A pilot-scale experiment of natural rubber processing wastewater treatment was conducted using a combination system consisting of a two-stage up-flow anaerobic sludge blanket (UASB) and a down-flow hanging sponge (DHS) reactor for more than 10 months. The system achieved a chemical oxygen demand (COD) removal efficiency of 95.7% ± 1.3% at an organic loading rate of 0.8 kgCOD/(m3.d). Bacterial activity measurement of retained sludge from the UASB showed that sulfate-reducing bacteria (SRB), especially hydrogen-utilizing SRB, possessed high activity compared with methane-producing bacteria (MPB). Conversely, the acetate-utilizing activity of MPB was superior to SRB in the second stage of the reactor. The two-stage UASB–DHS system can reduce power consumption by 95% and excess sludge by 98%. In addition, it is possible to prevent emissions of greenhouse gases (GHG), such as methane, using this system. Furthermore, recovered methane from the two-stage UASB can completely cover the electricity needs for the operation of the two-stage UASB–DHS system, accounting for approximately 15% of the electricity used in the natural rubber manufacturing process.

INTRODUCTION

The natural rubber industry is concentrated in South-East Asia, particularly Thailand, Indonesia, and Malaysia. More than 60% of worldwide natural rubber production comes from these three countries (Mohammadi et al. 2010). Rubber latex, which is extracted from rubber trees, is used as a raw material for three intermediate forms of rubber products: ribbed smoked sheets (RSS), which are used as a raw material for making vehicle tires and industrial rubber parts; technically specified rubber, which is used as a raw material for making high viscosity products such as belts; and concentrated latex (CL), which is used as a raw material for making dipped products such as medical gloves. These three intermediate products are used in downstream rubber product industries (Tekasakul & Tekasakul 2006; Jawjit et al. 2010a, 2010b).

In a natural rubber processing factory, 20–35 m3 of wastewater, containing a high concentration of organic matter and nitrogen (particularly ammonia), is discharged for 1 ton of product (Nugyen & Luong 2012). This is because a large amount of ammonia is used for the preservation of rubber latex (Jawjit et al. 2010a, 2010b; Thongnuekhang & Puetpaiboon 2004). Furthermore, CL wastewater, which is the highest-strength wastewater, contains a high concentration of sulfate because sulfuric acid is used for the recovery of rubber from serum latex. In South-East Asia, wastewater from rubber processing is generally treated by conventional pond systems such as aerated ponds, anaerobic ponds, and facultative ponds (Chaiprapat & Sdoodee 2007; Jawjit et al. 2010a, 2010b). However, these systems require long retention times and large treatment areas. In addition, these systems consume large amounts of power to aerate and discharge huge quantities of excess sludge, and they cause environmental pollution, such as emissions of greenhouse gases (GHG; e.g., methane, carbon dioxide), malodor, and groundwater pollution (Rattanapan et al. 2009; Jawjit et al. 2010a, 2010b; Smitha et al. 2012). Conversely, anaerobic reactor systems have some advantages such as the prevention of GHG emission, lower energy consumption, and energy recovery as methane (Saritpongteeraka & Chaiprapat 2008). However, aerobic wastewater treatment systems are still the preferred choice for CL processing factories in Thailand (Jawjit et al. 2015).

As a new process for treating wastewater from natural rubber processing, a combination of a two-stage up-flow anaerobic sludge blanket (UASB) and a down-flow hanging sponge (DHS) reactor was developed. The UASB–DHS system exhibited high treatment performance, low excess sludge production, low operational costs, and energy savings (Machdar et al. 2000; Tanduker et al. 2007). In the two-stage UASB, staged reaction of both sulfate reduction and methane production occurs. In the DHS, oxidation of residual organic carbon, sulfide, and ammonia takes place (Takahashi et al. 2011).

In this study, an on-site pilot-scale experiment of a combined two-stage UASB and DHS system was conducted in the Von Bundit natural rubber processing factory in Srat Thani, Thailand. The pilot plant was operated for a period of more than 300 days to evaluate the process performance. Furthermore, the energy consumption and GHG emissions were compared with those of the present wastewater treatment system in the factory.

