Available at www.sciencedirect.com

Review

M. Maurer , W. Pronk, T.A. Larsen

8600 Dubendorf, Switzerland

Article history:

3152

3153

3155

3.2.1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3156

ARTICLE IN PRESS

WAT E R R E S E A R C H 40 ( 2006 ) 3151 3166Corresponding author. Tel.: +41 1 8235386.3.2.2. Evaporation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3156

3.2.3. Freeze-thaw . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3157

3.2.4. Reverse osmosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3157

0043-1354/$ - see front matter & 2006 Elsevier Ltd. All rights reserved.doi:10.1016/j.watres.2006.07.012

E-mail address: max.maurer@eawag.ch (M. Maurer).3.1.1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3155

3.1.2. Storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3155

3.2. Volume reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31562. Composition of urine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3. Treatment units. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3.1. Hygienisation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3152Received 20 December 2005

Received in revised form

7 July 2006

Accepted 11 July 2006

Available online 1 September 2006

Keywords:

Urine treatment

Process engineering

Wastewater

Source separation

Sustainable wastewater treatment

Struvite

P-recovery

N-recoveryThe separate collection and treatment of urine has attracted considerable attention in the

engineering community in the last few years and is seen as a viable option for enhancing

the flexibility of wastewater treatment systems. This comprehensive review focuses on the

status of current urine treatment processes and summarises the properties of collected

urine. We distinguish between seven main purposes of urine-treatment processes:

hygienisation (storage), volume reduction (evaporation, freeze-thaw, reverse osmosis),

stabilisation (acidification, nitrification), P-recovery (struvite formation), N-recovery (ion-

exchange, ammonia stripping, isobutylaldehyde-diurea (IBDU) precipitation), nutrient

removal (anammox) and handling of micropollutants (electrodialysis, nanofiltration,

ozonation). The review shows clearly that a wide range of technical options is available

to treat collected urine effectively, but that none of these single options can accomplish all

seven purposes. Depending on the overall goal of the treatment process, a specific technical

solution or a combination of solutions can be found to meet the requirements. Such

combinations are not discussed in this paper unless they are explicitly presented in the

literature. Except for evaporation and storage, none of the processes described have so

far advanced beyond the laboratory stage. Considerable development work remains to be

done to optimise urine-processing techniques in order to create marketable products.

& 2006 Elsevier Ltd. All rights reserved.a r t i c l e i n f o A B S T R A C TSwiss Federal Institute for Aquatic Science and Technology (Eawag),Treatment processes for source-separated urine

journal homepage: www.elsevier.com/locate/watres

. . .

. . .

. . .

. . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3158

. . .

. . .

. . .

. . .

. . .

. . .

ion

. . .

. . .

. . .

. . .

. . .

. . .

. . .

. . .

. .

. .

. .

less than one percent of the total wastewater volume.

(Wilsenach

Urine-sou

leaves many

me

org

lism (Esche

wa

ing

2006; Medila

flexibility a

prototypic w

du

illu

in this pap

rec

wa

household t

with todays

lon

pro

T

sep

and Gujer (1996) suggested local storage and transport in

ARTICLE IN PRESS

( 2g it would take a market economy to develop smart mass-

duced technology to do exactly that at competitive costs.

A large amount of data for urine is available in the medical

literature. Urine from collection systems differs from thesehere are

aration otreatment is seldom discussed. As an example,

reatment of urine seems inefficient and expensive

technology, but nobody has really examined how

2. Composition of urineycling, but the potential of mass-producing goods for

stewatery the many different treatment options discussed

er, ranging from nutrient removal to nutrienttions (e.g. manned space flights) as well as on experience in

the treatment of other high-strength liquid waste products.ced market goods (Larsen and Gujer, 2001). Flexibility is well

strated bexperience gained with urine treatment in different situa-r-scarce cities in emerging countries (Huang et al.,

nski et al., 2006). Furthermore, it offers increased

nd a possible shift away from investments in

astewater treatment plants towards mass-pro-

in agriculture is the most obvious application, but industrial

usage or simple nutrient removal are other possible options.

In this paper, we give an overview of the available technol-

ogies for treating source-separated urine, drawing on thestewater management when applied in the rapidly expand-

andwateand van Wijk-Sijbesma, 2005), the use of urine as a fertilizerropollutants originating from the human metabo-

r et al., 2006) and new ways of more efficientBecause the composition of urine reflects the average

requirement of nutrients for plant growth (Heinonen-Tanskintioned above, it also promises better ways of removing

anic micengineering options.tion, denitrification and phosphorus elimination

and Van Loosdrecht, 2004).

rce separation presents many advantages, but also

open questions. Besides the obvious advantages

Frohlich, 2002) and finally, on-site treatment may be possible

in the future, provided that the technical difficulties can be

overcome (Wilsenach et al., subm.). In the present paper, we

concentrate on the possibilities and difficulties of the processSubstantial separation of urine at source would thus allow

nutrient recycling from a concentrated nutrient solution and

at the same time obviate advanced nutrient removal, includ-

ing nitrifica

sewers over night; the concept developed in Sweden is long-

term local storage followed by truck transport (Hanaeus et al.,

1997); in some pilot projects, multiple piping is tested (Peter-3.3. Stabilisation . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3.3.1. Introduction . . . . . . . . . . . . . . . . . . . . . . .

3.3.2. Acidification . . . . . . . . . . . . . . . . . . . . . . .

3.3.3. Partial nutrification . . . . . . . . . . . . . . . . .

3.4. P-recovery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3.4.1. Introduction . . . . . . . . . . . . . . . . . . . . . . .

3.4.2. Struvite (MgNH4PO4) . . . . . . . . . . . . . . . . .

3.5. N-recovery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3.5.1. Introduction . . . . . . . . . . . . . . . . . . . . . . .

3.5.2. Ion exchange . . . . . . . . . . . . . . . . . . . . . .

3.5.3. Ammonia stripping . . . . . . . . . . . . . . . . .

3.5.4. Isobutylaldehyde-diurea (IBDU) precipitat

3.6. Nutrient removal (P and N) . . . . . . . . . . . . . . . . .

3.6.1. Introduction . . . . . . . . . . . . . . . . . . . . . . .

3.6.2. Anammox Process . . . . . . . . . . . . . . . . . .

3.7. Removal of micropollutants . . . . . . . . . . . . . . . . .

3.7.1. Introduction . . . . . . . . . . . . . . . . . . . . . . .

3.7.2. Electrodialysis. . . . . . . . . . . . . . . . . . . . . .

3.7.3. Nanofiltration . . . . . . . . . . . . . . . . . . . . . .

3.7.4. Ozonation and advanced oxidation . . . . .

4. Conclusions and outlook. . . . . . . . . . . . . . . . . . . . . . . . .

Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. Introduction

In the 1990s, various European groups began working on the

same basic idea that separating urine at source could

promote the sustainability of wastewater management

(Kirchmann and Pettersson, 1995; Larsen and Gujer, 1996).

All these approaches are based on the fact that urine contains

most of the nutrients in domestic wastewater but makes up

WAT E R R E S E A R C H 403152also many challenges in connection with source

f urine. Once urine has left the body it becomes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3158

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3159

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3159

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3159

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3159

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3160

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3160

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3160

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3160

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3160

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3161

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3161

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3161

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3161

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3162

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3162

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3163

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3163

an unpleasant, smelly and unstable solution. It is locally

produced and the present practise of dilution with large

amounts of water is actually a perfect way of neutralising

many of the more unpleasant aspects of urine. Furthermore,

centralized wastewater management is a system with inter-

dependent actors, and changing even a small part of it is

extremely difficult (Larsen and Lienert, 2003). Finally, the

question of transportation has not yet been solved. Larsen. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3157

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3157

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3157

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3158

006 ) 3151 3166data, because (a) the composition is averaged over time and

user group, (b) chemical alteration occurs in a non-sterile

ARTICLE IN PRESS

ste

kpl

H [2 CH S CH urine

.26

9.0

793

720

76

650

770

98

837

400

28

1.0

ted

mn

ita

), [6

(2environment, and (c) dilution with flushing water adds

elements such as calcium and magnesium that can further

alter the composition.

