prediction of soc breakthrough in granular activated
TRANSCRIPT
Prediction of SOC breakthrough in Granular Activated Carbon(GAC) Adsorption
Irene Slavik (2,3), Wolfgang Uhl (1,3)
(1) Norwegian Institute for Water Research (NIVA) Team Systems Engineering and Technology
(2) Wahnbachtalsperrenverband (3) Technische Universität Dresden Chair of Water Supply Engineering
Motivation
Raw water resources
Synthetic Organic Compounds
Removal of SOCs during
drinking water treatment
by activated carbon filtration?
Water treatment
Prediction of break-
through behavior of
SOCs during granular
activated carbon
filtration?
Water treatment using GAC
Breakthrough
Breakthrough prediction
1. Equilibrium
2. Kinetics c/c0
time
Predicting breakthrough: equilibrium
50
100
150
200
250
300
350
0 10 20 30 40 50
concentration c in mg/L
So
lid
ph
as
e c
on
ce
ntr
ati
on
q in
mg
/g
ncKq
Predicting breakthrough: kinetics
SFF cckn
qqkn SSKS
Filmdiffusion to the grain surface
Surface diffusion inside the grain
Two ways to predict breakthrough of new substances
Simulation
Equilibrium (Freundlich-parameters) Kinetics
Models:
• Polanyi-theory
• LSER
• normalized
Freundlich-
equation
Lab-scale filter test
• e.g. rapid small scale column tests (RSSCTs) Experiment: Model:
c/c0
t
c/c0
t
Objectives
Appropriate prediction of
breakthrough behavior possible?
c/c0
time
Approach
Simulation
Equilibrium (Freundlich-parameters) Kinetics
Model: Experiment: Models:
• Polanyi-theory
• LSER
• normalized
Freundlich-
equation
Prediction of breakthrough behavior
Lab-scale filter test
• e.g. rapid small scale column tests (RSSCTs)
Isotherm Freundlich coefficients from SOC’s properties
Models
• Polanyi-theory
• LSER
• normalized Freundlich-equation
to derive Freundlich-parameters
Data
• Literature data
• Own isotherm measurements
prediction of breakthrough
- Freundlich-parameters from models
- Kinetic models
Compounds investigated
Group of compounds
Alipha-X Ar / Ar-X Poly-FG Ar-Sul Amines Phenols Pesticides
All
Number of compounds
26 24 51 14 15 35 7 112
Correlation of isotherm data using the
Polanyi-model
m
eqeq
adsV
cSRTAV
qV
)/ln(lnlnln 0
S
V
TRAVK
m
lnexp0
mV
TRAn
1,0E-04
1,0E-03
1,0E-02
1,0E-01
1,0E+00
0 50 100 150 200 250 300 350 400
Vo
lum
e a
ds
orb
ed
in
cm
3 g
-1
Normalised adsorption potential in J cm-3
Amines obtained by [86]
EDTA, DDT and DDE
Atrazin and diuron obtained by [77]
polyfunctional organic compounds
Correlation of isotherm data using the
normalized Freundlich-equation
polyfunctional organic compounds
*
*
n
eq
eqS
cKq
*
1*
nSKK
*nn
1,0E-01
1,0E+00
1,0E+01
1,0E+02
1,0E+03
1,E-07 1,E-06 1,E-05 1,E-04 1,E-03 1,E-02 1,E-01 1,E+00 1,E+01
Am
ou
nt
of
so
lute
ad
so
rbe
d i
n m
g g
-1
Normalised concentration
DDT and DDE
Atrazin and diuron obtained by [77] and EDTA
Amines obtained by [86]
Comparison of adsorption isotherms obtained by theory and experiment
0
50
100
150
200
250
300
350
0 5 10 15 20 25 30 35
Ma
ss
so
lute
/mas
s c
arb
on
in
mg
/g
Concentration in mg/L
4-Nitrophenol
Isotherm experimental
Isotherm non-linear regression (FUV = 0.04)
Isotherm Polanyi (FUV = 1.10)
Isotherm modified Freundlich (FUV = 1.00)
Isotherm LSER (FUV = 0.04)
0
50
100
150
200
250
300
350
400
450
500
0 2 4 6 8 10 12 14
Mass s
olu
te/m
ass c
arb
on
in
mg
/g
Concentration in mg/L
Diuron
Isotherm experimental
Isotherm non-linear regression (FUV = 0.03)
Isotherm Polanyi (FUV = 0.76)
Isotherm modified Freundlich (FUV = 0.36)
Isotherm LSER (FUV = 0.04)
0
100
200
300
400
500
600
0 5 10 15 20 25
Mass s
olu
te/m
ass c
arb
on
in
mg
/g
Concentration in mg/L
2,4,6-Trichlorophenol
Isotherm experimental
Isotherm non-linear regression (FUV = 0.14)
Isotherm Polanyi (FUV = 3.56)
Isotherm modified Freundlich (FUV = 3.39)
Isotherm LSER (FUV = 1.85)
Can breakthrough be described using Freundlich coefficients from compound properties?
