Real-time magnetic nanothermometry: The use of magnetization of magnetic nanoparticles assessed under low frequency triangle-wave magnetic fields Jing Zhong, Wenzhong Liu, Ling Jiang, Ming Yang, and Paulo Cesar Morais Citation: Review of Scientific Instruments 85, 094905 (2014); doi: 10.1063/1.4896121 View online: http://dx.doi.org/10.1063/1.4896121 View Table of Contents: http://scitation.aip.org/content/aip/journal/rsi/85/9?ver=pdfcov Published by the AIP Publishing Articles you may be interested in Maneuvering the chain agglomerates of colloidal superparamagnetic nanoparticles by tunable magnetic fields Appl. Phys. Lett. 105, 183108 (2014); 10.1063/1.4901320 Temperature dependent dissipation in magnetic nanoparticles J. Appl. Phys. 115, 17B301 (2014); 10.1063/1.4853155 Relaxation of biofunctionalized magnetic nanoparticles in ultra-low magnetic fields J. Appl. Phys. 113, 043911 (2013); 10.1063/1.4789009 Formation and magnetic manipulation of periodically aligned microchains in thin plastic membranes J. Appl. Phys. 112, 083927 (2012); 10.1063/1.4759328 Real-time measurement of Brownian relaxation of magnetic nanoparticles by a mixing-frequency method Appl. Phys. Lett. 98, 213702 (2011); 10.1063/1.3595273
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 128.114.34.22 On: Sat, 22 Nov 2014 07:00:00
REVIEW OF SCIENTIFIC INSTRUMENTS 85, 094905 (2014)
Real-time magnetic nanothermometry: The use of magnetization of magnetic nanoparticles assessed under low frequency triangle-wave magnetic fields Jing Zhong,1,2 Wenzhong Liu,1,2,a) Ling Jiang,1 Ming Yang,1 and Paulo Cesar Morais1,3 1
School of Automation, Huazhong University of Science and Technology, Wuhan 430074, China Key Laboratory of Image Processing and Intelligent Control, Huazhong University of Science and Technology, Wuhan 430074, China 3 Universidade de Brasília, Brasília DF 70910-900, Brazil 2
(Received 24 August 2014; accepted 8 September 2014; published online 24 September 2014) In this study, we propose and demonstrate the usefulness of employing time-varying magnetization of a magnetic nanoparticle (MNP) based sample, induced by low frequency (f = 25 Hz) triangular-wave magnetic field, to achieve the approach of real-time recording of magnetization curve, which allows precise and noninvasive temperature probing with real-time performance. Moreover, the present report introduces the design and performed the test of a detection system for accurate and real-time recording of the magnetization curve of MNP-based samples. We found that by employing the magnetization curve of a magnetic fluid sample containing magnetite nanoparticles of about 30 nm in diameter the accuracy of the temperature probing is about 0.32 K (0.1% relative accuracy), with response time of 1 s. Furthermore, an increase in response time from 1 to 8 s improves the accuracy of temperature probing from 0.32 to 0.20 K. Finally, we envisage that breakthroughs in clinical hyperthermia, targeted drug delivery and basic cell research can be accomplished while using the approach reported in this study. © 2014 AIP Publishing LLC. [http://dx.doi.org/10.1063/1.4896121] I. INTRODUCTION
In recent years the material platform based on magnetic nanoparticles (MNPs) has attracted increasing attention in regard to the wide diversity of applications involved, ranging from the medical and biomedical fields, e.g., in clinical magnetohyperthermia,1–5 thermally and remotely controlled drug delivery,6–9 and thermogenesis in living cells10–13 up to industrial applications, as for instance in transformers’ refrigeration14, 15 and miniaturization.16, 17 In these regards, accurate, noninvasive and real-time monitoring, and control of temperature is of great significance to all the above-mentioned emerging technologies’ performance. Several optical approaches10, 18–21 have been reported for noninvasively, precise and in situ temperature probing, to name a few quantum dot nanothermometer10 and nanometer-scale thermometry based on nanodiamonds.18 In particular, the thermoplasmonic properties of nanosized particles revealed the correlation between maximum temperature increase and the reduction of the inter-particle distance,22, 23 which has the potential for noninvasive temperature probing. This approach offers the opportunity for local temperature probing while positioning the nanoparticles in the vicinity of a cell membrane. However, for the medical and biomedical applications, many of the existing approaches are limited to either in vitro temperature sensing or for superficial tissues’ probing. Therefore, the development of noninvasive, precise and real-time tools for in vivo temperature probing is of great interest to the medical and biomedical fields. a) Author to whom correspondence should be addressed. Electronic mail:
[email protected].
0034-6748/2014/85(9)/094905/8/$30.00
Magnetic nanothermometry10, 18, 24, 25 is one of the most promising technologies, which has the potential of assessing signal deep from tissues for in vivo temperature probing, precisely and noninvasively. Recently, our studies24, 26 reported a novel magnetic nanothermometry approach, employing susceptibility or magnetization data remotely recorded from MNPs. The MNP magnetization scales monotonically with temperature, thus being able to provide accurate and noninvasive temperature measurements, as long as an efficient algorithm for translation of the magnetic data into temperature is available. The literature26 reports that discrete magnetization data (Mi × Hi data) recorded under DC magnetic fields, measured in a commercial SQUID VSM and analyzed using the first-order Langevin function as well as the inverse calculation method based on the least square error algorithm, were employed to probe temperature at ultrahigh accuracy of about 0.017 ◦ C. The novel method has been proved to have potential for precise and noninvasive in vivo temperature probing. However, the magnetic nanothermometer based on the magnetic data recorded from standard SQUID VSM systems is far from being useful in practice. For routine applications, such as clinical magnetohyperthermia and thermally controlled drug delivery protocols, one of the key factors for temperature measurement is the real-time performance. The targeted tissue temperature should be controlled in real time, in the range of interest, such as 43–45 ◦ C.1 Thus, the appropriated thermometer should incorporate fast response times for real-time performance, in addition to higher sensitivity and precision. The reported magnetic nanothermometer employs SQUID VSM to record discrete magnetization (M) versus applied field (H) data (Mi × Hi data), which results in poor real-time performance for temperature (T) probing. The actual
85, 094905-1
© 2014 AIP Publishing LLC
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 128.114.34.22 On: Sat, 22 Nov 2014 07:00:00
094905-2
Zhong et al.
SQUID VSM technology takes about 10 min for recording a discrete magnetization curve comprising 20 points connecting the DC magnetic field and the corresponding magnetization values,26 meaning temperature probing with a response time of around 10 min. Therefore, it is crucial to propose a new approach and to design the corresponding system for real-time recording of the MNP magnetization data in order to fulfill the requirements of real-time magnetic nanothermometry. Differently from the reported magnetic nanothermometry and in order to achieve the real-time temperature probing, the present study proposes to employ time-varying magnetization data recorded at low frequency using triangular-wave magnetic field for real-time recording of the magnetization curve. Moreover, this study also reports on the design and the performance test of a home-made system for the generation of the low frequency triangular-wave magnetic field and the detection of the MNP’s time-varying magnetization. In addition R platform to algorithms implemented within the LabVIEW the proposed system allows for real-time magnetization curve measurement, providing precise and noninvasive temperature probing with real-time performance.
II. FUNDAMENTAL THEORY
In order to successfully perform real-time temperature (T) probing using MNPs we employed alternating low frequency applied magnetic field in the triangular-wave mode. Using this applied field while operating around room temperature the MNPs’ magnetization (M) follows the applied magnetic field (H), thus allowing real-time assessing to the Mi × Hi data. Figure 1 schematically shows the M × H monotonic correlation, indicating that one full cycle of applied magnetic field allows one to record twice the MNPs’ magnetization curve while in the superparamagnetic (SPM) regime. Moreover, the applied triangular-wave mode magnetic field operating at a frequency f allows for the measurement of the magnetization curve with a response time performance as short as 1/2f. Accordingly, this approach of temperature probing using the M × H data recorded from the MNPs has the same real-time performance (1/2f). A model describing the temperature dependence of the magnetization of the MNPs in the SPM regime is required for the magnetic nanothermometry accomplishment. As shown in Eq. (1) the first-order Langevin function is used to describe the static magnetic behavior, M(H, T), of non-interacting and
FIG. 1. Schematic for the magnetization curve of MNPs acquired at low frequency triangular-wave magnetic field.
