Multi-channel measurements with fine spatial resolution will make magnetoencephalograms (MEGs) possible with small animals using optically pumped magnetometers (OPMs). Therefore, we fabricated a 20-channel probe-beam detector that uses a K-Rb hybrid OPM to increase the spatial resolution. First, we investigated the sensitivity of the detector using the multi-channel measurements and demonstrated that the detector had a fine sensitivity (10–20 fT/Hz1/2 at 10 Hz). Subsequently, we measured magnetic field distribution generated from a loop coil and compared those measurements with analytically calculated distributions. The measurements were in good agreement with the theoretical predictions. The experimental results indicate that our newly developed multi-channel OPM detector has sufficient performance specifications for MEG measurements.
© 2018 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
Biomagnetic field measurement techniques, such as magnetoencephalograms (MEGs) , are widely used for elucidating biological functions. Highly sensitive sensors are required for measuring the very weak magnetic signals involved in biological systems (10−6–10−9 of terrestrial magnetism). At present, superconductive quantum interference devices (SQUIDs) are usually employed for bio-magnetic measurements. However, bio-magnetic measurement systems that use SQUIDs are not widely used in basic research and clinical studies because they require cryogenic cooling and their installation and maintenance costs are very high.
Recently, optically pumped magnetometers (OPMs) [2–4] have attracted much attention for these applications. They have an extremely high sensitivity (theoretically, 0.01 fT/Hz1/2) under the conditions in which the spin-exchange relaxation-free (SERF) regime is possible [3,4]. OPMs require no cryogenic cooling.
MEGs with small animals will be part of the next phase of research on human brain functions. MEGs for small animals have been measured with SQUIDs . Jensen et al. recently reported a MEG system with OPMs, which allowed single-channel measurements of MEG signals generated from the nerve impulses of frogs . For MEGs with small animals, multi-channel measurements with fine spatial resolution are required because the brains of these animals are much smaller than those of humans. For constructing multi-channel OPMs, researchers generally arrange multiple modules with a set of pump and probe beams [7,8]. However, this technique is limited because each sensor cell may have different characteristics. To address this limitation, we propose to use a large hybrid cell of potassium (K) and rubidium (Rb) atoms (5 cm3) with multiple pumps and probe beams. In a previous study, we demonstrated that this hybrid cell could achieve high spatial uniformity of the spin poralization [9,10].
To realize high-sensitivity multi-channel measurements, we must downsize and integrate the probe beam and detectors while maintaining their sensitivity. In this study, we fabricated a 20-channel OPM probe beam detector with a fine spatial resolution to increase the sensor density. To carry out multi-channel measurements, we need as many amplifiers as measuring points. However, commercially available amplifiers introduce problems of increasing size and cost. Therefore, we also fabricated a 20-channel amplifier. By measuring this OPM’s sensitivity and the magnetic field distribution, we demonstrate that our newly developed multi-channel OPM detector has performance specifications that allow it to be used for MEG measurements.
The experimental setup of the OPM we studied is shown in Fig. 1. This setup, with the exception of the probe beam detector, is similar to the setup we reported previously . The hybrid vapor cell is a cube (side, 5 cm), and it encloses K and Rb atoms in He and N2 buffer gases at a ratio of 10:1 and a total pressure of 150 kPa at room temperature. For the measurements, the vapor cell’s temperature was kept at approximately 180 °C and the densities of Rb and K atoms were 1.0×1018 m−3 and 1.6×1019 m−3, respectively. We used a vertical cavity surface emitting laser (VCSEL) for sensitivity measurement to increase the output signal, a titanium (Ti)-sapphire laser for magnetic field distribution measurements to obtain the distribution along the x-axis with a fine spatial resolution for the pump beam, and a diode laser for the probe beam. The pump beam density was determined to maximize the output signal strength.
