Principal component analysis on factors affecting the hyperthermia performance of magnetic nanoparticles

Magnetic hyperthermia is a complementary and novel type of cancer treatment that utilizes magnetic nanoparticles as heating agents with the intent of destructing cancer cells without damaging healthy tissues by locally raising the temperature of the tumor site. The effectiveness of magnetic nanoparticles in hyperthermia applications is quantified by the specific absorption rate (SAR). The average size, concentration as well as magnetic field frequency and intensity have been demonstrated to influence the corresponding hyperthermia performance of nanoparticles. Despite the numerous studies conducted on the impact of these factors on the hyperthermia performance of magnetic nanoparticles, a thorough statistical evaluation of factors and empirical models for estimation remains to be limited. Consequently, the essential aim of this study is to synthesize magnetic nanoparticles using diverse synthesis methods in the presence of various polymers and surfactants, with a view to reveal the effect of particle size, concentration and surface coating on the heating performance under various externally applied magnetic field frequency and field intensity combinations. The heating curves and PCA demonstrated that an increase in amplitude and frequency of the magnetic field resulted in an increase in SAR. Additionally, a positive correlation was observed between concentration and temperature increase. Nonetheless, the impact of concentration on SAR was determined to be more complex. Furthermore, the temperature enhancement of the partial oxidation-based nanoparticles was found to be more pronounced than that of the coprecipitation-based counterparts where the former exhibited mostly field amplitude, while the latter exhibited mostly frequency-dependent variations. Finally, the application of non-linear regression to the measured data revealed the unfeasibility of generating a comprehensive and holistic model. Conversely, empirical models have been developed for particular nanoparticle systems, allowing for the estimation of temperature changes with reduced absolute percent errors.