K VALUES FOR TRICKUNG FILTERS “Nth for flu but“ of M. S. MICHIGAN STATE UNIVERSITY km: E. Germain 1963 {HIIIC LIBRARY Michigan State University Q nationally the that entitled K VALUES FOR TRICK'LING FILTERS presented by James E. Germain .‘fiA'. ‘ ., ‘ ' ' ' -' 979..., has been accepted towards fulfillment “ ~ . -: 3| r- 4“; . of the reqmrements for M. S. degree in Sanitarz Engineering m..-“ ..' “ma. -._ : -. U - 1 . I l . . ‘ I r .. - I n I O ABSTRACT K VALUES FOR.THICKLING FILTERS by James E. Germain This thesis contains an analysis of the value "K" in - _KD/Q2/3 the expression p - 10 as determined from Operational data on trickling filters, where p = Fraction of BOD remaining D 2 Filter depth in feet Q = Hydraulic load in MGAD K 3 Constant Data from 19 Michigan rock trickling filter plants were analyzed and the K values determined. Several of these plants were analyzed on a daily basis to show the normal fluctuation of K at a single plant. K values were also determined for rock trickling filter plants in other regions of the country with different climatic conditions. In addition pilot plant and full scale plant data were analyzed which indicate that in a rock filter the K value decreases with increasing depth. In contrast the analysis of a considerable amount of data from trickling filter plants using open type plastic media demonstrated that in these cases K was constant and independent of depth. ;3 .. solar ULQLhLJ aiezleas SQ‘ ;ee*f ';; aux: a aneuoeb .. zeJi“ *“*" ‘ .. .. . «41 mil. '”' ‘3‘” - ...'._o -. r. 5J‘ .U 3! I. 8.5 3": 7.319013” :3 . eidusebtsnoo a To F" ;bss¢s138nomsb _s.£b : 5?: ~ \_ ”I 013885; 8:13;: nsqo gutsy 3::qu snap quaqano Alt-1 lethal II ant-'t .l.l\9‘r H._ r-Tv‘ ‘Lur-ih- "‘ 1'", ..-u '2 L n V 0’ '-' ue -:4.’ ' ~ ‘ . -~-- . f. “-2. l 0., I ' . ' r- i C o ...'.....- i' ~—————---IIllIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII'II'I James E. Germain Thus the validity of the expression used to determine K was proven correct for open type media. It is believed that the variations in K found for conventional filters employing crushed stone or rock as filter media are due to a lack of air supply and clogging and that this is especial- ly true in the deeper parts of the filter. Based on the results of this analysis an attempt was made to use the K value as an indicator of the treatability of various industrial and municipal wastes. 55:.) lo 2311:1131 out It!“ ° .3ij '10} Joe'noo nova-m OII' ._ .5: '1. :11 o'enomsi'anv 0d: 3m JJSCfQLo'XG gattollq-O Var giqqun qln lo 1081 I ' : «4»ec ed: a! sun: ‘1 a: o'nffi ;.i3 to evinced }_ 'iéni as as eulav ,;:i:.n bx- isr1aeubni ——————---IIlllIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII K VALUES FOR TRICKLING FILTERS by_ N» James Ef’Germain A THESIS Submitted to the College of Engineering Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Civil and Sanitary Engineering 1963 Approved 'Ji 830.1" x .. “3.513 c k/Q 5’ :}/-6f ACKNOWLEDGEMENT The author wishes to express his sincere appreciation to Dr. Karl L. Schulze of the Department of Civil and Sanitary Engineering for his valuable guidance and assistance in connection with this thesis: and also to the Michigan Department of Health, Mr. George Martin, Engineer - Manager of the Green Bay MetrOpolitan Sewerage District, and Dow Industrial Service for contributing a major part of the experimental data which were analyzed in this thesis. 11 ". a. d '50."...‘UB BUT F e._er~ J ital .10 ' L..~fauan '*‘ 3.3siainagxe Chapter II III IV VI VII TABLE OF CONTENTS Title Index of Tables Index of Figures List of Symbols Introduction Procedure Evaluation of K for Municipal Waste K as a Parameter for Treatability of Wastes Discussion Conclusions Bibliography 111 Page iv vi 13 18 20 21 - Kabul .70 leblli .‘ ——————----IIIIIIlIIllIIIIIIIIIIIIIIIIIIII'IIIIII'I'IIIIIII INDEX OF TABLES Table No. Title 1 K values from 10 State Standards Design Criteria, 90 percent Gross Removal 2 K values from 10 State Standards Design Criteria, 3O MGAD Dosage Rate 3 K values from 10 State Standards Design Criteria, 15 MGAD Dosage Rate 4 K values from 10 State Standards Design Criteria, 5 MGAD Dosage Rate Michigan Filters Trickling Filter, Sandusky, Michigan 7A Trickling Filter, Sparta, Michigan, Data Set No. 1 7B Trickling Filter, Sparta, Michigan, Data Set No. 2 8A Trickling Filter, Green Bay, Wisconsin, Data Set No. 1 8B Trickling Filter, Green Bay, Wisconsin,Data Set No° 2 9 High Rate Filters with Recirculation 10 Pilot Plant Data 11 Rock Filters Treating Industrial Wastes 12 Surfpac Tower, Battle Creek, Michigan 13 Surfpac Tower Treating Coke Plant Wastes l# Surfpac Towers Treating Industrial Wastes 15 Filters using Plastic Media Treating Frozen Food Wastes and Domestic Sewage 16 K values for various Industrial Wastes iv .' ‘ s: .11.:- 35L‘1‘7 I ~ - .L-L- W“ Wt 0" «m 1“" )1 - 'I‘fi ...)" 1‘ 1 9 t I. *l - . A} I I 13 J n . ‘1 , \ I ’ ’ 1.1L _ 1 t I "‘ . SI a1:-r" ' ' ... _ * ‘0 ‘ " ' ‘ I' I . 1 ' ' _ .. as ‘_ 4. ' . - c ‘1 . .