- Email: [email protected]
A high rate ponding unit operation linking treatment of tannery effluent and Arthrospira (Spirulina) biomass production. 1: Process development
b i o m a s s a n d b i o e n e r g y 5 1 ( 2 0 1 3 ) 1 8 3 e1 8 8
Available online at www.sciencedirect.com
http://www.elsevier.com/locate/biombioe
A high rate ponding unit operation linking treatment of tannery effluent and Arthrospira (Spirulina) biomass production. 1: Process development Peter Rose*, Kevin Dunn 1 Department of Biochemistry, Microbiology and Biotechnology, Rhodes University, PO Box 480, Grahamstown 6140, South Africa
article info
abstract
Article history:
Effluent from the production of wet blue leather has been shown to support the substantial
Received 9 January 2012
growth of Arthrospira biomass in tannery waste stabilization ponds, which is of interest in
Received in revised form
its use both in animal feeds and biofuels production. Here we report process development
4 January 2013
investigations which were undertaken in photobioreactor and outdoor high rate pond pilot
Accepted 7 January 2013
studies. Biomass productivities of 16 t ha1 yr1 (dry mass) were measured which compares
Available online 14 February 2013
broadly with yields reported for Arthrospira cultivated in other complex media. The specific growth rate m ¼ 0.05 d1 was somewhat lower than reports for the mixotrophic cultivation
Keywords:
of Arthrospira in defined media studies. This system operates under ammonia control and
Spirulina
may be relieved by recirculation of alkaline waters which accumulate in these waste sta-
Biofuels
bilization ponds. A substantial difference in total nitrogen and phosphorus removal be-
Animal feed
tween the experimental and theoretical yields due to biological activity suggests stripping
Tannery
may account for the largest fraction of ammonia removal, and precipitation for phosphate
Wastewater
removal, in this operation. Heavy metal contamination may be a problem with biomass
Waste stabilization ponds
production in industrial effluents and pretreatment of the tannery effluent in a primary facultative pond was shown to substantially reduce the heavy metal load. Process kinetic values were derived and have been used for the design and construction of a full-scale Arthrospira production operation using tannery effluent growth media, which is reported in a follow-up study. ª 2013 Elsevier Ltd. All rights reserved.
1.
Introduction
The resurgence of interest in microalgal biofuels development [1,2], has focused attention on the use of wastewaters as a means of reducing the cost of growth media formulation [3]. This remains one of the major production inputs determining the profitability of the biofuels enterprise [4], and the linkage of waste treatment and microalgal biomass production provides an opportunity to achieve reductions in both operational and capital costs [5].
Waste stabilization ponds (WSP) have been widely used in the treatment of domestic and industrial wastewaters [6] and their application in tannery effluent treatment, as zerodischarge systems, provides one of few treatment options available for leather production in highly water stressed areas [7,8]. While the appearance of massive near mono-species blooms of Arthrospira (Spirulina) have been described in these systems [9], little has been reported on factors regulating this growth in tannery effluents and these blooms remain unpredictable and unreliable in their appearance. The role of
* Corresponding author. Tel.: þ27 828011353; fax: þ27 466228627. E-mail address: [email protected] (P. Rose). 1 Current address: Orangeriestrasse 3, 67071 Ludwigshafen am Rein, Germany. 0961-9534/$ e see front matter ª 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.biombioe.2013.01.025
184
b i o m a s s a n d b i o e n e r g y 5 1 ( 2 0 1 3 ) 1 8 3 e1 8 8
microalgae in the successful operation of WSP has been welldescribed [10,11] and controlled use of Arthrospira growth could be of importance in the management of the tannery WSP, and particularly in the control of odour nuisance, to which they are prone [12]. The current investigation follows a previous evaluation of the potential use of tannery effluents as production media for the mass culture of Arthrospira biomass [13]. Here it was shown that in effluent from a wet-blue tanning operation, Arthrospira growth is under the direct control of the ammonia concentration in the growth medium. This indicated that an effective mass culture strategy in this medium would require the determination of a maximum effluent loading rate that operates as a function of the optimized ammonia removal rate. Determination of these kinetics under controlled conditions could provide a basis for the development of a full scale unit operation linking the treatment of tannery effluent and Arthrospira biomass production. This potential has been investigated in photobioreactor and outdoor pilot plant studies and is reported here.
2.
Materials and methods
The site for this study was a tannery WSP located in the Western Cape Province, South Africa, and received about 460 m3 day1 effluent from the processing of 1500 hides to wet blue leather daily. Combined tannery wastewaters passed through physico-chemical pre-treatment, following process stream segregation and including sulfide oxidation, balancing and aeration and flocculent-assisted sedimentation with removal of solids, before discharge to the WSP cascade. The medium used in the study was drawn at this point and is referred to hereafter as tannery effluent. The Arthrospira strain used in the study was sourced from the tannery WSP. It was provisionally identified as Arthrospira platensis and was previously shown to grow mixotrophically in this medium [13]. A 5L New Brunswick Bioflo 111 fermenter was configured as a photobioreactor with an array of 50 cm cool white fluorescent tubes arranged around its circumference. A 10% inoculum of the tannery WSP Arthrospira isolate was used for reactor startup in a 5% tap water dilution of tannery effluent (Table 1). Once photosynthetically generated dissolved oxygen (DO) reached 7 mg L1, feeding of the reactor commenced. The reactor was operated at 25 C under 12 h dark/light cycle with illumination at 158 mmol m2 s1, and in batch-fed mode with the addition of undiluted effluent at the equivalent of 3% of reactor volume daily. After the establishment of a stable system, the continuous-feed operation commenced at increasing rates of tannery effluent addition. Steady state was assumed for each loading rate after at least three replacements of reactor volume. Temperature, DO and pH were computer logged at 15 min intervals. A pond water alkalinity recirculation strategy was evaluated in flask studies. Tannery effluent in 1L Ehrlenmeyer glass flasks was inoculated with Arthrospira culture sourced from the steady state photobioreactor, fed at the 5% loading rate. Alkaline recirculation water, sourced from the terminal pond in the study site WSP, was added to triplicate flasks at 5%, 15% and 25% by volume. Photosynthetic carbon fixation was measured and results reflect a mean of triplicates.
Table 1 e Analysis of the tannery effluent used in this study (standard deviation in brackets). (mg L1) Chemical oxygen demand Ammonia as NH3 Phosphate as PO4 Calcium Chloride Sodium Sulphate Sulfide Total alkalinity (as CaCO3) Dissolved oxygen Salinity (gL1) pH
2474(1810) 731(98) 19(12.5) 226(11) 4048(202) 3090(198) 364(43) 1192(112) 525(49) 0.01(0.01) 10(0.039) 8.2(0.2)
Analyses of COD, nitrate, ammonia and phosphate followed [14]. Salinity was measured with an Atago refractometer. Chlorophyll a was measured following the method of Lichtenthaler [15]. Measurements of photosynthetic productivity used the [14C]-sodium bicarbonate CO2 fixation method modified by Oren [16]and sample measurement was undertaken in a Beckman LS3150T scintillation counter. Heavy metal and nutritional analyses were undertaken by the Animal and Poultry Science Laboratory, University of Kwa-Zulu Natal. Kinetic values for the process were derived in a continuous-feed operation at steady state. The specific growth rate m (d1) was derived from the dilution rate D (d1) with growth at steady state for each feed rate. Volumetric biomass productivity QX (mg L1 d1) was calculated as the concentration of biomass removed from the reactor as a function of time. The volumetric removal rates for total nitrogen (as N) QNT (mg L1 d1) and phosphate (as P) QP (mg L1 d1) were calculated as the difference between the feed and the reactor concentration as a function of the daily loading rate. Experimental removal yields in the reactor for nitrogen and phosphate, Y1NT and Y1P (dimensionless) were calculated as the ratio of amounts removed to those present in the feed. Theoretical removal ascribed to biomass uptake Y2NT and Y2P (dimensionless) was calculated by material balances from the data of cell mass production using an average dry biomass composition for A. platensis of C1.650 O0.531 N0.170 S 0.007 P 0.006 [17]. Total nitrogen removal was calculated according to Nrem ¼ NIN NH3 (NH3 þ NO3)SOL and assuming no significant further ammonification and zero level nitrate in the influent.
