This is the author s version of a work that was submitted/accepted for publication in the following source:

This is the author s version of a work that was submitted/accepted for publication in the following source: Kent, Geoffrey Alan (2013) Issues associated with using trash as a cogeneration fuel. In Hogarth, D.M. (Ed.) Proceedings of the International Society of Sugar Technologists, Sociedade dos Tecnicos Acucareiros e Alcooleiros do Brasil & The XXVIIIth ISSCT Organising Committee, Sao Paulo, Brazil, pp. 1759-1770. This file was downloaded from: http://eprints.qut.edu.au/77913/ c Copyright 2013 STAB & The XXVIIIth ISSCT Organising Committee Notice: Changes introduced as a result of publishing processes such as copy-editing and formatting may not be reflected in this document. For a definitive version of this work, please refer to the published source: http://www.issct.org/proceedings.html

FP3 ISSUES ASSOCIATED WITH USING TRASH AS A COGENERATION FUEL By G.A. KENT Queensland University of Technology, Brisbane g.kent@qut.edu.au KEYWORDS: Trash, whole crop, harvest, transport, process, recovery Abstract Considerable work has been undertaken to determine an economical process to provide sugarcane trash as a fuel for cogeneration. This paper reviews efforts to provide that trash fuel by harvesting, transporting and processing the trash with the cane. Harvesting trash with the cane has the advantage that cane that would otherwise be lost by extracting it with the trash is captured and sugar can be produced from that cane. Transporting trash with the cane significantly reduces the bulk density of the cane, requiring substantial changes and costs to cane transport. Shredding the trash at the harvester and compacting the cane in the bin prior to transport are possible methods to increase the bulk density but both have considerable cost. Processing trash through the sugar factory with the cane significantly reduces sugar recovery and sugar quality. Although considerable knowledge has been gained of these effects and further analysis has provided insights into their causes, much more work is required before whole crop harvesting and transport is an economically viable means of trash recovery. Introduction The concept of using trash as a fuel for cogeneration of electricity is one that has received a lot of attention in recent years. Botha and Van Den Berg (2009), Bocci et al. (2009) and Jais (2010) investigated scenarios where trash was transported to the factory and used for cogeneration. The purpose of the trash recovery is to maximise the amount of biomass available so that the cogeneration season can be extended as long as possible (Doolan and Lamb, 2009). More recently, there has been considerable discussion about the use of bagasse and trash for the production of second-generation biofuels (Dias et al., 2012 for example). In most of these studies, trash is considered as a substitute fuel for the boilers for steam and electricity generation so that more bagasse can be made available for biofuels production. This paper attempts to define the main issues associated with making trash available as a fuel for cogeneration and reviews recent work aimed at addressing those issues. In order to limit the scope, this paper focuses on whole of crop harvesting and transport, this being the pathway for which considerable research has been undertaken in Australia.

Leaving trash in the field Before taking trash from the field for use as a fuel for cogeneration, consideration needs to be given to its value in the field. The practice of burning cane before harvest is becoming less common as a result of environmental pressures (Dias et al., 2012). The burning of cane consumes much of the trash (Mitchell et al., 2000), significantly reducing its availability for cogeneration. There have been many studies reporting on the advantages of leaving trash in the field. Fortes et al. (2012) argued that trash left in the field plays an important role in preserving the fertility and sustainability of the soil. They reported that trash is an important source of carbon and nutrients (potassium, calcium and nitrogen) to the soil-plant system. Olivier and Singles (2012) reported that trash reduces evaporation from the soil surface. Several models have been developed to predict the benefit of leaving trash in the field (Thorburn et al., 2005; Purchase et al., 2008). Manechini et al. (2005) reported an attempt to experimentally determine the amount of trash to be left in the field. Of the 6.7 t/ha to 14.9 t/h total trash in the field, they found that 7.5 t/ha to 9.0 t/ha was required for weed control and that the amount of trash required to preserve yield varied considerably with cane variety, climate and pests. Trash pathways There are several variations to the process of making trash available for cogeneration that have a significant impact on the cost of providing the trash fuel. The main variations are presented in Figure 1. Trash can either be harvested with the cane or collected after the cane has been harvested. If trash is harvested with the cane, the trash can either be separated from the cane at a later stage (usually the factory) or processed through the factory with the cane. If the trash is processed separately to the cane, it can either be provided as a trash only fuel or mixed with bagasse.

