il l‘l‘M‘lM WWW WW WM W x ‘ 1 Ago—x IMO | (IND-[=- A COMPAMSON $5 FQUR FELTERING MAWRIALE USED EN TRééi‘iLffik'i} EELTERS {50R THE DiSi-Njfim. 0F BASE}? WASH Thesis for 1410 Dogma cf M. 3. MiCHéGAN STATE COLLEGa'E Fredrick 5314 Clay $950 1H“ . ., ‘ . u... :v 1’ , 1.; g,,,-.:.__ J‘Ahhx -__‘._.#- __‘L_.___—.-—_A___‘Ar \ :_1 This is to certify that the thesis entitled "A. Comparison of Four Filt ering Ihterials Used ; in Trickling Filters For me Disposal of Dairy Waste" presented by .. Frederick Earl Clay has been accepted towards fulfillment of the requirements for Jim—$.— degree in A311 cultural Engineering Major professor Date M 0-169 ——~_i ———— _. .V_ 4‘ A 4- . . ~ A COMPARISON OF FOUR FILTERING MATERIALS USED IN TRICKLING FILTERS FOR THE DISPOSAL OF DAIRY WASTE BY FREDRICK EARL CLAY A THESIS Submitted to the School of Graduate Studies of Michigan State College of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Agricultural Engineering 1950 THESIS TABLE OF CONTENTS Page INTRODUCTION...........................................l REVIEW OF LITERATURE...................................3 Preventing the Waste..............................3 Treating the Waste................................h STATEMENT OF THE PROBLEM...............................9 APPARATUS.............................................10 Filters..........................................lO Pumps.................................,..........12 Filtering materials..............................l5 EXPERIMENTAL PROCEDURE................................15 DISCUSSION OF ESULTS..........t......................18 CONCLUSIONS...........................................26 BIBLIOGRAPHY..........................................27 APPENDIX..............................................30 ACKI‘IOWLEDGELENTS......................................33 239279 INTRODUCTION The dairy industry although not the most valuable economically, is probably the most widely scattered industry creating waste disposal problems. It is estimated that between 115 and 116 billion pounds of milk are produced annually in the United States, the production being scattered throughout every state, but the seasonal character of the industry concentrates the waste dispoSal problem in the summer months. The trend in the dairy industry is for a greater portion of an increased milk production to be handled in larger plants equipped for better utilization of by-products. This increase in size leads to a greater concentration of unavoidable losses and so increases the industries waste problem. Investigation of milk plants indicates that even with careful handling, 'from 0.3% to 1% of the whole milk is lost. At the present time, many of the individual dairy plants are located along streams, and the method of diSposal is merely to allow the sewer pipes from the plant to empty into the stream. This method is satisfactory as long as the flow of the stream is great enough to prOperly dilute the sewage. However, if it is a small stream in comparision to the amount of sewage it is supposed to receive, the bacterial action on the organic material of the waste will deplete the oxygen supply of the stream. This has two detrimental effects. One is the continued action of bacteria under anerobic conditions, i.e., decomposition in the absence of air. The by-products of such decomposition are methane, ammonia and other foul smelling gases. Obviously, land owners along the stream become alarmed and try to eliminate this condition. The other effect is that after the oxygen supply has been de- pleted animal and plant life which depend on the oxygen supply of the stream are suffocated. The question that usually arises is, FWhy don't the fish and animal life of the stream.use the organic material as food?" The answer is that they do use it; however, they can not use it all and the effects mentioned above take place. Deanlo states that in the last congress, at least four bills on the subject of waste disposal and stream pollution control were introduced. Wieters35, Assistant Chief of the Division of water Pollution Control, U. 3. Public Health Service, states; "Generally speaking the wastes from dairy plants constitute a serious industrial waste problem. The milk solids decompose rapidly in water and, unless the receiving streams provide adequate dilution, the oxygen resources of the streams may be exhausted and foul putrefactive conditions result. For this reason, State water pollution control agencies are very much con- cerned with reSpect to discharges from dairy plants." Even though plants may be fortunate enough to be able to discharge their waste to a municipal sewage dis- posal system, they may still be sued together with the municipality, if the municipal plant is Operated in such a way that