Sugar Industry Waste as Wastewater Treatment

Surendra Kumar, Omprakash Sahu

  Open Access OPEN ACCESS  Peer Reviewed PEER-REVIEWED

Sugar Industry Waste as Wastewater Treatment

Surendra Kumar1,, Omprakash Sahu1

1Department of Chemical Engineering NIT Raipur


Treatment of gound waste water by biological denitrification experiment was conducted using sugar industry wastes, namely final molasses as a carbon source and bagasse charcoal pellets as supporting media for denitrifying bacteria.We employed an upflow fixed-bed reactor filled with the pellets and biofilm attached onto them.This was fed with potassium-nitrate and dilute-molasses solutions.Total nitrogen removals of more than 85 % were achieved at influent carbon-nitrogenratios between 2 and 4, and hydraulic residence times of more than 0.8 (h).This demonstrated final molasses could be used as an alternative carbon source.On the other hand, final molasses also contained some organic/ammonium nitrogen and refractory organic matter including colors, both of which were difficult to remove with the reactor.Accordingly, at higher C/N ratios, these substances caused major increases in effluent total-nitrogen and organic-carbon concentrations.Therefore an optimum C/N ratio was found to be around 2.

At a glance: Figures

Cite this article:

  • Kumar, Surendra, and Omprakash Sahu. "Sugar Industry Waste as Wastewater Treatment." Journal of Environment Pollution and Human Health 1.2 (2013): 21-25.
  • Kumar, S. , & Sahu, O. (2013). Sugar Industry Waste as Wastewater Treatment. Journal of Environment Pollution and Human Health, 1(2), 21-25.
  • Kumar, Surendra, and Omprakash Sahu. "Sugar Industry Waste as Wastewater Treatment." Journal of Environment Pollution and Human Health 1, no. 2 (2013): 21-25.

Import into BibTeX Import into EndNote Import into RefMan Import into RefWorks

1. Introduction

Waste water treatment is one of the big issue no w days in which groundwater pollution by nitrate has become increasingly serious in remote Raipur, as the city rely most of drinking and irrigation waters on groundwater. Such nitrate pollution is mainly attributed to human activities including agriculture. In the Raipur city for example, nitrogen in groundwater was estimated to derive from fertilizers (49 %), livestock wastes (25 %), domestic sewage (5 %) and natural sources (19 %) [1]. Such human activities have already had effect on groundwater quality in those islands. Most of the islands have quite permeable soils that originated from coral reefs, and therefore the effects of human inputs of nitrogen could quickly appear in groundwater. In the India, for instance, NO3-N concentrations between 10 and 19 (mg/L) have occasionally been observed in groundwater. In order to remove those nitrates, the local authority has equipped their water purification plant with a reverse osmosis process since 1997.Such an advanced technology would be reliable but rather energy-consuming and costly. One of the alternatives to such a technology is biological denitrification using fixed or fluidized bed reactors. For enhancing the denitrification reaction, various chemicals were usually added as organic carbon sources, such as methanol, ethanol, acetic acid or sucrose [2]. Although many experiences have been accumulated regarding this method, it still needs costs for chemicals. In order to reduce such costs, biomass or biomaterials have also been tried out as a carbon source. Some of them were solid ones such as wheat straw [3, 4, 5], newspaper [6] and cotton [7]. Nevertheless, denitrification rates using those materials seemed to be rather slow for practical purposes, due in part to slow release of available carbon from them.

By contrast, some studies employed high-strength liquid organic wastes as carbon sources, including brewery wastes, whey, yeast, other bio-industry wastes and silage effluents [8, 9]. In some selected cases, they achieved denitrification rates comparable to the processes using methanol. Similarly some textbooks have also proposed final molasses, a sugar-industry waste, as a carbon source [10, 11, 12]. Nevertheless, few experiences have been accumulated so far on denitrification using final molasses.

The present study therefore examines final molasses as an alternative carbon source (Figure 1). It is a viscous, dark-brown liquid byproduct that is produced while raw cane juice is refined into white sugar.It is particularly rich in organic matter (Table 1), which includes sugars. In addition, the authors also employed charcoal pellets made from a mixture of bagasse and final molasses, as supporting media for denitrifying bacteria (Figure 1). Bagasse is solid, fibrous residue left after raw cane juice was extracted.

Table 1. Properties of the materials used in this study

Thus, this study aims to employ final molasses and bagasse, both of which are byproducts from sugarcane processing, for a denitrification experiment (Figure 1). These materials were chosen because they were easily obtainable in Okinawa, where a lot of sugarcane factories are located for processing locally-produced sugarcane.

