Molecular Filtration of Rubber Derived Fuel

2nd International Conference on Environment, Agriculture and Food Sciences (ICEAFS'2013) August 25-26, 2013 Kuala Lumpur (Malaysia) Molecular Filtration of Rubber Derived Fuel Jefrey Pilusa, Edison Muzenda, and Mukul Shukla Murugan et al. (2008) [5] studied the modification of the crude tyre pyrolysis fuel and found that it could be modified in three stages (i) Moisture removal, (ii) De-sulphurisation, (iii) Vacuum distillation. It was found that distilled pyrolysis fuel had a 7% high heating value than the crude pyrolysis fuel, and this is attributed to the elimination of the impurities, moisture, carbon particles, sulphur and sediments. Majority of liquids mercaptans in crude fuels such as ethyl, n-propyl, n-butyl and carbon disulphide mercaptans are less reactive and have boiling points below 98oC [6]. These compounds can be removed along with the moisture during distillation process. Diesel fuel filter’s efficiency is described by Volkwein et al. (2004) [7] as the ability of the filter to remove contaminants at a given particle or molecule size. The type of media being used ultimately defines the filter's efficiency. Capacity is the measurement in grams of the total amount of containment a filter can retain at a rated flow and given endpoint. The type of media (i.e. glass, cellulose, synthetic, membranes etc.) and the active filtration area of media define capacity. Coppo et al. (2006) [8] explained the filter restriction in terms of the pressure drop across the element at a given flow, temperature, and fluid viscosity. Molecular Sieves are crystalline, three-dimensional molecules made up of silicon and aluminum atoms. The extensive networks that make up molecular sieves contain surface pores and channels which selectively absorb only molecules of a certain size and shape [9]. Positive ions, most commonly sodium, calcium, or potassium are added in order to balance the molecule [10]. The specific cation chosen influences the pore diameter and therefore the adsorptive properties of the molecular sieve. The sodium modified molecular sieve, with a pore diameter of ten angstroms will absorb those molecules with a critical diameter of less than 10 angstroms [11]. Eq. 1 show the chemical formula of the porous membranes selected as a filter media for the application described in this article. Abstract—The work presented in this article reviews the possible methods for removal of mercaptans from distilled rubber derived fuel. Distilled rubber derived fuel is a potential additive for diesel fuel; therefore reduction of sulphur compounds in this fuel is essential for application in compression ignition engines. Membrane filtration of distilled rubber fuel using 13-X molecular sieves has shown a significant reduction in sulphur content. It was observed that distilled rubber fuel can be effectively filtered via single pass to remove up to 53.67% of the fuel’s initial sulphur. Keywords—Diesel additive, Compression ignition, Membrane filtration, Pyrolysis fuel, Waste-to-energy Fuel, I. INTRODUCTION R UBBER fuel is derived by the slow pyrolysis of rubber containing waste products such as tyres. Crude fuel is further distilled to recover high aromatic fuel resembling the characteristics of diesel fuel. Previous research has shown that this fuel can be used directly in compression ignition engines without any engine modification. However high toxic exhaust emissions were realised due to the use of this fuel [1]. This fuel has also been reported to contain sulphur content of approximately 9,000–12,000 ppm [2]. This high sulphur content is as result of natural rubber vulcanisation in tyre manufacturing process [2]. High sulphur contents in fuels result in significant generation of sulphur oxides emissions which is a major environmental pollutant. Sulphur reduces catalyst efficiency in modern vehicles, and vehicles operating with higher sulphur fuels have higher emissions [3].The fuel derived from pyrolysis of rubber has been known as a material with excellent and consistent fuel properties with a high calorific value, thus maybe used directly as fuel or blended with other fuels [4]. A number of studies have been reported in literature relating to the