1

U. Zaher1 , C.O. Stockle1, S. Chen1 and P. A. Vanrolleghem2

Continuity Constrained Modeling of MultiContinuity Constrained Modeling of Multi--molecular molecular Transformations Transformations

forIntegrated MultiIntegrated Multi--scale Assessment of the Environment scale Assessment of the Environment

1Biological Systems Engineering, WSU, USA2modelEAU, Universite´ Laval, Canada

ACS NORM/RMRM, June 21, 2010, Pullman, WA

Good

ChallengeChallenge

Fair

TREATED WASTEWATER

INDUSTRY

Slide - 2Zaheru@wsu.edu

Poor

RAW SEWAGE

AGRICULTURE

MINING

2

OutlineOutline

Sustainabilitydecision support

LCA &

Bio-refinery

On-line Monitoring

& Control

Bio-remediation

Optimization

LCA & Economics

Integrated environmental systems

modeling

Environmental Biotechnology

Research

(III) Life Cycle Assessment

Slide - 3Zaheru@wsu.edu

& Control

Model based experimental design

Modeling kinetics

& biological systems

BiorefineryProcessDesign

Presentation’s methodology focus

Highlighted results

Classical Biological Model Formulation

Theoretical Chemical Oxygen Demand

, ,/ ( )n

kk in k j k j

j

dxQ V x x

dt

CSTR

CFD

TransportTransport

Slide - 4Zaheru@wsu.edu

ad- and desorptionDispersion

Advection

Diffusion

3

Continuity Constrained Modeling

BiochemicalChemical

Physical

l

, , 0j k j Compk

i , , , , ,with Comp Thod C N H O e

& other elements

Slide - 5Zaheru@wsu.edu

∆G <0

& other elements

S

N Particulate X0

HydrolysisNutrients release

On the Cell Scale: On the Cell Scale: Why continuity of nutrients !?

S NIn Vivo In Vitro

Soluble COD S1

VFA S2

CO2

X1

X2

Acidogenesis

H2

N t i t

Ex: solid waste degradation

Calibration

Slide - 6Zaheru@wsu.edu

CH4

Methanogenesis

X3Nutrients recovery

Nutrients uptake

Zaher et al., Applied Mathematical Modelling (2009), 33 (9), 3553–3564.Larger Scale

Experimental Design

4

22 1 2

12 2 21 1 1 1 2

4 3 22 1 3 2

2 1 2 31 3 2

1 1 2 1 2 3

2.303[ ].

[ ] 4 [ ]11

([ ] ) ([ ] [ ] )

[ ] 4 [ ] ( 9 ) [ ]

(4[ ] )

([ ] [ ] [ ]

l ma a a

i a j ai ja a a ai j

a a a a

a a a ak a

a a a a a a

H

H K H K KC K C K

H K H K H K K

H K H K K K H

H K K K KC K

H K H K K H K K K

21 )

n

k

S

N

Why continuity of charge ?Mono-protic Di-protic

k

Ac-+H+ HAc

CO2 + H2

pH=7

ATP

HAc

Ac-+H+

H+

pH>7pH<7

30

40

50

60

70

Ammonia

CarbonVFA

acit

y m

mol

/pH

/l

Tri-protic

Slide - 7Zaheru@wsu.edu

½ mol Glucose Fermentation

2ATP ADP + Pi

Zaher & Vanrolleghem, Analytical and BioanalyticalChemistry, (2005) , 383(4), 605-618.

0

10

20

3 4 5 6 7 8 9 10 11

Buf

fer

cap

pH

On the Pilot Scale: On the Pilot Scale: Example from Wine Industryultrafiltration membrane(0,14 µm)

Biogas

gasflowmeter

CH4/CO2 sensor

H2 sensor

cit

y m

eq

/l/p

H

VFABicarbonate

Temperaturesensor

Pump

1 m3Up-FlowFixed BedReactor Bu

ffe

r C

ap

ac

50

60

Slide - 8Zaheru@wsu.edu

Zaher et al., IWA, World Congress on Anaerobic Digestion (2004), Vol. 1, 330-336.

pump

liquidflowmeter

NaOH storage

OutputpH sensormixing

pump

0

10

20

30

40

50 60 70 80 90 100 110time h

CO

D lo

ad k

g/d

5

On the Full Reactor Scale On the Full Reactor Scale Bio-Physico-Chemical Complexity

