Pretreatment of Rubber Tapping (Heava brasiliensis) and Achi Mkpuru (Gossweilerodendron balsamiferum) Hardwood Sawdust for Bio-Oil Production: Drying and Particle Size Characteristics at Different Temperatures By Onuora Okorie,Utibe Benedict Edem, Ifeanyichukwu Edeh and Ayoade Kuye* Abstract Drying and particle size characteristics of Rubber Tapping (Heava brasiliensis) and Achi Mkpuru (Gossweilerodendron balsamiferum) hardwood sawdust for bio-oil production were investigated. The samples were dried at temperatures of 80, 90 and 105OC. The drying curves were then modelled using nineteen different equations, eighteen of which were obtained from the literature. The Modified Henderson and Pabis model gave the best fit for the drying characteristics of the rubber tapping and Achi Mkpuru sawdust. The coefficient of determination, used to measure the goodness of fit, were in the range of 0.97039 to 0.99997 for the Rubber Tapping sawdust and 0.9764 to 0.99973 for the Achi Mkpuru data. The particle size analysis indicated that about 73% and 61% of dried Rubber Tapping and Achi Mkpuru sawdust respectively were less than or equal to 1mm. Hence further grinding of these samples may not be necessary when they are used for bio-oil production. Keywords: Sawdust, Drying, Particle Size, Bio-oil. 1. Introduction The current over-dependence of the global population on fossil fuel as its primary source of energy as well as sustainability and emerging climatic challenges such as global warming, environmental degradation of fossil fuel has raised many questions concerning energy security. Consequently, eco-friendly, renewable and sustainable sources of energy are being aggressively explored in order to meet the global energy demand at minimum cost to the environment. Some of these alternative sources of energy include hydrogen, biofuel and synthesis gas with biofuel being the most eco-friendly. It should be noted that fossil fuels are carbon based energy sources such as coal and natural gas while bio fuels are made from plants such as wood sawdust and sugar cane. Biofuels can be obtained from bio oil. Bio oil is a dark-brownish organic liquid whose appearance is close to the conventional fossil crude oil. Bio oil can be produced from the fast pyrolysis of lignocellulosic biomass, such as wood sawdust1 Wood sawdust is readily available in Nigeria. Ohunakin2 estimates that Nigeria’s annual value of about 4.39 x 106 m3 of log split and plywood processed, roughly translates to about 42 tonnes of sawdust generated from every 100 tonnes of timber, further resulting in an annual generation potential of about 1.8million tonnes. At the moment, such an enormous amount of the wood sawdust are often discarded, used for landfill, used as a bulking agent for compost manufacture, * Department of Chemical Engineering, University of Port Harcourt, Port Harcourt, Nigeria. Czernik, S. and Bridgwater, A. (2004): Overview of Applications of Biomass Fast Pyrolysis Oil, Energy & Fuels, 18 (2), 590-598. 2 Ohunakin, S. O. (2010): Energy Utilization and Renewable Energy Sources in Nigeria; Journal of Engineering and Applied Sciences, 5(2), 171-177. 1 Petroleum Technology Development Journal (ISSN 1595-9104): An International Journal; January 2015 - Vol. 5 No. 1 66 burnt as wastes or used for cooking3,4,5,6. Also, the indiscriminate disposal and incineration of sawdust constitute an environmental nuisance. The incineration of sawdust produces hazardous pollutant such as polychlorinated dibenzo-dioxins and dibenzo-furans which are injurious of the health of plants and animals7,8. We are not aware of any systematic study on the use of Nigerian sawdust for the production of bio oil. However from literature, before the wood sawdust can be processed into bio