APPLICATION OF MEMBRANE BIOREACTOR FOR TEXTILE WASTEWATER TREATMENT: PILOT PLANT PROCESS MODELLING AND SCALE-UP

1.1) NECESSITY FOR MBR PRIMARY TREATMENT OF TEXTILE EFFLUENT

Pilot plant is important in the sense that it drastically improves our understanding of the functioning of the system for the given raw effluent characteristics and environmental conditions. It also helps to scale-up the results and optimize the size of the full-scale plant. In this respect, it is important that pilot plant tests shall be conducted before implementation of full-scale plant. This article describes how a MB pilot plant tests could be modelled based on the raw effluent characteristics, and the modelling results could be utilized for scaling-up. The selection of MB for te!tile waste water treatment is "ustified on the following lines# $% Membrane Bioreactor is an advancement over the conventional activated sludge process with the use of ultrafiltration or, microfiltration membrane, which helps to maintain high levels of M&'' concentration and better treated water (uality. )% *ctivated sludge process has been well documented, thoroughly e!perimented, and widely adopted for the treatment of industrial waste water, including that of te!tile waste water. +% ,hile conventional biological treatment systems may face problems due to# a. &ow -.M ratio /due to low levels of B01 in the raw effluent%2 b. 'ludge settleability in the clarifier2 c. 3rowth of algae due to higher levels of total dissolved solids in the raw effluent2 d. Increased operating costs due to low levels of M&'' concentration, increased aeration tan4 size, re(uirement for higher retention time2 e. Inability to withstand shocks, i.e., sudden changes in the raw effluent (uality2 f. e(uirement for more secured land fill area due to higher levels of sludge production2 g. Inefficient nutrient removal in the raw effluent that shall encourage growth of microbial organisms on the reverse osmosis membrane2 h. Poor treated water output (uality which shall impose operational restrictions on the secondary treatment system,

56P 607'8&TI73, 7ama44al.

www.ecpconsulting.in

Page $.$9

1.2 MODEL CALCULATIONS FOR PLANT DESIGN

1.2.1 Design principles
The pilot plant should simulate the full-scale plant. 6onsidering that the adopted full-scale plant implements MB with step-feed ano!ic.aerobic system$, the pilot plant configuration shall also follow the same line. In this regard, a single step-feed ano!ic.aerobic design for the pilot plant shall be employed. This means that there shall one ano!ic chamber followed by one aerobic chamber and a hollow fibre membrane chamber similar to the fullscale plant. The mathematical modelling of the biological processes for determination of various system parameters are described below in detail).

$ *ny redo! reaction involves an electron donor - electron acceptor pair. *s per definition, the electron donor is oxidized while electron acceptor is reduced. 3enerally, the terms anaerobic and ano!ic are used interchangeably to represent the absence of o!ygen in the waste water. :owever, there is an important difference as outlined here. 8nder biological o!idation.reduction processes, the terms anaerobic and ano!ic have different notions in the sense that anaerobic refers to the general condition where dissolved o!ygen is absent in the effluent, and any one or more of electron acceptors present in the waste water such as nitrate, nitrite and sulphate can act as electron acceptors. The term anoxic refers to the condition where only nitrate.nitrite are the only electron acceptors ta4ing part in the biological o!idation.reduction processes. 'ome bacteria can use o!ygen as electron acceptor when it is available, and in the absence of o!ygen the same bacteria can switch the respiration mode to utilize nitrate.nitrite as electron acceptors. These 4inds of bacteria are called facultative bacteria. These 4inds of bacteria can survive both in ano!ic as well as in aerobic conditions unli4e other 4inds of bacteria that can survive only in anaerobic or aerobic conditions. *s MB configuration involves ano!ic, aerobic, and membrane compartments with re-circulation from the membrane zone to the ano!ic zone, the ano!ic zone shall have very low levels of dissolved o!ygen /the condition chemically referred as sub-o!ic% brought bac4 by the recirculating effluent. Thus, facultative bacteria only dominate the scene in the MB . This technically means that in the ano!ic compartment, denitrifying bacteria shall dominate while, in the aerobic chamber other aerobic bacteria shall also present but with the dominance of facultative bacteria. This is an important factor to be considered while considering mathematical modelling of biological processes in the MB . The adopted configuration is meant to remove nitrogen in the ano!ic compartment, and B01 in the aerobic compartment. Thus, step-feed ano!ic.aerobic system considered here is suitable for modelling the adopted MB process. ) -or further discussion please refer Metcalf ; 5ddy /)<<+% ,astewater 5ngineering# Treatment and euse /-ourth edition%, Tata-Mc3raw :ill Publishing 6ompany &imited, 7ew 1elhi.
56P 607'8&TI73, 7ama44al. www.ecpconsulting.in Page ).$9

