Color removal from pulp and paper wastewaters by coagulation

Color removal from pulp and paper wastewaters by coagulation

Water Research VoI. 9. pp. 853 to 856,Pergamon Press 1975. Printed in Great Britain. COLOR REMOVAL FROM PULP AND PAPER WASTEWATERS BY COAGULATION MEr...

405KB Sizes 0 Downloads 35 Views

Water Research VoI. 9. pp. 853 to 856,Pergamon Press 1975. Printed in Great Britain.

COLOR REMOVAL FROM PULP AND PAPER WASTEWATERS BY COAGULATION MEr~'r G. OLTHOF* and W. WESLEY ECKENFELDER.JR. Environmental and Water Resources Engineering Vanderbilt University, Nashville, Tennessee, 37235, U.S.A. (Received 13 November 1973) INTRODUCTION

It is expected that the effluent standards for 1983 will require color removal from pulp and paper mill wastewaters in the United States. Very little color is removed by a conventional biological treatment process. The only economically feasible method at present for color removal is coagulation. Although laboratory studies using several coagulants have been reported [1--4] the most popular coagulant at present is lime. Lime has proven in pilot plant and full scale operation to be able to remove more than 90% of the color 1.5--8]. The sludge is thickened, dewatered and burned for recovery. Large dosages of lime are necessary for effective color removal. The effluent is neutralized with flue gas from the incinerator, pH control in this neutralization step is very critical since CaCO3 forms colloidal crystals at a pH lower than 11. These are almost impossible to separate in a sedingnltation basin 1.9]. Carbon absorption and reverse osmosis are alternative treatment methods to reduce the color content in the wastewaters 1"10,Ill. The economics of these processes, however, precludes total treatment and has application only for the removal of residual color after coagulation. It seems reasonable to expect that the use of [Fe2(SO~). 7H20] and alum I'AI2(SO,h. 18H20] as coagulants would require lower dosages and would be competitive to lime if sludge disposal does not materially affect the cost. Since alum sludge is in general more difficult to handle 1"12] and alum is a more expensive coagulant, Ferric salts and lime will be in most cases the more attractive coagulants. This paper describes the results of laboratory coagulation studies with wastcwater from three pulp and paper mills for color removal with iron [Fe2(SO,)3 • 71420], alum and lime. Two plants used bleached kraft pulping combined with groundwood pulping for the production of newsprint (Plants I and II). The third plant (III) was an unbleached kraft paperboard mill. The study was aimed at determining for all three coagulants: the optimum pH, the optimum coagulant dosage, and thickening and dewatering characteristics of the sludges produced. The results of the experiments were used for an economic evaluation of alternatives.

EXPERIMENTAL

PROCEDURES

The color of the samples (and also of the supernatant after coagulation) was measured with a Slx'etronic 20 at a wavelength of 420 nm. The pH was adjusted to 7.0 + 0.1 prior to color determination. The absorbance was convetted Pt-Co units with a calibration curve. Jar tests were run to determine the optimum pH range for coagulation with iron and alum. The optimum coagulant dosage was determined for iron, alum and lime. The optimum coagulant dosages were used to obtain sludge for further study. Sludge tests included thickening and vacuum filtration. RESULTS (a) Wastewater characteristics The wastewater characteristics of Plants L II and III are tabulated in Table 1. Significant amounts of suspended solids were present in the mixed wastewaters of the bleached kraft and groundwood pulping mills while the kraftboard plant had no settle,able solids. The alkalinity of the wastewaters is low. This means that the pH will drop rapidly after the addition of iron and alum. This can be advantageous if a low pH range is desirable for coagulation with these coagulants. The color of the settled wastewater is measured without filtration or centrifugation for removal of turbidity. The influence of filtration on turbidity and color measurement is shown in Table 2. The wastewater of Plant III still had 50°/, of the turbidity after filtration which causes the measurement of "appeared color." Filtration through a membrane filter (pore size 0.45 tan) will remove practically all turbidity, but as reported 1.15] this will also remove up

Table 1. Wastewater characteristics Pllr~ter

Plant I Nixed Settled

pH • coo. q l l

mg/1 TOC, ~ 1 $4ttlNbht Sol td$, 5o1"m~/'l,u.. SoSp~ded moll Total Solids, m g l l Tvrbtdlty, APHAuntts COlor, Pt-Co untts Alkatint t¥ as at/1 CaCO3 H4Pdr~sS aS m9/1 C4¢03

* Current address: Lancy, Nelienople. LA. 16063, U.S.A. 853

8.4 l~

Plant |! N l x a d Settled

8.4 776

8.8 1.o

262 70 500 1710

1322

8,8 ,so

Plant III m~ S~tleO 8.2 SOe

101 14S 50 580 1420

8.2 456 8) 14S

0 . 838

87 784

790

6tl

72

62

2250

1470

94O

147

105

74

206

128

85

~5-:

NIE[yI G OLTHOFand W. ~ESLEY ECKENFELDER, JR.

