Search Results (18 total)

We report on measurements of the spin lifetime of nuclear spins strongly coupled to a micromechanical cantilever as used in magnetic resonance force microscopy. We find that the rotating-frame correlation time of the statistical nuclear polarization is set by the magnetomechanical noise originating from the thermal motion of the cantilever. Evidence is based on the effect of three parameters: (1) the magnetic field gradient (the coupling strength), (2) the Rabi frequency of the spins (the transition energy), and (3) the temperature of the low-frequency mechanical modes. Experimental results are compared to relaxation rates calculated from the spectral density of the magnetomechanical noise.

When probing nuclear spins in materials on the nanometer scale, random fluctuations of the spin polarization will exceed the mean Boltzmann polarization for sample volumes below about (100  nm)³. In this Letter, we use magnetic resonance force microscopy to observe nuclear spin fluctuations in real time. We show how reproducible measurements of the polarization variance can be obtained by controlling the spin correlation time and rapidly sampling a large number of independent spin configurations. This allows significant improvement in the signal-to-noise ratio for nanometer-scale magnetic resonance imaging.

We cool the fundamental mechanical mode of an ultrasoft silicon cantilever from a base temperature of 2.2 K down to 2.9±0.3  mK using active optomechanical feedback. The lowest observed mode temperature is consistent with limits determined by the properties of the cantilever and by the measurement noise. For high feedback gain, the driven cantilever motion is found to suppress or “squash” the optical interferometer intensity noise below the shot noise level.

We have detected and manipulated the naturally occurring sqrt[N] statistical polarization in nuclear spin ensembles using magnetic resonance force microscopy. Using protocols previously developed for detecting single electron spins, we have measured signals from ensembles of nuclear spins in a volume of roughly (150 nm)³ with a sensitivity of roughly 2000 net spins in a 2.5 h averaging window. Three systems have been studied, ¹⁹F nuclei in CaF₂, and ¹H nuclei (protons) in both polymethylmethacrylate and collagen, a naturally occurring protein. By detecting the statistical polarization, we not only can work with relatively small ensembles, but we eliminate any need to wait a longitudinal relaxation time T₁ to polarize the spins. We have also made use of the fact that the statistical polarization, which can be considered a form of spin noise, has a finite correlation time. A method similar to one previously proposed by Carlson [Bull. Am. Phys. Soc. 44, 541 (1999)] has been used to suppress the effect of the statistical uncertainty and extract meaningful information from time-averaged measurements. By implementing this method, we have successfully made nutation and transverse spin relaxation time measurements in CaF₂ at low temperatures.

We have used the large gradients generated near the ferromagnetic tip of a magnetic resonance force microscope to locally suppress spin diffusion in a silica sample containing paramagnetic electron spins. By controlling the slice location with respect to the tip, the magnetic field gradient was varied from 0.01 to 36  mT/μm, resulting in a fourfold decrease in T₁^-1 and a similar decrease in T_1ρ^-1. The observed dependence of the relaxation rates on field gradient is consistent with the quenching of flip-flop interactions that mediate the transport of magnetization between slow and fast relaxing spins.

We report the detection of the sqrt[N] statistical polarization in a small ensemble of electron spin centers in SiO₂ by magnetic resonance force microscopy. A novel detection technique was employed that captures the statistical polarization and cycles it between states that are either locked or antilocked to the effective field in the rotating frame. Using field gradients as high as 5   G/nm, we achieved a detection sensitivity equivalent to roughly two electron spins, and observed ultralong spin-lock lifetimes, as long as 20 s. Given a sufficient signal-to-noise ratio, this scheme should be extendable to single electron spin detection.

We consider the process of spin relaxation in the oscillating cantilever-driven adiabatic reversals technique in magnetic-resonance force microscopy. We simulated the spin relaxation caused by thermal excitations of the high-frequency cantilever modes in the region of the Rabi frequency of the spin subsystem. The minimum relaxation time obtained in our simulations is greater than but of the same order of magnitude as one measured in recent experiments. We demonstrated that using a cantilever with nonuniform cross-sectional area may significantly increase spin-relaxation time.

Magnetic resonance force microscopy was used to study the behavior of small ensembles of unpaired electron spins in silica near a micrometer-size ferromagnetic tip. Using a cantilever-driven spin manipulation protocol and a magnetic field gradient greater than 10⁵  T/m, signals from as few as 100 net spins within a 20 nm thick resonant slice could be studied. A sixfold increase in the spin-lattice relaxation rate was found within 800 nm of the ferromagnet, while no effect due to silica surface proximity was detected. The results are interpreted in terms of Larmor-frequency magnetic field fluctuations emanating from the ferromagnet.

