Fuel 113 (2013) 415–419

Contents lists available at SciVerse ScienceDirect

Fuel

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

Potential application of anaerobic digestion to tobacco plant

0016-2361/$ - see front matter � 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.fuel.2013.06.006

⇑ Corresponding author. Tel.: +34 685764503.E-mail addresses: almudenagg@unex.es (A. González-González), cuadros1@

unex.es (F. Cuadros), aruiz@unex.es (A. Ruiz-Celma), ferlopez@unex.es(F. López-Rodríguez).

A. González-González a,⇑, F. Cuadros a, A. Ruiz-Celma b, F. López-Rodríguez c

a Dpto. de Física Aplicada, Universidad de Extremadura, Avda. de Elvas s/n, 06006 Badajoz, Spainb Dpto. de Ingeniería Mecánica, Energética y de los Materiales, Universidad de Extremadura, Avda. de Elvas s/n, 06006 Badajoz, Spainc Dpto. de Expresión Gráfica, Universidad de Extremadura, Avda. de Elvas s/n, 06006 Badajoz, Spain

h i g h l i g h t s

� Results of anaerobic digestion (AD) of the tobacco plant are presented.� Energetic benefits of AD are quantified and compared with other energy crops.� The amount of water recovered by biomethanation of the tobacco plant is quantified.� AD is an appropriate process to treat tobacco plant.

a r t i c l e i n f o

Article history:Received 22 August 2012Received in revised form 30 May 2013Accepted 4 June 2013Available online 20 June 2013

Keywords:Tobacco plantEnergy cropAnaerobic digestionBiofuelBiogas

a b s t r a c t

In this work, the energetic feasibility of using tobacco (not tobacco waste materials) as a raw material inthe process of anaerobic digestion has been studied in terms of methane production, in order to demon-strate the potentiality of tobacco plant to be used as an energy crop.

Long-term experiments have been performed at the laboratory scale with and without a regulation ofthe substrate pH and within the mesophilic range (38 �C). Methane production and the parameters thatcontrol the anaerobic digestion process have been monitored periodically. The highest methane produc-tion, 53.84 Nm3 methane/tonne of fresh tobacco, and a chemical oxygen demand reduction of 53.26%were achieved when the substrate with 15% tobacco (m/m) was treated with a hydraulic retention timeof 16 days.

In addition, biomethanisation of this substrate can recover 57% of the water contained in the mixture,which can be used for irrigation of the tobacco crop.

It should be noted that pH regulation is essential to ensure the stability of the biological process, sinceall the experiments was carried out with the natural pH showed signs of inhibition.

These outputs of methane are on the same order as those obtained from other substrates, such as Sudangrass, fodder beet, and millet and place an industrial plant for tobacco biomethanation at the edge ofeconomic profitability.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Extremadura is a region, located in the south-west of Spain andborder Portugal, where tobacco has been traditionally cultivatedfor decades, in fact, more than 93% of the dried tobacco of Spainis produced in Extremadura [1]. It has been a highly profitable cropbecause of the Common Agricultural Policy (CAP) (a European sub-sidy) has retained large expanses of fertile land unusable for theproduction of substrates or consumer products. Nevertheless, inrecent years this subsidy to the tobacco cultivation has been re-duced and it is scheduled to end in 2013, resulting in a drop in

the prices paid by tobacco companies. Thus, only crops thatimprove quality and greatly reduce production costs will be viable,but profits will be notably less than farmers currently receive.Therefore, the area dedicate to the tobacco production will be re-duced if other alternatives for tobacco cultivation are not found.

This work is focused on the analysis of the feasibility of a newapplication, the use of tobacco as a substrate to generate renew-able energy through anaerobic digestion (AD).

One of the major obstacles to the use of tobacco as biofuel isthat its cultivation is expensive and demanding when the leavesare produced for the tobacco industry. However, tobacco plantscan be grown by a high-density planting method similar to fodderproduction. This method can reach outputs of 150 t fresh tobacco/ha with a moisture content of 80%. Its high productivity is due tothat four or five cuts can be made during one growing season,which may extend to the end of October. Consequently, the

416 A. González-González et al. / Fuel 113 (2013) 415–419

potential of growing tobacco for biomass production and its usedas fuel is based on its high productivity and on the fact that it isnot a food crop.