Table 4. Required coagulant dosage for color removal

Table 2. Percentage color and turbidit~ remaining in sample after filtration Percen:a~e cclovr rer;~in'.n.~

Pi~.;

$1msa £ibe: f!l;er

~.emb~ame filler (C.~ ~,

?e.--~'en~.s~-~%urbid!ty z-ema~trl~ ~mbz~..e filter

31~.~ f i ~ r fil;er

65 53 53

4~

22

Io

~

42

2~

8

~II

66

35

*~

8

I

to 602/0 of the "'true" color. A pore size of 0.8/,an seems to be ideal: all turbidity is removed while the true color passes through the filter [15]. (b) Color removal by coagulation Before determining the required coagulant dosages the optimum pH range for coagulation with iron and alum was determined. Table 3 summarizes the results. The optimum pH ranges for iron is about one-half a pH unit lower than for alum, For Plant III the optimum pH range is about one pH unit higher than the other two plants. Good removal etficiencies were also obtained at high pH (8.5-9.5). The poorest results were obtained at a neutral pH range. The optimum coagulant dosages with the corresponding percentage color removal are tabulated in Table 4. All three coagulants effect good color removal. The final value for the true color in the effluent was 120-200 Pt-Co units for all three wastewaters. The required lime dosage is about 3--4 times that required for iron and alum. The required coagulant dosage increases with the color content of the wastewater. Turbidity is removed to less than 10APHA units. (c)

[email protected]: Number

Removal of organics by coagulation

i ZZ lit

i-~n %removal

~/] 5,30 Z75 250

?Z }i ~5

~,']

Alum ~ removal

400 250 250

Lime ~ eemoval

~(1

92 93 9I

]~O0 TO~ I O00

92 95 85

Table 5. COD removals at the optimum coagulant dosage for color removal Iron mg/l % removal coagulant

Alum ~g/I ~ rmaoval coa~u,lant

Lime mg/l % removal coa2ula~t --

[

500

60

400

1500

II Ill

275 250

53 53

250 250

Plant Number

53 48 a~

1000 ~000

38 45 40

after settling, the dewaterability and the cake solids after filtration. The sludge volume at the optimum coagulant dosages for color removal after 30 min settling time are shown in Table 6. Lime and alum yielded the smallest sludge volume for Plant III. The data in Table 6 provides a relative picture of the sludge volumes to be expected with the three coagulants. For Plant I coagulation studies were run with iron at high (10-11) and low (3.5--4.5) pH values. The required optimum dosage was the same for both ranges; however, the sludge volume was 25% less at the high pH range. This can be explained by the lime additior/to raise the pH. (b) Sludge thickening Sludge settling and thickening tests resulted in the design criteria as shown in Table 7. The first column shows the suspended solids concentration of the wastewater as it enters the clarifier/thickener (C3 and of the underflow (C,). Lime sludge will thicken to the highest degree.

Coagulation accomplishes more than removal of color and suspended solids. A fraction of degradable and non-degradable organics are also removed. The advantage of using coagulation as the first unit operation as opposed to using it as a polishing step at the end of the treatment sequence is that the load Table 6. Sludge volumes and quantities resulting from on subsequent treatment steps can be reduced. The coagulation* COD removal by coagulation can be 38-.60%, Ir0n , Alum , .L|.m depending on the wastewater and the coagulant as daSa~ dodge ~ I , III Plant ~ u ~ r m9/l ml/I mq/l m111 m i ll ml/l can be seen in Table 5. The percentage COD removal l 500 200 400 230 1500 120 is based on the COD of the settled wastewater. Iron It 275 200 250 ZOO I000 180 gave in all three eases the lowest final COD values. IlL 250 It5 250 190 |()00 tOO The average BOD removal is lower. The percentages BOD removal were for Plant I, 25%; Plant II, 179/o; *Sl.dge Volume a f t e r 30 Minutes S e t t l i n g a t Optimum Coagulant Oesage and Plant III, 27%. Table 7. Hydraulic loadings on clarificr/thickener SLUDGE

