Eichhornia crassipes used as tertiary color removal treatment for Kraft mill effluent

DESALINATION

www.elsevier.com/locate/desalDesalination 247 (2009) 46–55

Eichhornia crassipes used as tertiary color removal treatment forKraft mill effluent

Cesar Lagosa, Roberto Urrutiaa, Jacqueline Decapa, Miguel Martınezb,Gladys Vidala*

aEnvironmental Science Center EULA-Chile, Universidad de Concepcion, P.O. Box 160-C, Concepcion-ChilebDepartment of Microbiology, Biological Sciences Faculty, Universidad de Concepcion, P.O. Box 160-C, Concepcion

Tel: +56 (41) 2204067; Fax: +56 (41) 2207076; email: [email protected]

Received 14 October 2007; revised 28 December 2007; accepted 04 March 2008

Abstract

Chile annually produces 2 million tons of kraft mill pulp using pine and eucalyptus as raw materials. In spite ofthe primary and secondary treatment processes installed in almost all of the industries, the discharged effluentsstill contain color, affecting light transmission in aquatic systems. At present, 75 tons of color is produceddaily by industrial processes that require 57,000 m3/d of fresh water for processing.

This chapter evaluates a tertiary treatment with Eichhornia crassipes that is used to remove color and organiccompounds from kraft mill effluent.

E. crassipes removed 46–75% of organic matter and 11–17% of total phenolic compounds. Under experimen-tal conditions, E. crassipes was able to remove around 8.5% and 23.6% of color when the assays were done with50% and 10% kraft mill effluent, respectively.

Keywords: Eichhornia crassipes; Color; Aromatic compounds; Biodegradation; Kraft mill effluent

1. Introduction

Chile annually produces 2 million tons of kraftmill pulp using pine and eucalyptus as raw materi-als. In the last few years, kraft mills have substan-tially updated pulping and bleaching technologies,transforming the effluent’s biodegradation andtoxicity, although these effluents still require pri-mary and secondary treatments. Organic substrate

content in the kraft mill effluents may be reducedby secondary treatment [1], although it is wellknown that color and specific compounds remaineven after biological treatment due to diversearomatic compounds, ranging from simple mono-mers to high molecular weight (MW) polyphe-nolic polymers, often found in the aerobiceffluent [2–4]. At present, 75 tons of color is pro-duced daily by an industry that requires57,000 m3/d fresh water for the process. Thehigh MW fraction of the total organic chlorides*Corresponding author.

Presented at Multi Functions of Wetland Systems, International Conference of Multiple Roles of Wetlands, June 26–29, 2007, Legnario (Padova) Italy

0011-9164/08/$– See front matter � 2009 Elsevier B.V. All rights reserved

(MW > 1000 Da) is only slightly affected by anaerobic treatment, and chemical oxygen demand(COD) removal will generally not exceed 45%with these systems [5,6]. Still, Diez et al. [7] indi-cate that high MW aromatic components, such aslignin and tannins, are simply not biodegradable inaerobic environments. Similar results were shownby Rintala and Lepist€o [8], who worked with ananaerobic–aerobic sequence to treat thermome-chanical pulping whitewater. Recent studiessuggest that the color increase in kraft mill biolog-ical treatment systems may be caused by anaero-bic bacteria using high MW material frombleaching effluents as an electron acceptor forgrowth, resulting in material reduction, which inturn leads to nonreversible internal changes,such as intramolecular polymerization or the for-mation of chromophoric functional groups [9].

Since the color discharged could affect lighttransmission in aquatic systems, color removalby an alternative tertiary treatment or physico-chemical treatment must be explored [1].

Eichhornia crassipes, a dominant floating mac-rophyte present in natural lakes in Chile, has a highcapacity to incorporate nitrogen from both ammo-nium and nitrate sources [10,11] as well as a largesubmerged root system that supplies oxygen to therhizosphere. Previous studies show that E. cras-sipes is able to remove nitrogen in a range between50.1 and 163.0 mg N/g dry weight [10,11]. Addi-tional studies demonstrate that the biomass yieldincreases with an increase of phosphorus supplyup to 1.06 mg P/l; under that condition, phosphoruscontent from 3.8 to 4.3 mg P/g was measured in theplant tissue [12,13].