Although tobacco plant and its wastes have been used as sub-strate for power generation by pyrolysis [2], gasification [3], com-bustion [4], and the tobacco seeds can produce biodiesel [5,6], theirtreatment by AD has hardly been studied. However the high mois-ture content of the freshly cut tobacco plants makes possible theirenergy transformation by this technology, in fact it has been ap-plied to the treatment of several forage crops that can be comparedwith the tobacco plant [7]. Biogas production for some fodder isshown in Table 1.

With respect to the biomethanation of tobacco, we have notfound any literature related to the AD of fresh tobacco, hence theimportance of the results shown in this paper. Only one articlereferring to waste from the tobacco industry has been found [8],which showed methane production between 169 and 282 L/kg to-tal solids fed, day, depending on the ambient temperature, with ahydraulic retention time (HRT) of 15 days. Furthermore, we areaware of two tests that a private company performed on the bio-methanation of the tobacco plant; in both cases a TS content of8% was tested. The results that were obtained for tobacco diluted80% in water (the wet method) were 770 L biogas/kg organic mat-ter, i.e., 50 Nm3 biogas/t fresh tobacco (Kompogas, personal com-munication). For a 50% dilution (the dry method) the amount ofbiogas produced was lower because the TS content (8%) was far be-low the optimum content required in a biodigester that operates inthe dry mode, which should be approximately 30%. In agreementwith these previous experiments, the best option is to operate inwet mode, which allows to work with TS contents from 5 to 15%.

The main objective of this work is to analyse the environmentaland energetic feasibility of an AD system at a laboratory scale thatutilises the freshly cut tobacco plant as a substrate, which has beencultivated specifically for this purpose. Is the use of tobacco as abiofuel environmentally acceptable? Is the use of tobacco as a bio-fuel energetically feasible?

As mentioned, the idea is to take advantage of the excellent out-puts per hectare of tobacco that are achieved in the north ofExtremadura (Spain) and biodegradability of tobacco. Definitively,this study may offer an alternative to tobacco cultivation for hu-man consumption while preventing the elimination of jobs in theregion, improving living conditions in the area, attempting to allowthe rural settler to remain in their environment.

2. Materials and methods

2.1. Anaerobic inoculum, substrate and experimental design

The substrate that is analysed in this work, tobacco plant, lackssuitable microorganisms to activate a biodigestion process, thus anacclimated inoculum is necessary. It was taken from an anaerobicreactor located at the Wastewater Treatment Plant (WTP) inBadajoz which treat the sludge from the primary and secondary

Table 1Biogas production and percentage of methane obtained via the AD of various foddersubstrates.

Substrate Biogas production (L/kg) Percentage of methane (%)

Corn silage 202 52Grass silage 172 54Rye silage 163 52Sudan grass 128 55Fodder beet 111 51Millet 108 54Turnip leaves 70 54

treatments. 2 L of this sludge (usable volume of the reactor), with-out mixing with any other, were used to start up each experimen-tal digester.

Initially, the freshly cut tobacco plant were forced to undergo amechanical treatment to obtain a sufficiently small particle size[9], because the lower the particle size, the higher the efficiencyof the process, given the fact that a decrease in particle size impliesan increase of the surface on which bacteria might act. After thispre-treatment, crushed tobacco was mixed with tap water in orderto increase the moisture content of the mixture. Four mixtures offresh tobacco and water were tested: 5% fresh tobacco/95% water,10% fresh tobacco/90% water, 15% fresh tobacco/85% water, and20% fresh tobacco/80% water. The characterisation of these sub-strates is summarised in Table 2.

Despite the low pH of the substrates, the AD experiments werefirstly performed without regulating the acidity of the mixtures.After that, the four mixtures were treated with pH 7 in order todetermine the influence of pH in the AD process. To obtain pH val-ues of approximately 7, a small quantity of Ca(OH)2 was added tothe substrate mixtures. This compound is very inexpensive, has astrong alkalinity, and is chemically inert, non-toxic, and easy tohandle. The operational conditions for each experiment are shownin Table 3.

2.2. Experimental setup, start up and development of the anaerobicdigestion process

Fig. 1 shows a schematic of the experimental setup used to per-form the AD experiments in semicontinuous mode. This operationmode involves a daily extraction of a given volume of digestedsludge from the reactor, immediately followed by the introductionof the same volume of new substrate, so that a constant volume ofthe reactor is guaranteed.