oil, it has to be subjected to a combination of drying and size reduction steps. The drying process is primarily aimed at expelling the inherent free moisture from the raw sawdust. Bridgwater and Peacock9 posited that sawdust could contain an initial moisture content of about 30-60wt. %. The initial moisture content of the sawdust determines its behavior during the pyrolysis process and can have deleterious effects on the stability, corrosiveness, pH, fuel value of the bio oil produced10,11. The experimental data obtained from the drying process can equally be used in formulating an empirical model for predicting the drying behaviour of samples. Such models are often called “thin layer drying models”. These models provide an estimate of the drying times of the sawdust samples and their drying curves - which provide for an improved performance of drying system12. They can be formulated on theoretical, semi-empirical or empirical basis13,14. As stated earlier, another very important parameter to be considered in the utilization of the wood sawdust in the production of bio oil by fast pyrolysis is the particle size. Understanding the particle size distribution of wood sawdust prior to bio oil production will help to minimize the 3 Ajueyitsi, O. N. (2009): Optimization of Biomass Briquette Utilization as Fuel for Domestic Use: Ph.D Research Proposal Seminar, Dept. of Mechanical Engineering, Federal University of Technology (FUTO), Owerri. 4 Kuye, A. O. and Edeh, I. (2013): Production of Bio oil from biomass using fast pyrolysis: A Critical Review, Journal of Minerals Research 1(1): 1- 19. 5 Horisawa, S., M., Sunagawa, Y., Tamai, Y., Matsuoka, T., Miura, T. and Terazawa, M. (1999): Biodegradation of nonlignocellulosic substances II: Physical and chemical properties of sawdust before and after use as artificial soil. Journal of Wood Science 45: 492-497 6 Zavala, M.A.L., Funamizu, N. andTakakuwa, T. (2004): Modeling of aerobic biodegradation of feaces using sawdust as a matrix. Water Resources 38: 1327-1339. 7 Terazawa, M. (2003). Sawdust Saves Globe-Bioconversion of Biomass Wastes into Multifunctional Recyclates Using Sawdust as an Artificial Soil Matrix. In: Proceedings of the 1st International Symposium on Sustainable Sanitation, Nanjing, China (1): 9-12. 8 Frombo, F., R., Minciardi, M., Robba, F., Rosso, S. (2009): Planning woody biomass logistics for energy production: A strategic decision model. Biomass and Bioenergy 33: 372-83. 9 Bridgwater, A.V., and Peacock, G.V.C. (2000): Fast pyrolysis process for biomass, Renewable and Sustainable Energy Reviews, 4: 1-73. 10 Akhtar, J. and Amin, N. S. (2012): A review on operating parameters for optimum liquid oil yield in biomass pyrolysis, Renewable and Sustainable Energy Reviews, 16: 5101–5109. 11 Fagernäs, L., Brammer, J., Wilén, C., Lauer, M., and Verhoeff, F. (2010): Drying of biomass for second generation synfuel production. Biomass and Bioenergy, 34: 1267-1277. 12 Cihan et al. 2007 Cihan, A., Kahveci, K., and Hacıhafızoğlu, O. (2007): Modeling of intermittent drying of thin layer rough rice. Journal of Food Engineering, 79: 293-298 13 Özdemir, M. and OnurDevres, Y. (1999): The thin layer drying characteristics of hazelnuts during roasting. Journal of Food Engineering 42 (4); 225-223 14 Midilli, A., Kucuk, H., and Yapar, Z. (2002): A new model for single-layer drying. Drying Technology 20 (7): 1503-1513. Petroleum Technology Development Journal (ISSN 1595-9104): An International Journal; January 2015 - Vol. 5 No. 1 67 residence time of the pyrolytic reaction and problems arising from heat transfer15. The particles should be very small to allow for rapid heating and high yields of bio-oil16. Jahirul et al17 puts this size at <1mm. The size of the wood sawdust