Professor David C. Stuckey* Advanced Environmental Biotechnology Centre, Nanyang Environment and

Water Research Institute, Nanyang Technological University, Singapore.

AND *Department of Chemical Engineering, Imperial College London, UK.

Wastewater treatment plants as part of an

energy transition and resource recovery?

Keynote Address, IFAT, Munich, May 16th, 2018.

1) Background

- Water/Wastewater/Energy/Food (?) Nexus

- Paradigm shift - Single Pass vs “Circular Economy (CE)”, Resource Recovery

- Financial (market prices) vs Externalities, ie NPV vs LCA

2) Wastewater (sewage) Treatment

- what is removed and why?

- conventional treatment and future technology-energy/money efficiency?

- potential energy analysis of WWT-future technologies?

3) Wastewater (Industry) Treatment

- Food (easy), Biotech (Biopharma), Chemicals, Petrochemicals (difficult)

- Resource Recovery- water, energy, high value added (proteins)?

- “Bioaugmentation” for difficult compounds, eg PCP, PCE, PPCPs?

4) Global optimisation of treatment processes

- Are treatment processes as efficient as they could be-flowsheeting and LCA

5) Take Home Messages-Future?

- Anaerobic Digestion, N/P recovery, resource recovery (food?)-barriers???

- Pricing, Decentralisation? Apartment blocks?

OUTLINE

1) BACKGROUND • Crisis in energy currently water in some areas of the World future in food?

• Why? Market prices that do not “cost” goods accurately “ environmental

externalities”. Real cost based on C pricing, Life Cycle Analysis, “water footprint”.

• Recent paradigm shift to “Sustainable Development”, or “Circular Economies” and

Resource Recovery compared with current paradigm of “Single Pass”.

• Energy question shifted from availability and cost, to impact on Global Warming-

renewables – at the “tipping point” now?

• Water scarcity growing due to increasing populations, increasing incomes (

demand for better food, ie protein), and climate change (shift in where and how

much water precipitates). Water footprint of modern life increasing.

• Wastewater treatment processes used are 100 years old! Newer technology has

considerable potential, BUT resistance to change due to lifetime of concrete assets

(60 years), lack of innovation in this area, difference between Operating expenses

(OPEX) and Capital expenditure (CAPEX) means plants prefer to bleed to death!

Culture of operating plants-overdesigned (50x safety factor), poorly operated-lack

of modelling and control systems, compared to chemical processes?

2) Wastewater (sewage) Treatment Why treat?

1) remove bacteria/viruses to reduce spread of disease-Bow Street pump.

2) remove organics to improve river/ocean quality and smell.

3) remove N/P to stop eutrophication of water bodies algae, low DO

4) remove heavy metals, eg Hg, Cu, Zn, Cr and chlorinated organics

which are toxic/inhibitory and bioconcentrate up the food chain

5) remove endocrine disruptors (shampoo) to stop environmental damage

6) remove pharmaceuticals, eg antibiotics, personal care products

(PPCP) from damaging the environment-problem of ppb levels

7) water discharge standards getting tighter and tighter due to concerns

about the environment and human health

8) remove everything in aqueous phase (down to ppb) so that the

WATER can be recycled as drinking water! Thermodynamically useful?

Degree of removal of waste component

Cost p

er

un

it r

em

oved

Removal cost from WW is exponential-we can remove

anything but at what energy (financial) cost?

2) Wastewater (sewage-”used water”) Treatment • Aerobic “activated sludge” developed in the UK 100 years ago-still very

commonly used even now.

• Ponds where land is available, eg. Brazil, India, USA, Australia, BUT

often cannot meet strict WW discharge standards.

• Small scale plants can use “trickling filters” or rotating disc contactors.

• Any “biomass” (sludge) goes to anaerobic sludge digesters to reduce

the volume for disposal AND produce energy.

• Anaerobic digestion (AD) does not use air (early Earth) and produces

methane and carbon dioxide instead of water and carbon dioxide

because its biological metabolism is “inefficient”! Luckily this means we

can recover quite a bit of the potential energy embodied in ANY

wastewater. However, cells grow slowly, ie double every 4-5 days.