The composition of stored urine from different collection

Table 1 Concentration of urine from different collection sy

Parameter Unit

Source Household

S [1]School S [1] Wor

C

Dilution[a] () 0.33 0.33 0

pH () 9.0 8.9

Ntot (gNm3) 1795 2610 1

NH4++NH3 (gNm

3) 1691 2499 1

NO3+NO2

(gNm3) 0.06 0.07

Ptot (gPm3) 210 200

COD gO2 m3 1

K (gKm3) 875 1150

S (gSm3) 225 175

Na (gNam3) 982 938

Cl (gClm3) 2500 2235 1

Ca (gCam3) 15.75 13.34

Mg (gMgm3) 1.63 1.50

Mn (gMnm3) 0 0

B (gBm3) 0.435 0.440

The dilution[a] by the flushing water of the collection systems is extrac

urine composition of fresh urine (non hydrolysed) is listed in colu

(SO42S), [c]: value measured in undiluted, fresh urine, without precip

Ronteltap et al. (2003), [4]: Jonsson et al. (1997), [5]: Udert et al. (2005a

WATER RESEARCH 40systems is listed in Table 1. Data from a medical source (Ciba-

Geigy, 1977; last column) is given as a comparison for fresh

urine. During storage under non-sterile conditions, the urea

present in urine is hydrolysed to ammonia/ammonium and

carbonate due to microbial activity. This causes a pH increase

from around 6 to around 9 and triggers the precipitation of

calcium and magnesium in form of carbonates and phos-

phates (Udert et al., 2003a/b). This is clearly visible in the low

Ca and Mg concentrations in the stored urine. Depending on

the urine collection system, more or less flushing water is

present in the collected liquid. This water dilutes the overall

concentration of the substances and is considered in Table 1

with a dilution factor that expresses the fraction of undiluted

urine in the collection tank. Only the composition from

reference [5] in Table 1 presents concentrations from un-

diluted urine and therefore deviates substantially from all the

other concentrations.

The heavy metal content in urine is generally low (Kirch-

mann and Pettersson, 1995; Jonsson et al., 1997; Ciba-Geigy,

1977). Specific concentrations, i.e. relative to phosphorus or

nitrogen, are relevant if we want to use urine or products

derived from it as fertiliser. The comparison with commercial

fertilizer shows clearly that urine is always at the low or very

low end compared to commercially available fertilizer or

manure. Phosphate fertiliser in particular can have a high

heavy-metal content depending on the rock phosphate used.

Thus, analyses performed by Rogowski et al. (1999) and

McBride and Spiers (2001) show that agricultural bulk

fertilizers can have cadmium concentrations of up to36gCd kgP1, which are several magnitudes higher than those

of typical urine.

An important excretion pathway for hormones and many

pharmaceuticals is via urine. This is actively prepared in the

? 0.75 1 1

9.0 9.1 9.1 6.2

3631 9200 8830

4347 3576 8100 463

o0.1 0 154 313 540 8002000

6000 10000

3284 1000 2200 2737

273[b] 331 505[b] 1315

1495 1210 2600 3450

2112 1768 3800 4970

18 0 233

11.1 0 119

0.037 0.019

0.97

from the information given by the publications. For comparison, the

[6].Legend: [a]: defined as Vurine/(Vurine+Vwater), [b]: only sulfateStion, [1]: Kirchmann and Pettersson (1995), [2]: Udert et al. (2003a), [3]:

]: Ciba-Geigy (1977).ms

Concentration

ace]

Workplace[3]

Household[4]

Workplace[5]

Fresh[6],[c]

006) 3151 3166 3153liver by enhancing the water solubility of organic substances

adsorbed, so that they can be removed in the kidneys and

excreted via the urine (e.g. Ritschel and Kearns, 1999). Urine

might therefore contain a majority of the dissolved micro-

pollutants excreted by humans and due to their mobility also

the one most prone to transport and relevant for aquatic

ecosystems. However, it has to be emphasised that so far

there are no systematic evaluation of the medical literature

published that would back up this point statistically. Excep-

tions are estrogens, where 80% of the natural estrogens and

67% of the artificial hormone 17a-ethinyl estradiol areexcreted via urine (review in Christiansen et al., 2002).

Recent research indicates that the toxic effects of pharma-

ceuticals are additive. Silva et al. (2002) made it very clear that

in complex mixtures, such as wastewater or urine, threshold

values are very difficult to set. Escher et al. (2002 & 2005)

showed that the toxic effect of a mix of pharmaceuticals,

each without any specific mode of toxicity (baseline toxicity),

can be estimated by adding up the toxic effects of the single

substances.

3. Treatment units

The unique properties of urine mean that a wide variety of

technologies may be used to treat it. We defined seven main

purposes of a treatment unit: volume reduction, P-recovery,

N-recovery, stabilisation, hygienisation, removal of micropol-

lutants and biological nutrient removal. For practical

ARTICLE IN PRESS

Ta

ble

2

Overv

iew

of

the

treatm

en

tm

eth

od

sd

iscu

ssed

inth

ep

ap

er

Hygiene

Vol.

reduction

Stabilisation

P-recovery

N-

recovery

MP

elimination

NutrientMP

separation

Nutrient

elimination

Solidification

Needofpre/post-

treatm

ent

Info

literature

Hygienisation

Storage

+o

oo

oo

oo

(+)

-+

Volumereduction

Evaporation

+++

+++

++

oo

o++

+o

Freeze-thaw

?+

o++

++

oo

oo

o+

Reverseosm

osis

?+

o++

++

oo

oo

++

Stabilisation

Acidification

+o

++

oo

?o

oo

o+

Microfiltration

+o

++

oo

oo

oo

o+

Nitrification

+o

++

oo

?o

o(+)

o+

P-recovery

Struvite

o++

+++

+o

++

o++

o++

N-recovery

Ionexch

ange

o+

oo

++

o+

o++

o+

Struvite

o++

+++

++

o++

o++

o++

NH3stripping

o+

oo

++

o++

oo

o+

Isobutylaldehyde-

diurea

o+

oo

++

o+

o+

o+

Nutrientremoval

Anammox

+o

++

oo

?+

++

(+)

+++

Others

+o

+(+)

o?

o+

(+)

(+)

+

Micropollutionremoval

Electrodialysis

++

++

++

o+

oo

o+

Nanofiltration

++

o+

oo

o++

oo

++

Ozo

nation

+o

+o

o++

oo

oo

+

Thecolumnsrepresentthegoals

thatcanbeach

ievedwithasp

ecificprocess;therowslist

thetech

nologicalprocess.Legend:o:noeffect,+:positiveeffect,++:strongeffect,:

notapplicable.

WAT E R R E S E A R C H 40 ( 2006 ) 3151 31663154

ARTICLE IN PRESS

cri

Fig

(2Fig. 1 General template for the des

Fig. 2 Storage (seeWATER RESEARCH 40purposes, each treatment unit is assigned one of these seven

main purposes and is discussed under the corresponding

section. The assignment is somewhat arbitrary, because most

treatments attain several goals and other purposes might

have prevailed depending on the priorities set by the

evaluator. Table 2 gives an overview of the treatment methods

discussed in the paper. The efficiencies for specific goals

(columns) are roughly characterised. In order to achieve a

specific treatment target, it may be necessary to combine

several unit operations. For reasons of brevity, such combina-

tions are not discussed here unless they are explicitly

presented in the literature.

Each urine treatment process is roughly summarized in a

little scheme (Figs. 214). A general explanation of these

schemes is given in Fig. 1.

3.1. Hygienisation

3.1.1. IntroductionUrine from unhealthy humans may contain pathogenic

organisms (Santos et al., 2004; Vanchiere et al., 2005) as well

as prions (Reichl, 2002). Furthermore, faecal contamination

can result in high counts of faecal indicator organisms

(Hoglund et al., 2002b). Schonning et al. (2002) estimated a

faecal contamination of 9.175.6mg lurine1 in the urine collec-

tion system that they investigated. Although the exposition

pathways and effects associated with these pathogens have

not been investigated in detail, hygienic risks associated with

source-separated urine have to be avoided.. 1 for explanation).ption of urine treatment processes.

006) 3151 3166 3155Technically, there are many ways of pasteurising or

sterilising any solution (heat, pressure, UV, etc.), but none of

these methods have been tested for urine. Only storage has

been explicitly investigated for its ability to reduce the

amount of pathogens in source-separated urine, although

many of the treatment steps discussed in this paper are

expected to have an influence on its hygienic properties

(see Table 2).

3.1.2. StorageStorage (Fig. 2) offers a possible way of reducing the potential

health risks from faecal pathogens (See Fig. 1 for explana-

tion). Three storage parameters influence this process:

storage time, temperature and pH. Hoglund et al. (1998,

1999, 2000, 2002a/b) investigated the decay rates of bacterial

and viral indicator organisms in stored urine collected from

households. From the inactivation curves for rhesus rota-

virus, Campylobacter jejuni and Cryptosporidium parvum, the

authors concluded that if stored at 20 1C for at least 6

months, urine may be considered safe to use as a fertilizer for

any crop (Hoglund et al., 2002b). Their experiments showed

that temperature was the most crucial parameter influencing

the inactivation rates. Without pH control (pH49), 90% of the

rhesus rotavirus were inactivated after 35 days at 20 1C, but no

significant decrease could be detected at 4 1C. Combined with

pH control (see Fig. 3), pH values below 4 seem to result in

additional reduction in the number of pathogens (Hellstrom,

1999). Important side-effects of storing raw source-separated

urine are the precipitation of phosphorus compounds (Udert

ARTICLE IN PRESS

e F

e Fig.

