Zero Liquid Discharge (ZLD)Concept Evolution and Technology OptionsConcept, Evolution and Technology Options

Viatcheslav FregerWolfson Department of Chemical Engineering

Technion – Israel Institute of Technology, Haifa, Israel

“Z Li id Di h ” W k h G dhi J 27 28 2014“Zero Liquid Discharge” Workshop, Gandhinagar, January 27 ‐ 28, 2014

1ZLD Feb 2014

Outline

History and motivation History and motivation

Conventional ZLD

Hybrid ZLD

E i ZLD d ZLD lt ti Emerging ZLD and near‐ZLD alternatives

Outlook

ZLD Feb 2014 2

Some History ZLD sector was apparently born in 1970s in USA, driven by 

the regulator Tight federal regulations on salt discharge to surface waters 

introduced, especially, due to salinity problems in the Colorado River 

Regulations were mainly concerned with power plant discharges from cooling tower blowdowns and scrubbers (in the wake of previously introduced regulations on flu gas discharges)

Clean Water Act 1974,  revised 1977, 1982

First ZLDs installed were 500‐2,000 GPM units based on evaporation/crystallization 

Regulations are expected to keep tightening: new EPA’s guidelines (ELG) expected in 2017 and 2022 on various 

f d h ( h b )Freger ZLD Feb 2014 3

types of discharges (many have to be ZLD)Sources: GWI Report, 2009; G. Maller/URS, 2013

Current Drivers and Limitations Presently, the major driver for using ZLD are 

Environmental regulation on discharge of specific solutes (salt, toxic elements nitrate‐nitrite etc);elements, nitrate nitrite etc);

Water scarcity/water stress growing world‐wide along with still negligible rate of waste water recycling; 

Economics: recycled water becomes more affordable as the water supply from conventional sources becomes more expensive;

G i i l ibili d d i d f Growing social responsibility and education towards awareness of environmental issues

While ZLD cost is high in most cases, it might be a more economic solution g , gwhen waste needs to be transported in large volumes over long distances

Still ZLD has drawbacks, probably, the most significant are

Very high cost (both CAPEX and OPEX)

Custom‐design on case‐to‐case basis

Freger ZLD Feb 2014 4

Difficulties to deal with complex streams (e.g., petrochemical)

Current and Potential Markets for ZLD  

Treatment and recycling of industrial waste effluentsPower Petroleum and petrochemical• Power

• Synthetic fuels• Primary metals processing

• Petroleum and petrochemical• Oil refining• Steam Assisted Gravity DrainagePrimary metals processing

• Microelectronics• Chemical

Steam Assisted Gravity Drainage (SAGD) heavy oil recovery

• Cogeneration

• Pulp and paper• Coal mining

• Fertilizer• Solid waste (leachate and 

secondary sewage effluent)• Battery manufacturing• PVC manufacturing

Uranium mining

secondary sewage effluent)• Coal liquefaction• Ethanol production

• Uranium mining

Tertiary treatment of municipal waste effluents Inland desalination

Freger ZLD Feb 2014 5

Inland desalination…

Conventional Thermal ZLD Technology

The conventional ZLD is based on evaporation and crystallization operationscrystallization operations

Evaporation (MVC or live steam) usually aims at >90% water recovery y

crystallization may achieve 100% recovery

solids can be further dewatered on a filter‐press solids can be further dewatered on a filter press for landfill

Latent heat of evaporation  is partly recovered (especially, for MVC) 

Operational and capital costs are still very high d t hi h ti (20 40 kWh/ 3due to high energy consumption (20‐40 kWh/m3

vs. 2‐3 kWh/m3 in desalination), use of chemicals and expensive corrosion‐resistant materials.

Freger ZLD Feb 2014 6

MVC Evaporation (Falling Film)

Potential issues: - Tboil elevation (for MVC)

Prior removal of SS and Ca required

Freger ZLD Feb 2014 7

- Prior removal of SS and Ca required- Mg(OH)2 precipitation (scaling and corrosion)- High MgCl2 and CaCl2 solubility

Crystallization

Atmospheric Crystallization with Softening Pretreatment

(Tboil may be too high for MgCl2 and CaCl2)

Vacuum crystallization(lower Tboil, higher salt concentration)

Freger ZLD Feb 2014 8K. Jenkins et al/CH2M Hill, WaterWorld, 2013

Hybrid ZLD Technologies Due to the high cost there is a strong motivation to employ more energy‐

saving process to minimize the MVC/Crystallization share. (Compare with costs of desalination technologies:  RO << ED << Thermal.)