• This present opportunities but also problems for engineers!

7

“CONVENTIONAL TREATMENT TRAIN””

Raw

sewage

Screening Secondary

settlement

Final

treatment

Treated

Effluent

Biological

treatment

Primary

settlement

Surplus sludge disposal

Why was this process flowsheet chosen? Historical! - High energy demanding due to aeration-no resource recovery - High solids generating-Y~ 0.4 Takes lot of energy to dispose of sludge - Low organic loading rates~ 0.5 kgCOD/m3.d due to poor mass transfer

- Volatiles to atmosphere, AND CO2, CH4, N2O-GHG - Not “rational” today! ??? AD?

Digested sludge CH4

TYPICAL FLOWSHEET FOR SEWAGE TREATMENT

“activated

sludge”

aerobic Settles biomass

from clean water Settles solids from

sewage

Sand and grit

anaerobi

c

Clean

treated

water

Methane (~70%) and

CO2

2) Wastewater and Energy Content • WW treatment ~ 3% of the U.S. electrical energy. Net energy for domestic WWTP

using aerobic activated sludge and AD of sludge is 0.6 kWh/m3 of WW, about 1/2 for

aeration. With conventional aerobic treatment 1/4 to 1/2 plant’s energy comes from

biogas (CH4) produced during AD. IF full energy potential in WW was captured, and

less used for WWT, then WWT could become a NET energy producer.

• Table 1(following) summarises 3 energy-related characteristics of sewage: energy

contained in sewage; fossil-fuel requirements for N and P production; potential

energy from WW thermal content.

• Energy associated with N and P,∼10% of natural gas production used to fix

atmospheric N2 (Haber-Bosch) to satisfy agricultural demand for N. Less associated

with P production. Why do we use energy to remove these nutrients and then use

energy to replace them??

• Potential energy gained from thermal heat in WW, energy captured through heat

pumps for low-energy use (heating of buildings), sometimes used in cold climates

such as Sweden.

McCarty P. L. et al., Environ. Sci. Technol. 2011, 45, 7100–7106

11

METHANE + CO2

COMPLEX POLYMERSPROTEINS CARBOHYDRATES LIPIDS

AMINO ACIDS, SUGARS FATTY ACIDS, ALCOHOLS

INTERMEDIARY PRODUCTS

(propionate, butyrate, valerate, isovalerate)

ACETATE H2 + CO2

HY

DR

OL

YS

IS

AN

AE

RO

BIC

-O

XID

AT

ION

HOMOACETOGENESIS

ACETICLASTIC

METHANOGENESIS

REDUCTIVE

METHANOGENESIS

100 % COD

~ 21 % ~ 40 % ~ 5 %

~ 39 %

~ 34 %

66 %

20 %

34 %

~ 0 %

20 % 11 %

12 %

35 %

23 %

11%

34 %

? %

70 % 30 %

100 % COD

Clostridium,

PeptococcusBacteriodes,

Staphylococcus

Clostridium,

Micrococcus

Acetobacter,

Pseudomonas

Clostridium,

Pseudomonas

Syntrophomonas wolfeii,

Syntrophobacter wolinii

Clostridium aceticum

Methanosaeta,

Methanosarcina,

Methanospirillum

Methanobacterium,

Methanobrevibacterium

FE

RM

EN

TA

TIO

N

8 %

2) METHANE

FERMENTATION FROM

COMPLEX WASTE- many

biological steps taking

complex organics through to

small volatile fatty acids and

hydrogen which are then

converted to methane and

carbon dioxide by Archaea-

the oldest organism on

Earth. Hence the potential

energy of the organics

“trapped” in a usable form, ie

methane

What if we

treated ALL

WWs

Anaerobically,

and did not

use oxygen at

all?

McCarty et al., Environ. Sci. Technol. 2011, 45, 7100–7106

Very pure treated effluent stream (no bacteria/4-5 log removal of viruses-then need to post-treat to

remove N&P/PPCPs drinking water quality effluent after water treatment? Cost effectiveness??

