Hi i I 1 lwl \ I l 4 W I k 1 III 400—; 1c» \|\ (IMO-{>- CLEANENG MODEL STUDY: SUGAR SOIL REMOVAL BY ALKAUNE CLEANING SOLUTIONS Thesis {or the Dogma of M. 80 MC .{GRN STATE UNEVERSITY Carlos Borrero Angel i964 TH ESIS L [B R A R Y Michigan State { I University ABSTRACT CLEANING MODEL STUDY: SUGAR SOIL REMOVAL B! ALKALINE CLEANING SOLUTIONS By Carlos Borrero Angel Studies of the rate of sugar removal from a sugar soil model were performed using distilled water and alkaline cleaning solutions of varying concentration. Sugar soils were exposed to cleaning solutions for a defined time interval at constant temperature and rate of agita- tion and the quantity of the sugar soil dissolved was determined by ‘weight loss. Results suggest that in this model system the soluble sugar soil removal was controlled by the concentration gradient of the cleaning solution. Cleaning solutions having the smallest solute concentration exhibited the largest rate of soil removal. Cleaning solution additives were studied; an increase in trisodium phosphate concentration significantly decreased the rate of soil removal; tetrasodium.phosphate and sodium gluconate had no significant effect on the rate of soil removal; the addition of EDIA to the cleaning solution in concentrations of 2.5% and 51 by weight of caustic had no effect 'whereas the addition of 102 EDIA to the cleaning solution enhanced the rate of cleaning by 162. The rate of soil removal was increased by increasing the temperature and rate of agitation; the rate of soil removal was found to be constant ‘with respect to time, the amount of the soil with the cleaning solution. The commercially used cleaning solutions studied suggest that the increase in soluble solids not caustic of the cleaning solution‘with extensive usage is not large enough to decrease the efficiency of the cleaning solution. ii CLEANINGilflbe STUDY: SUGAR SOIL REMOVAL BYUALKALINE CLEANING SOLUTIONS By Carlos Barrera Angel A THESIS Submitted to the school for Advanced Graduate Studies of Michigan State University in partial fulfillment of the requirements for the degree of MISTER or SCIENCE Department of Food Science 1964 a mis padres iii ACKNCHLEDGEMENTS The author is greatly indebted and appreciative of his major professor. Professor 1. J. Pflug, for encouragement, guidance and con- structive criticism during his graduate work. ‘ The author also wishes to convey his sincere appreciation to Professor 0. W. Kaufmann and J. K. Jensen for their consideration and interest in their review of this manuscript. This project was initiated and sponsored by the American Battlers of Carbonated Beverages. This organisation has supported the work financially; however, members of the organization have individually contributed knowledge and testing materials. We appreciate the active support of this industry. Special recognition is due Mr. H. B.‘ Korab for his interest, counsel and enthusiastic support of this study. Thanks go to Professor B. S. Schweigert, Chairman, Food Science Department. for his interest and support of this program; to Mr. Allen Stewart and Mr. John Schamel for their assistance in the laboratory; to Dr. John Blaisdell for his helpful suggestions; to Hr. P. J. Pellets for proofreading the manuscript; and to Missb. Bang for the typing of the manuscript. iv 1. 2. 3. 6.. IABLE OF CONTENTS V Page Introduction 1 Review of Literature 3 a. Present cleaning specifications 3 b. Effect of variables on rate of cleaning 5 c. Longevity of solutions 7 d. Sequestering agents used and their chemical behavior 8 Experimental Design 13 a. General considerations 13 b. Description of test equipment 15 c. Test materials 15 (1) Experimental soils 15 (a) Sugar soil 15 (b) Sugar plus milk soil 15 (2) Cleaning solutions 16 (a) Neon solutions 16 (b) Neon solutions containing soluble solids 16 (c) 32 neon solutions containing sequestering agents 16 (d) Commercial cleaning solutions mode from proprietary cleaning compounds 17 (a) Used comercial proprietary compound cleaning solutions pd. Testing procedure 17 (1) Procedure for tests on the rate of soil removal 17 (2) Testing procedure for used cleaning solutions 19 Results 20 a. Apparatus performance tests 20 b. Tests of the effect of concentration 20 c. Study of some of the variables 21 17 Page d. Tests of commercially used cleaning solutions 21 5. Discussion 29 6. Summary 39 7. Conclusions and Recommendations 42 8. Recommendations 44 9. Appendix 54 10. References 61 vi Table 7. 8. 19. 1Q, LIST OF TABLES Analysis of variance table of weight loss of six batches of sugar soil, experimental procedure and agitator position. ‘Weight loss of sugar soil with increased sodium hydroxide concentration at 150°? and agitation speed of 147 rpm, mean of six replicates. Rate of sugar soil removal by distilled water and alkaline cleaning solutions at 150'? and agitator speed of 147 rpm, mean of six replicates. Rate of sugar soil removal by solutions of prOprietary cleaning compounds and 31 NaOH at 150’? and agitator speed of 147 rpm, means of six replicates. Rate of sugar plus milk soil removal by distilled water and alkaline cleaning solutions containing soluble solids at 150°! and agitator speed of 147 rpm.:means of six replicates. Analysis of variance table of the rate of soil removal of sugar soil of 31 sodium hydroxide cleaning solution with 2.52. 52 and 102 trisodium phosphate by weight of caustic. Analysis of variance table of the rate of soil removal of sugar soil of 32 sodium hydroxide cleaning solutions 'with 2. 5%. 5% and 10% tetrasodium pyrophosphate by weight of caustic. Analysis of variance table of the rate of soil removal of sugar soil of 32 sodium hydroxide cleaning solutions 'vith 2.5%. 52 and 10% sodium gluconate by weight of caustic. Analysis of variance table of the rate of soil removal of sugar soil of 31 sodium hydroxide cleaning solutions ‘with 2.51. 51 and 102 EDTA by weight of caustic. Rate of soil removal from sugar soil by 3% sodium hydroxide solutions with varying concentrations of four sequestering agents at 150°? and agitator speed of 147 rpm. 18 replicates per concentration. vii Page 22 22 23 23 24 24 25 25 26 26 Table 711. 1.3. 14. Rate of soil rmoval of sugar soil with varying tempera- ture and agitator speed of 147 rpm. Rate of soil removal of. sugar soil with varying agitation at 150’! using distilled water cleaning solutions. Analysis of variance table of rate of soil removal of sugar soil with respect to time. Normality and concentrations of soluble solids for 11 commercial cleaning solutions. viii Page 27 27 28 28 Figure l. 4. 5. 6. 7. 9. 10. LIST OF FIGURES Test equipment, agitator mechanism with beaker holder beneath it. Flask holding device and beaker holder. Rate of soil removal of sugar soil vs. increasing NaOH concentration. Rate of soil removal of distilled water cleaning solution vs. soluble solids concentration. Rate of soil removal of H O, 3%, 51 neon vs. Soluble solids concentrat on. Rate of soil removal of 3 prOprietary cleaning compounds using sugar soil. Rate of soil removal of sugar plus milk soil‘with H20 and 32 NaOH cleaning solutions vs. soluble solids concentration. Rate of soil removal of sugar soil by distilled water solutions at 3 temperatures vs. soluble soils concentration. Rate of soil removal by 32 NaOH cleaning solutions at 3 temperatures vs. soluble solid concentration. Rate of soil remgval of sugar soil by 51 neon cleaning solution at 3 temperatures vs. soluble solids concex tration. - Rate of soil removal of distilled water cleaning solutions at 3 rates of agitation vs. soluble solids concentration. Page 18 18 45 46 47 48 49 50 51 52 53 INTRODUCTION The cleaning of reusable carbonated beverage bottles is a relatively old operation, yet little is known about the process of soil removal from these containers. Analytical data regarding the effect of variables on the soil removal operation is practically nonpexistant. Increasing production costs and a decreasing profit margin in the food processing industry necessitates that the efficiency of processing equipment be increased in order for this process to be economically feasible. Returnable bottles are used and the bottle washing operation is undertaken because it is less expensive for the bottler to reuse the bottles than to use one trip bottles; however the washer is a large item of capital investment. The bottle washer must be used efficiently. The objective of this study is to investigate some of the variables involved in the bottle washing operation*which should be cons sidered to Optimize the design of bottle washer and the bottle washing operation. One way of increasing washer efficiency is to reduce bottle washing time'which increases bottle capacity of the machine; however this requires improved washing rate and perhaps a'more effective washing solution. If we knew when a cleaning solution no longer acted as an effective cleaning agent, it could be replaced to maintain maximum washer efficiency. If, however, bottle washing solutions do not wear out then the cost of recharging the soak compartment of the bottle washing machine could be saved. The variables of caustic concentration, soluble solid concentration temperature, rate of agitation and time were studied to detemmine their effect on the rate of soil removal. These variables are the controlling l factors in the bottle washing operation. The water used for cleaning reusable containers may'contain calcium and magnesium ions generally classified as hardness; many of the pro- prietary compounds used contain sequestering agents of either an organic or inorganic nature to remove these ions which effect bottle cleaning and final bottle appearance. The sequestering agents studied included: trisodium phosphate, tetrasodium perphosphate, sodium gluconate, and tetrasodium ethylenediaminetetraacetate. Tests were made using these compounds in 3% sodium hydroxide solution to determine the effect of an increase