A Novel Estimation Method for Temperature of Magnetic Nanoparticles Dominated by Brownian Relaxation Based on Magnetic Particle Spectroscopy

Zhongzhou Du, Gaoli Zhao, Zhanpeng Hua, Na Ye, Yi Sun, Wenjie Wu, Haochen Zhang, Longtu Yu, Shijie Han, Haozhe Wang, Wenzhong Liu, Takashi YoshidaIEEE Trans. Instrumentation & Measurement 2024 · DOI

TL;DR

This paper introduces a novel method for accurately estimating the temperature of magnetic nanoparticles (MNPs) by analyzing their AC magnetization response, specifically accounting for the significant impact of Brownian relaxation. By developing a compensation function that bridges the gap between theoretical models (Fokker-Planck and Langevin) and experimental realities, the researchers achieve a temperature estimation accuracy of less than 0.035 K, paving the way for improved MNP-based thermometry in applications like cancer hyperthermia.

What Problem Does This Paper Address?

Accurately measuring temperature is crucial for various biomedical applications, particularly in cancer therapy where magnetic nanoparticles (MNPs) are used to generate localized heat (hyperthermia). This process relies on precisely controlling the temperature of tumor tissues to induce cell death without harming healthy cells. Magnetic Particle Spectroscopy (MPS) offers a non-invasive way to monitor this temperature by analyzing the AC magnetization of MNPs. However, existing models, particularly the Langevin function, often fail to accurately capture the complex dynamics of MNPs when subjected to AC magnetic fields, especially when Brownian relaxation is a dominant factor. Brownian relaxation, a phenomenon where nanoparticles randomly reorient due to thermal motion, significantly influences the magnetization response and its harmonics. Previous approaches either neglected this effect or used simplified models that limited their accuracy and applicability, particularly at higher excitation frequencies. This paper aims to bridge this gap by developing a more robust temperature estimation method that explicitly accounts for Brownian relaxation.

Methods

The core of this work lies in developing a novel temperature estimation model. The authors begin by analyzing the differences between the magnetization response and harmonics predicted by the Fokker-Planck equation (which accounts for Brownian relaxation) and the simpler Langevin function (which largely ignores it). They then investigate how various factors, including Brownian relaxation time, temperature, magnetic field strength, core size, hydrodynamic size, and excitation frequency, influence these magnetization harmonics.

Based on this analysis, they propose a compensation function designed to correct the discrepancies between the Fokker-Planck and Langevin models, incorporating the influence of these key factors. This compensation function is then integrated into a temperature estimation model that utilizes the Langevin function. The model's parameters, specifically the amplitudes of the first and third harmonics of the MNP magnetization, are used to derive the temperature. The researchers employ a least squares algorithm to solve this model and estimate the temperature. The experimental setup involves a magnetic nanothermometer system that generates an AC magnetic field and measures the resulting magnetization harmonics of MNPs.

Datasets

The paper does not explicitly mention using pre-existing datasets for training. Instead, the researchers perform extensive simulations using MATLAB to model the behavior of MNPs under various conditions. These simulations are based on the Fokker-Planck and Langevin equations and are used to analyze the relationships between different physical parameters and the resulting magnetization harmonics.

For experimental validation, they use commercially available MNPs from Ocean NanoTech (SHP-30), which are iron oxide nanoparticles with carboxylic acid groups. The properties of these MNPs (e.g., core diameter, hydrodynamic size, saturation magnetization) are specified for both simulations and experiments.

Key Insights and Ideas

  • Brownian relaxation is a critical factor: The study highlights that neglecting Brownian relaxation leads to significant inaccuracies in temperature estimation, especially at higher excitation frequencies. The Fokker-Planck equation provides a more accurate description of MNP behavior under these conditions.
  • A compensation function is key: By developing a compensation function that quantifies the difference between the Fokker-Planck and Langevin models and incorporates various influencing factors, the authors create a bridge to more accurate MNP thermometry.
  • Harmonic amplitudes are sensitive indicators: The amplitudes of the first and third harmonics of the MNP magnetization are shown to be reliable indicators of temperature, and their relationship with temperature is well-defined within the proposed model.
  • Excitation frequency matters: The accuracy of temperature estimation is shown to be dependent on the excitation frequency, with higher frequencies leading to larger errors, likely due to the increasing influence of Néel relaxation, which is not fully captured by the current model.

Key Results

  • The proposed temperature estimation method achieves a maximum temperature error of less than 0.035 K in the temperature range of 310 K to 320 K.
  • Simulations demonstrate that the difference between the Fokker-Planck and Langevin models becomes more pronounced with increasing Brownian relaxation time, core size, hydrodynamic size, and excitation frequency.
  • The experimental system exhibits good stability, with harmonic amplitudes fluctuating within a narrow range over repeated measurements.
  • The lower detection limit for Fe in the MNP-based sample was found to be 1 µg, meeting the requirements for temperature measurement experiments.
  • The operating frequency range of the system was determined to be 0.2 kHz to 3 kHz.

Limitations and Future Work

The authors acknowledge that their model primarily considers Brownian relaxation and ignores Néel relaxation. They note that as the excitation frequency increases, the proportion of Néel relaxation becomes more significant, leading to increased temperature estimation errors. This suggests that future work should focus on developing models that can accurately describe MNPs dominated by Néel relaxation or a combination of both relaxation mechanisms.

Additionally, the paper mentions that the residual magnetic field can cause errors in the first harmonic calculation, and while a differential coil is used to mitigate this, further improvements in suppressing residual magnetization are desirable. The heating of excitation coils due to excessive magnetic field strength is also identified as a challenge, suggesting the need for cooling devices in future implementations. The proposed method is expected to improve the performance of MNP-based thermometers and find applications in magnetic nanoparticle-mediated hyperthermia.

Summary updated

BibTeX
@article{du2024novel,
  author = {Zhongzhou Du and Gaoli Zhao and Zhanpeng Hua and Na Ye and Yi Sun and Wenjie Wu and Haochen Zhang and Longtu Yu and Shijie Han and Haozhe Wang and Wenzhong Liu and Takashi Yoshida},
  title = {{A Novel Estimation Method for Temperature of Magnetic Nanoparticles Dominated by Brownian Relaxation Based on Magnetic Particle Spectroscopy}},
  journal = {IEEE Trans. Instrumentation \& Measurement},
  year = {2024},
  doi = {10.1109/tim.2024.3472772},
}
Figure