it causes a nuisance. There have been several cases where state agencies or consulting engineers have caused a municipality to discontinue treating waste from a dairy plant. The reason for this was increased pOpulation in the municipality or increased waste from the plant, or both, which results in over-loading the municipal plant. It is obvious, therefore, that definite action must be taken by dairy plant Operators in order to avoid future restrictions which would be costly to say the least. In many cases, the cost of constructing a treatment plant would exceed the cost of the manufacturing plant and many small Operators would be forced out of business. REVIEW OF LITERATURE H. A. Trebler29 of Sealtest Research Laboratories makes the following statement in a discussion of dairy waste: "It can not be repeated too Often that the all- important factor in the elimination of dairy-waste h nuisances is the elimination of the wastes in the plants themselves." The most important step is waste prevention follwed by waste saving and waste diSposal. Deanlo points out that management should adOpt and carry out a definite waste saving program. Management should determine where losses occur and take steps to modify or replace unsatisfactory equipment. Frequently, strict waste prevention carried out by every employee in the plant will reduce the strength and quantity of the waste to a point acceptable to the state authorities for discharge into a stream or the municipal treatment plant. At the present time, whey, buttermilk, and skim milk, should not be considered as waste products, because no company can afford to treat quantities of these materials in a waste disposal plant.’+ As an examplezz, a cheese factory with a milk intake of 50,000 pounds and a normal loss of 2% and no dumping of whey would Spend about $22,000 for a treatment plant. Since whey is one of the most difficult waste products to treat, it is apparent that if the whey is allowed to go through the treatment plant, the cost would be prohibitive. Therefore some other means must be found to dispose of these by—products and spdfled products. After all of the prevention and saving methods have been put into Operation and there is need for further treatment, a method Of treatment must be selected. The treatment may vary from simple aeration and equalization of flow to a somewhat complex and complete treatment by biOlogical oxidation. Equalization of flow consists of having a tank large enough to hold about 503 Of the days total discharge to the stream or sewer. The waste collected in this tank is aerated and disch rged to the stream in a steady flow so that the stream or sewers need not be burdened with a sudden overload and then receive nothing for several hours. In many cases, this is all the treatment that need be given to the waste. If further treatment is needed, there are a number of methods which can be employed. One method of treatment is known as the activated sludge process.8 This type Of treatment plant normally consists Of a holding tank, one or more aeration tanks and a final sedimentation tank. Aeration is by diffused air from blowers, or by mechanical agitation. In the activated sludge process, an accumulation of aerobic organisms with entrained organic matter develops in the aeration tanks and settles out in the final sedimentation tank. Some of the "activated sludge" is returned to the influent of the aerators to seed the incoming sewage. The activated sludge process offers promising possibilities; however, the cost of the treatment plantand 6 the need for a qualified supervisor limits its use. Another method is the chemical precipitation process. This consists of adding coagulating chemicals such as aluminum sulfate or iron sulfate(with lime) to the waste to precipitate the proteins. This gives a clear effluent, but since the milk sugar is not precipitated the oxygen demand reduction is usually not good enough for complete treatment. Barritt3states that chemical precipitation is not feasible because there is no practical method for removing sugars, and to remove the protein alone makes the sugars more difficult to purify than before. Oxygen demand reduction is not more than 50% although one investigator26 claimed reductions of 70 to 80%. The Operation of such a plant requires that the Operator have a fair understanding of chemistry. There are two methods of disposal which have been patented. They are the GuggenheimProcess12 and the Mallory Process 1,12,1h. The first process consists of chemical precipitation combined with the activated sludge process. It is more expensive and somewhat more complicated than the other methOds. The Mallory process is, in brief, a conventional activated sludge process under ‘ precise mathematical