| This work investigates the combined effects of processing parameters on electrospun fiber morphology as well as the processing parameter-to-electrospun fiber diameter relationship. In fact, electrospinning is a process by which nanofibers can be produced through spinning under the driving of an electric potential. Due to their extremely high surface to weight ratio, electrospun nanofibers exhibit special properties that have opened up a wide range of potential applications. Their morphology (fiber diameter and its distribution) is heavily dependent on processing parameters and solution properties and ambient parameters. Fiber diameter is regarded as the most important morphological property of electrospun fiber as it is the main parameter for quality control. Small fiber diameter and higher fiber uniformity are desired in many applications. Most of works have focused on the impacts of a particular factor separately on the electrospun fiber morphology. However, a study of the effects of combined processing parameters on electrospun fiber morphology can help provide insight into how to control and improve the design of the electrospinning process and nanofiber quality. This could also help determine the most important parameters necessary for the production of nanofibers of desired morphology. No many nanofiber morphology prediction systems have been reported yet. In fact, the complexity of the process makes empirical determination of parameter effects very difficult if not impractical. This work attempts to illustrate the interactive effects of processing parameters on electrospun fiber morphology.Chapter one deals with literature review, the objectives of the research and the innovations in this thesis. The literature review covered the researches and developments related to electrospun polymer nanofibers including processing, structure and property characterization, applications and modeling. Recently previous research works, other issues regarding the technology limitations, research challenges, and future trends have also been discussed. The conclusion of the literature review indicated that the contributions of each processing parameter on electrospun nanofiber morphology are difficult to isolate in general terms. All processing parameters interact in electrospun nanofiber production. From the recent previous works, it was observed that empirical techniques have been mainly used to establish the processing parameter-to-electrospun fiber diameter relationship. However, in complex multi-variate problems like electrospinning process, empirical model is not suitable. Therefore, there is still a need for a use of nonparametric approaches that are able to learn the relations between input variables and response directly from the observations without assuming any pre-specified functional form.Chapter two describes newly invented needleless electrospinning methods, namely, splashing, spiral wire coil and two straight rotary wires electrospinning, as well as to use them to produce nanofibers. A comparison between them in terms of electrospun fiber quality and processing parameters is also discussed. Average fiber diameter ranged from196nm to357nm was achieved and the range of standard deviation was found as33.4nm-105.3nm with the splashing method. An average of fiber diameter ranged between202nm-543nm and a relative standard deviation ranged between22.8nm-110.1nm were obtained with the spiral coil electrospinning. An average fiber diameter ranged from267nm to704nm has been achieved while the range of relative standard deviation was found as44.3nm-213.1nm with the two rotary wires methods. The fibers produced by these methods are comparable in diameter. Fibers electrospun with splashing electrospinning are finer than that of both spiral coil and straight wires electrospinning methods. Splashing electrospinning also provided higher production rate than its competitors. Thicker fibers are obtained by straight wires electrospinning methods compared to both splashing and spiral coil methods. Another observation is that higher voltage values between45kV to60kV are necessary for Splashing electrospinning while voltage values ranged between50kV to70kV and voltage values between40kV and60kV are enough for spiral coil and straight wires electrospinning, respectively. Moreover, the three methods could be scalable.Chapter three describes the application of three intelligent methods, namely, hybrid neuro-fuzzy inference systems (ANFIS), support vector machines (SVM) and gene expression programming (GEP) to control electrospinning process and how to use this approach for developing electrospun fiber quality prediction system. These methods have not yet been explored in the past in electrospinning engineering. These models have been applied to the use of electrospinning process parameters to study the relationship between electrospinning processing parameters and electrospun fiber morphology. Processing parameters such as high applied voltage (V), spinning distance (D), polymer solution concentration(C), wire diameter (W) and rotation speed (RS) were used as inputs. Electrospun fiber diameter (MFD) and fiber diameter standard deviation (FSD) were used as outputs. The prediction performances were compared. The SVMs model provided good prediction ability. However, results from both ANFIS and GEP models are also acceptable. The processing parameters were mathematically related with the electrospun fiber properties (fiber diameter and its distribution) using gene expression programming (GEP) technique. Nonlinear mathematical functions were derived based on the processing parameters.Graphs illustrating the relative importance of processing parameters for MFD and FSD were plotted. The models determined the most important processing variables. It was shown that the electrospun fiber morphology is influenced, to a greater or lesser degree, by processing parameters.Chapter four describes the application of intelligent control systems in electrospinning engineering as well as to use these approaches for optimizing processing conditions. A multi-objective optimization method based on gene algorithm (GA) has been proposed for the design and control of electrospinning process. Afterward, using a multi-objective optimization technique based on gene algorithm, optimal conditions were found in such a way that, mean fiber diameter and its distribution to be minimized. One of the most important advantages of the proposed multi-objective formulation is that it obtains several non-dominated solutions allowing the system operator (decision maker) to exercise his personal preference in selecting each of those solutions based on the operating conditions of the system. This advantage has not yet been explored in the past in electrospinning engineering.Combined effects of processing parameters on the nanofiber morphology are discussed in chapter five. The scheme was investigated by varying processing parameters including polymer solution concentration, distance between the electrode and the collector, applied voltage between the electrode and the collector, wire electrode diameter and rotation speed. The combined effects of processing parameters on the resulting fiber morphology were investigated. The analysis shows that in a multiple variable process like electrospinning, the interaction between the different processing parameters played an important role, rather than one parameter separately in obtaining desired nanofibers. Knowing the relative combined effects of processing parameters on fiber morphology should be useful for process control and prediction of electrospun fiber quality as it has been demonstrated in this study.Chapter six summarized the results obtained and briefly discussed the limitations of the research. Conclusion and recommendations were given. |
PDF Full Text Request