Elimination of phenols, ammonia and cyanide in wash water from ...

Q IWA Publishing 2009 Water Science & Technology—WST | 60.12 | 2009

3253

Elimination of phenols, ammonia and cyanide in wash water from biomass gasification, and nitrogen recycling using planted trickling filters Andreas Graber, Robert Skvarc and Ranka Junge-Berberovic´

ABSTRACT Trickling filters were used to treat wash water from a wood gasifier. This wash water contained

Andreas Graber Robert Skvarc Ranka Junge-Berberovic´ Institute of Natural Resource Sciences, Zurich University of Applied Sciences, Gruental, Waedenswil 8820, Switzerland E-mail: [email protected]; [email protected]; [email protected]

toxic substances such as ammonium, cyanide, phenols, and PAH. The goal was to develop a system that degraded toxic substances, and achieved full nitrification of ammonia. A 1 kW model wood gasifier plant delivered wash water for the experiments, which was standardised to a conductivity of 3 mS/cm by dilution. Toxicity was assessed by bacterial luminescence detection, germination test with cress (Lepidium sativum), and pot plants cultivated in a hydroponic setup irrigated continuously with the wastewater. Treatment experiments were done in both planted and unplanted trickling filters. Plant yield was similar to conventional hydroponic production systems. The trickling filters achieved complete detoxification of phenol, PAH and cyanide as well as full nitrification. The specific elimination rates were 100 g m23 Leca d21 for phenols and 90 g m23 Leca d21 for ammonium in planted systems. In unplanted trickling filters circulated for 63 h, phenol concentration decreased from 83.5 mg/L to 2.5 mg/L and cyanide concentration from 0.32 mg/L to 0.02 mg/L. PAH concentrations were reduced from 3,050 mg/L to 0.89 mg/L within 68 days. The assays demonstrated the feasibility of using the technique to construct a treatment system in a partially closed circulation for gasifier wash water. The principal advantage is to convert toxic effluents from biomass gasifiers into a non-toxic, nitrogen-rich fertiliser water, enabling subsequent use in plant production and thus income generation. However, the questions of long-term performance and possible accumulation of phenols and heavy metals in the produce still have to be studied. Key words

| biomass production, cyanide, effluent reuse, gasifier wash water, nutrient recycling, phenols, trickling filter

INTRODUCTION Biomass gasification offers an interesting alternative

gas exiting the combustion chamber passes through a

source of energy. A simple self-running gasifier technology

water curtain which serves to cool and eliminate particles

was developed by the Indian Institute of Science

and detrimental gases, especially sulphites that would

(Shara et al. 1997). In Switzerland, Giordano et al. (2001)

cause engine corrosion. This wash water contains varying

demonstrated that this technology can use a broad variety

concentrations of toxic substances (ammonium, cyanide,

of biomass for energy production: native wood chips or

phenols,

pellets, waste wood, pecan nut shells, coffee hull pellets

depending on biomass type, gasifier type, modus operandi

and even chicken litter pellets. In such gasifiers the raw

and water residence time in the wash water recirculation

doi: 10.2166/wst.2009.728

polycyclic

aromatic

hydrocarbons

(PAH)),

3254

A. Graber et al. | Elimination of phenols, ammonia and cyanide from biomass gasification wash water

Water Science & Technology—WST | 60.12 | 2009

sustainability

absorb simultaneously phenolic compounds, as well as Cu

of gasifier technology in a rural setting in the Indian

and Zn. In batch experiments, the plants incorporated

state of Karnataka by a multicriteria analysis, which

68 – 72% of added phenols within 24 h, but the removal

included technological, environmental and economic

trend levelled off after two days and remained low. He

analysis, as well as a sustainability assessment of the

showed that the roots played the crucial role in phenol

project. He found that the existing wastewater treatment,

removal,

consisting of four basins for sedimentation and recircula-

components.

system.