2. Materials and Methods

2.1. Preparation of Bagasse Charcoal Pellets and Apparatus

Bagasse and final molasses were obtained from Bhoramdev Suagr Indutries Ltd. Kawardha, India (Table 1). A 2.75 (L) (7 (kg)) of the bagasse was carbonized at 800 (C) for 2 (h) with a furnace DCP-31.Upon carbonization, oxygen within the furnace was purged with nitrogen gas.

Thereafter the volume of produced charcoal became about a half that of the raw bagasse.The charcoal was crushed to powders by hand, and then mixed with 500 (g) of final molasses and 2.5 (L) of tap water.The final molasses was supposed to act as binder in producing pellets in the following steps.

The above mixture was formed into pellets with a press pelletizer FMP-300.At this stage, the pellets were not hard enough for use in water treatment.

Therefore the pellets were again carbonized at 800 (C) for 2 (h) to increase their hardness. Finally, pellets with the following properties were obtained: a volume of 2.7 (L); a weight of 1.0 (kg); a diameter of 5 (mm); and a length of 5 to 10 (mm) (see also Table 1).

Meanwhile an upflow column was made with a transparent plastic pipe (Figure 2). This was filled with the bagasse-charcoal pellets at a depth of 80 (cm), and placed in a constant-temperature chamber.

2.2. Inoculation of Denitrifying Bacteria

A 12 (g) of final molasses and 17 (g) of potassium nitrate were dissolved with 20 (L) of tap water. About 10 (g) of surface soil sampled at a sugarcane farm in the Raipur city was added to this solution for seeding it with denitrifying bacteria.

The above solution was pumped into the upflow column, at a hydraulic residence time (HRT) of 6 (h). The effluent from the column was recirculated into the column. During this operation, appropriate amounts of final molasses and potassium nitrate were occasionally added to the solution in order to sustain bacterial growth.

The above operation was continued for 35 days. Consequently thin biofilm appeared on the surface of pellets in the column. At this stage, a reduction in nitrate concentration was also confirmed in the column, indicating that the biofilm would contain some denitrifying bacteria.

2.3. Continuous Denitrification Experiment

Next a continuous denitrification experiment was conducted for 77 days. The column was washed with air bubbles roughly once or twice a week throughout this experiment. Two kinds of substrate solutions, namely potassium nitrate and dilute final molasses, were prepared and stored in separate tanks (Figure 2), and injected to the column at a flow-rate ratio of 5:1. The influent to the column was sampled after these two solutions were injected and mixed (Figure 2). The influent NO3-N concentration was maintained around 20 (mg/L), which roughly corresponded to the highest groundwater concentration observed in the Raipur city. Meanwhile two parameters were varied in this experiment, namely the concentration of final molasses solution and the flow rate. This gave influent TOC concentrations ranging between 11 and 175 (mg/L), column flow rates between 5.3 and 58.1 (m/d), and hydraulic residence times (HRT) between 0.33 and 3.40 (h).

The experiment was conducted with a temperature controlled at 25±1 (C), and ambient pH and ORP in a range of 6.6-7.3 and 31-301 (mV), respectively.

2.4. Analytical Methods

Total organic carbon (TOC) of water samples was analyzed with TOC-5000, total nitrogen (T-N) with TN-301P, and inorganic ions with an ion chromatograph DX-320J. A total concentration of NO2-N and NO3-N was indicated as NOx-N, but NO3-N was a dominant form in all the samples. Total and suspended solids were measured according to the Standard Methods for the Examination of Water and Wastewater 2540 B and D, respectively [13]. Apparent color was measured with non-filtered samples, using a spectrophotometer. Thus, possible effects of light scattering by suspended solids were not eliminated in these measurements. The wavelength was set at 390 nm, which was designated for color measurement in the Indian Drinking Water Examination Methods. In order to convert absorbance values to color units, a calibration curve was prepared using a platinum-cobalt solution as described in the Standard Methods for the Examination of Water and Wastewater 2120 B [13].

