Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
ALL-OPTICAL SPATIAL LIGHT MODULATORS
Document Type and Number:
WIPO Patent Application WO/2021/011056
Kind Code:
A1
Abstract:
A spatial light modulator (SLM) comprised of a 2D array of optically-controlled semiconductor nanocavities can have a fast modulation rate, small pixel pitch, low pixel tuning energy, and millions of pixels. Incoherent pump light from a control projector tunes each PhC cavity via the free-carrier dispersion effect, thereby modulating the coherent probe field emitted from the cavity array. The use of high-Q/V semiconductor cavities enables energy-efficient all-optical control and eliminates the need for individual tuning elements, which degrade the performance and limit the size of the optical surface. Using this technique, an SLM with 106 pixels, micron-order pixel pitch, and GHz-order refresh rates could be realized with less than 1 W of pump power.

Inventors:
PANUSKI CHRISTOPHER (US)
ENGLUND DIRK (US)
Application Number:
PCT/US2020/032444
Publication Date:
January 21, 2021
Filing Date:
May 12, 2020
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
MASSACHUSETTS INST TECHNOLOGY (US)
PANUSKI CHRISTOPHER LOUIS (US)
ENGLUND DIRK ROBERT (US)
International Classes:
B82Y20/00; G02F1/01; H01L33/00; H01L33/02
Foreign References:
US20170307958A12017-10-26
Other References:
SHCHERBAKOV, MAXIM R., LIU SHENG, ZUBYUK VARVARA V., VASKIN ALEKSANDR, VABISHCHEVICH POLINA P., KEELER GORDON, PERTSCH THOMAS, DOL: "Ultrafast all-optical tuning of direct-gap semiconductor metasurfaces", NATURE COMMUNICATIONS, vol. 8, 12 May 2012 (2012-05-12), pages 17, XP055786726, Retrieved from the Internet [retrieved on 20200721], DOI: 10.1038/s41467-017-00019-3
MCKENDRY, JONATHAN AND JONATHAN DAVID MASSOUBRE; SHUAILONG ZHANG; BRUCE R RAE; RICHARD P GREEN; ERDAN GU; ROBERT K HENDERSON; KE: "Visible-Light Communications Using a CMOS-Controlled Micro-Light-Emitting-Diode Array", JOURNAL OF LIGHTWAVE TECHNOLOGY, vol. 30, no. 1, 2012, pages 61 - 67, XP011397320, Retrieved from the Internet [retrieved on 20200722], DOI: 10.1109/JLT.2011.2175090
CALUSINE, GREG; POLITI ALBERTO; AWSCHALOM DAVID D: "Silicon Carbide Photonic Crystal Cavities with Integrated Color Centers", APPLIED PHYSICS LETTERS,, vol. 105, no. 1, May 2014 (2014-05-01), XP012187903, Retrieved from the Internet [retrieved on 20200722], DOI: 10.1063/1.4890083
MINKOV, MOMCHIL; SAVONA VINCENZO; GERACE DARIO: "Photonic crystal slab cavity simultaneously optimized for ultra-high Q/V and vertical radiation coupling", APPLIED PHYSICS LETTERS, XP012222334, Retrieved from the Internet [retrieved on 20200722], DOI: 10.1063/1.4991416
DEMS, MACIEJ AND CHUNG IL-SUG, NYAKAS PÉTER, BISCHOFF SVEND, PANAJOTOV KRASSIMIR: "Numerical Methods for modeling Photonic-Crystal VCSELs", OPTICS EXPRESS, 14 July 2010 (2010-07-14), pages 3901 - 3903, XP055786737, Retrieved from the Internet [retrieved on 20200722]
ZHANG, YONG AND DANPING LI; CHENG ZENG; YU SHI; ZENGZHI HUANG; JINZHONG YU; JINSONG XIA: "Ultralow Power Nonlinear Response in an Si Photonic Crystal Nanocavity", vol. 5, no. 4, 25 July 2013 (2013-07-25), XP011522242, Retrieved from the Internet [retrieved on 20200722], DOI: 10.1109/JPHOT.2013.2273734
Attorney, Agent or Firm:
SUD, Dhruv et al. (US)
Download PDF:
Claims:
CLAIMS

1. A spatial light modulator comprising:

a layer patterned with a two-dimensional array of semiconductor cavities, the two- dimensional array of semiconductor cavities scattering signal light at a resonant wavelength; and at least one incoherent light source, in optical communication with the two-dimensional array of semiconductor cavities, to tune the resonant wavelength of at least one semiconductor cavity in the two-dimensional array of semiconductor cavities via optical free carrier injection.

2. The spatial light modulator of claim 1, wherein the semiconductor layer comprises silicon.

3. The spatial light modulator of claim 1, wherein each semiconductor cavity in the two- dimensional array of semiconductor cavities has a quality factor Q of at least 1000.

4. The spatial light modulator of claim 1, wherein the two-dimensional array of semiconductor cavities is comprised of HI photonic crystal cavities.

5. The spatial light modulator of claim 1, wherein the two-dimensional array of semiconductor cavities is comprised of L4/3 photonic crystal cavities.

6. The spatial light modulator of claim 1, wherein the two-dimensional array of semiconductor cavities is comprised of micropillar cavities.

7. The spatial light modulator of claim 1, wherein the two-dimensional array of semiconductor cavities has a pitch of about half the resonant wavelength to about the resonant wavelength.

8. The spatial light modulator of claim 1, wherein the at least one incoherent light source comprises a two-dimensional array of light-emitting diodes (LEDs).

9. The spatial light modulator of claim 8, further comprising: a control layer, operably coupled to the two-dimensional array of LEDs, to modulate LEDs in the two-dimensional array of LEDs at a rate of at least 1 GHz; and

a waveguide layer, in optical communication with the two-dimensional array of LEDs and the two-dimensional array of semiconductor cavities, to image optical free carriers emitted by the two-dimensional array of LEDs onto the two-dimensional array of semiconductor cavities.

10. The spatial light modulator of claim 9, further wherein the optical free carriers emitted by the two-dimensional array of LEDs are at a wavelength of less than 500 nm and the signal light is at a wavelength of more than 500 nm.

11. The spatial light modulator of claim 1, further comprising:

a signal light waveguide layer, in optical communication with the semiconductor layer, to guide the signal light to the at least one semiconductor cavity.

12. The spatial light modulator of claim 11, further comprising:

a signal light source, in optical communication with the signal waveguide layer, to launch the signal light into the signal waveguide layer.

13. A spatial light modulator comprising:

a resonant surface to reflect and/or transmit incident light at a resonant wavelength; and at least one light source, in optical communication with the resonant surface, to locally tune the resonant wavelength of the resonant surface via optical free carrier injection.

14. The spatial light modulator of claim 13, wherein the resonant surface comprises a surface defined by a photonic crystal slab guided mode resonator.

15. The spatial light modulator of claim 13, wherein the resonant surface includes a wavelength-scale patterned semiconductor slab.

16. The spatial light modulator of claim 15, wherein the resonant surface supports a spatially distributed resonant mode.

17. The spatial light modulator of claim 13, wherein the at least one light source comprises an incoherent light source.

18. The spatial light modulator of claim 17, wherein the incoherent light source includes a two- dimensional array of light-emitting diodes (LEDs).

19. The spatial light modulator of claim 18, further comprising:

a control layer, operably coupled to the two-dimensional array of LEDs, to modulate LEDs in the two-dimensional array of LEDs at a rate of at least 1 GHz; and

a waveguide layer, in optical communication with the two-dimensional array of LEDs and the two-dimensional array of semiconductor cavities, to image optical free carriers emitted by the two-dimensional array of LEDs onto the two-dimensional array of semiconductor cavities.