3.
Results and discussion
The process development investigation of a mixotrophic Arthrospira HRP unit operation in the treatment of tannery effluent was based on the results of preliminary flask studies [13], and was undertaken through scale-up from photobioreactor to pilot plant studies which are described here.
3.1.
Photobioreactor
In the start-up phase the photobioreactor was operated in batch-fed mode, COD levels were reduced from 2474 mg L1 to around 250 mg L1, DO remained well above saturation levels
185
b i o m a s s a n d b i o e n e r g y 5 1 ( 2 0 1 3 ) 1 8 3 e1 8 8
NH3
COD , NH3 (mg L-1)
3000
COD
PO4
NO3
50 45 40 35 30 25 20 15 10 5 0
2500 2000 1500 1000 500 0 Effluent
3%
5%
NO3, PO4 (mg L-1)
(12 mg L1) and ammonia stabilised below 60 mg L1, which had previously been shown to be a toxic threshold level in this system [13]. The reactor was maintained in this mode for several weeks of stable operation before it was changed to a continuous feed regime. In the continuous mode the reactor was fed successively with undiluted tannery effluent (pH 8.2) at loading rates of 3%, 5% and 8% of reactor volume daily (Fig. 1). Steady state was assumed after at least three replacements of reactor volume, with the exception of the 8% loading regime under which steady state could not be established, ammonia levels rose to 135 mg L1, Chl a levels fell below 1 mg L1, the culture became chlorotic and finally collapsed. The oxygen regeneration time was measured at each of the various feed rates and following sparging with nitrogen to DO < 1 mg L1. Recovery of DO was found to be a sensitive indicator of the ability of the microalgal/bacterial co-culture to cope with a specific effluent loading rate. Oxygen regeneration times of 20 h, 21 h and 40 h were measured for the 3%, 5% and 8% loadings respectively. The 5% loading proved the highest at which stable operation could be maintained and here COD was reduced by 86%, ammonia by 92% and phosphate was completely removed, possibly mainly by precipitation as calcium phosphate given the substantial difference between the experimental (Y1P) and theoretical (Y2P) yields (Table 2). Nitrogen removal in this system could be attributed to a number of mechanisms, possibly occurring simultaneously, and including ammonia stripping, nitrification and direct uptake by the culture itself. Given the large difference between the experimental (Y1NT) and theoretical (Y2NT) total nitrogen removal yields, it was assumed that ammonia stripping accounted for the largest proportion of nitrogen removal in the system. While the specific growth rate m ¼ 0.05 d1 was lower than m ¼ 0.096 d1 reported for a mixotrophic A. platensis culture with glucose as the organic substrate [18], it is closer to m ¼ 0.068 d1 recorded for a propionate carbon source in the same study, and is consistent with growth in the complex protein/amino acid-enriched tannery effluent medium. Volumetric cell growth (QX) in the tannery effluent medium of 31.3 mg L1 d1 was also lower than the 60 mg L1 d1 reported in the Lodi et al. study [18] for growth on propionate.
8%
Effluent loading as % reactor volume
Fig. 1 e Photobioreactor study of a continuously fed Arthrospira culture at steady state with various loadings of tannery effluent as daily percentages of reactor volume and showing changes in COD, ammonia, nitrate and phosphate.
Table 2 e Kinetic parameters for a steady state culture of Arthrospira in photobioreactor culture, continuously fed tannery effluent at various rates as a percentage of reactor volume and grown at 25 C in a 12 h dark/light cycle (standard deviation in brackets). Parametera Growth m (d1) QX (mg L1 d1) Nitrogen removal QNT (mg L1 d1) Y1NT () Y2NT () Phosphate removal QP (mg L1 d1) Y1P () Y2P ()
3%
5%
8%
0.03(0.006) 5.19(0.3)
0.05(0.007) 8.65(0.35)
e 7.25(0.3)
16.90(1.02) 0.938(0.06) 0.023(0.002)
15.20(1.00) 0.905(0.05) 0.043(0.002)
13.30(0.98) 0.804(0.05) 0.042(0.002)
0.54(0.045) 0.95(0.07) 0.057(0.002)
0.90(0.078) 0.95(0.07) 0.057(0.002)
1.44(0.25) 0.95(0.07) 0.03(0.002)
a Parameters defined in materials and methods.
3.2.
Pilot plant
The implementation of the above findings at larger scale was undertaken at the tannery study site where two 80 m2 paddle wheel operated High Rate Ponds (HRP) were constructed (volume 12 m3 each) adjacent to the WSP [19]. The start-up regime for the pilot ponds followed that used in the photobioreactor study with a 3% dilution of tannery effluent inoculated with 10% by volume of Arthrospira culture sourced from the WSP. The ponds were batch-fed with a loading rate at 3% of pond volume daily until stable operating conditions had been established. Thereafter the pilot ponds were continuously fed and loaded successively at 5%, 8% and 10% of total volume with undiluted tannery effluent. This regime was repeated for both summer and winter seasons. Environmental conditions were less easily controlled than in the photobioreactor and results here may be considered to reflect quasisteady state operation. Data were sourced as single measurements after operation for at least 30 days at each loading rate. Once again the 5% loading was found to be the highest rate providing stable HRP operation (ammonia 58 mg L1). Ammonia levels in the 8% and 10% loadings rose to 129 mg L1 and 199 mg L1 respectively with the cultures shifting into stationary phase, finally becoming chlorotic and crashing (Fig. 2). In the 5% loading, COD was reduced by 85% and ammonia by 93%. The pH remained constant at 8.2e8.4 at all loading rates. A notable feature was the substantially higher biomass production of 4.46 g m2 d1 (dry mass) measured in the outdoor ponds compared to the photobioreactor study giving an Arthrospira biomass yield of 16.29 t ha1 yr1 measured for the pilot pond operation. Photosynthetically available radiation (PAR) ranged between 500 and 1300 mmol m2 s1 from winter to summer conditions at the site. A photosynthetic carbon fixation rate of 7.022 g m2 d1 was measured as an average for summer and winter conditions. These results compare with 7.3 g m2 d1e9.5 g m2 d1 (dry mass) [20] and 8 g m2 d1e12 g m2 d1 carbon fixation reported for Arthrospira grown in sewage wastewaters [21]. Biomass yields reported in commercial systems range widely between 10 and 30 t ha1 yr1 [22]. The difference between biomass yield and
186
b i o m a s s a n d b i o e n e r g y 5 1 ( 2 0 1 3 ) 1 8 3 e1 8 8
carbon fixation values recorded here indicates the measurement of the latter under idealized conditions that do not effectively average light incidence throughout the water column in the reactor. Light extinction due to opacity of the medium and changing cell concentration will vary across the depth of the reactor with actual biomass yield providing a more accurate indication of the biomass production potential for the system. This will also account for the lower biomass yields observed for tannery effluent compared to literature values for sewage and defined media systems. The kinetic values derived from the pilot pond study are reported in Table 3 and show a substantially increased QX with production around fourfold higher than in the photobioreactor. These results were also in broad agreement with the observation that ammonia stripping probably accounts for the largest fraction of nitrogen removal in the system. Control of ammonia toxicity by manipulation of the effluent loading rate will be an important factor in further scale-up of the system.