1.2.2 Design approach

The modelling design adopts complete-mi! activated sludge treatment process for B01 removal+. The te!tile waste water (uality utilized for modelling was obtained through analysis of the raw effluent /Table-$.$%, and also calculated using necessary e(uations. In order to design the MB , the values of 4inetic coefficients /refer Table-$.)=% have to be obtained through pilot plant test data and parameter estimation through computer aided simulation. In the absence of pilot plant test results, the design approach here is to adopt recommended values of parameters for industrial waste water treatment /listed in Table-$.)%, and obtain full-scale plant design details that approaches the design with actual pilot plant test results. Table-1.1: Textile waste water quality utilized for modelling S.No. $ ) + = @ > 9 A ? $< B01 sB01 /soluble B01% 601 s601 /soluble 601% T'' B'' TC7 7:=-7 *l4alinity b601.B01 ratio Parameter Value )>$.) $?<.> @=@.@ +?A.) >>< +?>.9 +< ).+ A@@ $.> Unit g.m+ g.m+ g.m+ g.m+ g.m+ g.m+ g.m+ g.m+ g.m+

+ Most e!haustive computer aided simulations for parameter estimation are carried out with the use of *'M$ model of the International ,ater *ssociation. -or this purpose, ,5'T and *D8*'IM computer programmes were developed and demo versions of the software are available over the Internet. This model has been advanced into *'M) and *'M+. -or all practical purposes of scaling up pilot plant test results, the complete-mi! activated sludge treatment process described here shall be enough. In the present model, nitrification and denitrification were not considered as the main ob"ective is to address 601.B01 reduction. -urther, introduction of nitrification.denitrification does not alter the model and estimated values significantly. = *dopted from Metcalf ; 5ddy /)<<+% ,astewater 5ngineering# Treatment and euse /-ourth edition%, Tata-Mc3raw :ill Publishing 6ompany &imited, 7ew 1elhi.
56P 607'8&TI73, 7ama44al. www.ecpconsulting.in Page +.$9

1.2.3

Design conditions and assumptions

$% -ine bubble diffuser with an aeration clean water 0) transfer efficiency E )AF. )% &i(uid depth for the aeration basin @ m. +% The point of air release for the ceramic difference is <.@ m above the tan4 bottom. =% 10 in aeration basin E ) g/m3. @% 'ite elevation is +<=.) m above M'& /?AA ft% >% *eration E <.=@ for B01 removal2 E <.?@, and diffuser fouling factor - E <.?<. 9% The 4inetic coefficients are as given in Table-$.) A% ' T for B01 removal E )< days ?% 1esign M&'' GT'' concentration E $<,<<< g/m3 $<%TC7 pea4.average factor of safety -' E $.@ Table-1.2: Activated sludge kinetic coefficients for heterotrophic bacteria at 20C5. Parameter
Ma!imum specific growth rate half-velocity constant Biomass yield, mass of cell formed per mass of substrate consumed 5ndogenous decay coefficient -raction of cell mass remaining as cell debris values m 4d Cs 8nit less 8nit less 8nit less $.<+ $.<+ $.<< $.<A $.<A $.<< $.<9 $.<= $.<<

oeffi!ient
m Cs H 4d fd

Unit
g B''.g B'' . d g b601.m+ g B''.g b601

"ange
+.<< @.<< <.+< $+.)< =<.<< <.@< <.)< <.)<

Ty#i!al $alue
>.<<> )<.<< <.=< <.$) <.$@

g B''.g B'' . d 8nit less

<.<> <.<A

@ The values are converted to +<6 using appropriate e(uations where ever necessary. > * value of @.< was adopted for the calculation based on advice from the supplier of MB .
56P 607'8&TI73, 7ama44al. www.ecpconsulting.in Page =.$9