20. The spatial light modulator of claim 19, further wherein the optical free carriers emitted by the two-dimensional array of LEDs are at a wavelength of less than 500 nm and the signal light is at a wavelength of more than 500 nm.

21. The spatial light modulator of claim 13, wherein the at least one light source includes a laser light source configured to emit ultrashort pulses.

22. The spatial light modulator of claim 13, wherein a free carrier lifetime of the resonant surface is about 100 ps or less.

23. The spatial light modulator of claim 13, wherein a free carrier lifetime of the resonant surface is about 1 ns or more.

24. A spatial light modulator comprising:

a photonic crystal (PhC) cavity array to reflect and/or transmit incident light at a resonant wavelength, the PhC cavity having a ratio of quality factor Q to mode volume V of at least about 100; a two-dimensional array of light-emitting diodes (LEDs), in optical communication with the PhC cavity array, to locally tune the resonant wavelength of the PhC cavity array via optical free carrier injection; and

a two-dimensional array of complementary metal-oxide semiconductor (CMOS) controllers, operably coupled to the two-dimensional array of LEDs, to modulate LEDs in the two- dimensional array of LEDs at a rate of at least 10 MHz.

Description:
All-Optical Spatial Light Modulators

CROSS-REFERENCE TO RELATED APPLICATION(S)

[0001] This application claims the priority benefit, under 35 U.S.C. 119(e), of U.S. Application No. 62/873,232, filed on July 12, 2019, which is incorporated herein by reference in its entirety.

BACKGROUND

[0002] The generation and manipulation of structured light via spatial light modulators (SLMs) has become central to modern science and technology. Commercially available SLMs are ubiquitous, with common applications ranging from projectors to imagers and even manufacturing tools. In the laboratory, optical microscopy, metrology, imaging, and manipulation experiments leverage the dynamic reconfigurability of SLMs to control optical fields. For example, recent SLM-enabled demonstrations have included terahertz compressive imaging, deep-tissue imaging, trapping of single atoms in arbitrary lattices, and re-configurable integrated photonic switches.

[0003] However, despite their importance, modem commercial SLMs are limited by their operating principles. These devices can be broadly classified into two categories: liquid crystal on silicon (LCOS) SLMs and digital micromirror devices (DMDs). The frame rate of LCOS-based SLMs, which typically provide phase modulation by reorienting birefringent liquid crystals with an applied voltage, is limited to ~ 100 Hz by the LC natural response time. These LC response times are even longer at longer wavelengths, such as those used for infrared (IR) telecommunications, due to the need for thicker liquid crystal layers. DMDs offer on-off switching based on the electrically-controlled displacement of a MEMS mirror, and therefore enable modulation rates on the order of 10-100 kHz. However, their binary modulation significantly impairs the achievable diffraction efficiency (the ratio of energy in the first- and zeroth-order diffraction patterns) to roughly 30% of that of a typical LCOS SLM. MEMS deformable-mirrors (MEMS-DM) offer an alternative approach with excellent efficiency; however, the typical pixel pitch of a MEMS-DM is much greater than that of the controlled light. In fact, both LCOS and DMD techniques generally feature micron-order pixel pitches, which limits their use in applications such as IR beam steering where subwavelength (L/2) values are desired. [0004] Research devices have attempted to address these deficiencies. Due to its commercial feasibility and maturity, silicon photonics has attracted significant interest, yielding phase modulated arrays of 8x8, and more recently 32x32 vertical grating couplers. The demonstrated power efficiency and modulation rates, however, are both limited by the use of thermo-optic phase shifters, which have microsecond-order response times and require approximately 10 mW of power to generate a p phase shift. While the incorporation of a free-carrier based phase shifter within a micro-ring or micro-disc resonator could simultaneously reduce the switching energy to roughly femtojoules and increase the modulation rate to roughly gigahertz, the additional element would further increase the pixel pitch, which is already limited by the large (several square micron) sizes of vertical grating couplers. Given these limitations, the performance and scale of photonic integrated circuit (PIC) SLMs has yet to exceed that of other commercial solutions.

SUMMARY

[0005] A spatial light modulator (SLM) includes a layer patterned with a two-dimensional array of semiconductor cavities, the two-dimensional array of semiconductor cavities scattering signal light at a resonant wavelength. The SLM also includes at least one incoherent light source, in optical communication with the two-dimensional array of semiconductor cavities, to tune the resonant wavelength of at least one semiconductor cavity in the two-dimensional array of semiconductor cavities via optical free carrier injection. In some cases, the at least one incoherent light source comprises a two-dimensional array of light-emitting diodes (LEDs). In some cases, the SLM further includes a control layer, operably coupled to the two-dimensional array of LEDs, to modulate LEDs in the two-dimensional array of LEDs at a rate of at least 10 MHz, at least 1 GHz, and/or the like. In some cases, the SLM also includes a signal waveguide layer, in optical communication with the two-dimensional array of LEDs and the two-dimensional array of semiconductor cavities, to image optical free carriers emitted by the two-dimensional array of LEDs onto the two-dimensional array of semiconductor cavities.

[0006] A SLM includes a resonant surface to reflect and/or transmit incident light at a resonant wavelength, and at least one light source, in optical communication with the resonant surface, to locally tune the resonant wavelength of the resonant surface via optical free carrier injection. In some cases, the at least one incoherent light source comprises a two-dimensional array of light- emitting diodes (LEDs). In some cases, the SLM further includes a control layer, operably coupled to the two-dimensional array of LEDs, to modulate LEDs in the two-dimensional array of LEDs at a rate of at least 1 GHz. In some cases, the SLM also includes a signal waveguide layer in optical communication with the two-dimensional array of LEDs and the two-dimensional array of semiconductor cavities, to image optical free carriers emitted by the two-dimensional array of LEDs onto the two-dimensional array of semiconductor cavities.

[0007] A SLM includes a photonic crystal (PhC) cavity array to reflect and/or transmit incident light at a resonant wavelength, the PhC cavity having a ratio of quality factor Q to mode volume V of at least about 100. The SLM also includes a two-dimensional array of light-emitting diodes (LEDs), in optical communication with the PhC cavity array, to locally tune the resonant wavelength of the PhC cavity array via optical free carrier injection. The SLM further includes a two-dimensional array of complementary metal-oxide semiconductor (CMOS) controllers, operably coupled to the two-dimensional array of LEDs, to modulate LEDs in the two-dimensional array of LEDs at a rate of at least 10 MHz.

[0008] In some cases, a distributed resonator - such as a slab photonic crystal guided mode resonator - can be used instead of the two-dimensional array. Similar to the two-dimensional array, the resonant frequency of the guided mode resonator may be locally tuned via free carrier injection from an incoherent light source. The guided mode resonator can also have about unity reflection efficiency, which provides high diffraction efficiency due to the distributed resonant mode, and avoids the need to couple into individual resonators.

[0009] The incoherent light source may be a two-dimensional array of light-emitting diodes (LEDs). In this case, the SLM may include a control layer operably coupled to the two-dimensional array of LEDs and a two-dimensional signal waveguide layer in optical communication with the two-dimensional array of LEDs and the two-dimensional array of semiconductor cavities. The control layer modulates LEDs in the two-dimensional array of LEDs at a rate of at least 10 MHz, at least 1 GHz, and/or the like. The waveguide layer images optical free carriers emitted by the two-dimensional array of LEDs onto the two-dimensional array of semiconductor cavities. These optical free carriers emitted by the two-dimensional array of LEDs can be at a wavelength of less than 500 nm and the signal light can be at a wavelength of more than 500 nm.

