370 Chiang Mai J. Sci. 2014; 41(2)

Chiang Mai J. Sci. 2014; 41(2) : 370-382http://epg.science.cmu.ac.th/ejournal/Contributed Paper

Validation of a HPLC Method for the Determination ofBenzoic Acid and Sorbic Acid in NoodlesJankana Burana-osot*, Leelawan Arunsingkharat, Matsarin Naksuk, Supapan Naungnamjaiand Thanapon SaetunDepartment of Pharmaceutical Chemistry, Faculty of Pharmacy, Silpakorn University,Nakorn-pathom, 73000, Thailand.*Author for correspondence; e-mail: jankana@su.ac.th

Received: 23 August 2012Accepted: 26 December 2012

ABSTRACTA simple, sensitive and specific HPLC method was validated for the simultaneous

determination of benzoic acid and sorbic acid in noodles. Sample preparation involved theextraction with methanol and water (60:40, v/v). The separation was achieved on a Germini-C18 modified silica column (50 mm x 4.6 mm i.d., 5 μm particle diameter). The mobile phaseconsisted of 0.05 M ammonium acetate (pH 4.4) and methanol in the ratio of 60:40 (% v/v)at a flow rate of 1 mL/min and detection was performed at 234 nm using a diode-arraydetector (DAD). The method was validated with respect to specificity, linearity, accuracy,precision, limit of detection (LOD), limit of quantitation (LOQ) and robustness. The calibrationcurve for benzoic acid and sorbic acid was found to be linear in the concentration range of5-200 μg/mL (r2 0.9998) and 1-50 μg/mL (r2 0.9998), respectively. The values of LOD andLOQ were 0.42 and 1.14 μg/mL and 0.32 and 0.99 μg/mL for benzoic acid and sorbic acid,respectively. The accuracy determined by spike recovery measurement was 85.61-102.04%for benzoic acid and 83.62-102.47% for sorbic acid. The method and intermediate precisionfor benzoic acid and sorbic acid were demonstrated by the relative standard deviation of1.84% and 2.40% and 1.41% and 2.80%, respectively. The validation acceptance criteria weremet in all cases. The proposed method was successively applied for monitoring of benzoicacid and sorbic acid in 5 different kinds of noodles.

Keywords: benzoic acid, sorbic acid, noodles, HPLC-DAD

1. INTRODUCTIONNoodles (Kuy-Teaw) are generally

consumed in Thailand as oriental cultures andhave been exported to international markets.One of the most popular and favorite dishworldwide is Stir-fried rice noodle (Pad-Thai).Almost Thai noodles are made from rice;however, some kinds are made from wheat

flour (Ramen or Ba-mee) and starches derivedfrom mung bean (Mung bean vermicelli orWun Sen). Rice noodles are presented invarious size, shape and thickness such as Ricespaghettini (Sen lek), Rice fettuccine (Sen yai)and Rice vermicelli (Sen mee). The mainingredients are rice flour and water. Corn

Chiang Mai J. Sci. 2014; 41(2) 371

starch is also added in order to increase thegelatinous and chewy texture of the noodles[1]. All types of noodles are accessible fresh,semi-dried and dried. The moisture contentof fresh and semi-dried noodles leads tosensory deterioration during handling andstorage, producing loss of quality and limitingshelf-life [2].

Benzoic acid (C6H5COOH) and sorbicacid (C5H7COOH), organic acids, arepreservatives widely employed in food,pharmaceutical and cosmetic products(Figure 1). Benzoic acid and sorbic acidexhibit inhibitory activity against a widevariety of fungi, yeasts, molds and bacteria[3]. A broader spectrum of microbicidalactivity is achieved by using a combinationof them, which inhibit several bacterialstrains better than either of them alone [4].Under the notification of the Ministry ofPublic Health of Thailand ((No.281) B.E.2547(2004)) and announcement of the Thai Foodand Drug Administration (Thai FDA),the maximum level (ML) of benzoic acidand sorbic acid for noodles permitted bylaw is 1,000 parts per million (ppm) or1,000 mg/kg. According to international

standards adopted Codex AlimentariusCommission (CAC), the ML of benzoic acidfor noodles is 1000 mg/kg while sorbic acidis not allowed. The acceptable daily intake(ADI) values determined by the Joint FAO/WHO Expert Committee on Food Additives,for benzoic acid and sorbic acid are 0-5 and0-25 mg/kg body mass, respectively. A smallamount of benzoic acid is naturally excretedby the human body through the normalmetabolic processes as hippuric acid orbenzoyl glucuronide. Nevertheless, there is anevidence of health risks for consuming highamount of benzoic acid such as asthma,rhinitis, urticaria and anaphylactic shockin sensitive individuals. Cases of nausea,headache, weakness, burning and irritationof oesophagus have been reported as well[5]. Sorbic acid has a very low level ofmammalian toxicity. Few cases of non-immunological contact urticaria and pseudo-allergy have been reported [6,7]. A validatedquantification method for both acids issubsequently needed to facilitate consumerprotection and the production of good qualitynoodles [8].

