A 10 Megahertz Spatial Light Modulator

Abstract

Rapid and programmable shaping of light fields is central to modern microscopy, display technologies, optical communications and sensing, quantum engineering, and quantum information processing. Current wavefront shaping technologies face a fundamental dichotomy: spatial light modulators (SLMs) offer high pixel count but suffer from low refresh rates, while acousto-optic deflectors (AODs) provide moderate speed with restricted optical beam geometries. Though recent advances in photonic integrated circuits achieve fast switching, there is currently no tool that provides MHz-rate, continuous motion, and arbitrarily reconfigurable control over a set of diffraction-limited spots. Here we introduce a new class of spatial light modulator that provides both 2D pixel geometry and high speed. The device operates by encoding spatial information in frequency bins via a broadband optical phase modulator, and decoding them via a first-of-its-kind, high-resolution 2D spectrometer. The spectrometer, based on the architecture which we call the Re-Imaging Phased Array (RIPA), achieves its sensitivity through long path-lengths, enabled by intra-spectrometer re-imaging lens-guides. We demonstrate site-resolved optical pulsing with a 44(1)~ns rise time, corresponding to frame rates exceeding 10 million frames per second, as well as arbitrary, reconfigurable 2D addressing and multi-site operations, including asynchronous, independent beam motion, splitting, and recombination. Leveraging these tools opens new horizons in rapid optical manipulation of matter across science, from fast, scalable control that approaches the inertial and radiation limits of atoms in quantum processors, to dynamically programmable, microsecond-resolved illumination in microscopy and neuro-biological imaging.

Figures (9)