On the one hand, specific studies with aro-matic compounds show that E. crassipescould remove these compounds with adsorptionmechanisms [14] and/or biodegradation [15].Thus, basic dyes, such as methylene blue andVictoria blue, are removed when E. crassipesis used as sorbent. Maximum sorption capacitiesof water hyacinth roots for methylene blueand Victoria blue were 128.9 and 145.4 mg/g,

respectively [14]. On the other hand, 2.75 g drymatter of E. crassipes demonstrated the abilityto absorb 100 mg of phenol in 72 h [15]. More-over, Nor [16] shows that this aquatic plant isable to completely remove 200 mg/l of phenolsin the presence of trace metals (Cu, Zn) after anhydraulic retention time of 6 h. Still, no reportstudying color and aromatic compound behaviorin the presence of aquatic plants was found.

The objective of the present study is to eval-uate color removal and aromatic compoundbehavior in kraft mill effluents during tertiarytreatment by E. crassipes.

To corroborate aromatic behavior and colorremoval, the behavior of Poly R-478 as amodel compound was studied in parallel.

2. Materials and methods

2.1. Effluent

Effluent was obtained from a kraft mill thatuses elementary chlorine free pulp bleachingprocess. Effluent was collected after primarytreatment, which consisted in a settling tank toreduce fiber and suspended solid content.

2.2. Plants

Collection E. crassipes was collected at thebeginning of spring from the Tres Pascualaslagoon, VIII Region, Concepcion, Chile (36849’S; 738 03’W). New plants (with new greenlips from the spring season) were selected; thelongest and abundant leaves (four or five leavesfor each plant) were used as a classificationindex. The plant’s root length and size werealso considered. Plants were collected directlyfrom the lagoon and transported in an appropri-ate container with water from the lagoon.

Assay inoculation Each pond was inocu-lated with five plants. The biomass of each ofthe five plants from each pond was measuredat the inoculation, and then at 3 and 6 months

C. Lagos et al. / Desalination 247 (2009) 46–55 47

of operation time. The procedure to measurefresh biomass was the following: each plantwas weighed after being placed on adsorbent-water paper for 30 min.

2.3. Treatments

Figure 1 shows the experimental design for theplanted basins that receive kraft mill effluent(Fig. 1a: 100% (assay A), 50% (assay B) and10% (assay C) of kraft mill effluent) and a dyefor lignin (Fig. 1b: Poly R-478 in the concentra-tions 0.2 (assay A), 0.1 (assay B) and 0.05(assay C) g/l). Each basin was built in glass mate-rial (length�width� depth: 0.25� 0.12 0.2 m)with 6 l of volume. In each case, the controlassays (without plants) received kraft mill efflu-ent (Fig. 1a: 100% (a), 50% (b) and 10% (c) ofeffluent) or Poly R-478 (Fig. 1b: 0.2 (a); 0.1 (b)and 0.05 (c) g/l) and water with nutrients. Awater control (Fig. 1a–f) was also included. Alllaboratory-scale assays were performed in tripli-cate. For kraft mill effluent and Poly R-478,nine different lagoons were installed in the laband inoculated with 79+ 2 g of E. crassipes.The dilution water used was tap water. The efflu-ents and Poly R-478 solution were supplied withnutrients, whose dose was in the ratio N:P:K,1:1:0.6 (urea, P2O5 and K2O were used as N, Pand K sources, respectively).

Natural photoperiod (12 h light/12 h dark)was used, and the temperature was 20+ 28C.Experimental assays were performed from Sep-tember to March, where time zero was definedas the time of inoculation of each pond.

