Nuclear magnetic resonance gyroscope (NMRG) is the smallest atomic sensor in navigation level. Spin precession can be detected by measuring the optical rotation of the plane of an off-resonant linearly polarized probe beam. The optimal frequency and power of the probe beam counts for the performance of NMRG, which has been verified by our former experiments. The NMRG system would have higher sensitivity and lower consumption comparing to the circularly polarized probe beam. In this paper, we demonstrate the optimal frequency and power of an off-resonant linearly polarized probe beam by theoretical analysis and experimental verification. In theory, the off-resonant linearly polarized probe beam can be decomposed into two circularly polarized components of opposite helicity. Its plane of polarization will be rotated by an angle, due to the positive and negative helicity light experience different induces of refraction as it propagates through a birefringent medium. The off-resonant linearly polarized probe beam becomes partially absorbed by the alkali vapor as it propagates through the NMRG cell. The overall signal is determined by both optical signal and beam absorption. After optimizing the frequency and power of the probe, the magnetic field sensitivity was 2pT/Hz1/2.
Nuclear magnetic resonance gyroscope (NMRG) is the smallest atomic sensor in navigation level. Spin precession can be detected by measuring the optical rotation of the plane of an off-resonant linearly polarized probe beam. The optimal frequency and power of the probe beam counts for the performance of NMRG, which has been verified by our former experiments. The NMRG system would have higher sensitivity and lower consumption comparing to the circularly polarized probe beam. In this paper, we demonstrate the optimal frequency and power of an off-resonant linearly polarized probe beam by theoretical analysis and experimental verification. In theory, the off-resonant linearly polarized probe beam can be decomposed into two circularly polarized components of opposite helicity. Its plane of polarization will be rotated by an angle, due to the positive and negative helicity light experience different induces of refraction as it propagates through a birefringent medium. The off-resonant linearly polarized probe beam becomes partially absorbed by the alkali vapor as it propagates through the NMRG cell. The overall signal is determined by both optical signal and beam absorption. After optimizing the frequency and power of the probe, the magnetic field sensitivity was 2pT/Hz1/2.
We report a rubidium atomic magnetometer based on pump-probe nonlinear magneto-optical rotation. The rubidium vapor cell is placed in a five-layer magnetic shield with inner coils that can generate uniform magnetic fields along the direction of pump beam, and the cell is also placed in the center of a Helmholtz coil that can generate an oscillating magnetic field perpendicular to the direction of pump beam. The atoms are optically pumped by circularly polarized pump beam along a constant magnetic field in a period of time, then the pump beam is turned off and a /2 pulse of oscillating magnetic field for 87Rb atoms is applied. After the above process, the individual atomic magnetic moments become phase coherent, resulting in a transverse magnetization vector precessing at the Larmor frequency in the magnetic field. The linearly polarized probing beam is perpendicular to the direction of magnetic field, and can be seen as a superposition of the left and right circularly polarized light. Because of the different absorptions and dispersions of the left and right circularly polarized light by rubidium atoms, the polarization direction of probing beam rotates when probing beam passes through rubidium vapor cell. The rotation of the polarization is subsequently converted into an electric signal through a polarizing beam splitter. Finally, the decay signal related to the transverse magnetization vector is measured. The Larmor frequency proportional to magnetic field is obtained by the Fourier transform of the decay signal. The value of magnetic field is calculated from the formula:B=(2/) f, where and f are the gyromagnetic ratio and Larmor frequency, respectively. In order to measure the magnetic field in a wide range, the tracking lock mode is proposed and tested. The atomic magnetometer can track the magnetic field jump of 1000 nT or 10000 nT, indicating that the atomic magnetometer has strong locking ability and can be easily locked after start-up. The main performances in different magnetic fields are tested. The results show that the measurement range of the atomic magnetometer is from 100 nT to 100000 nT, the extreme sensitivity is 0.2 pT/Hz1/2, and the magnetic field resolution is 0.1 pT. The transverse relaxation times of the transverse magnetization vector in different magnetic fields are obtained, and the relaxation time decreases with the increase of the magnetic field. When the measurement range is from 5000 nT to 100000 nT, the magnetic field sampling rate of the atomic magnetometer can be adjusted in a range from 1 Hz to 1000 Hz. The atomic magnetometer in high sampling rate can measure weak alternating magnetic field at low frequency. This paper provides an important reference for developing the atomic magnetometer with large measurement range, high sensitivity and high sampling rate.
We propose an average Rb polarization model to analyze the influence of temperature on the spin polarization of Rb and 129Xe in a Nuclear Magnetic Resonance Gyroscope (NMRG) with low pump power. This model is essentially based on summing the Rb spin polarization along the direction of the pump beam and dividing the result by the cell length. We experimentally study the spin polarization of Rb and 129Xe atoms as a function of the cell temperature at low values of the pump power. The experimental results and the values calculated with the average Rb polarization model are in good agreement for both Rb and 129Xe. The spin polarization of Rb atoms decreases with increasing cell temperature, with a decreasing trend which is rapid at temperatures below 110 °C, and slower at temperatures above 110 °C. The experimental values of the 129Xe polarization, obtained with a pump power of 1 mW, first increase to a maximum P 129Xe−ave = 0.66 % at 118 °C, and then decreases as the temperature increases. Increasing the power of the pump beam shifts the temperature maximum to a higher value. Our model is suitable for the analysis of Rb and 129Xe polarization at high temperature and low pump power, i.e. when the power of the pump beam is completely absorbed within a few millimeters of the front window of the cell. Therefore, the present model can provide theoretical support for the improvement of the Signal-to-Noise-Ratio (SNR) of the NMRG, and to determine its optimal working temperature.
