PATRICK AND HILL: CHILLED JUICE MICROBIOLOGY 293 MICROBIOLOGICAL AND STORAGE STUDY OF CHILLED ORANGE JUICE1 Roger Patrick Florida Citrus Experiment Station and E. C. Hill Florida Citrus Commission Lake Alfred Chilled orange juice for national distribution has been successfully processed and packed in commercial Florida plants since about 195. United States standards for grades of chilled orange juice are available (1). Production figures were first available (2) during the 1954 55 citrus season, when slightly over 3 million boxes of Florida oranges were used in the pro duction of chilled orange juice. Production has been continuous since 1954 but the largest quan tity of oranges used for chilled juice was slightly under million boxes during the 12 sea son. A nontechnical review of problems related to the production and distribution of chilled orange juice was presented by Hamrick (3). He discussed the scope and importance, history, meth ods of preservation and production, distribution, the problem of an adequate supply of juice for summer distribution, adequate standards, con tainers and storage. The number of technical papers on chilled orange juice is limited (5,, 8). However, many papers have been published relative to various factors that pertain to prob lems that may arise during the production, stor age and distribution of this product. Wenzel et al. (11) reviewed briefly 32 of these publications, which were concerned primarily with the solution of problems relative to the frozen orange concen trate industry. Also available is a bulletin by Patrick and Hill (6) on the microbiology of citrus fruit processing, including chilled citrus juices. The purpose of this paper is to present re sults obtained from a storage study of chilled orange juices which were either unstabilized or stabilized by heating at different temperatures. Some of these products were prepared by recon stituting either 42 or 59 Brix frozen orange Florida Agricultural Experiment Stations Journal Series No. 2 lcooperative research by the Florida Citrus Experiment Station and Florida Citrus Commission. concentrate. Microbiological data were obtained and the retention of ascorbic acid was determined during the storage of the chilled orange juices at different temperatures. Procedures and Results Microbiological and analytical methods. Total plate counts were made using Difco orange serum agar, ph 5.5, with incubation at 3 C. for 48 hr. Diacetyl values, indicative of the amount of diacetyl and/or acetylmethylcarbinol resulting from the growth of microorganisms in orange juice, were determined as described by Hill and Wenzel (4). Ascorbic acid was determined by the dye titration method (1). Unstabilized chilled orange juice. Chilled or ange juices were prepared and packed in the pilot plant at the University of Florida, Citrus Experiment Station. Valencia orange juice was chilled immediately after extracting and finishing, using commercial equipment, and packed in quart waxed cartons. This juice was not stabilized by heat treatment. The cartoned chilled juice at approximately 4 F. was stored either at 3, 4, 5 or 6 F. Initial plate counts on orange serum agar for this juice are shown in Table 1, as are counts after storage of this product at the different temperatures for various periods of. The average for the initial counts was 64, colonies per ml., the diacetyl value 3.1 and the ph 3.9. After two days at 6 F., the juice had a count of 2,86, per ml. and a diacetyl value of 4.9 ppm. Microbial growth continued and after 5 days the count increased to 12,, per ml. which caused a very high diacetyl value of 3. ppm. Rapid spoilage occurred in the juice stored at 5 F.; after seven days it had a count of 28,, per ml. and a diacetyl value of.8 ppm. Chilled juice stored at 4 F. for 26 days showed no increase in diacetyl value. The microbial population in creased to 54, per ml. but offflavor, charac teristic of microbiological spoilage, was not de tectable. of juice for 26 days at 3 F. caused no increase in diacetyl value and the count decreased from 54,5 to 19,8 colonies per ml. Stabilized chilled orange juices stored at 4 F. Three packs of heatstabilized and one pack of
