ROUSE AND ATKINS: INSOLUBLE SOLIDS

METHODS

FOR

117

ESTIMATION OF INSOLUBLE

SOLIDS IN CITRUS JUICES AND CONCENTRATES1 A. H. Rouse and C. D. Atkins

Florida Citrus Experiment Station Lake Alfred

The water-insoluble solids in a citrus juice or concentrate are important because they determine some of the characteristics of these products. Rouse, Atkins and Huggart (4) found that water-insoluble solids, pectinesterase activity and pectin content generally increased as the pulp content was increased. The determination of "free and suspended pulp", using centrifugation, (5a) has been used by the Agricultural Marketing Service of the U. S. Department of Agriculture as an in dication of the water-insoluble substances in citrus juices and concentrates. The amount of pulp is a factor in ascertaining the U. S. grades (5a-g), and if excessive, is termed a physical defect. In the centrifugal method (5a) accuracy is sacrificed for rapidity. Meas urement is made of the volume of the pulp

particles without taking into consideration particle size. This is dependent upon mechan ical sizing during processing and the amount of water of hydration in the swollen particles. Olsen and Asbell (3) showed that the size of pulp particles influenced pulp content as determined by the centrifugal method. They determined the amount of coarse pulp by a flotation method and reported that the waterinsoluble solids increased when either the coarse pulp or total pulp was increased; how ever, no relationship was found between the coarse pulp content and the total pulp con tent after sizing of the particles in commercial concentrates. Thus it is evident that the centrifugal method, now used throughout the citrus processing industry for the determina tion of pulp content, is rapid but leaves much

to be desired from the standpoint of accuracy and reproducibility of results. This is especial ly true when it is applied to citrus concentrates in which the size of the pulp particles varies Cooperative research by the Florida Citrus Experi

ment Station and Florida Citrus Commission.

Agricultural 430.

Experiment

Station

Journal

Florida

Series,

No.

considerably. It therefore became necessary to find or develop another method for the estimation of insoluble solids in citrus juices and concentrates that would be accurate and capable of providing reproducible results; also a method that could be completed in a short period of time was desirable. Water-insoluble solids in citrus juices may

be determined by the quantitative A.O.A.C. methods I and II (1) as described for fruit and fruit products. In method I, the water-in soluble solids are dried overnight, while in method II a rapid drying device, such as the Moisture Teller Model 271T (manufactured by Harry W. Dietert Co., 9330 Roselawn Ave., Detroit 4, Michigan) is used which requires only 15 minutes; however, a long period of time is necessary in washing the pulp free of water-soluble substances.

The purpose of this paper is to present two

quantitative methods for the estimation of in

soluble solids in citrus juices and concentrates: one, a modified A.O.A.C. water-insoluble solids procedure; the other, a more rapid al cohol-insoluble solids procedure.

EXPERIMENTAL PROCEDURE Frozen concentrated citrus juices used for this investigation were commercial samples obtained from processors in Florida for a survey study.

Insoluble solids values are ex

pressed as milligrams per 100 grams of re constituted juice.

Preparation of Samples. — Two 6-oz. cans

of thawed 42° Brix concentrated citrus juice

were thoroughly mixed to insure uniformity in sampling. A 160 ml. aliquant was measured into a 250 ml. graduated cylinder, trans ferred to a blender, and comminuted for 3 min. The comminuted concentrate was pour ed into an 800 ml. beaker and 3 parts by volume of distilled water, 480 ml., was used to flush out the cylinder and blender. This brought the total volume of the reconstituted juice to 640 ml. The juice in the 800 ml. beaker was stirred mechanically at a moderate rate of speed for 3 min. This dispelled the

118

FLORIDA STATE HORTICULTURAL SOCIETY, 1955

air and resulted in a reconstituted juice in which the pulp particles were more uniform ly suspended.

Apparatus and Preparation of Crucibles for Quick Filtration. - A vacuum filtrator as described by the Fisher Scientific Company (2) was fitted with a prepared Gooch type crucible.