design and control. All the guess work of Operation is removed when using this process; however the cost is high and not many units have been installed. The most common type of treatment plant in Operation today is the trickling filter.31 A trickling filter is a bed of crushed stone, gravel, or slag Of relatively large size to which settled sewage is applied by Sprinkling or otherwise on the surface. The conventional beds are at least 5 ft. deep. The applied sewage trickles in a thin film over the surfaces of the filtering medium which have become coated with a zoogleal film. Fine suspended solids are held'by the film and colloidal material is adsorbed byit. Since air is present in the filter a large pOpulation Of aerobic bacteria will inhabit the film and work upon the suspended, colloidal and dissolved organic solids to bring about a reduction of oxygen demand. Theoretically, irregularity of the filtering media is desirable as providing more surface area on which the film can collect and increase bacterial action. However, for maximum efficiency the film should slough Off regularly, that isrthe filter is said to "unload". From this it is apparent that the filtering media dare not be too rough or the filter will clog up and the Operation will become useless. 8 'In the last few years more emphasis has been put on the recirculating filter1’12!l5!1892h. If the ordinary trickler is preceded by a tank which serves to equalize the flow and may also receive returned filter effluent, and if the recirculation is extended to a point where continuous dosage results; the trickler is then a high rate recirculating trickling filter. The recirculating filter provides a simple, efficient, economical, and comparatively flexible method for the treatment of many organic industrial wastes. It is simple to Operate since the control largely involves the mechanical care Of the equipment and the removal Of solids from the settling tank. It is efficient if prOperly designed and Operated and not overloaded for any appreciable duration beyond its capacity. It is flexible, at least to some extent, since by increased recirculation a higher degree of removal may be Obtained. Above all, it is dependa- ble, not subject to complete breakdown Of process when subjected to overload, and is quick to recover efficiency after the overloaded condition has been eliminated. The cost Of this type of filter is about 1/3 the cost of the standard filter. One great difficulty in Obtaining good performance and reliable design data is that all available dairy waste diaposal plants have a tremendous variation in load and flow from hour to hour, day to day, and season to season. NO plant has found it economically possible to run daily analysis on strength. Results of existing installations which have been printed in the literature are based on grab sampling or even continuous sampling for only a few days. The missing data regarding the most economical capacity and type of construction of tricklers, the rate of application which is more desirable, the size of the filtering media, the type of filtering media, etc., is very sorely needed by the industry. Eldridge states that "the size and type of stone is one of the most important items which influence the Operation of the filter." Trebler points out that dairy wastes, in particular, require a coarse filter material in order to avoid clogging. At the same time plenty of surface area is required for biological growth. He eXperimented with empty evaporated milk cans and found that they were a fair media. The biological coating seemed to protect the cans from rust so that they held up remarkably well. Steel lists the media used in order Of their pOpularity as stone, slag, coke, or Cinders, and gravel. STATEMENT OF THE PROBLEM With the need for information on relative efficiencies 10 of the various types Of filtering media so apparent, it was decided to construct a small trickling filter and use a synthetic waste under laboratory conditions. The fact that the waste would have to be synthetic, since it would be impossible to duplicate the average dairy waste, and that the filters would be Operated in a closed building so that they would not be eXposed to varying wheather elements, was given consideration. It was decided that even though the results might not give a true indication Of what would happen under actual conditions, the results would be of value. APPARATUS Four separate filters were constructed as shown in Fig.1. Two 55 gallon oil drums were cut in half to serve as equalizing and sedimentation tanks. Two pieces of tile were set over each of the half drums so that the synthetic waste could be applied at the tOp and allowed to trickle down through the filtering media and empty back into the half drums to be recirculated. The tile used was sewer tile and each piece was 3 feet long and 12 inches in diameter. The support for the tile was made out of 2"X6" lumber and the filtering media was held in the tile by means Of a false bottom. 