Using ONLY AD to

treat WWs-quite

common in Brazil

(warm)

2) Comparisons with Conventional

Activated Sludge Treatment.

• Figure 2a-with complete AD treatment a doubling of

CH4 prodn. over conventional aerobic treatment, and

energy production > energy needs for plant operation

(Figure 2c)-net energy producer. Digested sludge

from AD 60-80% less than aerobic (Figure 2b), so

less energy and cost.

• Anaerobic fluidized membrane bioreactor (AFMBR)

combines membrane with anaerobic fluidized bed

reactor (AFBR).

• Dilute wastewater (500 mg COD/L) at HRT of 5 h-

total energy was 0.058 kWh/m3 wastewater treated,

about 1/10 of energy required for a typical aerobic

membrane bioreactor. Effluent COD of 7 mg/L (99%

removal) and less than 1 mg/L of suspended solids.

• Life Cycle Analysis (LCA) may improve environ.

impact

• Climate change-dissolved CH4 in effluent be collected. Energy for gas stripping CH4

~0.05 kWh/m3, can also use solid phase membranes, reactive extraction, biodegrad.

• Sulfate (SO42-) reduction to sulfide (S2-), which competes with CH4 production and

produces a toxic and corrosive gas (H2S). Can add air/oxygen to AD to reduce SRB.

• Removal of nutrients N&P. Nitrifification/denitrification energy consuming and

wasteful. Less energy using Anammox (oxidizes NH3 with NO2- N2 gas). Does not

need organics for denitrification, BUT does not RECOVER nutrients. Can RECOVER

NH3 with membrane reactive extraction, electrochemistry. Chemical precipitation for

P or its conversion into struvite (NH4MgPO4.6H2O) for recovery as fertilizer. Treated

wastewater used for crop/landscape irrigation- water and nutrients reused. Less

energy than for potable reuse where RO required (why? Social acceptance?).

• Anaerobic secondary treatment to reduce energy and operating costs for

municipal wastewater treatment has considerable potential, more pilot as well

as fundamental studies to better explore options for effluent CH4 removal and

to optimize treatment. What is stopping the spread of this technology??

2) Issues that need addressing with AD

ANAEROBIC REACTOR DESIGNS

1. CSTR- first plant in Germany in 1927. = 15-25 days >> [xmin]lim of 4 days for

35oC. Exeter, UK – gas streetlights from horse manure using AD before 1900!

2. CONTACT- 1955 in Chicago. Sludge settled and recycled, but problems due to

gassing. For dilute wastewaters (COD of 1.3 g/L). = 0.5 day. 90-95% removal

at a load of 2-2.5 kg COD/m3.d.

3. FILTER – upflow developed in late 60’s. COD from 375-12,000 mg/L in feed. =

4-36 hrs. Plastic media used to have tall high columns. Downflow when high

H2S levels which can kill the culture if fed from bottom.

4. EXPANDED/FLUIDISED- 1982 cells attached to solid media so high

throughputs, eg 30 minutes even with recycle. Some loads can be up to 120 kg

COD/m3.d. A few full scale plants.

5. UASB- 1980s granules of cells develop which enable upflow velocities of 2 m/hr

in the reactor. Very common in Europe and good for soluble wastewaters.

6. OTHERS- baffled reactor which is simple and achieves high removals with both

strong and dilute WW (1983); SAMBR uses submerged membranes in reactor-

high removals (>97%) with low HRT=<3 hrs (2006)

Engineering Design

evolved over time

from 20-30 day

HRT, to as low as

2-3 hr (AnMBR),

and even as low as

30 mins (Upflow

attached growth

fluidised bed)! Key

was to lower HRT

while keeping SRT

high- physical

separation of

biomass was key.

SAMBR

MAN!

Raw

sewage

Screening Secondary

settlement

Final

treatment

Treated

Effluent

Biological

treatment

Primary

settlement

Surplus sludge disposal

Raw

sewage

Screening

Treated

Effluent

Anaerobic MBR processes

Renewable energy (CH4)

3% Sludge yield

WHY USE A SAMBR?