depends mostly on the type of wood and the size of the saw teeth18. According to Himmel et al.19 and Houghton et al20, wood sawdust is directly associated with the size of the saw teeth/knife. Thus, increase in the size of knife/saw teeth will produce sawdust that has larger particle size. The utilization of such sawdust with larger particle size in bio oil production will require size reduction. Particle size has been shown to affect the yield of bio oil21,22,23,24,25,26,27,28. The aim of this work was to experimentally study and model the drying characteristics as well as determine the particle size distribution of Rubber Tapping (Heava brasiliensis) and Achi Mkpuru (Gossweilerodendron balsamiferum) sawdust obtained from Port Harcourt area in Nigeria at temperatures of 80oC, 90oC and 105oC. 2 Materials and Methods 2.1 Raw Materials: Rubber Tapping and Achi Mkpuru sawdust samples were collected directly during the normal logging work at Aluu timber sawmill and Dagogo timber sawmill, Iloabuchi both within Port Harcourt, Rivers State, Nigeria. A Stanner and CD 6 band sawmill machines were used in the case of Rubber tapping and Achi Mkpuru logs respectively. The samples were bagged and 15 Putun, et al., 2007 Putun, A.E, Ozbay, N., Apaydin, V. E., Uzun, B.B., Ates, F. (2007): Rapid and slow pyrolysis of Pistachio Shell: effect of pyrolysis conditions on the product yields and characterization of the liquid product, International Journal of Energy Research 31: 506-14. 16 Bridgwater, A.V., and Peacock, G.V.C. (2000): Op. Cit. 17 Jahirul, I.M., Rasul, G.M., Chowdhury, A. A. and Ashwath, N. (2012): Biofuels Production through Biomass Pyrolysis - A Technological Review. Energies 5: 4952-5001. 18 Afuwape, F. K. (1983): Design and testing of a sawdust compactor, B.Sc. Thesis, Department of Agricultural Engineering; Obafemi Awolowo University; Ile-Ife, Nigeria. 19 Himmel, M., M. Tucker, J. Baker, C. Rivard, K. Oh, andK. Grohmann. (1985): Comminution of biomass: Hammer and knifemills. Biotechnology and Bioengineering Symposium 15: 39–58. 20 Houghton, J. I., J. E. Burgess and Stephenson, T (2002): Off-line particle size analysis of digested sludge. Water Resources 36: 4643-4647. 21 Bridgwater, A.V. (2003). Renewable fuels and chemicals by thermal processing of biomass. Chem. Eng. Journal 91: 87-102. 22 Kersten, S. R. A., Wang, X., Prins, W., and Van Swaaij, W.P.M. (2005): Biomass pyrolysis in a fluidized bed reactor: Part 1: Literature review and model simulations, Ind. Eng. Chem. Res.: 44, 8773. 23 Haykin-Acma, H., Yaman, S., and Kucukbayrak, S. (2006): Effect of heating rate on the pyrolysis yields of rapeseed. Renewable Energy 31: 803–810. 24 Park, H.J., Park, Y.K., and Kim, J.S. (2008): Influence of reaction conditions and the char separation system on the production of bio-oil from radiata pine sawdust by fast pyrolysis, Fuel Process Technol., 89: 797-802. 25 Park, H.J., Park, Y.K., Dong, J.I., Kim, J.S., Jeon, J.K., Kim, S.S., Kim, J., Song, B.H., Park, J., and Lee, K.J. (2009). Pyrolysis characteristics of Oriental white oak: Kinetic study and fast pyrolysis in a fluidized bed with an improved reaction system. Fuel Process Technol: 90, 186–195. 26 Jung, S.H., Kang, B.S., and Kim, J.S. (2008): Production of bio-oil from rice straw and bamboo sawdust under various reaction conditions in a fast pyrolysis plant equipped with a fluidized bed and a char separation system. J. Anal. Appl. Pyrolysis 82: 240–247. 27 Shen, D.K., and Gu, S. (2009): The mechanism for thermal decomposition of cellulose and its main products. Bioresource Technology 100: 6496–6504. 