11 Fig. l Twenty-two gallons of water were added to each tank daily. With this amount of water in the half drums, a space of about two inches was left between the bottom of the tile and the water surface. This amount of Space was considered ample to allow a natural circulation of air through the filter. The two pieces of tile were not sealed at the point of contact so that air could enter at this point if necessary. At first, the sewage which was trickling through the filter leaked out at this point 12 and ran down the outside of the second tile. But after a week of Operation, this leak stepped and all of the sewage trickled through the entire depth of the filter. Since it was considered necessary to keep a record of the temperature of the sewage, holes for thermometer wells were drilled about three inches from the bottom of the t0p piece of tile. A piece of 1/4“ diameter pipe about 10 inches long was inserted into the hole so that the temperature at the center of the filter could be obtained. Glass stemmed thermometers were inserted inside the pipe and the temperature of the filter bed was recorded. A record was also kept of the temperatures in the tank. In order to evenly distribute the sewage over the tOp of the filter, a can with several holes drilled in its bottom was used as a receptacle at the tep. An attempt was make to use Spray nozzles to distribute the sewage over the tep; however, with the low volumes required it was found that the nozZle tip was too small for this type of use and clogged up frequently. This clogging resulted in a continually changing rate of dosage. The recommended dosage rate for a recirculating filter is 350 gallon per square foot per day as a minimum. It was difficult to find pumps with a capacity low enough to give anything near this recommended minimum dosage, and still be reasonable in cost. Four 13 bronze gear pumps were finally obtained and the rpm's Of the pump suitably reduced by means of pulleys. A more accurate means Of controlling the rate Of flow was used by placing globe valves in the pipe line after the pump. A check on the rate Of flow was made frequently and all pumps standardized by means Of Opening or closing the valves. Unfortunately these gear pumps could not maintain the suction pressure needed to pump 350 gallon per square foot per day. When the rpmfls of the pumps were reduced to a point where the discharge was 350 gallons per square foot per day, they rarely picked up the sewage without being primed and even after priming they could not be depended upon to deliver the required amount for 2 4 hours ‘without losing their suction. Therefore it was necessary to speed them up to a rate Of dosage of 500 gallon per square foot per day. A main drive shaft was constructed and one one-half horsepower electric motor was used to Operate all four pumps. (Fig.2) Another pump was set up to run considerably faster than the other four pumps. The sole purpose Of this pump was to empty the circulated sewage from the tanks after the 2A hour period. This was necessary in order that fresh water could be added daily along with the milk sample and all filters start the 24 hour run with sewage Of the Fig. 2 same strength. Also during the 2k hour period, an accumulation of sloughed-Off zoogleal film will settle to the bottom Of the tank and this should be removed. The pump used to empty the tanks can be seen in the center Of the table in Fig. 2. The motor which was used to drive the four circulating pumps had a shaft on both ends and was placed on the table so that it could be made to drive the fifth pump which was used for emptying the tanks. At the end Of the 2A hour circulation period, the main drive shaft for the four circulating pumps was disconnected and the fifth pump was connected for Operation. 15 By the prOper selection of pulleys and with this quick change over procedure, the tanks could be emptied and refilled in about twenty minutes. Four different filtering materials were used. They were coke, Cinders, granite rock, and corn cobs. The coke and granite rock were about 3 inches in diameter. The Cinders were somewhat smaller-about 2 inches in diameter. The granite rock was used because Of its accepted ability in practice. The other three materials were used because they were porous, had considerable surface area and/or they were cheap. The depth of each filter bed was about 5% feet. \ EXPERIMENTAL PROCEDURE After all Of the apparatus had been assembled and the rate of dosage standardized, the filters were