CAPABLE OF “TURN UP/TURN DOWN” DURING PEAK FLOWS?

What other products can we recover from Sewage?

• Easy - BUT low returns- industrial water (low quality); energy (biogas),

• Harder – drinking quality water, eg Namibia, Singapore; nutrients (N, P, other trace

elements), eg Tianjin-small compared to fertiliser industry, Astara; PHA/PHB, but

low cost means separation is a big and costly problem! Also, very cheap product.

• Hard – Single Cell Protein ( food SCP) from ammonia recovered from sewage AND

hydrogen produced electrolytically from water (pilot plant in The Netherlands).

Recycling wastewaters for drinking water has the “yuch” factor (“toilet to tap”), ie

public acceptance, although this can be overcome through education. In the future this

has to be the way we go in many countries, eg Spain, Greece, Australia, SW-USA,

Africa.

Food from sewage -a “bridge too far” for now, and has issues around Public Safety.

Key issues-economics (which ones-NPV or LCA??)

RESOURCE RECOVERY IS VERY MULTI-DISCIPLINARY AND REQUIRES CLOSE

COLLABORATION BETWEEN MANY DISCIPLINES

MAYBE EASIER TO START WITH FOOD PROCESSING WW?

3) INDUSTRIAL WASTEWATER

• Food processing

- in Europe, brewing, soft drinks, food processing

- Easiest for energy recovery- no biological contamination, high strength easily

biodegradable, no toxins (metals, chlorinateds). Potential for Single Cell Protein

(SCP) production for animal food (fish?)

• Biotechnology

- biopharmaceuticals, eg antibiotics, dilute technology. Hence large (batch) flows

with mainly degradable organics.

• Chemicals/Petrochemicals

- challenging, but possible using both biological and physical/chemical combined.

Current Situation Little treatment and recycling in many industries, market prices of water below true

cost? A lot of Industry wastewaters discharged to sewer after partial treatment. Since

high water content like WW sludge, very high energy input then ash disposal problem.

Considerable energy/water recycling potential IF source separated and treated.

Different potential for different Industries;

4) GLOBAL OPTIMISATION Wastewater treatment plant (WWTP) synthesis

A range of promising treatment & recovery technologies-multiple choices

A number of possible interconnections- factorial choices of flowsheets

Trade-offs - economic viability vs. environmental objectives

ANNAMOX

24

OPTIMISING THE “POLYGENERATION” OF

OUTPUTS FROM THE TREATMENT OF WASTEWATER

Biorefineries-“Polygeneration is a relatively new concept which has the potential to

improve the efficiency of energy supply systems and, at the same time, integrate

the use of renewable energy sources to produce the required energy”.

In a WWTP we have many potentially useful treatment processes (Unit Operations),

and currently we do not know how to choose the “optimum” way of combining these

given a certain set of “boundary constraints”, ie what we want out in different

situations, including certain water standards. We use the idea of Polygeneration

with WWTP to maximise goals such as energy yield, GHG and solids minimisation,

and resource recovery, eg water, nutrients and proteins. Into this we incorporate

(with some complexity) LCA to minimise environmental damage-integration not

easy, BUT it enables us to ask “WHAT IF” questions.

What is Life Cycle Assessment (LCA)?

Phase 1 -LCA goal & scope

• Functional unit

• System boundary

• Impact categories

• Allocation approach

Phase 2- LCI analysis

• LCA data (energy and resource flow)

Phase 3- Impact assessment

• Classification

• Characterisation

• Normalisation, aggregation, weighting – optional

Phase 4 - LCA interpretation

• ‘Hotspot’ analysis

• LCA comparison

• Data quality analysis

• Other concerns (time horizon etc.)