Rev. Sci. Instrum. 85, 094905 (2014)
monodisperse MNPs in the SPM regime:27 Ms V H kT − , M (H, T ) = φMs coth kT Ms V H
(1)
where φ is the MNP fraction in the sampling volume (number of particles per unit volume), Ms is MNP’s saturation magnetization, V is the particle’s volume, k is the Boltzmann’s constant, and T is the absolute temperature. Equation (1) will be used to support the approach for real-time temperature probing. In general, in time-varying applied magnetic fields, the MNPs’ dynamic magnetization may present a rather complex susceptibility, which requires a complex model to describe the material’s response. To overcome the complexity of such a general scenario27, 28 the time-varying triangular-wave mode applied magnetic field was set to operate at low frequency (f) (around 25 Hz) and at field amplitude up to 200 Oe while keeping the MNPs within the SPM regime. Under these conditions the first-order Langevin function describes well the magnetic behavior of MNPs with an average diameter up to 30 nm, around room temperature. From the magnetization measurements, a set of magnetization points Mi , as well as the corresponding applied magnetic fields Hi , are acquired. Using the Mi × Hi data file as well as the first-order Langevin function the following equations can be written: ⎧
⎪ M1 = x coth yH1 − yH1 ⎪ ⎪ 1 ⎪ ⎪ ⎪ ⎨ M = x coth yH − 1 2 2 yH2 , (2) .. ⎪ ⎪ ⎪. ⎪ ⎪ ⎪ ⎩ M = x coth yH − 1 n
n
yHn
where x = φMs and y = Ms V /kT. Solving Eq. (2) allows for temperature probing using MNPs. III. SYSTEM DESIGN
The magnetic nanothermometer requires an excitation/detection system for recording the MNP’s magnetization curve, which includes generation of low frequency triangularwave magnetic fields and measurement of the time-varying magnetization. In this study, solenoid and pick-up coils are employed to generate the low frequency triangular-wave magnetic fields and to measure the time-varying magnetization, respectively. The system design overview is shown in Fig. 2. The wave-shape of the driving voltage is edited in the LabVIEW platform, outputted by the multifunctional DAQ
FIG. 2. Overview of the system’s design for temperature probing under the excitation of low frequency triangular-wave magnetic fields.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 128.114.34.22 On: Sat, 22 Nov 2014 07:00:00
094905-3
Zhong et al.
Rev. Sci. Instrum. 85, 094905 (2014)
NI-USB 6356 (National Instruments Corp., Austin, Texas, USA) and amplified to drive the solenoid coil using the linear power amplifier BOP 36-12ML (KEPCO, INC., Flushing, NY, USA). In addition, a pair of pick-up coils is employed to detect the MNP’s time-varying magnetization, which is sampled by NI-USB 6356 and processed by the LabVIEW software. The details are included in Secs. III A and III B.
sample resistance (Rs ) is placed in series with the solenoid coil for measuring the triangular-wave current (i), which is represented by the output voltage (Uout ) of the sample resistance (Rs ), as shown in Fig. 3(a). According to Fig. 3(a), the relationship between Uin and i is expressed as
A. Generating triangular-wave magnetic fields
It is assumed that the generated magnetic field (H) in the solenoid coil scales linearly with the driving current (i): H = Ki. In order to generate triangular-wave magnetic fields with amplitude H0 and frequency f the solenoid coil current (i) is expressed as ⎧ 4f H0 t/K, 0 ≤ t < 3/4f ⎪ ⎪ ⎪ ⎨
2H0 − 4f H0 t /K, 1/4f ≤ t < 3/4f . (4) i= ⎪
⎪ ⎪ ⎩ 4f H0 t − 4H0 /K, 3/4f ≤ t < 1/f
A low frequency triangular-wave magnetic field is applied in order to record the time-varying magnetization of MNPs for magnetic nanothermometry. The triangular-wave magnetic field generation requires a corresponding triangularwave current driving the solenoid coil. However, under the time-varying driven voltage the solenoid coil is described through its resistance RH and an inductance LH characteristics, which shows nonlinear behavior between the input voltage Uin and the corresponding driven current i. To generate the triangular-wave current (i) the solenoid coil input voltage (Uin ) is previously edited in the LabVIEW platform. Herein, to determine the generated triangular-wave magnetic field a
Uin = RH i + LH
B. Measuring the magnetization curve
As proposed in the present report magnetic nanothermometry requires the measurement of the magnetization curve of MNPs. Moreover, while measuring the magnetization curve requirements of real-time operation and accuracy directly and significantly affect the performance of mag-
(3)
Substituting Eq. (4) into Eq. (3), the input voltage (Uin ) is rewritten as
⎧ 4f H 0 ⎪ [(Rs + RH )t + LH ], 0 ≤ t < 3/4f ⎪ K ⎪ ⎪ ⎨ 2H (R +R ) 4f H 0 s H − K 0 [(Rs + RH )t + LH ], 1/4f ≤ t < 3/4f . Uin = K ⎪ ⎪ 4H (R +R ) 4f H ⎪ ⎪ ⎩ K 0 [(Rs + RH )t + LH ] − 0 Ks H , 3/4f ≤ t < 1/f
Using Eq. (5) and the measured solenoid coil resistance (RH ) and inductance (LH ) the input voltage Uin (as well as the output voltage Uout ) for generating the triangular-wave magnetic field is calculated, as shown in Fig. 3(a). Note that Uin is neither a regular sinusoidal wave nor a triangular wave-shape. Therefore, the power amplifier employed for driving the solenoid coil should be able to amplify any input signal. Figure 3(b) shows the measured wave-shapes of Uin and Uout . It indicates that with the edited input voltage (Uin ) the solenoid coil current is actually triangular waveshaped, which will allow for the generation of a triangularwave-shaped magnetic field. Our experimental results are in agreement with our calculation, meaning that the designed system allows for efficient generation of triangular-wave magnetic fields. For this study, the frequency of the triangularwave magnetic field was set at 25 Hz.
di + RS i. dt
(5)
netic nanothermometer. Notice that the measured magnetization curve is obtained from the time-varying magnetization, which follows (in phase) the applied triangular-wave magnetic field. In order to combine higher temperature probing precision with real-time performance the entire time-varying magnetization should be detected with high signal-to-noise ratio (SNR). However, while designing an accurate detection system for magnetization measurements two key factors should be taken into account.29 The first one is to amplify the output signal of the pick-up coils induced by the time-varying magnetization. Therefore, a precise detection system should be designed for measuring the MNPs’ time-varying magnetization with high SNR. The second key point is to suppress the influence of the applied triangular-wave time-varying magnetic field (residual magnetic field) onto the pick-up coils. In practice, the measured output signal of the pick-up coils is induced by a vector sum (coupled) of the residual magnetic field plus the time-varying magnetization associated to the MNPbased sample.29 Consequently, to account for these two key factors our detection system integrates two parts: hardware and software. Figure 4 shows schematically the hardware part of the detection system which integrates a series of analog signal processing for the measurement of the time-varying magnetization. The first step toward suppressing the influence of the
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 128.114.34.22 On: Sat, 22 Nov 2014 07:00:00
094905-4
Zhong et al.
Rev. Sci. Instrum. 85, 094905 (2014)
FIG. 3. Schematically (a) the equivalent circuit for driving the solenoid coil placed in series with the sample resistance Rs (note the overview of Uin and Uout ) and (b) the measured Uin (black solid line) and Uout (red solid line).