( 231Fig. 3 Evaporation (se

Fig. 4 Freeze-thaw (seet

tha

3.2

3.2Fro

con

wo

tio

ha

spa

3.2Ev

rem

the

Th

ap

su

WAT E R R E S E A R C H 4056al., 2003b) and possible evaporation of ammonia from tanks

t are not sufficiently sealed (Udert et al., 2005a).

. Volume reduction

.1. Introductionm the perspective of commercial fertilisers, the nutrient

tent in urine is small: N: 0.9%, P: 0.06%, K: 0.3% (Table 1). It

uld be beneficial to concentrate the nutrients for transporta-

n and storage purposes. Various water extraction techniques

ve been investigated and developed for long-term manned

ce flights (seeWieland, 1994 for an overview and references).

.2. Evaporationaporation is the most straightforward technology for

oving water from urine (Fig. 3). For space applications,

focus is on recycling water of the best possible quality.

e following list gives a condensed overview of the

proaches reported in the literature. Most of them are

mmarised and referenced in Wieland (1994):

Vapour compression distillation (VCD) recovers more than

96% of the water content and the energy requirement for a

T

los

can

tio

by

us

sp

Fig. 5 Reverse osmosis (see F1 for explanation).1 for explanation).ig.006 ) 3151 3166small-scale unit is 277396MJm3. NASA plans to install

this unit for processing urine in the international space

station in 2005.

Thermoelectric integrated membrane evaporation sys-

tems (TIMES). The urine is pre-treated with ozone (or

ideally UV) and sulphuric acid. It is then heated, pumped

through hollow fibre membranes and exposed to reduced

pressure so that it evaporates.

Air evaporation systems (AES). Pre-treated urine is

pumped through a particulate filter to a wick package.

Heated air is used to evaporate the water from the wick,

leaving the solids.

Lyophilization (Holland et al., 1992). Frozen urine sub-

limates under vacuum and is recovered at 90 1C toproduce ice with a total solid content of 51mgTSl

1.

he evaporation of urine presents two major challenges: (i)

s of ammonia and (ii) energy consumption. Ammonia loss

be avoided by using non-hydrolysed urine or by acidifica-

n (see below). The energy consumption can be minimised

energy recovery. Large-scale thermal desalination plants

e vapour compression distillation (VCD) and can reach

ecific energy requirements of 150180MJm3 distiled water

ig. 1 for explanation).

(Wood, 1982) compared to 2600MJm3 without any energy

recovery. The small-scale space VCD system for about four

adults requires less than 400MJm3, which is equivalent to an

energy recovery of 85%.

In our own laboratory-scale experiments (Mayer, 2002),

non-hydrolysed urine was evaporated at 200mbar and 78 1C.

A tenfold volume reduction was possible without any crystal-

lisation problems, producing a viscous liquid that contained

2000). The required dosages of these agents must be

determined experimentally as a function of the concentration

factor.

3.3. Stabilisation

3.3.1. Introduction

3.3.2. Acidification

ARTICLE IN PRESS

WATER RESEARCH 40 (2006) 3151 3166 31579.7% nitrogen by weight.

3.2.3. Freeze-thawLind et al. (2001) showed that by freezing urine at a

temperature of 14 1C, approximately 80% of the nutrientscan be concentrated in 25% of the original volume (Fig. 4).

Gulyas et al. (2004) confirmed these results by using falling-

film and stirred-vessel freeze concentrators. Using data from

commercial freeze concentrators, they calculated an energy

consumption of 1100MJm3 for a fivefold volume reduction.

These data indicate that evaporation is more efficient with

respect to energy efficiency and that the freeze-thaw process

will be an option only in places with cheap freezing energy

(a cold climate).

3.2.4. Reverse osmosisIn reverse osmosis membranes, the retention of ammonium

is better than its uncharged form (ammonia) and therefore

the retention performance depends strongly on the pH (Fig. 5).

Dalhammar (1997) acidified stored urine to pH 7.1 in order to

prevent permeation of ammonia. At a pressure of 50bar, a

maximum concentration factor of 5 could be achieved,

resulting in the following recoveries of nutrients in the

retentate: ammonium: 70%; phosphate: 73%; potassium:

71%. Similar results were achieved by Thorneby et al. (1999)

in reducing the volume of manure with reverse osmosis.

Fluxes in the range of 2025 lm2h1 could be obtained at

30bar (25 1C). The retention was greater than 98%, except for

ammonia, which was between 93% and 97%. Reverse osmosis

membranes have a high retention for micropollutants (Hof-

man et al., 1997) so that no separation between nutrients and

micropollutants can be expected in this process. The energy

consumption depends on operational and technical para-

meters and energy recovery systems can be installed in large-

scale applications (Avlonitis et al., 2003).

A limiting factor for the application may be the precipita-

tion of salts on or in the membrane, a phenomenon known as

scaling (Migliorini and Luzzo, 2004). In order to control

scaling, chemical formulations based on acids, surfactants or

combinations of both can be added (Jaffer, 1994; Al-Rammah,Fig. 6 Acidification (see FA way of preventing urea hydrolysis is to keep the pH in the

collection tank below 4 (Hellstrom, 1999). Experiments

showed that 60mmolH l1urine of a strong acid (e.g. 2:9g l

1urine

of concentrated sulphuric acid) successfully keeps the pH

below 4 for more then 250 days and prevents hydrolysis of

urea (Fig. 6).

The side effects of acidification are positive with respect to

hygiene due to detrimental effects on pathogenic organisms

at pH values below 4 (Hellstrom, 1999). Low pH values can also

have an impact on pharmaceuticals present in the urine. At

pH 2, an inactivation level of between 50% and 95% could be

found for antibiotics (sulfamethazin, sulfamethoxazol, tetra-

cyclin) and the anti-inflammatory drug diclofenac (Butzen et

al., 2005).

The prevention of urine hydrolysis is muchmore economic-

al than subsequent neutralisation. The neutralisation of

already hydrolysed urine requires 230mmolH l1urine (e.g.11:3g l1urine of concentrated sulphuric acid), approximatelyfour times more than preventive acid addition.Fresh urine contains salts, soluble organic matter and

ammonia bound in urea (Table 1). After microbial contamina-

tion, organic matter is degraded and urea hydrolysed.

Hydrolysis of urea releases ammonia and causes a pH

increase to about 9.2, resulting in more volatile NH3 and

precipitation of compounds with low solubility. True stabili-

sation of urine would thus prevent (i) degradation of organic

matter (causing odour), (ii) precipitation processes (clogging

pipes) and (iii) volatilisation of NH3 (with a number of

negative effects on air quality during storage, transport and

application of liquid fertilizer). Since microbial activity

triggers all these processes, prevention of microbial growth

would be the ultimate stabilisation process. Acidification,

microfiltration and ultrafiltration have been suggested to

achieve this, but only acidification has been studied in detail.

Information is available from the literature on urease

inhibitors to prevent urea hydrolysis, but Benini et al. (1999)

found that the reported efficiencies are low and negative side

effects have to be considered.ig. 1 for explanation).

ARTICLE IN PRESS

4) (s

(s

( 23.3.3. Partial nutrificationNitrification is a suitable method for lowering the pH. Since

no other relevant buffers are present in urine in significant

concentrations, nitrification of urine can only oxidise half the

available ammonium until nitrification stops due to low pH

conditions. High nitrite concentrations and low pH have a

specific detrimental effect on nitrite oxidisers, due to their

sensitivity to nitrous oxide (e.g. Hunik et al., 1993). Complete

Fig. 8 Struvite (MgNH4PO

Fig. 7 Partial nitrification

WAT E R R E S E A R C H 403158conversion of ammonia to nitrite is therefore often inhibited,

depending on the operational conditions. Experimental

results by Johansson and Hellstrom (1999) and Udert et al.

(2003c) confirm that the product of urine nitrification is either

an ammonium-nitrate or ammonium-nitrite solution with an

approximate 1:1 composition (Fig. 7).

Due to high concentrations in urine (mainly of salt,

ammonia and nitrous acid, see Udert et al., 2003c for a

detailed discussion), inhibition affects nitrification andmakes

a continuously operated technical process sensitive to

instabilities. A more detailed overview of the nitrifier kinetics

can be found in Hellinga et al. (1999) and Van Hulle (2005). An

attractive alternative for converting nitrite to nitrate is by

chemical oxidation with oxygen at low pH values (Udert et al.,

2005b).

Udert et al. (2003c) operated three continuous laboratory

systems: A moving-bed biological reactor (MBBR), a sequen-

cing batch reactor (SBR) and a continuously stirred reactor

(CSTR). Only the MBBR system was capable of producing

ammonium nitrate as a final product. The measured conver-

sion rates were 380gNm3 d1 (25.3 1C) at steady state. The

other two systems produced stable ammonium nitrite even at

high sludge ages or hydraulic residence times respectively.

