Abstract—In this study concept of experimental design is successfully applied for the determination of optimum condition to produce PP/SWCNT (Polypropylene/Single wall carbon nanotube) nanocomposite. Central composite design as one of experimental design techniques is employed for the optimization and statistical determination of the significant factors influencing on the tensile modulus and yield stress as mechanical properties of this nanocomposite. The significant factors are SWCNT weight fraction and acid treatment time for functionalizing the nanoparticles. Optimum conditions are in 0.7 % of SWCNT weight fraction and 210 min as acid treatment time for 1112.75 ± 28 MPa as maximum tensile modulus and in 216 min and 0.65 % as acid treatment time and SWCNT weight fraction respectively for 40.26 ± 0.3 MPa as maximum yield stress. Also after setting new experiments for test these optimum conditions, found excelent agreement with predicted values. Keywords—Polypropylene, carbon nanotube, nanocomposites, Central composite design I. INTRODUCTION OLYPROPYLENE is one of semi-crystalline thermoplastic materials that have many general uses. Completely obvious those physical properties of these materials such as Strength depend on their structure and morphology. In industry, polypropylene has several uses that these because of existence balance between the properties and the price of this material. So that process ability and low density are the tangible symbols of it. Since the discovery of CNTs by Ijima F. A. Y. Rostamiyan is with the Department of Mechanical Engineering, Islamic Azad university, Sari branch, Sari, Iran (phone: +989111537063;e- mail: yasser.rostamiyan@ yahoo.com). S. B. A. Fereidoon and Y. Rostamiyan are with the Department of Mechanical Engineering, Semnan University, Semnan, Iran (phone: e-mail: ab.fereidoon@ yahoo.com). T. C. Abdollah Omrani is with the Faculty of Chemistry, University of Mazandaran, Babolsar, Iran F. D. D.D.Ganji is with the Department of Mechanical Engineering, Babol University of Technology, Babol. F. E. N. Kordani is with the Department of Mechanical Engineering, Amirkabir University of Technology, Iran. S. F. Vahid Sharifi is with the Faculty of Chemistry, University of Mazandaran, Babolsar, Iran in 1991 [1], a lot of focus on the mechanical, electrical, thermal and chemical properties of these materials have been done. Both experimental and theoretical studies have shows CNTs exceptional properties. The tensile strength, modulus and Poisson's ratio of single walled carbon nanotubes (SWCNTs) have been reported to be in the range of 37- 100 GPa, 1-2 TPa and 0.14-0.28, respectively [2-5]. In our previous works due to the valuable properties of CNTs, some studies have been done on using these materials as nano-filler in polymer matrix [6-10]. The simplest way for length shortening and purification of CNTs is using acid treatment. These shortened CNTs were produced by carboxylic acid groups at the side walls and end of the nanotubes [11-14] and because of interfacial adhesion between CNTs and polymer matrix, cause mechanical and electrical properties Improvement [15]. Statistics is a scientific discipline devoted to the drawing of valid inferences from experimental or observational data. The study of variation, including the construction of experimental designs and the development of models which describe variation, characterizes research activities in the field of statistics [16]. According to surveys conducted in [17], in the titles of most articles used such words "optimization", "development" or "effect of", applied OVAT (one variable at time) model for studying the parameters in their experimental works and few of them like [18] for optimization use various models of "Experimental design". In our previous works [6- 10] as same as OVAT model qualitative study without any special experimental design were done. The optimization has been done by OVAT model can not guarantee that optimum point will be happen and it because that this approach is correct when variables play a role in optimizing are completely independent from each other. By comparing the results obtained from OVAT model and experimental design model can be said that [14], the experimental design considers the interaction between variables, while OVAT doesn't; the experimental design provides a global knowledge of experiment domain, while OVAT provide a local knowledge, the number of tests needed by an OVAT is greater than the number of experiments performed with experimental design approach, the quality of responses obtained by experimental design is depend on various parameters such as domain of Experimental and Statistical Study of Nonlinear Effect of Carbon Nanotube on Mechanical Properties of Polypropylene Composites Y. Rostamiyan, A. Fereidoon, Abdollah Omrani, D.Ganji, N. Kordani, Vahid Sharifi P World Academy of Science, Engineering and Technology International Journal of Mechanical and Mechatronics Engineering Vol:5, No:9, 2011 1806 International Scholarly and Scientific Research & Innovation 5(9) 2011 scholar.waset.org/1307-6892/9823 International Science Index, Mechanical and Mechatronics Engineering Vol:5, No:9, 2011 waset.org/Publication/9823