In order to measure a weak alternating magnetic field, an optically-pumped Rb magnetometer based on pump-probe structure is investigated and demonstrated. The pumping light and probing light propagate along the z axis and x axis, respectively. A constant polarization magnetic field along the pumping light is applied, which not only stabilizes the polarization of Rb atoms but also tunes resonance frequency of Rb atoms. When a weak alternating magnetic field is applied perpendicularly to the constant magnetic field, the magnetic moment will tip off the z axis and rotate around the z axis. And then the polarization plane of probing light is modulated correspondingly. The x component of the magnetic moment can be obtained with a balanced detector. As a result, a signal proportional to weak alternating magnetic field can be obtained.In order to obtain the magnetic response of the magnetometer, we analyze the signal amplitude as a function of polarization magnetic field strength B0 and transverse relaxation time 2 with numerical simulation. The amplitude-frequency response of the magnetometer is determined mainly by two parameters, namely cutoff frequency c=1/2 and resonance frequency 0= B0, where is the gyromagnetic ratio of Rb atom. At low frequency, that is a0 and a 0c2, the magnetometer has a flat response, here a is the frequency of the weak alternating magnetic field. If 0c, the signal amplitude will be large for large 0 or small c. For a given c, the peak response appears at 0=c. In the vicinity of resonance frequency, if c0, a peak will appear and if c 0, no peak occurs. At high frequency, the amplitude will decrease with the increase of a.We verify the above analyses in experiment. A vapor cell with a short transverse relaxation time is used to obtain large frequency response bandwidth. Through optimizing the powers and frequencies of pumping laser and probing laser, high polarization and detection sensitivity of atomic spin can be obtained. Moreover, through choosing an appropriate polarization magnetic field, the magnetometer can be maximally sensitive to the magnetic field to be measured. The experimental results show that the magnetometer has a sensitivity of about m 0.2; pT/HzHz and bandwidth of about 3.5 kHz. It can be used to detect low field magnetic resonance and high frequency abnormal physical phenomena.
Optical rotation detection systems are widely used in various kinds of atomic sensors, such as atomic magnetometers, co-magnetometers, and nuclear magnetic resonance (NMR) gyroscopes; and appear in many applications such as fundamental physics, biomagnetism, magnetism test, etc. The study analyses the frequency responses to three-axis dynamic magnetic fields in optical rotation detection systems (ORDS) of Spin-Exchange Relaxation-Free (SERF) magnetometers. Because of the imperfections of the optical elements, the probe light is actually elliptically polarised in ORDS, which can polarise the atom ensemble and cause the responses to the three-axis magnetic field. Theoretically it analyse the frequency responses to the sinusoidal magnetic field, and prove that the responses are closely associated with the DC magnetic field. The values of the DC magnetic fields are calculated with special frequency points, called ‘break points’, in the transverse responses. The relationships between the DC magnetic field and the sensitivities of ORDS, and effectively suppress the magnetic noise responses with the residual magnetic field compensation. The paper demonstrated the influence of elliptically polarised light on the sensitivity of ORDS in SERF magnetometers. The elliptically polarised probe light polarises the alkali atom ensemble, which leads to the responses of the magnetic noise in the shield. Finally, the sensitivity of ORDS is improved by approximately two times at 10-20 Hz. This method is beneficial for improving probe sensitivity in SERF magnetometers, co-magnetometers, and NMR gyroscopes.
We describe a dual-axis atomic spin comagnetometer for rotation sensing using a single laser beam. A circularly polarized laser is used to pump the co-located atomic ensembles of alkali metal and noble gas and probe their gyroscopic precession at the same time, which is different from the traditional two-beam orthogonal pump-probe configuration. Combined with magnetic field modulations perpendicular to the laser propagating direction, the comagnetometer could measure dual-axis rotation rates simultaneously. This system is demonstrated theoretically and experimentally. A rotation sensitivity of 5.1 × 10–7 rad s−1 Hz−1∕2 and a bias instability of 0.016 °/h have been achieved. The theory and method presented here shed light on the way to miniaturize the comagnetometer with high precision for practical applications.
Nonlinear resonances in alkali metal vapor, which are detected by the magneto-optical rotation of the linear polarization of light, are actively used in quantum magnetometry to fabricate atomic magnetometers. The magneto-optical rotation in most such sensors is due to magnetic birefringence, and rotation angles usually do not exceed tens of milliradians. In this work, an experiment where magneto-optical resonances of linear polarization rotation of a probe wave are due to strong dichroism induced in a medium by a counterpropagating pump wave has been proposed. Both waves are in resonance with the Fg = 2 → Fe = 1 optical transition in the 87Rb D1 line (λ ≈ 795 nm). Experiments have been carried out with a 2-cm-long cylindrical cell filled with a buffer gas, and the maximum rotation angle is ≈390 mrad (22°) at a width of resonance of about 300 nT. The results show that the configuration proposed for the observation of magneto-optical rotation is promising for the fabrication of compact high-sensitivity atomic magnetometers.