The measured nitrite formation rates were 790gNm3 d1

(30 1C; HRT 4.8d) for the CSTR and 280gNm3 d1 (24.5 1C;HRT 4d; SRT430d) for the SBR. Because the inhibitionkinetics are not yet fully understood, it is important forpractical purposes to monitor the nitrification process care-

fully. This can easily be done in technical systems by feeding

the system in such a way that the pH does not get too high or

too low. The product of these efforts is a stable solution

without the typical urine smell and with no easily degradable

substances.

3.4. P-recovery

ee Fig. 1 for explanation).

ee Fig. 1 for explanation).

006 ) 3151 31663.4.1. IntroductionPhosphate is produced from phosphate rock, a limited

resource. The estimated worldwide reserves range from 1.2

to 5 1010 tons, which would suffice for 50300 years (Steenand Steen, 1998; results also in Driver et al., 1999; Zapata and

Roy, 2005), depending on the assumed consumption scenario.

At the moment, the depletion of phosphorus reserves is less

of a concern than the decrease in their quality (and therefore

increase in price) and the strategic considerations of the

worlds phosphorus producers. If no new sources of high-

quality phosphate are identified, future phosphorus reserves

will contain less phosphate and higher levels of enrichment

by heavy metals, principally cadmium (Driver et al., 1999;

Smil, 2000; Isherwood, 2000).

Extensive reviews of phosphorus removal and recovery

technologies from liquid wastes are given in Brett et al. (1997),

Wilsenach and Van Loosdrecht (2002), Valsami-Jones (2004),

and De-Bashan and Bashan (2004). Many of the processes

described, especially those for the treatment of digester

supernatant, are also applicable to urine.

In the literature on phosphate recovery, precipitation of

struvite seems to be the predominant product, followed by

calcium phosphate. Udert et al. (2003b) investigated scales in

urine collection systems and reported struvite, calcium

phosphate (hydroxyapatite) and calcite as the predominant

forms of precipitation. Despite the many P-recovery options,

known as struvite, MAP or AMP, is an attractive precipitate

ARTICLE IN PRESS

ee

g (

(2because it conveys two dominant wastewater nutrients in

solid form (Fig. 8). Additionally, the product can be used as a

slow-release fertiliser (Bridger et al., 1961; Johnston andonly struvite precipitation can be found in the literature on

P-recovery from urine.

3.4.2. Struvite (MgNH4PO4)Magnesium ammonium phosphate (MgNH4PO4 6H2O), also

Fig. 9 Ion exchange (s

Fig. 10 Ammonia strippin

WATER RESEARCH 40Richards, 2003). Gaterell et al. (2000) suggest the conversion

of struvite into an enhanced struvite that contains two parts

of a slow-release fertiliser (magnesium phosphate, MgHPO4)

and one part of the easily soluble ammonium phosphate

((NH4)2HPO4): this product is claimed to have good market

potential. Technical struvite precipitation was extensively

investigated for the removal of N and P from digester

supernatant (e.g. Wu and Bishop, 2005), animal waste slurries

(e.g. Suzuki et al., 2002) and for the treatment of wastewater,

landfill leachate and abattoir effluent.

The pH of hydrolysed urine is optimal for struvite pre-

cipitation (Buchanan et al., 1994) and therefore no pH

adjustment is required. The precipitation is triggered by the

addition of magnesium, usually in the form of MgO, Mg(OH)2,

MgCl2 or bittern (the magnesium-rich brine from table-salt

production). As shown in Table 1, there is much more

ammonium than phosphate present in urine on a molar

basis. As a consequence, about 3% of the nitrogen can be

eliminated by magnesium addition only so that the effect on

the pH value is small. Ronteltap et al. (2006) investigated the

conditional solubility product and the equilibrium reactions

for urine. The simplified solubility product was determined

with [Mg] [NH4++NH3] [Portho] 107.6 M3 (pH 9 and a ionicstrength 0.68), where [Mg] is the concentration of dissolvedmagnesium, [NH4

++NH3] the measured ammonium+ammo-

nia, and [Portho] the dissolved ortho-phosphate. See alsoRonteltap et al. (2003) for an overview of the published

solubility products for struvite.

No relevant information is given about the kinetics of

struvite precipitation in urine. From the fact that stored urine

contains a plethora of fine particles such as micro-organisms

and precipitation products (Hoglund et al., 1998; Udert et al.,

2003b), it can be concluded that heterogeneous nucleation

could play a dominant role in the formation of struvite

crystals. Our own experience with urine shows that pre-

cipitation in batch experiments is fast and without any

Fig. 1 for explanation).

see Fig. 1 for explanation).

006) 3151 3166 3159perceivable lag.

3.5. N-recovery

3.5.1. IntroductionProcesses for the production of nitrogen fertiliser are based

on the fixation of atmospheric nitrogen in the Haber-Bosch

process. Depletion of resources does not therefore play a

major role in evaluating the environmental benefits of

nitrogen recovery processes, energy consumption being the

main parameter. An overview of current nitrogen recycling

technologies in general wastewater treatment is given in

Rulkens et al. (1998) and Maurer et al. (2002); corresponding

energy consumptions are summarised in Maurer et al. (2003).

3.5.2. Ion exchangeAn ion exchanger with a high affinity for ammonium is

clinoptilolite, a naturally occurring zeolite, and polymeric

macronet exchangers have recently also become available

which are suitable for this purpose (Jorgensen and Weath-

erley, 2003) (Fig. 9). Both materials as well as other zeolites

have been tested for the treatment of waste water effluents

(Liberti et al., 1981), and zeolites have been tested for the

removal of ammonia from urine, also combined with the

addition of MgO for recovering phosphate in the form of

struvite (Lind et al., 2000; Ban and Dave, 2004). The highest

If the aim of urine treatment is improved control of water

ARTICLE IN PRESS

see

s (s

( 23.5.3. Ammonia strippingStored urine was stripped under vacuum (0.4 bar, 40 1C) and

the gas stream was adsorbed in water at a pressure of 5 bar

and 20 1C (Behrendt et al., 2001). The resulting productrecovery rates were obtained at an MgO dosage of 0.5mg/l and

a zeolite dosage of 15g/l. The remaining supernatant

concentrations for P and N were 10gPm3 and 1000gNm

3,

respectively.

Fig. 12 Electrodialysis (Fig. 11 Anammox proces

WAT E R R E S E A R C H 403160contains 10% ammonia and is unstable at normal pressure.

No information is provided on the concentration of ammonia

remaining in the urine solution after stripping (Fig. 10).

Energy consumptions can be estimated from experiments

with digester supernatant (Siegrist, 1996). At 20 1C and 95%

ammonia removal, the energy consumption was reported to

be around 7kWhm3treated liquid. Vapour Phase Catalytic Am-

monia Removal (VAPCAR) combines vaporisation with high-

temperature catalytic oxidation of ammonia and other

volatile compounds. A two-step catalytic process is used to

produce nitrogen gas, carbon dioxide and water (Slavin and

Oleson, 1991).

3.5.4. Isobutylaldehyde-diurea (IBDU) precipitationIn fresh, non-hydrolysed urine, nitrogen is mainly present in

the form of urea. Urea forms a complex with isobutyralde-

hyde (IBU), resulting in the precipitation of isobutylaldehyde-

diurea (IBDU), a commercially available slow-release fertilizer.

Its industrial production requires a high urea concentration

(Behrendt et al. 2001, Reinhart, 2002) and even at these

concentrations an excess of urea or IBU results only in partial

complexation, so that relatively high fractions of urea remain

in the liquid phase after treatment. Production at the urea

concentration of approximately 1% present in urine is there-

fore not feasible (Reinhart, 2002). Experimental results with

urine confirm this statement (Behrendt et al. 2001). At a five-pollution, it may be desirable to remove N and P without

recovering them. Whereas biological P-removal has neverfold stoichiometric excess of IBU, about 75% conversion of

urea was obtained.

3.6. Nutrient removal (P and N)

3.6.1. Introduction

Fig. 1 for explanation).ee Fig. 1 for explanation).

006 ) 3151 3166been considered for the treatment of source-separated urine,

full nitrification can easily be achieved with an extension to

partial nitrification (see above). Denitrification (resulting in

N2) may be achieved in a number of ways: biological reduction

of nitrate with organic matter as the electron donor; biological

oxidation of ammonia with nitrite as the electron acceptor

(the anammox process) or electrochemical oxidation of

ammonia (NASA 1977). Of these technologies, the anammox

process has been studied in detail for urine.