CHARACTERISTICS

(a) Sludge quantities A substantial part of the total cost for wastc',cater treatment facilities is required for proper d ~ of sludges. Important features are the sludge volume Table 3. Optimum pH range for iron and alum for color removal Plant

Iron

Aium

I

3.5 - 4.5

4 . 0 - 5.'0

It Ill

3,5 - 4 . 5 4.5 - 5.5

4.0 - 5.5 5.0 - 6.5

c:t

at., z ZL"lm ,a. ~.,

ca

0.11

O.TO

0,23

4.0

o.,2 oao

"~.*:L.e"n 10,~ 9~l:i':t.x-

I : ~ , 1 ~ 10~1 tljJ-elmaex"

68O P94 772

103 130 5T7

~76o ~12o 1800

~o MO 495

59o 560 60o

72 46 57

zz

Z~

o.o3~ 0.?~

~

0.051

1.7?

~wZa.nt; IZZ Zz'on Llua ~

0,O92 0.074 0.172

0.5 0.4 1.4

Color removal by coagulation

855

The calculated hydraulic loadings for clarification Table 9. Cost equations for treating pulp and paper wastes by coagulation and thickening are shown in the Table. For Plant leon Alum Lime I, lime sludge produces the highest hydraulic loading ¢/I000 9al ¢II000 9al :/IOOQ cj~l and the highest underflow concentration. For Plants Pl ~tnt I 20.1 * O.21A 20.3 * 0.17B 15.8 4. 0.63C II and III iron provides a higher loading underflow 1 ~ l d S mgd 13,8 * 0.21A 13.9 * 0.178 9.Z + 0.63C concentration than the alum sludge. ZO ~ d 11.4 + 0.21A 11 .S + O.17El 6.6 ~" 0.63C Lime sludge for these plants will require a larger Plant IT 17.8 ÷ O.I1SA 19.4 * 0.1048 16.0 * 0.4ZC thickener than iron sludge but will result in a more 1 r~Jd 5 mgd 11.6 ~ 0.115A 1Z.3 * 0.I04B 9.4 * 0.4ZC concentrated underflow. The major difference in ZO mSd 9.5 * 0.11SA 10.~. * 0.1048 6.9 + 0.42C design loadings of the sludges for Plants I and II P l a n t 1 [ [ and the sludge of Plant III can be explained by the 1 ~Jd 1Z.3 + 0.I04A + 0.006C 1Z.8 ~" 0.1048 ~. 0.008C 16.0 ÷ 0.42C 7.7 + 0.104A + 0,006C 8.1 ," 0.1048 + 0,008C 9,~. ÷ 0.42C initial suspended solids concentration of the waste- S20aRd ~d 5.9 • 0.104A • 0,006C 6,5 * 0,104B • 0.008C 6,9 + 0.42C water. The mixed wastewater of Plants I and II conA: p r i c e Iron in S/ton B: p r i c e alura in S/ton rained about 500 nag 1- t of fibers while that of Plant C: petce lime in S/ton III was only 87 nag 1-t. So the percentage inorganic material in the sludge of Plant III is much higher and apparently results in a sludge that settles more tity to be dewatered is much larger than with iron and rapidly than sludges that contain a high percentage alum. part of this advantage will be offset. of fibers. (c)

Sludge dewateri~

ECONOMIC

Vacuum filtration was selected for the dewatering experiments in this study. Filter Ioadings can often be substantlaUy increased by the addition of polymers. Two polymers were used for reducing the specific resistance of the sludges. One was a cationic polymer (Primafloe C-7) and one was an anionic polymer (Primafloc A-10). Both are manufactured by Rohm & Haas, Philadelphia, Pa. The optimum polymer dosages were determined with the Buchner funnel test and are tabulated in Table 8. The dosages are expressed as mg I - t of wastewater as opposed to 1- i of sludge. In this case it is easier to compare dosages, and it is recommended to add the polymer at the end of the flocculation period to the total wastewater volume as it will improve the settleability of the sludges. Lime sludge dewaterability could in all three cases be improved with the acLdition of the anionic polymer. Primafloc C-7 was the best polymer for iron and alum sludges from the wastewaters with high suspended solids, while for Plant III the best polymer was Primafloc A-10. The filter loading based on the cycle time for the different sludges are tabulated in Table 8 together with the per cent cake solids one can expect after filtration. Lime sludge results in the highest filter loading in all three cases. Since the lime sludge quartTable 8. Vacuum filter loading I n f l ~ t Sol|ds CGocefltritlofl Ch %