MATERIALS AND METHODS

Two-stage UASB–DHS system

A schematic diagram of the two-stage UASB–DHS system is shown in Figure 1. The system consists of an acidification tank with a working volume of 1,180 L, a two-stage UASB reactor combined with 997 L of the first-stage UASB (first UASB) and 597 L of the second-stage UASB (second UASB), and a DHS reactor with a 195 L volume of sponge media. The UASB reactor and the DHS reactor were seeded with sludge obtained from an anaerobic pond and an aerated pond at the factory, respectively.
Figure 1

Schematic diagram of the two-staged system. P: pump; M: motor; AP: air pump.

Figure 1

Schematic diagram of the two-staged system. P: pump; M: motor; AP: air pump.

In the factory, three types of wastewater are discharged: the first type from CL production, the second type from RSS production, and the third type from standard Thai rubber (STR) production. Table 1 lists the chemical characteristics of CL, RSS, and STR wastewater. The CL wastewater is the highest-strength wastewater discharged from the factory. The CL wastewater was received in the acidification tank and was later fed to the first UASB as an influent. The pH of the influent was adjusted to 6.9–7.5 by addition of 2.5 N NaOH. Part of the UASB effluent obtained during the first UASB was returned to the influent at a recirculation ratio of 2:1. The operational period for CL wastewater treatment was divided into five periods (phases 1–5) according to the organic loading rate (OLR) and the hydraulic retention time (HRT). Table 2 shows the operating condition of each phase. In phase 1, OLR had increased by reduction of HRT from 4 days to 2 days for the first UASB. In phase 2, HRT of the first UASB was set to 4 days. In phases 3 and 5, OLR of the first UASB was controlled at 1.5 kgCOD/(m3.d) (COD: chemical oxygen demand) by dilution of influent wastewater. In phase 4, OLR of the first UASB had increased two-fold by reduction of HRT from 4 days to 2 days and stopping dilution of influent wastewater. After evaluation of the system performance of CL wastewater treatment, treatments of both RSS and STR wastewater were conducted using only the second UASB and DHS, respectively.

Table 1

Chemical characteristics of CL, RSS and STR wastewater

  Average ± SD*
 
Parameter Unit CL RSS STR 
pH – 5.54 ± 0.54 7.52 ± 0.42 7.84 ± 0.27 
Total COD   9,710 ± 2,600 1,360 ± 650 1,280 ± 770 
Soluble COD mg/L 8,220 ± 2,020 470 ± 222 284 ± 185 
Total biochemical oxygen demand   8,670 ± 2,750 423 ± 226 502 ± 232 
TSS   1,780 ± 1,260 651 ± 677 261 ± 116 
Sulfate mgS/L 1,430 ± 490 3 ± 8 4 ± 9 
Total nitrogen mgN/L 1,370 ± 480 233 ± 49 229 ± 33 
  Average ± SD*
 
Parameter Unit CL RSS STR 
pH – 5.54 ± 0.54 7.52 ± 0.42 7.84 ± 0.27 
Total COD   9,710 ± 2,600 1,360 ± 650 1,280 ± 770 
Soluble COD mg/L 8,220 ± 2,020 470 ± 222 284 ± 185 
Total biochemical oxygen demand   8,670 ± 2,750 423 ± 226 502 ± 232 
TSS   1,780 ± 1,260 651 ± 677 261 ± 116 
Sulfate mgS/L 1,430 ± 490 3 ± 8 4 ± 9 
Total nitrogen mgN/L 1,370 ± 480 233 ± 49 229 ± 33 

*Standard deviation.

Table 2

Operating condition of each phase

  HRT (days)
 
OLR (kgCOD/(m3.d))
 
Phase 1st UASB Whole system 1st UASB Whole system 
2.5 ± 0.9 4.5 ± 1.6 3.4 ± 1.3 1.9 ± 0.7 
3.9 ± 0.2 9.3 ± 1.7 2.3 ± 0.3 1.0 ± 0.2 
3.7 ± 0.5 10.3 ± 2.1 1.6 ± 0.2 0.6 ± 0.1 
2.0 ± 0.0 6.0 ± 0.5 3.5 ± 1.1 1.2 ± 0.4 
5.9 ± 0.6 11.5 ± 1.0 1.5 ± 0.1 0.8 ± 0.1 
  HRT (days)
 
OLR (kgCOD/(m3.d))
 
Phase 1st UASB Whole system 1st UASB Whole system 
2.5 ± 0.9 4.5 ± 1.6 3.4 ± 1.3 1.9 ± 0.7 
3.9 ± 0.2 9.3 ± 1.7 2.3 ± 0.3 1.0 ± 0.2 
3.7 ± 0.5 10.3 ± 2.1 1.6 ± 0.2 0.6 ± 0.1 
2.0 ± 0.0 6.0 ± 0.5 3.5 ± 1.1 1.2 ± 0.4 
5.9 ± 0.6 11.5 ± 1.0 1.5 ± 0.1 0.8 ± 0.1 

Average ± standard deviation.