294 FLORIDA STATE HORTICULTURAL SOCIETY, 14 Table 1. Microbiological plate counts of unstabilized chilled Valencia orange juice stored at different temperatures temperature F. 2 5 Plate counts per ml. of days juice1 X 13 26 3 5 6 695 58. 3. 58.5 36. 5.3 2,86 52.5 31. ij.,6 12,^ 1*5..9 28,^ Il3. 12. 18. 28 1.8 5U. Plated on orange serum agar, ph $. Discontinued because of spoilage. unheated chilled juices were prepared from the same Valencia orange juice, having a ph of 3.8. After extracting and finishing, the juices were heated at either 155, 15 or 195 F. A plate type heat exchanger was used for rapidly heating the juice to the stabilization temperature and cooling to approximately 8 F. The re quired in the plate heat exchanger to increase the temperature of the juice from room tempera ture to the stabilization temperature was 14 seeonce. The juice was held at the stabilization temperature for 13 seconds and then cooled to ap proximately 8 F. in 18 seconds. It was then chilled to approximately 4 F. and packed in both 8 and 32 oz. waxed cartons which were stored only at 4 F. The initial count of the juice not stabilized by heating was 6, per ml. and this increased in 24 days to 258, per ml. (Table 2). The three packs of juices, stabilized at 155, 15 and 195 F., never contained more than 5 or ganisms per ml. and after 24 days at 4 F. no increase in the microbial population was evident. The diacetyl value of the unheated juice was initially 3.6 ppm and there was no significant change in this value in either the unstabilized juice or in any of the stabilized products through out the storage period. No offflavors developed during storage that could be attributed to micro organisms. Chilled juices prepared from frozen orange concentrates. Chilled juices were also prepared from frozen orange concentrates. Both 42 and 59 Brix concentrates were used. They had been previously made from the same lot of Valencia oranges and stored at 8 F. for two months. When these concentrates were processed the evaporatorfeed juice was heated at 15 F. The 42 Brix concentrate contained unheated cutback juice but coldpressed orange oil was not added. The high density 59 Brix product contained added coldpressed orange oil but no cutback juice. Sufficient oil was added so that the reconstituted juice (11.8 Brix) would contain about 15 recoverable oil. These concentrates were mixed with chilled deionized water and the reconstitu ted juices were immediately packed in quart waxed cartons. The temperature of the juice in the cartons was approximately 4 F. when they were stored at either 4, 5 or 6 F. Table 2. Microbiological plate counts of stabilized chilled Valencia orange juices stored at l Stabilization temperature F. 155 15 195 6. h Plate 6k.$ 1 tf.5.1 Plated on orange serum agar, ph 55 counts per ml..5.3 1 255 of o'l days m.iuice1 X 13 U. lj. 2h 258..1.1
PATRICK AND HILL: CHILLED JUICE MICROBIOLOGY 295 Table 3 Microbiological plate counts of chilled Valencia orange juices prepared from 1*2 and 59 Brix stabilized frozen concentrated orange juices days temp. ~ "~" F* 1 3 h 8 1 15 1 22 p Plate counts per ml. reconstituted 1+2 Brix frozen concentrated orange juice X 1^ i 1*.3 2.3 1.9 33 68. 15 5 68. 6,23 6 22. 8,5 32,5OOJ 2 Plate counts per ml. reconstituted 59 o Brix frozen concentrated orange juice X 1 3 2.1 1.9 5 3 6 11. 28,5 3 13,.9. q 19,J.8.5. 2. 1.3.6 * Juice stabilized at 15 F. prior to concentration. Plated on orange serum agar, ph 55 3 Discontinued because of spoilage. Table 1*. of ascorbic acid in chilled Valencia orange juices stored at 1^ F. in 8 and 32 oz. waxed cartons1 days 3 1 Hi 1 2k ks 1*5 36 35 31* 91* 8 6 1* Stabilization temperature and ascorbic acid content 155 F. 1*2 3 36 31* 31 Juices stored in 8 oz. waxed cartons 91 8 8 1* 6 15 F, 3 3k 33 3 8 1* 2 65 195 F. mg/loqml k3 3$ 33 32 3 6 2 65 3 1 m 1 2l* Ii6 1*2 91. Juices 1*5 stored 91* in 32 oz. waxed k$ k3 cartons ' 1 6 * 31 8 1 Data obtained by R. L. Huggart and R.. Barron.