‘I 5 " : J) .' ,J ‘ .7 . ‘ ‘ - . ‘F' I v0 - -l l ‘ e I ‘ - “ ea- .2;; a.iv_f u..3vn8 - 31 ‘gigwnesoui SSIJASuI steak oiaaeii anion eusviii eaewsa oiiaouaa.anc casual _-;‘:-r. fi INDEX OF FIGURES Figure No. Title 1 Detention time versus D/Q3/3 after Sinkoff 2 K values from 10 State Standards Design Criteria 3 K versus Dosage Rate as expected from 10 State Standards for a 5 feet deep trickling filter u Sandusky, probability plot of K 5 Sparta, probability plot of K 6 Green Bay, probability plot of K 7 K versus Dosage Rate for operating Rock Filter Plants Treating Municipal Waste, Showing Influence of Depth 8 K versus Depth for Operating Rock Filter Plants treating Municipal Waste 9 K versus Depth from a 10 ft. Deep Rock Trickling Filter Pilot Plant 10 K versus Depth for Surfpac Pilot Plant at Battle Creek, Michigan 11 K versus Dosage Rate for Battle Creek, Michigan Surfpac Pilot Plant 12 K versus Dosage Rate for Industrial Wastes 13 K versus Dosage Rate for Industrial Wastes 14 K versus Dosage Rate for Industrial Wastes 15 K versus Dosage Rate for Industrial Wastes 16 K versus Dosage Rate for Industrial Wastes 17 K versus Dosage Rate for Industrial Wastes and Domestic Sewage 18 K versus Dosage Rate for Industrial Wastes and Domestic Sewage 3.33.71“ w'r? :013n63oflfi 7 seals? 1 susuev X _*1eh9838 .giaubnsa .; ". .2118 ‘ c 5 L Al: ._ ‘ ‘ .,k . " s ‘ V - ' . .I.. ..h" w; 3;...3153 x asaeew iniusaib ' 1' ' _ _ ’ all ’ r” '- . '1 we won m... 1 —- Y", ,."'._I , -" J ”_ ‘ "'.. ‘ '13-‘14» -: <93" _ m” II- .. Q — . - .: 1‘ 'E..1 LIST OF SYMBOLS Symbols are defined as they are used throughout this thesis, but for convenience common symbols are listed here also. BOD of settled filter effluent, mg/l BOD of incoming plant waste, mg/l BOD of mixed flow (incoming plus recirculation),mg/l A factor representing contact time on the filter Filter reaction rate ( to the base 10) Filter reaction rate (to the base 9) a constant Depth of Filter in feet DUGH‘Mdé-‘Sé‘ ll Hydraulic load to filter in MGAD, including recycle flow w\\ n a An exponent L p = Fraction of incoming BOD remaining 2 BE I p1 = Fraction of BOD remaining in filter effluent related to BOD of mixed flow (incoming plus recirculation) - LE r = Recycle ratio 3 .35 Q1 Flow to the plant in terms of MGAD Q1 QR = Recirculated flow in terms of MGAD vi .1 ".1133! A “eiIii 8 "MILE -"-'- ()8: 1‘: v " CHAPTER I - INTRODUCTION A trickling filter is a fixed bed through which a biologically treatable waste is passed and is broken down by the biological growth within the filter. The basic mechanism consists of some type of media (rock, slag, or artificial material such as plastic) on which a biological film grows as the waste is trickled over it. This can be illustrated diagrammatically as below. 5 biological film liquid waste Air / ~ -- _ 4- L“ / r y l I Filter 3 H28 l / , ; Media j | __ BOD / a 1 NH: l 02 x r g: ;=. / 2 I 0°2_ ; Anaerobic Aerobic 1 l / {31.1161 - I m '1—x;'§ i} a; 1e3111wh 1 v1' “1 siesw eld801018 1’Tt' 'r THV’J'TE “038010“ ‘70 eJeiauoo ..., _ ' 'e {11:18:15. Intel _’_ Q “ r C I " a1 «z: as awO'xa ‘ - : ‘wzstfl b33813lfli"‘ '-.qu;' €13 uealti sibel 2 The filter medium serves to support the biological growth, shown here as the anaerobic and aerobic layers. The aerobic layer is the effective portion of the filter as it oxidizes the waste which is captured out of the liquid waste stream. The air passing through the filter performs the very important function of supplying oxygen to the aerobic bacteria. The anaerobic layer is black in color and exists in the area devoid of oxygen. The organisms comprising this layer receive their oxygen from the mate» rial they reduce which in this case is the dead cells and metabolic products of the aerobic layer. By-products of the anaerobic layer such as H23, acids and ammonia are oxidized by the aerobic layer. The efficiency of removal of wastes from the waste stream is proportional to the time the waste is in contact with the biological growth. This has been substantiated by Rowland (1}), and Schulze (lk), and can be expressed mathematically by the following formula: L -—E- : e‘Kt (1) LI in which LE 3 Filter effluent BOD, mg/l Incoming BOD applied to filter, mg/l (without recirculation) ' LI t Z Contact time k 3 Reaction Rate § .0 -:_-.---y )3 am lath. I :rwoneens 353 I." ' , —‘ .. J "'0'!“ .‘I It 1.!“ 'i ' r 9.. till-W ad: I i '9'“ p, 4 «3953. ~. . I'm-mi 2‘!" ‘} . '2,’osd otdOCI. 7: «:3rxe bfli H”-e.1qm00 Ian ‘133539! all a r " 6013 JUBJROO = 3 ”a -. . 9-3" ; "" "— . ) ‘ -~ - .1 ' . 3 The mean contact time t through the filter media has been expressed by Schulze (l) and Howland (13) by the following relationship: t : CD/cn (2) Where C I A constant Depth of filter in feet D Q 3 Hydraulic load, MGAD n = an exponent, which has been shown to be 2/3 Hydraulic principles of trickling filters were investigated extensively by Bloodgood et a1 (12). The authors showed that the contact time on an inclined plane and on a sphere was inversely proportional to the 2/3 power of the liquid application rate. Experiments on an inclined plane with slime showed that the contact time increased up to 50% when the plane was flat (2.87 degrees) but at a 45 degree angle the contact time increased less than 10%. In additional eXperiments using 3/4 inch to 3 inch balls the contact time increased 12% but this was proven to be due to the increase in diameter of the balls caused by the biological film. :qinsne£¢° .. I 63” :4 - ‘ J.