residual magnetic field is to wire up the pair of pick-up coils oppositely to each other. The two impedance-matched pickup coils are symmetrically arranged to ensure that the signal induced by the triangular-wave applied magnetic field onto both coils should be cancelled as precisely as possible, which contributes to suppress the influence of the residual magnetic field with a high rejection ratio. Afterwards, to reconstruct the magnetization signal an integrator is employed to balance the derivative action of the pick-up coils. Herein, noise significantly affects the performance of the integrator, such as the SNR and the stability of the output signal. In order to measure the MNP magnetization with both high SNR plus high rejection ratio of the residual magnetic field the output signal from the pick-up coils is pre-amplified, notch filtered, and band filtered. The notch filter suppresses the interference of the noise with the outlet frequency (50 Hz in our case) whereas the band filter with a low cutoff frequency of 0.33 Hz and high cutoff
frequency of 1.8 kHz suppresses the noise close to DC and at higher frequencies (see Fig. 4). An additional amplifier is employed to adjust the output signal of the integrator whereas the amplifier output signal is digitized at 100k samples per second (SPS) using a DAQ (National Instrument, NI-USB 6356). However, the hardware part by itself is unable to completely suppress the influence of the residual magnetic field, as it will be discussed later on in this report. To improve the efficiency in suppressing the interference of the residual magnetic field onto the measured magnetization curve the detection system includes a secondary suppressor realized within the LabVIEW platform. The method implemented for the secondary suppressor records the pick-up coils signal (reference signal) induced by the residual magnetic field alone, before inserting the MNP-based sample into the equipment. The reference signal is then subtracted from the recorded output signal after inserting the MNP-based sample into the measuring system. The details for digital signal processing realized within LabVIEW to record the discrete magnetization curve (Mi × Hi data) are schematically shown in Fig. 5. Recording the magnetization curve requires simultaneous recording of the applied triangular-wave time-varying magnetic field (H) and the corresponding time-varying magnetization (M). Herein, H and M are acquired from the output signal of the sample resistance (Rs ) and from the pickup coils, respectively. The digital signal processing includes five units; the first one (averager) performs n-periodic signal average and m-point signal average. The averager improves the SNRs of the measured M and H, which contributes to the increase of the rejection ratio of the residual magnetic field while improving the accuracy of the measured magnetization. The second unit is a frequency spectrometer, which measures the spectrum of the response signal using a digital phase sensitive detector (DPSD) algorithm whereas the third unit is a reconstructor, which employs the measured spectrum to reconstruct the response signal by using Fourier series. Herein, the DPSD algorithm allows for the spectrum measurement with a high accuracy, thus contributing to suppress noise while providing an accurate measurement of the magnetization curve. The forth unit is a secondary suppressor performing further rejection of the residual magnetic field. It allows for an accurate recording of the time-varying magnetization. The fifth unit is a re-sampler, which outputs the discrete magnetization curve (Mi × Hi data) with inputting discrete magnetic fields Hi . The re-sampler output, meaning the discrete magnetization curve (Mi × Hi data), is acquired from the average of every 1/4-periodic wave-shape of the
FIG. 4. Analog signal processing for the measurement of time-varying magnetization.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 128.114.34.22 On: Sat, 22 Nov 2014 07:00:00
094905-5
Zhong et al.
Rev. Sci. Instrum. 85, 094905 (2014)
FIG. 5. Schematic representation of the digital signal processing for the measurement of the magnetization curve.
triangular-wave applied magnetic field and time-varying magnetization. Therefore, the detection system allows for the recording of a discrete magnetization curve with real-time performance of n/f. It indicates that at the low frequency triangular-wave applied magnetic field approach herein described the real-time performance for the magnetic nanothermometer can reach, in principle, n/f.
C. Analysis of the system performance
Figure 6(a) shows the one-period measurement of the residual magnetic field after the preliminary suppression whereas Fig. 6(b) shows the same measurement after the secondary suppression. Without inserting the MNP-based sample into the instrument the output signal is the residual magnetic field, with amplitude of about 1.1 V, as shown in Fig. 6(a). After the secondary suppression, performed within the LabVIEW, the amplitude of the residual magnetic field is about 0.38 mV. It means that the secondary suppressor allows for a rejection of the residual magnetic field with a ratio of about 65 dB. Figure 7 shows the measured applied magnetic field and the corresponding time-varying MNP magnetization. Using the triangular-wave magnetic field with amplitude of 200 Oe the output signal induced by the time-varying MNP magnetization has amplitude of about 0.33 V. The effect of the secondary suppression is clearly observed while comparing the
data shown in Figs. 6(b) and 7(b), from which we found the ratio between the residual magnetic field and the corresponding MNP magnetization of about 1/1000. Figure 8 shows the input discrete magnetic fields (Hi ) and the corresponding output discrete magnetization data (Mi ) as well as the background noise of the detection system. The measured Mi × Hi data sheet allows for the construction of a set of equations, like Eq. (1), for magnetic nanothermometry. Notice that the precise recording of the residual magnetic field by the hardware component contributes to the further suppression performed by the LabVIEW software. Therefore, all analog and digital signal processing significantly affect the rejection ratio of the residual magnetic field as well as the measurement accuracy of the MNP-based sample magnetization. Our experimental results show that the detection system herein reported is able to efficiently and precisely measure the MNP-based sample magnetization curve in real time mode, which has the potential for precise temperature probing with real-time performance. IV. RESULTS AND DISCUSSIONS
To validate the performance of the real-time temperature probing using time-varying magnetization induced by triangular-wave magnetic fields the discrete magnetization curves (Mi × Hi data) of a MNP-based sample were recorded at different temperatures. Measurements were performed in
FIG. 6. Measurement of the residual magnetic field (a) after the preliminary suppression and before the secondary suppression and (b) after the secondary suppression.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 128.114.34.22 On: Sat, 22 Nov 2014 07:00:00
094905-6
Zhong et al.
Rev. Sci. Instrum. 85, 094905 (2014)
FIG. 7. Measurement of the (a) applied magnetic field and (b) corresponding time-varying MNP magnetization.
the real-time regime and the inverse calculation method, based on the least square solution of Eq. (1), was used to assess the corresponding temperatures. The MNP-based sample used in this study was a commercial magnetic fluid (sample SOR-30) purchased from Ocean Nanotechnology Ltd. Corp. (Springdale, USA), consisting of magnetite nanoparticles with an average diameter of approximately 30 nm. The nanoparticles were surface-coated with oleic acid and dispersed in an organic solvent at a concentration of 25 mgFe/ml. The real-time discrete magnetization curves of the magnetic fluid sample were recorded at different temperatures using the home-made system described above. The experimental setup allows for one-time measurement of the magnetization curve using the two one-periodic wave-shapes of time-varying magnetization and applied magnetic fields.
FIG. 8. Measured MNP magnetization curve (red symbols) and the corresponding background noise (black symbols) after the two steps of suppression (hardware and software). Solid lines are only guide to the eyes.
It means that while applying the triangular-wave magnetic field at a frequency of 25 Hz the real-time measurement of the magnetization curve performs as good as 0.04 s. Figure 9 shows the experimental data for the real-time temperature probing recorded in a 1 s run, in the range from 310 to 320 K. As shown in Fig. 9, real-time measurement of MNPbased samples’ magnetization curve is indeed a very promising approach for real-time temperature probing. Figure 9(a) indicates that the probed temperature (PT) using the magnetization recorded from a MNP-based sample is in very good agreement with the reference temperature (RT) provided by
FIG. 9. Experimental results of the real-time temperature probing using MNPs’ magnetization recorded under triangular-wave magnetic fields. (a) The probed temperature (PT) data (red symbols) and the reference temperature (RT) curve (black solid line) provided by an optical fibre thermometer and (b) the corresponding temperature probing error. Red-dotted lines in (b) are only guide to the eyes.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 128.114.34.22 On: Sat, 22 Nov 2014 07:00:00
094905-7
Zhong et al.
FIG. 10. Measured (solid symbols) and calculated (dashed lines) magnetization curves at different temperatures (310 K, 315 K, and 320 K).
an optical fibre thermometer. Moreover, from the data shown in Fig. 9(b) the temperature probing approach using the magnetization curve has a maximum error of about 0.76 K, with a standard deviation of 0.32 K. Using the probed temperatures we were able to calculate the corresponding magnetization curves of the MNP-based sample, at different temperatures. Figure 10 shows the measured (solid symbols) and the calculated (dashed lines) magnetization curves at different temperatures. The data presented in Fig. 10 indicates that the agreement between the measured and calculated magnetization curve is excellent. Therefore, the real-time measurement of the MNP-based sample’s magnetization curve, realized by the time-varying magnetization and the corresponding applied magnetic field, allows for temperature probe in real time mode. Additionally, the temperature probing accuracy upon real-time performance was also investigated. In this study, the n-periodic average algorithm used in the digital averager determines the response time for the magnetization curve measurement as well as the real-time performance of the temperature probing. To study the accuracy of the temperature probing while using different real-time performance the magnetization curves were acquired using different response times, namely 1, 2, 4, 6, and 8 s. Our findings on the temperature probing error versus response time are shown in Fig. 11. Figure 11(a) shows the maximum error whereas Fig. 11(b) shows the standard deviation of the temperature probing error. The data shown in Fig. 11 indicate that when the response time of the magnetization curve measurement increases up to 8 s the maximum temperature probing error decreases down to 0.34 K and the standard deviation is 0.20 K. This finding means that a decrease in real-time performance by a factor of 8 leads to a 2-fold increase in the temperature probing accuracy. Therefore, as expected, the decrease in real-time performance improves the temperature probing accuracy. It is therefore of great importance to select the best suitable real-time performance for different applications in order to allow for the best accuracy in temperature probing.
Rev. Sci. Instrum. 85, 094905 (2014)
FIG. 11. The errors of temperature probing using the magnetization curves of the magnetic fluid sample at different response times. (a) The maximum error in temperature probing whereas (b) the standard deviation of the temperature probing error.