3.6.2. Anammox ProcessAnaerobic ammonium oxidation (Anammox) is a biological

process designed to eliminate nitrogen independently of a

carbon source (Strous et al., 1998) (Fig. 11). Under anaerobic

conditions, ammonium and nitrite are converted mainly to

nitrogen gas. As reported in Fig. 7, the formation of nitrite

stops halfway through the process, producing a 1:1 ammo-

nium/nitrite solution. Udert et al. (2003c) added this solution

to anammox sludge from a pilot plant treating digester

supernatant. At 30 1C they measured a denitrification rate of

1000gNm3d1 and the ratio of total ammonia to nitrite

elimination was 1:1.1870.07. The results of these experi-

ments show that nitrogen can be removed from source-

separated urine with anammox. A combination of nitrifica-

tion and anammox reactors could eliminate 7585% of the

nitrogen, leaving an ammonium nitrate solution.

3.7. Removal of micropollutants

3.7.1. IntroductionIncreasingly powerful analytical methods mean that a large

number of pharmaceuticals and natural hormones from the

human metabolism are now detected in the aquatic environ-

ment, but their environmental relevance is currently unclear.

Most of the effects that can be observed, e.g. the formation of

vitellogenin (a precursor of egg yolk proteins) in male trout

(Harries et al., 1997), are chronic effects without clear

consequences for the affected organism. It is generally

recognised that urine contains a significant amount of

excreted micropollutants. In general, a distinction must be

made between separation and elimination processes. The

separation of nutrients and micropollutants is relevant to the

production of a urine-based fertilizer, whereas the micro-

pollutants must be eliminated for water-pollution control.

Separation processes are primarily based on membranes or

precipitation whereas removal processes are based on oxida-

tion or adsorption (Larsen et al., 2004).

out (Pronk et al. 2006b) in order to manipulate the pH.

Ammonia was transferred across a hydrophobic membrane

from the basic into the acid concentrate. Batch experiments

confirmed that a pH decrease occurred in the acid concen-

trate, also known as the product compartment. However, at

higher conversions the pH rose again to its original value.

This pH increase can be attributed to carbon dioxide

transported from the basic concentrate across the gas-filled

membrane into the acid concentrate (Pronk et al. 2006b). The

use of an ammonium-selective gas-transfer membrane

instead of a hydrophobic gas-transfer membrane should in

principle solve this problem, but this has not yet been

investigated.

3.7.3. NanofiltrationNanofiltration (Fig. 13) has been tested for the retention of a

range of environmentally relevant compounds such as

pesticides (Van der Bruggen et al., 2001), disinfection by-

products and pharmaceutical compounds (Kimura et al.,

2004), phthalates (Kiso et al. 2001) and natural steroid

ARTICLE IN PRESS

see

WATER RESEARCH 40 (2006) 3151 3166 3161In addition to the processes presented here, it is possible in

principle to remove micropollutants by adsorption to active

carbon or other adsorbents. It can be expected that the

presence of high amounts of COD in urine strongly interfere

with the adsorption process (Quinlivan et al., 2005).

3.7.2. ElectrodialysisElectrodialysis membranes are ion-exchange membranes

made of functionalised polymers with a dense structure

(Strathmann 1992) enabling salts to be extracted and con-

centrated (Fig. 12). The apparent pore size is typically around

200Da (Kim et al. 2003) so that these membranes can

potentially retain micropollutants. Investigations showed

that electrodialysis may be used to selectively extract the

nutrients into a concentrated product stream while retaining

the micropollutants (pharmaceuticals) in the diluate (Pronk et

al., 2006a). Experiments with bipolar membranes were carried

Fig. 13 Nanofiltration (Fig. 14 Ozonation andhormones (Nghiem et al. 2004). For production of a urine-

based fertilizer, it is important for the micropollutants to be

retained and for mineral salts to be permeated in order to

obtain a product free of micropollutants. The removal of

micropollutants was tested with different nanofiltration

membranes (Pronk et al. 2006c). The efficiency of the

separation process depends strongly on the pH, demonstrat-

ing that electrostatic interactions with the membrane play

an important role in the separation of micropollutants.

Under optimised conditions, the removal rate of a set of

hormones and pharmaceutical compounds in urine exceeds

92% (Pronk et al. 2006c). Furthermore, it was shown that the

permeation of urea is almost complete, while 5080% of

the ammonia was retained, depending on the pH (Fig. 13). In

order to obtain high nitrogen recoveries, therefore, it is

important to use non-hydrolysed urine (see also Section

Urine Stabilisation).

Fig. 1 for explanation).advanced oxidation.

micropollutants from urine.

ARTICLE IN PRESS

( 24. Conclusions and outlook

In this paper, we have reviewed a number of unit processes

for treating human urine with respect to seven different

purposes: hygienisation, volume reduction, stabilisation, P-

recovery, N-recovery, nutrient removal and handling of

micropollutants. The review concentrates on processes that

have actually been tested with human urine at least on a

laboratory scale. Most of the urine treatment options found in

the literature are adapted from existing technologies and

have also been applied to other waste streams. However, the

unique chemical properties of urine make adaptation of

existing processes almost always inevitable. An example is

nitrification, where the conversion of nitrite to nitrate is

mostly inhibited and therefore makes the application of the

anammox process relatively simple.

Our evaluation made clear that a very large number of

technical options are available, with different strengths and

weaknesses. However, all seven purposes cannot be achieved

with a single unit process. Whether the aim is to concentrate

on a specific purpose such as nutrient removal or to combine

different process units to achieve a more comprehensive goal

depends on the circumstances and is not discussed in this

paper.

Hygienisation: Since actual effectiveness has only been

shown for storage, it is difficult to draw any final conclusions.

However, other processes will also be effective: membrane3.7.4. Ozonation and advanced oxidationMicropollutants can be oxidised with chlorine, chlorine

dioxide, ozone (O3), or OH radicals (advanced oxidation

processes, AOPs, see Prousek, 1996). In the case of ozone,

the reaction can take place directly with ozone or with the

secondary oxidants (e.g. OH-radicals) formed during ozona-

tion (Von Gunten 2003a; Von Gunten 2003b). In view of the

high COD content of urine (210g/l, see Table 1), oxidants

reacting specifically with micropollutants are preferred. As

most of the compounds tested show enhanced reactivity

towards ozone (Huber et al. 2003), use of ozone seems to be

preferable to advanced oxidation processes because a larger

fraction of the oxidant (OH radical) is lost to the matrix in the

latter (Fig. 14). From recent investigations with urine, it was

concluded that complete oxidation of a representative set of

micropollutants including pharmaceuticals and synthetic

hormones may be achieved (Pronk et al., 2006d). Despite the

quenching of oxidants by the organic matrix in urine, it was

shown that all the tested compounds could be transformed

completely. At an ozone dose of 1.1 g/l, fast-reacting com-

pounds such as ethinylestradiol were completely removed,

while removal of more recalcitrant compound such as

ibuprofen was 80% (Pronk et al, 2006d). Analysis of the results

showed that oxidation took place directly by ozone as well as

by OH radicals. Considering the high reactivity of the OH

radicals with most organic micropollutants, ozonation can be

regarded as a suitable method for removing a wide range of

WAT E R R E S E A R C H 403162processes (except for reverse osmosis), evaporation at high

temperature, acidification and biological processes all havethe potential to produce a hygienic product, but further

studies are required.

Volume reduction: Precipitation processes obviously repre-

sent the most efficient measures of volume reduction,

reducing the water content to a few percent. Evaporation is

almost as effective, resulting in a water content of 510%,

whereas freeze-thaw, electrodialysis and reverse osmosis are

considerably less effective.

Stabilisation: The most effective stabilisation processes are

acidification and biological processes: they prevent not only

ammonia evaporation but also the typical urine odour.

Struvite precipitation is quite efficient because it produces a

mineral product with little water and a low content of organic

material. Membrane processes (micro and nanofiltration plus

electrodialysis) are effective in preventing further microbial

growth, but require additional measures to prevent odour and

urea hydrolysis by the hydrolysing enzymes already present

in the dissolved urea.

P-recovery: We define a process as a P-recovery technique if

(a) a volume reduction has taken place and (b) the phos-

phorus is concentrated in a small volume. Consequently, the

relevant processes are struvite precipitation, the processes

recorded under volume reduction and electrodialysis, all of

them with a phosphorus recovery rate of between 90% and

100%.

N-recovery: We define N-recovery in an equivalent way to P-

recovery. N-recovery actually takes place in a large number of

processes, but with different yields. At least 90% recovery is

achieved with evaporation, electrodialysis, reverse osmosis

and struvite precipitation (with stoichiometric phosphate

addition). Between 80% and 90% recovery is achieved with

the freeze-thaw process and most likely with ammonia

stripping. An N-recovery of between 60% and 80% is achieved

with reverse osmosis, ion exchange with zeolite and IBDU

precipitation with a five-fold excess of IBU.

Nutrient removal: Only biological processes have been tested

for nutrient elimination. A nitrogen removal efficiency of

7580% is obtained with a combination of partial nitrification

and the annamox process.