put into Operation. NO reduction can be eXpected at first, at least not an acceptable reduction, until the develOpment Of the zoogleal film has reached its Optimum state Of growth. This usually requires about two to three weeks. In this study, the time for the filters to reach a_point of acceptable reduction and maintain this reduction was about four weeks. TO find the time when the reducing ability of the filters had reached an acceptable per cent Of about 80, 16 periodic checks were made on the influent and effluent. The amount Of milk added tO the 22 gallons Of water in the tanks each day during this "inoculation" period was 500 ml. After the periodic tests indicated that the filters were Operating at their peak efficiencies, a series Of daily tests was started. Each day the effluent was pumped.out Of the half drums, the drums rinsed out, the rinsings pumped out, 22 gallons Of fresh water added, and a definite amount Of milk put into each tank. Each tank was agitated sufficiently so that the milk and water became thoroughly mixed and a sample Of the resulting mixture was taken in a test tube. The circulating pumps were then started and the synthetic waste was allowed to circulate for 2h hours. Meanwhile, the test tube filled with a sample of the influent was taken into the testing laborata‘y and allowed to stand for a period of 2 hours. This holding period served as a sedimentation period and all Of the heavy particles settled to the bottom Of the tube. At the end of the two hour period, the prOper dilution Of the sample was made and the sample incubated for 5 days. After the original sewage had circulated for 2h hours, a sample was taken Of the final effluent, and the 17 same procedure followed, except for the percent Of dilution, as was followed in incubating the sample of original sewage. Since the number Of samples to be incubated for a 5 day period was large, 8 samples per day (duplicate determinations were made on each sample) plus a water blank, the size of the incubator was necessarily large. For this work, a farm type cream cooler was used. It was a two can size cooler and maintained a temperature range of 65° to 70°F. Which was considered accurate enough for this study. ‘ Five different concentrations of synthetic waste were used. (The milk product used for this synthetic waste was pasteurized skimmilk.) Each Of these concentrations was applied for four separate days to get an accurate picture of what each type Of filtering material was capable Of doing. .The concentrations used were 9h6 ml (one quart), 750 m1, 500ml, L00ml, and 250ml. This amount Of milk was diluted with 22 gallons Of water. Another common method of eXpressing concentrations in sewage disposal work is cubic feet Of filtering material per pound Of BOD per day. Using the accepted BOD of 73,000 ppm for skimmilk and converting the millilicensof milk in 22 gallons Of water to this eXpression, the results are 18 30 cu. ft./lb. of BOD/day for the 946 m1 dilution, 37.8 cu. ft./1b. or BOD/day for the 750ml dilution, etc. Pounds of 30D may be Obtained in the following way: ml Of milk X 52. 61‘. of milk X 73000 I grams of BOD rams of 1300‘ = lbs. of BOD 453 grams7Ib. A common value for design purposes is 80 cu. ft./ 1b. Of BOD/day. In this investigation, this value was exceeded in both directions in order to give an indication Of what would happen at the extreme limits. DISCUSSION OF RESULTS The sloughing action of the filters is Of a periodic nature and this action may perhaps explain the fluctuation Of the curves in Figs. 3, h, and 5. This sloughing action Of the filters deposits dead organic material in the tanks. Most of this organic matter settles to the bottom Of the tank, and thus the need for a sedimentation tank and a sedimentation period; however some Of it goes into solution and increases the BOD Of the sewage. An ideal filtering material would "shed" its zoogleal film uniformily from day to day and eliminate this fluctuation. The percent Of reducticn as shown by the e graphs is mUch higher than could be expected in practice. The 100; reduction, which was Obtained in several instances, is 19 n F“. toned .59.. ¢N coco are one of. to 3005 983 wcofiazfifiwz .votom SO... VN 9.0 3.55912 *3... xuvu EUDEDZ womb. _¢. 