Fig. 1 Process pathways in the preparation of trash as fuel for cogeneration The pathway being reviewed in this paper is the furthermost left pathway, involving harvesting, transport and processing of the trash with the cane to produce a bagasse and trash feed for cogeneration. Harvesting Mechanical harvesting of trash with the cane can be achieved by reducing the speed of the harvester extractor fans (Kent et al. 2010). Reducing the speed of the extractor fans has a benefit of reducing cane loss (Whiteing et al., 2001; Kent et al., 2003). Perhaps the most consistent data relating cane loss to trash content was reported by Linedale et al. (1993). Those data have been presented graphically in Figure 2. Of the six sets of data presented in Figure 2, all cases involved increasing the trash content by about three units and all cases saw the cane loss reduce to close to zero. The data suggest that cane loss can be effectively minimised by allowing an additional three units of trash into the cane supply. Reducing the extractor fan speed further to increase trash levels results in minimal additional recovery of cane.

16 loss (t/ha) 14 12 10 8 6 4 Ingham green Mossman green Mulgrave green Proserpine green Proserpine burnt Bundaberg green 2 0 2 4 6 8 10 12 14 16 18 Trash content (% cane) Fig. 2 Effect of trash content on cane loss (data from Linedale et al., 2003) Transport It is well established that higher trash contents in the cane supply reduce the bulk density of the cane and hence reduce the mass of cane that can be transported each trip. Bernhardt et al. (2000) and Eggleston et al. (2012a), for example, provided evidence of a reduction in payload with an increase in trash. One of the difficulties with payload measurements is ensuring that the same volume is occupied for each mass measurement. For the burnt cane and whole crop measurements discussed by McGuire et al. (2011) for example, mass limits enforced by the local road authority prevented the full volume being occupied for their low trash results. Similar volumes were occupied for the data reported by Kent et al. (2003) and those data are presented graphically in Figure 3.

9.5 9.0 8.5 Payload (t) 8.0 7.5 7.0 6.5 6.0 0 2 4 6 8 10 Trash content (%) Fig. 3 Effect of trash content on payload during transport (data from Kent et al., 2003) Considerable research has been undertaken to try and increase the bulk density and hence the payload for cane supplies with high trash content. Barnes et al. (2009) aimed to take advantage of the expectation that cane of shorter billet length has a higher bulk density (Foster et al. 1977). They considered a different chopping system to that of a conventional cane harvester to enable billet length to be varied over a greater range and tested the new design over a range of parameters to determine its effect on cane and juice loss during chopping. While a comparison against a conventional cane harvest system found reasonably similar losses at a billet length of 200 mm, it was expected, based on work by Hockings et al. (2000), that the new system would perform much better relative to the conventional system at shorter billet lengths and as the chopper blades wore. Spinaze et al. (2002) and Hassuani et al. (2005) reported on shredder fan designs to be incorporated in harvester primary extractors for the purpose of separating trash from cane, shredding it to improve its bulk density and then depositing it into a second in-field transporter for separate transport of the trash. Inderbitzin and Beattie (2012) investigated a range of options for increasing bulk density including shredding the trash (but depositing it back with the cane instead of separating it), reducing the billet length, compacting the cane in the bin, vibrating the bin and topping the cane during harvest. They reported a target bulk density of 250 kg/m 3. They distinguished between the in-field haulage task of taking cane from the harvester to the road transport pad and the road transport task of taking the cane from the road transport pad to the factory. The increase in bulk density achieved by the various systems is summarised in Table 1.