Handling of micropollutants: separation from nutrients and

removal: Only chemical oxidation has so far proved effective

for the actual removal of micropollutants from urine, and this

is not entirely beyond doubt. Much work remains to be done.

Results from biological oxidation are still outstanding.

Struvite precipitation, ammonia stripping and nanofiltration

have proved to be highly effective for the separation of

micropollutants and nutrients. IBDU precipitation and elec-

trodialysis are only partially effective; electrodialysis optimi-

sation is proceeding.

Energy consumption: Besides the technical description of the

various options, we also estimated the effectiveness and

energy consumption of these processes. It is obvious that we

had to make a number of assumptions because very few of

these processes have actually been optimised for full-scale

application. As an example, we assumed realistic but

challenging energy recovery schemes resulting in estimated

energy requirements mostly in the order of 20100MJm3

(without the energy used to construct the treatment device),

006 ) 3151 3166with a few exceptions. If we assume a production of 2 l

urine per day and person (including flushing water), this

the job in a realistic urine source-separation scenario. Many

of the techniques described in this paper need some sort of

pre-treatment (e.g. evaporation) or only deal with a specific

for substituting phosphorus recovered from wastewater

ARTICLE IN PRESS

(2problematic fraction in the urine (e.g. struvite precipitation).

Urine treatment solutions will most probably consist of a

combination of treatment processes. An example is the

recovery of phosphate by struvite precipitation followed by

a biological process designed to eliminate the organic

pollutants and nitrogen. This enhanced flexibility for urban

wastewater treatment is one of the great benefits of urine

source separation and makes the development of processes

for urine treatment attractive. However, it is impossible to

discuss process combinations without a scenario as context

and since an explicit discussion of scenarios would go way

beyond the scope of this review, a second publication dealing

with this issue is in preparation.

A general assessment of NoMix technology is not possible

on the basis of process engineering technology alone.

However, as we have shown in this paper, the limitations

for NoMix technology will not be found in a lack of NoMix

process engineering options. In most cases, our expectation

that such technologies might be too energy intensive has not

come true either. Although other problems connected to

NoMix technology may be more difficult to solve, one should

not underestimate the efforts of research and development of

NoMix process engineering technologies. Except for evapora-

tion and storage, none of the processes described have yet

advanced beyond the laboratory stage. Considerable develop-

ment work remains to be done in order to enhance urine-

processing techniques into marketable products.

Acknowledgements

The authors wish to thank Urs von Gunten and Detleff

Knappe for their input with respect to the unit operations of

oxidation and activated carbon.

R E F E R E N C E S

Al-Rammah, A., 2000. The application of acid-free antiscalant tomitigate scaling in reverse osmosis membranes. Desalination132, 8387.

Avlonitis, S.A., Kouroumbas, K., Vlachakis, N., 2003. Energycorresponds to approximately 0.52.5W per person. This is

within a reasonable range compared to the approximately

4W per person that we currently apply for nitrification in an

ordinary wastewater treatment plant. A number of other

relevant aspects such as cost and robustness were not

discussed.

The review shows clearly that a wide range of technical

options is available for the effective treatment of collected

urine. Depending on the overall goal of the treatment process,

a specific technical solution or a combination of solutions can

be found to meet the requirements. Additional research is

needed to find promising process combinations that can do

WATER RESEARCH 40consumption and membrane replacement cost for seawaterRO desalination plants. Desalination 157, 151158.Ban, Z.S., Dave, G., 2004. Laboratory studies on recovery of N and P

from human urine through struvite crystallisation and zeolite

adsorption. Environ. Technol. 25 (1), 111121.Behrendt, J., Arevalo, E., Gulyas, H., Niederste-Hollenberg, J.,

Niemiec, A., Zhou, J., Otterpohl, R., 2001. Production of value

added products from separately collected urine. IWAWorldWater Congress, Berlin.

Benini, S., Rypniewski, W.R., Wilson, K.S., Miletti, S., Ciurli, S.,

Mangani, S., 1999. A new proposal for urease mechanismbased on the crystal structures of the native and inhibited

enzyme from bacillus pasteurii: Why hydrolysis costs twonickels. Structure 7 (2), 205216.

Brett, S., Guy, J., Morse, G.K., Lester, J.N., 1997. Phosphorus

Removal and Recovery Technologies. Selper Publication,London, ISBN: 0 94841 110 0.

Bridger, G.L., Salutsky, M.L., Starostka, R.W., 1961. Metal ammo-

nium phosphates as fertilisers. In: Proceedings of the 140thMeeting of the American Chemical Society, Chicago, IL.

Buchanan, J.R., Mote, C.R., Robinson, R.B., 1994. Thermodynamics

of struvite formation. Trans. ASAE 37 (2), 617621.Butzen, A., Werres, F., Balsaa, P., 2005. Aufbau und Einsatz einer

problemorientierten Analytik mit dem Ziel eines Monitorings

ausgewahlter Pharmaka in Boden und Urin (Implementationand application of problem oriented analytical methods with

the goal of monitoring selected pharmaceuticals in soil and

urine, in German). in: Nahrstofftrennung und verwertung inder Abwassertechnik am Beispiel der ,,Lambertsmuhle,

Bonner Agrikulturchemische Reihe, Band 21, p. 2554, Bonn,

Germany, ISBN 3-937941-02-9. Source: Verein zur Forderungder Agrikulturchemie e.V., c/o Institut fur Planzenernahrung,

Karlobert-Kreiten-Strasse 13, 53115 Bonn, Germany.

Christiansen, L.B., Winther-Nielsen, M., Helweg, C., 2002. Femin-isation of fishthe effect of estrogenic compounds and their

fate in sewage treatment plants and nature. Environmental

Project no. 729, Danish Environmental Protection Agency, ISBN87-7972-306-3, download on the internet: http://www.mst.dk/

udgiv/publications/2002/87-7972-305-5/html/default_eng.htm.

Ciba-Geigy, 1977. Wissenschaftliche Tabellen Geigy, TeilbandKorperflussigkeiten (Scientific Tables Geigy. Volume: Body

Fluids), 8th ed. Basel. In German.Dalhammar, G., 1997. Behandling och koncentrering av huma-

nurin (Royal Institute of Technology, Stockholm, Department

of Biochemistry and Biochemical technology), Report, perso-nal communication.

De-Bashan, L.E., Bashan, Y., 2004. Recent advances in removing

phosphorus from wastewater and its future use as fertilizer(19972003). Water Res. 38 (19), 42224246.

Driver, J., Lijmbach, D., Steen, I., 1999. Why recover phosphorus

for recycling, and how? Environ. Technol. 20 (7), 651662.Escher, B.I., Eggen, R.I.L., Schreiber, U., Schreiber, Z., Vye, E.,

Wisner, B., Schwarzenbach, R.P., 2002. Baseline toxicity

(narcosis) of organic chemicals determined by in vitromembrane potential measurements in energy-transducing

membranes. Environ. Sci. Technol. 36, 19711979.

Escher, B.I., Bramaz, N., Maurer, M., Richter, M., Sutter, D., vonKanel, C., Zschokke, M., 2005. Screening test battery for

pharmaceuticals in urine and wastewater. Environ. Toxicol.

Chem. 24, 750758.Escher, B.I., Pronk, W., Suter, M.J.-F., Maurer, M., 2006. Monitoring

the removal efficiency of pharmaceuticals and hormones in

different treatment processes of source-separated urine withbioassays. Environ. Sci. Technol., in press.

Gaterell, M.R., Gay, R., Wilson, R., Gochin, R.J., Lester, J.N., 2000. An

economic and environmental evaluation of the opportunities

006) 3151 3166 3163treatment works in existing UK fertiliser markets. Environ.

Technol. 21 (9), 10671084.

ARTICLE IN PRESS

( 2Gulyas, H., Bruhn, P., Furmanska, M., Hartrampf, K., Kot, K.,Luttenberg, B., Mahmood, Z., 2004. Freeze concentration forenrichment of nutrients in yellow water from no-mix toilets.Water Sci. Technol. 50 (6), 6168.

Hanaeus, J., Hellstrom, D., Johansson, E., 1997. A study of a urineseparation in an ecological village in northern Sweden. WaterSci. Technol. 35 (9), 153160.

Harries, J.E., Sheahan, D.A., Jobling, S., Matthiessen, P., Neall, P.,Sumpter, J.P., Tylor, T., Zaman, N., 1997. Estrogenic activity infive United Kingdom rivers detected by measurement ofvitellogenesis in caged male trout. Environ. Toxicol. Chem. 16,534542.

Heinonen-Tanski, H., vanWijk-Sijbesma, C., 2005. Human excretafor plant production. Bioresource Technol. 96 (4), 403411.

Hellinga, C., van Loosdrecht, M.C.M., Heijnen, J.J., 1999. Modelbased design of a novel process for nitrogen removal fromconcentrated flows. Math. Comput. Model. Dyn. Syst. 5 (4),351371.

