Screening of factors influencing biobutanol production from pineapple waste using 2-level factorial design

Abstract

Eight local isolated bacteria strain were screened for high biobutanol production using modified Reinforced Clostridia Medium (RCM) at 37°C for 16 hours. Out of eight bacteria strain, only four strains were positive biobutanol producer. From four positive strains, UPM strain produced the highest biobutanol (0.141 g/L), analyzed using GC, which is three times higher as compare to other three positive bacteria. The productivity UPM strain 8.75x1 O'3 highest compared to other isolated bacteria, Nafizah, S I , and S2 strains. The liquid pineapple waste contain three types of reducing sugar which approximately 123.0 g/L of glucose, 2.23 g/L of sucrose and 15.86 g/L of fructose. Sugar analyzing was important prior to screening of factor that influenced the biobutanol production because the amount of sugar in pineapple waste concentration is one of the factors to be screened. Screening of factors influenced bio butanol production by strain UPM was conducted using Design Expert Software. The experimental design consisted of 37 experiments with replication at five central points for each stage using 2-level factorial. The screening experiments were carried out in static batch culture. The significant factors influencing biobutanol production was pineapple waste concentration, pH, NH4NO3 and yeast extract with ‘Prob > F’ value was less than 0.0001. Others factors such as temperature and inoculums concentration were not significant for biobutanol production. The suggested parameters for optimal biobutanol production gave predicted value of biobutanol 0.661 g/L with 99.1 % desirability towards biobutanol production. The each variable was suggested as initial pH 4.0, temperature 30 °C, 4.0 g/L yeast extract, 1.0 g/L NH 4NO3 , 10 % (v/v) inoculums concentration and 2.5% (v/v) pineapples waste concentration. The actual value of biobutanol using suggested factors produced 0.671 g/L which was significant to the predicted value. The differences between predicted value and the actual value was only 0.02%.