Table 3 e Kinetic parameters for the quasi-steady state outdoor culture of Arthrospira in the pilot plant HRP study continuously fed tannery effluent at various loading rates as a percentage of reactor volume and grown under a natural daylight regime. Data were sourced as single measurements after operation for at least 30 days at each loading rate. Parametera Growth m (d1) QX (mg L1 d1) Nitrogen removal QNT (mg L1 d1) Y1NT () Y2NT () Phosphate removal QP (mg L1 d1) Y1P () Y2P ()
3%
5%
8%
10%
0.03 21.90
0.05 31.3
0.025 18.70
0.02 17.3
17.30 0.96 0.097
27.2 0.91 0.088
39.40 0.82 0.036
42.80 0.72 0.031
0.54 0.95 0.242
0.90 0.95 0.208
1.44 0.95 0.078
1.80 0.95 0.057
a Parameters defined in materials and methods.
NH3-1
NH3-2 1200 1000 800 600 400 200
Effluent
Fig. 2 e Pilot plant study in which twin 80 mL2 high rate algal ponds inoculated with an Arthrospira culture were fed continuously at various loadings of tannery effluent as percentages of reactor volume, and showing changes in COD, ammonia, biomass and carbon fixation rates (annualized averages).
C fix2
NH3 (mg L-1)
C fixation (mg m -2 d -1)
C fix1 9000 8000 7000 6000 5000 4000 3000 2000 1000 0
5%
15%
25%
0
Effluent loading as % reactor volume
Fig. 3 e Flask study simulation of the effect of WSP terminal pond alkalinity recirculation on the photosynthetic productivity of an Arthrospira culture comparing C fix 1 (without recycle) and C fix 2 (with recycle) at various effluent loading rates. Ammonia levels in the study compare NH3 -1 (without recycle) and NH3 -2 (with recycle).
3.3.
Recirculation
A number of strategies have been proposed to deal with ammonia toxicity in microalgal ponding systems including the manipulation of loading rate [23], aeration stripping [24], addition of bicarbonate [25] and recirculation of oxygenated microalgae-enriched waters [26]. All of these may be applied in one degree or another in tannery WSP with terminal ponds in zero-discharge systems containing concentrated alkaline solutions of salts, including carbonates from photosynthetic activity, microbial biomass and residual slowly degradable organic compounds. Given the complexity of the solution, an empirical assessment of the recirculation alkalization option was investigated in flask studies first. Pipework constraints at the WSP study site had precluded the pumping of terminal pond water and thus the evaluation of this option during the pilot plant investigation. Flask studies were set up as described and compared the effect on Arthrospira production with increasing terminal pond recirculation loadings of 5%, 15%, and 25% pond water (carbonate alkalinity 2585 mg L1 as CaCO3). pH was elevated to 9.5 with the addition of the pond alkalinity, a substantial increase (w50%) in photosynthetic carbon fixation was observed and also a reduction in ammonia levels (Fig. 3).
Table 4 e Tannery effluent treatment performance in the reconfigured primary facultative pond (standard deviation in brackets).
COD (mg L1) Ammonia (mg L1) Sulphide (mg L1) Phosphate (mg L1) Suspended solids (mg L1)
Tannery effluent
After PFP
% Change
2474(1810) 731(98) 285(422) 19(12.5) 243(196)
1216(93) 452(105) 76(16) 1.7(2) 0.5(0.5)
50.8% 38.2% 73.3% 91% 99.8%
187
b i o m a s s a n d b i o e n e r g y 5 1 ( 2 0 1 3 ) 1 8 3 e1 8 8
Table 5 e Heavy metal concentrations in tannery effluent before and after treatment in the anaerobic compartment of a facultative pond unit located prior to the high rate pond, and also in Arthrospira biomass grown before and after the anaerobic treatment step. Feed standard as set out in Ref. [28] and subsequently consolidated in Ref. [29].
Chromium Cadmium Cobalt Iron Lead Nickel Zinc
Untreated effluent mg L1
Treated effluent mg L1
Biomass before treat mg kg1
Biomass after treat mg kg1
Feed standard mg kg1
4.68(0.1) 0.08(0.02) 0.35(0.03) 39.9(2.1) 0.76(0.03) 0.41(0.01) 1.51(0.1)
0.25(0.02) 0.02(0.01) 0.18(0.01) 0.41(0.03) 0.11(0.01) 0.21(0.01) <0.1
25.8(0.5) 5.96(0.23) 22.4(0.6) 2012(12.2) 219(2.3) 49.2(1.5) 218.5(2.8)
<1 <1 3.3(0.2) 795(13.4) 2.3(0.2) 17.5(0.8) 22.5(1.1)
10e15 1.0 4e6 1250e2500 10 10e15 250e500
Although ammonia levels at loading rates above 5% were higher than the 60e80 mg L1 toxic threshold, this was not found to be inhibitory up to the 15% loading, where DO levels remained well above saturation. These results suggest a number of mechanisms of ammonia detoxification including conversion of ammonia to a non-toxic ionized form at elevated pH 9.8 (pKa 9.25), removal by stripping and nitrification and possibly a small amount of direct uptake by the Arthrospira culture itself. The kinetic values derived in both the photobioreactor and pilot pond study support this conclusion (Table 2 and Table 3). The viability of a terminal pond recirculation strategy at the test site would have important process design implications by taking into account higher growth rates, higher effluent loading rates and therefore also a reduced HRP reactor size requirement. Based on these results a maximum recirculation rate of 0.2:1 was proposed as a design value which could be adjusted empirically during full-scale implementation.
3.4.
Primary facultative pond
The Advanced Integrated Algal Wastewater Ponding System (AIWPS) developed by Oswald [19] makes provision for a primary facultative pond (PFP) unit ahead of the HRP in the treatment of wastewaters. A pond in the study site tannery WSP system was retrofitted to operate as a PFP. The feed was relocated to the base of the pond providing an upflow anaerobic sludge zone, and shallow draught mechanical aerators were fitted to the pond surface, among other reasons, to control odour emissions. Methane gas production commenced soon after start-up and Table 4 reports the outcome of this intervention. COD was reduced by 50% and ammonia by 38% (assuming near complete ammonification, stripping and partial oxidation in the aerobic upper layer). Sulphate reduction was complete in the anaerobic zone (sulphate was reduced from 975 mg L1 to <1 mg L1), and partial re-oxidation of sulfide (73%) occurred in the aerobic zone. Suspended solids were nearly completely removed. The low residual phosphate levels could present a limitation for subsequent Arthrospira growth in the HRP.