Chemical analysis

Gas composition was measured using a gas chromatograph equipped with a thermal conductivity detector (GC-8A, Shimadzu). The solution pH was analyzed by using a pH meter (HM-20P, DKK). Sulfate, COD, total nitrogen, and sulfide were measured using a Hach colorimeter (DR890). Analysis of biochemical oxygen demand, total suspended solids (TSS), and volatile suspended solids was conducted according to Standard Methods (APHA 1998).

Microbial activity measurement

The methane-producing activity (MPA) and sulfate-reducing activity (SRA) of the retained sludge in the UASB reactor at 0 (as the seed), 87, 156, and 251 days were determined in duplicate. Test sludge was obtained from each stage of the UASB reactor at a vertical height of 0.5 m. The sludge was disintegrated under anaerobic conditions and was used for the activity tests. Sodium acetate (2,000 mgCOD/L) and H2/CO2 (80:20, v/v, 1.4 atm) were used as test substrates. All vials were incubated in a reciprocal shaker (radius = 4 cm) at 120 rpm and 35 °C. In the SRA test, sodium sulfate (200 mgS/L) and chloroform solution (5 mg/L) were added as an electron acceptor and a methanogenic inhibitor, respectively. Detailed procedures for the activity tests have been described in previous studies (Yamaguchi et al. 1997; Syutsubo et al. 2001).

Microbial community analysis

Sludge samples obtained from the first and second UASBs at 196 days were used for microbial community analysis. DNA was extracted from the washed sludge using an ISOIL Beads Beating Kit (Nippone gene), as described in the manufacturer's instructions. The extracted DNA was used for amplification of bacterial 16S rRNA gene fragments with primer pairs of EUB8F/1500r (Lane 1991; Ovreas et al. 1997). Bacterial 16S rRNA gene clone libraries were constructed using a TOPO TA cloning kit (Invitrogen). Classification and determination of the closest species of the obtained clones were performed using a classifier program of the Ribosomal Database Project (https://rdp.cme.msu.edu/).

Efficacy evaluation of the two-stage UASB–DHS system for wastewater treatment of a natural rubber processing factory

Figure 2 shows the flow of the currently used wastewater treatment system in the factory and the two-stage UASB–DHS system. Each wastewater stream was discharged from each production line and treated separately. The primary wastewater treatment system for both RSS wastewater and STR wastewater was anaerobic ponds. For high-strength CL wastewater treatment, aerated ponds were used as pretreatment. As a post-treatment, the combination of a facultative pond and a polishing pond was used for all types of wastewater. In order to evaluate the process performance, influent and effluent samples were obtained from each treatment step, and the water quality of the samples was analyzed. In addition, basic data regarding operational parameters such as power consumption, wastewater discharge volume, and excess sludge volume were collected to evaluate the performance of both the present treatment system and the two-stage UASB–DHS system. GHG emission from each wastewater treatment system was calculated using country-specific emission factors of electricity generation in Thailand (EGAT 2011; Krittayakasem et al. 2011), and a global warming potential of 25 was used for methane (IPCC 2006).
Figure 2

Flow chart for the present wastewater treatment system in the factory (upper) and the two-staged UASB–DHS system (lower).

Figure 2

Flow chart for the present wastewater treatment system in the factory (upper) and the two-staged UASB–DHS system (lower).

RESULTS AND DISCUSSION

Process performance of the two-stage UASB–DHS system for CL wastewater

Figure 3 shows the variation with time of the OLR and total COD removal of the first UASB and whole system. In phase 1, after increase of OLR by reduction of HRT (from 4 to 2 days for the first UASB), serious wash-out of retained sludge from both first and second UASB reactors to the DHS reactor occurred. As a result, COD removal in the first UASB sharply reduced to 25% at an OLR of 4.6 kgCOD/(m3.d). Consequently, COD removal of the whole system decreased to 60%.
Figure 3

Time course of (a) OLR and (b) total COD removal.

