Impurity Induced Slowing oF Nucleation in Emulsified Liquids
- PDF / 1,417,925 Bytes
- 12 Pages / 417.6 x 639 pts Page_size
- 17 Downloads / 203 Views
the determination of the size distribution for the emulsions is a large source of error in nucleation rate measurements [3]. Advances in emulsion synthesis techniques now makes it possible to create nearly uniformly sized emulsion particles. The average deviation in diameter from the mean size is only 10 to 15% [7-9]. By using such emulsions with narrow size distributions, the determination of the rate constant for any given emulsion radius is considerably improved. Also, this makes it possible to better test the predicted volume scaling of the nucleation rate. A number of groups have studied alkane nucleation through the use of emulsion samples [3,10-15]. The earliest work by Turnbull and Cormia [3] focused on C16, C17, C1S, C24 and C32. Theirs is the first evidence that alkane nucleation is unusual. First, they noted that there seemed to be an unusual spread in the melting temperatures. Because they performed isothermal nucleation studies they were sensitive to this spread in melting temperature. To analyze the nucleation behavior in this situation they focused on the early-time data with transformed fraction n < 0.5. This narrowed focus was meant to isolate the behavior of the sharp-melting, majority phase. The second anomaly in alkane nucleation is the ease with which the alkanes nucleate. Stated in terms of reduced undercooling, AT, = (Tmn - TN)/Tm, where TN is the point where the nucleation rate becomes significant and Tm is the thermodynamic melting temperature, ATr for the alkanes is about 0.05 whereas that for other materials is from 0.2 to 0.5 [16]. Turnbull and Cormia's analysis of the nucleation behavior in terms of the classical nucleation model showed that the nucleation barrier is small (9.64 mJ/m 2 for C1S; a similar value was found for C16 [10,11,16]) but that the pre-exponential factor is in rough agreement with that calculated from classical nucleation theory (experimental value = 1037.35±2 m- 3s-1 for C18). The small barrier accounts for the small undercooling temperature. The agreement of the pre-exponential with that from classical nucleation theory is itself unusual. Other materials typically exhibit values several orders of magnitude larger in value than the classical value [4]. In this paper, we report the results of a refined experimental and theoretical investigation of the nucleation rate in emulsified hexadecane (C16). The quality of experimental data has been significantly improved by using nearly monodisperse emulsions in a well controlled thermal environment and by using x-ray scattering to accurately monitor the volume of nucleated droplets during crystallization. Data were obtained for both fixed and linearly increasing undercooling as a function of time, henceforth referred to as isothermal nucleation and linear cooling, respectively. Thermodynamic melting curves have also been obtained for the samples in order to assess the influence of impurities introduced during the emulsification procedure. The theoretical analysis has concurrently been refined to account for the entire time-d
Data Loading...