3 w A m d : VO 1 0 ON 0300 .m m o? m. P afloo m .m cm .0 so U A \mcotw Wm axou ow “Eat . a . .3 LUflSO ~ x _ ivonmw On: seesaw €985. east... :on gamssses assertom 0.: .2. mu .. a... s .2 on... 0.: as mm . $5.5 is was 20' .v. m, N u. .V m .v 9.... worst So: VN rune c6 tam of .6 once: we; 30:32??an .OOIQL So: .VN $20 macwmoikmk 45mm. Loom LWQEDZ +mw._l N u. es by 00 c ow w. 38 \ [0N _ m£0+n \ f m IL“ J“ / Om 1w O OO cm .5350 / U x . d 9 .mp5... afoxou CON Du [up (K 09 338 asses west; re m9. aee38£es§ 2...... s .t mom of .9... MN. .25 .6? cos 0.: new mm . i=5. Econ 21 m .m; 60on .50: «N room we saw or: +0 opsE mica, mcofiszc 2&2 .8.le .50; ¢m wco ficwmwflm. v.03 Loom .3532 sea. '4‘ (T) N i s 9 43 63> 11483 43d - uoy+3npaa w V\ .VIIH 01.00 . . LwUCDMI H g mop\oom £ \utmem matte z; m: of .8 tom mm + «:5 .5253 22 practically an impossibility under actual conditions. The Optimum reduction is around 86;. One investigator strted thrt tile smaller *he filter the lower the reduction. The results obtained from this investigation would not bear out this statement; There are several explanations why the percent of reduction which was Obtained in this investigation is higher than the Optimum Of 86%. They are: 1. The test was conducted under laboratory conditions. This means that the temperature was held almost constant, the filter was not eXposed to the varying weather elements, and the composition Of the waste was practically constant. 2. The entire twenty-two gallons of sewage was recirculated for the 2b hour period, and this would not be true under actual conditions. Under actual conditions the sewage is recirculated as many times as possible; however, during the working part of the day, new waste is continually being added to the equalizing tank which means that some of the waste in the tank will overflow and leave the treatment plant. 3. The rate of application or the recirculation rate was considerably higher than the recommended minimum. However, this recommended minimum is not a standard rate, but it is a minimum rate under which the per cent Of reduction would probably be tOO low to be acceptable. A. There was no sudden change of concentration. The same concentration was applied for four days and then the concentration was changed. However, the change was only a slight one and it is certain that filters used in practice receive greater shocks than this. 23 The factor mentioned in number 1, composition of sewage, is a very important one. In a dairy plant, sour milk and alkaline washing powders are dumped to the sewers, and since bacteria are very sensitive to changes in pH, the acidity or alkalinity is very likely to effect their ability to decompose the milk solids. The pH factor is the important one in the diSposal of cheese whey, since whey has a rather high acidity. However, all Of the factors mentioned above were considered before this investigation was started. To Obtain information on a typical dairy waste would require at least two years of continuous sampling since the concentration Of the waste varies from day to day and season to season. Therefore, since all filters were to be Operating under the same conditions Of temperature, concen- tration, and rate of dosage, it was assumed that a reasonable comparison could be made. Inspection Of the graphs indicate that the four filtering materials can be rated cinder, coke, stone, and corn cobs as best, second best, third best, and poorest reSpectively. The efficiency of the corn cobs was very low and the use of corn cobs as filtering material would be doubt- ful to say the least. The low efficiency Of the corn cobs might be eXplained 24 by the fact that they are chemically unstable. When they become saturated with water, certain changes take place which add to the strength Of the sewage. The chemical reactions which occur here are not important as far as this work is concerned. The important fact is that they cannot by used with any degree Of success. The corn cobs also had a very Offensive Odor after they had become thoroughly saturated, which is another reason for not using them. The ranking of the other three materials might be ex- plained on the basis of their available surface area. Since all three of the remaining materials were chemically stable and by the time the tests were started, they had been eXposed to the synthetic waste for a period Of at least one month so that no adsorption or absorption could take place, the only remaining eXplanation seems to be the amount and type of surface area. The eXposed surface of the granite rock was quite smooth and fairly regular; that is, no sharp corners, crevices, or bends of any kind were existent on the surface. The surface of the coke was also fairly regular, but it was very porous and had a great number of holes present. At the end of a month, however, most of the holes had been filled with the dead organic matter and for all practical purposes the coke surface resembled that of the stone. 