(ISO, 2006)

Environmental assessment - midpoint vs. endpoint (UNEP/SETAC Life Cycle Initiative 2011)

An example – How to apply LCA to WWTP

Unit Process

Input Sand / fine inert

removal

Dewater

Hydrolysis (57-58ºC)

Digester

(Mesophilic 37ºC)

Air –Desulphurization

FeCl3 –Desulphurization

Water

NaOCl-for water treatment

Facilities

Electricity

Diesel

Output

Digestate-post-treatment

RDF and inert-landfill

Waste water-to drain

Waste water-recycle

Exhaust gas-bio-filter treatment

Decanter

CHP

Biogas recovery

Output

Renewable energy: Electricity

Biodegradable waste from

Ball Mill (MBT)

Water recycling

A wet (dry solid less than 15%), continuous-feeding multiple-stage digestion system operated at mesophilic

temperature

Parameters Inventory(statistical results not presented) Data sources

OLR of OFMSW 2.393g COD/L/day • A UK commercial AD

plant three month

operational data;

experimental results

• Waste and Resources

Assessment Tool for

the Environment

(WRATE) model

TSS 50.56±3.74 g /L

VSS 25.01±3.02g/L

Electricity for operation 1704.70 kwh/day

equivalent to 15.7% of the generated electricity

Thermal energy for

operation

3.37 MJ /kg bio-waste treated

Makeup water 150 m3/day

Equivalent to 3.34E-3 kg//kg bio-waste treated

Internally recycled water 150 m3/day

NaOCl 5.02 E-5 kg//kg bio-waste treated

Infrastructure Concrete, steel, cement, HDPE,

Life span 20 years

Exported surplus electricity 3.01 MJ /kg bio-waste treated

Exported heat 5.5 MJ /kg bio-waste treated

CO2 emissions 1.63 kg//kg bio-waste treated

NOx emissions 3.15 E-5 kg//kg bio-waste treated

Digestate cake 1.12 E-1 kg//kg bio-waste treated

(Guo et al., 2012)

Integration of LCA into WWTP Synthesis

GPS-XTM

(batch mode)

CapdetWorksTM

(batch mode)

Multi-objective

optimization

Regression

model

Performance

data

Costing data

LCA

Input-output flows Environmental profile

Validation

Validation

Detailed

design &

costing

Performance

model

Optimal

Superstructures

Input ranges

Configuration

Parameters

Superstructure modelling and optimization

Superstructure Synthesis problem

Given: A set of waste water streams

A set of water sinks and specification

A set of treatment & separation units

Determine optimal systems Units & interconnections

Flows and composition

Multi-objective optimization Maximize net present value (NPV)

Minimize environmental impacts for each performance

indicator kpi (𝐸𝐼𝑘𝑝𝑖)

𝑵𝑷𝑽 = 𝑺𝑨𝑳𝑬𝑺 − 𝑶𝑷𝑬𝑿

(𝟏 + 𝑹𝑨𝑻𝑬𝒅𝒊𝒔𝒄𝒐𝒖𝒏𝒕)𝒚𝒓− 𝑪𝑨𝑷𝑬𝑿

𝒍𝒊𝒇𝒆𝒕𝒊𝒎𝒆

𝒚𝒓=𝟏

𝑬𝑰𝒌𝒑𝒊 = 𝜶𝒌( 𝑬𝑰𝒇𝒄,𝒌𝒑𝒊𝑿𝒌→𝒋,𝒄𝒐𝒖𝒕

𝒄𝒌

𝑭𝒌→𝒋,𝒄𝒐𝒖𝒕 + 𝑬𝑰𝒇𝒓,𝒌𝒑𝒊𝒀𝒌,𝒓

𝒊𝒏 )

𝒓

(Puchongkawarin et al., 2015)

Case study – superstructure optimization results

Case 1 – Optimal superstructure Objective function : -13 M£

Bottleneck – abatement of TSS

GWP objectives mostly driven by

operation of activated sludge

(~80% N2O/CO2)

Beneficial effects of anaerobic

digestion (both NPV and GWP due

to energy recovery)

Case 2 – No anaerobic digestion (energy recovery) Objective function : -61 M£

Case 2 – No contribution from

anaerobic digestion and low

objective function is driven by

higher landfill cost

Minimal abatement in TSS

appears to be bottle neck

Case study – superstructure optimization results

OUTLOOK FOR SEWAGE AND

INDUSTRIAL WWT? • Early WWTP was cheap, used “basic technology”, and was reliable due to

overdesign. Our technology is NOT “Rocket Science”, even today, due to economics.