Table 1 Increase in bulk density achieved by Inderbitzin and Beattie (2012) System Bulk density Comment improvement (%) Shredder fan 14 Tests focussed on minimising pol loss Billet length 13 loss cost estimated to be greater than transport saving Compaction Large Improvement dependent on pressure applied. Benefit only for road transport (not in-field haulage) Vibration 11 Tests on in-field haulage Topping 7 Reduced extraneous matter by 20% Inderbitzin and Beattie (2012) presented the results of a financial analysis to assess the overall cost-effectiveness of each strategy and found the lowest costs were associated with the shredder fan, compaction or a combination of the two approaches. Processing with the cane Past work The low bulk density or high volume problem for transport of trash with cane also has capacity implications at the cane unloading station while the associated flow issues also affect the conveying of the cane to the cane preparation station. McGuire et al. (2011) examined the effect of trash on crop yield by comparing burnt cane harvesting with trash extracted against whole crop harvesting. The results showed an increase in crop yield of 24% with an increase in dry, ash-free fibre yield of 40%. Assuming no increase in season length, these higher rates have to be processed through the extraction station of the factory requiring an increase in capacity of cane preparation and extraction equipment. There has been considerable work to determine the effect of trash on sugar production and sugar quality. Kent (2007) provided a comprehensive review of factory-based research to determine the effect of trash. Kent et al. (2010) updated that review and provided further results from long (three-day) tests with different trash levels so that sugar production and sugar quality effects would be measured in the factory. Further work has since been reported by Eggleston et al. (2012b). The studies consistently show reductions in sugar recovery and sugar quality with greater levels of trash. 2010 experiment Following the methodology of Kent et al. (2010), a further experiment was conducted at Condong factory in Australia, investigating the effects of two cane supply types: burnt cane with trash extracted (BE cane) and green cane with half of the trash extracted (GE0.5 cane). An experiment consisting of four tests was designed, incorporating a four week period processing GE0.5 cane that had been previously arranged at Condong factory. Each test was conducted over two weeks. The tests, along with their measurement period, are presented in Table 2. Each test commenced and concluded at 08:00 on the dates shown.

Table 2 List of tests Test supply type Start End 1 BE 18 Jul 2010 01 Aug 2010 2 GE0.5 08 Aug 2010 22 Aug 2010 3 GE0.5 22 Aug 2010 05 Sep 2010 4 BE 12 Sep 2010 26 Sep 2010 The success of the experiment was reliant on the district harvest contractors supplying cane to the required specification. No extraneous matter analysis was conducted during the experiments. The success was judged through the cane fibre measurements. Figure 4 contains box plots showing the spread of results for each cane supply type during each experiment. The top and bottom of each box marks the minimum and maximum measured value and the horizontal white line through each box represents the median value. The figures show that the range of average fibre contents for each cane supply type do not overlap, indicating distinctly different cane supplies associated with each cane supply type. Fibre content (%) 15.0 16.0 17.0 18.0 BE GE0.5 Fig. 4 Average fibre content for each cane supply type Pol losses in bagasse, mud and molasses were individually measured, along with overall recovery. The results are shown in Figure 5. These results show increases in pol loss in bagasse, mud and molasses. While the differences are not all statistically significant (difficult to achieve with only four tests), they are consistent with the results reported by Kent et al. (2010) and are hence quite convincing.

Retents (%) 3.0 3.5 4.0 Pol in mud (% pol in cane) 0.6 0.8 1.0 1.2 1.4 1.6 BE GE0.5 BE GE0.5 Pol in molasses (% pol in cane) 9 10 11 12 Recovery (%) 83 84 85 86 87 88 BE GE0.5 BE GE0.5 Fig. 5 Sugar recovery results The three main sugar quality parameters for which possible effects of cane type were identified ash, filterability and colour are shown in Figure 6. While the ash and filterability results were consistent with the results of Kent et al. (2010), the ash results were not. It is noted that the sugar from three of the four tests had an ash content of 0.13% while the fourth, from a burnt cane test, had a higher ash content.