Various methods are available for thedetermination of benzoic and sorbic acid infoods products including the use oflanthanide-sensitized luminescence [9],potentiometric [10], spectrophotometric [11],capillary zone electrophoresis [12,13],high performance thin layer chromatography(HPTLC) [14], micellar electrokineticchromatography (MECC) [15], gaschromatography (GC) [16-18] and highperformance liquid chromatography (HPLC)[19-26]. The official AOAC GC methodrequires derivatization reaction. Some GC andHPLC methods use harmful organic solventfor extraction that can cause environmentalpollution and human health risk. In addition,almost of these methods are applied forfoods, fruits, beverages and drinks which are

Figure 1. Structure of benzoic acid and sorbicacid.

372 Chiang Mai J. Sci. 2014; 41(2)

not specific to noodle matrices. Only oneprevious non-validated HPLC method isoperated for noodle samples [21]. However,the lengthy sample preparation by streamdistillation is used. Hence, the present workwas conducted in order to develop a simpleand rapid HPLC-DAD method for thedetermination of benzoic acid and sorbic acidin noodles simultaneously. A simple extractionprocedure was applied in order to extract bothpreservatives and clean up samples in singlestep. The established method was validatedwith respect to specificity, linearity, precision,accuracy and robustness.

2. MATERIALS AND METHODS2.1 Chemicals and Reagents

Standards of benzoic acids and sorbicacid with purity greater than 99% werepurchased from Sigma/Fluka (Schnelldorf,Germany). Ammonium acetate and acetic acidwas from Univar Ajax Finechem (Australia).The HPLC-grade methanol and acetonitrilewere purchased from Merck (Darmstadt,Germany). Other chemicals used were ofanalytical grade. High purity water wasprepared by using Milli-Q RO system(Millipore, Bedford, MA, USA).

2.2 Noodle SamplesFive kinds of noodles; Rice spaghettini,

Rice fettuccine, Rice vermicelli, Mung beanvermicelli and Ramen, were purchased from6 noodle shops around Silpakorn Universityin Amphur Muang, Nakorn prathomprovince, Thailand. A total of 150 sampleswere collected incessantly for 5 weeks andkept at 4°C in a refrigerator.

2.3 Instrumention and ChromatographicConditions

The HPLC-DAD system consisted of anAgilent 1100 series pump, an on-line solventdegasser, an autosampler, a diode-array

detector (DAD) and a Chemstation softwareversion A.08.01 (Agilent Technologies, USA).A reversed-phase column, 50 mm × 4.6 mmpacked with 5 μm, Germini-C18 modifiedsilica (Phenomenex, USA) and a guardcolumn, 20 mm × 3.9 mm packed with5 μm, C18 were used. The column wasoperated at ambient temperature. Theseparation was carried out under isocraticelution with 0.05 M ammonium acetatebuffer (pH 4.4 with acetic acid) and methanolin the ratio of 60:40 (% v/v). The flow ratewas 1 mL/min and the wavelength wasmonitored at 234 nm. The injection volumewas 10 μL. The mobile phase was filterthrough 0.45 mm Chrom Tech Nylon-66filter and degassed in an ultrasonic bathprior to use. The samples were analyzedusing a DAD detector in scan mode coveringthe range of 200-400 nm.

2.4 Preparation of Stock and StandardSolutions

Stock solutions of benzoic acid andsorbic acid in methanol were prepared at aconcentration of 2.5 and 1.25 mg/mL,respectively. A series of standard mixturesolutions was prepared by the appropriatedilution of the stock standard solutions withmethanol to concentration range of 5-200 μg/mL for benzoic acid and 1-50 μg/mL forsorbic acid.

2.5 Preparation of Analytical SampleNoodle samples were cut into small

pieces and dried in a hot air oven at 50°Cfor 30 min. The dried samples were thenhomogenized to fine powder. One gram ofpowdered sample was accurately weighedinto a 10 mL volumetric flask and added apart of extraction solvent; methanol: water(60:40, v/v). The mixture solution was shakenvigorously for 1 min by vortex mixing andmade up to volume with extraction solvent.