‘ . ,- . * - . 1 eilau m .. .1‘-.-——I ' I ' I in... -' I I . I I V ‘ _ . v 1‘ K" con '1 ‘ . in - ‘ - n. ‘ 6‘-" c3 11.." q' . .m If ierrgafczj 633 {d hoarse. ‘I‘. . and: don: LEGEND @001. B 2 D = u.1 feet ' (glass spheres, 3/4 " Diam.) Col D 2 D = 9.0 feet (porcelain spheres, 3 " Diam.) m g I :3 c: H s , c: ) ” 1 d‘ p “’10 w a ! E3 / 1 g AiC;Columh B :8 [*4 3 "3‘7 / L 36 :5 .. / L :35 / 4 [f6 r" A//’ A”,,Ar’<:3:Column D 2 5/7 + l o ’§%:)¥I”drr 2.0 3.0 u.6 D/Q2/3 FIGURE 1 — DETENTION TIME VERSUS D/Q3/3 AFTER SINKOFF I LEGEND Col. B 2 D = 4.1 feet I (glass Spheres, ' 3/h " Diam.) Col D : D = 9.0 feet (porcelain spheres, 3 " Diam ) in Minutes ’ H O O\ Liquid Residence Time tr Nb»! «PJU'I p 0 . - 0 1.0 2.0 3.0 4.0 D/02/3 FIGURE 1 — DETENTION TIME VERSUS D/Q2/3 AFTER SINKOFF fi a Figure l as plotted from data by Sinkoff et a1 (11) indicates the linear relationship between detention time and D/Q2/3. It can be seen that column D has a shorter residence time (tr) than column B. This absolute residence time can be computed by the formula Offered by Sinkoff et al.: tr - c H Vl/2 3m ‘1—I7‘g3‘q— Where Cl a constant, 3.0 for porcelain Spheres and 1.5 for glass spheres H = Height v = Kinetic viscosity of fluid g = Acceleration due to gravity 8 = Specific surface of media I Eggéiuggrggggpfggq_ Q = Hydraulic loading rate m = a constant, 0.83 for porcelain Spheres 0.53 for glass spheres The m in the above formula for porcelain and glass spheres corresponds to the n determined by Schulze (l) and Howland (13) for sewage on Operating trickling filters. __________---IIIlIlllIIIIIIIIIIIIIIIIIIIII'III'IIIIIIIIIIIII 5 Schulze and Howland_determined that n = 2/3 and thus we can say t = CD/Q3/3; (3) and substituting for t in equation I we have LE - c 2/3 L‘f‘ekD/Q° changing from base e to base 10 gives the basic equation: Combining the constants k and C and LE _ 2/3 '—- = 10 KD/Q as originally preposed by Schulze (5). l" H And since p = EE we then have I p = lO-KD/Q3/3 (u) Extensive work using radioactive tracers for measuring detention time in trickling filters was done by G. E. Eden and K. V. Melbourne (24). They found the losses by adsorption of Nag“, Rb86, and K)"2 to be great. More satisfactory results were obtained with Brae, and Co58 and 0060. However these authors did not vary the hydraulic load on their experimental filter so that it was not possible to obtain a series of K values from their data. Temperature is another factor that has an effect on the performance of trickling filters. In his paper on this subject Howland (10) used the following approach: Assume kt - k20 C(T ‘ 20) Where kt - BOD reaction coefficient at any temperature, common logarithms R20 = BOD reaction coefficient at 200 C; common logarithms efeb hinfwofl ',r P’- 3" ’. "a "r\\U-' : J L C 1 ‘ L 5HIJUJIJO'; 1 “PW“? d . IxiewA axed! :Clfiggp63_tas :u snsioillsoo 80L30881 008 - ‘;2:g‘ \- : , ‘ ' D l . i l - I I ' ‘h *‘I . o -- a ’ ‘.-g - q. ‘l _ .. - ———————---IIIIIIIIIIIIIIIIIIIIIIIIIIIIIII'IIIIIIIIIIIIIII 6 - A number chosen to fit experimental data (l.O#7 O l is used in BOD calculations) T = Temperature in °C with the value kt appearing in: = 1.3 lo‘ktt (5) Where L = Ultimate first stage BOD at time t La = Ultimate first stage BOD at time 0 t = Time, in days BOD remaining can be expressed by L/La thus: -—I£:; = 10-ktt and substituting for kt we have —%g = lO‘thO ° adwfi om ® owwmpod Qwfl .QmpHHQ on BOHQ m :m m- a... oapmm oaozomm n m am .H\ms.oom passacam Hmcaa mm .H\ms.mom pcmzaccm .sanm H\we.aom pcmzaksH A.m>ofimm mmomw azmummm Om .¢HmmeHmU onme mnm¢mz M I H qu<9 l. It ‘J t, .I .3 33m oh 8m...— u.....:.. C . r4 .1... F... J I a H/Maa a... .. r. “moth. , .-uh when m... HQ“. _. .A H m.......... Q :m m a: m mm m :00 m Amxmav Hm MMMHI.M mw.m . mm.m mw.m mo.m a m\ma mmn. cam. maa. mmo. m woa mms. Nam. was. mmm. Hm mos. Nam. map. mmm. Q m.mm m.mm m.mm m.mm ass\ao coca\# .omsmon oacsmpo o~m cam cam cam ams\* .pmpfiaa on smog oficmmso omm.~ omm.~ omm.~ omm.~ so .gon_m n spams o ossao> mmmo. mmmo. mmmo. mmmo. mass on o ammouoe oo.H oo.H oo.H oo.H mas .smpaaa 0p scam rm u m o o o o mu.mm oapmm odomomm m.mm :N m.mm .HHH jm .H\ms .mom passaaam Hausa *ow &Om fibm fio: Hamam>o .pCmapmona & omH oma oma oma mm .H\ms .oom passages .sHsm mwa mma mma mma axma .mom pcmsaacH awsa mesa awed mesa A.m> M u N mqm mo .pCmapmmaB & mm .H\ma .mom pcmzaccm .sapm H\ms .mom passaacH A.m> M I N Mdmde H3. mo.m mmm. mm:. mm:. mm ooo.H oom.HH mmmo. oo.H m.mm sea omH me amzH QQHQ ma new: mH o swamped Q32 .QmpHHQ op BOHQ H m II mil Hm I mm OHpmm maozomm Hm .H\ma.oom pcmsHQQm HmcHQ Hm>osmm & mm .H\ms.mom pcosHQQa .aHQQ H\me.mom pcmsHQcH A.w>dv pcmHQ 0p BOHQ mH .HHmeHmo szme mQQQQZHQm HQQQm 0H somQ mmaq<> H I m mHmQQ L J. .. . .. ....v in . , . . _. ...- ram .2393 + ...w... .2; 1’ 0.23.3... H E 3093...... Chg _fiilt ..f.....-.. .m I IQanmmi .. ; ..QQQ 0Q m_ ..aa .....c..«. .. u H h I '1 522.5% .mwa. F' ..II II . 'II I’IIDI‘",I L I‘ 0 mm m mm m mm m Qm\m00 Hm WMMQI.0 00.0 00.0 00.0 HQm\m0 NQH. NNH. QNH. m on Hmm. Hnm. mmm. HQ an. :mH. mom. Q m.mH Q.Hm m.m0 Q00\Q0 000H\* .mmsmon OHQsmso :mm.H 0Qm.H Qm0.H Q00\* .nmpHHQ 0p 0004 0Hcmmpo 00m.ww 000.0: 00m.0H Q0 .30 I _m u an00 o mssHo> QoH. mmmH. QHwo. 0¢0s mH s swampoe H.0 mQ.m NH.H Q02 .QmpHHQ 0» QOHQ H.m mQ.H NH.0 m u m .