Buergi

(2003)

investigated

the

tion of wash water, did not produce satisfying treatment

suggesting

Oliveira

&

phenolic

Fernandes

oxidases

(1998)

as

reported

active phenol

for release in the environment, and concluded that further

elimination rates for petrochemical wastewater in Typha

biological or chemical wastewater treatment is necessary.

angustifolia with a retention time of 72 h. The system was

Constructed wetlands that feature aquatic plants would be

capable to efficiently eliminate phenol at inflow concen-

an interesting low-tech possibility for this task.

trations as low as 3.5 mg/L. Polprasert et al. (1996) evaluated the potential of a free water surface constructed wetland system in treating phenolic wastewaters. Within a

The role of plants

temperature range of 22– 308C, and at hydraulic retention

Aquatic plants can contribute directly or indirectly to the

times of 5 – 7 days, the constructed wetland units removed

removal of pollutants from water and sediment. According

more than 99% of the input phenol concentration of

to Crowley et al. (1997), sporadic reports have appeared

400 mg/L when they were operated at organic loading

in the literature since the late 1970s demonstrating that

rates of 40– 140 g COD ha21 d21.

plants may enhance the degradation of several different

This paper investigates the possibilities of combining

compounds, including organophosphates, parathion, PAH,

wastewater treatment in constructed wetlands with pro-

pentachlorophenol and organic solvents such as trichloro-

duction of usable plant biomass by using trickling filters as

ethylene. Phytovolatilisation, microbial activity in the

substrate and production area for plants, thus combining

rhizosphere (Crowley et al. 1997), transpiration and soil

the processes of nutrient transformation and nutrient

matrix storage (Kuschk et al. 1999) are of varying import-

recycling with detoxification. Trickling filters are widely

ance for remediation in wetlands. Aquatic plants can

applied to treat wastewaters of different sources and

enhance microbial activity in sediments (Karjalainen et al. 2001) and remediate wastewaters containing excess nutrients, organic pollutants and heavy metals. Kuschk et al. (1999) showed that plants are able to detoxify various xenobiotics, but in this process plants play a minor role as compared to microbes. First tests on removal and tolerance of phenolic compounds by macrophytes were done by Seidel (1968). Further studies used Scirpus lacustris (Kickuth 1970) and

compositions, where their main purpose is to provide nitrification and BOD removal. Unfortunately, the conventional trickling filter approach does not make use of nutrients in the waste water, but merely transforms them to non-toxic (H2O, NO3) or gaseous forms (CO2, N2). To evaluate the potential of planted trickling filters for treating toxic wastewaters, we adapted the system used for nitrification in Aquaponic (Graber & Junge 2009) to

Lemna gibba (Barber et al. 1995). Water hyacinths,

treat wastewater effluent from a biomass gasifier. The goal

Eichhornia crassipes, in open water units were tested by

was to develop a system that degrades toxic substances and

several researchers (Wolverton & McKown 1976; O’Keeffe

achieves full nitrification of ammonia. The treated water

et al. 1987). Vaidyanathan et al. (1983) measured 89%

should be used as a nitrogen fertiliser either for direct

phenol

concentration,

irrigation of crop plants or as fertiliser in fish ponds. In

whereas unplanted control units achieved only 13%

order to define a treatment system applicable at large scale,

removal. The purification performance improved over 13

basic parameters such as the composition of wash water

days and remained stable after this time. Nor (1994)

resulting from gasifier operation and the volume to be

found that Eichhornia crassipes has a vast capacity to

treated were determined.

removal

at

75 mg/L

influent

3255



In this paper, all LECA used was bought as “Blaehton

METHODS

rot” from Ricoter, a Swiss supplier (for use in hydro-

A model wood gasifier plant with a capacity of 1 kW

ponics, size 8 –16 mm, density 440 g/L). It is heavy

(IISc open top, Netpro Renewable Energy Ltd, India)

enough to provide secure support for the plants’ root

was operated half-automated during daytime. As the

systems and was successfully used in hydroponic systems

gasifier wastewater composition varies over time, the

(Graber & Junge, 2009). To ensure the build-up of an

raw wastewater was standardised by dilution to a salt

active biofilm on the LECA before starting the exper-

concentration equal to an electric conductivity of 3 mS/cm.

iment, the filters were inoculated four times during 55

This value was found to be the upper limit for plant

days by adding ammonium chloride to set a concentration

growth from previous experiments (Staudenmann &

of 10 mg NH4-N/L.

Junge-Berberovic 2003; Graber & Junge-Berberovic 2008).