Chiang Mai J. Sci. 2014; 41(2) 373

The mixture solution was then placed ina sonicator bath for 30 min to affectcomplete extraction. The extracted solutionwas centrifuged for 5 min at 4,000 rpm.An aliquot supernatant was filtered through a0.45 μm nylon membrane filter and subjectedinto HPLC column. For concentratedsamples, further dilution with mobile phasewas performed.

3. RESULTS AND DISCUSSION3.1 Sample Preparation

Several types of extraction solvent; 100%methanol, 100% acetonitrile, methanol: water(60:40, (v/v)), acetonitrile: 20 mM sodiumacetate (50:50, (v/v)) and 1.0 N NaOH wereinvestigated. A blank noodle sample wasspiked with a standard mixture solutioncontaining 50 μg/mL of benzoic acidand 10 μg/mL of sorbic acid, mixed andprepared as previous above. The recoveryand standard deviation were calculated andcompared. The average recovery rate forbenzoic acid ranged from 66-78% using100% methanol, 100% acetonitrile, acetonitrile:20 mM sodium acetate (50:50, (v/v)) and1.0 N NaOH while for sorbic acid rangedfrom 89-105%. The mixture of methanol andwater (60:40, v/v) was found to be the mostsuitable extracting solvent in this study withthe average recovery between 96% and 103%.The result achieved indicated that there wereno matrix effects on benzoic acid and sorbicacid determination.

3.2 Optimization of the HPLC ConditionThe chromatographic conditions were

optimized so as to obtain a good separationbetween benzoic acid, sorbic acid and matrixinterferences in noodles. Acetate andphosphate aqueous solutions with differentpH and organic solvent (methanol andacetonitrile) were investigated and systemsuitability test (SST) parameters were

performed during the development andoptimization of method. Firstly, the separationwas attempted and compared on a C18column (150 mm × 4.6 mm, 5 μm) and aC18 column (50 mm × 4.6 mm, 5 μm) with awavelength at 224 nm (for sorbic acid),234 nm (for benzoic acid) and 254 nm usingacetonitrile and 5 mM phosphate buffer(pH 7) in the ratio range from 5:95 - 30:70.Both acids were eluted quickly and could notseparate from matrix interferences. Methanolwas then replaced in the same ratio range butthe peaks of both acids were co-eluted.Increasing of phosphate buffer concentration(5-30 mM) led to increase in analytes’ retention,decrease in tailing factor and resolution value.Consequently, ammonium acetate aqueoussolution was employed instead of phosphateto achieve the best resolution. Poor resolutionand peak tailing was obtained when acetonitrilewas used. Separation of benzoic acid andsorbic acid was accomplished by varying theconcentration of ammonium acetate (5-30mM) and the ratio of methanol. An increasein retention and peak broadening of benzoicacid and sorbic acid was observed when higherconcentration of ammonium acetate at higherpH was utilized. Hence, the aqueouscomponent of the mobile phase was selectedto be ammonium acetate 5 mM, pH 4.4.Methanol in varying ratio was studied and theresolution (R) between two acids andinterferences, and the run time of thechromatogram was taken into consideration.The optimization condition was performedusing an isocratic elution at the proportion of60:40 (v/v) 5 mM ammonium acetate aqueoussolution adjusted to pH 4.4: methanol at aflow rate of 1 mL/min. The C18 columnwith 5 cm length was preferred as thecomplete separation was occurred within 3min and the mobile phase operated was also3 times decreased.

Under this condition, benzoic acid was

374 Chiang Mai J. Sci. 2014; 41(2)

well separated from sorbic acid with goodpeak shape (resolution 3.62). Tailing factor was0.96 and 1.00 and the number of thetheoretical plates (efficiency) was 3029 and3599 for benzoic acid and sorbic acid,respectively. The typical retention time ofbenzoic acid and sorbic acid was 1.82 and2.32 min, respectively (Figure 2). Moreover,

peaks of matrix interference in noodles wereobserved to separate from the peak ofbenzoic acid and sorbic acid (resolution > 2)(Figure 2 (B) and (D)). The overlaid diode-array spectrum showed good UV absorbanceat 234 nm for benzoic acid and sorbic acid,and then this wavelength has been chosen fordetection.