To find the values of m, 4d, Cs at a temperature different from )<6,

1.2.4 Design parameters

The following parameters shall be considered for designing calculations. 'ome of them re(uire analysis of the effluent, while many of them could be determined using applicable e(uations. In the following Table-$.+, the list of all parameters and their values are provided. Table-1.%: &esign #arameters for modelling !al!ulations S.No.
$ ) + = @ > 9 A ? $< $$ $) $+ $= $@ $>

Parameter
Biodegradable 601 7on-biodegradable 601 5ffluent soluble 601 7on-biodegradable B'' Biodegradable particulate 601 Particulate 601 'oluble B01 'oluble 601 Inert T'' Duantity of waste water feed Biomass yield Influent B01 or, bs601 concentration 6oncentration of growth limiting substrate in solution :alf-velocity constant 5ndogenous decay coefficient 'olid etention Time b601 nb601 s601e nbB'' bp601 p601 sB01 s601 iT'' D H 'o ' Cs 4d ' T

Notation

Value
5stimated in the model 5stimated in the model 5stimated in the model 5stimated in the model 5stimated in the model 5stimated in the model *nalysed value /raw effluent% Through raw effluent analysis 5stimated in the model $@<< m+.day 5stimated in the model *nalysed value /raw effluent% 5stimated in the model ecommended value ecommended value )< days

Unit
g.m+ g.m+ g.m+ g.m+ g.m+ g.m+ g.m+ g.m+ g.m+ m+.day g B''.g b601 g.m+ g.m+ g.m+ day-$ day
Page @.$9

56P 607'8&TI73, 7ama44al.

www.ecpconsulting.in

Table-1.%: &esign #arameters for modelling !al!ulations

$9 $A $? )< )$ )) )+ )= )@ )> )9 )A )? +< +$ +) ++ += +@ +> +9 Ma!imum specific growth rate Mass of B'' in the aeration tan4 Mass of T'' in the aeration tan4 Biomass as B'' wasted per day 5ndogenous decay coefficient for nitrifying organisms m PG,B'' PG,T'' PG,bio 4dn Cg.d gB''.g B''.d g.m+ g.m+ g.m+ 4g.day 4g.day g.m+ g.m+ g.m+ g.m+.day g.m+ m+ bs 601 . g B''.d or, g B01.g B''.d day g.g g.m+ g.m+ ecommended value g B''.g B''.d

6oncentration of 7:=-7 in the influent flow that is 70! nitrified Influent waste water T'' concentration Influent waste water B'' concentration 7et waste activated sludge produced each day, measured in terms of total suspended solids *mount of B'' produced and wasted daily 5ffluent 7:=-7 concentration Influent TC7 concentration 7itrogen o!idized 'oluble substrate utilization rate T''o B''o PG,T'' PG,B'' 7e TC7o 70! rsu

Mi!ed li(uor biomass concentration in the aeration G tan4 *eration tan4 volume -ood to biomass ratio :ydraulic retention time -raction of cell mass remaining as cell debris Influent waste water T'' concentration Influent waste water B'' concentration B -.M fd T''o B''o

56P 607'8&TI73, 7ama44al.