The second treatment experiment compared the

Toxicity of the standardised gasifier wastewater was

treatment performance of three planted trickling filter

assessed by EC50 (half maximal effective concentration) and

towers (Xanthosoma sp., Cyperus alternifolius, Carica

NOEL (no observed effect level) values using bacterial

papaya), and one unplanted control. Two boxes were

luminescence detection (Hach Lange LCK 491), a germina-

stacked one upon the other, planted on all outsides

tion test with cress (Lepidium sativum) (OECD 2000,

(eight sides of the two boxes plus the top surface) and

five replicates, 20 seeds per assay), and four pot plant

placed in a metal tray of 1 m2. Washwater was continuously

species cultivated in a hydroponic setup and irrigated

pumped (Q ¼ 1,000 L/h) to the top of the tower and

continuously with the wastewater. Each plant (two –three

distributed evenly over the surface with perforated tubes;

replicates) was cultivated in a control solution and in

the outflow water (total volume 20 L) was collected in the

three different concentrations (20, 50 and 100%) of the

metal tray first and finally in 30 L plastic boxes placed

standardised wastewater. To exclude growth limitation by

underneath the trays, where the pump was installed.

insufficient nutrient supply, all four trials were fertilised

For the planted filters, boxes from the first experiment

with nitrogen, phosphorus, potassium, calcium, magnesia

were used, whereas the control was started with fresh LECA

and trace elements. The wastewater was renewed twice per

without any inoculation. All experiments were done in a

week; the experiment lasted 26 days.

greenhouse (temperature 23 ^ 38C) at the institute in

The first experiment assessed a possible limitation of

Waedenswil, Zurich.

treatment performance by the lack of essential nutrients or minerals in the wash water. The treatment performance was compared between trickling filters irrigated with a

Monitoring and balancing of recirculation systems

standardised wash water diluted to 50% either by

Electric conductivity (EC), pH, redox potential, and

rainwater collected from the greenhouse roof (pH 7.13,

dissolved oxygen (DO) were either measured continuously

EC 178 mS/cm) or by water from a fish tank (pH 7.4,

and logged every 15 min with SC1000 sensors (Hach Lange

EC 340 mS/cm, TN 20 mg/L, TP 15 mg/L), or by water

sc-sensors 3798-S, phD-S, 1200-S, LDO), or measured daily

supplemented

fertiliser

with a hand-held multi-electrode meter (Multiline 350i,

for hydroponic vegetables (pH 5.5, EC 2,500 mS/cm,

WTW). Dissolved ions were determined by photometry

TN 250 mg/L, TP 50 mg/L). Three replicates were used for

(LCK tests, Cadas 30, Hach Lange).

with

micronutrients

from

a

each treatment. Each experimental unit consisted of a closed

recirculation

system

consisting

of

one

box

Plant growth was assessed by measuring growth length of the leaves.

(0.4 £ 0.6 £ 0.4 m, green PVC) holding 40 L of LECA,

In gasifier wash water, 16 PAH on the priority list of US

placed over a water tank holding 25 L of wastewater with a

EPA were measured by EMPA laboratory according to

submersed aquarium pump (Q ¼ 1,000 L/h) running con-

ISO/IEC 17025 in a filtered sample (0.45 mm): 2- and 3-ring

tinuously. LECAY (Light Expanded Clay Aggregate) is a

PAH were measured by GC-MS, 4-5-6-ring PAH were

type of clay which is super-fired to create a porous medium.

detected by HPLC. Metals were measured by ICP-OES by


3256

BMG Labors, Zurich. Since all experiments were designed

Table 1

|


Analysis of wash water from a 1 kW wood gasifier diluted to 3 mS/cm

as batch systems, samples were taken in the water reservoir only and elimination rates were calculated.

RESULTS Gasifier wastewater To maintain a steady EC value of 3 mS/cm in the wash water circulation of the gasifier, 3.8 L of wash water needed

Parameter

Unit

Value

Parameter

Unit

Value

pH

–

7.21

TN

mg/L

770

EC

mS/cm

3,000

NH4-N

mg/L

374

TOC

mg/L

393

NO2-N

mg/L

0.1

DOC

mg/L

307

NO3-N

mg/L

2.8

COD

mg/L

912

PO4-P

mg/L

0.3

TSS

mg/L

,0.5

K

mg/L

16

BOD5

mg/L

,20

Cyanide CN

mg/L

0.28

BOD21

mg/L

,40

Total Phenols

mg/L

182

PAH

mg/L

3.05

to be replaced per kg biomass burned (Figure 1). The resulting wastewater contained high concentrations of ammonium and phenols (Table 1). The phenol compounds contributed to a specific “smoky” smell of the wastewater. Detailed analysis revealed that 1.7% of phenols consisted of PAH, 2- and 3-ring PAH making up 3.033 mg/L or 99.6% of total PAH. As additional possible plant toxicant, cyanide

indicating that toxicity had been reduced by a factor of 2,000. The four tested plant species grew as well in a wastewater concentration of 50% as in the control, only in pure wastewater growth was reduced slightly (Figure 2).