Figure 2. Typical HPLC chromatograms of (A) a standard mixture solution of benzoic acid100 μg/mL and sorbic acid 20 μg/mL (B) a blank noodle sample (C) a blank noodle samplespiked with a standard mixture solution of benzoic acid 100 μg/mL and sorbic acid 20 μg/mL and (D) a noodle sample containing benzoic acid.

Chiang Mai J. Sci. 2014; 41(2) 375

3.4 SpecificityThe specificity of the method was

established through the study of resolutionfactor of benzoic acid, sorbic acid and peakfrom the nearest resolving peak. Benzoicacid and sorbic acid were completelyseparated from all interference matrixes withthe satisfactory resolution greater than 2.

Peaks were identified with retention timescompared with standards and confirmed thecharacteristic spectra by photodiode arraydetection (range 200-400 nm) in both sampleand standard solutions. DAD spectral analysiswas employed to verify the homogeneity ofthe benzoic acid and sorbic acid peaks in allsolutions. The peak purity value was greaterthan the threshold value of 995 and thusestablishing the selectivity of the assay methodfor noodle samples.

3.5 Linearity and RangeLinearity was established by least squares

linear regression analysis of the calibrationcurve and evaluated by a determinationcoefficient (r2) using data analysis MS excelsoftware (2007) (ANOVA; P < 0.05). Standardmixture solutions of benzoic acid and sorbicacid at seven different concentrations (ranging

from 5 to 200 μg/mL and 1 to 50 μg/mL,respectively) were prepared and analyzed intriplicate to prove the linearity of system.Peak area of benzoic acid and sorbic acidsolution was plotted against their respectiveconcentrations. The calibration curve ofbenzoic acid and sorbic acid showed goodlinearity over the concentration range with r2

0.9998 and 0.9998, respectively. The statisticaldata of the regression equation is presentedin Table 2.

3.6 Determination of Limit of Detectionand Quantitation (LOD and LOQ)

Limit of detection (LOD) and limit ofquantitation (LOQ) of benzoic acid and sorbicacid were calculated from the residualstandard deviation of the regression line (d)of the calibration curve of each compoundstaged above and its slope (S) in accordanceto the following equation: LOD = 3.3(d/S)and LOQ = 10(d/S). The results given inTable 2 suggested that this method couldbe used for the determination of very smallconcentrations of benzoic acid and sorbicacid in noodle samples.

Table 1. System suitability test data.

Parameter Benzoic acid Sorbic acid Preferable levelCapacity factor (k′)Selectivity factor (α)Resolution (Rs)Theoretical plate (N)Tailing factor (T)

1.70

2.4230290.96

2.681.583.6235991.00

1.0 to 10.01.0 to 2.0

> 2> 2,500

1

3.3 System SuitabilityThe system suitability test (SST)

parameters were performed during thedevelopment and optimization of the methodas well as through the validation procedure.SST parameters including capacity factor (k′),selectivity (a), resolution (Rs), column efficiency(number of theoretical plates, N) and tailing

factor (T) listed in Table 1 were establishedby ten replicates. All parameters weresatisfactory with good specificity for thequantitation of benzoic acid and sorbic acidin noodles. The RSD% values for thecalculated SST parameters for 10 replicateswere less than 2%.

376 Chiang Mai J. Sci. 2014; 41(2)

Table 2. Linearity, statistical data of the regression analysis, LOD and LOQ.Parameter Benzoic acid Sorbic acid

Calibration range (μg/mL)Limit of detection (μg/mL)Limit of quantitation (μg/mL)Regression equation Slope Standard error of mean (S.E.) of slope 95% confidence limit of slope Intercept Standard error of mean (S.E.) of intercept 95% confidence limit of intercept Standard error of estimate Regression coefficient (r2)

5-2000.421.14

32.21900.198231.7096-32.728437.469920.5838-15.4424-90.382236.88000.999811

1-500.320.99

55.23380.327554.3918-56.07575.38787.4843-13.8513-24.626814.01470.999824

3.7 AccuracyThe accuracy of the method was

performed on blank noodle samples spikedwith known quantities of preservatives.Five known amount of benzoic acid andsorbic acid were added in blank samples,mixed and all samples were treated andanalyzed as described above. Three sampleswere prepared at each concentration. Thepercentage recovery of added preservativeswas calculated by comparing the peak area

of the test samples with that of the standardsolutions. The results obtained from 15samples are summarized in Table 3. Themean recoveries ranged from 98.92-102.04%for benzoic acid and 97.66-102.47% forsorbic acid excepting at the lowest spiked levelof each compound that was 85.61 and83.62%, respectively. The RSD value of eachspiked concentration was < 1% and all RSDvalues (n=15) were < 8%.