Models
• Polanyi-theory
• LSER
• normalized Freundlich-equation
to calculate Freundlich-parameters
Data
• Literature data
• Own isotherm measurements
Simulation of breakthrough
using calculated Freundlich-
parameters
Plot of breakthrough
as determined in experiment
Lab-scale filter test
Comparison
Breakthrough in SSCT from experiment and predicted using the normalized Freundlich equation
0,0
0,1
0,2
0,3
0,4
0,5
0,6
0,7
0,8
0,9
1,0
0 5 10 15 20 25 30 35 40
Filterlaufzeit [h]
c/c
0 [
-]
4-MP exp 3-NP exp 3-CP exp 4-NP exp 2,4-DCP exp 2,4,6-TCP exp
4-MP sim 3-NP sim 3-CP sim 4-NP sim 2,4-DCP sim 2,4,6-TCP sim
Phenols
Filter run time [h]
Results
Comparison of breakthrough at c/c0 = 0.1
Compound Exp. break-
through
Simulated breakthrough
Polanyi-model
LSER- model
normalized Freundlich-
equation
[h] [h] [Δh] [h] [Δh] [h] [Δh]
4-MP 3-CP 3-NP 4-NP 2,4-DCP 2,4,6-TCP
11 14 16 18 19 28
7 9 13 12 17 26
-36 % -36 % -19 % -33 % -11 % -7 %
9 12 14 12 16 19
-18 % -14 % -13 % -33 % -16 % -32 %
9 10 14 13 16 23
-18 % -29 % -13 % -28 % -16 % -18 %
-24 % -21 % -20 %
Results
Comparison of breakthrough at c/c0 = 0.8
Compound Exp. break-
through
Simulated breakthrough
Polanyi-model
LSER- model
normalized Freundlich-
equation
[h] [h] [Δh] [h] [Δh] [h] [Δh]
4-MP 3-CP 3-NP 4-NP 2,4-DCP 2,4,6-TCP
16 18 22 24 29 39
17 19 23 25 31 54
6 % 6 % 5 % 4 % 7 % 38 %
18 22 25 22 29 36
13 % 22 % 14 % -8 % 0 % -8 %
17 18 24 24 28 41
6 % 0 % 9 % 0 % -3 % 5 %
11 % 11 % 4 %
Conclusions
• Using the huge dataset and the models helps to
identify non reliable data
• Using Freundlich coefficients derived using
compound properties
breakthrough behavior is best described using
Freundlich coefficients using the normalized
Freundlich equation
- begin of breakthrough (10 %) was predicted
about 22 % too early
- 80 %-breakthrough is predicted well with an average
deviation of 8 %
Conclusions
SOC breakthrough prediction using Freundlich-coefficients determined using compound properties is appropriate for a conservative risk assessment
Thanks for your
attention!
Slavik, Uhl, Börnick, Worch
Assessment of SOC adsorption prediction in activated
carbon filtration based on Freundlich coefficients
calculated from compound properties
RSC Advances 2016, 6, 19587-19604