,
methylparaben and propylparaben) in foodstus using high-perfor-
Taoda soy sauce 78.15 2.90 98.30
mance liquid chromatography. Journal of Chromatography A, 1073,
Beikang dark soy sauce 60.98 2.95 98.25
393397.
Haitian light soy sauce 60.27 2.97 98.21
Sarkar, S. K., & Malhotra, S. S. (1979). Gas-liquid chromatographic
Beikang vinegar 63.50 2.85 98.90
method for separation of organic acids and its application of pine
Beikang rice vinegar 65.24 2.86 98.85
needle extracts. Journal of Chromatography A, 171, 227232.
Donghu mature vinegar 112.31 2.87 98.83
Shui, G. H., & Leong, L. P. (2002). Separation and determination of
Zilin mature vinegar 89.22 2.86 98.85
organic acids and phenolic compounds in fruit juices and drinks by
Beikang avor vinegar 90.04 2.85 99.01
high-performance liquid chromatography. Journal of Chromatography
Mean values of ve determinations (n = 5). A, 977, 8996.
Other conditions as in Fig. 4. Thompson, A. C., & Hedin, P. A. (1966). Separation of organic acids by
thin-layer chromatography of their 2,4-dinitrophenylhydrazide deriv-
atives and their analytical determination. Journal of Chromatography
Table 5 A, 21, 1318.
The results for CE analysis of sorbic acid in the samples Vonach, R., Lendl, B., & Kellner, R. (1998). High-performance liquid
Sample The average value RSD Recovery chromatography with real-time Fourier-transform infrared detection
found (mg L 1) (%) (%) for the determination of carbohydrates, alcohols and organic acids in
wines. Journal of Chromatography A, 824, 159167.
Taoda soy sauce 17.94 2.67 99.00 Yang, M. H., & Choong, Y. M. (2001). A rapid gas chromatographic
Haitian light soy sauce 15.21 2.70 98.20 method for direct determination of short-chain (C2C12) volatile
Hubao seasoning sauce 23.46 2.65 99.18 organic acids in foods. Food Chemistry, 75, 101108.
Haitian seafood soy sauce 15.34 2.65 99.03 Zhao, J. C., Guo, Z. A., Chang, J. H., & Wang, W. (2001). Study on
Donghu mature vinegar 24.01 2.60 99.24 reserved-phased high performance liquid chromatography separa-
Mean values of ve determinations (n = 5). tion conditions determination method of organic acids. Sepu, 19,
Other conditions as in Fig. 5. 260263.