Table 3. Recovery analysis of benzoic acid and benzoic acid.Actual

concentration(μg/mL)

Calculatedconcentration

(μg/mL)

Recovery(%)

Meanrecovery

(n=3)

RSD(%)

Benzoic acid

5.00

10.00

20.00

50.00

100.0

4.304.284.279.909.7810.0019.7620.2220.0051.0550.5750.07101.79102.15102.17Total

85.9385.5585.3898.9297.77100.0098.78101.10100.00102.09101.13100.15101.79102.15102.17

(n = 15)

85.61

98.92

99.96

101.12

102.04

97.53

0.28

0.92

0.95

0.78

0.17

6.25

Chiang Mai J. Sci. 2014; 41(2) 377

Table 3. continue.

Method accuracy was also demonstratedby plotting the amount of benzoic acid andsorbic acid found against the amount added.Method accuracy of benzoic acid (y = 1.026x-0.615) and sorbic acid (y= 1.032x-0.172)showed good linearity with r2 0.9990 and0.9999, respectively. The result is found to besatisfactory for intended purpose and isadequate for routine analysis.

3.8 PrecisionPrecision was evaluated in term of system

precision (repeatability), method precision(intra-day) and intermediate precision (inter-day).3.8.1 System precision

System precision or repeatability wasdetermined by performing ten consecutiveinjections (n=10) of a sample solution and astandard mixture solution containing 100 μg/mL of benzoic acid and 20 μg/mL of sorbicacid. The RSD% of peak area response andretention time was calculated and showed thesatisfactory precision of the system (< 1 %)(Table 4).

Actualconcentration

(μg/mL)

Calculatedconcentration

(μg/mL)

Recovery(%)

Meanrecovery

(n=3)

RSD(%)

Sorbic acid

1.02

2.05

5.12

10.24

20.48

0.860.860.852.001.982.025.035.145.1010.4510.3810.3020.9321.0321.00Total

84.2583.7682.8597.7396.5398.7298.31100.4699.61102.07101.32100.62102.18102.68102.55

(n = 15)

83.62

97.66

99.46

101.34

102.47

96.91

0.69

0.92

0.89

0.59

0.21

7.10

3.8.2 Method precisionThe precision of method was assessed

by determining the spiked samples. Blanknoodles were spiked with four known levelsof each compound (10, 20, 50 and 100 μg/mL for benzoic acid and 2.05, 5.12, 10.25,and 20.48 μg/mL for sorbic acid). Threereplicate of each level were prepared andanalyzed within the day (intra-day). Themethod was found to be precise with %RSDvalues within 0.85-1.05% for benzoic acid and0.43-1.55% for sorbic acid. The data is shownin Table 4.

3.8.3 Intermediate precisionThe RSDs for intermediate precision was

evaluated by analysis of benzoic acid andsorbic acid at the same concentration as it wasperformed for method precision, repeatedfor two different days by two analysts (inter-day). The %RSD values were 1.57-1.84%(overall = 2.40%) for benzoic acid and 1.42-1.62 (overall = 2.80%) for sorbic acid.

The precision was also investigated byanalysis of selected noodle samples for 9

378 Chiang Mai J. Sci. 2014; 41(2)

Table 4. System precision, method precision and intermediate precision data.

determinations on the same day (intra-day)and on three consecutive days (inter-day). Onlybenzoic acid was detected with %RSD valuesof 2.00 and 3.61% for intra-day and inter-

day assay, respectively. All results of methodand intermediate precision confirmedexcellent precision of the proposed HPLC-DAD method (Table 4).

Benzoic acid(mean + SD) %RSD

System Precision(n= 10)Peak AreaRetention time

Sorbic acid(mean + SD) %RSD

3317.43 + 2.141.80 + 0.01

0.070.28

1113.01 + 4.722.32 + 0.01

0.420.38

Spiked sample analysis Benzoic acidSpiked level(μg/mL)

%RSDDay 1Analyst 1

Sorbic acidSpiked level(μg/mL)

Day 2Analyst 2

%RSDDay 1Analyst 1

Day 2Analyst 2

102050100(n = 12)(n = 24)