distribution of the hydrocarbons composition of the pyrolytic fuel at various operating temperatures. It has been reported that the aromatic hydrocarbons were about 34.7- 75.6w/w% when the pyrolysis temperature was varied from 300oC to 700oC while the aliphatic hydrocarbons were about 19.8 59.2w/w% [4]. Na2O. Al2O3 .0.254 SiO2 .xH 2O (1) Molecular sieves are suitable for drying, purifying, and separating a wide variety of compounds, such as inorganic gases, hydrocarbons, halogenated hydrocarbons, and sulphurous compounds. They are being used to adsorb, and temporarily isolate molecules. When loaded on a molecular sieve [12]. A saturated molecular sieve can be restored to its original capacity by regeneration, the principle of which involves changing the conditions surrounding the adsorbent to correspond to a very low equilibrium capacity. In general, the greater the difference between the equilibrium capacities of adsorption and regeneration, the more rapid and complete Jefrey Pilusa is with the Department of Mechanical Engineering Science at the University of Johannesburg, Auckland Park, South Africa (corresponding author: phone: +27 10 210 4813; fax: +27 10 210 4800; email: [email protected] Edison Muzenda is a Professor in the Department of Chemical Engineering Technology at the University of Johannesburg, Doornfontein, South Africa e-mail: [email protected] Mukul Shukla is a Professor in the Department of Mechanical Engineering MNNIT, ALLAHABAD, 211 004, UP, India ,e-mail: [email protected] 121 2nd International Conference on Environment, Agriculture and Food Sciences (ICEAFS'2013) August 25-26, 2013 Kuala Lumpur (Malaysia) the regeneration. The maximum regeneration temperature for Silica is approximately 300oC [12]. The type of media and general filter construction defines restriction and fuel flow through the active area of the filter media. The flow rate of diesel fuel system can be estimated by the following equation as described by Ensfield, (2002) [13]. mf = Pb .3600 cvl .ηbt B. Screening Tests Samples of distilled rubber derived fuel were added into porous membranes in closed sample containers for 24 hours as shown in Fig. 3. A discoloration in both porous membranes and fuel were observed. A continuous fuel filtration system was developed to explore the effectiveness of single pass adsorption at various flow rates. (2) P b is brake power in kW m f is fuel mass flow rate in kg/h. ηbt is engine efficiency cv1 is the lower calorific value of the fuel in kg/kJ Fig. 3 Distilled rubber fuel undergoing molecular filtration This work investigates the possibility of using porous membranes for the removal of mercaptans in distilled rubber derived fuel. C. Continuous single pass filtration An experimental set-up consisting of stainless steel filter housing tube packed with porous membranes connected to a vacuum system is shown in Fig. 4. Fuel samples of known initial sulphur content were passed through a layer of porous membranes at various flow rates. Diesel fuel, distilled rubber fuel and 50% vol. diesel-rubber fuel blends were passed over a porous layer of molecular sieves with a packing density of 853.41 kg/m3 and active filtration area of 0.13m2. Each fuel sample was added into a sample container with a porous layer. Samples were collected every 30 seconds for 120 seconds. The same porous layer was re-used for filtration of diesel-rubber fuel blend and pure rubber fuel over the same time interval. It was observed that the majority of the sulphurous compounds in the fuels are removed during the first 30 seconds of contacting the fuel with the micro porous layer. The residence time for fuel contacting with the porous filter media and the final sulphur content were recorded. The results for these tests are presented in table IV. II. METHOD A. The Sample Characterisation A representative sample of crude rubber fuel derived through a slow pyrolysis process was obtained from Pace fuels (Pty) Ltd in South Africa. Physical properties of the crude fuel were measured with reference to diesel fuel and presented in Table I. The crude fuel was heated at 100oC using the apparatus shown in Fig.1. The temperature was maintained at 100oC while condensing the vapours to remove moisture and