ions were detected. A screening of heavy metals revealed concentrations below 0.05 mg/L for 16 elements, 0.19 mg/L for manganese and 0.13 mg/L for zinc. The BOD:COD ratio

Treatment experiments

was very low, 4% if BOD is set to 40 mg/L. Phosphorus and

In the first experiment, there was no significant difference

potassium were low compared to total nitrogen content, the

between the treatments, i.e. different additives to the

N:P:K ratio being 2,500: 1: 53. Of total nitrogen, only 49%

washwater, indicating that nitrification in pure wash water

was attributed to ammonia, nitrite and nitrate.

is not limited by the lack of nutrients. To get an overview about the ongoing processes in the trickling filters, results

Toxicity tests

of all units were pooled (Figure 3). Phenol was removed within 70 h from 174 ^ 3.6 mg/L to 9.1 ^ 2.0 mg/L; ammonia

Toxicity testing followed the typical pattern that higher

removal was slower, reaching 6.7 ^ 7.6 mg/L after 29 days.

organisms tolerate higher concentrations of toxicants

Nitrification was in the range of 50– 141% of the initial

(Table 2). Additionally, toxicity was assessed in wash

ammonia nitrogen. PAH concentrations were reduced from

water that had been treated during four weeks in an

3,050 mg/L to 0.89 mg/L within 68 days.

unplanted LECA trickling filter. This pre-treated wash water had no effect on bacteria cells in a dilution of 1:10,

In the second experiment, planted filter towers achieved higher elimination rates than unplanted filters (Figure 4). Nitrate build-up remained low and peaked at 44 mg/L (4– 36% nitrification, uptake by plants not included) in planted and at 8 mg/L (3% nitrification) in unplanted Table 2

|

Toxicity of raw gasifier wash water, diluted to a salt concentration equal to EC 3 mS/cm

NOEL (dilution factor)

EC50 (dilution factor)

,1: 20,000

1: 1,000

Cress

1: 1,000

1: 8

Macrophytes

1: 5

–

Bacteria

Figure 1

|

EC increase in the wash water and cumulative biomass burned.


3257

Figure 4

|

Figure 2

Length growth of pot plants irrigated with different concentrations of gasifier wash water.

|


Elimination of ammonium and phenol in planted (n ¼ 3) and unplanted trickling filters.

could grow at cyanide levels up to 20 mg/L. The highest systems. The specific elimination rates in planted systems were

100 g m23 LECA d21

for

phenols

and

90 g m23

phenol concentration in the 20% assay was 28 mg/L; at this concentration plant growth might have been stimulated by

LECA d21 for ammonium. Cyanide was reduced from

the presence of phenols. Huebner et al. (2000) tested the

0.37 mg/l to 0.10 mg/l in the unplanted filter and from

germination and growth response of Phalaris arundinacea

0.29 mg/l to 0.04 mg/l in the planted filters within 16 hrs.

and Phragmites australis when exposed to different phenol and phenanthrene concentrations. Both plants showed stimulated growth, more intensive green colouring

DISCUSSION Toxicity of gasifier wash water

and increased shoot production up to phenol concentrations of 100 mg/L. A damaging effect occurred only in treatments with higher doses. The stimulating effect of

Plants thrived well and no acute toxicity was observed even

phenol was explained by its uptake and conversion into

when irrigated with pure wash water, as long as EC was

the plant biomass.

below 3 mS/cm. Thus acute toxicity levels for plants are expected only at higher salinity levels. A growth depression

Detoxification by trickling filters

indicating a chronic toxicity was observed at 50% and 100% assays compared with the control, whereas no difference

The assays demonstrated the detoxifying potential of

was observed for the 20% assay. The start concentrations in

planted trickling filters and proved the feasibility of using

the 100% assay were 84 mg/l for phenols and 0.32 mg/l for

the technique to construct a partially closed treatment

cyanide. At these levels cyanide was possibly not a limiting

system for gasifier wash water. Removal of phenols might be

factor, as Granato (1993) found that Eichhornia crassipes

due to adsorption processes to plant roots or filter medium, as well as evaporation, which could be promoted by continuous spraying of the water over the trickling filter.