in the sequestering agent concentration on the cleaning rate. This investigation is part of a study sponsored by thekaerican Bottlers of Carbonated Beverages (A308) to evaluate the effect of some of the variables which may affect the rate of soil removal in reusable carbonated beverage bottles by caustic cleaning solutions. This thesis describes an experimental procedure for studying the rate of removal of a soluble sugar soil by caustic solutions; the effect of increasing the concentration of sequestering agents on the rate of soil removal, and the effect of the variables of time, temperature, and the rate of agitation on the rate of soil removal. REVIEW OF LITERATURE Present Cleaning Specifications The cleaning of reusable carbonated beverage bottles is one of the most important processes in the manufacture of carbonated beverages. This process has suffered little change in the past 70 years. The recommendations appearing in the 1893 Bottle washing Bulletin of the Nordberg Manufacturing Co. of Milwaukee, Wisconsin, were: "The cleaning solution consists of caustic soda dissolved in water. This solution can be made of any strength, but for economical reasons about 32 caustic in a tank full of water is ordinarily used. Heating the solution adds to the efficiency of the soaking solution." These recommendations are similar to those of the A303 (1951), "unclean bottles shall be exposed to 32 solution of which not less than 601 is caustic (sodium hydroxide) for a period of not less than.5 minutes at a temperature of not less than 130'F, or to an equivalent cleansing and sterilizing process." These conditions express the minimum conditions to which an unclean con- tainer should be submitted. Forty-three states have requirements con- trolling the type and strength of bottle washing solution and 26 of these states restrict the bottle washer Operators to one solution con- centration, a minimum.temperature and a minimum contact time. A general rule of thumb used by bottlers states that for each 10'! increase in solution temperature of 501 increase in the caustic solution concentra- tion, the necessary contact time will be reduced by one-half. Jennings (1963) defines soiling as "the result of a decrease in the free energy of a system", implying that removal of soil necessitates 4 work to add enough energy to reverse the process. This work is generally supplied to the system.in two forms: (1) mechanical energy, and (f) physiochemical energy. The mechanical energy in the bottle washing pro- cess is supplied by means of turbulence of the cleaning solution, abrasion (brushes), or hydroSprays, whereas the physiochemical energy comes from the detergent and additives in the cleaning solution. Pflug et al. (1961) using radioactive soils determined that soil build up was greatest on a dirty surface than on a clean surface. This would be assumed from the former statement that soiling results from a decrease in the free energy of a system. The initial soiling would then be the most difficult to incur, and subsequent soiling once the first soil was deposited would not involve as great an energy change, there- fore it would recur at a faster rate. various techniques have been suggested for estimating soil deposits. Jensen (1946) used light transmittance readings by a spectrophotometer, Kaufmann et al. (1960a,b) used the bacterial count, with or without swab test for determining soil deposits. Jennings (1960), and Pflug et a1. (1961) used P32 labeled milk soils in their cleaning studies. Jennings (1961) found a high correlation between the soil removed and the radioactivity removed. Pflug et a1. (1961) studied the effect of stainless steel finish on the rate of soil removal using dry milk residues. His data indicated no significant differences in the rate of radioactive soil removal among the finishes. These results were confirmed by Kaufmann et al. (1960a) in their studies of the bacteriological cleanability of stainless steel finishes. Pflug et a1. (1961) and Jenning (1960) have both shown that soil removal from.a surface is a complicated process that requires increasing quantities of energy to remove the last remaining traces of soil from the critical surface. In reusable bottle washing the cleaning solution should remove the soil from the container and sterilize the bottle through the action of heat and caustic during washing or subsequent rinsing. The containers from the washer may be slightly contaminated; therefore, they are classified as commercially sterile. The disinfectant action of alkalies, as summarized by Carpenter (1961), was found to be due largely to the presence of hydroxyl ions in the solution, with the greatest degree of dissociation having the most effective germicidal action. The additives in the caustic solution affect the permeability of the cell membrane causing either a loss of protOplasmic material or enhancement of the effect of the sodium hydroxide. An example is trisodium phosphate inp creasing the germicidal efficiency of the solution. Effect of Variables on Rate of Cleaning Ruff and Becker (1955) listed the following factors influencing the effectiveness of the cleaning sclution: (l) the concentration of caustic in the solution, (2) the temperature of the solution, (3) the time of contact of the solution and container, (4) the composition of the cleaning solution (type and composition of additives), (5) the type of washer, (6) the condition of the containers to be cleaned, (7) the nature of the water supply, and (8) the amount of soil residues in the solution. Information pertaining to cleaning and soil removal in carbonated beverage bottles is practically nonpexistant, but there has been a great deal of work done on other systems such as the cleaning of dairy equipment, textiles, etc. In studying the effect of detergent con- centration, Ladewig (1955) stated that the presence of excessive amounts of caustic in the cleaning solution inhibited the efficiency of the cleaning solution; however, there was no quantitative data given, nor was the degree of inhibition stated. Parker et a1. (1953) summarized the effect of temperature on C1? (cleaning in place) systems by stating that the higher temperatures gave more efficient cleaning. The work of Jennings (1959) ascertained that the removal of cooked-anemilk films from CIP lines by solutions of sodium hydroxide exhibited a QIB'F of 1.6 within the temperature range llS-lBG'F. He speculated that "increasing the temperature of the solution continues to increase the cleaning efficiency until one reaches a point where detergents decompose or the vapor pressure of the fluid interferes with the operation." Shand (1958) explains the effect of temperature on glass by saying that "when glass is suddenly heated, the initial stress developed is compressive so that the hazards of fracture are slight, but when suddenly chilled, the stresses are tensile so that the probability of fracture is increased greatly." Fracture because of increased tensile stress due to expansion or contraction produced by temperature change is called thermal shock. Dnngfelder (1957) found thermal heat shock breakage was higher ‘when hot glass was chilled than when cold glass was subjected to a hot environment. He estimated a maximum temperature difference between the glass and the cooling solution of 35-45’? and that very slight breakage occurred upon introducing the bottles in the hot solution. It is the heat shock effect which imposes limitations on the Operating temperatures of bottle washers. The washing cycle includes the immersing of the bottles in a hot cleaning solution and.then the rinsing of the bottles to remove the detergent solution from the bottle. The rinsing Operation is carried on at a lower temperature than the soaking Operation, thus the cooling of the bottles is one of the cons trolling factors in determining an Operating temperature. Buchanan and Levine (1939) reported a great variation in the temperatures of cleaning solutions. They found temperatures from IDS-170’? in usage throughout the bottling industry. Ruff and Becker (1955) confirmed the preceding temperature data, but indicated a majority of washers Operating between 140-170'F. When bottles are cooled in a step process by immersing in cooler soaker solutions where changes in temperature are within a AT of 35.3, a washing temperature of under 170°F is not critical in glass breakage. Longevity of Solutions There is no definite way to know when a cleaning solution is no longer effective due to soil pick up during usage. iMoCallion (1961) tested two soaker solutions using different fonmulations to determine when they became ineffective; the testing criteria was the ability of the cleaning solution to clean glass without an increase in the reject rate. NO difference was Observed over an extended period of time. Dormuth (1956) summarised the extent Of the knowledge on the life Of cleaning solutions when he stated that "when the contamination reaches a certain point it is necessary to dump the solution from the tanks and recharge with fresh solution." From this statement it is evident that the solution is discarded whenever the Operator judges it to be ineffective. Sequestering Agents Used and Their Chemical Behavior Cleaning solutions used for bottle washing in the carbonated beverage industry are adversely affected by the action of hard water salts with the sodium.hydroxide in the solution. The free calcium and magnesium salts present in the solution react with the sodium hydroxide to form insoluble magnesium hydroxide and calcium carbonate. These precipitates form scale on.washer tank, heat transfer surface and moving parts of the machine. The Operating efficiency of the washer is lowered and costly damage to the equipment is the end result. Sequester- ing agents are used to prevent the precipitation of the aforementioned salts from the solution by forming soluble complexes with the ions of calcium