www.ecpconsulting.in

Page >.$9

1.2.5 Equations and parameter relationships

Table-$.= provides necessary e(uations and parameter relationships that are utilized in the model calculations. Table-1.': (quations and #arameter relations)i#s
$ ) + = @ > 9 A ? PG,B'' E TC7 /Total C"eldahl 7itrogen% E organic I ammonia nitrogen JTC7 g.m+K PG,bio / E PG,B''% $.' T E /specific biomass growth rate% E H /rsu.G% L 4d Mass of M&'' E /GT''% /B% E /PG,T''% ' T Mass of M&B'' E /GB''% E /PG,B''% ' T B''.T'' E <.><$ -.M E /Total applied substrate rate.Total microbial biomass% E D'o.BG -.M E 'o. G JD.B E ] DH /'o-'% /$ 4g. $<+ g% ---------------------------------J$ I /4d% ' TK DHn /70!% /$ 4g. $<+ g% ---------------------------------J$ I /4dn% ' TK I /fd% /4d% DH/'o-'% ' T /$ 4g. $<+ g% ------------------------------------------------J$ I /4d% ' TK D /nbB''% /$ 4g. $<+ g%

$< $$ $) $+ $=

b601 E $.> /B01% /g.m+% nb601 E 601 L b601 /g.m+% s601e E s601 L $.> /sB01% /g.m+% nbB'' E /$-bp601.p601% B'' /g.m+% bp601 --------p601 E /b601.B01% /B01 L sB01% -------------------------------------/601 L s601%

56P 607'8&TI73, 7ama44al.

www.ecpconsulting.in

Page 9.$9

$@ $> $9 $A $? )< )$

iT'' E T'' L B'' J7ote# B''.T'' E <.A@ /range# <.A< L <.?<%K = /B.D% /day%

M$-/bp601 p601%NB'' Jbp601 p601K E //b601 B01% /B01 L sB01%% /601-s601%

nbB'' E E Temperature activity coefficient. 1esign ' T E )< days.
o

E D /'o L '% L $.=) PG,bio

1.2.6 Steps involved in the design calculations

1) Development of aste ater characteristics needed for design.

a% -inding b601# b601 E $.> /B01% E $.> /)>$.) g/m3% E =$9.? /g/m3% b% -inding nb601 /non-biodegradable 601%# nb601 E 601 L b601 E /@=@.@ L =$9.?% g/m3 E $)9.> /g/m3% c% -inding effluent s601e /assumed to be non-biodegradable%# s601e E s601 L $.> /sB01% E +?A.) - $.> /$?<.>% g/m3 E ?+.)= /g/m3% d% -inding nbB'' /non-biodegradable B''% nbB'' E /$-bp601.p601% B'' /g/m3%

56P 607'8&TI73, 7ama44al.

www.ecpconsulting.in

Page A.$9

bp601 ---------p601

/b601.B01% /B01-sB01% ------------------------------------ E /601 L s601%

/=$9.?.)>$.)% /)>$.)-$?<.>% -----------------------------------/@=@.@ L +?A.)%

$$).?@ E ---------- E <.9>>A /g/m3% $=9.+<

e% -inding the iT'' /inert T''% iT'' E T'' L B'' /g/m3% E />>< L +?>.9% E )>+.+ /g/m3%
2) Designing suspended gro th s!stem for "#D removal onl!.

a% 1etermining biomass production# P!,vss E DH /'o-'% fd 4d DH /'o L '% ' T ----------------- I ----------------------------/$ I 4d ' T% /$ I 4d ' T%

1efining input data for above e(uation# D E $@<< m3.day. H E <.=@+ g B''.g b601 'o E =$9.> g b601.m3 ' E effluent substrate concentration /g b601.m3% 6alculation of effluent substrate concentration /'%# Cs /$ I /4d% ' T ' E ------------------------' T /m L 4d% L $ Cs E )< g/m3 /Typical value from Table-$.)%

56P 607'8&TI73, 7ama44al.

www.ecpconsulting.in

Page ?.$9

m,T E m/T-)<% m,T E /@.< g.g .d% /$.<9%/+<-)<% E ?.A= g.g.d 4d,T E 4)< /T-)<% 4d,+< E /<.$) g.g. d% /$.<=%/+<-)<% E <.$A g.g. d 6alculating '# /)< g/m3% J$ I /<.$A g.g.d% /)< d%K M/)< d% J?.A=- <.$A% g.g.d L $KN 92 $9+.) E <.@+ g b601.m3 'olving for PG,B'' /$@<<% /<.=@+% /=$9.? L <.@+% /$ 4g. $<+ g% PG,B'' E ------------------------------------------------J$ I /<.$A% /)<% K /<.$@% /<.$A% /$@<<% /<.=@+% /=$9.? L <.@+% /)<% /$ 4g. $<+ g% I --------------------------------------------------------------------------J$ I /<.$A% /)<%K E >$.>$ I ++.)9 E ?=.AA /4g.day%
3) Determining the mass of $SS and %SS in the aeration &asin.