1.050.971.020.851.84

1.131.441.561.051.572.40

2.055.1210.2420.48

0.720.740.431.551.41

1.131.090.670.251.622.80

Method precisionIntermediate precision

Noodle analysis Benzoic acidMeasuredconcentration(μg/mL) %RSD

Sorbic acidMeasuredconcentration(μg/mL) %RSD

385.08+8.09

400.77+8.59400.00+6.80394.83+26.05398.53+14.41

2.00

2.141.706.603.61

nd

ndndndnd

Analyst 1 Day 1 (n=9)

Analyst 1 Day 2 (n= 3)Analyst 2 Day 3 (n= 3)Analyst 3 Day 4 (n= 3)Total (n= 9)

-

----

3.9 RobustnessThe robustness is a measure of its

capacity to remain unaffected by small, butdeliberate variations in method parametersand provides an indication of its robustnessduring normal usage. The robustness wastested during the development phase by aslightly vary of volume of injection (+ 5 μL),wavelength (+ 2 nm), flow rate (+ 0.1 mL/min), the percent of organic modifier (+2% absolute) and pH of the mobile phase(+ 0.2 pH unit). The robustness was

performed on a standard mixture solutionof benzoic acid and sorbic acid and theresult was evaluated by calculating the SST(system suitability test) values. The SSTparameters kept on unaffected overdeliberate changes in the chromatographicconditions (Table 5), illustrating that themethod was robust. Moreover, the noodlesamples were considered under differentparameters and it was observed that therewas no change in the area under the peak,confirming the robustness of the method.

Chiang Mai J. Sci. 2014; 41(2) 379

Table 5. Robustness data of benzoic acid (BA) and sorbic acid (SA).

Conditions

Injectionvolume (μL)

Wavelength(nm)

Flow rate(mL min-1)

Percent ofmethanol

Mobilephase pH

Range

510a

15

232234a

236

0.91.0a

1.1

3840a

42

4.24.4a

4.6

T

BA SA

0.960.960.96

0.960.960.96

0.960.960.97

0.980.960.96

0.980.960.95

1.001.001.00

1.001.001.00

1.001.001.01

1.021.000.99

1.011.001.00

N

BA SA

Rs

BA SA

Measured concentration(μg/mL) (n=3)BA SA

329630292706

302930293029

336130293523

309030292901

324030292898

369535993124

359935993599

371435994004

363635993714

386135993457

6.806.776.70

6.776.776.77

6.966.776.68

6.856.776.63

7.186.776.59

3.653.623.60

3.623.623.62

3.733.623.55

3.843.623.39

3.053.623.67

381.12 ± 0.11380.22 ± 0.34381.26 ± 0.43

380.08 ± 0.32380.22 ± 0.34380.12 ± 0.43

380.77 ± 0.38380.22 ± 0.34379.02 ± 0.20

380.31 ± 0.17380.22 ± 0.34380.22 ± 0.17

383.67 ± 1.12380.22 ± 0.17384.31 ± 1.28

ndndnd

ndnd

ndndnd

ndndnd

ndndnd

a HPLC method conditions.T = tailing factor, N = theoretical plates and R = resolution.nd = not detected.

3.10 Stability of Standard SolutionsThe stability of benzoic acid and sorbic

acid standard was assessed by analyzing thesolutions at 0, 1, 2, 4, 6, 8, 12 and then every 4h until 48 h after preparation. The %RSDvalue of the assay of benzoic acid and sorbicacid was 0.07% and 0.42%, respectively (<1%). The chromatogram showed no peakcorresponding to the degradation productsand there was no significant change in the peakarea response of benzoic acid and sorbic acid.The results reveal that standard solution isstable in methanol for at least 48 h.

3.11 Analysis of Noodle SamplesFive kinds of noodles were collected

over a period of five weeks from 6 noodle

shops and analyzed by the proposed validatedmethod. No concentration level ofpreservative was declared on their label. Theaverage level of benzoic acid and sorbic acidfound in noodle samples is presented in Table6. The presence and purity of benzoic acidand sorbic acid was confirmed by DADspectra. The benzoic acid content wassimultaneously detected in all Rice spaghettiniand Rice fettuccine samples throughout thefive weeks in a wide range of concentration.Over 90 % of both rice noodles containedbenzoic acid exceeded the approvedregulation limits (<1,000 mg/kg). Higherquantity of benzoic acid added was probablyto extent their shelf life of these products witha higher demand. Nevertheless, benzoic acid

380 Chiang Mai J. Sci. 2014; 41(2)

found in Rice vermicelli, Mung bean vermicelliand Ramen was within the allowable regulationlimits. Amounts of benzoic acid added variedamongst the kinds of noodle and with time.Wide fluctuation can be attributed due tothe lack of quality control and inconsistentin benzoic acid usage. Sorbic acid wasdetected only in 6 samples of Ramen noodle

(from totally 150 analyzed samples) in thelow concentration ranging from 2.68 to5.79 mg/kg. Moreover, the proposedmethod was applied for the analysis ofbenzoic acid and sorbic acid in 2 differentnoodle samples produced for exportingwhich were labeled as “no preservativesadded”. No preservative was detected.