Composites

Li id

CO 2

CH 4

H 2

H 2O GasOverload

Inerts

Proteins Carbohydrates Fats

MSAA

HAc, HPr, HBu, HVa LCFAAc -, Pr -, Bu -, Va -, HCO 3

-, NH 4+, LCFA -

Death/Decay

Liquid

Bio

chem

ical

Gas

lactate

Toxicity

Disintegration

Hydrolysis

Acidogenesis

Acetogenesis

DissociationVFA

Slide - 9Zaheru@wsu.edu

HAc H 2

CO2

CH4

H 2O

Microbes

3HCO-

NH 3 NH 4+

Physico-chemical

Gas

Toxicity

e.g. Ammonia, Cyanide,…

Acetogenesis

Methanogenesis Gas stripping

In view of the International Water Association (IWA) ADM1

Continuity Resolves the Complexity Maintaining the Bio-Physico-Chemical Cascade

Sva Sbu Spro Sac SIC SINSh2 Sch424 components

tio

ns

19 r

eact

Slide - 10Zaheru@wsu.edu

6

On the Scale of Process ChainOn the Scale of Process Chain

Anaerobic DigesterConventional Activated Sludge

50%CO97-80%

O2=Energy inputEnergy output

Aeration

COD100%

COD100%

50%CO2 CH4+CO2gy p

Settling

Return

Waste

Slide - 11Zaheru@wsu.edu

50% sludge3-20% sludge

Transformer models: Transformer models: Continuity for integrated assessment

Influx Outflux

Slide - 12Zaheru@wsu.edu

, , , , , , ,0j k j Comp with Comp Thod C N H O ek

i , 1,

1

n

j k j k for k Pj

Influx

, 1,1

n

k j k j for k P P Qj

Outflux

Developed at UGENTUpgraded at WSUApplications: UGENT : Plant-wide modelingWSU : High solids digestion

7

The Activated Sludge Benchmark Simulation Model No.1

European project COST action 624

On the PlantOn the Plant--Wide scaleWide scale

IWA taskgroup on benchmarking WWT control strategies

European project COST action 624

TransformersASM1

Slide - 13Zaheru@wsu.edu

Sludge treatment facilities

ADM1

Developed at UGENT

Interface

AS IWA ADM1

On the treatment centers’ scaleOn the treatment centers’ scaleCo-digestion in Biogas plant

Anaerobic

Digester

Model ADM1

Input

Slide - 14Zaheru@wsu.edu

Waste Characteristics

8

Waste practical characteristics ADM1 Input

Transformer: general ADM1 interface

bohy

drat

es

Pro

tein

s

Lipi

ds

Car

b

Slide - 15Zaheru@wsu.edu

C

N

P

Co-digestion Model for Biogas Plants

Slide - 16Zaheru@wsu.edu

Zaher et al., IWA Publishing, Water Research (2009), 43(10), 2717-2727

9

Co-digestion optimization

WSU in cooperation withINRA, FranceLUND University, Sweden

Gas

flo

w (

m3 /d

)

Slide - 17Zaheru@wsu.edu

Kitchen waste ratio

HRT days

G

Whatcom

Energy production EIA

Soil fertility/pollution

Impacts

On the EcoOn the Eco--System ScaleSystem ScaleLife Cycle Assessment

GIS – Resource inventory Processing

Yakima

King

Okanogan

Grant

Ferry

Lewis

Chelan

Clallam

Kittitas

Lincoln

Stevens

PierceAdams

Benton

Klickitat

Whitman

JeffersonDouglas Spokane

Snohomish

Pacific

Skamania

Grays Harbor

Cowlitz

Franklin

Mason

Clark

Walla Walla Asotin

Garfield

Kitsap

Thurston

SkagitPend Oreille

Columbia

Island

San Juan

Wahkiakum

Thermochemical

Biochemical

Physicochemical

Interface Wastewater

Solid waste/Residuals

off-Gas

Treatment Models

Interface

Interface

Interface

Interface Feedstock

Air pollution

Bio-refinery Process Models

InI I

river water pollution

Yakima

King

Okanogan

Grant

Ferry

Lewis

Chelan

Clallam

Kittitas

Lincoln

Stevens

PierceAdams

Benton

Klickitat

Whitman

JeffersonDouglas Spokane

Snohomish

Pacific

Skamania

Grays Harbor

Cowlitz

Franklin

Mason

Clark

Walla Walla Asotin

Garfield

Kitsap

Thurston

Skagit

Whatcom

Pend Oreille

Columbia

Island

San Juan

Wahkiakum

MSW_paper

Slide - 18Zaheru@wsu.edu

Eco

nom

ic m

odel

s

Interface

Interface

nterface

Transportation cost Tipping fees

…..