-.. m 0Hpmm machomm mm.m m.mH 0.Qm Hm .H\ma.nom pcmsHQQm HacHQ mmm aOm fiow Hm>oamm g ONH omH omH mm .H\ma.mom pcmzamym .EHQQ mmH mmH me mee.000 pcmsHQcH Q02H Q02H 002H A.m>¢v pcmHQ on SOHQ AQHDZHBZOOV medm mw¢mon 0462 mHQ¢HmHBHmO ZOHmHQ QO M I M H4m<9 erlx 0H. 00.0 mmm. 00:. 00:. mm 000.H 000.0: 00.0 0.H 0.00 000 00H 00H 002H 0040 004000 0<00 m . 0402 m @ wwwmaod 002 .QmpHHQ 00 50HQ m- 0.. 00 OHpmm 0Homoom :0 .H\05 .000 pcmsHQQQ chHQ H0>oamm R 00 .H\00 .000 pcmsHQQm .eHQQ H\00 .000 pcmsHmcH A.m> M i # mdm¢e 00. 00. 00. Hh0 n 0 0 0 000000 H mwwu .0 00.0 00.0 00.0 H 0\00 00:. 00:. 00:. m 00H 000. 000. 000. H0 000. :0H. 000. 0 00.0 0.0H 0.H0 000\00 000H\* .000000 0000000 :00.H 000.H 000.H 000\* .000H00 00 0000 0Hc0000 000.:00 000.00H 000.0: 00 .=0 I .0 u 00000 0 0asH0> .0.0H 00.0 :0.0 0400 0 0 0000000 H.0 00.0 0H.H 000 .000HH0 00 00H0 H.0 00.H 0H. 0 n 0 OHpmm machomm 00.0 0.0H 0.00 :0 .H000 .000 0000H000 H0000 fimm fibm $00 Hw>oamm $ 00H 00H 00H 00 .H\00 .000 0000H000 .sHpm 00H 00H 00H H\00 .000 0000H00H 000H 002H 000H 0.0000 000H0 00 son AQHDZHBZOOV Hadm mw< owmnonfi mucomuumoa .0... nmpdkdaoo momasooa QmHCCdpm owompo Hamscadflm consumoz mxmsocmm pcoaonh nomom nonnwm comudpmn aoxastonom mpnmmm apflo mmao ammooo mama ocdawm ammo: poamhpo NH. NH. 0.0H O.m o.N mm.m mm.m Om.H m\ma ms. Nm. ms. mm. :N. mH. Hm H H woa 9 NM. mm o.m mm. mm o.oH mH. mm o.m c socmHOHccm * d AmmazHezoov mmmeqHa z< pCHHh mPCoamHo undo: comHam cmmHROHs anozpsom TABLE 6 - TRICKLING FILTER, SANDUSKY, MICHIGAN (31, 32) Data obtained from the Michigan Department of Health Q D I! 15 MGAD including variable recirculated flow 6.0 feet Date, June and July, 1958 using 8 hour composite samples. Flow MGD % Efficiency p r pl log %1 K .22 59 .41 2.58 .72 .14 .14 .22 73 .27 2.58 .57 .24 .24 .22 76 .24 2.58 .53 .27 .27 .23 66 .34 2.43 .64 .19 .19 .24 75 .25 2.16 .51 .29 .29 .24 75 .25 2.29 .53 .28 .28 .24 52 .48 2.29 .76 .12 .12 .24 69 .31 2.29 .60 .22 .22 .25 66 .34 2.6 .62 .21 .21 .25 59 .41 2.16 .67 .17 .17 .26 71 .29 2.04 .55 .25 .25 .26 53 .47 2.04 .73 .14 .14 .26 60 .40 2.04 .67 .17 .17 .26 58 .42 2.16 .70 .15 .15 .28 56 .44 1.82 .68 .17 .17 .28 54 .46 1.82 .71 .15 .15 TABLE 6 - TRICKLING FILTER, SANDUSKY, MICHIGAN (31, 32) (CONTINUED) Flow 1 MGD % Efficiency p r pl log ~ K p1 .29 71 .29 1.72 .53 .27 .27 .29 65 .35 1.73 .60 .22 .22 .30 58 .42 1.63 .66 .18 .18 .31 63 .37 1.55 .60 .22 .22 .32 73 .27 1.47 .48 .32 .32 .49 78 .22 0.61 .31 .51 .52 a so Bond HquHm one cans mmoq mH m 08He no fi H m.o H. m.mm mm mm mm om om oN om On 0: on om 0H m m ——~,.._ __ . onus m.o 0.0 .v 1.1... . e. .. :2. .... _w. _. _. . ..ictud +11... . TABLE 7A - TRICKLING FILTER,SPARTA, MICHIGAN DATA SET NO. 1 (31, 32) Data obtained from the Michigan Department of Health Q = 3.86 MGAD including variable recycle flow D = 6.0 feet Date, April thru August, 1958 using 8 hour composite samples. Flow MGD % Efficiency p r p1 log % K .146 74 .26 1.06 .42 .38 .12 .149 79 .21 .95 .34 .47 .14 .159 85 .15 .83 .24 .62 .19 .161 73 .27 .81 .40 .40 .12 .165 79 .21 .76 .32 .49 .15 .168 77 .23 .72 .34 .47 ..14 .169 68 .32 .42 .40 .40 .12 .171 79 .21 .70 .31 .51 .15 .175 86 .14 .66 .21 .68 .21 .181 78 .22 .60 .31 .51 .15 .206 76 .24 .41 .31 .51 .15 .218 71 .29 .28 .34 .47 .14 .238 74 .26 .22 .30 .52 .16 .253 82 .18 .13 .20 .70 .21 .271 73 .27 .07 .28 .55 .17 .328 77 .23 — .23 .64 .20 TABLE 73 — TRICKLING FILTER,SPARTA, MICHIGAN DATA SET N0. 2 (31, 32) Data obtained from the Michigan Department of Health Q = 3.86 MGAD including variable recycle flow D = 8.0 feet Date, April thru August, 1958 using 8 hour composite samples. flégw’% Efficiency p r p1 log %1 K .149 96 .04 .95 .075 1.12 .34 .166 95 .05 .75 .085 1.07 .32 .174 94 .06 .67 .10 1.00 .30 .175 95 .05 .67 .08 1.10 .33 .178 93 .07 .63 .11 .96 .29 .182 95 .05 .59 °08 1.10 .33 .182 90 .10 .60 .15 .82 .25 .188 88 .12 .54 .17 .77 .24 14n0131 - ET :2 ATAO '? .23'1‘1 boa .L r ‘J. - Fulani “‘0' Joel QA . .A 0113 ftxqt.. a, writ)?!“ * " m.mm mm mm csosm 03Hc> 0 he swap mmmq ow 0N om om 0: on om M so 904m weHHHm one came mmoq mH M mEHa Mo m mm mm mm mm om ow ON ow on o: om om 0H m m H m.o H.0 I ‘ _ 0.0 m1... 8 m page H.0 H4 p m mprL. 848...: 6 1T N8 +. 1. 0 o m.o n... + O O_ .0 +0 +0 + IT .3 0 +6 .+ m.o m.0 .._u :fil‘lixm 11 Figures 4, 5, and 6 demonstrate that the K value for each filter fluctuated over a definite range. The median values and standard deviations of K for these plants are as follows: Location of Plant K value Range Sandusky, Michigan 0.20 i .06 Sparta, Michigan Dats Set #1 0.15 f .03 Data Set #2 0.30 f .05 Green Bay, Wisconsin Data Set #1 0.37 i .10 Data Set #2 0.36 f .10 Each of these trickling filters had a constant dosage rate throughout the 5 months test. Table 9 consists of operating data obtained by Rankin (22) and from Infiloo Corporation (23) regarding high rate filters with recirculation. The K values for all plants discussed so far including those from Table 9 were plotted versus depth and hydraulic load as shown in Figures 7 and 8. Figure 7 shows that regardless of hydraulic load the shallow filters of 3 and 4 feet depth had the highest K. with an average value of 0.81, whereas 6 and 7 feet deep filters produced an average K of only 0.26. Figure 8 is a plot of K versus filter depth for a series of filters treating municipal wastes (22, 23, 31, 32). 