28 Heo, H.S., Park, H.J., Park, Y.K., Ryu, C., Suh, D.J., Suh, Y.W., Yim, J.H., and Kim, S.S. (2010): Bio-oil production from fast pyrolysis of waste furniture sawdust in a fluidized bed. Bioresource Technology 101: S91-S96. Petroleum Technology Development Journal (ISSN 1595-9104): An International Journal; January 2015 - Vol. 5 No. 1 68 immediately analysed at the PTDF Bio-oil Research Laboratory located at the Department of Chemical Engineering, University of Port Harcourt, Rivers State, Nigeria. 2.2 Drying Experiments The equipment used in conducting the drying experiments consisted of a muffle furnace and an analytical balance (Radwag WLC 6/A2). The muffle furnace was switched on for 45minutes for the air and a cleaned stainless steel pan within the furnace to attain the chosen drying temperature (80, 90 or 105 ). The stainless pan was later withdrawn from the furnace and allowed to cool down to the ambient temperature, after which it was weighed with the analytical balance to the nearest 0.1g. 120g wet sawdust samples were evenly distributed on the pan and weighed to the nearest 0.1g. The sample was withdrawn from the furnace periodically (every 20min) and reweighed. At each interval, the change in the weight of the sample resulting from a gradual loss of moisture was recorded until there was no further change in the sample weight. The procedure was repeated three times; the average values were computed and used for analyses. The moisture content (on dry basis) was obtained using the equation: (1) where (g water/g dry solids) is moisture content at a time t, is the total product weight at each time and is weight of the dry solids. The moisture content can be converted to a non dimensionless quantity referred to as moisture ratio (MR) by dividing by the initial moisture content (g water/g dry solids), that is: (2) An empirical drying model was formulated to fit the experimental data as follows: (3) Eighteen other empirical and semi empirical drying models were also considered in the study. These models are: Lewis29 Page30 Modified Page31 (4) (5) (6) Henderson and Pabis32 (7) 29 Ayensu, A. (1997): Dehydration of food crops using a solar dryer with convective heat flow. Solar Energy 59: 121–126. 30 Karathanosa, V.T., and Belessiotis, V.G. (1999): Application of a thin layer equation to drying data of fresh and semi-dried fruits. Journal of Agricultural Engineering Research 74: 355–361. 31 Yaldiz, O., Ertekin, C., and Uzun, H. B. (2001): Mathematical modelling of thin layer solar drying of sultana grapes. Energy 26: 457–465. 32 Akpinar, E.K., Bicer, Y., and Yildiz, C. (2003): Thin layer drying of red pepper. Journal of Food Engineering 59: Petroleum Technology Development Journal (ISSN 1595-9104): An International Journal; January 2015 - Vol. 5 No. 1 69 Logarithmic33 Two-term34 Two-term exponential35 (Approximation of diffusion) (8) (9) (10) (11) 36 Wang & Singh Diffusion approach37 Verma et al.38 Modified Henderson & Pabis39 Midilli et al.40 Weibull distribution41 Aghbashlo et al42 Logistic43 Jena & Das44 Demir et al.45 Alibas46 (12) (13) (14) ⁄ ( ( ⁄ √ ) ) (15) (16) (17) (18) (19) (20) (21) In Eqns. 3 to 21, = moisture ratio; coefficients; drying constants; drying time. Non-Linear regression analyses of these models [Eq (3)-Eq (21)] were made using Microsoft Office Excel 2007 Solver. The purpose of the non-linear regression analysis was to calculate the parameters; of the models. The coefficient of determination ( ) was used to select the best equation from Eqs (3) to (21) that provides the best fit to the experimental sawdust drying data at each temperature. ( ) is given by: 99–104. 33 Yagcioglu, A., Degirmencioglu, A., and Cagatay, F. (1999): Drying characteristic of laurel leaves under different conditions. In: 170: Proceeding of the 7th International Congress of Agricultural Mechanization and Energy, (Eds.: A Bascetincelik) Cukurova University, Adana, Turkey, 26-27 May, pp.565-569. 