We need to consider the cross-talk between neighboring channels when designing the multi-channel detector. The spatial resolution was determined based on diffusion constants of Rb and K atoms. The cross-talk free distance lD is given as follows :
Photographs of the fabricated probe-beam detector and the arrangement of the photo diodes are shown in Figs. 2 and 3, respectively. The outer dimensions of the detector are 7×7×7 cm3. The detector consists of a polarizing beam splitter (PBS) and two printed circuit boards, on each of which 20 photodiodes are printed (S10356-01, Hamamatsu Photonics, Sizuoka, Japan). Each photodiode has a receiving area of 3 mm × 3 mm. Each column consists of 10 channels at intervals of 1 mm along the pump beam so that the sensor density is 2.5 cm−1. The distance between the 2 columns is 27 mm for the sake of differential measurements. In this study, we used only the upper columns (10 channels). A photograph of the fabricated amplifier and a schematic of the amplifier circuit are shown in Figs. 4 and 5, respectively. The outer dimensions of the amplifier are 17 cm×12 cm×7 cm. We selected the parts of the amplifier circuit, such as low-noise operational amplifiers (TL074, Texas Instruments, Texas, America) and twisted pair cables, to reduce the electrical noise. The amplifier was enclosed in a metal shield to eliminate the influence of external magnetic fields.
The circularly polarized pump beam tuned to the D1 resonance wavelength polarizes the spins of the Rb atoms in the vapor cell. This spin polarization is transferred to the K atoms by spin-exchange collisions  that rotate the polarization around an external magnetic field perpendicular to the pump-probe plane. The polarization plane of the linearly polarized probe beam, now slightly detuned from the D1 resonance wavelength, is rotated optically by the spin polarization of K atoms. The photodiode arrays detect such optical rotations.
First, to assess whether the fabricated detector has a sufficiently fine sensitivity for MEGs, we measured the sensitivity of the sensor using many channels. Subsequently, we measured the magnetic field distribution generated from a loop coil to evaluate the performance of the detector.
3. Results and discussion
The noise spectrum densities and the magnetic field sensitivities of the 10-channel OPM probe beam detector measured at the same time are shown in Figs. 6 and 7. The pump and probe beam power densities were 452.1 mW/cm2 and 2.0 mW/cm2 and were 50 mm in diameter and 50 mm×2 mm. The pump beam power density was much higher than that for a single alkali-metal cell. This is because the spin polarization of probed atoms arises via spin exchange collisions between pumped and probed atoms . The sensitivities of the OPM were 10–20 fT/Hz1/2 at 10 Hz, which were obtained by averaging the noise from 9 to 11 Hz. The half widths at half maximum of the magnetic resonance lines were 6.3–6.4 Hz. In our previous study, the system captured MEG signals from human subjects with a sensitivity of 21 fT/Hz1/2 . Therefore, these results indicated that the multi-channel detector’s sensitivity is sufficiently fine for the MEG signals. Three factors improve sensitivity when using multi-channel measurements. First, the polarimeter detectors in this system yield high signal intensities. Second, using K atoms with their small spin destruction relaxation rate for the probe atoms enhances the coherence of the sensor volume. Finally, we introduced noise suppression techniques to reduce the electrical noise in the amplifier circuit.
Then, we compare the contributions of different noise sources to the output of Ch 1 [Fig. 8]. The optical and electrical noises were measured by blocking the pump and probe beam and by blocking the probe beam, respectively. Those noises were converted by the scale factor from the voltage to the magnetic field of the magnetometer to evaluate the performance of our amplifier. Magnetic noise, optical noise and electrical noise accounted for 18.4, 10.1, and 4.0 fT/Hz1/2 at 10 Hz, respectively. To compare clearly, all of the noise were converted to the magnetic signals with the same reference signal obtained when the pump beam and probe beam were not blocked. The electrical noise was smaller than the others, indicating that the amplifier had sufficient performance for our magnetometer. Furthermore, these results indicate that the sensitivity may be further improved by using differential measurements and improving the stability of the probe beam.
For testing magnetic field distribution measurements, we used a loop coil (diameter, 10 mm) as a signal source. It was installed 60 mm away from the sensing volume along the y-axis. A test signal was generated from the loop coil by applying a sinusoidal current of 0.4 mA. The pump and probe beam power densities were 495.4 mW/cm2 and 2.8 mW/cm2 with diameters of 3 mm and 50 mm×2 mm, respectively. The experimental procedure is shown in Fig. 9. Measuring the one-dimensional magnetic field distribution and shifting the loop coil location at intervals of 4.5 mm along the x-axis, we obtained the two-dimensional magnetic field distribution. Therefore, the magnetic fields along the z-axis were measured simultaneously and those along the x-axis were measured successively. To evaluate the experimental results quantitatively, we introduced the following g value :Figs. 10(a) and (b). The dotted circles in Fig. 10(b) are measuring locations. The measured results are in good agreement with the theoretical results. The g value was 98.8%. We can, therefore, confirm valid distributions along the z-axis, which is shown in Fig. 9(b) as simultaneous measurements. On the other hand, distortions appear along the x-axis. We consider that these are caused by shifts in the location of the loop coil and changes to the magnetic field within the magnetic shielding when we changed the location of the loop coil. However, such a high value of g indicates that we can carry out valid multi-channel measurements with the fabricated detector.