A circular disc of absorbent cotton, ca. M-in. in diameter, was placed on the bottom of a Pyrex, Gooch type, fritted glass crucible of 15 ml. capacity with coarse porosity disc. This was placed in the crucible holder of the filtrator and the cotton pad and crucible elutri ate with 250 ml. of hot distilled water, fol lowed by a flush with 15 ml. of 99.5% isopropyl alcohol. This was dried for 3 hr. at 105°C., cooled in desiccator, and weighed. Method for the Estimation of Water-In

soluble Solids. — Into a 250 ml. beaker, weigh accurately on a torsion balance 50 g. of the prepared sample of reconstituted juice. The prepared juice should be agitated at the time of sampling to insure complete suspension of the free and suspended pulp. Cover beaker containing juice with a coverglass and place on an electric hot plate with rheostat set so as to boil gently for 20 min. Transfer the hot juice to a 50 ml. graduated, short conical-bot tom, centrifuge tube. The beaker is rinsed with hot distilled water and the volume made up to 50 ml. in the tube. Centrifuge at 2100 r.p.m. for 3 min. and decant the supernatant solution through a washed, tared crucible. Add ca. 20 ml. hot distilled water and triturate the residue in the bottom of the centrifuge tube with a rubber policeman. Rinse off policeman and make contents to a volume of 50 ml. with hot distilled water. Repeat the centrifugation and decantation. Add ca. 20 ml. of 66% isopropyl alcohol (2 parts of alcohol by volume to one part of distilled water), again triturate, make contents to 50 ml. with alcohol, and centrifuge as before. Pour off supernatant alcoholic solution through crucible and with a plastic squeeze bottle containing 66% alcohol direct a fine stream of alcohol around the edge of the residue packed at the bottom of the tube. This will loosen the solid material with out breaking it into fine pieces. Transfer all of the residue to the crucible with alcohol, suck dry, place in oven at 105°C. for 3.5 hr., cool in desiccator for 30 min. and weigh.

Method for the Estimation of Alcohol-In soluble Solids. — Into a tared 50 ml. centrifuge tube, weigh accurately on a torsion balance 25g. of the prepared sample of reconstituted juice. Add 99.5% isopropyl alcohol to a volume of 45 ml. and stir with a rubber police man. Rinse policeman with 99.5% alcohol and make contents to a volume of 50 ml. Centrifuge the tube at 2100 r.p.m. for 3 min. and decant the supernatant liquid through a washed, tared crucible. Add ca. 20 ml. of 66% isopropyl alcohol to the tube and triturate the residue using a rubber policeman. Rinse policeman and make contents to a volume of 50 ml. with 66% alcohol. Centrifuge and de cant as before. Transfer the residue from the centrifuge tube to the crucible as described above for water-insoluble solids. Dry in oven at 105 °C. for 3 hr., cool in desiccator for 30 min. and weigh. RESULTS AND DISCUSSION

Preliminary data had shown that samples of commercial and Station packed citrus con centrates, when comminuted, resulted in 5.3 to 35.5% lower water-insoluble solids than noncomminuted samples. This range of difference is probably due to the occlusion of varying amounts of soluble solids by pulp particles of different size present in citrus concentrates. These result from varying commercial prac tices, such as extraction methods, finishing op erations, or type of pumps used in circulat ing the juice during processing. The data in Table 1 show a comparison of the insoluble solids in reconstituted juices, calculated on a basis of 12°Brix juice, when samples of frozen orange concentrates were prepared for analysis under different condi tions. Water-insoluble solids did not vary much when a sample was prepared in three different ways, as long as it was comminuted. The differences in water-insoluble solids found in the 3 samples prepared from any single concentrate averaged 10.6% with a maximum difference of 14%. According to the data in

Table 1 concerning alcohol-insoluble solids, it made little difference whether the sample was comminuted or not. The greatest difference found between the alcohol-insoluble solids in the samples prepared in three different ways was less than 8% for any individual concen trate. The quantity of alcohol-insoluble solids in orange juices (Table 1 and 2) was found

ROUSE AND ATKINS: INSOLUBLE SOLIDS

119

TABLE 1

Comparison of Insoluble Solids in Reconstituted Juices (12°Brix) When Erozen Orunge Concentrates are Prepared for Analysis Under Different Conditions. Water-insoluble solids Concentrate Concentrate

number

comminuted

Concentrate reconstituted

and

and

Alcohol-insoluble solids

Concentrate

Concentrate

comminuted

comminuted

Concentrate

Concentrate

reconstituted

reconstituted

and not

and reconstituted

comminuted

ag./lOOg.

comminuted

mg./lOOg.

mg./lOOg.

reconstituted

comminuted

reconstituted

mg./lOOg.

mg./lOOg.

mg./lOOg.

and

and not

1

228

223

244

547

2

106

115

99

409

404

411

3

115

123

129

476

473

451

4

91

100

95

350

337

349

5

249

245

237

575

533

558

6

75

86

81

365

369

383

7

118

123

131

464

441

AA9

1 8

1U

153

151

475

510

494

to be from 2 to 5 times the amount of waterinsoluble solids, indicating no relationship be tween the two kinds of insoluble solids. The

547

541

juice, were within ± 4.5% of the mean value.