3.5.
will be of importance, especially where a feed grade product is targeted [27]. An analysis of heavy metal levels in Arthrospira biomass harvested from the WSP at the study site is reported in Table 5 [9]. Given high levels of sulfates and sulfides in the PFP noted above, its use as a pretreatment step for the removal of contaminating heavy metals was also investigated. Table 5 reports the reduction in effluent heavy metal concentration during this operation, and also the changes in Arthrospira biomass levels grown before and after treatment of the effluent in this unit. The WSP Arthrospira strain was found to be unaffected by the high concentrations of sulfide (1100 mg L1) from the anaerobic pond unit, which were rapidly oxidized back to sulphate in the HRP (<1 mg L1). Chromium in all samples analysed over a three-year period was in the Chromium III form. Chromium VI was not found in the system.
4.
Conclusions
1. The scale-up studies have demonstrated technical viability in the use of tannery effluent as a growth medium for Arthrospira production. 2. Arthrospira growth yields around 16 t ha1 yr1 were within the range reported for open pond commercial production systems. 3. Control of ammonia toxicity will determine loading rates and therefore the HRP pond size required. For successful operation, the loading rate will need to be directly coupled to the ammonia removal rate. 4. Accumulated alkalinity in tannery WSP may be recirculated to the HRP to raise the pH and shift ammonia to the ionized form and thereby increase the effluent loading rate with important implications for pond sizing and costs. 5. Heavy metals contamination of the Arthrospira biomass may be a problem in these wastewaters but may be substantially reduced in an anaerobic pretreatment step. 6. The outcomes of this study provide design values for the fullscale implementation of the process for Arthrospira production in tannery wastewaters. Process scale-up has been undertaken and the design, construction and operation of the high rate ponding operation at industrial scale has been reported in the Part 2 follow-up to this study.
Heavy metals removal
Given the wide range from which hides are sourced, and a diversity of commercial chemicals used in the tanning operation, these wastewaters may contain heavy metals. Microalgal biomass is at risk of accumulating a proportion of these which
Acknowledgements The authors acknowledge the support of Messrs. RE Newson and R Smith formerly of Mossop-Western Leathers Co. Pty.
188
b i o m a s s a n d b i o e n e r g y 5 1 ( 2 0 1 3 ) 1 8 3 e1 8 8
Ltd., Rhodes University and the Water Research Commission for financial support and permission to use data from WRC Report TT188/02.
references
[1] Chisti Y. Biodiesel from microalgae. Biotechnol Adv 2007;25: 294e306. [2] Greenwell HC, Laurens LML, Shields RJ, Lovitt RW, Flynn KJ. Placing microalgae on the biofuels priority list: a review of the technological challenges. J R Soc Interface 2010;7:703e26. [3] Christensen L, Sims R. Production and harvesting of microalgae for wastewater treatment, biofuels, and bioproducts. Biotechnol Adv 2011;29(6):686e702. [4] Harun M, Singh M, Forde GM, Danquah MK. Bioprocess engineering of microalgae to produce a variety of consumer products. Renew Sust Energ Rev 2010;14:1037e47. [5] Su Y, Mennerich A, Urban B. Municipal wastewater treatment and biomass accumulation with a wastewaterborn and settleable algal-bacterial culture. Water Res 2011; 45:3351e8. [6] von Sperling M. Waste stabilisation ponds. London: IWA Publishing; 2007. [7] Rowswell RA, Cooper DA, Shuttleworth SG. Evaporation ponds: a solution for tannery effluent disposal. J Soc Leath Tech Ch 1984;69:123e9. [8] Tadesse I, Greene FB, Puhakka JA. The role of sulphidogenesis in anaerobic treatment phase of tannery wastewater treatment in advanced integrated wastewater pond system. Biodegradation 2003;14:219e27. [9] Maart BA. The biotechnology of effluent grown Spirulina and applications in aquaculture nutrition. Grahamstown: Rhodes University MSc Thesis 1993. [10] Mara DD. Domestic wastewater treatment in developing countries. London UK: Earthscan Publications; 2004. [11] Munoz R, Guieysse B. Algal bacterial processes for the treatment of hazardous contaminants: a review. Water Res 2006;40(15):2799e815. [12] Rose PD, Maart BA, Dunn KM, Rowswell RA, Brits P. High rate algal oxidation ponding for the treatment of tannery effluents. Water Sci Technol 1996;33:219e27. [13] Dunn KM. The biotechnology of high rate algal ponding systems in the treatment of saline tannery wastewaters. Grahamstown; Rhodes University PhD Thesis 1998. [14] APHA. In: Clesceri LS, editor. Standard methods for the examination of water and wastewater. 17th ed. Washington, DC: American Public Health Association; 1989.
[15] Lichtenthaler HK. Chlorophylls and carotenoids: pigments of photosynthetic biomembranes. Method Enzymol 1987;148: 350e71. [16] Oren A. Bacterial activities in the Dead Sea: 1980e1991 a survival at the upper limits of salinity. Int J Salt Lake Res 1992;1:7e20. [17] Cornet JF, Dussap CG, Cluzel P, Dubertret G. A structured model for simulation of cultures of the cyanobacterium Spirulina platensis in photobioreactors: II. Identification of kinetic parameters under light and mineral limitations. Biotechnol Bioeng 1992;40:826e34. [18] Lodi A, Binaghi L, De Faveri D, Carvalho JCM, Converti A. Fedbatch mixotrophic cultivation of Arthrospira (Spirulina) platensis (Cyanophycea) with carbon source pulse feeding. Anal Microbiol 2005;55(3):181e5. [19] Oswald WJ. Large-scale algal systems (engineering aspects). In: Borowitzka MA, editor. . In: Borowitzka LJ, editor. Microalgal biotechnology. Cambridge UK: Cambridge University Press; 1988. [20] Saxena PN, Ahmad MR, Shyam R, Amla DV. Cultivation of Spirulina in sewage for poultry feed. Experientia 1983;39: 1077e83. [21] Fox R. Nutrient preparation and low cost basin construction for village production of Spirulina. In: Stadler T, editor. Algal biotechnology. Amsterdam: Elsevier; 1988. [22] Jime´neza C, Cossı´ob BR, Niella FX. Relationship between physicochemical variables and productivity in open ponds for the production of Spirulina: a predictive model of algal yield. Aquaculture 2003;221:331e45. [23] Abeliovitch A. The effects of unbalanced ammonia and BOD concentrations on oxidation ponds. Water Res 1983;17(3): 299e301. [24] Pano A, Middlebrooks EJ. Ammonia nitrogen removal in facultative wastewater stabiliztion ponds. J Water Control Fed 1982;54:344e51. [25] Chaudari PR, Krishnamoorthi KP, Vittal Rao M. Growth potential of Spirulina, a blue green alga in sewage. Indian Acad Sci 1980;89(3). [26] Pescod MB. The role and limitations of pond systems. Wat Sci Tech 1996;33(7):11e22. [27] Yi-Chao Lee Y-C, Chang S-P. The biosorption of heavy metals from aqueous solution by Spirogyra and Cladophora filamentous macroalgae. Bioresour Technol 2011;102: 5297e304. [28] German Ministry of Agriculture Law Collection No 117; 1987. [29] European Union. Regulation (EC) No 882/2004 of the European parliament and of the Council of 7 May 2002 on undesirable substances in animal feed.