During the whole operational period, eachpond’s efficiency was estimated for COD,BOD5, total phenolic compounds, color, ligno-sulfonic acids, aromatic compounds and lignin-derived compounds.

2.4. Analytical methods

The samples were collected from each pond’scenter. COD, BOD5, phenols, lignin, tannins, total

suspended solids (TSS), total nitrogen Kjeldhal(TNK) and phosphate were measured followingAPHA-AWWA-WPCF [17]. The total phenoliccompound (UV phenols) concentration was meas-ured by UV absorbance in a 1-cm quartz cell at215 nm, pH 8.0 (0.2 M KH2PO4 buffer) and trans-formed to concentration using a calibration curvewith phenol as standard solution. All the sampleswere membrane filtered (0.45mm). Spectrophoto-metric measurements of filtered samples wereprincipally performed at wavelengths of 436 nm(color), 346 nm (lignosulfonic acids), 254 nm(aromatic compounds) and 280 nm (lignin-derived compounds) in a 1� 1 cm quartz cellusing a Genesys UV-VIS spectrophotometer(Spectronic Unicam, Genesis 10 UV) and accord-ing to the Cecen [18] procedure. Color was meas-ured according to APHA-AWWA-WPCF [17]using U Pt/Co.

2.5. Statistical analysis

Batch experiments were performed in tripli-cate. Analysis of variance (ANOVA), followedby multiple comparison using the FriedmanANOVA procedure at the 5% error level, wasperformed with SPSS version 9.0 software.

3. Results and discussion

The characteristics of the kraft mill effluent areshown in Table 1. The pH ranged from 3.5 to10.6. The average soluble COD was 1196 mg/l.According to BOD5 values, only 32% of theorganic matter is aerobically biodegradable.Total phenolic compounds ranged between 190and 350 mg/L, whereas 18–23% corresponded tothe tannin and lignin degradation products and0.3–0.5% corresponded to the phenols com-pounds. The average color was 1100 U Pt/Co,and average phenol compound was 1.0 mg/l.

Figure 2 and Table 2 show the COD, BOD5

and total phenolic compounds removal efficien-cies in the A, B and C assays for kraft mill

48 C. Lagos et al. / Desalination 247 (2009) 46–55

Controls

A

B

C

a

b

c

Concentrations

200 mg/l

100 mg/l

50 mg/l

Poly R-478 Water + Nutrients

a)Planted basins Controls without plants

A

B

C

a

b

c

f

Dilutions

100 %

50 %

10 %

Kraft mill efluent Water + Nutrients

Planted basins b)

Fig. 1. Experimental design: (a) Kraft mill effluent and (b) Poly R-478 assays.


effluent. In both cases, COD and BOD5 remov-als were between 54.6–60.7% and 31.3–67.8%,respectively, for assays A and B. In thosecases, total phenolic compounds were removed16.4% and 17.1% on average, respectively. Inthe case of assay C, the COD (45.9%) andtotal phenolic compound removals (10.5%) arelower than in assays A and B.

On the other hand, E. crassipes showed nocolor removal capacity in the assay with 100%

effluent (assay A), and color intensity evenincreased by 2.8%. However, in the assayswith 50% and 10% of kraft mill effluent, acolor removal of 8.5 and 23.6% on averagewas observed. Considering these values, it isvery likely that in the case of 100% kraft milleffluent, polymerization by aerobic bacteria onE. crassipes roots occurred due to the high phe-nolic compound (313 mg/l) presence and oxygenconcentration. Oxygen concentrations in the

Table 1

Physical – chemical characteristics of Kraft mill effluent

Parameter Unit Average Range

pH 7.0 3.5–10.6COD Mg/l 1196 850–1540BOD5 Mg/l 380 223–540Total phenolic compounds UV215 Mg/l 313 190–350Phenols Mg/l 1.0 0.9–1.2Tannin and lignin Mg/l 52.2 44.0–64.0Color U Pt/Co 1100 950–1400TSS g/l 1.8 1.2–3.1TNK Mg/l 6.0 3.0–9.0Phosphate Mg/l 0.8 0.4–1.3

CO

D, B

OD

5,T

otal

phe

nolic

s co

mpo

unds

rem

oval

(%

)

Assay A Assay B Assay C0

20

40

60

80

100

Fig. 2. Removal efficiency for COD (&); BOD5 ( ) and total phenolic compounds (�) from kraft mill effluent treatedby E. crassipes.