[0010] The SLM may also include a signal light waveguide layer, in optical communication with the semiconductor layer, to guide the signal light to the semiconductor cavity. A signal light source, in optical communication with the signal waveguide layer, can launch the signal light into the signal waveguide layer.

[0011] All combinations of the foregoing concepts and additional concepts discussed in greater detail below (provided such concepts are not mutually inconsistent) are part of the inventive subject matter disclosed herein. In particular, all combinations of claimed subject matter appearing at the end of this disclosure are part of the inventive subject matter disclosed herein. The terminology used herein that also may appear in any disclosure incorporated by reference should be accorded a meaning most consistent with the particular concepts disclosed herein.

BRIEF DESCRIPTIONS OF THE DRAWINGS

[0012] The skilled artisan will understand that the drawings primarily are for illustrative purposes and are not intended to limit the scope of the inventive subject matter described herein. The drawings are not necessarily to scale; in some instances, various aspects of the inventive subject matter disclosed herein may be shown exaggerated or enlarged in the drawings to facilitate an understanding of different features. In the drawings, like reference characters generally refer to like features (e.g., functionally similar and/or structurally similar elements).

[0013] FIG. 1A shows a transmission-mode all-optical spatial light modulator (SLM).

[0014] FIG. IB shows a reflection-mode all-optical SLM.

[0015] FIG. 1C shows a reflection-mode all-optical spatial light modulator (SLM) using a guided mode resonator to shape reflected light.

[0016] FIG. 2A shows a transmission-mode all-optical SLM with micropillar cavities.

[0017] FIG. 2B shows one of the micropillar cavities of FIG. 2A in greater detail.

[0018] FIG. 3A shows an HI PhC cavity array for an all-optical SLM.

[0019] FIG. 3B shows an L4/3 PhC cavity array for an all-optical SLM.

[0020] FIG. 3C shows a heterostructure cavity with a resonant mode at the gamma point that is vertically confined by symmetry protection.

[0021] FIG. 4 shows a thermal model for the nonlinear resonator cavity array.

[0022] FIG. 5A shows a transient numerical solution of Eqns. 14-16 for the default parameters listed in Table 1 assuming a square wave pump input (dashed line), for carrier density.

[0023] FIG. 5B shows a transient numerical solution of Eqns. 14-16 for the default parameters listed in Table 1 assuming a square wave pump input (dashed line), for membrane temperature. [0024] FIG. 5C shows a transient numerical solution of Eqns. 14-16 for the default parameters listed in Table 1 assuming a square wave pump input (dashed line), for resonance shift.

[0025] FIG. 6A shows a periodic steady state solution to Eqns. 14-16 for the default parameters of Table 1, for carrier density.

[0026] FIG. 6B shows a periodic steady state solution to Eqns. 14-16 for the default parameters of Table 1, for membrane temperature.

[0027] FIG. 6C shows a periodic steady state solution to Eqns. 14-16 for the default parameters of Table 1, for resonance shift.

[0028] FIG. 7A shows the periodic steady state solution to Eqns. 14-16 when t ox is reduced to 0.5 mnh to minimize membrane heating, for carrier density.

[0029] FIG. 7B shows the periodic steady state solution to Eqns. 14-16 when t ox is reduced to 0.5 mnh to minimize membrane heating, for membrane temperature.

[0030] FIG. 7C shows the periodic steady state solution to Eqns. 14-16 when t ox is reduced to 0.5 mth to minimize membrane heating, for resonance shift.

[0031] FIG. 8A shows the periodic steady state solution to Eqns. 14-16 for a shorter 10 ns pump period T, showing the possibility of fast switching when T ~ t h , for carrier density.

[0032] FIG. 8B shows the periodic steady state solution to Eqns. 14-16 for a shorter 10 ns pump period T, showing the possibility of fast switching when T ~ t h , for membrane temperature.

[0033] FIG. 8C shows the periodic steady state solution to Eqns. 14-16 for a shorter 10 ns pump period T, showing the possibility of fast switching when T ~ t h , for resonance shift.

[0034] FIG. 9A shows the periodic steady state solution to Eqns. 14-16 for T = t h , which limits switching contrast, for carrier density.

[0035] FIG. 9B shows the periodic steady state solution to Eqns. 14-16 for T = t h , which limits switching contrast, for membrane temperature.

[0036] FIG. 9C shows the periodic steady state solution to Eqns. 14-16 for T = t h , which limits switching contrast, for resonance shift.

[0037] FIG. 10A shows the periodic steady state solution to Eqns. 14-16 for a smaller unit cell membrane (A m = 7a X 7a), which results in substantial temperature deviations but maintains switching contrast, for carrier density. [0038] FIG. 10B shows the periodic steady state solution to Eqns. 14-16 for a smaller unit cell membrane (A m = 7a X 7a), which results in substantial temperature deviations but maintains switching contrast, for membrane temperature.

[0039] FIG. IOC shows the periodic steady state solution to Eqns. 14-16 for a smaller unit cell membrane ( A m — 7a X 7a), which results in substantial temperature deviations but maintains switching contrast, for resonance shift.

[0040] FIG. 11A shows the periodic steady state solution to Eqns. 14-16, with an enhanced switching contrast due to an enhanced carrier density when is reduced to 100 ps, for carrier

density.

[0041] FIG. 1 IB shows the periodic steady state solution to Eqns. 14-16, with an enhanced switching contrast due to an enhanced carrier density when is reduced to 100 ps, for membrane

temperature.

[0042] FIG. l lC shows the periodic steady state solution to Eqns. 14-16, with an enhanced switching contrast due to an enhanced carrier density when is reduced to 100 ps, for resonance

shift.

[0043] FIG. 12A shows bandgap width (left plot), center position (center plot), and design deviation (right plot) (Eqn. 17) for a 1.55 mm center wavelength for a lattice design for t sla = 220 nm-thick silicon slab with a surrounding oxide cladding.

[0044] FIG. 12B shows a model of an example design lattice with r = 0.262a and a = 410 nm.

[0045] FIG. 12C shows the band structure of even and odd guided modes for r = 0.379a and a = 505 nm.

[0046] FIG. 12D shows a model of another example design lattice with r = 0.379a and a = 505 nm.

[0047] FIG. 12E shows the band structure of even and odd guided modes for r = 0.379a and a = 505 nm.

[0048] FIG. 13 A shows bandgap width (left plot), center position (center plot), and design deviation (right plot) (Eqn. 17) for a 1.55 mm center wavelength for a lattice design for t sla = 220 nm-thick silicon slab with a surrounding air cladding.

[0049] FIG. 13B shows a model of another example design lattice with r = 0.267 a and a = 445 nm. [0050] FIG. 13C shows the band structure of even and odd guided modes for r = 0.267a and a = 445 nm.

[0051] FIG. 14 shows the mode profile of an optimized oxide-clad silicon HI PhC cavity produced via numerical optimization of quality factor in MEEP.

[0052] FIG. 15 shows the mode profile of an optimized oxide-clad silicon L4/3 PhC cavity produced via numerical optimization of quality factor in MEEP.

DETAILED DESCRIPTION

[0053] The spatial light modulators (SLMs) as described here address many of the deficiencies of conventional SLMs. Such an SLM may include a layer (e.g., a silicon layer) patterned with a two- dimensional array of semiconductor cavities and at least one incoherent light source, in optical communication with the two-dimensional array of semiconductor cavities. In operation, the two- dimensional array of semiconductor cavities scatters signal light at a resonant wavelength. And the incoherent light source tunes the resonant wavelength of at least one semiconductor cavity in the two-dimensional array of semiconductor cavities via optical free carrier injection.