Available online at www.sciencedirect.com
http://www.elsevier.com/locate/biombioe
A high rate ponding unit operation linking treatment of tannery effluent and Arthrospira (Spirulina) biomass production. 1: Process development Peter Rose*, Kevin Dunn 1 Department of Biochemistry, Microbiology and Biotechnology, Rhodes University, PO Box 480, Grahamstown 6140, South Africa
article info
abstract
Article history:
Effluent from the production of wet blue leather has been shown to support the substantial
Received 9 January 2012
growth of Arthrospira biomass in tannery waste stabilization ponds, which is of interest in
Received in revised form
its use both in animal feeds and biofuels production. Here we report process development
4 January 2013
investigations which were undertaken in photobioreactor and outdoor high rate pond pilot
Accepted 7 January 2013
studies. Biomass productivities of 16 t ha1 yr1 (dry mass) were measured which compares
Available online 14 February 2013
broadly with yields reported for Arthrospira cultivated in other complex media. The specific growth rate m ¼ 0.05 d1 was somewhat lower than reports for the mixotrophic cultivation
Keywords:
of Arthrospira in defined media studies. This system operates under ammonia control and
Spirulina
may be relieved by recirculation of alkaline waters which accumulate in these waste sta-
Biofuels
bilization ponds. A substantial difference in total nitrogen and phosphorus removal be-
Animal feed
tween the experimental and theoretical yields due to biological activity suggests stripping
Tannery
may account for the largest fraction of ammonia removal, and precipitation for phosphate
Wastewater
removal, in this operation. Heavy metal contamination may be a problem with biomass
Waste stabilization ponds
production in industrial effluents and pretreatment of the tannery effluent in a primary facultative pond was shown to substantially reduce the heavy metal load. Process kinetic values were derived and have been used for the design and construction of a full-scale Arthrospira production operation using tannery effluent growth media, which is reported in a follow-up study. ª 2013 Elsevier Ltd. All rights reserved.
1.
Introduction
The resurgence of interest in microalgal biofuels development [1,2], has focused attention on the use of wastewaters as a means of reducing the cost of growth media formulation [3]. This remains one of the major production inputs determining the profitability of the biofuels enterprise [4], and the linkage of waste treatment and microalgal biomass production provides an opportunity to achieve reductions in both operational and capital costs [5].
Waste stabilization ponds (WSP) have been widely used in the treatment of domestic and industrial wastewaters [6] and their application in tannery effluent treatment, as zerodischarge systems, provides one of few treatment options available for leather production in highly water stressed areas [7,8]. While the appearance of massive near mono-species blooms of Arthrospira (Spirulina) have been described in these systems [9], little has been reported on factors regulating this growth in tannery effluents and these blooms remain unpredictable and unreliable in their appearance. The role of
* Corresponding author. Tel.: þ27 828011353; fax: þ27 466228627. E-mail address: [email protected] (P. Rose). 1 Current address: Orangeriestrasse 3, 67071 Ludwigshafen am Rein, Germany. 0961-9534/$ e see front matter ª 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.biombioe.2013.01.025
184
b i o m a s s a n d b i o e n e r g y 5 1 ( 2 0 1 3 ) 1 8 3 e1 8 8
microalgae in the successful operation of WSP has been welldescribed [10,11] and controlled use of Arthrospira growth could be of importance in the management of the tannery WSP, and particularly in the control of odour nuisance, to which they are prone [12]. The current investigation follows a previous evaluation of the potential use of tannery effluents as production media for the mass culture of Arthrospira biomass [13]. Here it was shown that in effluent from a wet-blue tanning operation, Arthrospira growth is under the direct control of the ammonia concentration in the growth medium. This indicated that an effective mass culture strategy in this medium would require the determination of a maximum effluent loading rate that operates as a function of the optimized ammonia removal rate. Determination of these kinetics under controlled conditions could provide a basis for the development of a full scale unit operation linking the treatment of tannery effluent and Arthrospira biomass production. This potential has been investigated in photobioreactor and outdoor pilot plant studies and is reported here.
2.
Materials and methods
The site for this study was a tannery WSP located in the Western Cape Province, South Africa, and received about 460 m3 day1 effluent from the processing of 1500 hides to wet blue leather daily. Combined tannery wastewaters passed through physico-chemical pre-treatment, following process stream segregation and including sulfide oxidation, balancing and aeration and flocculent-assisted sedimentation with removal of solids, before discharge to the WSP cascade. The medium used in the study was drawn at this point and is referred to hereafter as tannery effluent. The Arthrospira strain used in the study was sourced from the tannery WSP. It was provisionally identified as Arthrospira platensis and was previously shown to grow mixotrophically in this medium [13]. A 5L New Brunswick Bioflo 111 fermenter was configured as a photobioreactor with an array of 50 cm cool white fluorescent tubes arranged around its circumference. A 10% inoculum of the tannery WSP Arthrospira isolate was used for reactor startup in a 5% tap water dilution of tannery effluent (Table 1). Once photosynthetically generated dissolved oxygen (DO) reached 7 mg L1, feeding of the reactor commenced. The reactor was operated at 25 C under 12 h dark/light cycle with illumination at 158 mmol m2 s1, and in batch-fed mode with the addition of undiluted effluent at the equivalent of 3% of reactor volume daily. After the establishment of a stable system, the continuous-feed operation commenced at increasing rates of tannery effluent addition. Steady state was assumed for each loading rate after at least three replacements of reactor volume. Temperature, DO and pH were computer logged at 15 min intervals. A pond water alkalinity recirculation strategy was evaluated in flask studies. Tannery effluent in 1L Ehrlenmeyer glass flasks was inoculated with Arthrospira culture sourced from the steady state photobioreactor, fed at the 5% loading rate. Alkaline recirculation water, sourced from the terminal pond in the study site WSP, was added to triplicate flasks at 5%, 15% and 25% by volume. Photosynthetic carbon fixation was measured and results reflect a mean of triplicates.
Table 1 e Analysis of the tannery effluent used in this study (standard deviation in brackets). (mg L1) Chemical oxygen demand Ammonia as NH3 Phosphate as PO4 Calcium Chloride Sodium Sulphate Sulfide Total alkalinity (as CaCO3) Dissolved oxygen Salinity (gL1) pH
2474(1810) 731(98) 19(12.5) 226(11) 4048(202) 3090(198) 364(43) 1192(112) 525(49) 0.01(0.01) 10(0.039) 8.2(0.2)
Analyses of COD, nitrate, ammonia and phosphate followed [14]. Salinity was measured with an Atago refractometer. Chlorophyll a was measured following the method of Lichtenthaler [15]. Measurements of photosynthetic productivity used the [14C]-sodium bicarbonate CO2 fixation method modified by Oren [16]and sample measurement was undertaken in a Beckman LS3150T scintillation counter. Heavy metal and nutritional analyses were undertaken by the Animal and Poultry Science Laboratory, University of Kwa-Zulu Natal. Kinetic values for the process were derived in a continuous-feed operation at steady state. The specific growth rate m (d1) was derived from the dilution rate D (d1) with growth at steady state for each feed rate. Volumetric biomass productivity QX (mg L1 d1) was calculated as the concentration of biomass removed from the reactor as a function of time. The volumetric removal rates for total nitrogen (as N) QNT (mg L1 d1) and phosphate (as P) QP (mg L1 d1) were calculated as the difference between the feed and the reactor concentration as a function of the daily loading rate. Experimental removal yields in the reactor for nitrogen and phosphate, Y1NT and Y1P (dimensionless) were calculated as the ratio of amounts removed to those present in the feed. Theoretical removal ascribed to biomass uptake Y2NT and Y2P (dimensionless) was calculated by material balances from the data of cell mass production using an average dry biomass composition for A. platensis of C1.650 O0.531 N0.170 S 0.007 P 0.006 [17]. Total nitrogen removal was calculated according to Nrem ¼ NIN NH3 (NH3 þ NO3)SOL and assuming no significant further ammonification and zero level nitrate in the influent.