V. CONCLUSION
In conclusion, we report on the design of a system to measure the magnetization curve of magnetic nanoparticlebased samples using a low frequency triangular-wave applied magnetic field in order to perform noninvasive and precise temperature probing in the real time recording mode. Moreover, a prototype which allows for the generation of the low frequency triangular-wave applied magnetic field and precise measurement of magnetization curves is herein introduced and tested. The low frequency (f = 25 Hz) triangular-wave applied magnetic field induces a time-varying magnetization in the magnetic-based sample which follows the typical behavior described by the Langevin’s function. The present approach allows for real-time measurement of the magnetization curve, as well as real-time temperature measurement. Based on the Langevin’s function while describing the magnetization curve of magnetic nanoparticle-based samples we found that our approach of temperature probing provides an accuracy of about 0.32 K (0.1% relative accuracy) with a real-time performance of 1 s. Furthermore, with an increase in the response time from 1 to 8 s the temperature probing error decreases down from 0.32 to 0.20 K. Finally, the prototype herein introduced is pioneering in this field while employing low frequency triangular-wave applied magnetic field to record the magnetization curve for remotely and precisely probing the temperature in real time mode. We envisage that the content of our report can potentially support breakthroughs in clinical hyperthermia, targeted drug delivery, and basic cell research.
ACKNOWLEDGMENTS
This work was supported by projects of 11104089(NSFC), 2013CFA017(HBSTD), and 2014070504020238(WUHAN).
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 128.114.34.22 On: Sat, 22 Nov 2014 07:00:00
094905-8 1 L.
Zhong et al.
C. Branquinho, M. S. Carrião, A. S. Costa, N. Zufelato, M. H. Sousa, R. Miotto, R. Ivkov, and A. F. Bakuzis, “Effect of magnetic dipolar interactions on nanoparticle heating efficiency: Implications for cancer hyperthermia,” Sci. Rep. 3, 2887 (2013). 2 A. M. Falqueiro, F. L. Primo, P. C. Morais, E. Mosiniewicz-Szablewska, P. Suchocki, and A. C. Tedesco, “Selol-loaded magnetic nanocapsules: A new approach for hyperthermia cancer therapy,” J. Appl. Phys. 109, 07B306 (2011). 3 K. Murase, H. Takata, Y. Takeuchi, and S. Saito, “Control of the temperature rise in magnetic hyperthermia with use of an external static magnetic field,” Phys. Med.: Eur. J. Med. Phys. 29, 624 (2013). 4 F. L. Primo, M. M. A. Rodrigues, A. R. Simioni, Z. G. M. Lacava, P. C. Morais, and A. C. Tedesco, “Photosensitizer-loaded magnetic nanoemulsion for use in synergic photodynamic and magnetohyperthermia therapies of neoplastic cells,” J. Nanosci. Nanotechnol. 8, 5873 (2008). 5 H. F. Rodrigues, F. M. Mello, L. C. Branquinho, N. Zufelato, E. P. SilveiraLacerda, and A. F. Bakuzis, “Real-time infrared thermography detection of magnetic nanoparticle hyperthermia in a murine model under a nonuniform field configuration,” Int. J. Hypertherm. 29, 752 (2013). 6 A. Chan, R. P. Orme, R. A. Fricker, and P. Roach, “Remote and local control of stimuli responsive materials for therapeutic applications,” Adv. Drug Delivery Rev. 65, 497 (2013). 7 T. Hoare, J. Santamaria, G. F. Goya, S. Irusta, D. Lin, S. Lau, R. Padera, R. Langer, and D. S. Kohane, “A magnetically triggered composite membrane for on-demand drug delivery,” Nano Lett. 9, 3651 (2009). 8 M. K. Jaiswal, A. Pradhan, R. Banerjee, and D. Bahadur, “Dual pH and temperature stimuli-responsive magnetic nanohydrogels for thermochemotherapy,” J. Nanosci. Nanotechnol. 14, 4082 (2014). 9 P. Pradhan, J. Giri, F. Rieken, C. Koch, O. Mykhaylyk, M. Döblinger, R. Banerjee, D. Bahadur, and C. Plank, “Targeted temperature sensitive magnetic liposomes for thermo-chemotherapy,” J. Controlled Release 142, 108 (2010). 10 J. M. Yang, H. Yang, and L. W. Lin, “Quantum dot nano thermometers reveal heterogeneous local thermogenesis in living cells,” ACS Nano 5, 5067 (2011). 11 G. Kroemer, L. Galluzzi, and C. Brenner, “Mitochondrial membrane permeabilization in cell death,” Physiol. Rev. 87, 99 (2007). 12 W. Droge, “Free radicals in the physiological control of cell function,” Physiol. Rev. 82, 47 (2002). 13 T. D. Vreugdenburg, C. D. Willis, L. Mundy, and J. E. Hiller, “A systematic review of elastography, electrical impedance scanning, and digital infrared thermography for breast cancer screening and diagnosis,” Breast Cancer Res. Treat. 137, 665 (2013). 14 W. R. Viali, G. B. Alcantara, P. P. C. Sartoratto, M. A. G. Soler, E. Mosiniewicz-Szablewska, B. Andrzejewski, and P. C. Morais, “Investigation of the molecular surface coating on the stability of insulating magnetic oils,” J. Phys. Chem. C 114, 179 (2010). 15 L. A. T. Reino, T. M. Lima, A. S. Costa, A. F. Bakuzis, C. M. B. Santos, P. P. C. Sartoratto, and P. C. Morais, “Investigation of colloidal stability and
Rev. Sci. Instrum. 85, 094905 (2014) insulation characteristics of magnetic oils,” J. Nanosci. Nanotechnol. 12, 9319 (2012). 16 T.-H. Tsai, L.-S. Kuo, P.-H. Chen, D.-S. Lee, and C.-T. Yang, “Applications of ferro-nanofluid on a micro-transformer,” Sensors 10, 8161 (2010). 17 T.-H. Tsai, P.-H. Chen, D.-S. Lee, and C.-T. Yang, “Investigation of electrical and magnetic properties of ferro-nanofluid on transformers,” Nanoscale Res. Lett. 6, 264 (2011). 18 G. Kucsko, P. C. Maurer, N. Y. Yao, M. Kubo, H. J. Noh, P. K. Lo, H. Park, and M. D. Lukin, “Nanometre-scale thermometry in a living cell,” Nature (London) 500, 54 (2013). 19 F. Vetrone, R. Naccache, A. Zamarron, A. Juarranz de la Fuente, F. SanzRodriguez, L. Martinez Maestro, E. Martin Rodriguez, D. Jaque, J. Garcia Sole, and J. A. Capobianco, “Temperature sensing using fluorescent nanothermometers,” ACS Nano 4, 3254 (2010). 20 J. S. Donner, S. A. Thompson, M. P. Kreuzer, G. Baffou, and R. Quidant, “Mapping intracellular temperature using green fluorescent protein,” Nano Lett. 12, 2107 (2012). 21 K. Okabe, N. Inada, C. Gota, Y. Harada, T. Funatsu, and S. Uchiyama, “Intracellular temperature mapping with a fluorescent polymeric thermometer and fluorescence lifetime imaging microscopy,” Nat. Commun. 3, 705 (2012). 22 M. Essone Mezeme, S. Lasquellec, and C. Brosseau, “Subwavelength control of electromagnetic field confinement in self-similar chains of magnetoplasmonic core-shell nanostructures,” Phys. Rev. E 84, 026612 (2011). 23 M. Essone Mezeme and C. Brosseau, “Engineering nanostructures with enhanced thermoplasmonic properties for biosensing and selective targeting applications,” Phys. Rev. E 87, 012722 (2013). 24 J. Zhong, W. Liu, Z. Du, P. C. de Morais, Q. Xiang, and Q. Xie, “A noninvasive, remote and precise method for temperature and concentration estimation using magnetic nanoparticles,” Nanotechnology 23, 075703 (2012). 25 J. B. Weaver, A. M. Rauwerdink, and E. W. Hansen, “Magnetic nanoparticle temperature estimation,” Med. Phys. 36, 1822 (2009). 26 J. Zhong, W. Liu, L. Kong, and P. C. Morais, “A new approach for highly accurate, remote temperature probing using magnetic nanoparticles,” Sci. Rep. (in press). 27 F. Ludwig, D. Eberbeck, N. Lowa, U. Steinhoff, T. Wawrzik, M. Schilling, and L. Trahms, “Characterization of magnetic nanoparticle systems with respect to their magnetic particle imaging performance,” Biomed. Eng.-Biomed. Te. 58, 535 (2013). 28 T. Yoshida and K. Enpuku, “Simulation and quantitative clarification of AC susceptibility of magnetic fluid in nonlinear brownian relaxation region,” Jpn. J. Appl. Phys. 48, 127002 (2009). 29 J. Zhong, W. Liu, M. Zhou, and L. He, “AC magnetization measurement employing a pair of differentiating coils for magnetic nanoparticle noninvasive and remote temperature estimation,” Proceedings of International Instrumentation and Measurement Conference Technology, Graz, Austria, 13–16 May 2012 (IEEE, 2012), pp. 1780–1783.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 128.114.34.22 On: Sat, 22 Nov 2014 07:00:00
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 128.114.34.22 On: Sat, 22 Nov 2014 07:00:00