Ash (%) 0.130 0.134 0.138 BE GE0.5 Filterability (%) 70 72 74 76 78 80 82 BE GE0.5 Colour (IU) 1200 1250 1300 1350 BE GE0.5 Fig. 6 Sugar quality results Overall effects To summarise the sugar production effects of increasing trash content, taking into account the reduced cane loss during harvest as discussed earlier, the effect of trash on the cane pol yield and on the produced sugar pol yield were determined, based mainly on the results provided by McGuire et al. (2011), Linedale et al. (1993), Kent et al. (2010) and the 2010 results presented in this paper. Figure 7 presents that summary. It is noted that, as the trash content initially starts to increase, there is an expected increase in the cane pol yield, corresponding to a reduction in cane loss. Once that cane loss has been eliminated, there is no further increase in pol yield. Although trash has been found to contain pol, that pol is presumed to be a result of juice sprayed onto the trash during the harvesting operation. Tests reported by McGuire et al. (2011) found that the pol on trash deteriorates quickly and is lost within typical cut to crush delays. Considering the sugar pol yield, the reduced cane loss largely compensates for the reduced recovery, indicating that increasing the trash in the cane supply by several units may be reasonably cost-effective, providing extra fuel for cogeneration without impacting significantly on sugar production.

11.0 10.5 Sugar Pol yield (t/ha) 10.0 9.5 9.0 8.5 5 10 15 20 25 30 35 trash content (%) Fig. 7 Predicted effect of trash on cane pol and sugar pol yield Kent et al. (2010) and Figure 6 also confirmed a reduction in sugar quality with increasing trash levels. The two parameters for which sugar quality reductions were consistently measured were filterability and colour. Figure 8 and Figure 9 show the results of the three sets of factory experiments where filterability and colour respectively were measured. In both sets of tests, cane fibre content was measured as an indicator of trash content.

Broadwater 2009 Condong 2009 80 80 70 70 Filterability (%) 60 50 Filterability (%) 60 50 40 40 30 16 18 20 22 30 16 18 20 22 fibre content (%) fibre content (%) Condong 2010 80 70 Filterability (%) 60 50 40 30 16 18 20 22 fibre content (%) Fig. 8 The measured effect of trash on sugar filterability

Broadwater 2009 Condong 2009 2400 2400 2200 2200 2000 2000 Colour (IU) 1800 1600 Colour (IU) 1800 1600 1400 1400 1200 1200 16 18 20 22 fibre content (%) 16 18 20 22 fibre content (%) Condong 2010 2400 2200 2000 Colour (IU) 1800 1600 1400 1200 fibre content (%) Fig. 9 The measured effect of trash on sugar colour 16 18 20 22 In addition to the overall effects on sugar production and sugar quality, processing of juice from cane with high trash contents has proven to cause significant clarification and pan boiling problems (Moller et al., 2010). Some of these problems may have resulted from specific factory limitations at Broadwater factory (including limited clarification and filtration capacity) and these limitations may also have affected the results presented in Figure 8 and Figure 9. Further work has been undertaken in an effort to understand the effect of trash on the clarification of juice in an effort to reduce processing difficulties and improve recovery and quality (Thai et al., 2012; Thai and Doherty, 2012). In particular, slow settling of flocculated mud impurities has been identified with juice from high trash cane supplies, caused by high levels of proteins, organic acids and polysaccharides.

