Engineering Conferences International
ECI Digital Archives
Wastewater and Biosolids Treatment and Reuse:
Bridging Modeling and Experimental Studies
Proceedings
Spring 6-12-2014
Modeling Reactions Between Activated Sludge
Fractions and Ozone to Optimize Biosolids
Reduction Processes
Dominic Frigon
McGill University
Follow this and additional works at: http://dc.engconfintl.org/wbtr_i
Part of the Environmental Engineering Commons
Recommended Citation
Dominic Frigon, "Modeling Reactions Between Activated Sludge Fractions and Ozone to Optimize Biosolids Reduction Processes" in
"Wastewater and Biosolids Treatment and Reuse: Bridging Modeling and Experimental Studies", Dr. Domenico Santoro, Trojan
Technologies and Western University Eds, ECI Symposium Series, (2014). http://dc.engconfintl.org/wbtr_i/31
This Conference Proceeding is brought to you for free and open access by the Proceedings at ECI Digital Archives. It has been accepted for inclusion in
Wastewater and Biosolids Treatment and Reuse: Bridging Modeling and Experimental Studies by an authorized administrator of ECI Digital Archives.
For more information, please contact [email protected].
Modeling Reactions Between
Activated Sludge Fractions and Ozone
to Optimize Biosolids Reduction Processes
Dominic Frigon
Microbial Community Engineering Laboratory
Department of Civil Engineering
McGill University
June 12, 2014
1
Introduction
• Cost of biosolids disposal is rising in North America due to
disposal options and environmental taxes.
– For our partner wastewater treatment facility, disposal cost more
than doubled over the last 7 years.
• In some Canadian jurisdictions (e.g., Province of Quebec),
landfilling of biosolids will be banned by 2020.
Thus, facilities want to reduce biosolids production.
• Air Liquide tries to open the North American market, but
imprecision in performance predictions remain an obstacle.
• New Canadian laws require proper nitrification even during
winter (water temperature <<10°C )
– Is nitrification affected by RAS-ozonation?
2
Schematics of RAS Ozonation Unit
Aeration Basin
Return Activated Sludge
Clarifier
WAS
Ozone
Contactor
3
Objectives
• Develop and validate a mathematical model to improve
performance predictions.
• Perform a global sensitivity analysis to understand the
impact of biological processes on biosolids reduction.
• Perform a scenario analysis on nitrification stability to
identify threatening operation conditions.
4
Concept of Ozone Unit Performance
• Technical/economic performance
Mass of Sludge
Reduction per day
Mass of
O3 per day
• True performance evaluation steps
Mass of
O3 per day
Mass of
Sludge Solubilized
Reactions
in O3 Contactor
Mass of Sludge
Reduction per day
Re-degradation
in Treatment Basin
5
O3 Contactor Performance by COD Solubilization
Note: total COD and total suspended solids (TSS) are related
COD Solubilization (%)
Solubilization:
g-COD/L
COD solubilization performance vary 1 to 3
Factors:
Solubilization Performance:
• Soluble COD
g-CODsolubilized/g-O3
• Bubble size
• Ozone transfer rate
Ozone dosage (g-O3/g-TSS)
Ozone dose: g-O3/L
6
Folodari, Andreottola and Ziglio (2010) Sludge Reduction Technologies in Wastewater Treatment Plants
1.0
500
0.8
400
Constant SRT
0.6
0.4
Sludge Prod. Ratio = e–4.85bO3
Sludge Prod. Ratio = 1 – 6.55bO3
300
200
Constant MLVSS
0.2
100
2
3
SRT
= 10
+ 103b
SRT
for
constant
MLVSS
O3 – 1221b
O3 + 20768bO3
0.0
0.00
0
0.05
0.10
0.15
0.20
Solubilization Rates (COD Solubilized/COD
Solids Inventory/day)7
bO3 (day–1)
SRT SRT
for Constant
MLVSS
for Constant
MLVSS(day)
(day)
Sludge Production
Biosolids Production
RatioRatio
CONTROL
OZONATED
Biosolids Reduction with Increasing Solubilization Rate
(COD Solubilized/Solids Inventory/day)
Model of Activated Sludge + RAS-Ozonation
Influent COD
Outlet COD (Effluent+Sludge)
Model development:
Effluent
• Description of Inactivation/transformation
• Determination COD Fractions
NonDegradable
NonDegradable
Substrate
Sludge
cBOD5
Yield (Y)
Substrate
Biomass
(1-Y) O2
(1-YN) O2
NH4
fsoluble,degradable
+ fparticulate,degrad.