Treatment efficiency Nitrification occurred in all systems, but with varying efficiency. The findings in unplanted trickling filters suggest that ammonia might be adsorbed to LECA and released later as nitrate (Figure 3). Still, the complete recovering of ammonia-N as nitrate-N (104 ^ 29% for all nine filters) Figure 3

|

Elimination of ammonium and phenol and build-up of nitrate in unplanted trickling filters (n ¼ 9).

shows that ammonia was not lost in gaseous form. The high variability in the nitrification ratio and the low efficiency in

3258



the filter towers in the second experiment indicate that full

nitrate, a fertiliser commonly used in agriculture. A planted

nitrification cannot always be expected. The role of

trickling filter designed to keep a constant water quality

particulate nitrogen, which made up 51% of total nitrogen

would be able to treat 140 L washwater m23 Leca d21. From

in the washwater (Table 1), remains unclear, and it could be

a 100 kWe gasifier, a size typically installed in rural villages

the source of nitrogen leading to 141% nitrification

in India, 1 to 4 kg nitrogen d21 could be recycled (270 kg/h

observed in the first experiment. The fast elimination of

biomass burned leads to 25 m3/d treated wastewater with a

ammonia and phenol in the unplanted filter in the second

nitrogen concentration of 50 – 200 mg/L). The questions of

experiment, that had no biofilm on the LECA, might also be

long-term effects and possible accumulation of phenols and

explained by adsorption (Figure 4).

heavy metals in vegetables still have to be studied.

In all filter units tested, even after 34 days of constant phenol loading no saturation of the filter capacity was reached. Phenol elimination rates reached 20 g m22 d21, whereas Polprasert et al. (1996) found 32 g m22 d21, possibly

ACKNOWLEDGEMENTS

due to the higher inflow concentrations (380 mg/L versus

The authors would like to thank the Swiss Federal Office

180– 200 mg/L in this study). Wiessner et al. (1999) treated

for the Environment for financing this study.

wastewater from lignite pyrolysis which was deposited untreated in open ponds during the long era of industrial lignite exploitation in eastern Germany. Using a planted subsurface-flow constructed wetland of 125 m2 to reach the same goals as in this paper, they were able to remove 5–12 g COD g m22 d21 and nitrify 0.3–1.2 g NH4-N m22 d21. PAH degradation was possibly performed by microbial activity, as PAH with 2 and 3 rings were degraded in earlier experiments using porous organoclay (Ake et al. 2003). The PAH concentrations in unplanted trickling filters almost reached drinking water limits of 0.1 mg/L.

CONCLUSIONS This paper affirms the study of Polprasert et al. (1996) that a continuously operated large scale filter system to treat wash water from biomass gasifiers should be technically feasible. Trickling filters have the capability to evaporate considerable

amounts

of 23

evaporated 100– 400 L m

water: 21

LECA d

the

filter

towers

depending on the

season (Graber & Junge 2009). Because plants are able to reduce the conductivity in the partially closed water recirculation system by nutrient uptake, the required addition of fresh water could be reduced substantially below the 3.8 m3 per ton of biomass burned found for unplanted filters. After detoxification of phenols and cyanide, the effluent could be used as a nitrogen fertiliser in agriculture, as most nitrogen would be converted into