REVIEW OF SCIENTIFIC INSTRUMENTS 85, 094905 (2014)
Real-time magnetic nanothermometry: The use of magnetization of magnetic nanoparticles assessed under low frequency triangle-wave magnetic fields Jing Zhong,1,2 Wenzhong Liu,1,2,a) Ling Jiang,1 Ming Yang,1 and Paulo Cesar Morais1,3 1
School of Automation, Huazhong University of Science and Technology, Wuhan 430074, China Key Laboratory of Image Processing and Intelligent Control, Huazhong University of Science and Technology, Wuhan 430074, China 3 Universidade de Brasília, Brasília DF 70910-900, Brazil 2
(Received 24 August 2014; accepted 8 September 2014; published online 24 September 2014) In this study, we propose and demonstrate the usefulness of employing time-varying magnetization of a magnetic nanoparticle (MNP) based sample, induced by low frequency (f = 25 Hz) triangular-wave magnetic field, to achieve the approach of real-time recording of magnetization curve, which allows precise and noninvasive temperature probing with real-time performance. Moreover, the present report introduces the design and performed the test of a detection system for accurate and real-time recording of the magnetization curve of MNP-based samples. We found that by employing the magnetization curve of a magnetic fluid sample containing magnetite nanoparticles of about 30 nm in diameter the accuracy of the temperature probing is about 0.32 K (0.1% relative accuracy), with response time of 1 s. Furthermore, an increase in response time from 1 to 8 s improves the accuracy of temperature probing from 0.32 to 0.20 K. Finally, we envisage that breakthroughs in clinical hyperthermia, targeted drug delivery and basic cell research can be accomplished while using the approach reported in this study. © 2014 AIP Publishing LLC. [http://dx.doi.org/10.1063/1.4896121] I. INTRODUCTION
In recent years the material platform based on magnetic nanoparticles (MNPs) has attracted increasing attention in regard to the wide diversity of applications involved, ranging from the medical and biomedical fields, e.g., in clinical magnetohyperthermia,1–5 thermally and remotely controlled drug delivery,6–9 and thermogenesis in living cells10–13 up to industrial applications, as for instance in transformers’ refrigeration14, 15 and miniaturization.16, 17 In these regards, accurate, noninvasive and real-time monitoring, and control of temperature is of great significance to all the above-mentioned emerging technologies’ performance. Several optical approaches10, 18–21 have been reported for noninvasively, precise and in situ temperature probing, to name a few quantum dot nanothermometer10 and nanometer-scale thermometry based on nanodiamonds.18 In particular, the thermoplasmonic properties of nanosized particles revealed the correlation between maximum temperature increase and the reduction of the inter-particle distance,22, 23 which has the potential for noninvasive temperature probing. This approach offers the opportunity for local temperature probing while positioning the nanoparticles in the vicinity of a cell membrane. However, for the medical and biomedical applications, many of the existing approaches are limited to either in vitro temperature sensing or for superficial tissues’ probing. Therefore, the development of noninvasive, precise and real-time tools for in vivo temperature probing is of great interest to the medical and biomedical fields. a) Author to whom correspondence should be addressed. Electronic mail:
[email protected].
0034-6748/2014/85(9)/094905/8/$30.00
Magnetic nanothermometry10, 18, 24, 25 is one of the most promising technologies, which has the potential of assessing signal deep from tissues for in vivo temperature probing, precisely and noninvasively. Recently, our studies24, 26 reported a novel magnetic nanothermometry approach, employing susceptibility or magnetization data remotely recorded from MNPs. The MNP magnetization scales monotonically with temperature, thus being able to provide accurate and noninvasive temperature measurements, as long as an efficient algorithm for translation of the magnetic data into temperature is available. The literature26 reports that discrete magnetization data (Mi × Hi data) recorded under DC magnetic fields, measured in a commercial SQUID VSM and analyzed using the first-order Langevin function as well as the inverse calculation method based on the least square error algorithm, were employed to probe temperature at ultrahigh accuracy of about 0.017 ◦ C. The novel method has been proved to have potential for precise and noninvasive in vivo temperature probing. However, the magnetic nanothermometer based on the magnetic data recorded from standard SQUID VSM systems is far from being useful in practice. For routine applications, such as clinical magnetohyperthermia and thermally controlled drug delivery protocols, one of the key factors for temperature measurement is the real-time performance. The targeted tissue temperature should be controlled in real time, in the range of interest, such as 43–45 ◦ C.1 Thus, the appropriated thermometer should incorporate fast response times for real-time performance, in addition to higher sensitivity and precision. The reported magnetic nanothermometer employs SQUID VSM to record discrete magnetization (M) versus applied field (H) data (Mi × Hi data), which results in poor real-time performance for temperature (T) probing. The actual
85, 094905-1
© 2014 AIP Publishing LLC
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 128.114.34.22 On: Sat, 22 Nov 2014 07:00:00
094905-2
Zhong et al.
SQUID VSM technology takes about 10 min for recording a discrete magnetization curve comprising 20 points connecting the DC magnetic field and the corresponding magnetization values,26 meaning temperature probing with a response time of around 10 min. Therefore, it is crucial to propose a new approach and to design the corresponding system for real-time recording of the MNP magnetization data in order to fulfill the requirements of real-time magnetic nanothermometry. Differently from the reported magnetic nanothermometry and in order to achieve the real-time temperature probing, the present study proposes to employ time-varying magnetization data recorded at low frequency using triangular-wave magnetic field for real-time recording of the magnetization curve. Moreover, this study also reports on the design and the performance test of a home-made system for the generation of the low frequency triangular-wave magnetic field and the detection of the MNP’s time-varying magnetization. In addition R platform to algorithms implemented within the LabVIEW the proposed system allows for real-time magnetization curve measurement, providing precise and noninvasive temperature probing with real-time performance.
II. FUNDAMENTAL THEORY
In order to successfully perform real-time temperature (T) probing using MNPs we employed alternating low frequency applied magnetic field in the triangular-wave mode. Using this applied field while operating around room temperature the MNPs’ magnetization (M) follows the applied magnetic field (H), thus allowing real-time assessing to the Mi × Hi data. Figure 1 schematically shows the M × H monotonic correlation, indicating that one full cycle of applied magnetic field allows one to record twice the MNPs’ magnetization curve while in the superparamagnetic (SPM) regime. Moreover, the applied triangular-wave mode magnetic field operating at a frequency f allows for the measurement of the magnetization curve with a response time performance as short as 1/2f. Accordingly, this approach of temperature probing using the M × H data recorded from the MNPs has the same real-time performance (1/2f). A model describing the temperature dependence of the magnetization of the MNPs in the SPM regime is required for the magnetic nanothermometry accomplishment. As shown in Eq. (1) the first-order Langevin function is used to describe the static magnetic behavior, M(H, T), of non-interacting and
FIG. 1. Schematic for the magnetization curve of MNPs acquired at low frequency triangular-wave magnetic field.