Figure 3

Time course of (a) OLR and (b) total COD removal.

In phase 2, the HRT of the first UASB returned to 4 days. As a result, the outflow of TSS from both the first and second UASB decreased, although the total COD removal efficiency in the first UASB was 30.9% ± 13.7%. However, the total COD removal efficiency of the whole system was 91.5% ± 10.8%. This shows that organic matter was sufficiently treated in the second UASB and the DHS. In phase 3, the OLR of the first UASB was controlled to 1.5 kgCOD/(m3.d) by dilution of influent wastewater. As a result, the total COD removal efficiency of the first UASB increased and stabilized to 72.6% ± 3.9% (95.7% ± 1.3% for the whole system). To evaluate the effect of shock loading, the OLR of the first UASB was increased to two times higher than phase 3 by reducing the HRT of the first UASB to 2 days, and stop of the wastewater dilution for 10 days (phase 4). As a consequence, the total COD removal efficiency of the first UASB dropped to approximately 50%. Subsequently, the OLR of the first UASB was retuned to 1.5 kgCOD/(m3.d) in phase 5. As a result, the total COD removal efficiency of the first UASB recovered to 76.2% ± 0.7% within 5 days. Some previous studies investigated the effect of shock loading on UASB performance. They reported that the performance of a mesophilic UASB exposed to 1.5–3.0 times higher shock loading condition for 4 days was able to recover within 6 days after returning the OLR to normal condition, by acclimation of sludge for more than 1 year in advance (Ramakrishnan & Gupta 2008; Sawaiker et al. 2012). These results suggest that the two-stage UASB–DHS system was able to resist shock loading due to acclimation of sludge by increasing the activity and concentration of the retained sludge.

Based on the experimental results described above, the maximum acceptable OLR of the first UASB is approximately 2.5 kgCOD/(m3.d) during phase 2 (i.e., ca. 1.0 kgCOD/(m3.d) for the whole system). The optimum OLR of the first UASB is approximately 1.5 kgCOD/(m3.d) (i.e., ca. 0.8 kgCOD/(m3.d) for the whole system) because a sufficient COD removal efficiency of 76.2% ± 0.7% was achieved in the first UASB during phase 5. In the present studies, the OLR and the COD removal efficiency of the anaerobic system treating CL wastewater ranged from 0.6 kgCOD/(m3.d) to 3.3 kgCOD/(m3.d) and from 62% to 83%, respectively (Boonsawang et al. 2008; Kanyarat & Sumate 2008; Kongjan et al. 2014). In terms of the OLR and the COD removal efficiency, the two-staged UASB–DHS system achieved similar or better performance than that of the other anaerobic treatment system for CL wastewater. Conversely, the OLRs of mesophilic UASB reactors treating non-sulfate-rich natural rubber processing wastewater were higher (Anotai et al. 2007; Jawjit et al. 2010a, 2010b; Hatamoto et al. 2012) because of the high concentration of sulfate in CL wastewater. As reported by Lens et al., the MPA of granular sludge was inhibited by a high concentration of hydrogen sulfide (Lens et al. 1998). In the first UASB, more than 90% of sulfate was consumed for organic removal (i.e., ca. 80% of removed COD) during phase 2, and consequently the generation and inhibition of hydrogen sulfide in the second UASB was reduced. Therefore, it is suggested that the two-stage UASB system achieved good performance in CL wastewater treatment by separation of the stages of organic removal by sulfate-reducing bacteria (SRB) and methane-producing bacteria (MPB).