3.
Results and discussion
The process development investigation of a mixotrophic Arthrospira HRP unit operation in the treatment of tannery effluent was based on the results of preliminary flask studies [13], and was undertaken through scale-up from photobioreactor to pilot plant studies which are described here.
3.1.
Photobioreactor
In the start-up phase the photobioreactor was operated in batch-fed mode, COD levels were reduced from 2474 mg L1 to around 250 mg L1, DO remained well above saturation levels
185
b i o m a s s a n d b i o e n e r g y 5 1 ( 2 0 1 3 ) 1 8 3 e1 8 8
NH3
COD , NH3 (mg L-1)
3000
COD
PO4
NO3
50 45 40 35 30 25 20 15 10 5 0
2500 2000 1500 1000 500 0 Effluent
3%
5%
NO3, PO4 (mg L-1)
(12 mg L1) and ammonia stabilised below 60 mg L1, which had previously been shown to be a toxic threshold level in this system [13]. The reactor was maintained in this mode for several weeks of stable operation before it was changed to a continuous feed regime. In the continuous mode the reactor was fed successively with undiluted tannery effluent (pH 8.2) at loading rates of 3%, 5% and 8% of reactor volume daily (Fig. 1). Steady state was assumed after at least three replacements of reactor volume, with the exception of the 8% loading regime under which steady state could not be established, ammonia levels rose to 135 mg L1, Chl a levels fell below 1 mg L1, the culture became chlorotic and finally collapsed. The oxygen regeneration time was measured at each of the various feed rates and following sparging with nitrogen to DO < 1 mg L1. Recovery of DO was found to be a sensitive indicator of the ability of the microalgal/bacterial co-culture to cope with a specific effluent loading rate. Oxygen regeneration times of 20 h, 21 h and 40 h were measured for the 3%, 5% and 8% loadings respectively. The 5% loading proved the highest at which stable operation could be maintained and here COD was reduced by 86%, ammonia by 92% and phosphate was completely removed, possibly mainly by precipitation as calcium phosphate given the substantial difference between the experimental (Y1P) and theoretical (Y2P) yields (Table 2). Nitrogen removal in this system could be attributed to a number of mechanisms, possibly occurring simultaneously, and including ammonia stripping, nitrification and direct uptake by the culture itself. Given the large difference between the experimental (Y1NT) and theoretical (Y2NT) total nitrogen removal yields, it was assumed that ammonia stripping accounted for the largest proportion of nitrogen removal in the system. While the specific growth rate m ¼ 0.05 d1 was lower than m ¼ 0.096 d1 reported for a mixotrophic A. platensis culture with glucose as the organic substrate [18], it is closer to m ¼ 0.068 d1 recorded for a propionate carbon source in the same study, and is consistent with growth in the complex protein/amino acid-enriched tannery effluent medium. Volumetric cell growth (QX) in the tannery effluent medium of 31.3 mg L1 d1 was also lower than the 60 mg L1 d1 reported in the Lodi et al. study [18] for growth on propionate.
8%
Effluent loading as % reactor volume
Fig. 1 e Photobioreactor study of a continuously fed Arthrospira culture at steady state with various loadings of tannery effluent as daily percentages of reactor volume and showing changes in COD, ammonia, nitrate and phosphate.
Table 2 e Kinetic parameters for a steady state culture of Arthrospira in photobioreactor culture, continuously fed tannery effluent at various rates as a percentage of reactor volume and grown at 25 C in a 12 h dark/light cycle (standard deviation in brackets). Parametera Growth m (d1) QX (mg L1 d1) Nitrogen removal QNT (mg L1 d1) Y1NT () Y2NT () Phosphate removal QP (mg L1 d1) Y1P () Y2P ()
3%
5%
8%
0.03(0.006) 5.19(0.3)
0.05(0.007) 8.65(0.35)
e 7.25(0.3)
16.90(1.02) 0.938(0.06) 0.023(0.002)
15.20(1.00) 0.905(0.05) 0.043(0.002)
13.30(0.98) 0.804(0.05) 0.042(0.002)
0.54(0.045) 0.95(0.07) 0.057(0.002)
0.90(0.078) 0.95(0.07) 0.057(0.002)
1.44(0.25) 0.95(0.07) 0.03(0.002)
a Parameters defined in materials and methods.
3.2.
Pilot plant
The implementation of the above findings at larger scale was undertaken at the tannery study site where two 80 m2 paddle wheel operated High Rate Ponds (HRP) were constructed (volume 12 m3 each) adjacent to the WSP [19]. The start-up regime for the pilot ponds followed that used in the photobioreactor study with a 3% dilution of tannery effluent inoculated with 10% by volume of Arthrospira culture sourced from the WSP. The ponds were batch-fed with a loading rate at 3% of pond volume daily until stable operating conditions had been established. Thereafter the pilot ponds were continuously fed and loaded successively at 5%, 8% and 10% of total volume with undiluted tannery effluent. This regime was repeated for both summer and winter seasons. Environmental conditions were less easily controlled than in the photobioreactor and results here may be considered to reflect quasisteady state operation. Data were sourced as single measurements after operation for at least 30 days at each loading rate. Once again the 5% loading was found to be the highest rate providing stable HRP operation (ammonia 58 mg L1). Ammonia levels in the 8% and 10% loadings rose to 129 mg L1 and 199 mg L1 respectively with the cultures shifting into stationary phase, finally becoming chlorotic and crashing (Fig. 2). In the 5% loading, COD was reduced by 85% and ammonia by 93%. The pH remained constant at 8.2e8.4 at all loading rates. A notable feature was the substantially higher biomass production of 4.46 g m2 d1 (dry mass) measured in the outdoor ponds compared to the photobioreactor study giving an Arthrospira biomass yield of 16.29 t ha1 yr1 measured for the pilot pond operation. Photosynthetically available radiation (PAR) ranged between 500 and 1300 mmol m2 s1 from winter to summer conditions at the site. A photosynthetic carbon fixation rate of 7.022 g m2 d1 was measured as an average for summer and winter conditions. These results compare with 7.3 g m2 d1e9.5 g m2 d1 (dry mass) [20] and 8 g m2 d1e12 g m2 d1 carbon fixation reported for Arthrospira grown in sewage wastewaters [21]. Biomass yields reported in commercial systems range widely between 10 and 30 t ha1 yr1 [22]. The difference between biomass yield and
186
b i o m a s s a n d b i o e n e r g y 5 1 ( 2 0 1 3 ) 1 8 3 e1 8 8
carbon fixation values recorded here indicates the measurement of the latter under idealized conditions that do not effectively average light incidence throughout the water column in the reactor. Light extinction due to opacity of the medium and changing cell concentration will vary across the depth of the reactor with actual biomass yield providing a more accurate indication of the biomass production potential for the system. This will also account for the lower biomass yields observed for tannery effluent compared to literature values for sewage and defined media systems. The kinetic values derived from the pilot pond study are reported in Table 3 and show a substantially increased QX with production around fourfold higher than in the photobioreactor. These results were also in broad agreement with the observation that ammonia stripping probably accounts for the largest fraction of nitrogen removal in the system. Control of ammonia toxicity by manipulation of the effluent loading rate will be an important factor in further scale-up of the system.