34 Wang, X., Chen, H., Luo, K., Shao, J., and Yang, H. (2008): The influence of microwave drying on biomass pyrolysis. Energy & Fuels, 22: 67-74. 35 Yaldiz, O., Ertekin, C., and Uzun, H. B. (2001): Op. Cit. 36 Yaldiz, O., Ertekin, C., and Uzun, H. B. (2001): Op. Cit. 37 Özdemir, M. and Onur Devres, Y. (1999): Op. Cit. 38 , O., Ertekin, C., and Uzun, H. B. (2001): Op. Cit. 39 Karathanosa, V.T., and Belessiotis, V.G. (1999): Op. Cit. 40 Midilli, A., Kucuk, H., and Yapar, Z. (2002): Op. Cit. 41 Babalis, S.J., Papanicolaou, E., Kyriakis, N., and Belessiotis, V.G. (2006): Evaluation of thin-layer drying models for describing drying kinetics of figs (Ficuscarica). Journal of Food Engineering 75: 205-214. 42 Aghbashlo, M., Kianmehr, M.H., Khani, S., and Ghasemi, M. (2009): Mathematical modeling of carrot thin-layer drying using new model. Int. Agrophysic 23: 313-317. 43 Chandra, P.K., and Singh, R.P. (1995): Applied numerical methods for food and agricultural engineers. CRC Press, Boca Raton, FL: pp. 163-167. 44 Jena, S., and Das, H. (2007): Modeling for vacuum drying characteristics of coconut press cake. Journal of Food Engineering 79:92-99. 45 Demir, V., Gunhan, T., and Yagcioglu, A.K. (2007): Mathematical modeling of convection drying of green table olives. Biosystems Engineering 98(1): 47-53. 46 Alibas, I. (2012). Selection of the Best Suitable Thin-Layer Drying Mathematical Model for Vacuum Dried Red Chili Pepper. J. Biol. Environ. Sci. 6(17): 161-170. Petroleum Technology Development Journal (ISSN 1595-9104): An International Journal; January 2015 - Vol. 5 No. 1 70 ∑ ∑ (22) where; is the experimental moisture ratio, is the predicted moisture ratio, is average value of the experimental moisture ratio, NTotal is the total number of experimental data points, and represents the various constants within each model. 2.3. Determination of Particle Size Distribution The particle size distribution was determined using a digital sieve shaker and an analytical weighing balance. After each of the sieves had been cleaned and weighed, they were stacked above one another in order of increasing sieve particle sizes (0.300, 0.425, 0.500, 0.710 and 1.000) mm. 50g of the dried sawdust samples was carefully emptied into the top sieve and the lid placed on it. The stack of sieves was gently mounted on the digital sieve shaker and set for 10min at an amplitude of 0.9mm. After a 10min agitation the stack of sieves was gently carried off the shaker and carefully dissembled. Each sieve with its retained sawdust were weighed and recorded. The total sawdust retained was computed by adding the individual weights retained on each sieve. 3. Results and Discussion The parameters for the nineteen empirical and semi-empirical drying models (Eqns 3 - 21) for the experimental drying characteristics of the two sawdust samples were calculated using MS Excel Solver. For the Rubber Tapping sawdust, the R2 ranged from 0.97039 to 0.99997 while that for the Achi Mkpuru ranged from 0.9764 to 0.99973. The specific values for each model can be found in Okorie47. For both saw dusts, the modified Henderson and Pabis (Eqn 14) gave the highest value of R2 at each drying temperature. The parameters for Eqn 14 are summarized in Tables 1 and 2 for Rubber Tapping and Achi Mkpuru saw dusts respectively. Table 1 Modified Henderson and Pabis model parameters for Rubber Tapping Sawdust a b c g h k R2 80 0.485964 0.65127 -0.137273 0.034056 0.799911 0.010640 0.99961 90 2.237306 -1.27593 0.038627 0.036043 1.009900 0.027925 0.99959 105 0.768766 8.55477 -0.188213 0.203605 1.000000 0.034879 0.99997 Temp, OC Table 2 Modified Henderson and Pabis model parameters for Achi Mkpuru Sawdust Temp, OC a b c g h k R2 80 3.541554 -2.522919 -0.018620 0.008274 