Reverse Osmosis* (RO) – rejects salt, passes water,  2‐4 kWh/m3

Nanofiltration* (NF) – similar to RO, but passes some salt

Electrodialysis* (ED) or ED reversal (EDR) – removes ion, costs i t di t t RO d MVCintermediate to RO and MVC

Natural Evaporation – slow, large footprints

Another possible motivation is presence of organics, volatiles, colloids etc., which complicates the treatment and water reuse. Available solutions:

Conventional bioremediation Conventional bioremediation

MBR/UF pretreatment 

Freger ZLD Feb 2014 9*RO, NF and ED will be covered in detail on 2nd day

ZLD Combined with RO RO is presently the best and most energy‐saving available technology for desalting. 

The purpose is then to use RO to recover as much water as possible before MVC. The ZLD cost drops as RO recovery increasesThe ZLD cost drops as RO recovery increases. 

The recovery in RO is however limited by 3 main factors

Osmotic pressure becomes too high for TDS ~ 80 000 ppm Osmotic pressure becomes too high for TDS   80,000 ppm Scaling by sparingly soluble salts (Ca, Mg, SO4, PO4, silica), maybe alleviated  

to some degree using anti‐scalants Fouling (by organics colloids biofilms etc ) Fouling (by organics, colloids, biofilms etc.)

90.0

100.0

uct BC -150

Cost of Brine Concentration for BWRO

60.0

70.0

80.0

nts

/m3

prod

u

BC -100

30.0

40.0

50.0

70 80 90 100C

en

BC - 50

Freger ZLD Feb 2014 10

70 80 90 100% Recovery

Glueckstern, Proc. 6th IDS, 2003

RO Limitations on Recovery

Brine Conc140

160180

, atm 2.5 atm

8.3 atm Conc

Scaling Onset

80100120140

pres

sure

20406080

Exit

conc

RO Recovery0

20

50 60 70 80 90 100

E

O %

Scaling Potential vs. Recovery

RO Recovery, %

Brine Osmotic Pressure vs. Recovery

Jv = Lp(P – LSI = SP/SPc ~ Cn, n ~ 2-5

Freger ZLD Feb 2014 11

Increasing RO Recovery: 2‐Stage RO/NF  A simple 2‐stage (different membranes & pressures used at each stage)

Interstage softening/precipitation (more chemicals used)

Rahardianto, et al., JMS 2007; EST, 2008; Des. 2010, Sanciolo et al., Chemosphere, 2008.

Interstage softening/precipitation (more chemicals used)

Sanciolo et al., Chemosphere, 2008.

Qf

Yprimary = 85%3000 mg/L

ProductYsec = 67%

18100 mg/L

0 05 Q

A/S19440 mg/L

FB0.15 Qf

0.05 Qf54000 mg/L

Brine Treatment

Freger ZLD Feb 2014 12

FB Crystallizer

High Efficiency RO (HERO) Process

• High Silica Water• Cooling Tower

BlowdownBlowdown• Tertiary Treated Effluent

(Sewage)• High/TOC Biologically

By removing Ca and carbonate hardness RO can run at

g g yActive Water

By removing Ca and carbonate hardness RO can run atpH >10.5 High pH creates a “cleaning environment” => low fouling High pH creates a  cleaning environment   > low fouling Silica solubility very high, hardness removed  => low scaling Salt rejection and flux are increased Salt rejection and flux are increased

Recovery >90% However high chemical costs add ~$0 13/m3 overall product

Freger ZLD Feb 2014 13

However, high chemical costs add  $0.13/m overall product

Source: FEMP Bulletin, DOE/EE – 0294; aquatech.com

ZLD Combined with ED ED is not limited by osmotic pressure and thus it can achieve a 

much higher recovery.T i ll ED d l i i hi h h RO b l h Typically, ED desalting cost is higher than RO but lower than MVC/crystallization. The optimal placement of ED is then between RO and evaporationbetween RO and evaporation.