and magnesium. The precipitation of magnesium hydroxide on the bottle surface causing a cloudiness or haze condition can be alleviated by use of chelated caustic solutions (Ladewig, 1955). The inorganic phosphates or the organic sequestering agents such as the sodium salts of the organic acidso-sodium gluconate or sodium salts Of the aminopolycarboxylic acids EDTA are the materials used to sequester the calcium and magnesium ions present in hard water. These compounds act on the metal salts by forming a soluble complex, which in turn prevents the formation of precipitates on both the washer and the surface of the bottles. These sequestering agents also improve the wetting ability and the free rinsing qualities of the solution. Schwartz et a1. (1958) defines free rinsing as "freedom from ‘waterbreak, a condition.where water drains in a continuous film with- out breaking into droplets or streams". This condition indicates a 9 surface free of soil and reduces the carry over by the bottles of austi 3 solution from one tank to another. Bottles are clear and shiny indicating a clean bottle whereas if bottle washing soiutions are H) Ire.e rinsing, Opaque Spo ts on the bett1e can be used to indicate an -nelean and undesirable tattle. The inorganic phosphates have a wider usage at pre sent as free rinsing agents because Of their lower cat as compared to the organic sequestering agents (C';.ahereck and )0 The compounds to be examined in this study are: trisodium U: \CI WittEl-ig 1;; phosphate, tetrasodium perthSphate, sodi' gluconate, and ethy1.eneo diaminetetraacetate (EDTA). Triscdi um pho osp hate has been widely used because of its cleaning properties; however, when added to a caustic solution prepared using hard water, ions are removed as insoluble phOSphat e salts. Trisodium phosphate is different from the sequestrants in ha.t it p ecipitates th calcium salts rather than tying them.in a soluble chelate (Schwartz, et a1. (1958). Overman (1964) indicated that normsiiy used concentre tions of tris odium phos Mpha e ran from 0~lGE of the caustic present in solution. The inorganic poiyphosphates have had extensive usage due to th “1 sequestering properties and1 ow cost 0 Martell and Calvin (1953) state that "the relative effectiveness of the polyphosphates increase with increasing chain length." Audisadvantage of the polyphosphate sequestering agents is their tendency to hydrolyze to orthophOSphate (reversion), caus_ng precipitation of the insoluble salts of the phOSphates. Chshereck and Hartell (1959) indicate that "the degree of hydrolysis of poiyphosphates increases with an increase in chain length." They also found tetrasodium pyrophoSphate to have the 10 slowest rate of reversion of the polyphosphates at 100°C. Reversion is highly accelerated at the high temperatures used for bottle washing, and is undesirable because upon reversion there is precipitation of the in- soluble phosphates of the metal salts in the solution. In their des- cription of the polyphosphates they describe an additional effect encountered with the usage of these compounds, namely the "threshold effect". The "threshold effect" is described as the use of sequestering agents in small concentrations to prevent the formation of insoluble salt precipitates. Thus it is possible to maintain higher quantities of calcium and magnesium ions in solution that would stoichiometrically combine with the available polyphosphate. The "threshold effect", however, does not work when the concentration of available calcium is greater than 20 ppm. Mehltretter et al. (1953) reported that another advantage derived from the use of polyphosphates, in addition to the deactivation of metal ions, was "the ability to defloculate and suspend water insoluble substances such as clays and hard soils." Benson (1856) states that polyphosphates promote the wetting of glass by the caustic solution and also promote the free drainage of water allowing for lower carry over loss and generally cleaner bottles. Use of tetrasodium perphosphate depends on the conditions to be met. Cone centrations from 0-122 (expressed as a percentage of the caustic) were reported by Korab (1964) as being in use by the industry. One of the major considerations in the choosing of the concentra- tion of additives to be used appears to be the cost of the cleaning compound ingredients. The two main organic sequestering agents used in the bottle washing industry today are sodium gluconate and tetra- sodium ethylenediaminetetraecetate salts. 11 Chabereck and'Marteil (1959) stated that the industrial value of sodium gluconate salts lies in the fact that at pH values greater than 11 the sodium gluconate salts become very efficient sequestering agents in the presence of free caustic. Also, that in the presence of free sodium hydroxide, sodium.giuconate compared with tetraphosphate and citric acid was the most effective sequestering agent in the normal pH range over a wide range of sodium.hydroxide concentrations. In a similar manner to the polyphoSphates, sodium gluconate also exhibits the "threshold effect" in caustic solutions. Pfizer Gluconates in Caustic Bottle Washing (1959) listed the recommended amounts of sodium gluconate to be used in caustic solutions based on the hardness of the water used. These values are tabulated in.Table 1. Table 1. Average levels of gluconate required in practice. Sodium Gluconate Required Per 100 Pounds Caustic Soda at Caustic Concentratioas_of Water Hardness .1Ef—-_——_§Z. 3% 41 51 grains/gal. 1b. 1b. 1b. 1b. 1b. 1 to 5 2.0 1.0 1.0 0.5 0.5 6 to 10 5.5 3.0 2.0 1.5 1.0 11 to 15 9.0 4.5 3.0 2.0 2.0 16 to 20 12.5 6.0 4.0 3.0 2.5 20 or over 14.0 7.0 5.0 3.5 3.0 The organic chelatiag agents offer some advantages over the poly- phosphates, especially as regards thermal stability in aqueous solutions. The metal chelates have a higher stability than the polyphosphates, and they remove rust by forming stable chelates with ferrous and ferric ides (Mahltretter et al., 1953). The maximum efficiezmcy of ethylene- diaminetetraacetate (EDTA) is attained at a pH greater than 8. The 12 action of EDTA differs from the polyphosphates in that it forms a chelate 'with the metal ions. Hartell and Calvin (1953) define a chelate as formed by the combination of a metal with a substance containing two or more donor groups so that one or more rings are formed. The high cost of this compound has inhibited its more widespread usage even.with its greater efficiency as a sequestering agent. EXPERIMENTAL DESIGN General Considerations The design of this experiment required a homogenegus test soil ‘which‘would be representative of soils which might be encountered in the commercial practice of washing reusable bottles. The selection of'a soil for use in carbonated beverage bottle cleaning studies is a critical part of this cleaning study. more than 99% of the returned carbonated beverage bottles contain only the residual beverage plus extraneous materials such as straws, cigarettes and silt or dust from the air. A sugar soil was selected because the problem of washing carbonated beverage bottles is mainly one of removing the sugar soil from the- bottle's inner surface. The sugar glass used in this study is a soil similar to that encountered in the bottles after the water has evaporated and is therefore a logical soil to use to simulate the cone ditions inside the bottle. A.sugar glass is more difficult to remove than a crystallinepsugar. Bottlers will generally set aside bottles containing oils, tar, heavy clays or cement to wash them separately or they may destroy these bottles. Therefore, the washer Operation is geared for normally soiled returned bottles and for these conditions this study should provide data to improve this Operation. A preliminary experiment was made in.which a sugar glass was cut into rectangular pieces 3" x 1-1/2" x 1/4". ‘Wire loOps were imbedded in the sugar glass and these pieces were suspended in the solution and the 'weight loss determined by difference. The nature of the sugar glass, namely its fluidity at high temperatures was as encountered in these tests giving poor results which led to the improved beaker test method where area is accurately controlled and it is necessary to only remove 13 l4 soil from a horizontal surface. A second preliminary test was made in which the soils were produced in the bottom of 250 ml beakers. A sufficient amount of soil was added to assure the exposure of a constant surface area to the cleaning solu- tion in each test vessel. The sides and bottom of the soil layer adhered to the glass, therefore were not exposed to the action of the cleaning solu ion. The soil could be separated from the solution and reweighed to determine the weight loss by decanting the cleaning solution from the beaker following a test. Preliminary tests made using these vessels with 200 ml of cleaning solution at lSO'F under 150 rpm of agitation for 5 mizutes, indicated that this method was unsatisfactory. The agitation conditions in the beaker caused the formation of several vortexes in the solution. The resulting effect was uneven removal of soil from the surface, with a greater amount of soil removed where the vortex inpinged on the soil. The lack of reproducibility of results made the preceding method of testing undesirable. Further experimenta- tion showed that when the soil was poured into 600 ml heakers, conditions of turbulence did net appear, thus an even surface, free of indentations, was maintained and reproducible results could be obtained. A carbonated beverage contains acid and flavor in addition to sugar. In preparing the synthetic soil we can start with a sugar-acidc flavor mix; however, acid and flavor will be evaporated by the time the end of the cook was reached. To add acid and flavor to the sugar glass it is best to add it in the dry form and blend it into the hot molten glass by mixing. In general these additions will not make the soil any \ harder o 15 Description of Test Equipment The testing equipment consisted of three basic parts: the stirring unit with six agitators; the beaker holder; and a constant temperature bath. The stirring unit was designed so the blade in each beaker would produce equal agitation. The system was assembled to minimize the "without agitation" contact time of the solution with the soil. It was important to place the agitator on the test heakers immediately upon their immersion in the bath and at the end of the time of agitation to remove the unit allowing decantation of the cleaning solution. The beaker holder maintained the six test beakers equi- distant from each other and directly beneath the agitator blades (see Figure 1). This holder also enabled the Operator to place and remove the six beakers simultaneously from the constant temperature bath. The flask holding unit with the six 500 nl Erlenmeyer flasks was used to pour 500 m1 of cleaning solution at the test temperature into the beakers in the holder. The flasks were spaced such that solution could be poured directly into each beaker (see Figure 2). The constant temperature bath was equipped with a thermostat which maintained a test temperature 1: l'F‘. Test Materials Experimental Soils Sugar Soil. The sugar soil was prepared by mixing 1600 g of sucrose and 400 g of corn sugar with 1500 ml of distilled water. This solution was heated until the temperature reached 310°F. A hard creek consistency resulted upon tooling. 