[0054] Each semiconductor cavity in the two-dimensional array of semiconductor cavities can have a quality factor Q of at least 1000. The two-dimensional array of semiconductor cavities can be comprised of HI photonic crystal cavities, L3 photonic crystal cavities, L4/3 photonic crystal cavities, or micropillar cavities. It may have a pitch equal to about half the resonant wavelength to about the resonant wavelength.

[0055] All-optical control can overcome the fundamental scaling limit associated with individual control elements, which can be described generally as follows. The number of pixels that can be placed within an array of area A scales directly with A, but the number of control elements that fit through the perimeter scales as \[A. However, with all optical control, electronic controls don’t need to be routed through a single layer to each pixel. When the electronic control employs CMOS control of the light source, 3D control electronics can be employed while keeping a single optical layer. Optical control can also allow for the emitters to be spread out more, as they can be imaged onto the semiconductor layer with variable magnification. These all-optical SLM designs and techniques are different than optically-addressed spatial light modulators, which can feature optically-written control masks but are still fundamentally limited by the slow response of the liquid crystal control medium.

[0056] A fast, efficient, low-pitch, and large scale SLM can directly impact a variety of existing fields and unveil entirely new avenues of research. Real-time, sub-diffraction-limited optical microscopy, ultra-dense optical interconnects, high-rate LiDAR, optical neural networks, topological quantum optics, and large-scale control of atomic ensembles for quantum science are a few envisioned applications that illustrate the diversity of foreseeable uses.

[0057] Example Spatial Light Modulators

[0058] FIG. 1A shows an example transmission-mode all-optical spatial light modulator (SLM) 105. The SLM 105 includes a cavity array 110 of semiconductor cavities 115, a signal light waveguide/waveguide array 120, an incoherent light projector array 125, and a CMOS control array 130.

[0059] The cavity array 110 can generally be structured as a 2D semiconductor cavity array of high-index cavities 115 - with high experimental quality factor ( Q ) to mode volume (V ef f ) ratios - creates a narrow-pitch lattice of localized modes when resonant signal light illuminates the surface of the cavity array 110. The quality factor ( Q ) can be about 1000, about 10,000, about 100,000, about 500,000, about 1,000,000, including all values and sub-ranges in between. The mode volume (L eff ) can be from about 0.1(l/h) 3 to about 5(l/h) 3 , including all values and sub-ranges in between. The ratio Q/V eff can be from about 100 to about 10 7 , including all values and sub-ranges in between. FIG. 1A shows the array 110 as a photonic crystal cavity (PhC) array, but other vertically-coupled resonator types (e.g., micropillar, Fabry-Perot, and/or the like) can also be employed.

[0060] With an electro-refractive substrate material such as silicon, the amplitude and phase of each resonator’s highly-confined mode can be modified via the plasma dispersion effect by injecting free carriers into the cavities 115. Since the strength of such free carrier nonlinear effects scales with Q/V eff , fJ-order switching energies are possible for each mode. Free carriers can be injected optically at a wavelength below the modulation wavelength such that individual control elements on the optical surface of the cavity 110 - which can both limit the spacing between and quality of optical resonators - are eliminated. The pitch concerns associated with traditional PIC components (thermo-optic phase shifters and grating couplers) can also be mitigated by introducing nanoscale integrated gratings into the PhC hole lattice, which simultaneously affords high-Q and efficient vertical coupling of each resonator mode out of the cavity 110. The viability of producing foundry-based PhC cavities has been established via photolithographic patterning. The 2D resonator array 110 can leverage such recent manufacturing advances to harness the enhanced light-matter interaction characteristic of PhC cavities, yielding an ultra-dense, energy- efficient modulator array.

[0061] The 2D signal light waveguide 120, also sometimes referred to simply as a waveguide, is optional, and can be employed, particularly in the fully integrated (transmission-mode) format of FIG. 1 A, to efficiently distribute coherent light signal from a single signal laser to the PhC cavities 115, although any suitable distribution optics may be employed. For example, given a large enough array, an incident free space signal beam could be reflected from (depending on the coupling conditions) the PhC surface, as illustrated and described for FIG. IB. When the array 120 is used, both the coherent signal beam as well as the incoherent control light can pass through the array 120

[0062] In this modality, the SLM 105 can serve as a high-performance, multimode modulating retro-reflector. In some cases, the external signal light can be removed altogether by integrating active layers into the photonic crystal surface, thus forming an array of nanocavity light-emitting diodes (LEDs). The waveguide 120 can include, for example, a two-dimensional silicon waveguide array or an appropriately designed oxide slab. This slab or waveguide array can be any suitable design that equally distributes signal light power to each of the pixels.

[0063] The incoherent light projector array 125 can be employed to generate short- wavelength, incoherent, control/pump light to optically tune each resonator via the free carrier dispersion effect. Short wavelengths (e g., less than about 500 nm) can be useful to achieve sufficient absorption in the thin PhC cavity array 110, which could, for example, be fabricated in the 220 nm silicon layer offered in standard SOI CMOS processes. In some cases (not shown), this control light could be simply imaged onto the surface of the array 110 using a high-resolution external display. For example, the high-resolution display light can be collected by a lens and imaged with variable magnification onto the cavity array 110. Alternatively, the high-resolution display light can be columnated by a collection lens and imaged through an infinity-corrected objective onto the cavity array 110. FIG. 1A illustrates a more integrated solution that includes the projector array 125 as, for example, a 2D array of incoherent emitters 128 such as short wavelength emitters. For example, the emitters 128 can be LEDs, vertical cavity surface emitting lasers (VCSELs), and/or the like. As one example, gallium nitride (GaN) LEDs, with ~450 nm emission wavelengths, may be employed. Individual GaN LEDs with GHz modulation bandwidths have been demonstrated, along with arrays with integrated micro-lenses and pitches of 10 /mm or less.

[0064] The CMOS control array 130 is optional, and is employed for electronic control of individual emitters of the light projector array 125. The control array 130 can be configured to drive the emitters of the projector array 125 at a modulation rate of about 1 Hz, about 1 MHz, about 10 MHz, about 100 MHz, about 500 MHz, about 1 GHz, including all values and sub-ranges in between. Each pixel of the control array 130 can include, for example, a CMOS transistor to switch the connected emitter. The control array 130 pixels can also include local memory and/or digital-to-analog converters (DACs) to enable greyscale modulation of the incoherent emitters using locally stored patterns.

[0065] During use, in the illustrated transmission mode in FIG. 1A, a coherent signal light 118 is coupled into the array 110 of the high-quality factor ( Q ) (e.g., a Q of at least about 1000), low mode volume (V e ff) semiconductor cavities 115 using the 2D signal waveguide 120, which allows the transmitted electromagnetic field to be shaped. The incoherent pump light projector 125 - illustrated in FIG. 1 A as a gallium nitride LED array emitting light at 450 nm - is imaged onto the cavity array 110 to control each resonant mode via the free carrier dispersion effect. Photoexcited carriers from the absorption of short- wavelength pump light enable fast (e.g., GHz range), low- energy (0-order) switching of signal modes due to the high Q/V e ff ratio (e.g., a ratio of at least about 100) of the semiconductor cavities. Silicon photonic crystal cavities, which enable record high Q/V e ff ratios using foundry fabrication processes, are shown in FIG. 1A as an example implementation of this cavity array 120. However, other semiconductor cavities - including micro posts and vertical Fabry-Perot resonators - could also be switched by the control projector array 125. FIG. 1A also illustrates a plane 140 where, for example, a detector or other sensing device may be placed to receive the output light signal from the SLM 105.