25 The surface Of the cinder was irregular and porous. Therefore the only significant difference between the surface areas of the three materials after a month of use was the irregularity of the Cinders, and this difference must account for the consistently better results which the Cinders gave. The greatest Objection to a rough surface seems tO be that it would inhibit the sloughing action of the filter. However, it seems logical that the Opposite of this would be true. If the surface were irregular, which would result in more turbulent flow, it would aid the sloughing action by not allowing the thickness of the zoogleal film to become too great and by unloading or warhing the sloughed off film out of the_filter more quickly and more regularly. The basis for the preceding statement was apparent at the end of the 2b hour run. When the tanks were pumped out, the settled mass at the bottom of the tanks consisted of larger lumps in the case of the stone filter as compared to the cinder filter, thus indicating more irregular cleaning. 26 CONCLUSIONS 1. The filtering materials rank Cinders, coke, stones, and corn cobs as best, second best, third best,and poorest respectively. 2. Corn cobs are unsuitable as a filtering material. They could undoubtedly be improved upon if they were seasoned and then thoroughly saturated before use. 3. The nature of the surface area is the most important factor to consider in selecting a filtering material. Some irregularity is definitely desirable to insure a more regular sloughing action. A. The trickling filter is the best method to use in disposing of dairy plant waste because it is: 1. Simple to Operate. 2. Efficient. 3. Capable Of responding quickly to fluctuations in loading. h. Requires very little attention. 5. Low in first cost. 5. The problem of disposing of the waste from many plants can be entirely eliminated by giving strict attention to preventing the loss of the milk solids down the sewers. 27 BIBLIOGRAPHY l. 9. 10. ll. 12. 13. ‘A Critical Review of the Literature of 19A? on Sewage and Waste Treatment and Stream Pollution. Sewage Works Jour. 20:22h-272. 1948 Babbitt, H. E. Sewage and Sewage Treatment. hth ed. London, John Wiley and Sons, Inc. 1932 Barritt, H. W. The Purification Of Dairy Waters. Sewage Purification 1:118. March 1939 Bloodegood,Don E. Milk Waste DiSposal. Sewage Works Jour. 20:695. 19h8 Buswell, A. M. Treatment Of Milk Waste. Iilk Plant Monthly 26:28-30. January 1937 Buswell, A. M., Boruff, C. S., and Wiesman, C. K. Anerobic Stabilization of Milk Waste. Ind. Eng. Chem. 2h:1h23-1A25. 1932 Charles, C. Agar. Practical Methods Of Preventing Dairy Waste Nuisance. Sewage Works Jour. 10:115 1938 Cavanaugh, G. W. Dairy Waste Saving and Disposal. Proc. Intern. Assoc. Milk Dealers, Plant ~ Section. 131-137. 1935 Davy, P. S. How Much Money Are You Pouring Down the Drain? National Butter and Cheese Jour. 36-37. June l9h8 Dean, W. A. Jr. Dairy Waste Saving and DiSpOsal. Milk Plant Monthly 36:h2-50. May 19h7 Elder, A. L. and Stahl, G. W. Ferric Oxide Hydrosols as Coagulants in Waste Treatment, Ind. Eng Chem. 31:925. 1939 Eldridge, E. F. Industrial Waste Treatment Practice. 1st ed. New York, MoGraw Hill Book Company Inc. 19h2 Eldridge, E. F. The Disposal Of Wastes From Milk Products Plants. Mich. State Agr. EXP. Sta. Bull. 272. 14. 15. 18. 19. 20. 21. 22. 23. 2A. 25. 26. 28 Eldridge, E. F. Milk Waste Treatment by the Mallory Process. Water Works and Sewage Jour. 88:h57-h63. 1941 Eldridge, E. F. The Recirculating High Rate Filter In Industrial Waste Treatment. Hater Horks and Sewage Jour. 88:483-A90. l9hl Fleming, J. R. Water Supply and Sewage DiSposal for Dairies. The Milk Dealer. 52-5h. 19Al Halverson, H. 0. AerO Filtration of Sewage and Industrial Wastes. Water Works and Sewage 83:307-313. 1936 Imhoff, Karl. The Design Of High Rate Trickling Filters. Sewage Works Jour. 18:89-92. 19A6 Jones, Frank W. The Trickling Filter and its Operation. Sewage Works Jour. 18:705-711. 1938 Knechtges, 0. J. A Study Of Trickline Filter Loadings. Sewage Works Jour. 10:93 -938. 1938 Levine, Max and SOppeland, Lulu. Bacteria in Creamery Wastes. Iowa State Eng. Exp. Sta. Bull. 77. 1932 McKee, Frank J. Cheese Factory Waste Disposal National Butter and Cheese Jour. 38. Feb. l9h8 Metcalf, L. and Eddy, H. P. American Sewage Pratice. 2nd8ed. New York, McGraw Hill Book Company 192 Oeming, Loring F. Milk Waste Treatment at the Detroit Creamery Company Condensery. Ovid Mich. Sewage Works Jour. 20:512-51h. 19A8 Sandborg, 0. W. and Stiffen, H. L. Waste Saving and . DiSposal, American Butter Review 9:40-A3. June 19h? Southgate, B. A. Treatment Of Milk Washings By Addition Of Coagulants, Sedimentation, and Biological Filtration. Dairy Industries (Brit.) March 1948 27. 28. 29. 30. 31. 320 33. 3h. 35. 360 37. 29 Steel, E. W. Water Supply and Sewage. 2nd ed. NeW'York, Trebler, H. Trebler, H. Trebler, H. Warrick, L. McGraW'Hill Book Company Inc. 1947 A. and Harding, H. G. Dairy Waste DiSposal. A.,Ernsperger, R. P., and Roland, C. T. Dairy Waste Elimination and Sewage Disposal. Sewage Works Jour. 10:868. 