The setup basically consists of a 2 L glass flask with a rim at-tached to a central tube immersed in the reaction medium andwith an input opening for the insertion of the substrate and theextraction of the digested sludge, and an output one for the collec-tion of the biogas generated in the process.

The digestion unit was submerged in a water tank maintainedat 38 �C by a thermostat. The substrate inside the reactor washomogenised using a magnetic stirrer. Experimental design guar-anteed that the temperature was uniform throughout the reactorvolume. The operating conditions should therefore be regardedas optimal. A 5 L tank attached to the biodigester was used todetermine the volume of methane generated during the AD pro-cess. A squeeze bottle containing a sodium hydroxide solution(20% by weight) was placed between the digester and the gas tank,with the aim of retaining the carbon dioxide generated during thedigestion process. The methane generated during the experimentdisplaced the water in the tank, which was collected in a measur-ing cylinder. Thus, the volume of the displaced water could be usedto calculate the volume of methane generated daily in each exper-iment at ambient temperature (23 ± 1 �C) and atmospheric pres-sure (1017 ± 5 hPa).

The start up of the AD process begins with the introduction of2 L of inoculum into the digesters after that, they are sealed so thatthe biodigestion process take place in the absence of oxygen andinoculum is subjected to gentle agitation to promote the contactwith the microorganisms. The day after the load starts the feedingprocess, increasing volumes of substrate are daily added to theinoculum until reach the flow of substrate to be treated so thatthe bacteria could acclimate to the substrate.

Each flow was treated for a period of time that can ensure thatreliable results were obtained, in terms of the level of degradationand the methane production. The optimum time for each assay wascalculated based on the HRT which is the period of time that the

Table 2Physicochemical characterisation of substrates.

Parameter Tobacco 5% Tobacco 10% Tobacco 15% Tobacco 20%

Initial COD (g/L) 11.70 ± 2.04 15.85 ± 0.88 26.27 ± 4.97 35.14 ± 9.60VFA (g/L) 0.57 ± 0.13 0.72 ± 0.02 0.99 ± 0.40 0.67 ± 0.32Alkalinity (g/L) 0.10 ± 0.03 0.10 ± 0.05 0.15 ± 0.07 0.30 ± 0.14VSS (g/L) 16.44 ± 5.90 12.98 ± 7.21 30.55 ± 10.59 18.27 ± 6.01VDS (g/L) 1.92 ± 0.25 3.08 ± 0.06 3.16 ± 0.23 5.77 ± 0.38Total organic carbon (g/L) 4.45 ± 0.35 8.67 ± 0.82 14.89 ± 0.47 18.56 ± 0.96Total organic nitrogen (g/L) 0.24 ± 0.05 0.51 ± 0.02 0.76 ± 0.08 0.98 ± 0.04C/N 18.54 ± 1.56 17.00 ± 0.68 19.59 ± 2.75 18.94 ± 1.89pH 5.55 ± 0.06 5.28 ± 0.08 5.10 ± 0.01 5.02 ± 0.01

Table 3Operation conditions for anaerobic digestion experiments.

Experiment Substrate composition HRT (days) pH

1 5% fresh tobacco/95% water 14 Natural acidity2 10% fresh tobacco/90% water 20 Natural acidity3 20% fresh tobacco/80% water 20 Natural acidity4 10% fresh tobacco/90% water 20 7 ± 0.25 15% fresh tobacco/85% water 20 7 ± 0.26 15% fresh tobacco/85% water 16 7 ± 0.27 15% fresh tobacco/85% water 12 7 ± 0.28 20% fresh tobacco/80% water 20 7 ± 0.2

A. González-González et al. / Fuel 113 (2013) 415–419 417

waste must be in contact with the bacterial population. Thisparameter was calculated as the ratio between the volume of theanaerobic reactor and the volume of substrate added daily. Thus,for example, when a volume of 100 mL substrate/day is tested,the HRT is 20 days, so; in this case the experiment was maintained

1 Magnetic stirrer; 2 Digester; 3 Thermo6 NaOH squeeze bottle; 7 Methane mea

2

1

3

4

5

Fig. 1. Semi-continuous

for at least 40 days (twice the HRT). The first HRT stands for theacclimatization period, after which the reaction stabilizes. The sta-bilization of the anaerobic process was reached when effluentChemical Oxygen Demand (COD) was reduced and it was constantand the volume of methane produced daily was achieved a maxi-mum and also was constant.