'When the solution reached BIG'F, a sufficient quantity was poured into each of the 600 ml heakers to pr«duee 16 a soil thickness of 1/4 to 3/8 inch. After the soil had cooled to room temperature, the beakers with the soil were weighed to £0.01 3. ‘§ggar plus milk soil. 1600 g of sucrose, 400 g of corn sugar, 100 g of powdered milk, and 1500 g of water were mixed to make a solution which‘was heated to 310’F, at which time it was poured into 600 ml baskers to a depth of 1/4 to 3/8 inch. The beakers were weighed to t 0.01 g after cooling to room temperature. Cleaning Solutions The cleaning solutions used consisted mainly of neon solutions of varying concentrations, 32 NaOH solutions with varying concentrations of sequestering agents, and several commercial proprietary compounds. Enos solutions. The solutions of NaOH were prepared using dis- tilled water and anhydrous neon. Solutions were prepared by weight, 1.0. (32 neon - 3 g NaOH to 97 g H20). Solutions were prepared in the following concentrations: 3%, 5%, 101, and 20% neon. ugggg solutions containing_soluble solids. These solutions were prepared as above and to each concentration, 202, 402, 601 soluble solids (sucrose) was added, i.e. (to 3000 g of solution 600 3 sucrose for 202 soluble solids). The solutions were then placed in the 150'? bath and held at this temperature overnight. 0n the following day these solutions were restandardized to their respective normalities. ‘gzgnaon solutions containing sequestering agents. Solutions of 32 mass were prepared as above and to these solutions were added 2.5%, 52, and 102 of the sequestering agents based on the NaOH concentration i.e. (for a 102 sodium gluconate solution in 32 NaOH - 30 g NaOH/IOOO g of solution and 3 g of sodium gluconate were added to the 31 NaOH solution). The sequestering agents used in this study were trisodimn l7 phosphate, sodium perphosphate, sodium.glucenate and tetrasodium ethylenediaminetetraacetate. 'gggmercial cleaningsolutions made from proprietary compounds. Cleaning solutions were prepared using commercial prOprietary compounds (supplied by several manufacturers) plus distilled water to make a 3% caustic concentration. To these solutions were added 20%, 402, 60% soluble solids (sucrose), the resulting solutions being kept at 150'? overnight and restandardized to their original normality before using. Used commercial prOprietary compound cleaning solutions. Bottle washing solutions from the bottle washers of 11 commercial carbonated beverage bottlers were obtained from the A303. These solutions were tested to determine percent soluble solids and the normality of each solution. Testing Procedure grocedure for tests on the rate of soil remeval. The cleaning solution to be used in the test was first equilibrated in a large sen? tainer to approximately the test temperature and then poured into the 500 m1 Erlenmeyer flasks and allowed to equilibrate to the test temper- ature in the constant temperature bath. Six beakers‘with sugar soil 'were attached to the beaker holder, the solution poured in (Figure 2), and the beakers placed into the constant temperature bath. The stirrer unit was quickly lowered over the beakers and agitation begun. The elapsed time between pouring the solution into the beakers and the -start of agitation.was approximately 15 seconds and was constant for all tests. At the end of the five minute test period the stirrer was stopped and removed from.the beakers. The beakers were removed frem the bath and the solution was decanted. The interval from the end of 18 Figure 1. Test equipment, agitator mechanism with beaker holder beneath it. ‘ l ' [92‘ M .. r .\ . I " ‘/ ’ I I an. ‘ j 1“{“~fi_5 p_ ' x "I ." ‘ Figure 2. Flash holding device and beaker holder. l9 agitation to the decanting of the cleaning solution was approximately 20 seconds and was constant for all tests. After the beakers had cooled to room temperature they were weighed to 1; 0.01 g on a Sartorius Kilomat Balance. Six replicate beakers were run per test. Testigg procedure for used coumercial cleanirisolutions. The solutions obtained from the A303 were of varying age and degree of use. The total soluble solids in each of these solutions was determined by evaporating the water in a 200 3 sample of solution in a vacuum oven at '7 70'C then determining the soil content by the weight of the residue left in the beaker. The normality of the solution was obtained by titrating a 10 m1 sample of the caustic solution with 1 N H01 using phenophthalein indicator. The percent soluble solids not caustic was obtained by taking the difference between the total solids and the weight of caustic, as determined from solution normality, divided by the weight of the original sample. RESUBTS The results are reported under four headings: (1) Apparatus performance tests, (2) Effect of concentration gradient on the rate of soil removal, (3) Effect of temperature, time and rate of agitation, (4) Tests of commercially used cleaning solutions. Apparatus Performance Tests The results of tests to establish the reproducibility of the experimental procedure, equality of different batches of sugar soil, and reproducibility of agitation conditions at each agitator position ‘were treated using the analysis of variance method (Dixon and Massey, 1957). The results of the analysis of variance are tabulated in Table l. The data in Table 1 indicate no difference at the 5% level of significance between batches of soil, procedure, or agitator position. Tests of the Effect of Concentration The means of six replicate tests on the effect of sodium hydroxide concentration on the rate of soil removal are presented in Table 2; the mean values of six replicate tests on the effect of increasing the con- centration of soluble solids on the rate of soil removal of distilled ‘water, 3% sodium hydroxide, and 52 sodium hydroxide cleaning solutions are presented in Table 3; the effects of increasing the concentration of soluble solids in solutions prepared using commercial proprietary com- pounds are tabulated in.Tab1e 4; and the results of the study conducted to observe the effect of increasing the concentration of soluble solids in distilled water and alkaline cleaning solutions on the rate of soil removal of a sugar plus milk soil are listed in Table 5. 20 21 An analysis of variance was performed on the values obtained from tests on the effect of the addition of 2.52, 52, and 102 sequestering agents in sodium hydroxide cleaning solutions. The sequestering agents used were trisodium phOSphate, tetrasodium pyrophosphate, sodium gluconate, and tetrasodium ethylenediamdnetetraacetate; the results of the analyses are tabulated in.Tables 6, 7, 8, and 9, respectively. The means of six replicate tests performed on each of these concentrations of sequestering agent are presented in Table 10. Study of Time, Temperature and Rate of Agitation The results.pf a study conducted to observe the effect of an in- crease in temperature on the rate of soil removal of distilled water and alkaline cleaning solutions are listed in Table 11. The effects of varying agitation on the rate of soil removal of sugar soil by distilled water cleaning solutions of varying concentra- tions of soluble solids are tabulated in Table 12. An analysis of variance test was made on the results of tests made varying the contact time between the cleaning solution and the soil. Time intervals of 5, 10 and 20 minutes were used in this study. The solution used was distilled water. Table 13 represents the analysis Of thi‘ tests Tests of Commercially Used Cleaning Solutions From the cleaning solutions obtained from the A303 data were gathered to determine the percent concentration of soluble solids and the normality of the cleaning solution. Table 14 shows values for all treatments. TABLES Table 1. {Analysis of variance table of weight loss of six batches of sugar soil, experimental procedure, and agitator position. Degrees of Sum of Mean Source freedom squares squares F Batches of soil and procedure 5 1.26 .252 1.28 Position in agitator S .42 .084 Residual 25 4.92 .197 Total 35 6.60 (s Table:2. weight loss of sugar soil with increased sodium hydroxide concentration at 150°! and agitation speed of 147 rpm, mean of 6 replicates. NaOH Concentration Rate of soil removal Cpercent gglcmz min mean range 52 16.65 (16.10-17.24) 101 12.51 (11.67-14.15) 202 4.62 ( 4.36- 5.02) 22 23 Table .33. Rate of sugar soil removal by distilled water and alkaline cleaning solutions at 150°? and agitator speed of 147 rpm, means of 6 replicates. Rate of soil removal ( cm2 min 3 Soluble solids Distilled water 3% NaOH 5 51 mm 02 24.65 20.12 16.65 202 19. 