1.300000 0.009427 0.99938 90 4.605721 -3.535219 -0.070301 0.011244 1.011904 0.012407 0.99929 105 1.256541 -0.068221 -0.188213 0.006378 1.000000 0.025340 0.99973 The experimental moisture ratio values are shown in Figs 1 and 2 as symbols respectively for Rubber Tapping and Achi Mkpuru saw dusts at the different temperatures. The solid lines shown in these figures are the corresponding values calculated using Eqn 14. Figs 1 and 2 clearly confirm that the modified Henderson and Pabis model is a good fit for the experimental and validate the high values of R2 shown in Tables 1 and 2. 47 Okorie, O. (2014): Pretreatment of Rubber Tapping and Achi Mkpuru Hardwood Sawdust for bio-oil production: Drying and particle size Characteristics at Different Temperatures; M.Eng. Thesis, Chemical Engineering Department; Faculty of Process and Energy System Engineering, University of Port Harcourt, Port Harcourt, Nigeria Petroleum Technology Development Journal (ISSN 1595-9104): An International Journal; January 2015 - Vol. 5 No. 1 71 1 0.9 0.8 Moisture Ratio 0.7 0.6 80C 90C 0.5 105C 0.4 0.3 0.2 0.1 0 0 50 100 Drying Time, min 150 200 Fig 1: Moisture Ratio Curve of Rubber Tapping Sawdust at different drying temperatures. Petroleum Technology Development Journal (ISSN 1595-9104): An International Journal; January 2015 - Vol. 5 No. 1 72 1 0.9 0.8 0.7 Moisture Ratio 0.6 80C 0.5 90C 105C 0.4 0.3 0.2 0.1 0 0 50 100 150 200 Time, min 250 300 350 Fig 2: Moisture Ratio Curve of Achi Mkpuru Sawdust at different drying temperatures 3.1 Particle Size Distribution Particle size distribution of Rubber Tapping and Achi Mkpuru sawdusts at different temperatures are shown in Figs 3 and 4. These figures indicate that drying temperatures had minimal effect on the particle size distribution for both samples. Furthermore, it can be deduced that about 73% and 61% of Rubber Tapping and Achi Mkpuru respectively passed through the sieve diameter of 1mm at all three drying temperatures. Thus, about 27% of Rubber Tapping and 39% of Achi Mkpuru are greater than 1mm in size. The implication of these results is that Rubber tapping and Achi Mkpuru sawdust would only require drying when they would be used for bio oil production; Petroleum Technology Development Journal (ISSN 1595-9104): An International Journal; January 2015 - Vol. 5 No. 1 73 grinding may not be necessary. As noted earlier, the use of small sawdust particle sizes (<1mm) as feedstock for bio-oil production through fast pyrolysis is necessary for higher bio-oil yields48. Cumulative % of sawdust retained 100 80C 90C 105C 10 0.1 1 Sieve diameter (mm) Fig 3 Particle–Size Distribution Curve of Rubber Tapping sawdust at different temperatures Cumulative % of sawdust retained 100 80C 90C 105C 10 0.1 Sieve diameter (mm) 1 Fig 4 Particle–Size Distribution Curve of Achi Mkpuru sawdust at different temperatures 4. Conclusion The results obtained from this work indicate that the modified Henderson and Pabis model fits the experimental drying data for Rubber Tapping and Achi Mkpuru very well. The coefficient of determinations for any of the drying temperatures is greater that 0.999. The sieve analysis of the 48 Jahirul et al, 2012 Petroleum Technology Development Journal (ISSN 1595-9104): An International Journal; January 2015 - Vol. 5 No. 1 74 samples also indicate that more that 60% of the particles have diameters less than or equal to 1mm thus showing that there may not be any need for grinding when they are used for bio-oil production. 5. Acknowledgement The authors are grateful to the Petroleum Technology Development Fund (PTDF) for providing the necessary funds for this research work. Petroleum Technology Development Journal (ISSN 1595-9104): An International Journal; January 2015 - Vol. 5 No. 1 75