Production economicsFuel price elasticity

…Treatment cost

10

Goal and Scope

Inventory Analysis

Interpretation

SignificanceContribution Consistency

LCA: Computational Framework

Impact Assessment

….Reporting

n

aa

aa ,11,1

Processes

cono

mic

Flo

ws

Products

Fuels

R

…

f

f

1

s

1

Demand e.g. 1000 gal ethanol

Developing at WSU

Slide - 19Zaheru@wsu.edu

nmnm

n

nnn

bb

bb

aa

,,

,11,1

,1,

E

cE

mis

sio

ns

Resources

Gases

Liquids

Solids

Hazards

nf ns

mg

g

1

ns

s

1

Inventoryvector

LCA: Computational Framework

Goal and Scope

Interpretation

SignificanceInventory Analysis

Impact Assessment

SignificanceContribution Consistency

….Reporting

Gases Liquids Solids …

g1

EmissionsInventory

vector Impact vector

Slide - 20Zaheru@wsu.edu

Impa

ct

Cat

egor

ies

Gases Liquids Solids …

m

j

g

g

g

1

mkjkk

mj

qqq

qqq

,,1,

,1,11,1

Global warming

…

Eco-toxicity

kh

h

1

11

LCA Uncertainties Optimization

Goal and Scope

Interpretation

SignificanceInventory Analysis

Impact Assessment

SignificanceContribution Consistency

….Reporting

Gl b l

Mont CarloMont Carlo SimulationSimulationaa

Slide - 21Zaheru@wsu.edu

fAs Inventory

fAs 1 Bsg Qgh … w

ImpactsGlobal impact

Optimization & Optimization & OED techniquesOED techniques

aai,ji,j

bbi,ji,j

Outline

Sustainabilitydecision support

LCA &

Presentation’s methodology focus

Highlighted results

Bio-refinery

On-line Monitoring

& Control

Bio-remediation

Optimization

LCA & Economics

Integrated environmental systems

modeling

Environmental Biotechnology

Research

(III) Life Cycle Assessment

Slide - 22Zaheru@wsu.edu

& Control

Model based experimental design

Modeling kinetics

& biological systems

BiorefineryProcessDesign

12

Development of High Solids Anerobic Digestion(HSAD) Systems

Seed reactor

Gasholder for the seed reactor

Gasholder for solids reactor

Concentrated liquidFor nutrient recovery

Slide - 23Zaheru@wsu.edu

California Energy Commission –WSU High Solids Anaerobic Digestion

Leaching compartmen

t

Solids Reactor

solids reactor

Digested solid wastes

Feed from Leachate

HSAD System Design Criteria

Threshold Optimizationfor process

Design/performance parameter Optimization Typical HSAD system

installation with solid waste

recycle using

(1)New HSAD system with

augmentation

(2)Conventional

HSAD system withfor process

designKompogas

design

augmentation system with solid waste

recycleTotal COD g/L 200 200Optimization Thresholdm3 CH4/ton/day

39.7 39.7

Optimization results for feed rate of 1 ton/day:Methane production efficiency 96% 96%Solids reactor volume m3 17 25 38.3*PerformanceS lid di t l di t 0 06 0 04 0 026

Slide - 24Zaheru@wsu.edu

Solids digester loading rate ton/m3/day (lb/ft3/day)

0.06(3.75)

0.04(2.50)

0.026(1.63)*

Biogas production rate m3/ m3/day 4.62 3 2.8*Methane production rate m3/ m3/day 2.28 1.52Potential fertilizer:

kgN/ton waste 2.10 -- --kgP/ton waste 3.72 -- --

capital cost $/ton including post composting

18.9 27.8 48.6

13

Cost and economic benchmarks(1)

New HSAD system with

augmentation

(2)Conventional HSAD system

with solids recycle

Annual Savings of the new HSAD system

US$ %

Total cost production $/kWh 1.0750517 1.5538188 0.4787671 31%kWh from food waste in