[0066] The SLM 105 can efficiently generate incoherent modulated light from coherent modulated light based on several beneficial features. For example, the SLM 105 design permits fast modulation rates that can enable modem interconnects and novel applications. The micron-order pixel pitches can increase space-bandwidth product, increase fill factor, and enable 180° beam steering. The 0-order pixel tuning energy enables low-power control of large cavity arrays, and mature fabrication techniques can allow for scaling to several million elements/cavities. [0067] FIG. IB illustrates an alternative, reflection mode design of an SLM 205, where signal light can simply be reflected from the cavity array, and no signal light waveguide is required. This reflection-mode, all-optical SLM 205 is illustrated with a cavity array 210 (e.g., structurally and/or functionally similar to the array 110) of semiconductor cavities 215 (e.g., structurally and/or functionally similar to the cavities 115), an incoherent light projector array 225 (e.g., structurally and/or functionally similar to the light projector array 125), and an optional CMOS control array 230 (e.g., structurally and/or functionally similar to the control array 130). Here, the signal light waveguide 125 of FIG. 1A is omitted, and signal light 235 from a signal light source 240 instead reflects off the surface of the cavity array 210 and can be captured, such as by a sensing device at a plane 245. FIG. IB also illustrates separation of the specular and cavity reflections based on the off-axis incidence of the signal light 235. In some cases normal excitation (signal beam shone vertically down onto cavity array 210) can be employed such as, for example, using a polarizing beam splitter instead used to separate the direct reflection and cavity reflection, since these two reflections have orthogonal polarizations if the incident polarization angle is 45 degrees off-axis with the dominant cavity polarization.

[0068] FIG. 1C shows another reflection mode SLM 305. This reflection-mode, all-optical spatial light modulator 305 includes a resonant surface 310, an incoherent light projector array 325 (e.g., structurally and/or functionally similar to the light projector array 125), and an optional CMOS control array 330 (e.g., structurally and/or functionally similar to the control array 130). Generally, instead of a 2D semiconductor cavity array, a resonant surface can be used to reflect incident light with about near unity reflection efficiency. Free carriers can be optically injected in arbitrary patterns to provide the local resonance shifts desired to shape the amplitude or phase of the reflected light. The resonant surface can be any suitable sub-wavelength layer having a 2D layout of oscillators formed or deposited on it.

[0069] Here, the 2D microcavity array 210 of FIG. IB is replaced with a resonant surface 310. The resonant surface 310 can be structured as a wavelength-scale patterned semiconductor slab that supports a distributed resonant mode such, for example, guided mode resonance. The resonant surface can include, for example, a square or hexagonal lattice of holes in a semiconductor slab which is designed to support vertically-coupled guided mode resonances at the desired frequency.

[0070] The resonant surface 310 reflects signal light 335 from a light source 340 to generate an output signal that can be captured, such as by a sensing device at a plane 345. Generally, PhC cavities can provide high-Q diffraction-limited confinement of optical modes, but can require precise coupling and fabrication. Guided mode resonators (GMRs)on the other hand, such as the surface 310, while providing modest quality factors, can be formed with a defect-free PhC lattice, and be intrinsically vertically coupled. Coupled bilayer photonic crystal slabs can be used to improve the achievable GMR quality factor. The GMR can be locally tuned optically by injecting free carriers with arbitrary patterns. This enables the amplitude and phase of the reflected light to be tuned as desired at spatial resolution approaching l/2. The confined GMR modes exhibit large overlap with the injected free carriers and produce Fano reflection profiles due to interference between the direct reflection of incident light and vertical resonator leakage. Combined, these characteristics allow for efficient modulation of the reflection amplitude and phase with low power optical free carrier injection. While illustrated here for a reflection mode SLM design, a resonant surface can also be employed in transmission mode SLM designs, such as that illustrated in FIG. 1 A.

[0071] FIG. 2A shows a transmission-mode all-optical spatial light modulator 405 that includes a cavity array 410 (e.g., structurally and/or functionally similar to the array 110), a signal light waveguide 420 (e.g., structurally and/or functionally similar to the waveguide 120), an incoherent light projector array 425 (e.g., structurally and/or functionally similar to the light projector array 125), and an optional control array 430 (e.g., structurally and/or functionally similar to the control array 130). Here, the cavity array includes micropillar cavities 415 instead of the photonic crystal cavities 115 of FIG. 1A. The micropillar cavities 415 of FIG. 2A can also be used in a reflection- mode all-optical SLM, e.g., in the SLM 205, by eliminating the signal light waveguide 220 in FIG. IB and replacing the photonic crystal cavity array with these micropillar cavities. FIG. 2A illustrates the micropillar cavities 415 and incoherent light projectors of the projector array 425 as being arranged on the same lattice, with one micropillar cavity per incoherent light projector (e.g., per LED).

[0072] FIG. 2B shows a vertical micropillar cavity 415 suitable for use in the all-optical SLM 405 of FIG. 2A. The micropillar cavity 415 includes a central, low-bandgap cavity region 415a where the peak intensity region 415b of the optical mode is located and incoherent pump light is absorbed. Distributed Bragg reflectors (DBRs) 415cl, 415c2 comprising quarter- wavelength layers 415d of alternating high/low index dielectrics (e.g., TiO 2 /SiO 2 , InGaAsP/InP, or Si/SiO 2 ) are formed on top (the DBR 415cl) and bottom (the DBR 415c2) of the central cavity region 415b. The DBRs 415cl, 415c2 can confine light within the micropillar cavity 415 and set the desired coupling condition (under, over, or critically coupled) of signal light. These insulating DBRs 415cl, 415c2 also limit the diffusion of photogenerated charge carriers, increasing the overlap between free carriers and the optical mode. Towards the center of the cavity 415, the DBR layer thickness is adiabatically reduced to gently confine the mode, thus enhancing the achievable quality factor. The micropillar configuration of the cavity 415 can enhance vertical coupling at the cost of reduced quality factor. The cavity’s quality factor can be from about 1000 to about 10,000. Half wavelength spacing (as also illustrated in FIG. 2B) between each micropillar cavity is possible for signal wavelengths from the visible to the infrared (IR) (e.g., 1300 nm to 1700 nm).

[0073] The optically tuned array of nanophotonic resonators, shown in FIGS. 1A and IB as photonic crystal cavities and FIG. 2A as micropillar cavities, and tunable resonant surface (FIG. 1C) simultaneously achieve the desirable SLM performance metrics using the following components:

[0074] Generally referring to FIGS. 1A-1C, 2A-2B, the combination of these components - the cavity array (whether the optically tuned PhC arrays of FIGS. 1A and IB, the tunable resonant surface of FIG. 1C, or the micropillar cavities FIGS. 2A, 2B), the light projector array, the signal light waveguide (in some cases), and the optional projector control array - can enable fast, energy- efficient, all-optical conversion of incoherent light into a dense array of coherent, modulated signal modes. The SLM designs and signal generation approaches described can leverage the fabrication maturity of silicon photonics but circumvent the traditional limitations of PIC components by incorporating optically-controlled, vertically-coupled photonic crystal cavities.