distillable liquid mercaptans. Fig. 1 Set-up for batch distillation of crude rubber fuel The temperature was raised to the desired range to distil and condense the light fractions from the crude rubber fuel at 150, 200 and 250oC. The properties of the three light fractions obtained are presented in table II. Fig. 2 Distilled rubber fuel at 150, 200 and 250oC from left to right Fig. 4 Experimental set-up for single pass filtration 122 2nd International Conference on Environment, Agriculture and Food Sciences (ICEAFS'2013) August 25-26, 2013 Kuala Lumpur (Malaysia) to increase the density of a fuel and the heating value of the fuel resulting in improved cold flow properties [15]. On the other hand aromatics are restricted in diesel fuels because they reduce the centane number of the fuel and have been identified as contributors of nitrogen oxides (NO x ) emissions [16]. This will not be the case for distilled rubber fuel since the intensity of aromatic compounds present in the fuels is lower compared to diesel fuel as shown in Fig. 6. Other functional groups were identified in distilled rubber fuel which are not present in commercial diesel. These include carboxylic acids (O-H stretch) and ketones (C-C stretch). III. RESULTS AND DISCUSSIONS A. Fuel Characterisation The distillates obtained at 250oC being the fractions that resemble diesel fuel underwent further treatment to reduce the residual sulphur compounds. The compounds that are distillable at 150 to 250oC are napthas, at this temperature range sulphur is more easily removed because, lower boiling fuel fraction primarily contain sulphurous compounds that are in the form of mercaptans or lower member ring compounds which are relatively easier to de-sulphurise [14]. This is noticed by reduction in sulphur content from 9106 ppm to 7030 ppm when crude rubber fuel is distilled at 250oC. Due to its high viscosity, contamination and high sulphur content as shown in Table I, its application as a fuel in diesel engines is not recommended without refining. TABLE I PHYSICAL PROPERTIES OF CRUDE RUBBER FUEL AND DIESEL FUEL Property Density @ 20oC (kg/m3) Crude Rubber Fuel 926 Viscosity @ 40oC (cSt) 9 o Diesel Fuel 831 SANS 342 Specification 800-950 2.6 2.2-5.3 Flash Point ( C) 94 54 >55 Total Contamination (mg/kg) Total Sulphur (ppm) 143 9106 29 498 <24 <500 Water Content (vol. %) 3.54 0.05 <0.04 Gross Calorific Value (MJ/kg) 43 46 - Fig. 2 shows images of rubber fuel distillates obtained at 150, 200 and 250oC. The results presented in Table II reveals that the light fractions obtained at 250oC are near the properties of diesel fuel as per South African National Standards (SANS-342) presented in Table I. Total contamination of fuel is capped at 24 mg/kg according to SANS 342, crude rubber fuel has a total contamination value of 6 times the acceptable contamination limit. In contrast, distilling the fuel does not only increase the calorific value of the fuel but it also reduces total contamination by 95%. Majority of liquids mercaptans in crude rubber fuel were mostly removed along with the moisture during the pre distillation stage. Fig. 5 Common Functional Groups Identified in Distilled Rubber Fuel and Diesel Fuel C. Continuous single pass filtration Barth et al., 2004 [17] indicated that the average engine efficiency for heavy duty diesel trucks is about 47-49%. The maximum fuel flow through the ADE 407T truck engine to be used as a test model with technical specifications presented in table III was estimated using Eq.2. The maximum fuel flow rate through this engine under extreme operating conditions is approximately 700ml/min. The packing density of the porous membranes is defined as a function of membrane pellet mass and its diameter, this relationship is represented by Eq. 3. TABLE II PHYSICAL PROPERTIES OF DISTILLATES AT VARIOUS TEMPERATURES Property Density @ 20oC (kg/m3) Viscosity @ 40oC (cSt) Flash Point (oC) Total Contamination (mg/kg) Total Sulphur (ppm) Water Content (%) Gross Calorific Value (MJ/kg) 150oC 743 0.8 26 3.5 6923 0.04 38.6 200oC 789 0.9 35 5.5 6984 0.05 39.5 250oC 807 1.5 43 6.5 7030 0.05 43.7 TABLE III ENGINE TECHNICAL SPECIFICATIONS B. Identification of Functional Groups Analysis of the crude rubber fuel, diesel