Rev. Sci. Instrum. 85, 094905 (2014)
monodisperse MNPs in the SPM regime:27 Ms V H kT − , M (H, T ) = φMs coth kT Ms V H
(1)
where φ is the MNP fraction in the sampling volume (number of particles per unit volume), Ms is MNP’s saturation magnetization, V is the particle’s volume, k is the Boltzmann’s constant, and T is the absolute temperature. Equation (1) will be used to support the approach for real-time temperature probing. In general, in time-varying applied magnetic fields, the MNPs’ dynamic magnetization may present a rather complex susceptibility, which requires a complex model to describe the material’s response. To overcome the complexity of such a general scenario27, 28 the time-varying triangular-wave mode applied magnetic field was set to operate at low frequency (f) (around 25 Hz) and at field amplitude up to 200 Oe while keeping the MNPs within the SPM regime. Under these conditions the first-order Langevin function describes well the magnetic behavior of MNPs with an average diameter up to 30 nm, around room temperature. From the magnetization measurements, a set of magnetization points Mi , as well as the corresponding applied magnetic fields Hi , are acquired. Using the Mi × Hi data file as well as the first-order Langevin function the following equations can be written: ⎧
⎪ M1 = x coth yH1 − yH1 ⎪ ⎪ 1 ⎪ ⎪ ⎪ ⎨ M = x coth yH − 1 2 2 yH2 , (2) .. ⎪ ⎪ ⎪. ⎪ ⎪ ⎪ ⎩ M = x coth yH − 1 n
n
yHn
where x = φMs and y = Ms V /kT. Solving Eq. (2) allows for temperature probing using MNPs. III. SYSTEM DESIGN
The magnetic nanothermometer requires an excitation/detection system for recording the MNP’s magnetization curve, which includes generation of low frequency triangularwave magnetic fields and measurement of the time-varying magnetization. In this study, solenoid and pick-up coils are employed to generate the low frequency triangular-wave magnetic fields and to measure the time-varying magnetization, respectively. The system design overview is shown in Fig. 2. The wave-shape of the driving voltage is edited in the LabVIEW platform, outputted by the multifunctional DAQ
FIG. 2. Overview of the system’s design for temperature probing under the excitation of low frequency triangular-wave magnetic fields.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 128.114.34.22 On: Sat, 22 Nov 2014 07:00:00
094905-3
Zhong et al.
Rev. Sci. Instrum. 85, 094905 (2014)
NI-USB 6356 (National Instruments Corp., Austin, Texas, USA) and amplified to drive the solenoid coil using the linear power amplifier BOP 36-12ML (KEPCO, INC., Flushing, NY, USA). In addition, a pair of pick-up coils is employed to detect the MNP’s time-varying magnetization, which is sampled by NI-USB 6356 and processed by the LabVIEW software. The details are included in Secs. III A and III B.
sample resistance (Rs ) is placed in series with the solenoid coil for measuring the triangular-wave current (i), which is represented by the output voltage (Uout ) of the sample resistance (Rs ), as shown in Fig. 3(a). According to Fig. 3(a), the relationship between Uin and i is expressed as
A. Generating triangular-wave magnetic fields
It is assumed that the generated magnetic field (H) in the solenoid coil scales linearly with the driving current (i): H = Ki. In order to generate triangular-wave magnetic fields with amplitude H0 and frequency f the solenoid coil current (i) is expressed as ⎧ 4f H0 t/K, 0 ≤ t < 3/4f ⎪ ⎪ ⎪ ⎨
2H0 − 4f H0 t /K, 1/4f ≤ t < 3/4f . (4) i= ⎪
⎪ ⎪ ⎩ 4f H0 t − 4H0 /K, 3/4f ≤ t < 1/f
A low frequency triangular-wave magnetic field is applied in order to record the time-varying magnetization of MNPs for magnetic nanothermometry. The triangular-wave magnetic field generation requires a corresponding triangularwave current driving the solenoid coil. However, under the time-varying driven voltage the solenoid coil is described through its resistance RH and an inductance LH characteristics, which shows nonlinear behavior between the input voltage Uin and the corresponding driven current i. To generate the triangular-wave current (i) the solenoid coil input voltage (Uin ) is previously edited in the LabVIEW platform. Herein, to determine the generated triangular-wave magnetic field a
Uin = RH i + LH
B. Measuring the magnetization curve
As proposed in the present report magnetic nanothermometry requires the measurement of the magnetization curve of MNPs. Moreover, while measuring the magnetization curve requirements of real-time operation and accuracy directly and significantly affect the performance of mag-
(3)
Substituting Eq. (4) into Eq. (3), the input voltage (Uin ) is rewritten as
⎧ 4f H 0 ⎪ [(Rs + RH )t + LH ], 0 ≤ t < 3/4f ⎪ K ⎪ ⎪ ⎨ 2H (R +R ) 4f H 0 s H − K 0 [(Rs + RH )t + LH ], 1/4f ≤ t < 3/4f . Uin = K ⎪ ⎪ 4H (R +R ) 4f H ⎪ ⎪ ⎩ K 0 [(Rs + RH )t + LH ] − 0 Ks H , 3/4f ≤ t < 1/f
Using Eq. (5) and the measured solenoid coil resistance (RH ) and inductance (LH ) the input voltage Uin (as well as the output voltage Uout ) for generating the triangular-wave magnetic field is calculated, as shown in Fig. 3(a). Note that Uin is neither a regular sinusoidal wave nor a triangular wave-shape. Therefore, the power amplifier employed for driving the solenoid coil should be able to amplify any input signal. Figure 3(b) shows the measured wave-shapes of Uin and Uout . It indicates that with the edited input voltage (Uin ) the solenoid coil current is actually triangular waveshaped, which will allow for the generation of a triangularwave-shaped magnetic field. Our experimental results are in agreement with our calculation, meaning that the designed system allows for efficient generation of triangular-wave magnetic fields. For this study, the frequency of the triangularwave magnetic field was set at 25 Hz.
di + RS i. dt
(5)
netic nanothermometer. Notice that the measured magnetization curve is obtained from the time-varying magnetization, which follows (in phase) the applied triangular-wave magnetic field. In order to combine higher temperature probing precision with real-time performance the entire time-varying magnetization should be detected with high signal-to-noise ratio (SNR). However, while designing an accurate detection system for magnetization measurements two key factors should be taken into account.29 The first one is to amplify the output signal of the pick-up coils induced by the time-varying magnetization. Therefore, a precise detection system should be designed for measuring the MNPs’ time-varying magnetization with high SNR. The second key point is to suppress the influence of the applied triangular-wave time-varying magnetic field (residual magnetic field) onto the pick-up coils. In practice, the measured output signal of the pick-up coils is induced by a vector sum (coupled) of the residual magnetic field plus the time-varying magnetization associated to the MNPbased sample.29 Consequently, to account for these two key factors our detection system integrates two parts: hardware and software. Figure 4 shows schematically the hardware part of the detection system which integrates a series of analog signal processing for the measurement of the time-varying magnetization. The first step toward suppressing the influence of the
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 128.114.34.22 On: Sat, 22 Nov 2014 07:00:00
094905-4
Zhong et al.
Rev. Sci. Instrum. 85, 094905 (2014)
FIG. 3. Schematically (a) the equivalent circuit for driving the solenoid coil placed in series with the sample resistance Rs (note the overview of Uin and Uout ) and (b) the measured Uin (black solid line) and Uout (red solid line).
residual magnetic field is to wire up the pair of pick-up coils oppositely to each other. The two impedance-matched pickup coils are symmetrically arranged to ensure that the signal induced by the triangular-wave applied magnetic field onto both coils should be cancelled as precisely as possible, which contributes to suppress the influence of the residual magnetic field with a high rejection ratio. Afterwards, to reconstruct the magnetization signal an integrator is employed to balance the derivative action of the pick-up coils. Herein, noise significantly affects the performance of the integrator, such as the SNR and the stability of the output signal. In order to measure the MNP magnetization with both high SNR plus high rejection ratio of the residual magnetic field the output signal from the pick-up coils is pre-amplified, notch filtered, and band filtered. The notch filter suppresses the interference of the noise with the outlet frequency (50 Hz in our case) whereas the band filter with a low cutoff frequency of 0.33 Hz and high cutoff
frequency of 1.8 kHz suppresses the noise close to DC and at higher frequencies (see Fig. 4). An additional amplifier is employed to adjust the output signal of the integrator whereas the amplifier output signal is digitized at 100k samples per second (SPS) using a DAQ (National Instrument, NI-USB 6356). However, the hardware part by itself is unable to completely suppress the influence of the residual magnetic field, as it will be discussed later on in this report. To improve the efficiency in suppressing the interference of the residual magnetic field onto the measured magnetization curve the detection system includes a secondary suppressor realized within the LabVIEW platform. The method implemented for the secondary suppressor records the pick-up coils signal (reference signal) induced by the residual magnetic field alone, before inserting the MNP-based sample into the equipment. The reference signal is then subtracted from the recorded output signal after inserting the MNP-based sample into the measuring system. The details for digital signal processing realized within LabVIEW to record the discrete magnetization curve (Mi × Hi data) are schematically shown in Fig. 5. Recording the magnetization curve requires simultaneous recording of the applied triangular-wave time-varying magnetic field (H) and the corresponding time-varying magnetization (M). Herein, H and M are acquired from the output signal of the sample resistance (Rs ) and from the pickup coils, respectively. The digital signal processing includes five units; the first one (averager) performs n-periodic signal average and m-point signal average. The averager improves the SNRs of the measured M and H, which contributes to the increase of the rejection ratio of the residual magnetic field while improving the accuracy of the measured magnetization. The second unit is a frequency spectrometer, which measures the spectrum of the response signal using a digital phase sensitive detector (DPSD) algorithm whereas the third unit is a reconstructor, which employs the measured spectrum to reconstruct the response signal by using Fourier series. Herein, the DPSD algorithm allows for the spectrum measurement with a high accuracy, thus contributing to suppress noise while providing an accurate measurement of the magnetization curve. The forth unit is a secondary suppressor performing further rejection of the residual magnetic field. It allows for an accurate recording of the time-varying magnetization. The fifth unit is a re-sampler, which outputs the discrete magnetization curve (Mi × Hi data) with inputting discrete magnetic fields Hi . The re-sampler output, meaning the discrete magnetization curve (Mi × Hi data), is acquired from the average of every 1/4-periodic wave-shape of the
FIG. 4. Analog signal processing for the measurement of time-varying magnetization.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 128.114.34.22 On: Sat, 22 Nov 2014 07:00:00
094905-5
Zhong et al.
