Materials and Sample Preparation

Polymer samples of PS-h standards (MW = 5700, 8300, 10200, 11600, 13000, 28000, 51500, 115000, 200000, 257900, 515000, 533000) and and PS-d standards (Mw = 10500, 11200, 119000, 205000, and 520000 of polydispersity generally MW/MN < 1.06 were obtained from Polymer Laboratories, USA. PS-h (Mw = 22200, 73000) was obtained from Polymer Source, Canada. PDMS-h (Mw -22500, 47700, and 79900 of polydispersity Mw/Mn < 1.03) and PDMS-d (Mw -27600, Mw/Mn < 1.11 and 49600, Mw/Mn < 1.03) were synthesized and characterized at the Max Planck Institute for Polymer Research, Germany. PDMS-d (75600, Mw/Mn < 1.3) was purchased from Polymer Standards Service GmbH, Mainz, Germany. Protonated MCH-h and BrBz-h as well as deutero substituted solvents CH-d, AC-d, MCH-d, and BrBz-d with D/(H+D)=0.995 were obtained from Sigma Chemical and dried over molecular sieves prior to preparing solutions. CO2 (SFC purity 99.99%) was obtained from Matheson Gas products, Inc., USA.

PS - MCH and PS - CH solutions were prepared at the critical volume fraction ^ c of the polymer which was assigned in accordance with the empirical power law ^ c=6.65(MW)-o.379 [ 14] and c=7.69(Mw)-o.385 [ 15], respectively. PS AC-d solutions were prepared at near critical concentration C=20.3 wt%. Solutions of PDMS in SC CO2 and in BrBz were prepared at the overlap concentration C*=0.1397 and 0.0959 g/ml for Mw 22500 and 47700, respectively calculated using C*=3Mw/4 n (Re)3Na, where Na is the Avogadro number. The overlap concentration is approximately the same as the critical concentration of a polymer solution [16]. The liquid solutions, PS-CH, PS -MCH, and PDMS-BrBz, were filtered through Millipore filters (0.22 ^m) and thoroughly homogenized either in a 10 mm cylindrical optical cell (DLS) or in a 2 mm thick quartz cells (SANS)~ 10 degrees above the © temperature of PS-CH (40 OC), PS-MCH (~ 78 oC) PDMS-BrBz (~ 68 OC) [17]. In PDMS - SC CO2 experiments, the polymer was loaded into a stainless steel cylindrical cell having three optically polished sapphire windows one of which was located at the scattering angle 9 =90 O and used for the DLS measurements. The samples were pressurized with CO2 at T=50 oC, stirred thoroughly for 10 min until a transparent homogeneous solutions was obtained after which the variation of the static correlation length (SANS) or the dynamic correlation length (DLS) were measured at various temperatures and/or pressures. The temperature of the samples was controlled to better than ±0.l K. The range of T covered in the experiment was from the © temperature down to Tc of each solution. The critical temperature of phase demixing was identified as a sharp maximum in the integral neutron count rate as a function ofT (SANS) or as the appearance of the meniscus (DLS). This allowed an estimate of Tc to better than ±0.2 K.

Small-Angle Neutron Scattering

Measurements were performed on the 30-m SANS spectrometer at the Oak Ridge National Laboratory. The neutron wavelength was X -4.75 A (AX/X-0.06) and the range of scattering vectors was 0.005<©-4rnX-i sin 9 <0.05 A-1, where 29 is the scattering angle. The data were corrected for scattering from the empty cells, detector sensitivity and beam-blocked background and placed on an absolute scale using pre-calibrated secondary standards after radial averaging to produce functions of the intensity I vs. Q. Procedures for subtracting the incoherent background have been described previously [ 18]. The functions Ss(Q,T) were obtained using Eqs. 1,2 (PS-AC-d, PDMS-BrBz-d, PDMS-SC CO2) and Eq.5 (PS-CH-d) and used to extract the (z-averaged) Rg(T) by fitting the Debye form factor Eq.6. The functions St(Q,T) were obtained using Eqs.1,3 and used to extract the correlation length ^(T) using Eq.9.

Dynamic Light Scattering

DLS measurements were made in the self-beating (homodyne) mode using a setup the details of which were described previously [19]. The cylindrical scattering cells with liquid polymer solutions were sealed and immersed in a large-diameter thermostated bath containing decalin placed at the axis of the goniometer. Measurements were made at different angles in the range 30 to 135o as a function of temperature (80 < T/OC < 0). Analysis of the data was performed by fitting the experimentally measured g2(t), the normalized intensity autocorrelation function, which is related to the electrical field correlationfunction,gi(t)bytheSiegertrelation [20]:

where p is a factor accounting for deviation from ideal correlation. For polydisperse samples, g 1 (t) can be written as the inverse Laplace transform (ILT) of the distribution of relaxation times, xA(x):

0 0

Post a comment