Finally, when each experiment ends the digested effluent wascollected and decanted to separate solid and liquid fractions aswell as to determine the volume of water that can be recovered.

2.3. Analytical methods

A periodic sampling protocol was applied to the digesters dur-ing the experiments to record data for the following parameters:COD, Volatile Fatty Acids (VFA), alkalinity, Volatile Suspended Sol-ids (VSS), Volatile Dissolved Solids (VDS) and pH. All these param-eters were analysed weekly, except pH that was performed daily.

meter; 4 Sampling; 5 Heating unit; suring unit.

1

2

4

6

7

anaerobic digester.

418 A. González-González et al. / Fuel 113 (2013) 415–419

Total organic carbon, and total organic nitrogen were determinedonly on the substrates.

VFA, alkalinity, pH, VSS and VDS were analysed according tostandard methods [10]. On the other hand, Nanocolor� kits and aportable PF-12 spectrophotometer (all from Macherey–Nagel Com-pany, Duren, Germany) were used for data analysis of the COD andtotal organic carbon and nitrogen levels.

The monitoring of all of these analytical parameters allowed theAD process to be tracked and possible anomalies to be detected.

3. Results and discussion

3.1. AD tests without pH regulation

Experiment 1 yielded an average methane production of22.90 L/kg fresh tobacco and an average reduction in COD of43.36%. In this experiment, methane production during the firststage of the biomethanation process was high and then declinedgradually until the reaction was inhibited 31 days after the startof the steady state.

This same behaviour was observed in all of the other tests. Forexperiment 2, the average methane production value was 23.58 L/kg fresh tobacco, with an average decrease in COD of 59.14%. In thiscase, inhibition was reached 43 days after the start of the steadystate. Experiment 3 reached an average methane production valuewas 26.97 L/kg fresh tobacco with an average decrease in COD of68.01%. In this case, inhibition began 27 days after the start ofthe steady state.

In all of the experiments performed without daily pH regulationof the substrate, the bioreaction was inhibited. This result indicatesthat to maintain the AD process over time, the pH should be ad-justed daily until a value close to neutral.

3.2. AD test with pH regulation

Experiment 4 was performed using the same substrate and HRTthat 2 but a small amount of calcium hydroxide was added to thesubstrate in order to maintain its pH around seven. The purpose ofthis test was to confirm whether the bioreaction was self-regulat-ing over time and whether some variation in methane productionalso occurred with respect to the analogous experiment withoutdaily pH regulation of the substrate. The average methane produc-tion was 42.91 L/kg fresh tobacco, which is almost twice that pro-duced without substrate pH regulation (23.58 L/kg fresh tobacco).The reduction in COD was 44.60%, and the duration of the steadystate phrase of the reaction was 108 days (5.4 times the HRT),without signs of inhibition (see Table 4).

Experiment 8 repeated experimental conditions of experiment3 but including pH regulation in the substrate. The total durationof the steady state was 230 days without inhibition. The averagemethane production was 28.92 L/kg fresh tobacco. In this case,

Table 4Results of AD experiments.

Experiment (LCH4/kg freshtobacco)

pH VFA (g/L) Alkalinity(g/L)

VSS (g/

1 22.92 ± 10.18 6.92 ± 0.74 0.38 ± 0.09 3.16 ± 0.85 2.67 ± 02 23.58 ± 10.64 6.79 ± 0.17 0.87 ± 0.12 1.33 ± 0.22 3.98 ± 23 26.97 ± 8.29 7.13 ± 0.15 1.39 ± 0.77 2.87 ± 0.68 7.25 ± 34 42.91 ± 13.14 7.32 ± 0.10 1.82 ± 1.23 2.74 ± 0.63 3.56 ± 15 51.93 ± 18.40 7.31 ± 0.15 1.64 ± 1.02 4.10 ± 1.14 8.80 ± 26 53.84 ± 15.48 7.26 ± 0.10 0.49 ± 0.10 2.37 ± 0.26 5.41 ± 17 39.33 ± 14.06 7.35 ± 0.17 1.24 ± 0.87 3.38 ± 0.25 3.87 ± 08 28.92 ± 8.18 7.27 ± 0.13 1.37 ± 0.98 2.90 ± 0.62 7.69 ± 2

a Denotes inhibition of the reaction.

there is little difference from the output obtained when therewas no pH regulation (26.97 L/kg fresh tobacco). However, thereaction was prolonged up to 11.5 times the HRT, indicating thatthe reaction self-regulates for long periods of time. The character-istic parameters of this experiment are shown in Table 4.