Freger ZLD Feb 2014 14

Increasing ED recovery for ZLD As in RO, precipitation of sparingly soluble salts in the brine 

limits recovery. Proposed solutions includeOff k i i i ( d d) Off‐stack precipitation (seeded) 

EDM in place of regular ED

on (%

)

35

40

m O

vers

atur

atio

20

25

30

Time (Hrs)0 2 4 6 8

Gyp

sum

10

15

Time (Hrs)

Onset of precipitation

Freger ZLD Feb 2014 15

O se o p ec p oplace crystallizer in brine loop

R. Bond et al, 2011, Florida Water Res J; J. Gilron, Wetsus, 2013.

ED Metathesis

Formation of sparinglyFormation of sparingly soluble salts prevented

using a stack of 4-compartment units

Freger ZLD Feb 2014 16R. Bond et al, 2011, Florida Water Res J; T. Davis, USBR Rpt. 135.

RO+EDM+Off‐Stack Precipitation ZLD Process

T. Davis, USBR Rpt. 135.

Freger ZLD Feb 2014 17

Biological (Pre‐)Treatment( ) d l d l Removes TOC (most organics) as CO2 and sludge, may leave some 

recalcitrant organicss MBR/UF is significantly more expensive, but offers a smaller footprint / g y p , p

and a more robust process

Tirupur ProjectTirupur Project Textile Effluent, 54 MLD, 2007

S. Prakash, GWI, Barcelona, 2007

Ambur–VaniyambadiTannery Effluent7 MLD, 2007

Freger ZLD Feb 2014 18

Emerging and State‐of‐the‐Art ZLD Solutions

Several alternative technologies or hybrids are in use or being i d f ZLDexamined for ZLD. 

SPARRO (Seeded RO)

ARROW (O’Brien and Gere, 2007) – pH elevation + IX + RO

VSEP (by New Logic Rerearch Inc.) – membranes vibratedS (by e og c e ea c c ) e b a es b ated

HEEPM (by EET Corporation) – ED treats the feed to RO 

F d O i (FO) Forward Osmosis (FO)

Molecular distillation (MD)

Wind‐assisted intensified evaporation (WAIV)

Freger ZLD Feb 2014 19Mickley, WaterReuse Foundation, 2008

SPARRO Process

Developed for treating hard waste water from mining industry. 

Freger ZLD Feb 2014 20

Forward Osmosis FO is used today for treating produced water in oil industry 

(generating a larger volumes of waste water – no ZLD)FO d l i RO Vi bl l h FO was proposed as an alternative to RO. Viable only when a waste energy (heat or osmotic) is available.

Gases to 

Vaporcompressor

NCG 

Concentrated seawater

adsorb out

Diluted 

Proposed concept(M C t h t l 2005)

Feed seawater

draw solution

Freger ZLD Feb 2014 21

(McCutcheon et al., 2005) Concentrated draw solution

Product water

Sagiv and Semiat

WAIV (enhanced natural evaporation)

Evaporation ponds (EP) are widely used as part of ZLD, but their footprint may be excessively large.

WAIV may offer a 1/15 land and 1/3 CAPEX of EP for the same evaporation rate

2

1.5

2

ed

wind1

kWh/

m3

fee

Brine MgtROII

water0.5E

sp, k

ROI

Courtesy, Lesico Ltd.0

BC WAIV Evap Pond

Freger ZLD Feb 2014 22J. Gilron, Wetsus, 2013

Outlook

Efforts continue to find alternatives to energy‐intensive evaporator/crystallizer systems. 

Hybrids systems with increased recovery are and will be the dominant approach

Progress is being made in lowering capital costs; a total installed cost factor is down from 5 to 1.8‐2.

“… industry analysts predict a cumulative annual growth rate for recovery/ reuse systems in excess of 200% over the next d d f hi h i ifi t ti ld b t d fdecade, of which a significant portion could be accounted for by ZLD capacity. … The economic and regulatory climate is such that ZLD or near zero discharge is going to continue tosuch that ZLD or near zero discharge is going to continue to grow rapidly…” [G. Cope, “From zero to hero”, globalwaterintel.com]

Freger ZLD Feb 2014 23

Thanks and Acknowledgements

Prof. Jack Gilron (Zukerberg Institute for Water research) Ben‐Gurion University)(Zukerberg Institute for Water research), Ben Gurion University) Prof. Rafi Semiat(Chemical Eng Department Technion – IIT)(Chemical Eng. Department, Technion – IIT)

ZLD Workshop OrganizersZLD Workshop Organizers

Freger ZLD Feb 2014 24