Rev. Sci. Instrum. 85, 094905 (2014)
FIG. 5. Schematic representation of the digital signal processing for the measurement of the magnetization curve.
triangular-wave applied magnetic field and time-varying magnetization. Therefore, the detection system allows for the recording of a discrete magnetization curve with real-time performance of n/f. It indicates that at the low frequency triangular-wave applied magnetic field approach herein described the real-time performance for the magnetic nanothermometer can reach, in principle, n/f.
C. Analysis of the system performance
Figure 6(a) shows the one-period measurement of the residual magnetic field after the preliminary suppression whereas Fig. 6(b) shows the same measurement after the secondary suppression. Without inserting the MNP-based sample into the instrument the output signal is the residual magnetic field, with amplitude of about 1.1 V, as shown in Fig. 6(a). After the secondary suppression, performed within the LabVIEW, the amplitude of the residual magnetic field is about 0.38 mV. It means that the secondary suppressor allows for a rejection of the residual magnetic field with a ratio of about 65 dB. Figure 7 shows the measured applied magnetic field and the corresponding time-varying MNP magnetization. Using the triangular-wave magnetic field with amplitude of 200 Oe the output signal induced by the time-varying MNP magnetization has amplitude of about 0.33 V. The effect of the secondary suppression is clearly observed while comparing the
data shown in Figs. 6(b) and 7(b), from which we found the ratio between the residual magnetic field and the corresponding MNP magnetization of about 1/1000. Figure 8 shows the input discrete magnetic fields (Hi ) and the corresponding output discrete magnetization data (Mi ) as well as the background noise of the detection system. The measured Mi × Hi data sheet allows for the construction of a set of equations, like Eq. (1), for magnetic nanothermometry. Notice that the precise recording of the residual magnetic field by the hardware component contributes to the further suppression performed by the LabVIEW software. Therefore, all analog and digital signal processing significantly affect the rejection ratio of the residual magnetic field as well as the measurement accuracy of the MNP-based sample magnetization. Our experimental results show that the detection system herein reported is able to efficiently and precisely measure the MNP-based sample magnetization curve in real time mode, which has the potential for precise temperature probing with real-time performance. IV. RESULTS AND DISCUSSIONS
To validate the performance of the real-time temperature probing using time-varying magnetization induced by triangular-wave magnetic fields the discrete magnetization curves (Mi × Hi data) of a MNP-based sample were recorded at different temperatures. Measurements were performed in
FIG. 6. Measurement of the residual magnetic field (a) after the preliminary suppression and before the secondary suppression and (b) after the secondary suppression.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 128.114.34.22 On: Sat, 22 Nov 2014 07:00:00
094905-6
Zhong et al.
Rev. Sci. Instrum. 85, 094905 (2014)
FIG. 7. Measurement of the (a) applied magnetic field and (b) corresponding time-varying MNP magnetization.
the real-time regime and the inverse calculation method, based on the least square solution of Eq. (1), was used to assess the corresponding temperatures. The MNP-based sample used in this study was a commercial magnetic fluid (sample SOR-30) purchased from Ocean Nanotechnology Ltd. Corp. (Springdale, USA), consisting of magnetite nanoparticles with an average diameter of approximately 30 nm. The nanoparticles were surface-coated with oleic acid and dispersed in an organic solvent at a concentration of 25 mgFe/ml. The real-time discrete magnetization curves of the magnetic fluid sample were recorded at different temperatures using the home-made system described above. The experimental setup allows for one-time measurement of the magnetization curve using the two one-periodic wave-shapes of time-varying magnetization and applied magnetic fields.
FIG. 8. Measured MNP magnetization curve (red symbols) and the corresponding background noise (black symbols) after the two steps of suppression (hardware and software). Solid lines are only guide to the eyes.
It means that while applying the triangular-wave magnetic field at a frequency of 25 Hz the real-time measurement of the magnetization curve performs as good as 0.04 s. Figure 9 shows the experimental data for the real-time temperature probing recorded in a 1 s run, in the range from 310 to 320 K. As shown in Fig. 9, real-time measurement of MNPbased samples’ magnetization curve is indeed a very promising approach for real-time temperature probing. Figure 9(a) indicates that the probed temperature (PT) using the magnetization recorded from a MNP-based sample is in very good agreement with the reference temperature (RT) provided by
FIG. 9. Experimental results of the real-time temperature probing using MNPs’ magnetization recorded under triangular-wave magnetic fields. (a) The probed temperature (PT) data (red symbols) and the reference temperature (RT) curve (black solid line) provided by an optical fibre thermometer and (b) the corresponding temperature probing error. Red-dotted lines in (b) are only guide to the eyes.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 128.114.34.22 On: Sat, 22 Nov 2014 07:00:00
094905-7
Zhong et al.
FIG. 10. Measured (solid symbols) and calculated (dashed lines) magnetization curves at different temperatures (310 K, 315 K, and 320 K).
an optical fibre thermometer. Moreover, from the data shown in Fig. 9(b) the temperature probing approach using the magnetization curve has a maximum error of about 0.76 K, with a standard deviation of 0.32 K. Using the probed temperatures we were able to calculate the corresponding magnetization curves of the MNP-based sample, at different temperatures. Figure 10 shows the measured (solid symbols) and the calculated (dashed lines) magnetization curves at different temperatures. The data presented in Fig. 10 indicates that the agreement between the measured and calculated magnetization curve is excellent. Therefore, the real-time measurement of the MNP-based sample’s magnetization curve, realized by the time-varying magnetization and the corresponding applied magnetic field, allows for temperature probe in real time mode. Additionally, the temperature probing accuracy upon real-time performance was also investigated. In this study, the n-periodic average algorithm used in the digital averager determines the response time for the magnetization curve measurement as well as the real-time performance of the temperature probing. To study the accuracy of the temperature probing while using different real-time performance the magnetization curves were acquired using different response times, namely 1, 2, 4, 6, and 8 s. Our findings on the temperature probing error versus response time are shown in Fig. 11. Figure 11(a) shows the maximum error whereas Fig. 11(b) shows the standard deviation of the temperature probing error. The data shown in Fig. 11 indicate that when the response time of the magnetization curve measurement increases up to 8 s the maximum temperature probing error decreases down to 0.34 K and the standard deviation is 0.20 K. This finding means that a decrease in real-time performance by a factor of 8 leads to a 2-fold increase in the temperature probing accuracy. Therefore, as expected, the decrease in real-time performance improves the temperature probing accuracy. It is therefore of great importance to select the best suitable real-time performance for different applications in order to allow for the best accuracy in temperature probing.
Rev. Sci. Instrum. 85, 094905 (2014)
FIG. 11. The errors of temperature probing using the magnetization curves of the magnetic fluid sample at different response times. (a) The maximum error in temperature probing whereas (b) the standard deviation of the temperature probing error.
V. CONCLUSION
In conclusion, we report on the design of a system to measure the magnetization curve of magnetic nanoparticlebased samples using a low frequency triangular-wave applied magnetic field in order to perform noninvasive and precise temperature probing in the real time recording mode. Moreover, a prototype which allows for the generation of the low frequency triangular-wave applied magnetic field and precise measurement of magnetization curves is herein introduced and tested. The low frequency (f = 25 Hz) triangular-wave applied magnetic field induces a time-varying magnetization in the magnetic-based sample which follows the typical behavior described by the Langevin’s function. The present approach allows for real-time measurement of the magnetization curve, as well as real-time temperature measurement. Based on the Langevin’s function while describing the magnetization curve of magnetic nanoparticle-based samples we found that our approach of temperature probing provides an accuracy of about 0.32 K (0.1% relative accuracy) with a real-time performance of 1 s. Furthermore, with an increase in the response time from 1 to 8 s the temperature probing error decreases down from 0.32 to 0.20 K. Finally, the prototype herein introduced is pioneering in this field while employing low frequency triangular-wave applied magnetic field to record the magnetization curve for remotely and precisely probing the temperature in real time mode. We envisage that the content of our report can potentially support breakthroughs in clinical hyperthermia, targeted drug delivery, and basic cell research.