12 These data show a definite tendency for K to increase as the depth decreases. This, the author believes, is due to the fact that the deep filters have a zone of low activity in the middle region caused by the lack of air. To further illustrate this, data from pilot plant studies by Abdul - Rahim, Hindin, and Dunstan (9) were analysed for K. The results are shown in Table 10 and plotted in Figure 9. The experimental rock filter used by Abdul — Rahim et al. was 10 feet high and 5 feet in diameter. The hydraulic dosage rate was varied from 10 to 35 MGAD, and grab samples at filter depths from 1 to 10 feet were taken semi—weekly. The curves clearly illustrate how K decreases with increas— ing depth on a rock filter. MN._ o.m New. Nmm. mm.H mm. o.Nm o.mm MH.H emepm eem MN. o.m ozm. NNm. mN.H mm. o.Nm m.Nm MH.H Show ewwewmpwmeq mm. o.m mMH. 0mm. mN.m 0mm. o.mm H.4m Sm.o mmepm eem Aewepm mm. o.m mmm. mmm. Nm.H 0mm. o.mw m.HH em.o .ana .mwmwmewmmem oeav on. o.m mwm. mHm. em.H mmm. m.oN m.HH mmm.o emepm eem em. o.m Hom. oom. H:.H mam. m.ON H.w mmm.o .aea .mmemmMmmewwme Ne. o.m nee. mm. 0.H omm. o.mN m.mH NH.mH.eHao .apHo eeoeeHao mm. m.: mmm. 1:. MN.m MNH. N.mm m.mH NH.m cosmeHnmes .eeeHeeHm Hm. m.m NNm. me. m.H mmm. m.NN o.mH mN.H 0Hno .peoaeea He. o.w ems. mNm. :H.m omH. 0.:w m.Hm mm.H eeOH .eaeq enoem mm. m.m mos. mm. mo.m SMH. w.ww w.om mm.o eHenoeHHeo .HemHHe Aemsem Nm. o.m owe. mm. Hm.m me. :.Nw o.mH mm.o eeeeeH< .eeeeom eHmeHmv om. 0.: Now. Hm. mH.H mmH. H.mw mm.N mm.o .w .z .aeez .eew mo.H o.m omN. mH. mm.o 80H. :.mw mm.m oH.o .os.eHHSEpeeo magmamm M Q Hm woa HQ 9 m mocoHonmm on. .0 amwm cogmooq Amm .mmv vHOHeddbomHomm mBHB mmmBAHb meg mem ... m m4m popmonocHa H mMEMEom M Q m moa an a Q mocmHOHMMH & .3 avg COHpcooq eOHE AQMDZHBZOOV ZOHB50 I. 9 €> a a a O <> <> ... 0.0 . 0 5 10 15 20 25 30 Dosage Rate, Q in MGAD FIGURE 7 - K VERSUS DOSAGE RATE FOR OPERATING ROCK FILTER PLANTS TREATING MUNICIPAL WASTE, SHOWING INFLUENCE OF DEPTH LEGEND 4-Michigan Filters (31, 32) °.After Rankine (22) CIAccelo Filters (23) 1.4 1.2 7K\ 0 \ 1.0 00 \1 mo \ 0 8 v \ . \ \ 0'6 o \\ o (3 EJ \\\ 0 m 0.4 \\ 9 w \b‘ \ f \ \ \\ \§E. EJ m 0. . 2* 2 + + \ '4‘ \ 0.0 .1 0 2 4 6 8 10 Filter Depth, in Feet FIGURE 8 - K VERSUS DEPTH FOR OPERATING ROCK FILTER PLANTS TREATING MUNICIPAL WASTE TABLE 10 - PILOT PLANT DATA (9) Q 1% Efficiency pl log %1 D K 10 43 .57 .243 1 1.13 10 55 .45 .346 2 0.80 10 62 .38 .418 3 .65 l0 66 .34 .468 4 .55 10 67 .33 .481 5 .45 10 68 .32 .494 6 .38 10 71 -29 -537 7 ~35 10 73 .27 .568 8 .33 10 76 .24 .619 9 .32 10 78 .22 .657 10 .30 20 23 .77 .113 1 ~83 20 47 .53 -27“ 2 1°02 20 52 .48 .313 3 .77 20 55 .45 .346 4 .64 20 58 .42 .376 5 «57 20 60 .40 .397 5 .49 20 53 .37 .426 7 «“5 20 65 .35 ~454 8 ““1 20 67 .33 ~“31 9 .40 20 70 .30 .522 10 .39 TABLE 10 - PILOT PLANT DATA (9) 1 Q, % Efficiency pl log 51 D K 10 43 .57 .243 1 1.13 10 55 .45 .346 2 0.80 10 62 .38 .418 3 .65 10 66 .34 .468 4 .55 10 67 .33 .481 5 .45 10 68 .32 .494 6 .38 10 71 .29 .537 7 °35 10 73 .27 .568 8 .33 10 76 .24 .619 9 .32 10 78 .22 .657 10 .30 20 23 .77 °113 1 .83 20 47 .53 .274 2 1.02 20 52 .48 .313 3 O7? 20 55 .45 .346 4 °64 20 58 .42 .376 5 .57 20 60 .40 .397 5 °49 20 63 .37 .426 7 ~45 20 65 .35 «45” 8 °41 20 67 .33 ~“31 9 .40 20 7O .30 .522 10 939 TABLE 10 - PILOT PLANT DATA (CONTINUED) Q % Efficiency pl 10% %l D K 20 51 ~49 ~308 1 2.28 30 56 .44 .362 2 1,34 20 64 .36 .441 3 1,09 30 69 .31 .505 4 93 20 71 -29 .530 5 .81 20 72 .28 .554 6 .64 20 72 .28 .554 7 ,59 20 74 .26 .582 8 .54 _ 20 75 .25 .606 9 .50 20 77 .23 .637 10 .47 35 25 .75 .121 1 1.30 35 35 .65 .188 2 1.01 35 47 .53 .270 3 .96 35 52 .48 .322 4 86 35 57 .43 .365 5 .78 35 59 .41 .390 6 .69 35 62 .38 .416 7 .63 35 64 .36 .442 8 .59 35 66 .34 .472 9 .56 35 70 .30 .525 10 .56 2.0 LEGEND (9) 9, 0.: 35 MGAD 1'8 T B, Q= 20 MGAD \ o . Q = 20 MGAD 1 6 _ A, 0.: 10 MGAD 1.4 \ \k\ 10 \ N: \ \ 8 o .51 \ ‘ 6 c\ u A \» ° " II o > . n J .4 , . . _ L“4‘~—Mi .2 .0 fl 0 2 4 6 8 10 Filter Depth, in Feet FIGURE 9 - K VERSUS DEPTH FOR A 10 FEET DEEP ROCK TRICKLING FILTER PILOT PLANT CHAPTER IV - K AS A PARAMETER FOR TREATABILITY OF WASTES It can be asSumed that municipal wastes are more or less similar over the country, whereas industrial wastes certainly have very different characteristics. These differences could be measured by K as the degree of treat- ability; the higher the K value, the greater the treatabil- ity of the particular waste as long as the apparatus is the same. A comparison of K values obtained from pilot plants using an identical type of plastic media and distributor should therefore provide an index of treatabil- ity. It should be noted that the K factor is much more applicable to the Open type of filter media since in this case K does not vary appreciably with depth. Table 11 shows a series of K values computed for filters using stone as a medium and treating various industrial wastes. The first set of data shown in table 11 relates to a full scale plant using rock media and treating a Fine Chemical Waste (25). The plant was operated at a recycle ratio of 9 with a depth of 4 feet. The average BOD removal was 53 percent. The low K with an average value of .06 is an indication that this waste was difficult to treat. The second set of data in table 11 relates to a .004 MGD pilot plant treating Phenolic Waste (26). The K values for this pilot plant ranged from .11 to .23. 