3. Results and Discussions

3.1. Nitrogen Removal
Figure 3. Effect of influent TOC/T-N ratio on nitrogen removal and concentration

The Figure 3 (a) shows a relationship between the T-N removal and the influent TOC/T-N ratio (C/N ratio). The C/N ratio, expressed on a weight basis, was roughly propotional to the amount of final mollasses injected into the column, as influent NOx-N concentrations were kept constant. In this and the subsequent figures, the data were categorized into 4 groups by the HRT. When the HRT was between 0.3 and 0.6 (h), the T-N removal stayed below 60 % (Figure 3 (a)), although slightly better performances were achieved at higher C/N ratios. This indicated the HRT was too short for completing denitrification reactions regardless of the C/N ratio. On the other hand, when the HRT was 0.8 (h) or longer, the T-N removal largely followed a similar trend, with the maximum removal of above 85 % achieved at C/N ratios between 2 and 4 (Figure 3 (a)). Accordingly the effluent T-N (Figure 3 (b)) showed the lowest concentrations at the same range of C/N ratios. By contrast, the effluent NOx-N (Figure 3 (c)) kept low concentrations at C/N ratios of 2 or more. A difference between the effluent T-N and NOx-N concentrations was particularly significant at C/N ratios of more than 4 (Figure 3 (b) and Figure 3 (c)).

Figure 4. Effect of influent TOC/T-N ratio on TOC removal and concentration

These results suggested that, at C/N ratios of less than 2, carbon source was not enough for denitrification, and a lot of NOx-N remained in the effluent. By contrast, at C/N ratios of more than 2, carbon source seemed to be sufficient for denitrification to maintain low effluent NOx-N concentrations (Figure 3 (c)). At the same time, however, increasing the C/N ratio (i.e. influent molasses concentration) also meant the increase in influent organic/ammonium nitrogen that derived from final molasses (Table 1). Because the column was under an anaerobic condition, the influent organic/ammonium nitrogen would largely be carried away to the effluent, except for a small amount assimilated by biofilm in the column. This might explain the increase in effluent T-N concentrations at C/N ratio of more than 4 (Figure 3 (b)).

3.2. Organic Carbon and Color Removal

Unlike T-N removal, TOC removal showed no clear trend in terms of the C/N ratio or the HRT (Figure 4 (a)), and stayed below 65 % in every condition. Accordingly the effluent TOC concentration increased proportionally to the C/N ratio (Figure 4 (b)). This implied that final molasses would contain some refractory organic matter.

Final molasses is known to contain various forms of sugars as well as colored materials such as melanoidine, caramel and polyphenolic compounds [14]. Also in this experiment, a slight yellowish color was increasingly visible both in the influent and effluent, as the influent molasses concentration was increased. This was confirmed by spectrophotometric measurements of apparent colors (Figure 5). On average, the color removal remained only 17 %, which was significantly lower than the TOC removal of 50 %. The difference was statistically significant at level 0.01 (P-value according to the two-sided t-test was 2.710-18). This implied that colored materials were not readily consumed by the biofilm and tended to remain in the effluent.

Figure 5. Relationship between influent TOC/T-N ratio and effluent apparent color
3.3. Effect of Influent TOC on Nitrate Removal Rates

In order to examine the effect of influent TOC on nitrate removal rates, observed nitrate removal rates were simulated by a Monod model (Figure 6) using the following equations. For this simulation, data at HRT of 0.3 to 0.6 (h) were excluded because of the incomplete denitrification.



rv is volumetric nitrate removal rate (g m-3 d-1),

k is a maximum denitrification rate constant (g m-3 d-1),

Sc is an influent TOC concentration (mg/L),

Ks is a half-saturation constant (mg/L),

SN0 is an influent NOx-N concentration (mg/L),

SN is an effluent NOx-N concentration (mg/L),

Q is a flow rate (m3/d), and

V is a volume of the column (m3).

Figure 6. Effect of influent TOC concentration on nitrate removal rate (Plots show observed values while lines indicate simulated values with a Monod model)

Table 2. Simulated parameters by the Monod model

The simulation showed that maximum denitrification rate constants were in a range of 195 to 479 (g m-3 d-1), with the largest one achieved at HRT of 0.8-1.0 (h) (Table 2). This would be comparable to reported denitrification rates using methanol, between 290 and 1600 (g m-3 d-1) for downflow packed-bed reactors (USEPA, 1993), but considerably smaller than the rates between 2000 and 6400 (g m-3 d-1) for upflow fluidized-bed reactors (USEPA, 1993). USEPA (1993) attributed the faster denitrification rates of fluidized-bed reactors in comparison to fixed-bed reactors, to a larger surface area of media per unit reactor volume provided by fluidized-bed reactors. Therefore future studies should investigate fluidized-bed reactors for improving denitrification rates using final molasses. Meanwhile the half-saturation constants were found to be between 6.7 and 21.9 (mg/L) (Table 2), which were larger than reported general values for methanol between 0.1 and 6.0 (mg/L) (USEPA, 1993). This means that larger amounts of organic carbon need to remain in the effluent in order to achieve sufficient denitrification, when compared with methanol. It might also be another evidence that final mollase contains some refractory organic matter.