It was found that the most satisfactory pro cedure for sample preparation for the determ

higher values for the alcohol-insoluble solids

ination of insoluble solids in citrus concen

are mainly due to the presence of pectin in the juice. Results from duplicate samples of either water-insoluble or alcohol-insoluble solids, when taken from the same prepared

trates was to comminute the product prior to

reconstitution.

Results of insoluble solids and

pulp analyses are presented in Table 2 for

TABLE 2

Insoluble Solids Found in Commercial Samples of Frozen Concentrated Citrus Juices0 Reconstituted Juices

Concentrate

Type

numbtr

of concentrate

9 10 11 12

Orange tt it n

13 14 15 16

Orange

17 18

Orange

Pulp ,ty volume

unsized

sized

A*

%

Water-insoluble solids

mg./lOOg.

13.5

12.0

247

12.0

560

228

10.0 10.0

231

544 491 424

10.5

212

9.0

it

9.0

9.0

8.0 7.0

7.0 7.0

6.5

6.5

6.0

101

5.0

77

it

19

ti

20

tt

solids

mg./lOOg.

13.0 12.0

tt it

Alcohol-insoluble

6.0 6.0 5.5

172 151 147 130

402 424

336 361

5.5

101

5.0

98

316 286 295 291

21

Grapefruit

5.0

5.0

36

215

22

Gft.-Orange

7.0

6.0

96

315

blend

120

FLORIDA STATE HORTICULTURAL SOCIETY, 1955

12 commercial concentrated citrus juices. The quantity of pulp, when unsized, in the recon stituted orange juices ranged from a high of 13.5 to a low of 5.5% by volume, whereas the amount of sized pulp in the juices varied from 12 to 5%.

However, the data did not indicate a rela tionship between descending insoluble solids values and decreased pulp contents except in a very general way. This confirmed the data of Olsen and Asbell (3) in which the pulp content by centrifugal and flotation methods showed no relationship to water-insoluble solids found in reconstituted commercial frozen citrus concentrates. However, a relationship of pulp content and water-insoluble solids has been shown by Olsen and Asbell (3) and by Rouse, Atkins and Huggart (4) when packs were prepared, with varying amounts of pulp, from the same lot of fruit using the same pro cessing techniques.

No correlation was found between alcoholinsoluble solids and water-insoluble solids in citrus concentrates nor between the insoluble solids and free and suspended pulp; never theless it is believed that the amount of al cohol-insoluble solids could be established as an indication of physical defects such as ex cessive juice cells, particles of membrane, core, peel, seeds and portions thereof. The determination of alcohol-insoluble solids, as an indication of these defects, has the ad vantage over the determination of pulp by the centrifugal method. This is true because it is based upon weight, rather than volume measurements, and therefore the size and de gree of hydration of the pulp particles does not affect the determination. Alcohol-insolu ble solids may also be determined more rapid ly than water-insoluble solids and comminution of the concentrate or juice is not necessary.

SUMMARY

Data presented in Table 2 show that the quantity of sized pulp in the reconstituted concentrates, as determined by the centrifugal method, was usually smaller than the amount of pulp found in the same products when the pulp particles were not sized by comminution; however, in some of the concentrates—for example sample 13—the percentage of pulp by volume was the same before and after comminuting. The data in Table 2 also in dicates that the pulp content, as determined by the centrifugal method, of a reconstituted juice containing unsized pulp particles is not a true criterion of the amount of water-in soluble solids in the product. Thus, although samples 13 and 14 had the same pulp con tent prior to comminution, the water-insoluble solids in sample 13 was approximately 14% greater than that in sample 14. Results ob tained using samples 18 and 19 also illustrate this fact. The time required by the method for the estimation of water-insoluble solids is approxi

mately 5 hr., whereas the method for alcoholinsoluble solids required about 4 hr. The alcohol-insoluble solids method requires less time because the sample is not boiled and 30 min. less drying time is needed. Since 3 to 3.5 hr. are required for drying the insoluble

solids, the use of a device such as the Moisture Teller would reduce the time of these methods to 1 hr. or less.

A modified A.O.A.C. method for the rapid determination of water-insoluble solids in citrus juices and concentrates is presented and dis cussed. Centrifugation replaces the slow gravity filtration during the washing pro cedure and is followed by an alcohol rinse for quick drying. Also presented for consideration in estab lishing standards for physical defects in citrus products is a method for estimating alcoholinsoluble solids, which is more rapid than the

water-insoluble solids procedure. A means of quick drying is suggested which would reduce the time of these methods to less than an hour. Data are presented on both the water- and

alcohol-insoluble solids in 14 frozen concentrated citrus juices, firms previous observations that ship exists between pulp*content

commercial which con no relation by volume, and water-insoluble solids in commercial con

centrates. LITERATURE CITED 1. Association of Official Agricultural Chemists. 1950. Official Methods of Analysis. 7th Ed., 320-321 Washington, D. C 2. Fisher Scientific Company. 1952. Modern Lab oratory Appliances. 410. Pittsburgh, Pennsylvania. 3. Olsen, K. W. and Dorothy M. Asbell. 1951. De termination of the pulp content of concentrated citrus juices. Proc. Florida State Hort. Soc, 64th Ann. Meeting: 171-174.