13 14 This fluctuation in K values is to be eXpected from a waste of this type since it is non—uniform in makeup and subject to shock loads. The third group of data was obtained from a plant (27) treating the same phenolic waste which was handled by the pilot plant. The K value of 0.14 for this case is in agreement with the K values listed for the pilot plant. Data from a trickling filter pilot plant which was Operated at the municipal treatment plant in Battle Creek, Michigan (20) are shown in table 12, and plotted on Figure 10. This filter used “Surfpac" which is the trademark of The Dow Chemical Company for their plastic biological oxidation media. This is one of several open type filter media which is commercially available. The lines drawn through the K values for each depth in Figure 10 indicate the consistancy of these values in regard to depth. Thus it may be concluded that the K values in an open media type filter such as Surfpac do not vary with depth. AucmaQ mHmom Hasmv 09003 oHHoeccm :H. 0.0H N:m. m:. u m:. mm 0.0 0.NH HH. m.0H m:m. mNm. : mNm. m.m: N.0H :00. ApecHe BOHHQV peace OHHoecnm mm. m.0H 0N:. mmm. u mmm. 0.00 N.0H :00. mm. m.0H m0:. 0mm. u 0mm. 0.00 N.0H :00. opens HceHEceo ccHa 00. 0.: :0. H0. 0.0 N:. mm 0.:H 0H.H H mxamacm M a m woa HQ 9 Q NoCmHOHmmm x. d Q62 BOHrm ANN .0m .mmv meme; anmemsozH ezHeemma mmmeqHQ Moom I HH Hdm¢e TABLE 12 - SURFPAC TOWER, BATTLE CREEK, MICHIGAN (20)* Flow MGD 01% Efficiency p r pl log 51 D K .04 20 33 .67 ~ .67 .174 10.5 .12 .04 20 59 .41 - .41 .390 21.0 .14 .04 20 79 .21 - .21 .676 31.5 .16 .04 20 88 .12 - .12 .925 42.0 .16 .12 51 57 .43 - .43 .366 21 .24 .12 51 79 .21 — .21 .68 42 .22 .23 102 22 .78 - .78 .108 10.5 .22 .23 102 40 .60 — .60 .222 21.0 .23 .23 102 39 .61 - .61 .215 31.5 .15 .23 102 57 .43 — .43 .367 42.0 .19 .47 206 40 .60 1.0 .75 .125 21.0 .21 .47 206 51 .49 1.0 .65 .19 42.0 .16 .49 213 10 .90 - .90 .046 10.5 .15 .49 213 12 .88 - .88 .058 21.0 .10 .49 213 14 .86 - .86 .068 31.5 .08 .49 213 23.5 .765 — .765 .116 42.0 010 * This is a combination of municipal and industrial waste LEGEND <> 213 MGAD 0 206 MGAD e 102 MGAD B 51 MGAD A 20 MGAD 0.3 I L O O 2 ‘\:>-<§:‘ A ‘ T:::==——=::f [L \ \‘ 0.1‘ <> 0.0 . 0 10,5 21.0 31.5 42.0 Depth in Feet FIGURE 10 - K VERSUS DEPTH FOR SURFPAC PILOT PLANT AT BATTLE CREEK, MICHIGAN 15 The same data were replotted in Figure 11 to relate the K values to the hydraulic dosage rate. The K values range from 0.08 to 0.23 while the hydraulic dosage rate varies from 20 to 213 MGAD. These data indicate that K is independent of hydraulic dosage rate. Db» .O L v . O 50 100 150 200 250 Dosage Rate, 0, in MGAD FIGURE 11 - K VERSUS DOSAGE RATE FOR BATTLE CREEK 1110111an SURFPAC PILOT PLANT ’ 16 Data from the Surfpac Technical Bulletin of Plastic Biological Oxidation Media (21) and other sources (28, 29) are analyzed in Tables 13, 14 and 15, and shown graphically in Figures 12 through 18. These data include the results of at least several months of testing at each location and are taken from full scale plants or from pilot plant Oper- ations set up at the reSpective industrial waste treatment plants. All results are from composite samples. The individual K values computed for each type of waste were then plotted versus hydraulic load as shown in Figures 12 through 18. A summary of these data is contained in Table 16 where the different types of wastes are listed accord— ing to their decreasing degree of treatability as measured by K as a parameter. From the table it can be seen that the waste with the greatest treatability was Domestic Sewage (29) with a K value of 0.75. This K value was obtained for a 10.5 feet deep roughing filter which removed 50% of the primary effluent BOD at a hydraulic load of 188 MGAD. This is certainly an outstanding performance for a filter and this is demonstrated by the high K value. The waste that gave the next highest value of treatability was the Hydrocarbon Waste (21) with a K value of 0.56. This plant employed a recycle ratio of 4.5 and obtained 91 percent BOD removal. 17 A K value of 0.30 was measured for a municipal sewage when the pilot plant was not operated as a roughing filter but for 82% BOD removal. One filter pilot plant treated pulp and paper wastes at the same locality with varied operating conditions and produced different K values (21). A K value of 0.27 resulted when the waste was treated at a tempera~ ture of 48 - 54° C and a K value of 0.17 was obtained at a temperature of 39 - 41° C. At a Coke Plant the Ammonia Still waste produced a K of 0.30, but when the same plant was treating Final Cooler Mixture a K of 0.19 was produced. The most difficult to treat waste was a Paper Mill Waste (21) which was from a source other than the Pulp and Paper waste above, and produced K values of 0.15 and 0.09 when operated on separate filters with depths of 31.5 feet and 42 feet respectively. 