Conclusion Harvesting, transporting and processing trash with the cane provides the advantage of minimising the number of process steps involved in trash recovery but there are substantial costs incurred in doing so. The most economical pathway to providing trash as a cogeneration fuel has not been identified. Harvesting trash with the cane has the advantage that cane that would otherwise be lost by extracting it with the trash is captured and sugar can be produced from that cane. Increasing the trash content in the cane supply by about three units by reducing the harvester extractor fan speed is most likely sufficient to recover this cane. It is likely that the increased sugar production from the recovered cane is sufficient to balance the sugar production inhibiting characteristics of the three units of trash. Transporting trash with the cane significantly reduces the bulk density of the cane, requiring substantial changes and costs to cane transport. Shredding of the trash at the harvester and compacting the cane in the bin prior to transport are possible methods to increase the bulk density but both have considerable cost. Processing trash through the sugar factory with the cane significantly reduces sugar recovery and sugar quality. Although considerable knowledge has been gained of these effects and further analysis has provided insights into their causes, much more work is required before whole crop harvesting and transport is an economically viable means of trash recovery. Acknowledgements The author particularly acknowledges the efforts undertaken by the New South Wales sugar industry, particularly New South Wales Sugar Milling Co-operative Limited, in initiating and conducting much of the research discussed in this paper. The author also acknowledges the Sugar Research and Development Corporation for funding much of the experimental work discussed in this paper, most notably the 2010 experiment. REFERENCES Barnes, M.G., Loughran, J.G., Whiteing, C. and Lamb, B.W. (2009). Development and testing of a sugarcane harvester single drum chopper system. Proc. Aust. Soc. Sugar Technol., 31: 546-555. Bernhardt, H.W., Pillay, V. and Simpson, A. (2000). Impacts of greencane harvesting on sugar factory operation at Sezela. Proc. S. Afr. Sug. Technol. Ass., 74: 369-372. Bocci, E., Di Carlo, A. and Marcelo, D. (2009). Power plant perspectives for sugarcane mills. Energy, 34: 689-698. Botha, D.H. and Van Den Berg, M. (2009). Linking agronomic and economic models to investigate farm-level profitability under a bioenergy-oriented sugar industry in South Africa. Proc. S. Afr. Sug. Technol. Ass., 82: 67-82. Dias, M.O.S., Junqueira, T.L., Jesus, C.D.F., Rossell, C.E.V., Maciel Filho, R. and Bonomi, A. (2012). Improving second generation ethanol production through optimization of first generation production process from sugarcane. Energy, 43: 246-252.