HSAD Systems Economic Analysis

California

kWh from food waste in California 1,229,000,000 1,229,000,000Total cost utilizing all food waste in California (annual savings) $1,321,238,572 $1,909,643,305 $588,404,732 31%kWh from all digestible waste in California 5,130,000,000 5,130,000,000Total cost utilizing all digestible waste (annual savings) $5,515,015,359 $7,971,090,442 $2,456,075,083 31%

rr : relative efficiency ~ 0.8: relative efficiency ~ 0.8 Mixing is 75% of the costMixing is 75% of the cost

Slide - 25Zaheru@wsu.edu

rr yy

SG: specific gravity 1.2SG: specific gravity 1.2

µ: viscosityµ: viscosity

G: mixing intensityG: mixing intensity

gg

California Energy Commission – WSU High Solids Anaerobic Digestion

Washington State Ecology dept.– WSU Energy and Fertilizers from Solids Waste

~35% CO2 : 65% methaneBoeing– WSU LCA of biofuel for aviation

Seed reactor

Gasholder for the seed reactor

Gasholder for solids reactor

Concentrated liquidFor nutrient recovery

CO2 for phototrophic

growth

Purified methane

Slide - 26Zaheru@wsu.edu

Leaching compartment

Solids Reactor

Digested solid wastes

Feed from Leachate

Extraction

Oil for biodiesel / hydrocarbons for

gasoline Note: multi-objective optimization

e.g. using genetic algorithms

14

To agricultural drainTo agricultural drain

Grit and Oil removal

Slide - 27Zaheru@wsu.edu

1,000,000 m3/day average flowfrom 17 km tunnel + 14.7 & 4 left station

Gabal El-Asfar WWT/Biorefinery Plant (GEAWWTP), Cairo, Egypt

GEAWWTP OptimizationNitrify or NOT?

What are the Env. Impacts?•Water pollution•Eutrophication?•Land use•…

What are the Economic Impacts?•Energy in/out•Capital cost•Operation costs

AT FC

RAS

F tE ( )

MLSSM

X

tFtF

T

toi

in

41

)()(

FSL(t)FW(t)FRS(t)

Layer/ zone (1)

Layer (2)

Layer (3)

Layer (4)

•Tariffs / profit•Sustainability

Slide - 28Zaheru@wsu.edu

• M.Sc. (2000), Cairo University, Egypt• Water Intelligence (2004), IWA publishing (on-line)

15

GEAWWTP Improvements

1. Capital cost savings on plant extension

2. Operation cost saving due to not nitrifying

3. Saving > 50 GWh/year

4. Biogas increase due to no NO3

-

5. Improved land use desert reclamation

6. Biodiesel from oil-plants BOT contract

1

2&3

4

5&6

Slide - 29Zaheru@wsu.edu

7. …. 4

A biorefinery…,indeed,

From Built to Natural Environment

Wheat

Slide - 30Zaheru@wsu.edu

Using the CropSyst®

model

16

0.8

[1] WW‐SB‐SW (high) ‐ CT

[2] WW‐SB‐SW (high) ‐RT

[3] WW‐SB‐SW (high) ‐NT

[4] WW‐SB‐FY (middle) ‐ CT

[5] WW‐SB‐FY (middle) ‐NT

[6] WW‐FY     (low) ‐ CT

[7]  WW‐FY      (low) ‐RT

[LCA Scenario   No.]  Rotation(Rain fall zone) ‐ Tillage  ‐‐‐>

N t i i

0

0.2

0.4

0.6

g CO2 e ha‐1  y ‐1

Net emissions

Sequestration

Nitrous oxide emissions

Fertilizer production

Fuel consumption

Slide - 31Zaheru@wsu.edu

‐0.8

‐0.6

‐0.4

‐0.2Mg

Main MessagesMain Messages

• Considering continuity of elemental mass, chargeand energy advances modeling of biochemical multi-molecular transformation

• Applying the continuity constraints expandsmechanistic modeling to multi-scales from the celllevel to the ecosystem

• Integrated modeling on these multi-scales lead tosustainable environmental solutions

• Tackling water energy and climate challenges in such

Slide - 32Zaheru@wsu.edu

• Tackling water, energy and climate challenges in suchintegrated manner is the tripod that supportssustainable environmental decisions

17

QuestionsQuestions

Slide - 33Zaheru@wsu.edu