and selected distilled rubber fuel chemical composition were conducted using a Fourier Transform Infra-Red Spectroscopy (FTIR) to identify the functional groups present. Fig 5 shows specific functional groups present in both diesel fuel and rubber fuel distillates obtained at 250oC. The distilled rubber fuel has noticeable fractions of aromatic compounds which are said 123 Engine Owner: University of Johannesburg Engine Type: ADE 407T Stationary truck engine Aspiration Turbocharged Operation 4 stroke diesel Dynamometer 668mm torque arm Froude Hydraulic Max Torque & Power(kW): 1140N.m @ 1200rpm & 206kW Wet Mass(kg) 815 Bore/stroke(mm) 125/155mm 2nd International Conference on Environment, Agriculture and Food Sciences (ICEAFS'2013) August 25-26, 2013 Kuala Lumpur (Malaysia) Volume displacement(ml)  2.526.mi  ∂ρ =   3  d i .π  υ = 0.14475.π .lt .Dt 2 0.6315.D t .lt 3 di κ= 2.526.π .lt .D 2 t di t= (3) (4) 2 n= analysis on the fuel samples before and after molecular filtration indicated a 23.6 % reduction in total sulphur. The porous membranes beads of 0.1nm were used for adsorbing mercaptans and other sulphur compounds below 0.1nm. Active surface area of each spherical porous bead is an important parameter ensuring that selected molecules such as mercaptans are adsorbed effectively. This is seen in Fig. 6 whereby 9.3%, 63% and 44.9% reduction is sulphur content was observed in diesel, diesel-rubber fuel blend and rubber fuel respectively. Further adsorption time does not seem to have any further effect on fuel de-suphurisation. The selected membrane sieves have shown effective removal of specific molecular range of sulphur compounds. A 9.3% sulphur removal in diesel fuel was observed compared to higher sulphur reduction in high sulphur concentration fuels. Accumulations of the porous media have not been identified in these tests. All the fuels were passed through the same porous media without regeneration and the adsorption efficiencies during the first 30 seconds were consistent for pure distilled rubber fuel as shown in Fig 6. 11,416 (5) (6) 0.10525.π .lt .D 2 t Φ (7) ∂ p Packing density in (kg/m3) mi Average mass of porous membrane bead (kg) d i Average diameter of porous membrane bead (m) υ n Voids volume between porous membranes (m3) Number of porous membrane beads lt Active length of filter media housing tube (m) κ Active filtration area of filter media (m2) t Fuel residence time (s) Φ Fuel flow rate (m3/s) Dt Inside diameter of filter media housing tube (m) The porous membranes with nominal pore sizes of 10 Armstrong were obtained from Sigma Aldrich, South Africa. These spherical membranes were 3.2 mm diameter and 34.7 mg per pellet. The packing density of porous membranes in this study as shown in Table IV was calculated to be 853.41kg/m3 using Eq. 3 for a given average membrane bead mass and diameters of 34.8mg and 3.2mm respectively. The voids volume and active surface area were 24.22 ml and 0.13m2 respectively. This was calculated using Eqs. 4 and 6 for a given tube length of 110 mm and inside diameter of 22 mm. Fig. 6 De-sulphurisation of fuel over porous membranes TABLE IV TEST RESULTS FOR MOLECULAR FILTRATION OF DISTILLED RUBBER DERIVED FUEL Fuel flow (m3/s) 0.0025 0.0050 0.0075 0.0117 0.0067m3/s Residence time (ms) 7.042 3.521 2.347 1.509 Average 3.6046 ms Filter media area(m2) Flux(L/m2.s) Tube Volume (ml) Void Volume (ml) Number of membranes sieves Volume micro porous layer(ml) Mass of porous layer(g) Filter Media Packing Density(kg/m3) E. Filtration fluxes Eq. 5 was developed to estimate the number of beads required as a function of filter housing diameter, length and bead diameter. Eq. 7 describes the relationship between the filter media packing and fuel residence time. The filtration flux was computed as the fuel flow rate through a packed porous layer per unit active area. The experimental set-up presented in Fig. 4 has an average filtration flux of 50.495 l/m2.s. Sulpher removed (%) 58.34% 56.32% 52.36% 47.67% 53.67% 0.13 50.494 41.81 24.22 1025 17.6 35.685 853.41 IV. CONCLUSION The sulphur content in the fuel was reduced by 53.67% via a single pass molecular filtration using 13X molecular sieves. Adsorption of mercaptans and sulphur compounds occurs very