Heterotrophic Inactivation process
O3
Biomass
Biomass Transformation
Nonprocess
Degradable Debris
NonDegradable
+
fsoluble,non-deg.
Substrate
< 5%
Soluble
Particulate
fmineralized
Nitrifying
Biomass
O3
Inactivation process 8
Pilot-Scale Parallel Reactors
O3 Contactor
(1 m3 Aeration Tank)
Distribution
Control
Well (HRT≈1min)
Reactor
Test
Reactor
Aeration Tank
(HRT ≈12 hours)
Waste from
Aeration
(SRT ≈7 days)
Portion
of RAS
Clarifier
9
Modeling Development and Validation
• Description of inactivation
– kinetics and stoichiometry
• Model prediction of pilot-scale experiments’ data
– Same installation studied in 3 different years
10
Modeling Effects of Ozone on RAS Biomass
SOUR/SOUR0
Is biomass inactivated at low O3 dose?
• Others reported dose threshold
• Others reported linear inactivation
Respirometer
Lab-scale
• Conclusion:
contactor
NO O3 threshold
Exponential inactivation
Inactivation occurs at low O3 doses
Does inactivation solubilize biomass?
• From literature: ozone solubilizes
cellular content…
• Conclusion:
Little COD solubilization upon
inactivation
Ozonator Scale/Solids source COD solubilization
(mg-COD/mg-O3)
Lab-scale: RAS
Pilot-scale
Fresh RAS
Lab-scale: Fresh RAS
Lab-scale: sonicated RAS
Pure culture: R. jostii
1.0
0.8
0.6
0.4
0.2
0.0
0
50
100
150
O3 dose (mg/L)
200
11
3 Pilot-Scale Experiments (3-5 months each)
Year 1
Year 2
Aeration
SRT
Ozone Dose
Biosolids
Reduction
+Ozonated Reactor (day)
g-O3
Control Rule
kg-VSS Inventory·day
Aerobic
6
0 to
0 to
+Constant MLVSS
6.5
46%
Aerobic
6
10.3
53%
+Constant SRT
Year 3
Phase 1 • Anoxic/Aerobic
Phase 2 • Aerobic
Phase 3 • Aerobic
+Constant MLVSS
• 12
• 12
• 6
• 7.3
• 8.9
• 11.4
(lower COD
solubilization)
• 22%
• 19%
• 18%
12
Year 1 – Calibration of Model
• Independent calibration of inactivation parameters
• Fitting of inventory with transformation parameters.
• Good fit of the observed state variable.
Total Solids Inventory
Ozonated reactor
Effluent Soluble COD
Control reactor
13
Improvements in State-Variables’ Predictions
for Different Descriptions of Inactivation
Linear
Exponential
Inactivation
Fractions: Inactivation
Same
Same
vs. Transformation
Predicted State-Variables Relative Squared Errors
Exponential
Independent
Conclusions
• Exponential inactivation works better
• Inactivation solubilizes <10% of cellular COD
14
Calibrated Parameters of Year 1 Satisfactorily
Predicted Year 2 and Year 3 Observations
Year 2 - Total Solids Inventory
Ozonated reactor
Year 2 – Nitrification Activity
Control reactor
15
Model Global Sensitivity Analysis
• Trends in biosolids reduction performance
• Generate implementation guideline
16
Daily Solubilized COD fraction (d-1)
0
BOD5/COD
SRT
0.25
0.1
Ozone particulate
substrate fraction
0.08
Influent
substrate…
0.6
Temperature
0.75
Ozone
inactivation…
High Sensitivity
Medium Sensitivity
Low Sensitivity
Ozone soluble
inert fraction
0.8
True yield
Constant MLVSS
Decay rate
0.4
Hydrolysis half
saturation…
0.06
Substrate half
saturation…
0.04
HRT
1
Influent COD
0.02
Maximum
growth rate
0
Maximum
storage rate
Sludge Production Ratio
0.5
Maximum
hydrolysis rate
Absolute Standardized regression coefficient
Sensitivity of Biosolids Reduction to Model and
Operation Parameters
1
Operational
Parameters
Sludge Production Ratio slope variation
Ozone Reaction
Parameters
Biomass
Parameters
17
Operation Parameters Most Influential of
Biosolids Reduction
Sludge Production Ratio
0.2