a) eter!ination of "#$%&& and "#$T&&: Mass E P! /' T% P!,B'' E ?=.AA 4g.d I D /nbB''% g.day E ?=.AA I J/$@<<% /<.9>>A% /$ 4g.$<+ g% E $.$@ 4g.dayK E ?>.<+ 4g.day PG,T'' E /?=.AA 4g.day%.<.><$ I $.$@ 4g.day I D /T''o L B''o% ' 15(.)( * 1.15 * +1500 ,--0 . /0-.()1 ' 55/.0( kg2da3

56P 607'8&TI73, 7ama44al.

www.ecpconsulting.in

Page $<.$9

b) Calculating the !ass of %&& and T&& in the aeration basin. i% /GB''% /B% E E /PG,B''% ' T /?>.<+ 4g.day% /)< day% E $?)<.>< 4g /PG,T''% ' T /@@+.?9 4g.day% /)< days% E $$<9?.=< 4g

ii% /GT''% /B% E E

4) Selecting a design '(SS concentration and determination of the aeration tan) volume and detention time using the %SS mass computed in step *3).

a% 1etermining the aeration tan4 volume using the relationship from step-/+b%. /B% /GT''% E $$<9?.=< 4g *t GT'' E $<,<<< g/m3 B E J/+>,?+<.A 4g% /$<+ g.4g%K $<,<<< /g/m3% E $$<9.?= m3 b% 1etermining the aeration tan4 detention time E B.D E /$$<9.?= m3% /)= h.day% /$@<< m3.day% E $9.9+ h c% 1etermination of M&B''# -raction B'' E $?)<.>< 4g B'' $$<9?.=< 4g T'' E <.$9 M&B'' E <.$9 /$<,<<< g/m3% E $9++ g/m3

56P 607'8&TI73, 7ama44al.

www.ecpconsulting.in

Page $$.$9

5) Determination of +,' and "#D volumetric loading.

a% 1etermining -.M ratio# -.M E /D'o.GB% E 4gB01 4g M&B''.d E J$@<< m3.day% /)>$.) g/m3%K J/$9++ g/m3% /$$<9.?= m3%K E <.)<= 4g.4g.d b% 1etermining volumetric B01 loading# B01 loading E /D'o B% E 4g B01 m3.d E /$@<< m3.day% /)>$.) g/m3% /$$<9.?= m3% /$<+ g/kg% E <.+@= kg/m3.d.
6) Determining the o&served !ield &ased on %SS and $SS.

a) 4bserved 3ield based on T&& 0bserved yield PG,T'' b601 removed E E E E E g T''.g b601 E 4g T''.4g b601 @@+.?9 4g.day D /'o-'% /$@<< m3.day% J/ =$9.? L <.@+% g/m3K /$ 4g.$<+ g% >)>.<@ kg/d.

Hobs,T'' E /@@+.?9 4g.d% />)>.<@ 4g.d% E <.AA@ 4g T''.4g b601 E /<.AA@ g T''.g b601% E /<.AA@ g T''.g b601% /$.> g b601.g B01% E $.=$> g T''.g B01

56P 607'8&TI73, 7ama44al.

www.ecpconsulting.in

Page $).$9

b) 4bserved 3ield based on %&& Hobs,B'' E B''.T'' E <.><$ E /@@+.?9 4g.d% />)>.<@ 4g.d% E <.AA@ g T''.g b601 E /<.AA@ g T''.g b601% /<.><$ g B''.g T''% E <.@+) g B''.g b601 E /<.@+) g B''.g b601% /$.> g b601.g B01% E <.A@$ g B''.g B01
-) .alculate the #2 demand.