87 14. 98 11. 16 40% 15.63 7.85 3. 25 601 10.36 4.07 .93 . Table 4. Rate of sugar soil removal by solutions of proprietary cleaning compounds and 3% NaOH at 150'! and agitator speed of 147 rm. means of 6 replicates. Rate of soil removal (ngcm2 min) 2 soluble solids 3% Neon §glution 1 Solution Solution 3 0% 20.12 19.05 19.01 17.31 202 14. 98 12.40 10. 14 11. 16 401 7.85 7.09 4.43 7.05 602 4.07 .690 1.53 .70 24 Table 5. Rate of sugar plus milk soil removal by distilled water and alkaline cleaning solutions containing soluble solids at 150’! and agitator speed of 147 rpm, means of 6 replicates. Rate of soil removal Salami, min) .7: Soluble solids Distilled water 32 New 07. 24.70 20.12 201 19.14 13.51 401 12.02 7.15 602 5.30 1.59 Table '6. Analysis of variance table of the rate of soil removal of sugar soil of 31 sodium hydroxide cleaning solution with 2.52, 51, and 102 trisodium phosphate by weight of caustic. Sum ‘ of Hoan £225.29. if. squares squares 1' Concentration 2 11.08 5.54 13. 19** Within 51 22.04 .42 Total 53 33. 12 r 2 51 - 3.19 95h) Table 70 Table 8. Analysis of variance table of the rate of soil removal of sugar soil of 3z.sodiumxhydroxide cleaning solutions with 2.51, 51, and 10% tetrasodium pyrophosphate by weight of caustic. Source m Concentration Within Total 21. 2 51 53 Sum of squares 1.11 15.91 17.02 F.95 (2,51) - 3.19 Eben squares .555 .31 1.79 Analysis of variance table of the rate of soil removal of sugar soil of 31 sodium‘hydroxide cleaning solutions with 2.52, 5%, and 10% sodium gluconate by weight of caustic. nun: 2: Concentration 2 Within 44 Total 46 ’.95 (2:44) Sum of squares 4.44 16.71 21.15 I 3.23 Mela squares 2.22 .38 P 5.8** 26 Table T9. Analysis of variance table of the rate of soil removal of sugar soil of 3% sodium hydroxide cleaning solutions with 2.51. 51. and 102 tetrasodium ethylenediaminetetraacetate by weight of caustic. Source Concentration Within Total P .9 Table 10. Rate of soil solutions‘wi agents at 15 replicates p Sequestering agent Sum of Mean .23 squares squares 2 8.5 4.25 8.67** 51 25.13 .49 53 33.63 5 (2,51) I 3.19 removal from sugar soil by 32 sodium; hydroxide th varying concentrations of four sequestering 0°? and agitator speed of 147 rpm. 18 or concentration. Rate of soil removal (mg/cm? min} concentration as I of cuastic Ha3P0g Na4P207 Na gluconate Na4 EDTA 0 20.12 20.12 20.12 20.12 2.52 19.09 20.15 18.25 19.96 5% 16.04 21.41 20.36 20.87 101 15.31 21.20 20.11* 23.35 *11 replicates for this test. 27 Table 11. Rate of soil removal of sugar soil with varying temperature at agitator speed of 147 rpm. 1 Rate of soil removal $330332 min) Soluble #135'3 T-lSO F - . T-170'P solids H20 37. NaOH 5% Neon H20 3% Neon 51 NaOH H20 32 NaOH 52 NaOH 3 0% 20.51 18.58 13.16 24.65 20.12 16.65 27.64 26.11 21.64 20% 14.95 9.78 8.33 19.87 14.98 11.16 24.25 18.18 16.51 402 8.33 4.07 2.04 15.63 7.85 3.25 19.09 13.35 10.95 601 1.89 .98 -2.83 10.36 4.07 .93 12.51 8.25 4.95 Table 12. Rate of soil removal of sugar soil with varying agitation at 150'! using distilled water cleaning solutions. gte of soilIremoval (mgécmz min) 0 RPM 65 RPM 14 RPM 1 Soluble solids 02 10.79 13.38 24.65 201 5.31 8.51 19.87 401 2.22 5.78 15.63 801 -2.83 -1.34 10.36 28 Table 13. .Analysis of variance table of rate of soil removal of sugar with respect to time. Degrees of Sum. of Mean .3guggg freedom squares squares F Contact time 2 75.47 37.74 .0384 Within 6 5,849.72 982.86 Total 8 5,925.19 P.95 (2,6) - 5014 Table 14. Normality and concentration of soluble solids for 11 commercial cleaning solutions. 44¢¢¢4¢¢¢4¢ Normality .276 .84 .97 .80 1.03 1.21 1.38 1.04 .93 .60 .61 Z Caustic 1.10 3.30 3.90 2.80 4.10 4.80 5.50 4.20 3.70 2.40 2.40 2 Soluble solids not caustic 1.03 2.4 2.5 3.2 2.0 3.7 3.5 2.2 2.3 2.2 1.7 DISCUSSION The results of the analysis of variance tabulated in Table l indi- cate no significant difference at the 5% level in the experimental pro- cedure between runs, different batches of soil, or beaker position in the agitator system. Having established the reproducibility of procedure and soil. studies of the other variables were undertaken. Studies on the effect of increasing sodium hydroxide concentration indicated a decrease in the rate of soil removal with an increase in the sodium hydroxide concentration. Figure 3 is a plot of the data of Table 2 showing the effect of the increase in the sodium hydroxide concentration on the rate of soil removal. The results from this study indicate that distilled water is more effective sugar soil removing agent than sodium.hydroxide. The role of caustic in the cleaning solution appears to be twofold: (1) To act as a germicidal agent during the cleaning cycle, the combined effect of temperature and caustic are bacteriocidal and maintain the soak tank solution in a sterile condition. and (2) To saponify and solubilize any oil or fat residues left in the bottle. When a sugar soil is to be removed, water is the most effective cleaning agent due to its larger concentration gradient. The mechanism of soil removal is one where a boundary layer is formed between the sugar soil and the solution, consisting of a saturated sugar solution layer in which there is'movement of sugar to the solution and water to the soil. The concentration gradient between the solution and the film is con- trolling and diffusion of the soil to the solution is greater than water to the soil. An increase in the sodium hydroxide concentration decreases this concentration gradient thus lessening the driving force 29 30 of the sugar to the solution with a subsequently lower rate of soil re- moval. To determine more precisely the effect of this factor, a study was designed covering a large range of concentration gradients. To dis- tilled water. 3% and 5% sodium.hydroxide cleaning solutions. 20%. 40%. and 60% soluble solids were added. Table 3 lists the mean values re- sulting from these tests and the graphical representation of this data in Figure 4. indicates that as the concentration gradient is increased, the rate of soil removal is decreased. Figure 4a is the linear regression line drawn from the data obtained from the test, the equation for this line is Y - -.234 X'+ 24.57. indicating a linear correlation for the rate of soil removal and the concentration of soluble solids present in distilled water cleaning solutions. The plot of rate of soil removal versus percent concentration of soluble solids not caustic for alkaline solutions demonstrates a non-linear relationship. This relationship indicated that the sodium hydroxide content. while lower- ing the rate of soil removal of the solution at each concentration of soluble solids, also affected the relationship of the percent soluble solids with the rate of soil removal. Linearity was found no longer applicable to this system. An increase in the percent concentration of soluble solids will increase the viscosity of the solution thus making the rate of soil removal also dependent on the viscosity. Increasing the viscosity of a cleaning solution reduces the rate of soil removal. In a highly viscous system the‘motion of the atoms is slower, and their tendency to go into solution is smaller due to the higher shear forces and resistance to movement. Solutions of higher soluble solids not caustic have notably lower rates of soil removal than those of lower 31 concentrations; this decrease is in part due to their higher viscosity. Three proprietary cleaning compounds and 32 sodium hydroxide cleaning solution.were tested using the sugar soil. The proprietary compounds were formulated to have a 3% caustic strength with the same three concentrations of soluble solids not caustic added to each solution. The results of these tests are listed in Table 4. The data of Table 4 is represented graphically in Figure 5 indicating 3% sodfmm‘hydroxide to have the highest rate of soil removal of the four solutions. The three preprietary compounds were observed to have approximately the same rate of soil removal for each of the four concentration gradients. Since the concentration of additives in these solutions was not known. further analysis could not be made. In order to establish a comparison for the sugar soil a second type of soil was devised. For this soil the same fonmulation as for ' the sugar soil was used with the addition of 100 g of powdered skimmed milk. This soil represented a different type of surface and removal of this soil included removal of insoluble particles. Tests were made using distilled water and 3% sodium hydroxide cleaning solutions with the four concentrations of soluble solids. Table 5 shows that an in- crease in soluble solids decreases the rate of soil removal, again indicating the concentration gradient to be a controlling factor in this system. Due to a smaller concentration gradient for increasing soluble solids concentration. the 3% sodium hydroxide displayed a smaller cleaning ability than the distilled water cleaning solutions at the four soluble solid concentrations. Figure 6 is a graphical representa- tion of Table 5. Studies were made to determine the effect of increasing the concen- 32 tration of sequestering agents on'the rate of soil removal of 3% sodium hydroxide cleaning solution. For these studies. concentrations of trisodium phosphate. tetrasodium pyrophosphate. 806111!!! gluconate, and tetrasodium ethylenediaminetetraacetate of 2.52.. 