Table 3 e Kinetic parameters for the quasi-steady state outdoor culture of Arthrospira in the pilot plant HRP study continuously fed tannery effluent at various loading rates as a percentage of reactor volume and grown under a natural daylight regime. Data were sourced as single measurements after operation for at least 30 days at each loading rate. Parametera Growth m (d1) QX (mg L1 d1) Nitrogen removal QNT (mg L1 d1) Y1NT () Y2NT () Phosphate removal QP (mg L1 d1) Y1P () Y2P ()
3%
5%
8%
10%
0.03 21.90
0.05 31.3
0.025 18.70
0.02 17.3
17.30 0.96 0.097
27.2 0.91 0.088
39.40 0.82 0.036
42.80 0.72 0.031
0.54 0.95 0.242
0.90 0.95 0.208
1.44 0.95 0.078
1.80 0.95 0.057
a Parameters defined in materials and methods.
NH3-1
NH3-2 1200 1000 800 600 400 200
Effluent
Fig. 2 e Pilot plant study in which twin 80 mL2 high rate algal ponds inoculated with an Arthrospira culture were fed continuously at various loadings of tannery effluent as percentages of reactor volume, and showing changes in COD, ammonia, biomass and carbon fixation rates (annualized averages).
C fix2
NH3 (mg L-1)
C fixation (mg m -2 d -1)
C fix1 9000 8000 7000 6000 5000 4000 3000 2000 1000 0
5%
15%
25%
0
Effluent loading as % reactor volume
Fig. 3 e Flask study simulation of the effect of WSP terminal pond alkalinity recirculation on the photosynthetic productivity of an Arthrospira culture comparing C fix 1 (without recycle) and C fix 2 (with recycle) at various effluent loading rates. Ammonia levels in the study compare NH3 -1 (without recycle) and NH3 -2 (with recycle).
3.3.
Recirculation
A number of strategies have been proposed to deal with ammonia toxicity in microalgal ponding systems including the manipulation of loading rate [23], aeration stripping [24], addition of bicarbonate [25] and recirculation of oxygenated microalgae-enriched waters [26]. All of these may be applied in one degree or another in tannery WSP with terminal ponds in zero-discharge systems containing concentrated alkaline solutions of salts, including carbonates from photosynthetic activity, microbial biomass and residual slowly degradable organic compounds. Given the complexity of the solution, an empirical assessment of the recirculation alkalization option was investigated in flask studies first. Pipework constraints at the WSP study site had precluded the pumping of terminal pond water and thus the evaluation of this option during the pilot plant investigation. Flask studies were set up as described and compared the effect on Arthrospira production with increasing terminal pond recirculation loadings of 5%, 15%, and 25% pond water (carbonate alkalinity 2585 mg L1 as CaCO3). pH was elevated to 9.5 with the addition of the pond alkalinity, a substantial increase (w50%) in photosynthetic carbon fixation was observed and also a reduction in ammonia levels (Fig. 3).
Table 4 e Tannery effluent treatment performance in the reconfigured primary facultative pond (standard deviation in brackets).
COD (mg L1) Ammonia (mg L1) Sulphide (mg L1) Phosphate (mg L1) Suspended solids (mg L1)
Tannery effluent
After PFP
% Change
2474(1810) 731(98) 285(422) 19(12.5) 243(196)
1216(93) 452(105) 76(16) 1.7(2) 0.5(0.5)
50.8% 38.2% 73.3% 91% 99.8%
187
b i o m a s s a n d b i o e n e r g y 5 1 ( 2 0 1 3 ) 1 8 3 e1 8 8
Table 5 e Heavy metal concentrations in tannery effluent before and after treatment in the anaerobic compartment of a facultative pond unit located prior to the high rate pond, and also in Arthrospira biomass grown before and after the anaerobic treatment step. Feed standard as set out in Ref. [28] and subsequently consolidated in Ref. [29].
Chromium Cadmium Cobalt Iron Lead Nickel Zinc
Untreated effluent mg L1
Treated effluent mg L1
Biomass before treat mg kg1
Biomass after treat mg kg1
Feed standard mg kg1
4.68(0.1) 0.08(0.02) 0.35(0.03) 39.9(2.1) 0.76(0.03) 0.41(0.01) 1.51(0.1)
0.25(0.02) 0.02(0.01) 0.18(0.01) 0.41(0.03) 0.11(0.01) 0.21(0.01) <0.1
25.8(0.5) 5.96(0.23) 22.4(0.6) 2012(12.2) 219(2.3) 49.2(1.5) 218.5(2.8)
<1 <1 3.3(0.2) 795(13.4) 2.3(0.2) 17.5(0.8) 22.5(1.1)
10e15 1.0 4e6 1250e2500 10 10e15 250e500
Although ammonia levels at loading rates above 5% were higher than the 60e80 mg L1 toxic threshold, this was not found to be inhibitory up to the 15% loading, where DO levels remained well above saturation. These results suggest a number of mechanisms of ammonia detoxification including conversion of ammonia to a non-toxic ionized form at elevated pH 9.8 (pKa 9.25), removal by stripping and nitrification and possibly a small amount of direct uptake by the Arthrospira culture itself. The kinetic values derived in both the photobioreactor and pilot pond study support this conclusion (Table 2 and Table 3). The viability of a terminal pond recirculation strategy at the test site would have important process design implications by taking into account higher growth rates, higher effluent loading rates and therefore also a reduced HRP reactor size requirement. Based on these results a maximum recirculation rate of 0.2:1 was proposed as a design value which could be adjusted empirically during full-scale implementation.
3.4.
Primary facultative pond
The Advanced Integrated Algal Wastewater Ponding System (AIWPS) developed by Oswald [19] makes provision for a primary facultative pond (PFP) unit ahead of the HRP in the treatment of wastewaters. A pond in the study site tannery WSP system was retrofitted to operate as a PFP. The feed was relocated to the base of the pond providing an upflow anaerobic sludge zone, and shallow draught mechanical aerators were fitted to the pond surface, among other reasons, to control odour emissions. Methane gas production commenced soon after start-up and Table 4 reports the outcome of this intervention. COD was reduced by 50% and ammonia by 38% (assuming near complete ammonification, stripping and partial oxidation in the aerobic upper layer). Sulphate reduction was complete in the anaerobic zone (sulphate was reduced from 975 mg L1 to <1 mg L1), and partial re-oxidation of sulfide (73%) occurred in the aerobic zone. Suspended solids were nearly completely removed. The low residual phosphate levels could present a limitation for subsequent Arthrospira growth in the HRP.