Hellstrom, D., Johannson, E., Grennberg, K., 1999. Storage ofhuman urine: acidification as a method to inhibit decom-position of urea. Ecol. Eng. 12, 253269.

Hofman, J., Beerendonk, E.F., Folmer, H.C., Kruithof, J.C., 1997.Removal of pesticides and other micropollutants with cellu-lose-acetate, polyamide and ultra-low pressure reverse os-mosis membranes. Desalination 113, 209214.

Hoglund, C.E., Stenstrom, T.A.B., 1999. Survival of cryptospor-idium parvum oocysts in source separated human urine. Can.J. Microbiol. 45, 740746.

Hoglund, C., Stenstrom, T.A., Jonsson, H., Sundin, A., 1998.Evaluation of faecal contamination and microbial die-off inurine separating sewage systems. Water Sci. Technol. 38 (6),1725.

Hoglund, C., Vinneras, B., Stenstrom, T.A., Jonsson, H., 2000.Variation of chemical and microbial parameters in collectionand storage tanks for source separated human urine. J.Environ. Sci. Health Part A-Toxic/Hazard. Substances Environ.Eng. 35, 14631475.

Hoglund, C., Ashbolt, N., Stenstrom, T.A., Svensson, L., 2002a.Viral persistence in source-separated human urine. Adv.Environ. Res. 6, 265275.

Hoglund, C., Stenstrom, T.A., Ashbolt, N., 2002b. Microbial riskassessment of source-separated urine used in agriculture.Waste Manage. Res. 20, 150161.

Holland, P.J., Bird, D.M., Miller, C.L., 1992. Extraction of potablewater from urine for space applications. In: Sadeh, W.Z., Sture,S., Miller, R.J. (Eds.), Engineering, construction, and operationsin space III: Space 92; Third International Conference, vol. 2.American Society of Civil Engineers (ASCE), Denver, CO andNew York, USA, pp. 16801689.

Huang, D., Bader, H., Scheidegger, R., Schertenleib, R., Gujer, W.,(2006). Confronting limitations: new solutions required inurban water management of a Chinese mega-city. J. Environ.Manage, in press.

Huber, M.M., Canonica, S., Park, G.Y., Von Gunten, U., 2003.Oxidation of pharmaceuticals during ozonation and advancedoxidation processes. Environ. Sci. Technol. 37, 10161024.

Hunik, J.H., Meijer, H.J.G., Tramper, J., 1993. Kinetics of nitrobacteragilis at extreme substrate, product and salt concentrations.Appl. Environ. Microbiol. 40 (23), 442.

Isherwood, K.F., 2000. Mineral fertilizer use and the environment.International Fertilizer Industry Association/United NationsEnvironment Programme, Paris, 106 pp.

Jaffer, A.E., 1994. The application of a novel chemical treatmentprogram to mitigate scaling and fouling in reverse-osmosisunits. Desalination 96, 7179.

WAT E R R E S E A R C H 403164Johansson, E., Hellstrom, D., 1999. Nitrification in combinationwith drying as a method for treatment and volume reductionof stored human urine. In: Johansson, E. (Ed.), Urine separatingwastewater systems: design experiences and nitrogen con-servation. Licentiate Thesis. Lulea University of Technology,Lulea, Sweden.

Johnston, A.E., Richards, I.R., 2003. Effectiveness of differentprecipitated phosphates as phosphorus sources for plants.Soil Use Manage. 19 (1), 4549.

Jonsson, H., Stenstrom, T.A., Svensson, J., Sundin, A., 1997. Sourceseparated urine-nutrient and heavy metal content, watersaving and faecal contamination. Water Sci. Technol. 35 (9),145152.

Jorgensen, T.C., Weatherley, L.R., 2003. Ammonia removal fromwastewater by ion exchange in the presence of organiccontaminants. Water Res. 37 (8), 17231728.

Kim, D.H., Moon, S.-H., Cho, J., 2003. Investigation of theadsorption and transport of natural organic matter (NOM) inion-exchange membranes. Desalination 151, 1120.

Kimura, K., Toshima, S., Amy, G., Watanabe, Y., 2004. Rejection ofneutral endocrine disrupting compounds (EDCs) and phar-maceutical active compounds (PhACs) by RO membranes. J.Membrane Sci. 245, 7178.

Kirchmann, H., Pettersson, S., 1995. Human urinechemicalcomposition and fertilizer use efficiency. Fertilizer Res. 40,149154.

Kiso, Y., Kon, T., Kitao, T., Nishimura, K., 2001. Rejection propertiesof alkyl phthalates with nanofiltration membranes. J. Mem-brane Sci. 182, 205214.

Larsen, T.A., Gujer, W., 1996. Separate management of anthro-pogenic nutrient solutions (human urine). Water Sci. Technol.34 (34), 8794.

Larsen, T.A., Gujer, W., 2001. Waste design and source control leadto flexibility in wastewater management. Water Sci. Technol.43 (5), 309318.

Larsen, T.A., Lienert, J., 2003. Societal implications of re-engineer-ing the toilet. Water Intelligence Online. UNIQUE ID:200303006. http://www.iwaponline.com/wio/2003/03/de-fault001.htm.

Larsen, T.A., Lienert, J., Joss, A., Siegrist, H., 2004. How to avoidpharmaceuticals in the aquatic environment. J. Biotechnol.113 (13), 295304.

Liberti, L., Boari, G., Petruzzelli, D., Passino, R., 1981. Nutrientremoval and recovery from wastewater by ion exchange.Water Res. 15, 337342.

Lind, B.B., Ban, Z., Byden, S., 2000. Nutrient recovery from humanurine by struvite crystallization with ammonia adsorption onzeolite and wollastonite. Bioresource Technol. 73 (2), 169174.

Lind, B.B., Ban, Z., Byden, S., 2001. Volume reduction andconcentration of nutrients in human urine. Ecol. Eng. 16 (4),561566.

Maurer, M., Muncke, J., Larsen, T.A., 2002. Technologies fornitrogen recovery and reuse. In: Lens, P., Pol, L.H., Wilderer,P., Asano, T. (Eds.), Water Recycling and Resource Recovery inIndustry. IWA Publishing, London, pp. 491510.

Maurer, M., Schwegler, P., Larsen, T.A., 2003. Nutrients in urine:energetic aspects of removal and recovery. Water Sci. Technol.48 (1), 3746.

Mayer, M., 2002. Thermische Hygienisierung und Eindampfungvon Humanurin. Diplomarbeit des Institut fur Umweltechnikder Fachhochschule beider Basel, Muttenz, Schweiz (Thermaldisinfection and evaporation of human urine. Diploma workof the Institute for Environmental Technology, Fach-hochschule beider Basel, Muttenz, Switzerland).

McBride, M.B., Spiers, G., 2001. Trace element content of selectedfertilizers and dairy manures as determined by ICP-MS.Commun. Soil Sci. Plant Anal. 32 (1&2), 139156.

Medilanski, E., Chuan, L., Mosler, H., Schertenleib, R., Larsen, T.A.,

006 ) 3151 3166(2006). Wastewater Management in Kunming, China: Feasi-bility and Perspectives of Measures at the Source from aStakeholder Point of View. Environ. Urban., 18(2), in press.

ARTICLE IN PRESS

(2Migliorini, G., Luzzo, E., 2004. Seawater reverse osmosis plant

using the pressure exchanger for energy recovery: a calcula-

tion model. Desalination 165, 289298.NASA, 1977. Electrolytic pretreatment of urine. Prepared by

Lockheed Missiles & Space Co., NASA Report no. NASA-CR-

151566, Johnson Space Center, USA.Nghiem, L.D., Schafer, A.I., Elimelech, M., 2004. Removal of

natural hormones by nanofiltration membranes: measure-

ment, modeling, and mechanisms. Environ. Sci. Technol. 38,18881896.

Peter-Frohlich, A., 2002. Sanitation concept for separatetreatment (SCST), anton.peter-froehlich@bwb.de, http://

www.kompetenz-wasser.de/dt/projekte/proj_scst.html.

Pronk, W., Biebow, M., Boller, M., 2006a. The application ofelectrodialysis for the recovery of salts from a micropollu-

tant-containing urine solution. Environ. Sci. Technol., in press.

Pronk, W., Biebow, M., Boller, M., 2006b. Treatment of source-separated urine by a combination of bipolar electrodialysis

and a gas transfer membrane. Water Sci. Technol. 53 (3),

139146.Pronk, W., Palmquist, H., Biebow, M., Boller, M., 2006c. Nanofiltra-

tion for the separation of pharmaceuticals from nutrients in

source-separated urine. Water Res. 40 (7), 14051412.Pronk, W., Dodd, M., Zuleeg, S., Escher, B., Von Gunten, U., 2006d.