1938 A. and Harding, H. R. Industrial Wastes in the Dairy Industry. Industrial and Eng. Chem. 608-613. May 19h? F. New Developments in Treatment of Milk Wastes. Food Ind. 12:46-h8. Sept. l9h0 Washington Pollution Control Commission Suggestions for Waste PreVention. National Butter and Cheese Jour. 3h-35. July 19h9 Webb, Byron H. and Whittier, E. O. The Utilization of Whey. Jour. Dairy Science 31:139-16h Feb. l9h8 Whittier, E. O. and Rodgers, L. A. Continuous Wieters, H. Fermentation in the Production of Lactic Acid. Ind. Eng. Chem. 23:532-53h. 1931 H. Public Health Service, Washington, 25 D. C. Dairy Plant Wastes-Private Communication- 19h9 Wittmer, Earl F. Alternating Two-Stage Filtration of Sewage and Milk Waste. Sewage Works Jour. 20:846-8h8. l9h8 Young, Lewis A. Waste DiSposal Problems of Milk Products Plants. Milk Plant Monthly 31:28-30. January 19h2 30 APPENDIX A BOD In any discussion of a sewage treatment plant, the term BOD is frequently used. These letters stand for "Biological Oxygen Demand" or "Biochemical Oxygen Demand". Almost all industrial wastes, like domestic wastes, contain organic matter which is not in its most stable condition. This organic matter consists of complex chemical compounds which require oxidation in order to be reduced to their most stable form. The term BOD is therefore a numerical value which indicates how much oxygen is required to stabilize the waste. Complete stabilization requires more than 100 days at 20°C., but such long periods of incubation are impracticable. Consequently a much shorter period of incubation is used. Incubation for l, 2, 5, 10, or 20 days at 20°C. is customary, and the 5 day BOD is a standard which is used by most workers. Conversion of the data from one incubation period to another or from one temperature to another, may be approximated. The procedure for obtaining the numerical value for the BOD consists of diluting the sewage sample in prOperly prepared dilution water. The dilution water is made by taking-a good grade of distilled water which has no oxygen demand of its own and which is not germicidal, and dissolving certain recommended chemicals in it. The dilution water is incubated and the temperature maintained at 2000, for 5 days, 31 At the end of the five day period the sample is removed and the amount of dissolved oxygen in the sample is determined. By running a blank bottle, or a bottle with no sewage added to it, at the same time; the difference in the amount of dissolved oxygen in the two samples can be determined and this determination will give the amount of oxygen required to stabilize the sewage. APPENDIX B It was the intent of the author to include in this report a section for dairy plant owners who would like to calculate their own requirements as to the size of a treat- ment plant or to determine whether a treatment plant would be required for their particular plant. However, there are a great many factors which would have to be considered and an evaluation of these factors should be made by an experienced engineer. To give some idea of the complexity of the problem, some of the factors will be given here. They are: 1. An accurate account must be made of the amount of milk lost down the drain or sewer. This can best be done by installing a continuous sampling device in the sewer outlet of the plant. Using this device, the waste from the plant should be sampled for aslong a period as practical. 32 If it is found that some means of treatment will be required, an eXperienced engineer may be able to reduce the cost of the treatment plant by by- passing some of the cooling water to reduce the volume of waste handled, or he may be able to reduce the rate of discharge to the stream by means of equalizing tanks. There are a great number of steps that can be taken, but it would require an eXperienced engineer to pick the cheapest and most reliable to fit the situation. If a stream flows near by and it is convienient to empty the sewage into the stream, a stream pollution survey would have to be made. This would require a determination of the rate of flow and of the amount of dissolved oxygen present in the stream. Permission would also have to be obtained from the prOper authority to dispose of the sewage in any particular stream. If a competent and recognized engineer has surveyed the plant, permission will undoubtedly be obtained more easily. It would be practically impossible for a layman to design his own treatment plant, such as motor and pump sizes, rate of flow to use, etc. 33 ACKNOWLEDGEMENTS The author wishes to express appreciation to Dr. Walter M. Carleton and Professor A. W. Farrall for the many helpful suggestions and guidance during the course of this work. ' 0'- g-. hail u‘Lbun.