Noncontact friction between a Au(111) surface and an ultrasensitive gold-coated cantilever was measured as a function of tip-sample spacing, temperature, and bias voltage using observations of cantilever damping and Brownian motion. The importance of the inhomogeneous contact potential is discussed and comparison is made to measurements over dielectric surfaces. Using the fluctuation-dissipation theorem, the force fluctuations are interpreted in terms of near-surface fluctuating electric fields interacting with static surface charge.

Cantilever magnetometry with moment resolution better than 10⁴μ_B was used to study individual nanomagnets. By using the fluctuation-dissipation theorem to interpret measurements of field-induced cantilever damping, the low frequency spectral density of magnetic fluctuations could be determined with resolution better than 1μ_B Hz^-1/2. Cobalt nanowires exhibited significant magnetic dissipation and the associated magnetic fluctuations were found to have 1/f frequency dependence. In individual submicron rare-earth alloy magnets, the dissipation/fluctuation was very small and not distinguishable from that of a bare silicon cantilever.

Electron-spin resonance of E^′ centers in vitreous silica was studied using force-detection techniques at temperatures down to 5 K. Cyclic adiabatic inversion of electron spins was performed by frequency modulation of the applied microwave magnetic field. This produced an oscillatory magnetic force between the electron spins and a nearby permanent magnet, resulting in the vibration of a cantilever on which the sample was mounted. By pulsing the microwave field prior to the force-detection sequence, nutation of the spins could be observed. Spin echoes were observed mechanically using a modified echo-pulse sequence. The decay of magnetization during cyclic adiabatic inversion was also studied and is discussed in terms of utility for future single-spin detection experiments.

Recent initial experiments in magnetic resonance force microscopy (MRFM) have detected the magnetic force exerted by electrons and nuclei in microscopic samples. The experiments generate a force signal by modulating the sample magnetization with standard magnetic resonance techniques. Sample sizes of a few nanograms generate readily detected force signals of order 10^-14 to 10^-16 Newtons. This article describes the present status of MRFM technology, with particular attention to the feasibility of detecting single-electron magnetic moments, and the possible applications of MRFM in biological imaging.

A Comment on the Letter by J. I. Pascual et al. Phys. Rev. Lett. 71, 1852 (1993).

Magnetic resonance usually is detected inductively, using resonant circuits. Recent experiments have detected magnetic resonance mechanically, using the magnetic force acting between a sample and a nearby force microscope cantilever. This article compares the sensitivity of inductive and mechanical methods for detecting magnetic resonance. We show that, as mechanical oscillators are made smaller, their ability to detect magnetic resonance signals improves, such that existing force microscope cantilevers provide a viable alternative to inductive methods for detecting magnetic resonance.

A mechanical degenerate parametric amplifier has been devised which greatly increases the motional response of a microcantilever for small harmonic force excitations. The amplifier can improve force detection sensitivity for measurements dominated by sensor noise or backaction effects and can also produce mechanical squeezed states. In an initial squeezing demonstration, the thermal noise (Brownian motion) of the cantilever was reduced in one phase by 4.9 dB.

We have demonstrated that a gold scanning-tunneling-microscope tip can be used as a miniature solid-state emission source for directly depositing nanometer-size gold structures. The emission mechanism is believed to be field evaporation of tip atoms, which is enhanced by the close proximity of the substrate. The technique has been used in air on Ag(111) surfaces to write several thousand features with no apparent degradation of the tip’s ability to emit atoms.

One of the oldest unresolved problems in physics is the mechanism of charge exchange between contacting surfaces when at least one of them is insulating. We describe a new technique, using force microscopy, for studying this problem with greater lateral resolution than has been previously possible. The force microscope is shown to have 0.2 μm lateral resolution and the sensitivity to detect 3 electronic charges. In contact-charging experiments between the microscope tip and polymethyl methacrylate, the charged region was much larger than the expected contact area and bipolar charge exchange was observed.

This paper presents results of a new and highly accurate technique for measuring low-energy phonon dispersion in liquid ⁴He. The technique is based on the behavior of ultrasonic second-harmonic generation in a lossless, dispersive medium. By using frequencies in the low gigahertz range and measuring second-harmonic intensity as a function of propagation distance, the coherence length for the harmonic generation can be determined. The coherence length is, in turn, related to the phonon dispersion curve in a simple way. The results are interpreted in terms of the series expansion ε(k)=c₀ℏk(1+α₁k+α₂k²+α₃k³+⋯), where ε and k are phonon energy and wave number, respectively. By using measurements taken at two different fundamental frequencies, we find |α₁|<10^-3 Å at saturated vapor pressure (SVP) and 6.3 bars, and α₂=(1.56±0.06) Ų at SVP. If α₁ is assumed to be zero, the α₂ can be determined from a measurement at a single frequency, and we find α₂=(1.55±0.01) Ų at SVP. At higher pressures, α₂ decreases. Since the excitation spectrum is probed with such low-momentum phonons (k<0.011 Å^-1), the analysis is insensitive to assumed values of α₄, α₅, etc., and is only slightly sensitive to the assumed value of α₃.