In this study, we fabricated a 20-channel OPM probe beam detector with a fine spatial resolution to make progress toward measuring MEGs from small animals. We achieved noise levels below 20 fT/Hz1/2 and demonstrated the validity of multi-channel measurements with the detector. The detector shows sufficient performance for use with MEGs. We believe that the sensitivity can be further improved by taking differential measurements. In future, we plan to implement pump-beam modulation to increase the number of measurement points along the probe beam . We also plan to actually measure MEGs on small animals with the detector.
Ministry of Education, Culture, Sports, Science and Technology (15K06106, 15H01813, 16K13114); Nakatani Foundation for Advancement of Measuring Technologies in Biomedical Engineering.
References and links
1. M. Hamalainen, R. Hari, R. J. Ilmoniemi, J. Knuutila, and O. V. Lounasmaa, “Magnetoencephalography theory, instrumentation, and applications to noninvasive studies of the working human brain,” Rev. Mod. Phys. 65, 413–497 (1993). [CrossRef]
4. D. Budker and M. V. Romalis, “Optical magnetometry,” Nat. Phys. 3, 227–234 (2007). [CrossRef]
5. G. B. Christianson, M. Chait, A. Cheveigné, J. F. Linden, and J. Neurophysiol, “Auditory evoked fields measured noninvasively with small-animal MEG reveal rapid repetition suppression in the guinea pig,” J. Neurophysiol. 112, 3053 (2014). [CrossRef] [PubMed]
6. K. Jensen, R. Budvytyte, R. A. Thomas, T. Wang, A. M. Fuchs, M. V. Balabas, G. Vasilakis, L. D. Mosgaard, H. C. Stærkind, J. H. Müller, T. Heimburg, S.-P. Olesen, and E. S. Polzik, “Non-invasive detection of animal nerve impulses with an atomic magnetometer operating near quantum limited sensitivity,” Sci. Rep. 6, 29638 (2016). [CrossRef] [PubMed]
7. G. Bison, N. Castagna, A. Hofer, P. Knowles, J.-L. Schenker, M. Kasprzak, H. Saudan, and A. Weis, “A room temperature 19-channel magnetic field mapping device for cardiac signals,” Appl. Phys. Lett. 95, 173701 (2009). [CrossRef]
9. Y. Ito, H. Ohnishi, K. Kamada, and T. Kobayashi, “Sensitivity improvement of spin-exchange relaxation free atomic magnetometers by hybrid optical pumping of potassium and rubidium,” IEEE Trans. Magn. 47, 3550–3553 (2011). [CrossRef]
10. Y. Ito, D. Sato, K. Kamada, and T. Kobayashi, “Measurements of magnetic field distributions with an optically pumped K-Rb hybrid atomic magnetometer,” IEEE Trans. Magn. 50, 4006903 (2014). [CrossRef]
11. Y. Ito, D. Sato, K. Kamada, and T. Kobayashi, “Optimal densities of alkali metal atoms in an optically pumped K-Rb hybrid atomic magnetometer considering the spatial distribution of spin polarization,” Opt. Express 24, 15391–15402 (2016). [CrossRef] [PubMed]
12. K. Kim, S. Begus, H. Xia, SK. Lee, V. Jazbinsek, Z. Trontelj, and M. V. Romalis, “Multi-channel atomic magnetometer for magnetoencephalography: A configuration study,” Neuroimage 89, 143–151 (2014). [CrossRef]
13. H. G. Dehmelt, “Spin resonance of free electrons polarized by exchange collisions,” Phys. Rev. 109, 381–385 (1958). [CrossRef]
14. K. Kamada, D. Sato, Y. Ito, H. Natsukawa, K. Okano, N. Mizutani, and T. Kobayashi, “Human magnetoencephalogram measurements using newly developed compact module of high-sensitivity atomic magnetometer,” Jpn. J. Appl. Phys. 54, 026601 (2015). [CrossRef]
15. Y. Mamishin, Y. Ito, and T. Kobayash, “A novel method to accomplish simultaneous multilocation magnetic field measurements based on pump beam modulation of an atomic magnetometer,” IEEE Trans. Magn. 53, 4001606 (2017). [CrossRef]