52 and 102 by weight of caustic were used. An analysis of variance (Table 6) was performed on the experimental values from the soil removal tests for trisodium phosphate. It was found that the concentration of trisodium phosphate affects the rate of soil removal. with an increase in the trisodium phosphate concentration causing the decrease in the rate of soil removal. Table 10 lists the mean values for these tests. A concentration of 2.5% trisodium phos- phate had no apparent effect on the rate of soil removal. but an in- crease in the concentration from 2.5 to 5% reduced the rate of soil removal from 19.09 to 16.04 lug/cal2 min. A further increase in the tri- sodium phosphate concentration from 5 to 102 did not have as radical an effect on the rate of soil removal as an increase from 2.5 to 51; however. it did lower it. The effect of trisodium phosphate on the rate of soil removal is not unlike that of sodium hydroxide in that it decreases the rate of soil removal with an increase in concentration. The trisodium phosphate effect on the rate of soil removal is more marked than the sodium hydroxide. i.e. The rate of soil removal of 3‘1 sodium hydroxide with 10% trisodium phosphate cleaning solution is 15.31 rug/cm2 min. compared to that of 52 sodium hydroxide with no phosphate which'is 16.65 tug/c1112 min. This again could be caused by its effect on the concentration gradient. The solubility of sodium hydroxide at O'C is 42 g/ 100 m1 of water. whereas that of trisodium phosphate is 8.8 g/100 ml of 33 water. These data explain.why the addition of 102 trisodium phosphate by weight of caustic to 32 sodium hydroxide cleaning solution has a lower rate of soil removal than a 52 sodium hydroxide cleaning solution. The effect of increasing the concentration of trisodium phosphate is five fold that of sodium hydroxide. The effect in solution of the trisodium phosphate on the 04"" ions present due to water hardness is to form insoluble precipitates with these ions. Trisodium phosphate is in this manner different in its treatment of water hardness from the other sequestering agents. The fact that it decreases the rate of soil removal may be due to its not forming soluble complexes as the other sequestering agents. Tests using tetrasodium pyrophosphate as the sequestering agent ‘were performed for the three concentrations. Table 7 represents an analysis of variance for the experimental results. and at the 51 level of significance there was no difference between the three concentrations used. There was also no significant difference between the 32 sodium hydroxide solution and the solution‘with the sequestering agents. Mean values for these tests are listed in Table 10. A study of the effect of increasing the concentration of sodium gluconate on the rate of soil removal was performed and the experimental results were analyzed by the analysis of variance method (Table 8). This table indicates significantly different means for the three con- cantrations. The rate of soil removal for the 2.51 solution was lower than for the 5 and 101 solutions. The mean results of these tests are tabulated in Table 10. A contrast between the 2.51 sodium gluconate solution and the 31 sodium hydroxide solution indicated a difference 34 between the two rates of soil removal. with the 31 sodium hydroxide solution being the more efficient cleaning solution. The decrease in efficiency between the 2.52 sodium gluconate solution and the 31 sodium hydroxide solution was in the order of 97.. At the higher sequestering agent concentrations such as the 5% and 107. concentrations there was no significant effect on the rate of soil removal when compared with the value from the 37. sodium hydroxide solution. The effect of tetrasodium ethylenediaminetetraacetate on the rate of soil removal was studied using the three concentrations of sequestering agent in32 sodium hydroxide solution. An analysis of variance (Table 9) on the experimental data indicates a significant difference between the mean values for the three concentrations. There was no effect on the cleaning rate when 2.51 and 51 concentrations of EDTA were added; however. for the 10% concentration. a notable effect was observed. The values listed in Table 10 showed an enhancement in the rate of soil removal of 16% by the addition of 10% M. The effect of the variables of temperature was studied using dis- tilled water, 31 sodium hydroxide, and 51 sodium hydroxide cleaning solutions. These three concentrations (0-51 sodium hydroxide) were used because they represented the causticity range used in the bottle washing industry. Temperatures of 130, 150, and 170‘! were used to study the effect of temperature on the rate of soil removal for the three cleaning solutions at the four concentrations of soluble solids. The data of Table 11 indicate that the rate of soil removal for the three temperatures for any given sodium hydroxide or soluble solids concentration was highest at 170’1'. Figures 7, 8 and 9 graphically represent the data of Table 11. From these figures a contrast can be 35 ‘made between distilled water and the alkaline cleaning solutions for the four concentrations of soluble solids. An increase in the temperature of a cleaning solution decreases the viscosity of the solution and increases the solubility of sugar in the solution, thus increasing both the diffusion rate and the concentration driving force. The in- crease in the diffusion rate is linked to the effect of an increase in temperature on the molecular motion. An increase in the temperature will increase the motion of the molecules due to higher energy thus increasing the transfer rate of material from the film to the solution. The effect of sodium.hydroxide concentration on the rate of soil removal at each temperature was the same; an increase in sodium.hydroxide con- centration or soluble solids concentration decreased the rate of soil removal at all three temperatures tested. I The effect of agitation on the rate of soil removal was studied using distilled water cleaning solutions. The agitation speeds used were 0, 65, and 147 rpm. Agitation was maintained at these low velo- cities because of the difficulty in reproducing the tests when higher agitation speeds, which caused turbulent conditions in the test vessel, were used. The data in Table 12 showed a greater rate of soil removal for the four concentrations of soluble solids used as the result of a greater rate of agitation. Figure 10 represents the data of Table ”A It is interesting to note that at 0 and 65 rpm the cleaning solution with 601 soluble solids caused a gain in.weight in the test vessel. This can only mean that the diffusion of water into the sugar soil was greater than the rate of diffusion of the sugar into the fluid. The low speed of agitation would cause a larger mass transfer fihm thickness which would effectively cause a greater resistance to the sugar diffusion. 36 However, the concentration gradient in the sugar was large enough to cause a driving force for the water toward the soil causing a gain in weight of the soil in the test beaker. The values obtained would indi- cate the advantages to be gained from the added energy for soil removal which the fluid supplies when it is under a higher velocity. Turbulent conditions, while not ideal for this model would be ideal in an industrial cleaning system. These conditions would cause a higher rate of soil removal because of greater energy being available for dislodging the soil particles attached to the glass surface. Tests were conducted to determine the effect of the variable of time on the rate of removal of sugar soil using distilled water cleaning solutions. An.analysis of variance (Table 13) performed on the rate of soil removal data found the rate to be independent of time; however, the amount of soil removed was prOportional to the soil solution con- .tact time. The test model is such that for the contact times used the increase in concentration due to soil removal can be considered negligible. This is due to the large amount of solution'with respect to the amount of soil removed per unit of time. Since the concentration gradient remains constant, the rate of soil removal is also a constant with the amount of soil removed being prOportional to the soil solution contact time. The time factor is important in bottle cleaning because of both the required sterilizing time and the time necessary to remove the soil from.the glass surface. Determination of the soil pick up of cleaning solutions with respect to age and usage were conducted on 11 commercially used clean- ing solutions which were provided by the A303. These solutions were ‘made up from various proprietary compounds as used by the processing 37 plants. Appendix 1 lists the information which was supplied with each of the cleaning solutions. The cleaning solutions are listed from4A to K and for each of these solutions there is given age of solution, compound used, type of waSher, number of bottles washed, frequency of make-up, etc. Percent soluble solids and normality was calculated for each solution. The concentration varied from 1.03 to 3.5% soluble solids not caustic as shown in Table 14. There is no apparent correla- tion between age of solution, number of bottles washed and the percent soluble solids. This result was expected because the conditions en- countered in each plant are rarely reproducible. The data of.Appendix 1 would indicate that age or use are not criteria for discarding the solu- tion. However, these data do indicate that periodic make up of solution with both.water and caustic to maintain the causticity and volume of the solution constant, do not allow a soluble solids to build up to a level where they will substantially affect the rate of soil removal. The type of washer greatly influences the cleaning of the con- tainers because of the differences in solution contact time, the use of mechanical energy to assist in soil removal, and the temperature of operation. The differences between brands and models of the same brand are great enough so that any study performed on a particular washer will he pertinent only to that model of washer. Thus, because of the dis- similarity between washing Operations, this study was performed in a laboratory under ideal conditions with