2 FLORIDA STATE HORTICULTURAL SOCIETY, 14 The rate of growth of microorganisms in the chilled juices, ph 3., during storage was about the same for those products prepared from either 42 or 59 Brix frozen orange concentrates (Table 3). Since the juices were prepared from concentrates, the diacetyl values of the reconsti tuted juices were at the very low levels of.1 and.4 ppm. After storage for one day at 4 F., the counts were less than 5, per ml. Fermen tation was detectable organoleptically by the third day in the juices stored at 6 F. of these products at 5 F. resulted in spoilage on or be fore the eighth day. The diacetyl value increased to 6.2 and 4. ppm in eight days at 5 F and to 115.9 and 6. ppm in 4 days at 6 F. There was no increase in the diacetyl values of the juices after storage at 4 F. for 22 days. Some deteri oration in flavor of both of these reconstituted juices was detectable after storage for 22 days but is was not microbiological in character. of ascorbic acid in chilled orange juices stored at 6 F. The ascorbic acid content of the three chilled juices, stabilized at 155, 15 and 195 F., and also of an unstabilized juice was determined at various s during their storage for 24 days at 4 F. The retention of ascorbic acid was less in the juices in the 8 oz. cartons than in those packed in the 32 oz. car tons (Table 4). For example, retention was found in the juice, stabilized at 15 F., after storage for 24 days when it was packed in 32 oz. cartons, as compared to 65 retention when the 8 oz. cartons were used. Calculations showed that there was 8.4 ml. of juice per 1 sq. cm. of juice surface exposed to air in the headspace when the 8 oz. cartons were used. However, with the 32 oz. cartons this relationship was 2.5 ml. of juice per 1 sq. cm. of juice surface exposed to the headspace air. Thus, oxidation and con sequently the loss of ascorbic acid would be greater in the 8 oz. cartons. Also, the area of waxed paperboard in contact with the juice was greater for the 8 oz. cartons than that for the 32 oz. cartons. Calculations indicated that there was.8 sq. cm. of paperboard per milliliter of juice when 8 oz. cartons were used, whereas there was.5 sq. cm. of paperboard per milliliter of juice in the 32 oz. cartons. If waxed cartons are to some extent permeable to atmospheric oxygen, then the above relationship could cause more oxi dation of the ascorbic acid in juices packed in the 8 oz. cartons. Rushing and Senn (8) reported that their data on retention of ascorbic acid in chilled orange juice in waxed 32 oz. cartons would predict an average loss of about as compared to an average 14 loss in 3 weeks found by Tucker, Lime and Griffiths (9) when chilled orange juice was stored in sterile glass con tainers. There was a 15 loss when stabilized juices were stored at 4 F. for 3 weeks in 32 oz. waxed cartons, which is about the same as that found by Tucker et al. (9) when glass containers were used (Table 3). However, before drawing conclusions about the permeability to air of waxed paperboard, details other than those men tioned by Rushing and Senn (8) concerning the study by Tucker et al. (9) must be known and taken into consideration. Neither heating nor the temperature of stabi lization affected the loss of ascorbic acid to any major amount. Summary The most important factor in preventing mi crobiological spoilage of chilled orange juice was found in this investigation to be storage temperature. Products packed in waxed cartons and stored at 5 and 6 F. spoiled rapidly, but fermentation did not occur in juices held at 3 and 4 F. for three weeks. The total number of microorganisms was re duced greatly by stabilization of juice at 155, 15 or 195 F. before chilling. However, the initial plate counts and those after the samples had been stored at 4 F. for 24 days were about the same, regardless of the stabilization tempera ture. Chilled juices prepared from stabilized frozen 42 or 59 Brix frozen orange concentrates had similar microbiological characteristics and showed little difference in rate of microbiological spoil age from those prepared from freshly extracted juice, which was stabilized before chilling and packaging. of ascorbic acid in stabilized chilled orange juices, packed in 32 oz. waxed cartons, was after storage at 4 F. for three weeks. However, when 8 oz. cartons were used, there was a greater loss of ascorbic acid. LITERATURE CITED 1. Association of Vitamin Chemists. 1951. Methods of Vitamin Assay 2nd ed. 31 pp. Interscience Publishers, Inc., New York, N. Y. 2. Florida Canners Association, Winter Haven, Florida. 3. Hamrick, David O. 1956. Problems related to the production and distribution of cartoned orange juice. Trans actions Citrus Engineering Conference, Florida Section, American Society of Mechanical Engineers, Winter Haven, Florida, also Citrus Industry : (1) 226; (11) 2931; and (12) 2224, 32. 195. 4. Hill, E. C. and F. W. Wenzel. 195. The diacetyl test as an aid for quality control of citrus products. I. Detection of bacterial growth in orange juice during con centration. Food Technol. 11: (4) 24243.