4. Conclusions

This study demonstrated that final molasses could be used as a carbon source for biological denitrification. At least an influent C/N ratio of 2 and HRT of 0.8-1.0 (h) were required to achieve enough denitrification. On the other hand, final molasses contained some organic/ammonium nitrogen and refractory organic matter including colors, which were difficult to remove with the reactor. Accordingly, at higher C/N ratios, those substances caused major increase in effluent T-N and TOC concentrations. This would discourage excessive input of final molasses to the influent. Therefore, an optimum C/N ratio for denitrification using final molasses was found to be around 2, which was comparable to a usual C/N ratio of 3 for methanol and various biological wastes. However, even if the C/N ratio is kept around the above optimum, the effluent would still contain some residual substances including organic/ammonium nitrogen, phosphorous and refractory organic matter. Therefore, when one attempts to apply this study to water treatment, those substances must be removed at downstream processes such as coagulation and sand filtration. When applied to tertiary wastewater treatment, the effluent might as well be reused as irrigation water, because farm soils might be effective in degrading or absorbing those substances as the effluent passed through them. Anyway further studies are required to investigate these issues.


[1]  Nakanishi, Y. and Ikema, M., Origins of nitrogen in drinking water and estimated capacity of nitrogen removal by forest in Miyako Iskand, Okinawa, Japanese J. of Soil Science and Plant Nutrition, 72 (3): 372-378, 2001.
In article      
[2]  Kapoor, A. and Viraraghavan, T., Nitrate removal from drinking water-review, J. Environ. Eng., ASCE, 123 (4): 371-380, 2007.
In article      CrossRef
[3]  Avnimelech, Y., Diab, S. and Kochba, M., Development and evaluation of a biofilter for turbid and nitrogen rich irrigation water, Wat. Res., 27 (5):785-790, 2003.
In article      CrossRef
[4]  Soares, M.I.M. and Abeliovich, A., Wheat straw as substrate for water denitrification, Wat. Res., 32 (12):3790-3794, 2008.
In article      CrossRef
[5]  Aslan, Ş. and Türkman, A., Biological denitrification of drinking water using various natural organic solid substrates, Wat. Sci. Tech.,48 (11-12): 489-495, 2003.
In article      
[6]  Volokita, M., Belkin, S., Abeliovich, A. and Soares, M.I.M., Biological denitrification of drinking water using newspaper, Wat. Res., 30 (4): 965-971, 2006a.
In article      CrossRef
[7]  Volokita, M., Abeliovich, A. and Soares, M.I.M. Denitrification of groundwater using cotton as energy source, Wat. Sci. Tech., 34 (1-2):379-385, 2006b.
In article      CrossRef
[8]  Monteith, H.D., Bridle, T.R. and Sutton, P.M., Industrial waste carbon sources for biological denitrification, Prog. Wat. Tech., 12:127-141, 2012.
In article      
[9]  Skrinde, J.R. and Bhagat, S.K., Industrial wastes as carbon sources in biological denitrification, J. WPCF, 54 (4): 370-377, 2010.
In article      
[10]  Eckenfelder, W.W., Industrial water pollution control, 2nd edition, McGraw-Hill Book, New York, 178, 1989.
In article      
[11]  Horan, N.J., Biological wastewater treatment systems, theory and operation, John Wiley & Sons, Chichester, England, p.226, 1990.
In article      
[12]  USEPA Manual: nitrogen control, EPA/625/R-93/010, Office of Water, United States Environmental Protection Agency, Washington DC, 211, 1993.
In article      
[13]  APHA, AWWA and WEF Standard Methods for the Examination of Water and Wastewater, 20th edition, American Public Health Association, American Water Works Association and Water Environment Federation, Washington, DC, 1998.
In article      
[14]  Nagano, A., Nakamoto, C. and Suzuki, M. Decolorization treatment by electrolysis of molasses wastewater –characteristics of colored components and effects of electrolytes, J. Japan Soc. Wat. Environ, 23 (1): 34-40, 2000.
In article      CrossRef
  • CiteULikeCiteULike
  • Digg ThisDigg
  • MendeleyMendeley
  • RedditReddit
  • Google+Google+
  • StumbleUponStumbleUpon
  • Add to DeliciousDelicious
  • FacebookFacebook
  • TwitterTwitter
  • LinkedInLinkedIn