• Massive and fundamental changes in WWTP design and aims in the last 30 years

has started to shift the dominant paradigm from “Single Pass” to “Resource

Recovery” (Circular Economy), but this is very slow.

• Where are we evolving technically?? A difficult prediction-I think we will eventually be

using short HRT anaerobic processes (3-6 hr) to treat both sewage and Industrial

WW (even at low temps-8o C ) to produce energy and reduce sludge production to a

minimum. Post treatment (adsorption, reactive extraction, ion exchange,

electrochemistry) to remove and recycle N and P, and remove PPCPs to recyclable

levels. WWTPs will become energy “neutral” (exporting?) and generate income from

N&P exports. All these advances need increases in technical capabilities.

• What inhibits a rapid transition to a CE?? Background that virtually ALL technologies

have a half-life of market “penetration” of 20-40 years (exception in consumer

electronics), eg CAMBI-lab work in the late 70s, commercial in ~2005.

• Recent survey (Kirchherr et al., Ecological Economics, 150 (2018) 264–272) found that cultural barriers,

particularly a lack of consumer interest and awareness as well as a hesitant

company culture, are considered the main circular economy barriers by businesses

and policy-makers.

• These are driven by market barriers which, in turn, are induced by a lack of

synergistic governmental interventions to accelerate the transition towards a circular

economy.

• Meanwhile, not a single technological barrier is ranked among the most pressing

circular economy barriers, according to their research.

• Overall, their work suggests that circular economy is a niche discussion among

sustainable development professionals at this stage.

• “Significant efforts” need to be undertaken for the concept to maintain its

momentum. For example, government cross-subsidies from fossil fuels, C tax, RR

subsidy?

OUTLOOK FOR SEWAGE AND

INDUSTRIAL WWT (cont.)?

34

POSSIBLE ANAEROBIC FLOW-SHEET

What would a “new” WW flowsheet look like?

This flowsheet reduces energy use, produces very few solids, enables N and

P removal, and recovers water. Joint Cranfield/Imperial College study on

flowsheeting an AD plant.

TAKE HOME TECHNICAL MESSAGES

• Technology currently exists (which is not very sophisticated) which can reduce

energy inputs to sewage WWTPs to zero, or even to become a net energy exporter.

The quality of the water produced would be close to recyclable via a reservoir, and

maybe we could use a “water pinch” approach.

• Technology is in its early days at present to be able to recover and recycle N and P

from sewage. However, based on LCA it would be worth doing both.

• Industrial WWs can be treated and recycled, recovering both water, energy and at

times other valuable resources (food, polypeptides, fine chemicals). This would

have a significant impact on both water use, energy consumption, and cost.

• Treatment options should be based as much on LCA as market prices since the

later does not allow for the environmental impact of the process to be included.

• Flowsheeting of treatment trains using Global Optimisation to minimise LCA impact

(and maximise NPV) should be carried out on new WWTPs.

POLICY QUESTIONS

• Why is technical change so slow in the WWT sector? Few hard drivers like price,

competition in the sector, innately conservative due to the ramifications of failure?

• IF water prices were raised, how would water consumption react in terms of price

elasticity of demand? Similarly, if the COST of discharge were increased

significantly, would this enhance water recycling significantly?

• Cross cultural issues between disciplines in the field -primarily Civil/Environmental

Engineers with little background in reactor design, flowsheeting, separations (of

valuable products), chemical/biotech industry, Global Optimisation?

• Also, more proper studies need to be done on people’s perceptions of drinking

recycled WW-does this change, what is needed to help people understand issues?

• Recycling not their “core business” so Industry is not that interested, even if it

saves them money! Takes companies into technical areas they are not used to.

Thank you for your attention

QUESTIONS?