0H. mN.mm 0MH. 0mN. 0m.0 mam. 0.0m 0.0mm mmo. NH. mN.mm 00H. mmN. mm.0 0mm. H.0m 0.mHm N:0. mm. mN.mm mNm. mmm. mm.0 m0:. m.mm 0.NNH mmo. Acnsetz neHooo HceHa mH. mN.mm N:H. NHN. Hm.0 mmm. m.:m 0.NNH wmo. Hum choaa4 HHpech caoov mm. mN.mm Nmm. HN:. mm.0 HH:. 0.0m 0.Nw 0H0. mm. mN.mm HHN. 0H0. 00.H m::. m.mm 0.:0m wmo. 0m. mN.mm Hon. 0m. : Hom. 0.0: 0.NmH mmo. mm. mN.mm Nmm. 0:. mm.0 «mm. 0.00 0.NMH mmo. HHpm cHeoee4 AH pech exoov 0m. mN.mm 00:. :H. mm.0 mam. 0.0N 0.Nw 0H0. H exudaom M a m mOH HQ .H Q hocmHoChm m“ d 902 Bofim HHNV mane; azenm umoo czHaeame_mmgoa 04mmm00 - HH names 0H. Hm 00H. 00N. 00. N00. 0.H0 0.0:H A0 H :H. Hm 0NH. 0:N. 00. 000. 0.00 0.0HH 0 on 0 cm 0H. Hm 00H. :00. N0. 000. 0.0: 0.00 .0880 aceHHe meHnwsom H0. H0 mum. 000. 00. 00:. 0.H0 0.N0 09003 achQ 0:8 aHemv 0H. H0 000. 000. 00.H 00:. :.N0 0.NN 0H. Hm 00H. 000. 00.0 :00. 0.H0 0.00 000. m: 0H0. 000. 00. N0:. 0.00 N.NN HH. 0: 000. 00:. 00. N00. 0.00 0.00 Hounds HHHS thch mmo. m: mmm. NNm. om. 1H3. 0.wm N.wm 00. 0: :NH. 0N0. 00. 0H0. 0.0: 0.00 NH. 0.H0 000. 000. 0N. :0:. 0.00 0.00H Ampmmg HHHz acacmv 0H. 0.H0 000. 0N:. 00.H 000. H.00 N.NN 0H. 0.H0 0NH. 000. 00.H 00:. 0.00 N.NN .IIIIHqumai. 000808080000 00. 0.0H 00:. 000. 0.: 000. H0 mHH IIIIHuuqqauHuA. ponds uzom acne peangmmv 0H. 00 000. 00:. 0.0 0H0. N.00 N.0m H 0Mpcaom M a m wOH HQ 9 Q moccHoHuuw &. d AHNV 00004: anmemanzH 0zHaemma 000000 oememam . :H 040:5 NH. H0 0N0. 00. 0H.0 00. :0 0.H0 AnopHHa 00. H0 N00. ::. 00.H H0. 00 0.0N mcHnwsom 09003 00H0000 00000 00. H0 000. H0. 00. 00. HN 0.00 N0. H0 00H. 00. 0H.0 N0. 00 0.00 00. H0 00H. 000. 0H.0 000. :.:0 000 00. H0 00H. N:0. 0N.0 0N0. H.0N 0N0 00. H0 000. 00:. 0N.0 :HH. 0.00 00H A0000: HHH8000V 00. H0 0H0. 0H0. 00.: ::0. 0.0N HOH ::. H0 000. 00:. 00.0. :0H. 0.00 0:H 00. H0 000. 000. 00.0 000. :.0N 00H 00. H0 :00. ::0. 00.H 0N0. 0.00 0HH 00. H0 0NH. 000. 00. 000. H.0: 0.000 A0 0 m on 0 cm: . ace nepHH0 00. H0 00H. N:0. 00. 000. 0.0: 0.:NH mcHnwsom 3003 08000 000 0H000 00. H0 N0H. 000. 00. N00. 0.H: 0.:NH 00.0.0620 M n Hm moH HQ .H Q zocmHOHmmm 00 a 10002H92000 000000 A .8 $3 (2) .7 <> <> .6 <> a:,5 .A .3‘ """_ -_-‘__’ B a (l) .2 ___- __ m <> .1 .0 150 200 250 LEGEND (1) Domestic Sewage and Food Processing Wastes (21) K = 0.18 Z .ou8 (2) Domestic Sewage, Midland (21) (High degree of Treatment) K = .30 Z .osu .6 .5 -_2_ h“-3 (‘> O O __- _ ::.L._ _-_._. _-__ _A_ 2 (l) A .1 A ' 0 50 100 150 200 250 Dosage Rate, Q in MGAD FIGURE 18 — K VERSUS DOSAGE RATE FOR INDUSTRIAL WASTES AND DOMESTIC SEWAGE TABLE 16 - K VALUES FOR VARIOUS INDUSTRIAL Waste Domestic Sewage; (Roughing Filter) Hydrocarbon Waste Ragmill Waste Domestic Sewage; (high degree of treatment) Coke Plant Ammonia Still Pulfi and Paper; (Roughing Filter) 8 - 54° 0 Frozen Food Waste Meat Packing Waste; (Roughing Filter) Coke Plant Ammonia Still, Final Cooler Mixture Domestic Sewage and Food Processing Waste Pulp and Paper; (Roughing Filter) 39 - 41°C Frozen Food Processing Wastes Paper Mill Waste, 78 MGAD and above Effluent from Sour Water Stripper Paper Mill Waste, 78 MGAD and below OOOOO WASTES .75 .56 .38 .30 .30 .27 .25 .20 .19 .18 .17 .17 .15 .13 .09 K / l‘k l‘k l‘k l‘k I‘L. l\ l\ I‘k l\ |\ l‘k .288 .047 .084 .006 .021 .061 .068 .066 .048 .031 .047 .021 CHAPTER V - DISCUSSION The data for this thesis were obtained from the operat- ing results of many different trickling filter plants, treating many different types of wastes from municipal to purely industrial waste. The filters used for comparison not only varied in height from three to fourty feet but also in the type of media and distributers used. Prior to the development of the new filter equations by Howland (13) and Schulze (14) there was no way of com— paring trickling filter plants having such different physical structure and treating wastes of such different chemical nature. Up to date the design of conventional filters was based exclusively on empirical criteria geared toward the treatment of municipal waste. Experience has shown that these empirical formulations could not be used successfully as a design basis for the treatment of industrial wastes° It is believed that the new equations as they are applied here not only make possible the comparison of dif- ferent types of filter plants but also of different types of wastes. This means that the value of the trickling filter constant K can be used as a parameter to indicate the degree of treatability of the particular waste, if the same physical tYPe of plant is used to treat the different wastes. 18 19 Once a series of K values has been established for a number of specific wastes these would provide a basis for a more rational design of trickling filter plants than has been available so far. In addition to the dosage rate, depth of media, and type and size of media there are other factors which affect the performance of a trickling filter. Prominent among these are temperature, and type of distributor. However with the data presently available it is not possible to properly evaluate these factors. CHAPTER VI CONCLUSIONS The data presented in this thesis demonstrate the value _KD/QE/B of the equation p = 10 for predicting the treatment that domestic or industrial wastes will receive on a trickl- ing filter. The depth and the hydraulic load are physical factors which can be clearly established. For the conditions analyzed K varied from 0.09 for the most difficult to treat industrial waste to 0.75 for municipal waste, and within a much narrower range for each filter or each specific waste. The newly established equation gives a better correla- tion between filters using open type plastic media than it does between filters using rock media. The author believes that this is due to the oxygen deficiency that occurs within the rock filter. This conclusion is supported by the results Showing the effect of depth on the performance of filters using conventional media. 