As noted in Fig. 3, bio-oil from canola had lower viscosity than that of aspen. Among the bio-oils produced from corn cob, the viscosity of bio-oil with catalyst (corn cob 4) was lower than other bio-oils due to more water content. The viscosity of bio-oil produced from corn cob without catalyst was similar to that of liquid phase of canola and it was lower than that of liquid phase of aspen.

All the bio-oils from different feedstocks were behaved as a non-Newtonian fluid. Similar behaviors have been reported for bio-oils (upper layer) obtained from forest residues (Garcia-Perez et al., 2006a) and pine and oak bark (Ingram et al., 2008). However, Ingram et al (2008) reported Newtonian behavior at 25°C for the bio-oils produced from pine and oak wood through auger reactor but at higher temperatures (50 and 80°C) they showed mild shear thinning behaviors. Rheological data of shear rate and shear stress of the bio-oils were fitted according to the Power-Law model

A = ky-1 (1)

where, g is the viscosity (Pa-s), k is the consistency coefficient (Pa. sn), у is the shear rate, 1/s and n is the flow behavior index of the fluid (dimensionless). The power law conststants for different bio-oils are presented in Table 1. The flow behavior indexes n less than 1 suggests that presence of the pseudoplastic behavior (shear thinning). A possible reason might be breakdown of (waxy) structure would result in low viscosity at high shear rate. In general, the values of flow behavior index are more reliable than that of consistency coefficient (Johnson, 1999).The deviation of flow behavior index from ‘unity’ indicates the degree of deviation from Newtonian behavior. For shear thinning, the index value can be anywhere between 0 and 1. The smaller the value of n, the greater is the degree of shear thinning (Chhabra & Richardson, 1999). Considering the above points, canola aqueous phase

exhibited strong shear thinning behavior than that of the rest. In general, the bio-oils from corn cob approaches Newtonian behavior as the n values were close to unity.

2.2 Comparison of viscosity viscosity measurements and bio-oil viscosities

Accurate measurement of the viscosity of bio-oil/fuel is essential for the proper operation of fuel supply systems and atomisers. The viscosity of bio-oil can be measured according to the ASTM D 445 using the following equation

q = жРг1 t / 8lv = Tthpgr4t / 8lv (2)

where n is the viscosity (dynes/cm2 or poise), v is the volume of liquid (c. c.), t is the liquid flowing time (s), r is the radius of narrow tube (cm), l is the length of narrow tube (cm). This is most widely followed method, as evident from table 2. The viscosity of bio-oil can be measured using capillary or rotational viscometers and they are reported as kinematic (cSt) or dynamic viscosity (mPa. s). The kinematic viscosity of the bio-oil can be converted into dynamic viscosity if the density (kg/ dm3) of bio-oil is known at a given temperature using the following formula

According to ASTM D445, the viscosity of standard fuels, which are Newtonian fluids, is typically measured as kinematic viscosity. Leroy et al (1988) conducted extensive studies on rheological characterization of several bio-oils from wood and concluded that those bio-oils exhibited an essentially Newtonian behavior at the shear rate range of 1 to 1000/s. In contrary, bio-oils used in this study showed shear thinning behavior. For Newtonian fluid, the viscosity remains constant with increasing shear rate. Radovanovic et al (2000) reported a procedure to measure bio-oil viscosity using falling ball viscometer. Recently, Osamaa et al (2009) recommended using Cannon-Fenske viscometer tubes because the flow direction in these tubes compared to Ubbelohde tubes ensured more accurate results with dark coloured liquids. No prefiltration of the sample is required if the bio-oil is visually homogenous. Elimination of air bubbles before sampling and an equilibration time of 15 min are essential for viscosity measurement at a given temperature. Comparison of viscosity measurement and viscosity of bio-oils from produced different feedstocks through different pyrolysis reactors are presented in Table 2. Bio-oil density measurement and density values are also included for converting the viscosity from kinematic to dynamic units. According to ASTM D445-88, viscosity should be measured at 20 and 40°C, as seen from the table the viscosity was reported at different temperatures ranging from 20 to 100°C. The viscosity of bio-oils used in this study had lower viscosity than the viscosities listed in the table irrespective of temperatures.