• Figure 1: Operation of a 10 Mega-FPS spatial light modulator.a, We introduce the Re-Imaging Phased Array (RIPA), a device that enables a first-of-its-kind highly-dispersive, two-dimensional spectrometer for high-bandwidth spatial addressing using controlled frequency multiplexing. An input beam is converted into a 2D beam array by the first and second RIPAs (insets i and ii). Each RIPA duplicates and displaces the beam over each round trip, during which the beams acquire relative propagation phases $\varphi_y$($\varphi_x$), determined by the round-trip path lengths of the first (second) RIPA and the laser frequency. Beam diffraction over the long round trips is suppressed by refocusing optics, yielding a uniform transverse mode across the 2D phased array. Interfering these beams with a focusing lens performs frequency-to-position mapping. When the coherent input beam is phase-modulated by a high-bandwidth electro-optic modulator (EOM) to generate designed frequency tones $\nu_k$ (shades of red after EOM), the RIPA-spectrometer directs each tone to a specific location $(x_k,y_k)$. Increasing the frequency rasters the spot through the Brillouin zone, wrapping around at the edge. Programmable RF modulation therefore synthesizes arbitrary 2D images at the spectrometer output (inset iii). b, Detailed optical layout of the RIPA system. Diffraction-less propagation is accomplished using a running-wave pseudo-cavity geometry with: an intra-path microlens array (MLA) in the first RIPA, and an identical MLA + two 4f telescopes in the second RIPA (to add propagation distance without spatial inversion).
• Figure 1: Detailed schematic of the RIPA system. The experiment operates with a 780 nm source (red paths), while an auxiliary 785 nm laser (pink paths) provides active path-length stabilization. Input frequency tones are generated by phase-modulation using an electro-optic modulator (EOM) driven by an arbitrary waveform generator (AWG). A second-order optical filter isolates the $+1$ modulation sidebands and inject them into the first RIPA. The first RIPA converts the input into a 1D beam array using a microlens array (MLA). This array is transferred via a 4f relay which contains angled elevator mirrors and a Dove prism for spatial alignment (tilt, displacement, and roll) to the second RIPA. The second RIPA employs the same MLA and V-coated 4f telescopes to generate a 2D beam array. A pellicle beamsplitter out-couples the light to a spatial light modulator (SLM) for phase calibration (see Ext. Data Fig. \ref{['fig:fige2']}) and final imaging (with static phase mask). Active locking of the second RIPA path length is achieved via quadrant photodiodes and piezo-actuated mirrors. Pairs of mirrors mounted on translation stages allow for the independent tuning of round-trip path length and beam separation to match the MLA pitch.
• Figure 2: 1D phased-array frequency-to-position mapping. For a single frequency input, a 1D RIPA produces, at its output a, an array of parallel beams with pitch $p$ and a controllable phase step between adjacent beams $\varphi_y$. A lens with focal length $f$ interferes these beams to produce a focused intensity pattern whose envelope (dashed curve) is set by the single-beam size and whose peak position $y$ is set by $\varphi_y$. The insets illustrate the expected linear relationship between optical frequency detuning $\Delta\nu$ (modulo one free spectral range, FSR) and output position $y$, within the first Brillouin zone (unshaded) spanning $\pm \lambda f / 2p$. b, Measured focal-plane intensity distribution for a single-frequency input. The dashed vertical line indicates the line-cut shown to the right, where the red curve is the measured line-cut profile and the dashed black curve is the Gaussian envelope corresponding to a single beam in the phased array. c, Intensity profiles along the line-cut in b for increasing laser frequency, showing continuous peak translation across the image plane consistent with the designed frequency-to-position mapping.
• Figure 2: Interferometric calibration of SLM phase mask.a. Schematic of the phase calibration procedure. Localized blazed gratings are applied to specific sub-regions of the SLM to steer pairs of beams from the 2D beam array, causing them to interfere at the focal plane. This process compensates for static wavefront aberrations and ensures phase uniformity across the array. b. The resulting calibrated phase mask. Each grid tile consists of $100\times100$ SLM pixels (covering an $800\times 800~\mu\mathrm{m}^2$ area), providing independent phase control for each beam in the array. c. Representative interference fringes captured at the focal plane. d, One-dimensional intensity line-cut (dots) of the fringes in c with a sinusoidal fit (solid line), from which the relative phase is extracted for precise compensation.
• Figure 3: Spectrally programmed 2D image synthesis and static characterization.a, Principle of 2D frequency-to-position mapping: Sweeping the laser frequency $\nu$ (color scale) drives the addressing spot along a tilted raster trajectory across the Brillouin zone row-by-row. b, Injection of a single optical tone generates a (measured) intensity distribution corresponding to a diffraction-limited spot. c, Synthesis of an "S"-shaped pattern via simultaneous injection of multiple optical tones. d, Frequency-domain synthesis of the pattern in c. The pattern can be decomposed into 11 independent spots, each corresponding to a unique optical frequency $\nu_k$ that the RIPA+lens system maps to a distinct spatial location. This optical spectrum is generated by RF driving of a high-bandwidth electro-optic phase modulator, which in practice produces symmetric optical sidebands on the input light, and leaves a strong carrier - a custom, second-order flat-top optical filter Li2025Filter transmits only the +1 order blue sidebands, suppressing all other orders and the carrier. e, The crosstalk in this system comes directly from the discreteness of the phased array, resulting in power-law tails on each spot. Here we plot the measured (blue circles) and predicted (blue curve) crosstalk between two addressed sites as a function of their separation $d$, in units of the $1/e^2$ intensity radius $w_0^\prime$. Scaling to a $100\times 100$ array (red curve) predicts crosstalk levels below $10^{-4}$ at moderate separations. Inset: Crosstalk is calculated through azimuthal averaging (along the dashed circle) the integrated intensity within one beam waist radius (solid circle). f–h System-wide uniformity maps across the first Brillouin zone illustrating f, peak intensity and g,h, fitted beam waists $w_x$ ($w_y$) in the $x$- ($y$-) directions. The system exhibits high uniformity, with relative standard deviations ($\sigma$) of $2.6\%$ for peak intensity and $3.1\%$ ($1.9\%$) for the $w_x$ ($w_y$).