MURDOCK AND DENNIS: DIACETYL 29 5. Lamden, M. P., C. E. Schweiker and H. P. Pierce. 1. Ascorbic acid studies on chilled, fresh and fermented orange juice. Food Research 25: 1922. 6. Patrick, Roger and Elmer C. Hill. 1959. Microbiology of Citrus Fruit Processing. Fla. Agr. Expt. Sta. Bui. 618.. Purko, M., R. V. Hussong and C. W. Kaufman. 1956. Bacteriological flora in frozen and chilled orange juice. Food Research : 5588. 8. Rushing, N. B. and Vincent J. Senn. 14. Shelf life of chilled orange juice with heat treatment and preser vatives. Food Technol. 18: (8) 122224. 9. Tucker, D. M., B. J. Lime and F. P. Griffiths. 1959. Private communication to Rushing and Senn. Food Technol. 18: (8) 1224. 1. U. S. Department of Agriculture, Agricultural Mark eting Service, Washington, D. C. United States standards for grades of chilled orange juice. June, 1959; also, Amendment. December 1, 1959. 11. Wenzel, F. W., E. L. Moore, C. D. Atkins and Roger Patrick. 1955. Chilled citrus products. Proc. Florida State Hort. Soc. 68: 16. Also, Citrus Magazine 18: (6) 14 19. 1956. DETECTION OF DIACETYL AND ACETYLMETHYLCARBINOL IN PROCESSING FROZEN CONCENTRATED ORANGE JUICE A PRELIMINARY REPORT D. I. MURDOCK AND R. E. DENNIS Minute Maid Company A Division of The CocaCola Company Orlando Since the introduction of frozen concentrated orange juice in 1946, there has been concern with the production of offflavors characterized as be ing similar to "buttermilk." Early investigators, namely Hays (3) and Murdock (9), were of the opinion that the bacteria responsible were prin cipally those belonging to the genus Lactobacillus. It soon became evident that a rapid method was needed for the detection of the metabolic by products of these organisms. The application of the VogesProskauer (VP) test was first intro duced to the citrus industry by Hill et al. in 1954 (4). Further modifications were reported during that same year by Byer (1). Further studies were reported by Hill and Wenzel in 195 (). This procedure and modifications thereof are essentially for the detection of diacetyl and acetylmethylcarbinol (AMC). The test is highly specific for these compounds in concentrations of 1 ppm or less (8). The VP test, as applied to the citrus industry, is based on the development of a pink to ruby color after the addition of alphanaphthol and 4 KOH creatine solutions to a juice distillate sample. The color reaction is measured on a Lumetron Colorimeter at the end of a given interval which may vary from 1 to 1 or more minutes. Most processors of frozen concentrated orange juice now use some form or modification of the VP test for the detection of diacetyl and AMC in orange juice. A considerable variation exists in the industry as to the procedure employed for the detection of these compounds. For example, some plants read the color reaction 1 minute after the addition of the VP reactants, others at 8, 1 minutes or intermediate periods. Even the distillation of 3 fractions is employed by some processors. Confusion also exists as to the importance of AMC in the juice sample. Diacetyl content is generally used as one index of sanitation of the processing operation. AMC, on the other hand, is considered by most food technologists to have little or no sanitary significance. AMC, which is tasteless, is found naturally in citrus fruit and increases as the fruit matures (5). According to Hill et ah (5), it may reach levels as high as 4 ppm in oranges. AMC is also produced by a large number of organisms including lactic acid bacteria. The presence of AMC in fruit as it is re ceived at the plant makes it difficult to determine whether or not an abnormally high colorimetric value in the juice distillate is an indication of poor sanitary practices. Hill (6) states that in 1 minute the VP reaction for diacetyl has prac tically attained maximum color development while the reaction for AMC is only around 1 complete. Therefore, it appears that the VP reaction, as applied to the citrus industry, should reflect the maximum amount of diacetyl and the the minimum concentration of AMC in the juice. Also, since almost every citrus processor has his own "pet" modification of the VP test for the de tection of diacetyl, the need for a standard pro cedure becomes quite evident. This would be especially helpful where concentrate is purchased on the open market, and diacetyl is one of the specifications for quality. This paper concerns the methods used by the