A comparison of the values of methane production for the 10%tobacco mixture with the 20% tobacco mixture revealed a decreasefrom 42.91 L/kg fresh tobacco to 28.92 L/kg fresh tobacco. This re-sult suggests that such high proportions of tobacco do not allowmethanogenic bacteria to adequately develop.

Therefore, several experiments (5–7) were performed with asubstrate formed by 15% fresh tobacco and 85% water with dailypH regulation and trying different HRT. The results (Table 4) con-firmed that a this proportion of tobacco by weight is optimal withregard to the methane production. Experiment 5 achieved an aver-age of 51.93 L methane/kg fresh tobacco and a reduction in COD ofthe influent of 35.28%. In experiment 6 the methane productionand the COD reduction were increased to 53.84 L methane/kg freshtobacco (72 ± 6% methane in the biogas) and 53.26% respectively.However, in experiment 7, when the HRT was lowered to 12 daysor if the feeding volume was increased to 1/6 L/day, the averagemethane production decreased considerably to 39.33 L methane/kg fresh tobacco, but the reduction in COD was the highest at60.01%.

Therefore, the optimum proportion of substrate is 15% byweight of fresh tobacco/85% by weight of water and the maximumproduction of energy, 53.84 L methane/kg fresh tobacco, was ob-tained for a HRT of 16 days. Figs. 2 and 3 show the time evolutionof methane production and the pH respectively for the optimummixture, once the stability of the biological process has beenreached. It can observe in Fig. 2 that methane production presentsa strong variability, even during the steady state; this feature wasobserved in all the experiments conducted. However, this factcannot be associated with the presence of changes in operatingconditions as the pH remains almost constant in steady state(Fig. 3).

For these optimum conditions of biomethanation, decanting ofthe digested effluent allows to recover 57.5% of the water con-tained in the substrate. Thus, for each cubic meter of substrateinfluent, 557 L of water would be recovered, taking into accountthat the fresh tobacco has a moisture content of 80%.

On the other hand, methane production achieved by the AD offreshly cut tobacco plant can be comparable to that obtained withother mono-substrates, such as Sudan grass, fodder beet, and mil-let, which were on the order of 70.40, 56.61, and 58.32 L methane/kg respectively.

So, this result places the generation of renewable energy (bio-gas) via the industrial biomethanation of tobacco at the edge ofprofitability. The comparison with these other crops would be evenmore favourable for tobacco if we compare the methane produc-tion per hectare of crop per year.

L) VDS (g/L) Final COD (gO2/L)

COD reduction(%)

Duration of steady state(days)

.76 1.53 ± 0.23 6.86 ± 1.53 43.36 ± 3.21 31a

.17 1.63 ± 0.36 6.48 ± 0.90 59.14 ± 5.69 43a

.25 2.45 ± 0.86 16.80 ± 5.43 68.01 ± 8.00 27a

.00 1.45 ± 0.35 9.10 ± 2.70 44.60 ± 13.84 108

.24 3.08 ± 1.58 16.90±.349 35.28 ± 13.32 211

.12 1.43 ± 0.21 12.28 ± 1.07 53.26 ± 4.07 40

.42 2.01 ± 0.28 10.50 ± 0.49 60.01 ± 1.86 52

.97 1.99 ± 0.91 15.86 ± 5.00 54.86 ± 14.23 230

Fig. 2. Variations in methane production under steady-state conditions with 15%tobacco and a HRT of 16 days.

Fig. 3. Variations in the pH under steady-state conditions with 15% tobacco and aHRT of 16 days.

Table 5Average methane production for different energy crops, obtained through AD, and theaverage annual production of methane per hectare of crop.

Substrate (Lmethane/kg)

Average crop production(tonne/ha year)

Methane production(Nm3/ha year)

Sudangrass

70.40 12.12 853.25

Fodderbeet

56.61 71 4019

Millet 58.32 0.794 46.31Tobacco 53.84 135 7268

A. González-González et al. / Fuel 113 (2013) 415–419 419

According to Meister [11], the production of Sudan grass variesbetween 11 and 13 t/ha year, with an average value of 12.12 t/ha -year. The corresponding values for fodder beet oscillate between55.5 and 86.5 t/ha year [12], with an average value of 71 t/ha year,and for millet is on average 0.794 t/ha year [13]. Tobacco produc-tion per hectare per year in Extremadura (Spain) can reach150 tonnes. For comparisons with the previous crops, we will usemore conservative value of 135 t/ha year.