Tab

le2

Rem

ov

alef

fici

ency

for

org

anic

com

po

un

ds

inK

raft

mil

lef

flu

ent

and

po

lyR

-47

8w

hen

trea

ted

by

E.

cra

ssip

es

CO

D(%

)B

OD

5(%

)T

ota

lp

hen

oli

cco

mp

ou

nd

s(%

)C

olo

r(%

)

Ass

ays

Aver

age

Ran

ge

Aver

age

Ran

ge

Aver

age

Ran

ge

Aver

age

Ran

ge

Kra

ftm

ill

effl

uen

tA

58

.45

7.3

–6

0.7

75

.05

3.5

–6

7.8

16

.49

.3–

18

.3–

2.8

�3

.5–�

1.1

B5

5.6

54

.6–

57

.16

2.5

31

.3–

50

.01

7.1

5.7

–3

1.0

8.5

1.3

–1

5.1

C4

5.9

32

.8–

47

.1–

#–

10

.57

.4–

10

.32

3.6

23

.6P

oly

R-4

78

A�

2.7

�4

.7–

6.9

––

––

6.6

4.4

–9

.8B

�1

3.7

�1

8.1

–�

3.0

––

––

1.3

�4

.4–

6.2

C�

39

.4�

39

.4–

––

–1

.6�

0.9

–4

.6

#N

ot

det

erm

ined

.

experiment ranged from 3.5 to 5.0 mg/l. Thisphenomenon was also observed by Diez et al.[7] and Milestone et al. [9]. In the case of lowkraft mill effluent concentrations (assays B andC), sorption phenomena could be occurring[14,15].

In the assays with Poly R-478, the CODincreased, probably due to the E. crassipes bio-mass missing in fall season; meanwhile a littlecolor removal was observed in the trials A(6.6%), B (1.3%) and C (1.6%) (Table 2). Mostof the Poly R-478 removal could be attributedto the sorption phenomena [14,15].

Table 3 shows the aromatic compound behav-ior in the kraft mill effluent assays. The behaviorof lignosulfonic acid (VIS346 /COD) content inthe kraft mill effluent shows a range of 0.035–0.065 for assay A, whereas for assays B and Cthe ranges were 0.038–0.051 and 0.032–0.040,respectively. The relationship VIS346 /CODincreased, indicating low biodegradation ofthese compounds by E. crassipes. In the sameway, lignin behavior (measured as UV280 /COD)shows a range between 0.026 and 0.051 in the dif-ferent assays, indicating that E. crassipes miner-alizes the lignin compounds to a lesser extentthan organic matter, measured as COD.

The UV254/UV280 relationship is used asan indicator of the presence of lignin-derivedcompounds in wastewaters, where low values

indicate a higher percentage of these compounds[18]. During the treatment by E. crassipes, theUV254/UV280 relationship in the different assayswas around 1.206–1.340. Similarly, Cecen [18]shows that UV254/UV272 did not undergo asignificant change (ranging between 1.1 and1.13). These results suggest that the residualCOD consisted in lignin compounds, whichwere also the major aromatic species in theseeffluents. The same results were found byChamorro et al. [19].

E. crassipes’s low removal efficiency forlignin-derived compounds and color could indi-cate the impossibility of this treatment. How-ever, the E. crassipes yield in the presence ofkraft mill effluent, even when it was not diluted,confirms its possible application to removenutrient content in kraft mill effluent. Similarresults were obtained by Dellarossa et al. [10]and Casablanca [20].