pl,.tl

Opttmm Po|~aer Doslge ~/1

F t l t e e load L¢

lblftZ/hr

Percent Cake Solids 18

leon

0.7

2 C-7

0.40

Alum

0.9

i . S C-7

4.45

18

Lira Plant II

4.0

Z.S6

24

leon

O.S

I C-7

0.37

21

Alum

0.4

1 C-7

0.24

20

1.4

4 A-10

1.26

29 18

Llw4

Plqnt

4 A-IO

I11

leon

0.75

2 A-IO

0.21

Alum

O.SS

4 A-IO

O.lS

14

Lime

1.77

2 A-IO

1.54

27

EVALUATION

The cost for color removal from pulp and paper mill wastewaters have been estimated based on the results of the experiments with the wastewaters from Plants I, II and llI. The economic evaluation includes the cost for a solids contact clarifier, a vacuum filter and the coagulants with the polymers and chemicals required to operate in the optimum pH range. The treatment costs are evaluated for a 1, 5 and 20 million gal day- t plant. The cost information used is presented in references 16 and 17 for the solids contact clarifier and the vacuum filter. All costs are updated to 1973 levels and include capital, operating and maintenance costs. In Table 9 the results are presented with the price of the coagulants left as a variable. The price of coagulants depends on the location of the treatment plant since transportation costs can influence the price significantly. Reuse practices of the chemicals can also influence the price paid for the coagulants. The fraction of the treatment costs for the purchase of chemicals increases with plant size because no economies of scale are applicable as exist for the solids contact clarifier and the vacuum filter. The capital cost for the vacuum filter is in every case lowest for lime. However, the operating costs are based on the amount of sludge to be handled, and this is poorest for lime. Therefore the operating and maintenance costs of the vacuum filter are the highest when lime is used. Current prices for the three coagulants are substituted into the equations in Table 9 and the results are tabulated in Table 10. It can he seen that iron is the least expensive coagulant to use for wastewater fi'om Plants II and III and that treatment cost of wastewater from Plant I is about the same for iron and lime. Treatment costs will depend on the initial color and fiber content. The higher the color content the more chemicals are required. Fibers in the wastewater tend to produce a sludge which settles slowly, which will increase the costs for a solids contact clarifier.

)vIFI',T G OLTHOF and VV VVESLE~ECKE~kELDER. Jl~.

$5(>

Table i0, Cost of coagulation ~ith estimated prices tbr coagulants iron ¢II000 ~al

~]um c,"!CC0 la:

Lime :II000 ~a~

1 rr~d

28.5

30.5

23.~

5 ~d

e2.z

e-'.~

z~.8

20 mgd

19.8

2:.7

19.2

Plant .Ij 1 mgd

22.4

25.6

2a.3

S mgd

16.2

13.5

17.8

20 mgd

14.1

]5.5

15.3

1 mgd

16.6

1g,2

19.8

S mgd 20 mgd

12.0 9.2

14,5 12,9

16.0 14.3

Plan: I

Plant I l l

Estimated prices:

Iron - $40.O0/ton Alum - $60.O0/ton Lime - SZO.00/ton

DISCUSSION OF RESULTS The results of this study show that the use of iron salts for color removal from pulp and paper mill wastewaters can be an attractive alternative to lime treatment, The required coagulant dosage for iron salts is only 25-33~ of the lime dosage. This reduction in chemical use is the most important factor which makes iron salts the cheapest coagulant to be used in two out of the three mills tested. For the third mill, iron salts and lime gave comparable results. The economic evaluation did not include an attempt to calculate cost for recycling the coagulated chemicals. The cake after vacuum filtration can he incinerated and in the case of lime, the coagulant can be recovered and reused. Only about 10-20°~ of the chemical dosage has to he purchased for make up in this case. It is not yet known whether one has the same option if iron salts are used as a coagulant, This is probably the most important subject which should he researched as a continuation of this study. The advantage of incineration is not only the possibility of recovery of the chemicals, but at the same time it also solves the final sludge disposal problem. Some plants, however, will be too small to operate their own incinerators, and the filter cake will probably be trucked away for land disposal. In this case no chemical recovery is possible and iron salts are an economical, attractive solution for color removal. Another area which would probably he worthwhile to look into is the use of iron salts at high pH levels (8.5-9.5). The required coagulant dosage is about the same as for color removal at low pH levels (3.5--4.5), but advantages can he exl~Cted in the sludge treatment. The cost of lime for operation at this high pH has to be compared with the savin~ on sludge treatment facilities. Another aspect which is not reflected in the economic evaluation is the effluent quality. Lime will produce an diluent with a high pH and a great deal of calcium in solution. Common practice is to recarbonate the effluent which will reduce the pH and recover the calcium as CaCO3. This requires an additional treatment step, which may be difficult

because of the tendenc? of CaC03 to form colloidal crystals if the pH is reduced below I t. Iron salts and alum will produce an effluent with a slightly acidic character. The level of required neutralization (if any) will depend on the buffer capacity of the biological treatment. In the event that alkali addition becomes necessary, no separate sedimentation tank would be required as is the case with lime addition. Acknowledgement--This research has been supported by a grant from Cities Service Co.

REFERENCES [1] Thirumurlki D. et al. (1969) BOD and color removal from kraft mill wastes by alum. War Sewage Wks 12. [2] Smith S. E. and Christman R. F. (1969) Coagulation of pulping wastes for the removal of colon J. War Pollut. Control Fed. 41, 22. [3] Tejera N. E. and Davis M. W. Jr. (1970) Removal of color and organic matter from kraft mill caustic extraction waste by coagulation. TAPPI 53, 10, 1931. [4] Clarke Y. and Davis M. W. Jr. (1900) Color Removal From Pulp Mill Bleaching Waste. Dept. of Chemical Engineering, University of South Carolina, Columbia, South Carolina. [5] Davis D. L. Y. (1969) Tertiary treatment of kraft mill effluent including chemical coagulation for color removal. TAPPI 52, 11, 2132. [6] Gould M. (197l) Lime based process helps deeolor kraft wastewater. Chem. Engng 55. [7] Interstate Paper Corp. (1900) Color removal from kraft pulping effluent by lime addition, Env. Prot. Ag., 12040 ENC 12/71, War Pollut. Control Res. Set. [8] Oswalt Y. H. and Land Y. G. Jr. (1900) Color removal from kraft pulp mill effluents by massive lime treatment. Env. Pror Ag. Technol. Set. EPA-R273-086. [9] National Council of the Paper Industry for Air and Stream Improvement, Inc. (NCASY) (1962) A process for removal of color from bleached kraft effluents through modification of the chemical recovery system. N C A S Y Technical Bull. No. 157. June. 1"10] Morris D. C., Nelson W. R. and Walraven G. O. (1900) Recycle of papermill wastewaters and application of reverse osmosis. Env. Prot. Ag., 12040 FUB 01/72, War Pollut. Control Res. Set. [ l l ] Timpe W. G., Larry E. and Miller R. L. (19001 Kraft pulping effluent treatment and refuse, state if the art. Env. Pror Ag. Technol. Ser. EPA-R2-73-164. [12] Kreissl J. F. and Westrick Y. Y. (1972) Municipal waste treatment by physical-chemicalmethods. Applications of New Concepts of Physical-Chemical Wastewater Treatment. International Association on Water

Pollution Research. [13] Standard Methods for the Examination of Water and Wastewater. 13th edition, American Public Health Association [14] The Development of Design Criteria for Wastewater T r e a ~ t Processes, Vanderbilt University~ Nashville, Tenr~ssee, April (1973). [15] National Council of the Paper Industry for Air and Stream Improvement (NCASY) (1971) An investigation of improved procedures for measurement of mill effluent and receiving water color. Tech. Bull No. 253, December. [16] Burns and Roe, Inc. (1971) Process design manual for suspe'nded solids removal. Env. Prot. Ag. Technol. Trans. October. [17] Associated Water and Air Resources Engineers. Inc. (1973) Analysis of national industrial water pollution control costs. Env. Pror Ag. May.