The average daily production values of methane per tonne ob-tained via AD of the different abovementioned energy crops andthe average annual methane production per hectare per year arelisted in Table 5. The methane production per hectare per year of

tobacco (7268 m3) is greater than obtained with the other cropsanalysed.

Comparing our results with those of Meher et al. [8] and recall-ing that the moisture content of fresh tobacco is 80%, the optimummethane production obtained in the present work relative to theTS would be equal to 223.53 L/kg TS. This amount is much greaterthan that obtained by Meher et al. for the biomethanation of tobac-co waste, which varied between 100 and 170 L/kg TS.

The results obtained in the present work are 7.68% greater thanthose obtained through the AD of the tobacco plant via the wetmethod by KOMPOGAS, using their own technology.

4. Conclusions

The AD experiments without daily pH regulation are inhibitedafter a time 1.5 to 2 times the Hydraulic Residence Time (HRT)but if a small amount of Ca(OH)2 is added, the AD process is notinhibited. Moreover, the optimum energy production(53.84 L methane/kg fresh tobacco) is obtained for the mixture of15% tobacco/85% water with a HRT of 16 days. Additionally, 557 Lof water could be recovered for each cubic meter of digested sub-strate. Finally, it should be noted that tobacco enables better en-ergy yields than other energy crops, so the AD of tobacco couldbe economically profitable.

Acknowledgments

This work was supported in part by the Government ofExtremadura through Project PCJ 100201 and GR10045. A. Gon-zález-González thanks the Ministry of Education and Science forthe grant from the FPU (reference AP2008-02546).

References

[1] Ministry of Agriculture, Food and Environment. Advances in agricultural areasand crops. <http://www.magrama.gob.es/es/estadistica/temas/estadisticas-agrarias/cuaderno_Mayo2012_tcm7-214649.pdf>; [cited 12.03.13].

[2] Yang Y, Li T, Jin S, Yang H. Catalytic pyrolisis of tobacco rob: kinetic study andfuel gas produced. Bioresour Technol 2011;102:11027–33.

[3] Yang Y, Jin S, Lin Y, Huang S, Yang H. Catalytic gasification of tobacco rob insteam-nitrogen mixture: kinetic study and fuel gas analysis. Energy2012;1:509–14.

[4] Yang Z, Zhang S, Liu L, Li X, Chen H, Yang H, et al. Combustion behaviours oftobacco stem in a thermogravimetric analyser and a pilot-scale fluidized bedreactor. Bioresour Technol 2012;110:595–602.

[5] Usta N, Aydogan B, Çon AH, Uguzdogan E, Özkal SG. Properties and qualityverification of biodiesel produced from tobacco seed oil. Energ ConversManage 2011;52:2031–9.

[6] Veljkovic VB, Lakicevic SH, Stamenkovic OS, Todorovic ZB, Lazic ML. Biodieselproduction from tobacco (Nicotiana tabacum L.) seed oil with a high content offree fatty acids. Fuel 2006;85:2671–5.

[7] Proyecto probiogas. <www.probiogas.es>; [cited 12.03.13].[8] Meher KK, Panchwagh AM, Rangrass S, Gollakota KG. Biomethanation of

tobacco waste. Environ Pollut 1995;90(2):199–202.[9] Izumi K, Okishio Y, Nagao N, Niwa C, Yamamoto S, Toda T. Effects of particle

size on anaerobic digestion of food waste. Int Biodeter Biodegr 2010;64:601–8.[10] APHA, AWWA, WPCF. Standard methods for the examination of water and

wastewater. 18th ed. Washington, DC: American Public Health AssociationPublications; 1992.

[11] Meister HS. Sample cost to establish and produce Sudangrass. UC CooperativeExtension, Imperial County; 2004. <http://coststudies.ucdavis.edu/files/sudangrass04.pdf>; [cited 12.03.13].

[12] Roth G, Fier MC, Antle M. Crop management research report. Department ofcrop and soil sciences. Penn State University; 2008.

[13] Food and Agriculture Organization of the United Nations (FAO). <http://www.fao.org/docrep/T0818e/T0818E05.htm>; [cited 12.03.13].