HENDRICKSON AND KESTERSON: CRUDE HESPERIDIN 4. Rouse, A. H., C. D. Atkins and R. L. Huggart.

1954. Effect of pulp quantity on chemical and physi cal properties of citrus juices and concentrates. Food Technol.

8:

5. U. S. grades of Agricultural (a) Frozen

431-435.

Dept. Agriculture. U. S. standards for canned citrus juices and concentrates. Marketing Service, Washington, D. C: concentrated grapefruit juice, Dec. 1951;

121

(b) Frozen concentrated blended grapefruit juice and orange juice, Dec. 1951; (c) Canned concentrated grapefruit juice, Nov. 1945; (d) Canned grapefruit juice, Oct. 1954; (e) Canned blended grapefruit juice and orange juice, Oct. 1954; (f) Frozen concentrated orange juice, Sept. 1950; (g) Canned orange juice, Oct.

1954.

PURIFICATION OF CRUDE HESPERIDIN R. Hendrickson and J. W. Kesterson Florida Citrus Experiment Station

.

Lake Alfred

Hesperidin—the principal glucoside of or anges and predominant constituent of such products as "citrin", citrus Vitamin P and citrus bioflavanoids— continues to find ex tensive use as a therapeutic agent. It is ad ministered in the treatment of abnormal capil lary fragility and permeability associatedt with hemorrhagic disorders. The growing interest in hesperidin has point ed to the need for additional information on the recovery and purification of this bytproduct. A previous report (2) evaluated the conditions and procedures necessary for maxi mum recovery of crude hesperidin. It is the

purpose of this report to discuss the methods of purifying crude hesperidin.

The essential features of the previously re ported recovery procedure (2) were an al kaline extraction of chopped orange peel fol lowed by neutralization, and the addition of filter-eel and heat to promote crystallization and aid filtration. The crude hesperidin/sub sequently isolated and dried, was found to vary in hesperidin content from a high of 42 percent to a low of 25 percent, depending on the time of year and the variety extracted. The investigation of methods of purifying crude hesperidin was carried out with a com posite sample of the following composition: 30 percent hesperidin, 40 percent filter-eel, and 30 percent ballast impurities. Extraction

Technique

for

Purification.

—

The general extraction technique was confined initially to using a minimum of 25 grams of crude hesperidin per 500 ml. of alkaline solu tion or solvent. After stirring 15 to 60 minutes, the sample was filtered, neutralized (diluted Florida Agricultural Series, No. 415.

Experiment

Station

Journal

in the case of some solvents), and allowed to crystallize for 24, 48, or 72 hours. The sample was then filtered, dried, and analyzed for yield and purity. Hesperidin recovery in the latter part of the investigation was improved by slurrying the crude with only 350 ml. of solvent and using the other 150 ml. to either reslurry or wash the remaining insolubles. Initial experience indicated that an economi cal commercial purification of crude hesperidin would be dependent on an alkaline extraction followed by filtration and neutralization. Cal cium and magnesium hydroxides were found to have an inadequate hesperidin solubility for a concentrated extraction. Dilute ammon ium hydroxide and sodium carbonate solutions did not show adequate solubility for hesperidin even at elevated temperatures.

It was soon

evident that a strong alkali such as sodium hydroxide would be required. When a 0.2N sodium hydroxide solution was used a very satisfactory extraction of hes peridin was obtained, but the subsequent fil tration from the insoluble and inert material proved exceedingly slow. Further investiga tion led to the addition of isopropyl alcohol to the extracting solution as a means of in creasing the filtration rate.

Effect of Alcohol on Filtration Rate. - Vary ing quantities of isopropyl alcohol were used to replace the water of a 0.2N sodium hy droxide solution and the rate of filtration measured against concentration of alcohol. As can be seen in Fig. 1, the rate of filtration pro portionately increased with the alcoholic con tent. The improvement is brought about by the greater degree of dehydration and coagula tion of high-molecular-weight colloidal im purities. The filtration was sufficiently im proved with 40 to 50 percent alcohol to war rant further consideration.

It was necessary,

however, to evaluate the effect of alcoholic content on hesperidin recovery.