High Sensitivity
Medium Sensitivity
1
Constant MLVSS
0.8
0.6
0.4
0
0.02 0.04 0.06 0.08
0.1
Daily Solubilized COD fraction (d-1)
0
-0.2
-0.4
-0.6
Anoxic/Oxic vs. Temperature Influent COD
Aerobic
Degradability
Sludge Production Ratio
Standardized regression coefficient
Sludge Production Ratio slope variation
Initial
SRT
SRT
Sludge Production Ratio slope variation
1
0.8
0.6
0.4
0
Constant MLVSS
0.02 0.04 0.06 0.08
0.1
Daily Solubilized COD fraction (d-1)
-0.8
18
Nitrification Scenario Analysis
• Explaining “inconsistent” data on nitrification rates
• Identification of operation conditions threatening
nitrification
• Developing strategies to protect nitrification
Nitrifiers
Ammonia Oxidizers / Nitrite Oxidizers
NH3 (-III)
NO2- (+III)
NO3- (+V)
Change in Specific Nitrification Rates
• Anoxic/Oxic conditions less detrimental than fully aerobic
• Specific nitrifcation rates can increase in some cases:
influent TKN/COD < 0.1 g-N/g-COD
Specific Nitrification Rate
Change in Specific Nitrification Rates
(mg-N/(mg-VSS·day)
(ln[Ozonated/Control])
Year 1 Year 3 Year 3
Aerobic Aerobic Anoxic/
Oxic
Influent TKN/COD Ratio (g-N/g-COD)
20
Operation Conditions Threatening Nitrification
• Safety Factor (SF) = SRToperation/SRTmin
SRT min of nitrifiers =( μ ANO,max − bANO − bANO,O3)−1
• 3 simulation studies of 1,500 simulations:
variable reductions, 40% and 60% reduction
• Lower sludge reduction shows higher risk
4
Ln (SF /SF
)
O3
Ctrl
ln[SFO3/SFcontrol]
Variable
3
refrence line
Constant-40%
Constant-60%
2
1
0
Reduction in predicted
nitrification stability
-1
0
20
40
60
Biosolids Reduction (%)
Sludge reduction(%)
80
100
21
Operation Conditions Threatening Nitrification
(Constant 40% Biosolids Reduction)
Temperature
SRT
Temperature (°C)
ln[SFO3/SFcontrol)
Inactivation rate constant
Inactivation rate
constant (day−1)
SRT (day)
Potential problem: 32%
• Temperature < 15°C
and
• SRT < 16 days
and
• Inactivation rate
constant > 0.09 day−1
How to minimize inactivation rate?
• Biosolids production is function of
overall daily solubilization
• Overall daily solubilization:
– Solubilization in contaction
– Proportion of inventory treated
• Same factors for overall inactivation
Minimize proportion of inventory
Ozone
Contactor
Solubilization in Contactor
Inactivation in Contactor
Sludge Production Ratio slope variation
1
SOUR/SOUR 0
Sludge Production Ratio
Biosolids Production
vs. Overall Solubilization
Constant MLVSS
0.8
0.6
0.4
0
0.02
0.04
0.06
0.08
0.1
Daily Solubilized COD fraction (d-1)
0
250
500
750
O3 dose (mg/L)
1000
Pilot-scale RAS
Lab-scale: Fresh RAS
1.0
0.8
0.6
0.4
0.2
0.0
0
50
100
150
23
O3 dose (mg/L)
Conclusion
• Bioprocess conditions influence greatly performance of
RAS-ozonation units for biosolids reduction.
• Developed a model capable of predicting performance
based on inactivation and COD solubilization.
• Nitrification stability is generally enhanced.
• Nitrification can be negatively impacted at lower
temperature, SRT and higher overall inactivation.
– Contactor operation can be adjusted to alleviate problems.
24
Acknowledgements
• Students
Overall Operation
Siavash
Isazadeh
PhD
Pinar
Ozcer
PDF
Inactivation Sensitivity
Min
Feng
MEng
Luis
Urbina Rivas
MEng
Lab-scale reactors
Theresa
Luby
MEng
Shameem
Jaffur
MEng
Funding Organizations
• Regie d’Assainissement des Eaux du Bassin LaPrairie
• Air Liquide Canada
• Natural Science and Engineering Research Council of Canada
25