ACKNOWLEDGMENTS
This work was supported by projects of 11104089(NSFC), 2013CFA017(HBSTD), and 2014070504020238(WUHAN).
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 128.114.34.22 On: Sat, 22 Nov 2014 07:00:00
094905-8 1 L.
Zhong et al.
C. Branquinho, M. S. Carrião, A. S. Costa, N. Zufelato, M. H. Sousa, R. Miotto, R. Ivkov, and A. F. Bakuzis, “Effect of magnetic dipolar interactions on nanoparticle heating efficiency: Implications for cancer hyperthermia,” Sci. Rep. 3, 2887 (2013). 2 A. M. Falqueiro, F. L. Primo, P. C. Morais, E. Mosiniewicz-Szablewska, P. Suchocki, and A. C. Tedesco, “Selol-loaded magnetic nanocapsules: A new approach for hyperthermia cancer therapy,” J. Appl. Phys. 109, 07B306 (2011). 3 K. Murase, H. Takata, Y. Takeuchi, and S. Saito, “Control of the temperature rise in magnetic hyperthermia with use of an external static magnetic field,” Phys. Med.: Eur. J. Med. Phys. 29, 624 (2013). 4 F. L. Primo, M. M. A. Rodrigues, A. R. Simioni, Z. G. M. Lacava, P. C. Morais, and A. C. Tedesco, “Photosensitizer-loaded magnetic nanoemulsion for use in synergic photodynamic and magnetohyperthermia therapies of neoplastic cells,” J. Nanosci. Nanotechnol. 8, 5873 (2008). 5 H. F. Rodrigues, F. M. Mello, L. C. Branquinho, N. Zufelato, E. P. SilveiraLacerda, and A. F. Bakuzis, “Real-time infrared thermography detection of magnetic nanoparticle hyperthermia in a murine model under a nonuniform field configuration,” Int. J. Hypertherm. 29, 752 (2013). 6 A. Chan, R. P. Orme, R. A. Fricker, and P. Roach, “Remote and local control of stimuli responsive materials for therapeutic applications,” Adv. Drug Delivery Rev. 65, 497 (2013). 7 T. Hoare, J. Santamaria, G. F. Goya, S. Irusta, D. Lin, S. Lau, R. Padera, R. Langer, and D. S. Kohane, “A magnetically triggered composite membrane for on-demand drug delivery,” Nano Lett. 9, 3651 (2009). 8 M. K. Jaiswal, A. Pradhan, R. Banerjee, and D. Bahadur, “Dual pH and temperature stimuli-responsive magnetic nanohydrogels for thermochemotherapy,” J. Nanosci. Nanotechnol. 14, 4082 (2014). 9 P. Pradhan, J. Giri, F. Rieken, C. Koch, O. Mykhaylyk, M. Döblinger, R. Banerjee, D. Bahadur, and C. Plank, “Targeted temperature sensitive magnetic liposomes for thermo-chemotherapy,” J. Controlled Release 142, 108 (2010). 10 J. M. Yang, H. Yang, and L. W. Lin, “Quantum dot nano thermometers reveal heterogeneous local thermogenesis in living cells,” ACS Nano 5, 5067 (2011). 11 G. Kroemer, L. Galluzzi, and C. Brenner, “Mitochondrial membrane permeabilization in cell death,” Physiol. Rev. 87, 99 (2007). 12 W. Droge, “Free radicals in the physiological control of cell function,” Physiol. Rev. 82, 47 (2002). 13 T. D. Vreugdenburg, C. D. Willis, L. Mundy, and J. E. Hiller, “A systematic review of elastography, electrical impedance scanning, and digital infrared thermography for breast cancer screening and diagnosis,” Breast Cancer Res. Treat. 137, 665 (2013). 14 W. R. Viali, G. B. Alcantara, P. P. C. Sartoratto, M. A. G. Soler, E. Mosiniewicz-Szablewska, B. Andrzejewski, and P. C. Morais, “Investigation of the molecular surface coating on the stability of insulating magnetic oils,” J. Phys. Chem. C 114, 179 (2010). 15 L. A. T. Reino, T. M. Lima, A. S. Costa, A. F. Bakuzis, C. M. B. Santos, P. P. C. Sartoratto, and P. C. Morais, “Investigation of colloidal stability and
Rev. Sci. Instrum. 85, 094905 (2014) insulation characteristics of magnetic oils,” J. Nanosci. Nanotechnol. 12, 9319 (2012). 16 T.-H. Tsai, L.-S. Kuo, P.-H. Chen, D.-S. Lee, and C.-T. Yang, “Applications of ferro-nanofluid on a micro-transformer,” Sensors 10, 8161 (2010). 17 T.-H. Tsai, P.-H. Chen, D.-S. Lee, and C.-T. Yang, “Investigation of electrical and magnetic properties of ferro-nanofluid on transformers,” Nanoscale Res. Lett. 6, 264 (2011). 18 G. Kucsko, P. C. Maurer, N. Y. Yao, M. Kubo, H. J. Noh, P. K. Lo, H. Park, and M. D. Lukin, “Nanometre-scale thermometry in a living cell,” Nature (London) 500, 54 (2013). 19 F. Vetrone, R. Naccache, A. Zamarron, A. Juarranz de la Fuente, F. SanzRodriguez, L. Martinez Maestro, E. Martin Rodriguez, D. Jaque, J. Garcia Sole, and J. A. Capobianco, “Temperature sensing using fluorescent nanothermometers,” ACS Nano 4, 3254 (2010). 20 J. S. Donner, S. A. Thompson, M. P. Kreuzer, G. Baffou, and R. Quidant, “Mapping intracellular temperature using green fluorescent protein,” Nano Lett. 12, 2107 (2012). 21 K. Okabe, N. Inada, C. Gota, Y. Harada, T. Funatsu, and S. Uchiyama, “Intracellular temperature mapping with a fluorescent polymeric thermometer and fluorescence lifetime imaging microscopy,” Nat. Commun. 3, 705 (2012). 22 M. Essone Mezeme, S. Lasquellec, and C. Brosseau, “Subwavelength control of electromagnetic field confinement in self-similar chains of magnetoplasmonic core-shell nanostructures,” Phys. Rev. E 84, 026612 (2011). 23 M. Essone Mezeme and C. Brosseau, “Engineering nanostructures with enhanced thermoplasmonic properties for biosensing and selective targeting applications,” Phys. Rev. E 87, 012722 (2013). 24 J. Zhong, W. Liu, Z. Du, P. C. de Morais, Q. Xiang, and Q. Xie, “A noninvasive, remote and precise method for temperature and concentration estimation using magnetic nanoparticles,” Nanotechnology 23, 075703 (2012). 25 J. B. Weaver, A. M. Rauwerdink, and E. W. Hansen, “Magnetic nanoparticle temperature estimation,” Med. Phys. 36, 1822 (2009). 26 J. Zhong, W. Liu, L. Kong, and P. C. Morais, “A new approach for highly accurate, remote temperature probing using magnetic nanoparticles,” Sci. Rep. (in press). 27 F. Ludwig, D. Eberbeck, N. Lowa, U. Steinhoff, T. Wawrzik, M. Schilling, and L. Trahms, “Characterization of magnetic nanoparticle systems with respect to their magnetic particle imaging performance,” Biomed. Eng.-Biomed. Te. 58, 535 (2013). 28 T. Yoshida and K. Enpuku, “Simulation and quantitative clarification of AC susceptibility of magnetic fluid in nonlinear brownian relaxation region,” Jpn. J. Appl. Phys. 48, 127002 (2009). 29 J. Zhong, W. Liu, M. Zhou, and L. He, “AC magnetization measurement employing a pair of differentiating coils for magnetic nanoparticle noninvasive and remote temperature estimation,” Proceedings of International Instrumentation and Measurement Conference Technology, Graz, Austria, 13–16 May 2012 (IEEE, 2012), pp. 1780–1783.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 128.114.34.22 On: Sat, 22 Nov 2014 07:00:00