Figure 3 shows the kinetic behavior of COD,BOD5 and pH during kraft mill effluent treat-ment by E. crassipes during the 6 months ofthe experiment. The pH decreased during theassays, and there was a significant differencebetween assays and the control (p < 0.05). Inthe assays, the pH ranged between 7.5 and 6.6;whereas in the control assays, it ranged between7.8 and 7.0. In the fall, it was observed that E.crassipes leaves died off and decomposed at

Table 3

Behavior of aromatic compounds during biological treatment of Kraft mill effluent by E. crassipes

AssaysLignosulfonic acid

VIS346/CODAromatic compounds

UV254/CODLignin

UV280/CODLignin-derived compounds

UV254/UV280

A Mean 0.052 0.014 0.042 1.254Range 0.035–0.065 0.010–0.018 0.028–0.051 1.250–1.276

B Mean 0.047 0.012 0.038 1.246Range 0.038–0.051 0.010–0.013 0.030–0.040 1.250–1.251

C Mean 0.038 0.008 0.028 1.260Range 0.032–0.040 0.007–0.010 0.026–0.030 1.240–1.340

aAverage and range of each parameter are shown.


the tank’s bottom. In these assays, organic mat-ter increased and pH falls to 6.6 due to biodegra-dation. Figure 3c evidences the higher BOD5

values at the assay’s end.

Table 4 shows the E. crassipes biomass yieldafter 3 and 6 months. The variation of the bio-mass yield in the case of kraft mill assays duringthe first 3 months of assays is highest for assay

0

10

Con

cent

ratio

n (m

g/l)

pH

Time (month)

200

400

600

800

1.000

1.200

0

200

400

600

800

1.000

1.200

0 1 2 3 4 5 6 0 1 2 3 4 5 60

200

400

600

800

1.000

1.200

0

200

400

600

800

1.000

1.200

0

200

400

600

800

1.000

1.200

0

200

400

600

800

1.000

1.200

4

6

8

10

a')

4

6

8

b')

4

6

8

10

10

4

6

8

10

4

6

8

4

6

8

10

c')c)

b)

a)

Fig. 3. Behavior of COD (�), BOD5(&) and pH (¤) during kraft mill effluent treatment (a) assay A, (b) assay B, and (c)assay C and Poly R-478 (a’) assay A, (b’) assay B, and (c’) assay C by E. crassipes.


C, probably due to the lower concentrations ofBOD5 and phenolic compounds in the kraftmill wastewater.

4. Conclusions

Organic matter removal by E. crassipesranged between 46 and 75%, whereas total phe-nolic compound removal was between 11 and17%. Under experimental conditions, when theassays were performed with 50% and 10%kraft mill effluent, E. crassipes removed around8.5% and 23.6% of color, respectively.

Poly R-478, used as a model compound toevaluate the aromatic behavior in the system,presented similar results in terms of colorremoval with kraft mill effluent. Under experi-mental conditions, E. crassipes biomass wasmissing during the fall season.

Acknowledgements

This work was supported by the FONDECYTGrant 1070509.

References

[1] G. Vidal, M. Belmonte, M. Calderon and S.Chamorro, Environmental advances of the bleachingkraft Chilean mill, Induambiente, 15(2007), 6–30.

[2] R. Konduru, S. Liss and G. Allen, Recalcitrantorganics emerging from biological treatment ofkraft mill effluents, Water Qual. Res. J. Can.,36(2001), 737–757.

[3] G. Vidal, S. Videla and M.C. Diez, Molecularweight distribution of Pinus radiata kraft millwastewater treated by anaerobic digestion, Biore-source Technol., 77(2001), 183–191.

[4] G. Vidal, R. Navia, L. Levet, M.L. Mora and M.C.Diez, Kraft mill anaerobic effluent colourenhancement by a fixed-bed adsorption system,Biotechnol. Letters, 23(2001), 861–865.