no attempt being made to simulate industrial conditions. The analytical solution to this problem could not be solved because of the model chosen for this study. The beaker test vessel presents a very difficult system to analyze due to the velocity gradient toward the 38 periphery of the beaker. The velocity is greatest near the sides of the beaker and least in the center. This velocity gradient causes a gradient in the mass transfer film thickness which is preportional to the rate of soil removal. There were several attempts made at trying to determine the thickness of the mass transfer film by analogy with the heat transfer film. These tests were unsuccessful because the equipment at hand was not sensitive enough to measure the very short came up times encountered in the test model. Because of insufficient data the values for the diffusivity of sugar in.this system.cou1d not be determined, nor was there an available relationship for the mass transfer coefficient of the system.under study. SUMMARY Studies of the effect of several variables on the rate of soil removal have shown significant differences. The experimental procedure developed in this study allowed for the determination of the rate of soil removal when some of the variables affecting the system were changed. An analysis of variance indicated that the results of tests made using this procedure were reproducible with respect to batches of soil, beaker position in the agitator and enperimental procedure. It was observed that an increase in the sodium hydroxide concen- tration would cause a marked decrease in the rate of soil removal. The range of caustic concentrations used in the bottling industry (0.52) were studied extensively. The addition of 51 caustic to a dis- tilled water cleaning solution caused a decrease in the rate of soil removal of 32.51. The use of higher concentrations of sodium hydroxide further decreased the rate of soil removal. The most effective clean- ing agent for removal of soluble solids was observed to be distilled water since this system is controlled by the concentration gradient and distilled water cleaning solutions have the largest concentration gradient of the cleaning solutions used. . The presence of soluble solids in the cleaning solutions lowered substantially the rate of soil removal. However, for all concentra- tions of soluble solids used in distilled water and alkaline cleaning solutions, distilled water was the most effective cleaning agent at each of the soluble solids concentrations used. Proprietary cleaning compounds were tested and compared to 32 sodium.hydroxide cleaning solutions. The rate of soil removal was highest for the 32 sodimm hydroxide cleaning solutions; further analysis could not be made 39 40 because the additive effects in the proprietary cleaning compounds were not known. The rate of soil removal by 31 sodium hydroxide was markedly larger at the highest soluble solids concentrations. Tests using sugar plus milk soil showed that distilled water was again the more efficient cleaning agent as demonstrated by the sugar soil. This soil offered a greater resistance to removal at the high soluble solids cleaning solutions. The high viscosity of the high soluble solids cleaning solutions was evidently a deterrent in the removal of the in- soluble soil particles. The addition of sequestering agents in concentrations of 2.5, 5 and 102.by weight of caustic to 32 sodium.hydroxide cleaning solutions was studied to determine any variation in the rate of soil removal. The tests using trisodium phosphate indicated that the addition of this agent lowered the rate of soil removal. The solution containing the highest concentration of trisodimm phosphate exhibited the lowest rate of soil removal. The tests with tetrasodium pyrophosphate showed no effect on the rate of soil removal upon the addition of three concentra- tions of this sequestering agent. The means of the sodium gluconate tests indicated no significant effect for 5 and 10% concentrations of sodium gluconate in the cleaning solution. However, for the 2.52_sodiun gluconate cleaning solution there was a decrease in the rate of soil removal when the mean value was compared with that for the 31 sodium hydroxide cleaning solutions. The addition of EDTA sequestering agent in concentrations of 2.5 and 51 to the cleaning solution had no signifi- cant effect on the rate of soil removal. Addition of 10% of this sequestering agent enhanced the rate of soil removal by 162. In con- clusion, the experimental data indicate that the sequestering agents 41 used, with the exception of trisodium phosphate, have a very slight effect when used singly, on the rate of soil removal of 32 sodium hydroxide cleaning solutions. The variables of temperature, agitation, and time were also examined. Studies of distilled water and 3 and 51 sodium hydroxide cleaning solutions mixed with the four concentrations of soluble solids were made at 130, 150, and 170°F. The rate of soil removal was notably enhanced by an increase in the test temperature. The variable of agitation was studied using three agitator speeds. A higher rate of soil removal was demonstrated with the higher agitator speed. The tests were made at low agitation rates to eliminate irregularities caused by turbulence and thus assure reproducibility of the experimental results. The variable of time was found to be proportional to the amount of soil removed, however, due to large solution volumes and a negligible effect on the concentration gradient, the rate of soil removal was constant. The examination of the used comercial cleaning solutions with respect to soluble soil content did not give any definite criteria on when a cleaning solution becomes inoperative. From the experimental data, a very slight gain in soluble soils can be observed with use. This indicates that restitution of the losses of solution is sufficient to keep the soluble soil content from rising to a significant amount. The highest soluble soil content encountered in this test was 31 which would not affect the rate of soil removal significantly. CONCLUSIONS AND RECOMMENDATIONS The variables studied have a marked effect on the rate of soil removal of sugar soil. The results lead to the following conclusions. 1. Distilled water was the most effective soil removal agent ' tested. 2. An increase in the sodium hydroxide concentration decreases the rate of soil removal. 3. The presence of large concentrations of soluble solids other than caustic in the cleaning solution decrease the rate of soil removal substantially. 4. The three proprietary cleaning compounds tested were approxi- mately the same; however, the 3% sodium hydroxide cleaning solution ‘was a better soil remover. S. The effect of increasing the concentration of soluble solids not caustic in distilled water and 32 sodium hydroxide cleaning solutions was the same for sugar soil as for the sugar plus milk soil. 6. The addition of trisodium phosphate to the cleaning solution decreased the rate of soil removal. Trisodium phosphate has a greater effect on the rate of soil removal than increasing the sodium hydroxide concentration. 7. The addition of tetrasodium pyrophosphate to the cleaning solution has no effect on the rate of soil removal for the range of 0-102 as percent of caustic. 8. The addition of sodium gluconate sequestering agent has no noticable effect on the rate of soil removal except for the 2.5% as percent of caustic concentration which indicates a slight decrease in the rate of soil removal. 42 43 9. The addition of EDTA sequestering agent to the cleaning solu- tion has no effect on the rate of soil removal in concentration of 2.5% and 51 concentrations as percent of caustic. A concentration of 101 EDTA as percent of caustic increases the rate of soil removal significantly. 10. The rate of agitation of the cleaning solution affects the rate of soil removal with the greatest agitation having the largest rate of soil removal. 11. An increase in the test temperature significantly increased the rate of soil removal. 12. The amount of soil removed was proportional to the soil solution contact time; however, the rate of soil removal was constant. 13. The increase in the soluble solids (not caustic) concentra- tion of commercial cleaning solution is negligible. RECOMMATIONS The bottle washing Operation is dependent upon the variables of caustic concentration, temperature, time, agitation and additive or builder concentration. The data obtained in this study suggests that a more efficient cleaning operation than is now in use might be the use of a hot water soak before the bottles are immersed in the caustic solution. The hot water soak would remove the soluble soil in the bottles, and the caustic soak would be used mainly for sterilizing the bottles. To obtain cleaner bottles and insure the free rinsing qualities of the bottles, sequestering agents may be added to the water in the hot water soak tank. The use of-this soak would decrease the time necessary for the bottles to remain in the caustic tank because the bottles would then be essentially clean. The design of a different test model would facilitate obtaining mass transfer coefficients and diffusivities. For this experiment a system where a solution in laminar flow flows over a layer of sugar soil would eliminate the problem of a velocity gradient across the soil thus the mass transfer film thickness would remain constant. ‘The determination of the thickness of the mass transfer film'would yield the necessary data for calculating the mass transfer coefficients and diffusivities for the system. 2 min of soil removal mg/cm Rate 30 l J l l 0 5 10 15 20 1 Concentration neon Fig. 3 - Rate of soil removal of sugar soil with increasing NaOH concentration. 45 . f) c mg! nor min ¢ r8$UV&L 8011 10r- ~ Ya -.234X + 24.57 1 A1 1 0 2o 40 as Z Soluble solids Fig. 43 - Rate of soil removal of H 0 cleaning solution. 