A Research Note Determination of Organic Acids in Dairy Products by High Performance Liquid Chromatography A.E. BEVILACQUA, and A.N. CALIFANO ABSTRACT Calibration and calculations A high performance liquid chromatographic method was developed for the quantitative analysis of organic acids in dairy products. A reverse-phase C8 column at room temperature, a mobile phase of 0.5% w/v buffer ((NH,)zHPO, at pH 2.24 with H,PO,)-0.4% v/v acetonitrile, UV detection at 214 nm and 1.2 mUmin flow rate were utilized. Formic, acetic, pyruvic, propionic, uric, erotic, citric, lactic and butyric acids were quantitated for raw milk, yogurt, Blue, Pro- volone, Port Salut and Quartirolo cheeses. Recoveries greater than ‘85.3% were observed for all acids. Five aqueous calibration standards covering a broad concentration range were prepared for each of the organic acids analyzed (formic, acetic, pyruvic, propionic, uric, erotic, citric, lactic, and butyric acids). At lcast triplicate 10 uL injections were made. The resulting peak area counts were determined; then, the best fit standard cutvc (peak area vs concentration) was prepared for each of the organic acids using linear regression. Quantitation was based on the external standard method. Retention times were first determined separately for each acid. Af- terwards, individual standard solutions were added one by one to each sample to verify the correct identification of peaks. INTRODUCTION ORGANIC ACIDS information is important in understanding metabolism and quality of milk products. The acids appear as a result of. the hydrolysis of fatty acids, normal bovine meta- bolic processes, bacterial growth or direct addition as acidu- lants., They are believed to contribute to the flavor of most aged cheeses (Adda et al., 1982). Quantitative determination of these acids in dairy products is important to flavor studies, for nutritional reasons and as an indicator of bacterial activity. Sample preparation Numerous analytical methods including calorimetric analy- sis (Marshall and Harmon, 1978), gas chromatography (Gray, 1976) and various wet analytical techniques (AOAC, 1975; Davidson, 1948; Patton, 1953) have been employed to analyze organic acids in dairy products. Marsili et al. (1981) and La- vanchy and Steiger (1984) reported HPLC procedures for or- ganic acids in dairy products. Raw milk, freshly prepared yogurt, Blue, Port Salut, Quartirolo and two different brands of Provolone checscs were analyzed. The same procedure of sample preparation was followed for each dairy product (except milk): 7.OOg of samples were added to 50.0 mL of buffer-acctonitrile mohilc phase, homogenized, extracted for 1 hr, and ccntrifugcd at 7000 x g for 5 min. The supcrnatant was filtcrcd once through filter paper and twice through a 0.45 urn mcmhranc filter (Sartorius SM 11606); IO )LL were injected with a 25 uL Hamilton syringe (Hamilton Co., Rcno, NV). The same procedure of sample preparation was used for milk, except that IO mL were added to 40 mL of mobile phase. Duplicate analyses wcrc pcrformcd on all sam- ples. Recovery study The objective of this study was to develop an isocratic HPLC procedure for the quantitation of individual organic acids in dairy products. The efficiency of the extraction procedure was cvaluatcd by dctcr- mination of recovery of the organic acids from a spiked Blue cheese preparation. A known amount of standard solution of organic acids was added to 7.OOg of Blue cheese and extracted as dcscrihcd above. Duplicate injections were made for each sample of Blue chccsc and Blue checsc plus added acids. The amounts added wcrc roughly SO% of the actual concentration of the sample. MATERL4LS & METHODS Equipment and operating conditions A Waters Liquid Chromatograph (Waters Associates Inc., Milford, MA) was equiped with a Model U6K injector fitted with a 20 IJ.L sample loop, a Model 6000A solvent delivery system, a Model 450 variable wavelength detector and a Data Module M730. The detector was set at 214 nm or 285 nm. Operating conditions were: mobile phase was aqueous 0.5% w/v (NH4)?HP04-0.4% v/v acetonitrilc (at pH 2.24 with HXP04); flow rate 1.2 mL/min; ambient column temperature and chart speed I cm/ min. A reverse phase Beckman C8 (250 x 4.6 mm), ultrasphere- octyl column with a 5 p,m particle diamctcr was used. Mobile phase was prepared by dissolving analytical grade (NH4)?HP04 in distilled water, HPLC grade acetonitrile and HSPO,. HPLC grade reagents were used as standards (Sigma Chemical Co., St. Louis, MO). Solvents were degassed under vacuum. Both solvents and standard solutions were filtered through 0.2 pm and 0.45 km membrane filters (Sartorius SM 11607, SM I1606), rcspcctively. The authors are with Centro de lnvestigacion y Desarrollo en Criotecnologia de Alimentos (CIDCA), Fat. Ciencias Exactas, UNLP. 47 y 116, La Plats, 1900, Argentina, and members of the Career of Scientific Research of the Conseio National de Inves- tigaciones Cientificas y Tknicas de la Repbblica Argentina. RESULTS & DISCUSSION DIFFERENT OPERATING CONDITIONS were employed (mobile phase composition, pH, temperature, flow rate) to maximize the resolution of the peaks. Optimum wavelength for acid detection was chosen after obtaining the individual absorbancespectra on a Shimadzu double beam UV 150 digital spectrophotometer. The selected wavelength at 214 nm agrees with the one proposed by Picha (1985). Orotic and uric acids for Blue cheese were determined at 285 nm where lactic acid did not interfere. A typical chromatogram depicting the sepa- ration of an aqueousstandard solution of organic acids is shown in Fig. la. The relative average deviations of replicates was 3.14%. The correlation coefficients between area counts and concentration indicated excellent linear responses over wide concentration ranges of acids (Table 1). Organic acid composition of raw milk, freshly prepared yo- gurt, commercial Provolone, Port Salut, Quartirolo and Blue cheesesamples is presented in Table 1 as well as the Standard errors of the means. The detection limit values for the acids are also shown in Table 1. A typical chromatogram for yogurt is illustrated in Fig. lb. -Text continued on page 1079 107%JOURNAL OF FOOD SCIENCE-Volume 54, No. 4, 1989