The power dissipation level in the shake flask is indeed different from that of CSTR. It cannot be estimated independently like in the case of the impeller and hence, the observed/measured KLa values obtained in the shaker flask are used to get equivalent values in CSTR and no attempt has been made to develop a correlation for KLa for shake flask like the one developed in the case of CSTR.
The production of l-PAC was scaled up to the 5 l bioreactor from shake flask with same KLa as the criteria. First the KLa was estimated in the shake flask at different times during both, growth of the yeast and the biotransformation process. Initially the KLa in the shake flask were estimated by keeping them on shaker at 240 rpm and monitoring the dissolved oxygen concentration at regular intervals for both the media at different stages of growth. Before the estimation of the dissolved oxygen the broths were purged with nitrogen to bring down the DO to the minimum level. This was treated as zero time, DO was then measured with time, at regular intervals to estimate KLa during the various phases of growth and biotransformation. Similar values of mass transfer coefficients (KLa) were maintained in the 5 l bioreactor for the growth and biotransformation using the correlations developed earlier ( and ) and by the manipulation of power input and aeration rates. For the estimation of the power input and aeration rates, the oxygen respiration rates were estimated for growth and biotransformation, so that the desired DO level could be maintained with gas-flow and impeller speed combination giving a specific value of KLa.