REFERENCES Ake, C., Wiles, M., Huebner, H., McDonald, T., Cosgriff, D., Richardson, M., Donelly, K. & Philipps, T. 2003 Porous organoclay composite for the sorption of polycyclic aromatic hydrocarbons and pentachlorophenol from groundwater. Chemosphere 51, 835 –844. Barber, J., Sharma, H., Ensley, H., Polito, M. & Thomas, D. 1995 Detoxification of phenol by the aquatic angiosperm Lemna gibba. Chemosphere 31, 943– 959. Buergi, P. 2003 Sustainability assessment of bioenergy generation in the south-indian state of Karnataka—multicriteria analysis of a biomass gasification power plant. Diploma Thesis Swiss Federal Institute of Technology, Zurich, p. 179. www. southpolecarbon.com Crowley, D. E., Alvey, S. & Gilbert, E. S. 1997 Rhizosphere ecology of xenobiotic-degrading microorganisms. In: Kruger, E., Anderson, T. & Coats, J. (eds) Phytoremediation of Soil and Water Contaminants. American Chemical Society, Washington, DC, pp. 20 –36. Giordano, P., Hasler, P. & Dasappa, S. 2001 Test of Different Biomass Into the IISc Open Top Co-Current Gasifier. Swiss Federal Office for Energy, Berne, Final report of Project 39772. Graber, A. & Junge-Berberovic, R. 2008 Wastewater-fed Aquaculture Otelfingen, Switzerland: influence of system design and operation parameters on the efficiency of nutrient incorporation into plant biomass. In: Vymazal, J. (ed.) Wastewater Treatment, Plant Dynamics and Management in Constructed and Natural Wetlands. Springer Verlag, Berlin, p. 350. Graber, A. & Junge, R. 2009 Aquaponic systems: nutrient recycling from fish wastewater by vegetable production. Desalination 246(1 –3), 147 –156. Granato, M. 1993 Cyanide degradation by water hyacinths, Eichornia crassipes (Mart.) solms. Biotechnol. Lett. 15(10), 1085 –1090.

3259


Huebner, T. M., Tischer, S., Tanneberg, H. & Kuschk, P. 2000 Influence of Phenol and Phenanthrene on the Growth of Phalaris aurundinaceae and Phragmites australis. Int. J. Phytoremediation 2(4), 331 –342. Karjalainen, H., Stefansdottir, G., Tuominen, L. & Kairesalo, T. 2001 Do submersed plants enhance microbial activity in sediment? Aquat. Bot. 69(1), 1–13. Kickuth, R. 1970 Oekochemische Leistungen hoeherer Pflanzen. Naturwissenschaften 57, 55 –61. Kuschk, P., Wiessner, A. & Stottmeister, U. 1999 Biological processes in wetland systems for wastewater treatment. In: Rehm, H.-J. & Reed, G. (eds) Biotechnology Environmental Processes I: Wastewater Treatment. Wiley/VCH, Weinheim, pp. 241– 251. Nor, Y. 1994 Phenol removal by Eichhornia crassipes in the presence of trace metals. Water Res. 28(5), 1161 –1166. O’Keeffe, D., Wiese, T., Brummet, S. & Miller, T. 1987 Uptake and metabolism of phenolic compounds by the water hyacinth Eichhornia crassipes. Recent. Adv. Phytochem. 21, 101 –121. Oliveira, J. & Fernandes, A. 1998 Portugal. In: Vymazal, J., Brix, H., Cooper, P. F., Green, M. B. & Haberl, R. (eds) Constructed Wetlands for Wastewater Treatment in Europe. Backhuys Publishers, Leiden, pp. 227 –240. OECD 2000 Guideline for the testing of chemicals, TG208 Terrestrial plant test.


Polprasert, C., Dana, N. & Thayalakumarana, N. 1996 Application of constructed wetlands to treat some toxic wastewaters under tropical conditions. Water Sci. Technol. 34(11), 165 –171. Seidel, K. 1968 Elimination von Schmutz- und Ballaststoffen aus belasteten Gewaessern durch hoehere Pflanzen. Vitalst. Zivilisationskr. 4, 154 –160. Sharan, H. N., Mukunda, H. S., Shrinivasa, U. & Dasappa, S. 1997 IISc-DASAG biomass gasifiers: development, technology, experience and economics. Development in Thermochemical Biomass Conversion, IEA Bioenergy II, Bridgewater, A. K. & Boocock, D. G. B. (Eds), IEA, Vienna, pp. 1058 –1073. Staudenmann, J. & Junge-Berberovic, R. 2003 Recycling nutrients from industrial wastewater by aquaculture systems in temperate climates (Switzerland). J. Appl. Aqua. 13, 67 –103. Vaidyanathan, S., Kavadia, K., Rao, M., Basu, S. & Mahajan, S. 1983 Removal of phenol using water hyacinth in a continuous unit. Int. J. Environ. Stud. 21, 181– 191. Wiessner, A., Kuschk, P., Stottmeister, U., Struckmann, D. & Jank, M. 1999 Treating a lignite pyrolysis wastewater in a constructed subsurface flow wetland. Water Res. 33(5), 1296 –1302. Wolverton, B. & McKown, M. 1976 Water hyacinths for removal of phenols from polluted waters. Aqua. Bot. 2, 191– 202.