3.5.
will be of importance, especially where a feed grade product is targeted [27]. An analysis of heavy metal levels in Arthrospira biomass harvested from the WSP at the study site is reported in Table 5 [9]. Given high levels of sulfates and sulfides in the PFP noted above, its use as a pretreatment step for the removal of contaminating heavy metals was also investigated. Table 5 reports the reduction in effluent heavy metal concentration during this operation, and also the changes in Arthrospira biomass levels grown before and after treatment of the effluent in this unit. The WSP Arthrospira strain was found to be unaffected by the high concentrations of sulfide (1100 mg L1) from the anaerobic pond unit, which were rapidly oxidized back to sulphate in the HRP (<1 mg L1). Chromium in all samples analysed over a three-year period was in the Chromium III form. Chromium VI was not found in the system.
4.
Conclusions
1. The scale-up studies have demonstrated technical viability in the use of tannery effluent as a growth medium for Arthrospira production. 2. Arthrospira growth yields around 16 t ha1 yr1 were within the range reported for open pond commercial production systems. 3. Control of ammonia toxicity will determine loading rates and therefore the HRP pond size required. For successful operation, the loading rate will need to be directly coupled to the ammonia removal rate. 4. Accumulated alkalinity in tannery WSP may be recirculated to the HRP to raise the pH and shift ammonia to the ionized form and thereby increase the effluent loading rate with important implications for pond sizing and costs. 5. Heavy metals contamination of the Arthrospira biomass may be a problem in these wastewaters but may be substantially reduced in an anaerobic pretreatment step. 6. The outcomes of this study provide design values for the fullscale implementation of the process for Arthrospira production in tannery wastewaters. Process scale-up has been undertaken and the design, construction and operation of the high rate ponding operation at industrial scale has been reported in the Part 2 follow-up to this study.
Heavy metals removal
Given the wide range from which hides are sourced, and a diversity of commercial chemicals used in the tanning operation, these wastewaters may contain heavy metals. Microalgal biomass is at risk of accumulating a proportion of these which
Acknowledgements The authors acknowledge the support of Messrs. RE Newson and R Smith formerly of Mossop-Western Leathers Co. Pty.
188
b i o m a s s a n d b i o e n e r g y 5 1 ( 2 0 1 3 ) 1 8 3 e1 8 8
Ltd., Rhodes University and the Water Research Commission for financial support and permission to use data from WRC Report TT188/02.
references
[1] Chisti Y. Biodiesel from microalgae. Biotechnol Adv 2007;25: 294e306. [2] Greenwell HC, Laurens LML, Shields RJ, Lovitt RW, Flynn KJ. Placing microalgae on the biofuels priority list: a review of the technological challenges. J R Soc Interface 2010;7:703e26. [3] Christensen L, Sims R. Production and harvesting of microalgae for wastewater treatment, biofuels, and bioproducts. Biotechnol Adv 2011;29(6):686e702. [4] Harun M, Singh M, Forde GM, Danquah MK. Bioprocess engineering of microalgae to produce a variety of consumer products. Renew Sust Energ Rev 2010;14:1037e47. [5] Su Y, Mennerich A, Urban B. Municipal wastewater treatment and biomass accumulation with a wastewaterborn and settleable algal-bacterial culture. Water Res 2011; 45:3351e8. [6] von Sperling M. Waste stabilisation ponds. London: IWA Publishing; 2007. [7] Rowswell RA, Cooper DA, Shuttleworth SG. Evaporation ponds: a solution for tannery effluent disposal. J Soc Leath Tech Ch 1984;69:123e9. [8] Tadesse I, Greene FB, Puhakka JA. The role of sulphidogenesis in anaerobic treatment phase of tannery wastewater treatment in advanced integrated wastewater pond system. Biodegradation 2003;14:219e27. [9] Maart BA. The biotechnology of effluent grown Spirulina and applications in aquaculture nutrition. Grahamstown: Rhodes University MSc Thesis 1993. [10] Mara DD. Domestic wastewater treatment in developing countries. London UK: Earthscan Publications; 2004. [11] Munoz R, Guieysse B. Algal bacterial processes for the treatment of hazardous contaminants: a review. Water Res 2006;40(15):2799e815. [12] Rose PD, Maart BA, Dunn KM, Rowswell RA, Brits P. High rate algal oxidation ponding for the treatment of tannery effluents. Water Sci Technol 1996;33:219e27. [13] Dunn KM. The biotechnology of high rate algal ponding systems in the treatment of saline tannery wastewaters. Grahamstown; Rhodes University PhD Thesis 1998. [14] APHA. In: Clesceri LS, editor. Standard methods for the examination of water and wastewater. 17th ed. Washington, DC: American Public Health Association; 1989.
[15] Lichtenthaler HK. Chlorophylls and carotenoids: pigments of photosynthetic biomembranes. Method Enzymol 1987;148: 350e71. [16] Oren A. Bacterial activities in the Dead Sea: 1980e1991 a survival at the upper limits of salinity. Int J Salt Lake Res 1992;1:7e20. [17] Cornet JF, Dussap CG, Cluzel P, Dubertret G. A structured model for simulation of cultures of the cyanobacterium Spirulina platensis in photobioreactors: II. Identification of kinetic parameters under light and mineral limitations. Biotechnol Bioeng 1992;40:826e34. [18] Lodi A, Binaghi L, De Faveri D, Carvalho JCM, Converti A. Fedbatch mixotrophic cultivation of Arthrospira (Spirulina) platensis (Cyanophycea) with carbon source pulse feeding. Anal Microbiol 2005;55(3):181e5. [19] Oswald WJ. Large-scale algal systems (engineering aspects). In: Borowitzka MA, editor. . In: Borowitzka LJ, editor. Microalgal biotechnology. Cambridge UK: Cambridge University Press; 1988. [20] Saxena PN, Ahmad MR, Shyam R, Amla DV. Cultivation of Spirulina in sewage for poultry feed. Experientia 1983;39: 1077e83. [21] Fox R. Nutrient preparation and low cost basin construction for village production of Spirulina. In: Stadler T, editor. Algal biotechnology. Amsterdam: Elsevier; 1988. [22] Jime´neza C, Cossı´ob BR, Niella FX. Relationship between physicochemical variables and productivity in open ponds for the production of Spirulina: a predictive model of algal yield. Aquaculture 2003;221:331e45. [23] Abeliovitch A. The effects of unbalanced ammonia and BOD concentrations on oxidation ponds. Water Res 1983;17(3): 299e301. [24] Pano A, Middlebrooks EJ. Ammonia nitrogen removal in facultative wastewater stabiliztion ponds. J Water Control Fed 1982;54:344e51. [25] Chaudari PR, Krishnamoorthi KP, Vittal Rao M. Growth potential of Spirulina, a blue green alga in sewage. Indian Acad Sci 1980;89(3). [26] Pescod MB. The role and limitations of pond systems. Wat Sci Tech 1996;33(7):11e22. [27] Yi-Chao Lee Y-C, Chang S-P. The biosorption of heavy metals from aqueous solution by Spirogyra and Cladophora filamentous macroalgae. Bioresour Technol 2011;102: 5297e304. [28] German Ministry of Agriculture Law Collection No 117; 1987. [29] European Union. Regulation (EC) No 882/2004 of the European parliament and of the Council of 7 May 2002 on undesirable substances in animal feed.