Simultaneous Calibration of Size Exclusion Chromatography and Dynamic Light Scattering for the Characterization of Gelatin

A novel method of calibrating size exclusion chromatography (SEC or known as GPC) and dynamic light scattering (DLS or known as QELS or PCS) is proposed for the characterization of gelatin. A conventional calibration of SEC or DLS normally requires a set of narrowly distributed gelatin standards with different molecular weights. By contrast, the new method uses only one broadly distributed gelatin sample to calibrate SEC and DLS simultaneously in a single process in which the intrinsic connection between the measured elution volume in SEC and the translational diffusion coefficient in DLS (Le., both of them are related to the same hydrodynamic volume) is utilized. This new calibration method can also be used for the characterization of other polymers, where the conventional calibration is not applicable. Introduction Gelatin has been widely used in the biochemical, pharmaceutical, food, and photographic industries as a binder, stabilizer, and gelling agent. For example, gelatin is used as an important stabilizer ingredient for pulverulent formulations of carotenoids and vitamin A. Quality control of gelatin is very important for its various applications. Characterization of the molecular weight distribution of gelatin is an essential part of quality control. Different methods, such as ultracentrifugation,l viscosimetry? size exclusion chromatography (SEC)? osmometry$ and static light scattering: have been used to characterize gelatin. For characterizing molecular weight distribution, SEC is a convenient and established method. However, there are a number of difficulties associated with the use of SEC to characterize the molecular weight distribution of gelatin. One of them is to calibrate an SEC column. In a conventional calibration method, the elution volumes of a set of narrowly distributed standards with known molecular weights are measured. However, it is very difficult to obtain such a set of gelatin standards in practice. Therefore, one hae to use other methods to calibrate the SEC columns for the characterization of gelatin. Most of the reported calibration methods”’O require at least two polymers with different molecular weights or one sample with either two different molecular weight averages or one molecular weight plus intrinsic viscosity data which implies the use of one additional instrument, such as an osmometer or a viscometer. In this article, we report a newly developed method that simultaneously calibrates both SEC and DLS by using only one broadly distributed gelatin, wherein the intrinsic connection between the elution volume in SEC and the translational diffusion coefficient in DLS (Le., they are related to the same polymer hydrodynamic volume) is utilized. Basic Principles An elution volume distribution C( V) and a translational diffusion coefficient distribution G(D) can be respectively measured in SEC and DLS. We can convert C(V) or G(D) into the molecular weight distribution with the calibration (1) V = A + B log(M) or D = k , , i P D (2) where A, B, k D , and CYD are the calibration constants. It i.e., log@) = lo&,) aD log(M) 0024-9297/93/2226-5423$04.00/0 should be noted that in eqs 1 and 2 we have assumed that both Vand log@) are linear functions of log(M), Le., the first-order approximation, because this wil l simplify, but not affect, the following discussions. However, if the sample has a special molecular weight distribution or V and log@) cannot be linearly scaled by log(M), eqs 1 and 2 have to be properly modified. In that case, additional information about the molecular weight distribution and the dependence of V and log@) on log(M) is required. The main task of the calibration is to find A and B or k~ and ag. Traditionally, the two-sample methods have been used to calibrate SEC or DLS. In this paper, we present a new method to obtain A, B, k D , and C ~ D simultaneously by using only one broadly distributed polymer sample. The principle of this one-sample method is in the following: by combining eqs 1 and 2, we have V = A + B log@) (3) where A = A + B log(kD)/arD and B = -B/cYD. Further, by taking the square of both sides of eq 3, we obtain v2 = A’ + 2AB log@) + B2 log2(D) (4) After integrating both sides of eqs 3 and 4, we have (V) = A + B (log@)) (5) and ( v2) = A’ + 2AB (log@) ) + B2 ( log2(D)) (6) where