[5] J.R. Franta and P.A. Wilderer, Biological treat-ment of paper mill wastewater by sequencingbatch reactor technology to reduce residual organ-ics, Wat. Sci. Tech., 35(1997), 129–136.

[6] G. Thompson, J. Swain, M. Kay and C.F. Forster,The treatment of pulp and paper mill effluent: Areview, Bioresource Technol., 77(2001), 275–286.

[7] M.C. Diez, G. Castillo, L. Aguilar, G. Vidal andM.L. Mora, Operational factors and nutrientseffect on activated sludge treatment for phenoliccompounds degradation from Pinus radiata kraftmill effluent, Bioresource Technol., 83(2002),131–138.

Table 4

E. crassipes biomass yield after 3 and 6 months of assays

AssaysInitial biomass

(kg/m2)Biomass after

3 months (kg/m2)Biomass after

6 months (kg/m2)Total biomassyield (kg/m2)

Kraft mill effluentControl 3.1 3.4 3.4 0.3A 3.4 3.5 3.8 0.4B 3.3 3.1 3.7 0.4C 3.2 3.1 3.8 0.6

Poly R-478Control 0.77 – 0.78 0.01A 0.74 – 0.76 0.02B 0.75 – 0.78 0.03C 0.78 – 0.80 0.02


[8] J. Rintala and S. Lepist€o, Anaerobic treatment ofthermomechanical pulping whitewater at 35–708C, Water Res., 10(1992), 1297–1305.

[9] C.B. Milestone, T.R. Stuthridge and R.R. Fulthorpe,Role of high molecular organic colour formationduring biological treatment of pulp and paperwastewater, Wat. Sci. Tech., 55(2007), 191–198.

[10] V. Dellarossa, J. Cespedes and C. Zaror, Eichhor-nia crassipes-based tertiary treatment of kraft pulpmill effluents in Chilean Central Region, Hydro-biologia, 443(2001), 187–191.

[11] S. Pertuz, J. De La Rotta, N. Jimenez, I. Araujo,L. Herrera, L. Chapın, de Bonilla, A. Del Villar,P. Molleda and G. Toro, Effluent treatment byaquatic hyacinth (In Spanish), Intercencia,24(1999), 120–124.

[12] Y. Xie, D. Yu and B. Ren, Effects of nitrogen andphosphorus availability on the decomposition ofaquatic plants, Aquat. Bot., 80(2004), 29–37.

[13] K.R. Reddy, M. Agami and J.C. Tucker, Influ-ence of phosphorus on growth and nutrient stor-age by water hyacinth (Eichhornia crassipes(Mart.) Solms) plant, Aquat. Bot., 37(1990),355–365.

[14] K.S. Low, C.K. Lee and K.K. Tan, Biosorption ofbasic dyes by water hyacinth roots, BioresourceTechnol., 52(1995), 79–83.

[15] B.C. Wolverton and M.M. McKown, Water hya-cinths for removal of phenols from pollutedwaters, Aquat. Bot., 2(1976), 191–201.

[16] Y. Nor, Phenol removal by Eichhornia crassipes inthe presence of trace metals, Water Res., 28(1994),1161–1166.

[17] APHA-AWWA-WPCF. Standard Methods forExamination of Water and Wastewater, 16th edn.Washington, 1985.

[18] F. Cecen, The use of UV-VIS measurements in thedetermination of biological treatability of pulpbleaching effluents; 7th International Water Asso-ciation Symposium on Forest Industry Waste-waters, Seattle, USA, 2003.

[19] S. Chamorro, C. Xavier and G. Vidal, Behaviourof aromatic compounds contained in the kraftmill effluents measurements by UV-VIS, Biotech-nol. Progr., 21(2005), 1567–1571.

[20] M.L. Casablanca, Large-scale production of Eich-hornia crassipes on paper industry effluent, Biore-source Technol., 54(1995), 35–39.


Eichhornia crassipes used as tertiary color removal treatment for Kraft mill effluent

Documents