[0075] Example Analysis of Performance of an All-Optical SLM

[0076] The device performance of any of the SLMs described here can be approximated generally as follows. Assume that the absorbed pump light (e.g., from a light projector array) excites /V abs free carrier pairs within a cavity’s mode volume V eff. According to perturbation theory, the free carrier dispersion effect then (nearly instantaneously) shifts the cavity resonance frequency w 0 by

(see below for derivation)

where Qi = is the loaded quality-factor of the resonator (given a loaded decay rate and g

is a material dependent“scattering volume" that serves as the constant of proportionality between carrier density and the fractional index change . In silicon, g can be approximated in two

ways: using the high-frequency limit of the Drude model or by linearizing the empirical formula

where n is the free electron (hole) density, and p, q, r, and s, are wavelength

dependent coefficients. Eqn. (2) follows from absorption measurements in silicon for wavelengths between 1.3 and 14 mm. For a signal wavelength of 1.55 pm, these two methods yield g *

respectively. The latter value is employed in subsequent calculations

to yield a conservative approximation. The maximum allowable Qi for each SLM pixel is dictated by the desired modulation frequency. For modulation frequencies on the order of 10 GHz, Qi is limited to ~ 10 4 . Finally, a diffraction limited mode volume on the order of is

chosen. Given that PhC cavities with ultrahigh-Q values on the order of a million are routinely fabricated (even in foundry-based CMOS processes) and sub-diffraction limited mode volumes of have been demonstrated, the selected parameters represent a readily fabricable

PhC cavity. Combining the selected metrics, it is found that

free carrier pairs shift the frequency of the resonant cavity one linewidth . In terms

of pump energy, a one million pixel array operated at a 1 GHz modulation frequency

would then use

0.6 W (4) of absorbed light assuming a 450 nm pump wavelength . If one accounts for imperfect

absorption of the pump light in a 220 nm-thick silicon membrane (see Table 1 for parameter details), the total input pump power is

(5)

[0077] This is an upper bound on the pump power incident on the PhC cavity array, as it is assumed that every pixel is modulated on-off at 1 GHz. In practice, not all pixels would be tuned in every modulation period. With these metrics in mind, targeted performance parameters are an array size of at least 1 megapixel (MP), a pixel pitch of 1.55 mm, a refresh rate of 10 9 Hz, and a control power of less than 1 W. EXAMPLE 1

[0078] High-Q, Vertically-Coupled PhC Cavities

[0079] An all-optical SLM uses PhC cavities with two conflicting characteristics: high-Q and efficient vertical coupling. For example, a relatively higher Q cavity traps light for a longer time than one with a lower Q. However, good vertical coupling can mean that light can get in and out easily. An integrated grating as described here can optimize Q while simultaneously shaping the far field profile such that most of the light loss is collectable. High quality -factor PhC cavities are typically designed with“gentle confinement" to reduce out-of-plane leakage, resulting in a weak free-space coupling. However, several techniques have been proposed and demonstrated to overcome this limitation. Cavity-specific designs have shown that near-field shaping can produce Gaussian-like far-field radiation patterns which yield collection efficiencies greater than 80%. Alternatively,“band folding” with an integrated grating serves as a general, fabrication-tolerant approach to improving the collection efficiency of various PhC cavities. In this scheme, a periodic lattice perturbation - typically an increase in hole radius - at twice the lattice period a“folds" Fourier components located near the edge of the Brillouin zone down to the G ( 0) point, thereby enhancing vertical emission. Experimental implementations and theoretical analyses for L3, L5, and L7 cavities have demonstrated the possibility of achieving Q - 10 5 with 50% vertical coupling efficiency.

[0080] FIGS. 3A-3C show cavity arrays with integrated gratings, where shading differentiates several modifications to the hole lattice. FIG. 3 A shows the first array, which is composed of HI cavities, each formed by removing a single hole 310, optimized by shifting three hexagonal layers of holes, and vertically-coupled with the addition of an integrated grating.

[0081] FIG. 3B shows how the cavity array can be formed with the“L4/3" cavity, where four holes 320 (bold circles in FIG. 3B) are added in place of the three holes 330 removed for a standard L3 cavity. This design is more amenable to photolithographic fabrication than an L3 cavity (optical proximity effects are minimized by reducing location variation in hole density), and affords low volume, high-Q, resonant modes. The resonant wavelength can also be readily tuned by varying the radius of the four internal holes 320.

[0082] The cavity designs illustrated in FIGS. 3A and 3B have cavity pitches between one and two wavelengths for a 1.5 /mm signal wavelength, a control wavelength of less than about 500 nm, and a -400 nm lattice constant. While this spacing can reduce or minimize coupling between cavities, other designs could attain subwavelength pitches with the use of a cavity-specific vertical coupling technique. For example, micro-post cavities with integrated dielectric reflectors would enable subwavelength pitches.

[0083] FIG. 3C shows a heterostructure cavity with a resonant mode at the gamma point (vertical wavevector) that is vertically confined by symmetry protection (this is a so called“bounded in the continuum state”). Lateral confinement is provided by tapering the radii of the holes (r1, r2, and r3 in FIG. 3C) in a hexagonal pattern. Similar to the micropillar design in FIG. 2A, this cavity enhances vertical coupling at the expense of reduced quality factor (e.g., a quality factor on the order of 10,000). Since vertical coupling can be limited by symmetry (the optical mode and vertically propagating plane waves have opposite symmetry), the index contrast between the slab core and cladding can be reduced, allowing more material combinations to be used (one example is silicon nitride in a silicon dioxide cladding, which is often offered in foundry processes).

[0084] All-Optical Tuning

[0085] Each PhC cavity in the all-optical SLM is tuned via optical carrier injection. Here, the physical effects of optical carrier injection-based tuning of a photonic crystal cavity array are further explored, and a model to compare the effects of competing free carrier and thermal nonlinearities is described. With the results of this model, the architecture’s limitations and solutions to these limitations are discussed, and also described several enhancements and modifications of the structure that enhance performance.

[0086] Free Carrier Dispersion

[0087] Free carrier dispersion yields a complex permittivity shift that, according to first order

perturbation theory, induces a fractional cavity resonance shift

For a uniform, perturbative index change throughout the volume containing free carriers (V FC ) - assumed to be greater than or equal to the cavity mode volume

- Eqn. 6 simplifies to

where one can approximate and define as the constant of proportionality

between and carrier density

[0088] In silicon, An is primarily a result of coulomb interactions with the free carrier, while Burstein-Moss band-filling is negligible. Therefore, a simple Drude model analysis of this process estimates that the injection of carriers with a density n c shift the complex permittivity Î by

where q e is the electron charge, e 0 is the vacuum permittivity, ) s is the signal beam’s frequency, and t and m * are the effective mass and mean collision time of the free charge carriers (electrons for“e" subscripts, and holes for“/i" subscripts), respectively. The mean collision time governs the resulting behavior, and can be approximated using the experimentally measured mobilities m b * and effective masses

of free carriers in undoped silicon at room temperature, which yields t 6 =

0.22 ps and For a signal wavelength the high-

frequency limit of Equation 9 for a weakly absorbing medium yields a frequency shift

governed by

and an additional free carrier absorption loss

This corresponding loss can be represented with an additional absorptive quality factor

[0089] can be estimated directly from experimental absorption data, or from the Drude model, where for a cavity resonance at w in a material with a plasma frequency

resulting from the presence of a carrier density n c of individual charges q e with mass m - and mean collision time t as determined by the carrier mobility. This additional loss is negligible for the carrier densities required for switching, as Q abs is much larger than the desired loaded quality factors Qi (~ 10 4 ). One can neglect switching losses and consider the real component of the free carrier-induced index change.

Nonlinear Dynamics [0090] While the free carrier dispersion effect’s blueshift of cavity resonances enables all-optical tuning of the SLM, heating due to absorption and carrier recombination in most thermo-optic media causes an opposing redshift. The competition between these two nonlinearities is typically negligible for a single, short switching event, where the net recombination energy of carriers does not significantly change the temperature of the PhC membrane. However, repeated switching events can lead to long-term patterning effects due to the accumulation of thermal energy in the membrane.