Microbial activity of the retained sludge of the UASB reactor

Figure 4 shows the variation with time of MPA and SRA of the retained sludges of the two-stage UASB reactor. The MPA of both retained sludges in the UASB reactor increased during reactor operation except for H2/CO2-fed MPA (MPA-H2). After 196 days, acetate-fed methanogenic activity (MPA-Ac.) of the second UASB sludge became greater than that of the first UASB sludge. Here, the ratio of the acetate-fed activity of the second UASB sludge and first UASB sludge was 1.61 at day 251. This could have been caused by the low sulfate concentration in the influent in the second UASB. For the SRA of the UASB retained sludges, the H2/CO2-fed SRA (SRA-H2) was prominent for both stages. Acetate-fed SRA (SRA-Ac.) was minimal in the retained sludge of both the first and second UASB reactors. A slight increase in acetate-fed SRA was confirmed in the first UASB sludge after 196 days operation.
Figure 4

Time course of methane-producing and sulfate-reducing activities of the retained sludges of the first and second stages of the UASB reactor (N.D.: not detected).

Figure 4

Time course of methane-producing and sulfate-reducing activities of the retained sludges of the first and second stages of the UASB reactor (N.D.: not detected).

For an acetate substrate, MPA was higher than SRA for both UASB stages. Conversely, hydrogen-utilizing SRA was high compared to MPA after 196 days. At 251 days, hydrogen-fed activity was prominent in sulfate reduction, and the SRA/MPA ratios for the first and second UASB stages were 5.72 and 6.63, respectively. In addition, large numbers of clones related to the SRB families Desulfuromonadaceae, Desulfobacteraceae, and Desulfomicrobiaceae were detected in the retained sludge of both the first and second UASB by microbial community analysis targeting 16S rRNA genes. Hydrogen scavengers are indispensable for reduction of volatile fatty acids (VFA) such as butyrate and propionate to acetate under syntrophic association with VFA oxidizers (Imachi et al. 2000). Furthermore, SRB have a higher substrate affinity for hydrogen than MPB (Holmer & Kristensen 1994). These results indicated that SRB acted as hydrogen scavengers for reduction of VFA to acetate, and consequently MPB utilized acetate for methane production in both the first and second UASB.

During the shock-loading period, at day 251, all activities of the first UASB sludge were lower than the values at 196 days. Conversely, all activities of the second UASB sludge were increased, and the total COD removal efficiency of the whole system was maintained at more than 90% in spite of decreasing COD removal efficiency in the first UASB. These results indicated that the two-stage UASB system can resist shock loading conditions by maintaining at high MPB activity in the second UASB.

Efficacy evaluation of the two-stage UASB–DHS system for wastewater treatment of natural rubber processing wastewater

After evaluating the process performance for CL wastewater treatment, the two-stage UASB–DHS system was used to treat RSS and STR wastewaters. For the purposes of comparison, the assumption was made that the two-stage UASB–DHS system would treat all three wastewater streams. The first UASB (4 days of HRT) would treat the CL wastewater, the second UASB (16.8 hours of HRT) would be for both the RSS and STR wastewaters, and the DHS (4.8 hours of HRT) would provide post-treatment. Power consumption for operation of the two-stage UASB–DHS system was calculated from pump head for wastewater feeding to the first UASB and requisite amount of air supply for oxidation of organic matter, sulfate and nitrogen in DHS. As a condition for the calculation, up-flow velocity of the first UASB, dissolution efficiency of oxygen in DHS, efficiency of blower and total pressure of DHS were set as 5.0 m/hour (including recirculation water), 20%, 60% and 300 mmAq, respectively.

Figure 5(a) shows the power consumption of both wastewater treatment systems and Figure 5(b) shows GHG emission from both wastewater treatment systems. In the currently used wastewater treatment system in the factory, the CL wastewater treatment and post-treatment ponds consumed a large amount of power because of the aeration. In addition, a large amount of excess sludge was discharged from these aerated ponds. Furthermore, 19 t CO2e/d (CO2e: carbon dioxide equivalent) of methane gas was emitted from the anaerobic ponds. Emissions of methane gas were calculated from the COD removal in the anaerobic ponds. In case of this factory, GHG emissions from the production sector of natural rubber (including production of fresh latex, transportation of fresh latex to the mill, and production of primary rubber products) was estimated to be 71 tCO2/d using the CO2 emission factors for each rubber product (Jawjit et al. 2010a, 2010b). This result indicates that GHG emissions from wastewater treatment processes account for approximately 25% of total emissions from plantations and factories. On the other hand, the two-stage UASB–DHS system reduced the power consumption required for wastewater treatment by 95% and the excess sludge discharge by 98%. In addition, it is possible to prevent GHG emissions by recovering methane as an energy source from the two-stage UASB. These results prove that the two-stage UASB–DHS system is suitable for wastewater treatment in natural rubber processing factories.
Figure 5

(a) Power consumption of both wastewater treatment systems and (b) GHG emission from both wastewater treatment systems.