2 46 e of soil removal I‘- L‘- 30 in) U1 Ln 5% Nae-(2H Z Soluble solids Fig. 4- Rate of soil removal of H20, 32, 5% Neon. 47 min 2 Rate of soil removal mg/cm 20 Compound 1 ton-ound 3 Q 15 —' a 31 neon l l I l l 20 40 60 Z Soluble solids Fig. 5 - Rate of soil removal of 3 proprietary cleaning compounds using sugar soil. 48 min 2 l removal mgfcm 1 .5. ate of so" a.» A 30.. 3% Each Li? I 2 0 4 .9 6; {3 Z Soluble solids Rig. 6 - Rate of soil removal of milk soil with E20 and 31 Each cleaning solutions vs. sugar concentration. 49 soil ramawal mg! “3112 mi‘fi .e of Rat C) 20 H \H 2 Soluble smitda Fig. 7 - Rate of soil rammval of sugar sail iy distilled water s¢1uti¢ms at three temperatures. 50 soil remcvai 35 do 46 B_& r. (A) C.) 1 l 1 20 60 5.3L. 1 Soluble solids Fig. 8 - Rate 0f soil removal of sugar soil by 3% N303 salutimn at three temperatures. 51 25 - Bete of soil removal mg/cmz min 0 L J ! AL -+ J 20 4O 60 Z Soluble solids Fig. 9 - Rate of soil removal of sugar soil by 5% N803 solution at three temperatures. 52 Rate of soil removal mg/sm2 min I 0 35" 20 40 1 Soluble solids Fig.10 - Rate of soil removal with varying agitation. 53 GKL' o APPENDIX 1 Description of Cannercially Used Cleaning Solutions Saegle Nb. - A Age of solution Six (6) months Name of compound used Diversey Rely-0n Type of washer used Eeil - 8 wide Eunber of bottles washed to date 1,872,000 Frequency of make-up Weekly Amount of Caustic added at make-up Number of bottles washed between.meke-up 84,000 Type of labels used on bottles ACL * Capacity in gallons of Caustic Compartment Normality .276 Z Caustic 1.12 % Soluble solids 1.03% * - Applied Culer Label Sauple No. - B Age of solution Two (2) weeks (from No. l tank of washer) Kane of compound used Solvey - 761 Flake Caustic Soda Type of washer used Meyer Dumare - 16 wide Number of bottles washed to date 480,000 Frequency of make-up Daily Amount of Caustic added at make-up Made up with Caustic from No. 2 and No. 3 tasks Number of bottles washed between make-up 48,000 54 55 Type of labels used on bottles Paper Capacity is gallons of Cs.ustic Compartment 951 normality .84 Z Caustic 3.3” 1 Scluble solids 2 4% Sample No. - C AMe of solution Four (4) months Rene cf compound used Economics Laboratory, Inc. 3W - 61 - 1 Type of washer used Ladewig - 28 wide member of bottles washed to date 64,000,000 "requemey of make-up Every other bottling day &J.ouet of c.1ustis add ed at make-up 200 - 300 lbs. Nuaber of bottles mashed between meke-up 2,000,000 ”toe of labels used on bottles AOL * Ce s city in g;ellons of Caustic Campsrttemt 3,500 ”unlit" .9? % Cerstie 3..02 % So; his ids 2.5% *- Agpliei 0010: Label Sample NO. - D Ag;e of solution Five (5) months (takzn from.No. 2 compertmeut of sesEer) Name of compound used Diversey - Glo-Tak Type of washer used Barryfiweheiller - 40 wide 56 Number of bottles washed to date 29,160,000 Frequency of make-up Automatic makeuup equipment used Amount of Caustic added at make-up Number of bottles washed between make-up 1400-1500 bottles washed for each pound of make-up, Caustic Type of labels used on bottles ACL * Capacity in gallons of Caustic Compartment Normality .80 % Caustic 2.801 1 Soluble solids 3.2% * - Applied Color Label sample IJOQ " E Age of solution Five (5) months (taken from No. 2 compartment of washer) Name of compound used Diversey - Glo-Tak Type of washer used BarrySWehmiller - 28 wide Number of bottles washed to date 22,680,000 Frequency of make-up Automatic caustic strength control system Amount of Caustic added at make-up One pound/l400-1500 bottles Number of bottles washed between make-up Type of labels used on bottles ACL * Capacity in gallons of Caustic Compartment Normality 1.03 Z Caustic 4.102 X Soluble solids 2.0% * - Applied Color Label Sample No. - F Age of solution Eleven (11) months Name of compound used Diversey Spec-Tak-IOOO Type of washer used Ladewig - 24 wide Number of bottles washed to date 10,500,000 Frequency of make-up Daily Amount of Caustic added at make-up 30 lbso Number of bottles washed between make-up 48,000 Type of labels used on bottles ACL * Capacity in gallons of Caustic Compartment 2140 Normality 1.21 2 Caustic 4.8% Z Soluble solids 3.5% * - Applied Color Label Sample No. - 0 age of solution Twenty (20) months (taken from No. 3 tank of 6 tank washers) Name of compound used Diversey a Spec-Tak—lOOO Type of washer used buyerabumore - 24 wide Number of bottles washed to date 73,920,000 (402 are non-returnable bottles) Frequency of make-up made up with Caustic from No. l compartment Amount of Caustic added at make~up Number of bottles washed between make-up 168,000 Type of labels used on bottles ACL * Capacity in gallons of Caustic Compartment 922 58 Normality 1. 38 I Caustic 5.501 I Soluble solids 3.52 * - Applied Color Label Sample No. - H Age of solution Twenty (20) months (taken from No. 3 tank of S tank washer) Name of. compound used Divarsey - Spec-Iak-IOOO Type of washer used Meyer-human - 40 wide Number of bottles washed to date 126,000,000 Frequency of make-up Made up with Caustic from No. l compartment Amount of Caustic added at make-up Number of bottles washed between make-up 288,000 Type of labels used on bottles ACL * Capacity in gallons of Caustic Compartment 1490 Normality 1.04 2 Caustic 4. 201 Z Soluble solids 2. 22 * - Applied Color Label ample Ne. - 1 Age of solution One (1) day Name of compound used Diamond Alkali Hi-Test Alkali No. 3 Type of washer used ueyer-Dumora - 24 wide umber of bottles washed to date 56,000 (all non-returnable bottles) 59 Frequency of make-up Amount of Caustic added at mks-up Number of bottles washed between make-up Type of. labels used on bottles Paper Capacity in gallons of Caustic Compartment 1200 Normality .93 Z Caustic 3.7% 2 Soluble solids 2.3% Sample No. - J Age of solution Three (3) months Name of compound used Maryland Chemical Co. - Utopia Type of. washer used Ledewig - '28 wide Nmber of bottles washed to date 5,702,400 Frequency of make-up Daily Amount of Caustic added at mks-up 8 lbs. Number of bottles washed between make-up 86,400 Type of labels used on bottles AOL * Capacity in gallons of Caustic Canpartment 3,500 Normality .60 L 1 Caustic 2.402 1 Soluble solids 2.202 * - Applied Color Label 60 Sample Nb. - K Age of solution Three (3) months Name of compound used Sterling Chemical Co. CH-100 Type of washer used Ladewig - 24 wide Number of bottles washed to date 5,602,400 Frequency of make-up Daily Amount of Caustic added at make-up 8 lbs. NUmber of bottles washed between make-up 85,400 Type of labels used on bottles ACL * Capacity in gallons of Caustic Compartment 2140 Normality .61 Z Caustic 2.40% I Soluble solids 1.7% * - Applied Color Label BIBLIOGRAPHY Benson, W. 1'. 1956. The polyphospbate story. Proc. Third Ann. Meeting Soc. Soft Drink Technol., Washington, D. C. Bottle Washing Bulletin, 1893. Nordberg Manufacturing Co. Milwaukee, Wisconsin. Buchanan. J. 11., and H, Levine. 1939. Bottle washing and its control. Beverage Production and Plant Operation. 1958. American Battlers of Carbonated Beverages. Washington, D. C. Carpenter, Po Le 1961o HierObiologyo W. B. Saunders and COe Philadelphia. p. 432. Chaberek, J. and A. E. Martell. 1959. Organic sequestering agents. John Wiley and Sons. New York. p. 616. Dam. We Jo and to Jo Massey, Jre 1957a IntIOdUCtion to 'tatiatical analysis. HeGraw Hill Co.. Inc.. New York. p. 488. Dormth, L. C. 1956. Discussion of the longevity of bottle washing solutions including make up and caustic reclamation. Proc. Third Ann. Meeting Soft Drink Technol. Washington, D. C. Dungfelder, C. C. 1957. Importance of efficient rinsing in bottle washing equipment. Proc. Fourth Ann. Meeting Soc. Soft Drink TCChDOIo "38h1118tmp D. Go Jennings, W. C. 1959. Circulation cleaning Ill. The kinetics of a simple detergent systaa. l. Daig Sci” 42, 1763. Jennings, W. G. 1963. Detergency. J. Food Tech. 17 (7): 53. Jensen, J. M. 1946. Heasuring detergency functions as affected by various detergents and procedures against milk films by applica- tion of a mechanical washing apparatus. g. Daifl Sci. 29, 453. “um. 0o We. To Io Hedr1Ckp Io Jo Pflug, Go Go Phat]. and R. A. Keppeler. 1960a. Relative cleanability of various stain- less steel finishes after soiling with innoculated milk solids. 1. 239g §_c__1_. 43, 28. “um, as We; To 1o HCdrickp Io Jo Pflug find Go Go Pheil. 1960b. Relative cleanability of various finishes of stainless steel in farm bulk tanks. ,1. Milk Food Tech. 23, 377. Korab. H. B. 1964. Private communication. American Battlers of Carbonated Beverages. Washington, D. C. 61 62 Ladewig, A. 1955. Report of the bottle washing conmittee. Proc. Second Ann. Meeting Soc. Soft Drink Technol. Washington, D. C. Hartell, A. E. and M. Calvin. 1953. Chemistry of the metal chelate compounds. Prentice Hall. New York. McCalleon, J. J. 1961. Some chemical factors involved in bottle washing. Pros. Eighth Ann. Meeting Soc. Soft Drink Techno1., Washington, D. C. Nehltretter, C. L., B. 1!. Alexander, and C. B. Rist. 1953. Sequestra- tion by sugar acids. Ed. Ea. Chem, 45, 2782. Overman, 0. 1964. Private Communication. Cowles Chemical Co. Cleveland, Ohio. ' Parker. R. 3o. yo Re Elli-ken, Go To H.180“, Go A. Richardson and G. H. Wilster. 1953. Cleaning pipelines in place. Food Eng. 25 (1), 82. Pflus. Io Joy To Io Rodrick, 0o “- hum, Re A. Keppeler ‘nd C. C. Pheil. 1961. Studies on the deposition and removal of r‘d1NCt1VC 8011. lo Elk Food zeChe 24’ 390a Ruff, D. C. and x. Becker. 1955. Bottling and canning of beef. Siebel Publishing Co. Chicago, Illinois. Schwartz, A. 11., J. W. Perry and J. Berch. 1958. Surface active agents and detergents. Interscience Publishers, Inc. New York. p. 839. Shand, B. I. 1958. Glass Engineering Handbook. 1110er Hill Co., Inc. New YOtko Fe 482. Technical Bulletin No. 99. 1959. Pfizer gluconates in caustic bottle washing. Charles Pfizer and Co. New York. a 3. h ROG V I MICHIGAN STAT MN I I E U E 3 1193 0308 R I WWI IWITIWITI'ES 2 2 3 10