[0091] FIG. 4 shows a thermal model for the nonlinear resonator array. The composite device 410 is comprised of several (approximately) independent single-cavity unit cells 415. The device 410 generally includes a PhC cavity membrane 420, an oxide layer 425, and a metal heat sink 430. The PhC cavity in each cell 415 is patterned into the silicon membrane 420 of thickness t Si which lies above the buried oxide layer 425 of thickness (reflecting a CMOS SOI geometry). Incident

pump light 435 with power is absorbed within the cavity 420, generating free carriers that

are assumed to instantly diffuse to a uniform area where is governed by the ambipolar

diffusion coefficient and free carrier lifetime . Carrier relaxation heats the cavity to a

temperature as prescribed by its thermal capacitance Heat transfer to and from the membrane 415 (with a thermal capacitance at temperature ) as well as through the oxide layer 425 to

the metal heat sink 430 (temperature T 0 ) occur at characteristic rates

respectively. For the purposes of the model of FIG. 4, it is assumed that the control light is external and imaged onto the PhC cavity.

[0092] Since the all-optical SLM relies on repeated free carrier injection, a simple model based on the diagram in FIG. 4 is described here to explore the long-term feasibility of all-optical tuning. In this model, the PhC cavity array is decomposed into unit cells (e.g., the rectangular prism-like cell 415 in FIG. 4), each containing a single PhC cavity and assumed for simplicity to be independent of other unit cells. A periodic pump light pulse (e.g., the pump light 430 of FIG. 4) is projected onto the cavity, resulting in the generation of photoexcited carriers which we assume to instantaneously diffuse to an area (shown as the larger circle 440 in FIG. 4), where r n is equal

to the diffusion length for a given carrier lifetime. The resonant frequency of the cavity instantaneously shifts by an amount proportional to the resulting free carrier density, leading to a variation in the signal field emitted from the cavity. Excess energy from the absorption of above- bandgap pump light heats the PhC cavity. The corresponding increase in cavity temperature T c depends on the cavity’s material, thickness, and area (circle 445 in FIG. 4), which is assumed to be bounded by the surrounding holes. Carrier recombination, as dictated by the free carrier lifetime t h , also leads to heating within this cavity area 445. This thermal energy then diffuses (at a rate given by the material- and geometry-dependent thermal resistances; see arrows in FIG. 4.) to the surrounding membrane at temperature T m or through a buried oxide layer to a heat sink assumed to be at a constant temperature T 0 .

[0093] The described behavior is captured by the following system of first order differential equations:

relaxation pump absorption

relaxation heating

Cavity temp. :

oi pump

absorption heating

Membrane temp. :

[0094] The definitions of the model parameters, as well as their associated values, and material properties of the silicon-on-insulator (SOI) architecture of interest are listed in the Table 1. Eqn. 12 describes the evolution of the cavity signal field (where is energy normalized such that is the total amount of confined energy) which decays at a rate and is driven by (power normalized; = signal drive power). The carrier density evolution (Eqn. 14) is governed by

the absorption of the pump power and the carrier lifetime t The cavity and membrane

temperatures, as described by Eqns. 15 and 16, respectively, are coupled by conductive heat transfer according to the parameters in Table 1. For simplicity, photo-thermal oxidation, signal light effects (free carrier absorption, two-photon absorption, etc.), and other more complex long term effects are neglected.

Table 1

[0095] The following parameters in Table 1 are of particular interest:

• Oxide thickness t ox : controls the characteristic timescale of conduction from the membrane to the underlying heat sink

• Membrane area A m : determines the thermal capacity of the PhC slab within a unit cell (may be large enough to prevent cross talk between unit cells)

• Free carrier lifetime t h : sets the maximum frame rate of the device

• Pump period T : length, relative to , determines the carrier buildup over time

• Pump duty cycle h : sets the fraction of a period over which free carrier nonlinearities

dominate thermal effects

[0096] A portion of the transient numerical solution to Eqns. 14-16 for the default parameters listed in Table 1 is shown in FIGS. 5A-5C. A square wave pump excitation is assumed. Since T » t h , the carrier density (FIG. 5A) decays to ~0 during every cycle; however, the average membrane temperature (FIG. 5B) increases during each cycle. This accumulation occurs since the characteristic timescale of conduction through the buried oxide is much longer than

the pump period T. To prevent thermal accumulation when operating at modulation rates faster than 1/t oc ~ 500 kHz, a thinner oxide layer or supplemental heat sinking mechanism

is used. The slab heating causes the resonance to gradually shift (FIG. 5C) until equilibrium is reached. Despite this slab heating, the distinct contributions of both free carrier and thermal index changes are present throughout the five shown modulation periods. The initial increase in w 0 due to free carriers gives way to an overall decrease as heating from carrier recombination becomes dominant, and the average resonance shift within a modulation period decreases as the membrane is heated.

[0097] FIGS. 6A-6C show a numerical shooting-method solution for the periodic steady state response. The carrier density is shown as a function of time in FIG. 6A. The temperature is shown as a function of time in FIG. 6B. The default parameters yield a negative steady state resonance shift (FIG. 6C) of approximately one linewidth due to the ~ 3K temperature increase of

the membrane (see FIG. 6B); however, positive and negative deviations from this equilibrium are present due to free carrier and thermal effects. In a typical experimental setup, the signal laser would be biased about these values to achieve the desired switching.

[0098] FIGS. 7-10 show periodic steady state solutions for several variations of the“parameters of interest", and show, for each variation, carrier density (FIGS. 7A, 8A, 9A, 10A), membrane temperature (FIGS. 7B, 8B, 9B, 10B), and resonance shift (FIGS. 7C, 8C, 9C, IOC). FIG. 6B shows that, as expected, the mean temperature increase of the membrane is reduced by a factor of about 4 when the oxide thickness is halved to 0.5 mn . Further reduction of the temperature buildup may be possible by, as an example, by using thermally conductive vias connected directly between the PhC slab and underlying heat sink - since further thinning of the oxide layer would cause the underlying metal heat sink to interfere with the PhC cavity mode.

[0099] As illustrated in FIG. 7C, the pump period can be decreased to T = 10t h without significantly reducing the switching contrast. The switching contrast is substantially reduced when T = t h (FIG. 8), verifying that the switching rate is limited by the free carrier lifetime. Although a conservative value t h — 1 ns is assumed based on published recombination lifetimes in PhCs, the effective lifetime is typically much shorter - t h ~ 10 ps - possibly due to diffusion of carriers out of the optical mode, thus enabling switching rates of ~ 10 GHz.

[00100] Reducing the free carrier lifetime limits the timescales over which free carrier dispersion effects are dominant, as depicted in FIG. 11 A for t h = 100 ps. However, reducing the carrier lifetime also reduces the diffusion length, which in turn increases the effective density of carrier in our model and amplifies the free carrier dispersion effect. An all-optical SLM can include a charge confinement technique to increase or maximize the carrier density within the nanocavity and a carrier lifetime control mechanism to reduce t h to the maximum permissible value for a given modulation rate. For instance, surrounding the cavity with an insulator (such as an oxide barrier) that prevents the charges from diffusing would increase the carrier density. Other ways to increase the carrier density can include designing the cavity such that absorption happens in a small region with a sub-bandgap semiconductor, with the rest of the cavity material having a larger bandgap.

