廣義介電泳中虛部柯莫氏因子的量測及其在推導細胞介電性質的應用

Abstract

Generalized dielectrophoresis (gDEP), including conventional dielectrophoresis (cDEP), electrorotation (ER), and travelling wave dielectrophoresis (twDEP), are effective tools in the characterization and manipulation of particles and cells as they can exert force and torque to a particle via a non-contact manner by applying an appropriate electric field. The force and torque in twDEP and the torque in ER depend on the imaginary part of the Clausius-Mossotti factor, Ki, which plays a role in gDEP similar to viscosity in fluid mechanics. However, Ki cannot be evaluated according to its definition for many particles such as biological cells. Thus it is important to measure Ki in many practical situations. Current method in the literature for the measurement of Ki was based on torque balance using ER requiring (i) three-dimensional electrode system or optical tweezer for particle confinement, and (ii) numerical calculation of the electric field for the evaluation of the torque balance. The primary goal of this dissertation is to propose effective methods and fabricated the associated devices for measuring Ki, without the need of complicated devices (three-dimensional electrode system or optical tweezer) for particle confinement, and/or numerical calculation of the electric field. Two groups of methods were developed: (i) the twDEP method based on the force balance of a uniformly translating particle in a designed micro channel, and (ii) the ER method based on the torque balance of a confined particle in a designed electrorotation chamber. The devices of both methods were fabricated using standard flat microelectromechanical (MEMS) techniques, and thus complicated devices were avoided. Furthermore, analytical solution was employed, and thus numerical calculation of the electric field was not required for the twDEP method. Both the single and the dual frequency operations were applied for both methods. In particular, the dual frequency operation is necessary (for keeping the particle from adhering to the electrodes) for the measurement of Ki at a selected frequency when the particle exhibits positive dielectrophoresis. The force and torque balance of particles in the devices were studied in details analytically and numerically for different conditions, for understanding the physical reasoning of the methods, the design geometry of the devices, and the operating parameters of the experiments. The method and the device for the experiments were validated using sephadex particles in KCL solution and lipid particles in de-ionized water, in which values of Ki can be evaluated according to the theoretical definition. The methods were then demonstrated by measuring detailed Ki spectra (variation of Ki with applied electric frequency) of two lung cancer cells, CL1-0 and CL1-5, and one Colorectal cancer cell, Colo205, for medium conductivity from 0.01 – 1.2 S/m, using different developed methods. In general, the twDEP method is more effective than the ER method as numerical calculation of the electric field is not required, and a measurement can be completed within tens of seconds as the twDEP method employs a flow through device. However, the ER method has the advantage that it can be applied to measure Ki at lower applied frequency. Devices with four and eight electrodes were proposed and studied for the ER method. The differences of Ki for different cells indicate that Ki spectra may probably be served as phenotype for characterization and manipulation of cells of similar origin or at different metastatic stages, which may have significant medical application. The experimental results of Ki together with Kr in the literature were applied to derive cell (membrane and cytoplasm) dielectric properties, through the minimization of the difference between the experimental and theoretical values of the Clausius-Mossotti factor. It was found that even though the curve of regression using only Ki can fit accurately the experimental Ki data, the dielectric properties based on such a regression curve cannot fit the experimental data of Kr. Both the Kr and Ki spectra are required simultaneously for deriving appropriate cell dielectric properties.

廣義介電泳，含傳統介電泳、電旋轉及旅波介電泳，皆為粒子(細胞)操控及特性描述的有效工具，此因其可用非接觸方式對微粒施予力與力矩。旅波介電泳中的力與力矩、及電旋轉中的力矩，皆與虛部科莫氏因子(Ki)有關。Ki在旅波介電泳與電旋轉中扮演的角色如同流體力學中黏滯係數在剪應力與應變率間之關係。然而對於很多粒子(如細胞)，Ki無法從定義直接算得，因此量測Ki極為重要。現今文獻中是採用電旋轉方式量測Ki，其需要使用到:(i)立體電極系統或光鉗以將粒子侷限在特定區域內，(ii)數值方法來得到電場以計算介電力矩。 本論文旨在提出簡單且有效的方法、並製作相對設備來量測Ki因子，而不需使用立體電極或光鉗及數值計算。本文共發展兩系列方法:(i)旅波介電泳方法，其原理乃穩定平移中粒子旅波介電泳力與黏滯阻力的平衡。(ii)電旋轉方法，其原理乃穩定旋轉中粒子介電泳力矩與黏滯力矩的平衡。這兩種方法之晶片製造皆只需標準微機電製程技術、且無需複雜設備進行量測。此外，旅波介電泳方法中，因電場有解析解而不需數值計算。單頻與雙頻操作皆應用至此兩法中，其中雙頻率操作在實部科莫氏因子(Kr)為正的情況下是必需的，以避免粒子吸附在電極上。 所發展方法及設備先利用sephadex粒子在氯化鉀溶液與酯類(lipid)粒子在二次去離子水中兩種可算出理論值的實驗來作驗證。再利用相同方法量測兩種肺癌細胞(CL1-0與CL1-5)及大腸癌細胞(Colo205)在不同導電度溶液下(0.01–1.2 S/m)的Ki頻譜(Ki與電頻率的關係)。一般來說，旅波介電泳法比電旋轉法來得有效率，原因是其無需數值計算，且每一量測時間只為數十秒。但電旋轉法的可操作頻率較廣而為其長處，四電極與八電極設計均有在電旋轉法中採用。 不同細胞具不同的Ki頻譜，顯示其在醫療應用上或許可以被利用作不同期別細胞的辨識及操作。細胞之介電性質(含細胞膜及原生質的介電率及導電率)一般不易量測，但可利用本文所量得之Ki頻譜，配合文獻所得的Kr頻譜，將理論值與實驗值作最佳化處理，將其介電性質回歸得出。結果顯示僅由Ki頻譜回歸並不合理、而須將Ki與Kr頻譜一起考慮，方可求得較合理的細胞介電性質。