The ozonation of micropollutants in source-separated urine,

in preparation.Prousek, J., 1996. Advanced oxidation processes for water treat-

ment. Photochemical processes. Chem. Listy 90, 307315.

Quinlivan, P.A., Li, L., Knappe, D.R.U., 2005. Effects of activatedcarbon characteristics on the simultaneous adsorption of

aqueous organic micropollutants and natural organic matter.

Water Res. 39, 16631673.Reichl, H., 2002. Prion transmission in blood and urine: what are

the implications for recombinant and urinary-derived gona-

dotrophins? Fertil. Steril. 78 (Suppl. 1), 179.Reinhart, 2002. BASF Agricultural Products, Germany, Personal

communication.

Ritschel, W.A., Kearns, G.L., 1999. Handbook of basic pharmaco-kinetics. American Pharmaceutical Association, Washington,

DC.Rogowski, D., Golding, S., Bowhay, D. and Singleton, S., 1999.

Screening survey for metals and dioxins in fertilizer products

and soils in Washington StateFinal report. Washington StateDepartment of Ecology, Ecology Publication no. 99309,

Olympia, Washington, USA., Internet: http://www.ecy.wa.gov/

biblio/99309.html, last updated 11 May 2004)Ronteltap, M., Biebow, M., Maurer, M., Gujer, W., 2003. Thermo-

dynamics of struvite precipitation in source separated urine.

In: Second International Symposium on Ecological Sanitation,IWA, gtz, Lubeck, Baltic Sea, Germany, pp. 463470.

Ronteltap, M., Maurer, M., Gujer, W., 2006. Struvite precipitation

thermodynamics in source-separated urine. Submitted toWater Res.

Rulkens, W.H., Klapwijk, A., Willers, H.C., 1998. Recovery of

valuable nitrogen compounds from agricultural liquidwastes: potential possibilities, bottlenecks and future tech-

nological challenges. Environ. Pollut. 102 (Suppl. 1), 727735.

Santos, R.L.S., Manfrinatto, J.A., Cia, E.M.M., Carvalho, R.B.,Quadros, K.R.S., Alves-Filho, G., Mazzali, M., 2004. Urine

cytology as a screening method for polyoma virus active

infection. Transplant. Proc. 36 (4), 899901.Schonning, C., Leeming, R., Stenstrom, T.A., 2002. Faecal con-

tamination of source-separated human urine based on the

content of faecal sterols. Water Res. 36, 19651972.

WATER RESEARCH 40Siegrist, H., 1996. Nitrogen removal from digester supernatant

comparison of chemical and biological methods. Water Sci.

Technol. 34, 399406.Silva, E., Rajapakse, N., Kortenkamp, A., 2002. Something fromnothingeight weak estrogenic chemicals combined atconcentrations below NOECs produce significant mixtureeffects. Environ. Sci. Technol. 36, 17511756.

Slavin, T.J., Oleson, M.W., 1991. Technology tradeoffs related toadvanced mission waste processing. Waste Manage. Res. 9 (5),401414.

Smil, V., 2000. Phosphorus in the environment: natural flowsand human interferences. Annu. Rev. Energy Environ. 25,5388.

Steen, I., Steen, P., 1998. Phosphorus availability in the 21stcentury: management of a nonrenewable resource. Phos-phorus Potassium 217, 2531.

Strathmann, H., 1992. Ion-exchange membranes. In: Winston,W.S.S., Ho, K.K. (Eds.), Membrane Handbook. Chapman & Hall,New York, London, pp. 230245.

Strous, M., Heijnen, J.J., Kuenen, J.G., Jetten, M.S.M., 1998. Thesequencing batch rector as a powerful tool for the study ofslowly growing anaerobic ammonium-oxidizing microorgan-isms. Appl. Microbiol. Biotechnol. 50 (5), 589596.

Suzuki, K., Tanaka, Y., Osada, T., Waki, M., 2002. Removal ofphosphate, magnesium and calcium from swine wastewaterthrough crystallization enhanced by aeration. Water Res. 36(12), 29912998.

Thorneby, L., Persson, K., Tragardh, G., 1999. Treatment of liquideffluents from dairy cattle and pigs using reverse osmosis.J. Agric. Eng. Res. 73 (2), 159170.

Udert, K.M., Larsen, T.A., Biebow, M., Gujer, W., 2003a. Ureahydrolysis and precipitation dynamics in a urine-collectingsystem. Water Res. 37 (11), 25712582.

Udert, K.M., Larsen, T.A., Gujer, W., 2003b. Biologically inducedprecipitation in urine-collecting systems. Water Sci. Technol.:Water Supply 3, 7178.

Udert, K.M., Fux, C., Munster, M., Larsen, T.A., Siegrist, H.,Gujer, W., 2003c. Nitrification and autotrophic denitri-fication of source-separated urine. Water Sci. Technol. 48 (1),119130.

Udert, K.M., Larsen, T.A., Gujer, W., 2005a. Fate of majorcompounds in source-separated urine. Presented and pub-lished in the Proceedings of the Fourth World Water Congress,Marrakech, Morocco, September 2004; accepted for publica-tion in Water Sci. Technol.

Udert, K.M., Larsen, T.A., Gujer, W., 2005b. Chemical nitriteoxidation in acid solutions as a consequence of microbialammonium oxidation. Environ. Sci. Technol. 39, 40664075.

Valsami-Jones, E. (Ed.), 2004. Phosphorus in EnvironmentalTechnologyPrinciples and Applications. IWA Publishing,London, UK.

Vanchiere, J.A., White, Z.S., Butel, J.S., 2005. Detection of BK virusand simian virus 40 in the urine of healthy children.J. Med.Virol. 75 (3), 447454.

Van der Bruggen, B., Everaert, K., Wilms, D., Vandecasteele, C.,2001. Application of nanofiltration for removal of pesticides,nitrate and hardness from ground water: rejection propertiesand economic evaluation. J. Membrane Sci. 193, 239248.

Van Hulle, S., 2005. Modelling, simulation and optimization ofautotrophic nitrogen removal processes. Ph.D. Thesis, Facultyof Bioengineering Sciences, Ghent University, pp. 228. Down-load on the Internet: http://biomath.rug.ac.be/publications/download/vanhullestijn_phd.pdf.

von Gunten, U., 2003a. Ozonation of drinking water: part I.Oxidation kinetics and product formation. Water Res. 37,14431467.

von Gunten, U., 2003b. Ozonation of drinking water: part II.Disinfection and by-product formation in presence of bro-

006) 3151 3166 3165mide, iodide or chlorine. Water Res. 37, 14691487.Wieland, P.O., 1994. Designing for human presence in spacean

introduction to environmental control and life support

systems. NASA RP-1324, Appendix E/F, 227-251. Download onthe web: http://trs.nis.nasa.gov/archive/00000204/01/rp1324.pdf, e-mail author: paul.o.wieland@msfc.nasa.gov.

Wilsenach, J. A., Schuurbiers, C. A. H., van Loosdrecht, M. C. M.,(submitted) Phosphate and potassium recovery from sourceseparated urine through struvite precipitation. Submitted toWater Res.

Wilsenach, J., Van Loosdrecht, M., 2002. Separate urine collectionand treatmentoptions for sustainable wastewater systemsand mineral recovery. STOWA Report no. 2001-39, STOWA,Utrecht, ISBN 90-5773-197-5

Wilsenach, J.A., Van Loosdrecht, M.C.M., 2004. Effects of separateurine collection on advanced nutrient removal processes.Environ. Sci. Technol. 38 (4), 12081215.

Wood, F.C., 1982. The changing face of desalinationa consultingengineers viewpoint. Desalination 42, 1725.

Wu, Q.Z., Bishop, P.L., 2005. Enhancing struvite crystallizationfrom anaerobic supernatant. J. Environ. Eng. Sci. 3 (1), 2129.

Zapata, F., Roy R.N., 2004. Use of phosphate rocks for sustainableagriculture. FAO of the United Nations, Rome, p. 172, ISBN 92-5-105030-9. On the Internet: http://www.fao.org/documents/show_cdr.asp?url_file=/docrep/007/y5053e/y5053e00.htm.

ARTICLE IN PRESS

WAT E R R E S E A R C H 40 ( 2006 ) 3151 31663166

Treatment processes for source-separated urineIntroductionComposition of urineTreatment unitsHygienisationIntroductionStorage

Volume reductionIntroductionEvaporationFreeze-thawReverse osmosis

StabilisationIntroductionAcidificationPartial nutrification

P-recoveryIntroductionStruvite (MgNH4PO4)

N-recoveryIntroductionIon exchangeAmmonia strippingIsobutylaldehyde-diurea (IBDU) precipitation

Nutrient removal (P and N)IntroductionAnammox Process

Removal of micropollutantsIntroductionElectrodialysisNanofiltrationOzonation and advanced oxidation

Conclusions and outlookAcknowledgementsReferences