E E E E

D /'o - '% L $.=) PG,bio /$@<< m3.day% J/=$9.? - <.@+ g% g/m3K /$ 4g.$<+ g% L $.=) /?=.AA 4g.d% =$).>) 4g.d $9.$?+ 4g.h
rate at average design flo rate.

/) +ine &u&&le aeration design 0 determining air flo

'0T E 0T -------------------------------------------------MJ/ 6'T% L 6,K.6',)<N - //T-)<%% '0T 0T 6'T 6, 6',)< E E E E E E 'tandard o!ygen transfer rate /lb 0).h%. *mount of o!ygen re(uired /lb.h%B01 for o!idation of B01. The value relating o!ygen saturation in waste water compared to clean water L given by the ratio of o!ygen saturation in waste water /6s waste water% to o!ygen saturation in pure water /6s pure water%2 a value of <.?@ L $ is applicable for waste water. 0!ygen saturation concentration corrected for altitude and temperature /mg.&%. 0perating dissolved o!ygen concentration /)-= mg.& is typical range for waste water plants2 ) mg.& is adopted here%. 0!ygen saturation concentration for pure water at )<6 /mg.&%
www.ecpconsulting.in Page $+.$9

56P 607'8&TI73, 7ama44al.

 Therefore,

E E E

atio of o!ygen transfer efficiency /0T5% in waste water to 0T5 in pure water / E <.=@ for fine bubble diffuser%. -ouling factor, which accounts for membrane fouling /typically <.>@ to <.?2 a value of <.? was adopted for the calculation%. *rehenius constant used to correct for the effect of temperature /T% J E $.<)= at )<6K9.

'0T

E +9.?<= /lb.h% --------------------------------------------------------------------MJ/<.?@ A.$?% L )K.?.<)N <.=@ <.? $.<)=/+<6-)<6% E E +9.?<= <.+)? $$@.)$ lb.h /@).)@A 4g.h%.

6onverting the '0T to 'tandard 6ubic -eet per Minute /'6-M% of air re(uired# '6-M E '0T ./>< air '0T5 <.)+% where, '0T5 E air >< <.)+ '6-M E E E E E 'tandard o!ygen transfer efficiency E <.)A /)AF of 0) actually dissolves into water%2 this is dependent upon the depth of submergence and the type of diffuser /for diffused aeration system%. air density /<.<9+ lb.ft+ at an altitude of ?AA ft and +<6 for Tirupur%. number of minutes per hour. pounds of o!ygen per pound of air. $$@.)$ lb.h./>< <.<9+ lb.ft+ <.)A <.)+% =<A.==) ft+.min /$$.@9 m+.min%

air can be calculated using the following formulaA#

9 5ffluent temperature at the aeration tan4 is assumed to be +<6 for the calculations. A The air density was found to be $.$>= 4g.m+ /).@9 lb.m+ or, <.<9+ lb.ft+% for an altitude of ?AA ft /+<$ m% for Tirupur, from eGtreme high altitude calculator at# http#..bpesoft.com.s.wleizero.!hac.OMEd
56P 607'8&TI73, 7ama44al. www.ecpconsulting.in Page $=.$9

air where, p M T

p.M./ T%

E E E E

*ir pressure at the given altitude and temperature. Molecular weight of dry air /M E <.<)A?>== 4g.mol%. 8niversal gas constant /A.+$=@ 4Pa.C.mol% Temperature in Celvin /for calculations the ambient temperature of +<6 /+<+.$@ C% is adopted%.

The calculated air supply re(uirement for the MB is $$.@9 m+.min />?=.) m+.h%.

Authors: '.5swaramoorthi, C.1hanapal and 1.'.6hauhan. This article has resulted from the earlier wor4 carried out by the first author at 5PI6 Jwww.epicin.orgK, 6oimbatore. Hour comments and suggestions on this article can be forwarded to# infoPecpconsulting.in Mobile#I?$-?==+?-=<A$A

56P 607'8&TI73, 7ama44al.

www.ecpconsulting.in

Page $@.$9