Figure 5

(a) Power consumption of both wastewater treatment systems and (b) GHG emission from both wastewater treatment systems.

In terms of the costs and benefits of operation, installation of the first UASB has effectively reduced the power consumption of CL wastewater treatment. Furthermore, installation of the second UASB has significantly reduced GHG emissions. Normally, biogas is used for thermal energy and electricity in Thailand (Chaiprasert 2011). If the recovered methane were to be converted to electricity, the recovered methane from the first and second UASBs could provide the electricity for operation of the two-stage UASB–DHS system, corresponding to 15% of the power consumed in the rubber production process. Installation of the DHS as an aerobic post-treatment can reduce both power consumption (by 97%) and excess sludge (by 98%). In developing countries, it is necessary to reduce the operating costs of wastewater treatment. Therefore, it is confirmed that installation of the first UASB for CL wastewater treatment and DHS for post-treatment is effective for wastewater treatment in natural rubber processing factories.

CONCLUSION

The main degraders of organic matter in the first and second UASBs were SRB and MPB, respectively. More than 80% of sulfate was removed in the first UASB, and sulfide inhibition for the second UASB was relieved. In this way, the second UASB maintained high COD removal efficiency and methanogenic activity. This finding suggests that the separation of UASB into two stages is effective for the treatment of sulfate-rich wastewater such as CL wastewater. Finally, the remaining organic matter and sulfide were oxidized by the DHS.

MPA was higher than SRA for acetate utilization in both UASB stages. On the other hand, the hydrogen-utilizing activity of SRB was higher than that of MPB. These results indicate that SRB acted as hydrogen scavengers for reduction of VFA to acetate under syntrophic conditions, and consequently acetate was utilized to produce methane by MPB promptly in both stages of the UASB.

Compared with the present wastewater treatment system in the factory, the two-stage UASB–DHS system achieved better process performance. In addition, the UASB–DHS system reduced the power consumption for the wastewater treatment by 95%, and the volume of excess sludge discharged from the aerated ponds by 98%. Therefore, the two-staged UASB–DHS system can be used as an appropriate system for wastewater treatment in natural rubber processing factories.

ACKNOWLEDGEMENTS

This research was performed by ESCANBER (Establishment of Carbon-Cycle-System with Natural Rubber) project supported by the JST/JICA, SATREPS (Science and Technology Research Partnership for Sustainable Development), Japan. We wish to thank Von Bundit Co., Ltd and their staff for installation and operation of the two-staged system.

