Study On Autocorrelation Characteristics Of Scattered Light From Particles

Dynamic light scattering(DLS)is an effective method to measure submicron and nano particles.The method obtains particle size and distribution by inverting the Autocorrelation Function(ACF)of scattered light intensity signal of particles.Therefore,ACF is the key content in DLS technology.The variation of scattered light in different particle size distribution(PSD)systems is different,and the corresponding ACF also shows different characteristics.Studying the autocorrelation characteristics of scattered light can help to make full use of the particle size information in the ACF and improve the accuracy of PSD inversion.Based on the analysis of the autocorrelation characteristics of scattering light,the inversion algorithm is improved for ultra-low concentration and flowing particle system,and the inversion performance before and after the improvement is analyzed and compared.The main research contents of this thesis include:1.The processing method of scattering light ACF of stationary suspended particles is analyzed.The difference method is used to judge the polydispersity of the corresponding samples according to the difference between the original and reconstructed intensity ACFs.The slope method firstly divides the light intensity ACF into sections.For particle systems with the same average particle size,the slope of the light intensity ACF curve in the first section is also the same.The ACF after logarithm was analyzed,and the polydispersion of the samples is analyzed by comparing the straight line after taking the logarithm of the reconstructed unimodal ACF and the curve after taking the logarithm of the original ACF.And the differential method is used to differentiate the ACF.If the derivative function has two maxima,then the sample to be tested must not be unimodal.This method can be used to identify whether the unimodal small particle samples are mixed with large particle impurities.2.The autocorrelation characteristics of scattered light of ultra-low concentration particle system are studied.According to the influence of the concentration on the ACF,the Gaussian term representing the Brownian motion of particles is changed into the sum of the Non-Gaussian term representing the concentration,and the influence of truncation position on the inversion result is studied.When the concentration of the solution is ultra-low,the amplitude of the scattering field does not meet the Gaussian distribution,and the autocorrelation curve shows additional attenuation in the stage of large delay time.Aiming at the problem that the average particle size obtained by the classical cumulative method is larger than the real value of the ultra-low concentration ACF,the cumulative method combined with non-Gaussian term is proposed,which can eliminate the influence of concentration on the inversion results.The simulation and measurement results show that the cumulants method combined with non-Gaussian term can significantly improve the inversion effect.Finally,the influence of ACF truncation position on cumulative inversion results is studied when the measurement time is insufficient.It is found that the error of the results obtained by truncation at the minimum value of ACF derivative is larger than that obtained by truncation at 0.1 times of its maximum value.3.The autocorrelation characteristics of scattered light in the flowing particle system are studied,and the flowing velocity of particles is calculated from the ACF.According to the contribution of the flow velocity to the ACF,the cumulants method based on particle translation is proposed.By transforming the ACF model and fitting the ACF of the flow particle system,the velocity of the flow sample to be measured can be obtained respectively.Due to the influence of flow velocity,The ACF of scattering light of flowing particles decays faster than that of steady-state suspension,so the recovered particles with traditional cumulants method is smaller than the real particle size.In order to eliminate the influence of flow velocity on the inversion,the diffusion term representing the Brownian motion of particles is changed into the product of the diffusion term and the translation term,and a more accurate average decay rate is obtained,so that the inversion accuracy of particle size is improved.4.Aiming at the problem that the inversion results of flowing particles are not accurate due to the influence of flow velocity,a conditional preoptimization regularization inversion algorithm was proposed.The DLS inversion equation is a typical ill-posed equation.The directional flow of the measured particles will aggravates the ill-conditioned state of the inversion equation,and the error of the inversion results increases significantly with the increase of the flow rate.In order to improve the accuracy of DLS particle size inversion of flowing particles,a regularization inversion combined with conditional preoptimization is used,which can weaken the ill-condition state of the inversion equation and makes regularization less sensitive to velocities.The inversion results of simulated and measured data show that,compared with Tikhonov regularization inversion,the preconditioned Tikhonov regularization can overcome the limitation of regularization in flowing particle size inversion,significantly improve the inversion performance for flowing particles in DLS.