20 10. ll. 12. CHAPTER VII BIBLIOGRAPHY Schulze, K. L., "EXperimental Vertical Screen Trickling Filter,“ Sewage and Industrial Waste, 28, 4, 458, (April 19577. Egan, J. T., and Sandlin, M., “Evaluation of Plastic Trickling Filter Media,” Industrial Wastes, 3, 4, 71, (August 1960). Horton, R. K., Porges, R., Baity, H. G., "Studies on the Treatment of Sewage and Textile Wastes by Recircu- lating Filtration. II Domestic Sewage on a Continuous Basis,‘l Sewage Works Journal 14, 4, 818, (July 1942). Ingram, W. T., ”A New Approach to Trickling Filter Design,‘ Journal of Sanitary Engineering, Division of ASCE, 82, 3A3, 999, (June 19567? Schulze, K. L., "Load and Efficiency of Trickling Filters," Journal of Water Pollution Control Federation, 32, 3, 245, (March I960). Fairall, J. M., “Statistical Analysis of Trickling Filter Data," Sewage and Industrial Waste, 24, 11, 1359, (November 1952). Rankine, R. S. I'Performance of Biofiltration plants by Three Methods? Proceedings American Sgciety of Civil Engineers, Volume 79, Separate No. 336. Eckenfelder, W. W., "Trickling Filtration Design and Performance," Journal of Sanitary Engineering, Division of A.S.C.E., 87,SA4, 33,(July 1961). Abdul-Rahim, I. A., Hindin, E. and Dunstan, G. R., "Research of High Rate Trickling Filter Efficiency,u Public Works, 91, l, 87, (January 1960). Howland, W. E., "Effect of Temperature on Sewage Treat- ment Process,“ Sewage and Industrial Wastes, 25, 2, 161, (February 1953). Sinkoff, M., Porges, R. and McDermott, J., "Mean Residence Time of a Liquid in a Trickling Filter,“ Journal of Sanitary Engineering, Division of A.S.C.E., 85, 3A6, 51, (November 1959). Bloodgood, D. E., Telezke, G. R., and Pohland, F. G., "Fundamental Hydraulic Principles of Trickling Filters,“ Sewage and Industrial Waste, 31, 3,243, (March 1959). 21 13. 14. 15. 16. 17. 18. 19. 20: 21. 22. 23. 24. 22 Howland, W. E., Pohland, F. G. and Bloodgood, D. E., "Kinetics in Trickling Filters," 3rd Annual Confer— ence on Biological Waste Treatment, Manhattan College, (ApriI_1960). F‘ Schulze, K. L., "Elements of a Trickling Filter Theory," 3rd Annual Conference on Biological Waste Treatment, Manhattan COIlege, (April 1960). Sorrels, J. H., and Zeller, P. J. A., “Two—Stage Trick— ling Filter Performance," Sewage and Industrial Wastes 28, 8, 943, (August 1956). Fairall, J. M., ”Statistical Quality Control in Waste Treatment," Sewage and Industrial Wastes, 25, 11, 1282, (November 1953). Schulze, K. L., “Trickling Filter Theory," Water and Sewage Works, (March 1960). Fairall, J. M., "Correlation of Trickling Filter Data," Sewgge and Industrial Wastes, 28, 9, 1069, (September 1956). "Recommended Standards for Sewage Works,“ Great Lakes- Upper Mississippi River Board of State Sanitary Engi~ neers (revised 1960). Becher, A. E., and Bryan, E. R., "Study of the perfor- mance of Dowpao HCS when applied to the Treatment of Settled Sewage at the City of Battle Creek, Michigan,“ A report of the Project Steering Committee, (June 1958). "Surfpac Technical Bulletin, Plastic Biological Oxida- tion Media," Dow Industrial Service, Division of The Dow Chemical Company. Rankin, R. 8., "High-Rate Filters of Biofiltration Type and Their Application to Biologic Treatment of Sewage," Volume 1 81010 ical Treatment of Sewage and Industrial Wastes 5y Mo abe and EckenfeIder. (I959) "Infilco Accelo Filter" catalogue by Infilco Corporation. Eden, G. E., and Melbourne, K. V., ”Radioactive Tracers for Measuring the Periods of Retention in Percolating Filters," Journal of Applied Radiation and Isotopes, 3. 172, (1950). 23 25. Horne, W. R., Hartman, H. E., Rinaca, U. S., and St.Clair, R. L., "Biological Treatment of Fine Chemical Plant Wastes," Journal of Water Pollution Control Federation, 8, 43, 833 (August 1962). 26. Unpublished data from Dow Chemical Company Pilot Plant Trickling Filter treating chemical plant phenolic wastes. 27. Unpublished data from Dow Chemical Company Trickling Filter treating chemical plant phenolic wastes. 28. Morton Frozen Food Wastes - Koroseal Filter (PVC), Industrial Water and Wastes (July - August 1962). 29. Unpublished Data from Dow Chemical Company using Surfpac to treat Domestic Sewage. 30. Personal communication from Mr. George Martin, Engineer Manager, Green Bay Metr0politan Sewerage District. 31. Courchaine, C. L., "Performance of Twenty-Five Trickling Filter Plants in Michigan". Michigan Department of Health Report.(1958). 32. Personal communication from the Michigan Department of Health. ”'I’I'I‘IIIIII’I‘IIIIIIIIIII’IIIIIIIIII'“