Doolan, G.T. and Lamb, B.W. (2009). Whole-crop cane transportation: introduction and maintenance of lightweight high volume aluminium bins. Proc. Aust. Soc. Sugar Technol., 31: 556-567. Eggleston, G., Birkett, H., Gay, J., Legendre, B., Jackson, W., Schudmak, C., Monge, A., Andrzejewski, B., Viator, R. and Charlet, T. (2012a). How combine harvesting of green cane billets with different levels of trash affects production and processing. Part 1: Field yields and delivered cane quality. Int. Sugar J., 114: 91-98. Eggleston, G., Birkett, H., Gay, J., Legendre, B., Jackson, W., Schudmak, C., Monge, A., Andrzejewski, B., Viator, R. and Charlet, T. (2012b). How combine harvesting of green cane billets with different levels of trash affects production and processing. Part 2: Pilot plant processing to sugar. Int. Sugar J., 114: 169-178. Fortes, C., Ocheuze Trivelin, P.C. and Vitti, A.C. (2012). Long-term decomposition of sugarcane harvest residues in Sao Paulo state, Brazil. Biomass and Bioenergy, 42: 189-198. Foster, D.H., Cullen, R.N., James, R.A., Mason, V. and Meng, K.J. (1977). Harvesting trials in green and burnt cane an initial investigation. Proc. Qd. Soc. Sugar Technol., 44: 29-36. Hassuani, S.J., da Silva, J.E.A.R. and Neves, J.L.M. (2005). Sugarcane trash recovery alternatives for power generation. Proc. Int. Soc. Sugar Technol., 25: 394-402. Hockings, P.R., Norris, C.P. and Davis, R.J. (2000). Chopper systems in cane harvesters: B: Results of a test program. Proc. Aust. Soc. Sugar Technol., 22: 250-255. Inderbitzin, M. and Beattie, R. (2012). Improving the harvesting and transport of whole crop harvested sugar cane. Proc. Aust. Soc. Sugar Technol. (CD- ROM), 34: 11p. Jais, P.C. (2010). Communities auto sufficient in fuels for humans, transport and electric needs. Proc. Int. Soc. Sugar Technol. (CD-ROM), 27: 8p. Kent, G.A. (2007). The effect of trash on the operation and performance of a raw sugar factory. Proc. Int. Soc. Sugar Technol., 26: 1622-1628. Kent, G.A., Hoare, C.P., Miller, K.F. and Allen, W.J. (2003). Harvest and factory data to assist in evaluating the effect of extraneous matter on sugar production costs and revenues. Proc. Aust. Soc. Sugar Technol. (CD-ROM), 25: 15p. Kent, G.A., Moller, D.J., Scroope, P.D. and Broadfoot, R. (2010). The effect of whole crop processing on sugar recovery and sugar quality. Proc. Aust. Soc. Sugar Technol., 32: 559-572. Linedale, A.I., Ridge, D.R. and Chapman, F.L. (1993). A coordinated extension approach for maximising returns from mechanical harvesting. Proc. Aust. Soc. Sugar Technol., 15: 45-51. Manechini, C., Júnior, A.R. and Donzelli, J.L. (2005). Benefits and problems of trash left in the field. In: Biomass power generation. Sugar cane bagasse and trash, 2005. S.J. Hassuani, M.R.L.V. Leal and I. De Carvalho Macedo (Eds). Programa das Nações para o Desenvolvimento, Piracicaba pp. 27-35. McGuire, P.J., Inderbitzin, M., Rich, B. and Kent, G.A. (2011). The effect of whole crop harvesting on crop yield. Proc. Aust. Soc. Sugar Technol. (CD-ROM), 33: 12p.

Mitchell, R.D.J., Thorburn, P.J. and Larsen, P. (2000). Quantifying the loss of nutrients from the immediate area when sugarcane residues are burnt. Proc. Aust. Soc. Sugar Technol., 22: 206-211. Moller, D., Broadfoot, R., Bell, S. and Bakir, H. (2010). Whole crop processing at Broadwater mill: impacts on process operations. Proc. Aust. Soc. Sugar Technol., 32: 573-587. Olivier, F.C. and Singels, A. (2012). The effect of crop residue layers on evapotranspiration, growth and yield of irrigated sugarcane. Water SA, 38: 77-86. Purchase, B.S., Wynne, A.T., Meyer, E. and van Antwerpen, R. (2008). Is there profit in cane trash? Another dimension to the assessment of trashing versus burning. Proc. S. Afr. Sug. Technol. Ass., 81: 86-99. Spinaze, D., Harris, H. and Lamb, B. (2002). A harvester-mounted trash shredder and collection system. Proc. Aust. Soc. Sugar Technol., (CD-ROM), 24: 7p. Thai, C.C.D., Bakir, H. and Doherty, W.O.S. (2012). Insights to the clarification of sugar cane juice expressed from sugar cane stalk and trash. J. Agric. Food Chem., 60: 2916-2923. Thai, C.C.D. and Doherty, W.O.S. (2012). Characterisation of sugarcane juice particles that influence the clarification process. Proc. Aust. Soc. Sugar Technol., (CD-ROM), 34: 9p. Thorburn, P.J., Meier, E.A. and Probert, M.E. (2005). Modelling nitrogen dynamics in sugarcane systems: recent advances and applications. Field Crops Research, 92: 337-351. Whiteing, C., Norris, C.P. and Paton, D.C. 2001. (2001). Extraneous matter versus cane loss: finding a balance in chopper harvestered green cane. Proc. Int. Soc. Sugar Technol., 24: 276-282.