Table 6. Quantitative analysis of benzoic acid (BA) and sorbic acid (SA) in 5 types ofnoodles (mg/kg of noodle sample).

Noodle Type

Rice spaghettini

Rice fettuccine

Rice vermicelli

Mung bean vermicelli

Ramen

Nt

6

6

6

6

6

Week 1

BA SA

2082.32

2669.21

2.96

5.52

136.03

nd

nd

nd

nd

nd

Week 2

BA SA

2541.65

2137.78

3.10

7.77

40.20

nd

nd

nd

nd

nd

Week 3

BA SA

2534.80

4525.96

1.73

3.79

31.97

nd

nd

nd

nd

nd

Week 4

BA SA

2958.16

2763.54

0.87

5.51

47.60

nd

nd

nd

nd

nd

Week 5

BA SA

3023.24

2480.27

2.23

2.24

37.95

nd

nd

nd

nd

nd

Range found

BA SA

713.41-4607.41

192.04-10499.00

nd-42.04

nd-25.17

nd-487.29

nd-2.28

-

-

-

nd-5.79

Average (n=30)

BA SA

2414.09

3175.46

4.44

5.62

58.75

0.08

nd

nd

nd

0.83

Nt = number of analyzed samples per week. Each sample was determined in triplicate.

nd = not detected.

4. CONCLUSIONA simple isocratic HPLC-DAD

method was successfully developed for thedetermination of benzoic acid and sorbic acidin noodles. The sample preparation engagedthe extraction of sample with methanol: water(60:40, v/v) that was economic and easy toaccomplish without using harmful organicsolvent or time-consuming and expensivesolid-phase extraction as previous report[19,22,24]. The complete separation of theanalytes were achieved in shorterchromatographic run time (3 min) incomparison to all published literature [19-24]and found to be more specific for noodlesample. All validation parameters were withinthe acceptance range. This proposed methodwas applied for the analysis of benzoic acidand sorbic acid in the different kinds ofnoodles and shown satisfying results in termsof accuracy, repeatability and easiness.Additionally, reducing run time caused less inusing reagents. The advantage of utilizingfewer reagents could also bring thehealthier life and cleaner environment.

ACKNOWLEDGEMENTSThe authors gratefully acknowledge

the financial support of the Faculty ofpharmacy, Silpakorn University, Thailand.

REFERENCES

[1] Fu B.X., Asian noodles: History,classification, raw materials, and processing,Food Anal. Int., 2008; 41: 888-902.

[2] Rachtanapun P. and TangnonthaphatT., Effects of packaging types andstorage temperatures on the shelf lifeof fresh rice noodles under vacuumconditions, Chiang Mai J. Sci., 2011;38: 579-589.

[3] Colmenero F.J. and Solana J.B.,Additives: Preservatives; in NolletL.M.L and Toldra F., eds., Handbook ofProcessed Meats and Poultry Analysis,CRC Press, Florida, 2008; 91-107.

[4] Deshpande S.S., Deshpande U.S. andSalunkhe D.K., Food Acidulants; inMaga J.A. and Tu A.T., eds., Food AdditiveToxicology, Marcel Dekker, New York,1995; 53.

Chiang Mai J. Sci. 2014; 41(2) 381

[5] WHO 2000. Benzoic acid and sodiumbenzoate. Concise internationalchemical assessment documents(CICADs), 26, International programmeon chemical safety.

[6] Walker R., Toxicology of sorbic acidand sorbates, Food Addit. Contam., 1990;7: 671-676.

[7] Sofos J.N., Sorbic Acid; in Naidu A.S., ed., Natural Food AntimicrobialSystem, CRC Press, Florida, 2000:637-659.

[8] Joint FAO/WHO expert committeeon Food Additives (JECFA) 2004.General Standard for Food Additives,Codex Standard 192-1995, (Revised 5-2004), 2004; 79.