rapidly over an active porous membrane layer. A significant amount of sulphur can be removed from the fuel stream at a rate of 50.49 l/m2.h. The fuel properties of distilled fuel obtained at 250oC are nearly comparable to commercial diesel with high heating value of and low water content and total contamination. This work has also shown that the fuel cannot be used directly into compression D. De-suphurisation over porous membranes Since a fraction of mercaptans were removed from the fuel, the free sulphur remained in the fuel was reduced through adsorption over the porous membrane layer. Sulphur 124 2nd International Conference on Environment, Agriculture and Food Sciences (ICEAFS'2013) August 25-26, 2013 Kuala Lumpur (Malaysia) ignition engines in is pure form due to its higher sulphur content, low viscosity and low flash point. A blend of this fuel with diesel fuel may be considered to an extent whereby the overall viscosity and flash point of the blend fuel is at least 2.2cSt and 55oC respectively. The effect of packing density and voids between the membranes on the adsorption rate need to be investigated. Thermal reactivation of the membrane sieves for reuse and life span determination should be evaluated. ACKNOWLEDGMENT The authors are grateful to the National Research Foundation of South Africa (NRF), Departments of Mechanical and Chemical Engineering for supporting the research. REFERENCES [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] J. Shah, M.R. Jau, and F. Mabood, “Recovery of value-added products from the catalytic pyrolysis of waste tyre,” Energy Conversion and Management, vol. 50, pp. 991-994, 2009. A.M Cunliffe, and P.T Williams, “Composition of fuels derived from the batch pyrolysis of tyres,” Journal of Analytical and Applied Pyrolysis, vol. 44, pp. 131-152, 1998. M. P. Hirenkumar, and M.P. Tushar, “Emission Analysis of a Single Cylinder Fuelled with Pyrolysis Fuel Diesel and its Blend with Ethanol,” IJEST, vol. 4, pp. 2834-2838, 2012. M.F. Laresgoiti, B.M. Caballero, I. de Marco, A. Torres, M.A Cabrerro, and M.J. Chomon, “Characterization of the liquid products obtained in tyre pyrolysis,” Journal of Analytical and Applied Pyrolysis, vol. 71, pp. 917-934, 2004. S. Murugan, M.C Ramaswamy, and G. Nagarayan, “A comparative study on the performance, emissions and combustion studies of DI engine using distilled tyre pyrolysis fuel-diesel fuel blends,” Fuel, vol. 87, pp. 2111-2121, 2008. D. Prandhan, and R.K. Singh “Thermal Pyrolysis of Bicycle Waste Tyre Using Batch Reactor”, International Journal of Chemical Engineering Applications, vol. 2, pp. 322-336, 2011. J.C.Volkwein, S.E. Mischler, B. Davies, and C. Ellis, “Field Measurement of Diesel Particulate Matter Emissions”. Centers for Disease Control and Prevention, National Institute of Occupational Safety and Health, Pittsburgh, USA, 2004. M. Coppo, C. Dongiovanni, C. Negri, “Linear optical sensor for measuring needle displacement in common-rail diesel injectors”. Sensors and Actuators, 134: 366–373, 2006. P.T D. Giagou, G. Foudfuelis, V. Gekas, “Nano filtration: towards a breakthrough”, unpublished. William, and R.P Bottril, “Sulphur-polycyclic aromatic hydrocarbons in tyre pyrolysis fuel,” The University of Leeds: Department of Fuel and Energy, Leeds United Kingdom, 1995. L. Kramer, and W. Cappel, “Presentation on Mercaptans in Crude Fuel-Baker Hughes Incorporated 2011”, unpublished The Merck Index, “Twelfth Edition Merck & Co”, 1996. C. Roy, A. Chaala, and H. Darmstadt, “The vacuum pyrolysis of used tyres-end-uses for fuel and carbon black products,” Journal of Analytical and Applied Pyrolysis, vol. 51, pp. 201-221, 1999. P. Trongkaew, T. Utisthan, P. Ruebroycharen, and N. Hinchirama, “Photocatalytic desulphurisation of waste tyre pyrolysis fuel,” Energies, vol. 4, pp. 1880-1896, 2011. K. Unapumnuk, T.C. Keener, M. Lu, and F. Liang, “Investigation into the removal of sulphur from tyre derived fuel by pyrolysis,” Fuel, vol. 87, pp. 951-956, 2008. M.N Islam, M.F.R Khan, and M.Z Alam, “Production and Characterisation of Scrap Tyre Pyrolysis Fuel and its blend,” ICME, 2003. M.Barth, G.Scora, and T.Younglove, “A Modal Emission Model for Heavy Duty Diesel Vehicles”, Transportation Research Board, 2004 125