REFERENCES

REFERENCES
APHA
1998
Standard Methods for the Examination of Water and Wastewater
,
20th edn.
American Public Health Association/American Water Works Association/Water Environment Federation
,
Washington, DC
,
USA
.
Boonsawang
P.
Laeh
S.
Intrasungkha
N.
2008
Enhancement of sludge granulation in anaerobic treatment of concentrated latex wastewater
.
Songklanakarin Journal of Science and Technology
30
,
111
119
.
Chaiprasert
P.
2011
Biogas production from agricultural wastes in Thailand
.
Journal of Sustainable Energy and Environment (Special Issue)
,
63
65
.
EGAT
2011
Annual Environmental Report
.
Electricity Generation Authority of Thailand (EGAT)
,
Bangkok
,
Thailand
.
Hatamoto
M.
Nagai
H.
Takahashi
M.
Kawakami
S.
Choeisai
P. K.
Syutsubo
K.
Ohashi
A.
Yamaguchi
T.
2012
Rubber and methane recovery from deproteinized natural rubber wastewater by coagulation pre-treatment and anaerobic treatment
.
International Journal of Environmental Research
6
,
577
584
.
Holmer
M.
Kristensen
E.
1994
Coexistence of sulfate reduction and methane production in an organic-rich sediment
.
Marine Ecology Progress Series
107
,
177
184
.
Imachi
H.
Sekiguchi
Y.
Kamagata
Y.
Ohashi
A.
Harada
H.
2000
Cultivation and in situ detection of thermophilic bacterium capable of oxidizing propionate in syntrophic association with hydrogenotrophic methanogens in a thermophilic methanogenic granular sludge
.
Applied and Environmental Microbiology
66
,
3608
3615
.
IPCC
2006
Wastewater treatment and discharge
. In:
2006 IPCC Guidelines for National Greenhouse Gas Inventories, 5
(
Eggleston
H. S.
Buendia
L.
Miwa
K.
Nagara
T.
Tanabe
K.
, eds).
IGES
,
Japan
,
6.24
6.27
.
Jawjit
S.
Jawjit
W.
Liengcharernsit
W.
2010a
Anaerobic mesophilic and thermophilic treatment of concentrated latex processing wastewater in two-stage upflow anaerobic sludge blanket (UASB)
.
Journal of International Environmental Application and Science
5
,
329
341
.
Jawjit
W.
Kroeze
C.
Rattanapan
S.
2010b
Greenhouse gas emissions from rubber industry in Thailand
.
Journal of Cleaner Production
18
,
403
411
.
Jawjit
W.
Pavasant
P.
Kroeze
C.
2015
Evaluating environmental performance of concentrated latex production in Thailand
.
Journal of Cleaner Production
98
,
84
91
.
Krittayakasem
P.
Patumsawad
S.
Garivait
S.
2011
Emission inventory of electricity generation in Thailand
.
Journal of Sustainable Energy and Environment
2
,
65
69
.
Lane
D. J.
1991
16S/23S rRNA sequencing
. In:
Nucleic Acid Techniques in Bacterial Systematics
(
Stackebrandt
E.
Goodfellow
M.
, eds).
Wiley
,
Chichester, UK
, pp.
115
175
.
Lens
P. N. L.
Visser
A.
Janssen
A. J.
Hulshof Pol
L. W.
Lettinga
G.
1998
Biological treatment of sulfate-rich wastewater
.
Critical Reviews in Environmental Science and Technology
28
,
41
88
.
Machdar
I.
Sekiguchi
Y.
Sumino
H.
Ohashi
A.
Harada
H.
2000
Combination of a UASB reactor and a curtain type DHS (downflow hanging sponge) reactor as a cost-effective sewage treatment system for developing countries
.
Water Science and Technology
42
,
83
88
.
Mohammadi
M.
Man
H. C.
Hassan
M. A.
Yee
P. L.
2010
Treatment of wastewater from rubber industry in Malaysia
.
African Journal of Biotechnology
9
,
6233
6243
.
Nugyen
N. H.
Luong
T. T.
2012
Situation of wastewater treatment of natural rubber latex processing in the Southeastern region, Vietnam
.
Journal of Vietnamese Environment
2
,
58
64
.
Overas
L.
Forney
L.
Dae
F. L.
1997
Distribution of bacterioplankton in meromictic lake
.
Applied and Environmental Microbiology
63
,
3367
3373
.
Sawaiker
U. G.
Guptha
K. G.
Asolekar
S. R.
2012
Effect of effluent recycling and shock load on the biodegradation of organic solvents in UASB reactor
.
Journal of Environmental Research and Development
6
(
3
),
554
561
.
Smitha
H. S. S.
Raghavendra
M. P.
Shruthi
S.
Girish
K.
2012
Bioremediation of rubber processing industry effluent by Arthrobacter sp
.
International Journal of Research in Environmental Science and Technology
2
,
31
34
.
Syutsubo
K.
Shinthurat
N.
Ohashi
A.
Harada
H.
2001
Population dynamics of anaerobic microbial consortia in thermophilic granular sludge in response to feed composition change
.
Water Science and Technology
43
,
59
66
.
Tekasakul
P.
Tekasakul
S.
2006
Environmental problems related to natural rubber production in Thailand
.
Journal of Aerosol Research
21
,
122
129
.
Thongnuekhang
V.
Puetpaiboon
U.
2004
Nitrogen removal from concentrated latex wastewater by land treatment
.
Songklanakarin Journal Science and Technology
26
,
521
528
.
Yamaguchi
T.
Harada
H.
Tseng
I. C.
1997
Competitive exclusion of methane-producing bacteria by sulfate-reducing bacteria in anaerobic degradation of long chain fatty acids
.
Proceeding of the 8th International Conference on Anaerobic Digestion
Sendai, Japan, Volume 2. International Water Association
, pp.
362
370
.