[9] Aguilar-Caballos M. P., Gomez-Hens A. and Perez-Bendito D.,Simultaneous determination ofbenzoic acid and saccharin in softdrinks by using lanthanide-sensitizedluminescence, Analyst, 1999; 124: 1079-1084.

[10] Pezza L., Santini A.O., Pezza H.R.,Melios C.B., Ferreira V.J.F. andNasser A.L.M., Benzoate iondetermination in beverages by using apotentiometric sensor immobilized ina graphic matrix, Analyst, 2001; 433:281-288.

[11] Hamano T., Mitsuhashi Y., Aoki N.,Semma M. and Ito Y., Enzymaticmethod for the spectrometricdetermination of benzoic acid in soysauce and pickles, Analyst, 1997; 122:256-262.

[12] Han F., He Y.Z., Li L., Fu G.N., Xie H.Y.and Gan W.E., Determination of benzoicacid and sorbic acid in food productsusing electrokinetic flow analysis-ion pairsolid phase extraction-capillary zoneelectrophoresis, Anal. Chim Acta.,2008; 618: 79-85.

[13] Tang Y. and Wu M., A quick methodfor the simultaneous determinationof ascorbic acid and sorbic acid in fruitjuices by capillary zone electrophoresis,Talanta, 2005; 65: 794-798.

[14] Khan S.H., Murawski M.P. andSherma J., Quantitative high-performance thin-layer chromato-graphic determination of organic-acidpreservatives in beverages, J. Liq.Chromatogr., 1994; 17: 855-865.

[15] Boyce M.C., Simultaneous deter-mination of antioxidants, preservativesand sweeteners permitted as additivesin food by micellar electrokineticchromatography, J. Chromatogr. A.,1999; 847: 369-375.

[16] AOAC 2005. Benzoic Acid and SorbicAcid in Food, Gas-ChromatographicMethod (Method 983.16), in HorowitzW., ed., Official Methods of Analysis ofAOAC International, 18th Ed.,Association of Official AnalyticalChemists International, Gaithersburg,MD, 2005: 9.

[17] Dong, C. and Wang, W., Headspacesolid-phase microextraction applied tothe simultaneous determination ofsorbic and benzoic acids in beverages,Anal. Chim. Acta., 2006; 562: 23-29.

[18] Kokya, T.A., Farhadi, K. and Kalhori,A., Optimized dispersive liquid-liquidmicroextraction and determinationof sorbic acid and benzoic acid inbeverage samples by gas chromato-graphy, Food Anal. Meth., 2011; 4: 150-154.

[19] AOAC 2000. Benzoic Acid in OrangeJuice, Liquid ChromatographicMethod (Method 994.11), in HorowitzW., ed., Official Methods of Analysis ofAOAC International, 17th Ed.,Association of Official AnalyticalChemists International, Gaithersburg,MD, 2000: 10.

382 Chiang Mai J. Sci. 2014; 41(2)

[20] Ferreira I.J.M.P.L.V.O., Mendes E.,Brito P. and Ferrcia M.A.,Simultaneous determination ofbenzoic and sorbic acid in quince jamby HPLC, Food Res. Int., 2000; 33:113-117.

[21] Inchai S. and Sitthiopagul S., Benzoicacid and sorbic acid in noodles inNakhon Ratchasima municipality, J.Health Sci., 2008; 17: 744-750.

[22] Kamlert W., Quantitative determina-tion of benzoic acid, sorbic acid andsaccharin in preserved fruits, Bull.Dep. Med. Sci., 1992; 34: 31-36.

[23] Saad B., Bari Md. M., Saleh I., AhmadK. and Talib M.K.M., Simultaneousdetermination of preservatives (benzoicacid, sorbic acid, methylparaben andpropylparaben) in foodstuffs using

high-performance liquid chromato-graphy, J. Chromatogr. A., 2005; 1073:393-397.

[24] Techakriengkrai I and Surakarnkul,R., Analysis of benzoic acid and sorbicacid in Thai rice wines and distillatesby solid-phase sorbent extraction andhigh-performance liquid chromato-graphy, J. Food Compos. Anal., 2007;20: 220-225.

[25] Tfouni S.A.V. and Toledo M.C.F.,Determination of benzoic and sorbicacids in Brazilian food, Food Control,2002; 13: 117-123.

[26] Willetts P., Anderson S., Brereton P.and Wood R., Determinationof preservatives in foodstuffs:Collaborative trial, J. Assoc. PublicAnalysts, 1996; 32: 109-175.