[00101] FIGS. 10A-10C showcase two notable effects of reducing the membrane area (in this case to A m = la x 7a): 1) increased steady state temperatures (see FIG. 10B) due to a reduced membrane thermal capacitance, and 2) an accelerated roll-off of free carrier index change (see FIG. 10A) due to the fast competing temperature rise of the reduced thermal mass. Limitations

[00102] These examples lend insight into the dynamics of an experimental system. The experimental nonlinear switching response can be used to determine key coefficients governing the system behavior. The results can also make desirable the development of techniques to circumvent aspects such as potentially slow heat transfer between the PhC slab and underlying heat sink that can result in significant heating of the slab, especially for fast modulation rates. Another aspect to consider is that fast - as short as 3 ps - diffusion of carriers out of the nanocavity limits the impact of free carrier dispersion. Yet another aspect to consider is that excessively long carrier lifetimes of ~1 ns limit the achievable SLM frame rate to sub-GHz.

[00103] Each of these aspects can be addressed with the example enhancements described below.

[00104] Enhancements and Additional Techniques

[00105] PhC light sources: An external probe can be eliminated by using PhC light sources. Active PhC cavity emitters enable modulation rates in excess of 100 GHz due to Purcell effect enhancement of spontaneous emission rates, fJ/bit modulation energies, and tailored emission profiles.

[00106] Other semiconductor resonators: The use of PhC cavities, which afford high Q /C e ff ratios, reduces the pump tuning power. However, the optically tuned architecture can work with any semiconductor resonator, including Fabry -Perot nano-post resonators, which offer lower quality factors but enhanced vertical coupling. In such bases, the cavity array is formed from subwavelength-diameter micropillars with distributed Bragg reflectors on either end.

[00107] Passivation: If desired, the free carrier lifetime can be extended by coating the PhC slab with an A1 2 0 3 passivation layer, which reduces the surface recombination velocity.

[00108] p-i-n junction carrier control: Alternatively, shorter carrier lifetimes can be achieved by sweeping out carriers with an applied electric field. One approach is to integrate a reversed biased p-i-n junction, which has been used to reduce carrier lifetimes from 3 ns to 12 ps in silicon waveguides.

[00109] High performance claddings: Cladding PhC cavities increases thermal dissipation, reduces the impact of fabrication disorder, and increases structural durability. The cladding material can be chosen to hasten heat conduction out of the PhC slab, and the thickness can be reduced to the minimum feasible value with standard wafer thinning techniques. [00110] Passive charge confinement: To enhance carrier density within a nanocavity and minimize cross-talk between resonators, each resonator can be enclosed by a subwavelength insulating (oxide or air, for example) wall. Carriers generated by a pump beam focused on a nanocavity therefore remain within the cavity boundary regardless of diffusion speed.

[00111] Photo-conductive materials: The array could be built with a photo-conductive semiconductor to enable actuation of short wavelength signal beams with longer wavelength pumps. Taking optical dipole trap array formation as an example, a slab made of material sensitive to shortwave infrared radiation could be used to switch a visible wavelength with 1.55 mih telecommunications light.

[00112] Other material systems: Any material with a non-negligible free carrier dispersion effect can be used for the PhC slab. Gallium nitride, for example, has a large 3.4 eV bandgap (for controlling visible wavelengths) while offering electro-refractive tuning performance comparable to silicon.

[00113] Pump absorption enhancement: Numerous techniques are capable of improving the pump absorption efficiency. For example, a reflector below the PhC slab would afford doublepass pump absorption and could enable interferometrically-enhanced absorption. Coupling pump light into a higher-order resonance of the PhC cavity could also improve pump absorption.

[00114] Electrical control: Electrical control of PhC cavities through integrated p-i-n junctions is generally possible; however, contacting individual elements would require the development of new bump bonding techniques, as the standard pitch is around 50 /im.

[00115] Heat dissipation: Thermally conductive vias can be placed to connect the PhC slab to an underlying heat sink. Even silicon, which is two orders of magnitude more thermally conductive than oxide, in a CMOS process could be used for this purpose. Another method (at the expense of increased optical loss) would be to coat the PhC slab in a thin, thermally conductive material - such as graphene or diamond - to remove heat from the slab. For example, the thermal conductivity of graphene-on-Si0 2 is 600 W/m-K (~ 500 X that of oxide alone).

[00116] Coupled cavities: Sub-micron pitches between PhC cavities would be readily accessible if coupling between resonators can be tolerated. One approach would be to arrange the cavities such that the frequency resolved coupled modes generate the desired far-field emission patterns. Example All-Optical SLM Designs

[00117] Example designs for an all-optical spatial light modulator can include those with the following photonic crystal layers: arrays of optimized L3 cavities with swept lattice hole radius in oxide-clad silicon, released silicon, and released silicon nitride; arrays of optimized L4/3 cavities with swept lattice hole radius in oxide-clad and released silicon; arrays of optimized HI cavities with swept lattice hole radius oxide-clad silicon, released silicon, and released silicon nitride; and arrays of silicon, oxide-cladded, optimized HI cavities with integrated gratings for vertical coupling.

Optimized Lattices

[00118] To solve for the lattice parameters, various slab geometries were simulated with the open-source MIT Photonic Bands (MPB) software. For each geometry and material system, a 2D sweep of normalized slab thickness and hole radius was conducted, yielding the

bandgap width and normalized center frequency A desired resonant wavelength lies at the bandgap center when the condition

(17)

is satisfied for a given, constant slab thickness = 220 nm for silicon is assumed).

[00119] The simulations demonstrated that multiple designs satisfied this condition. The results of an oxide-cladded silicon slab, for example, are shown in FIGS. 12A-12E for a mid-gap wavelength The results of (a) indicate that the partial bandgap (bandgap under the

light cone) width is maximized for The grid cells in the third subplot of FIG.

12A represent feasible designs, each associated with a different lattice period a shown in the corresponding location within the second subplot.

[00120] This range of possible solutions to Eqn. 17 is significant for manufacturing tolerances. Two optimal lattices and their associated band structures are shown in FIGS. 12B-E. The first solution (FIGS. 12B and 12C), = 0.262 and a = 410 nm, is what is typically reported in literature. The alternative solution shown in FIGS. 12D and 12E - r/a = 0.379 and a = 505 nm - has a larger lattice constant, and therefore may be more amenable to photolithographic patterning.

[00121] FIGS. 13A-13C illustrate the same results for released silicon. FIG. 13A shows the bandgap tradespace, FIG. 13B shows the mid-gap frequency, and FIG. 13C shows the design offset (Equation 17) for 1550 nm signal light. Partially due to the course sweep mesh used, at least one feasible design with a > 400 nm was found for each slab. Future refinement of these simulations, combined with the experimental results of the fabrication run, may inform the design of an ideal, manufacturable PhC hole lattice for each desired material system.

Cavities

[00122] After selecting the hole lattice parameters, each cavity was designed using published, optimized hole shift values if possible, or optimized using 3D FDTD simulations in MEEP. The mode profiles of two resulting designs for oxide-cladded silicon slab PhC cavities are shown in FIGS. 14 and 15. FIG 14 shows a HI cavity formed by removing one hole from a hexagonal hole lattice in silicon. FIG 15 shows a L4/3 cavity formed by replacing three holes with four holes in a hexagonal silicon lattice.

Conclusion

[00123] While various inventive embodiments have been described and illustrated herein, those of ordinary skill in the art will readily envision a variety of other means and/or structures for performing the function and/or obtaining the results and/or one or more of the advantages described herein, and each of such variations and/or modifications is deemed to be within the scope of the inventive embodiments described herein. More generally, those skilled in the art will readily appreciate that all parameters, dimensions, materials, and configurations described herein are meant to be exemplary and that the actual parameters, dimensions, materials, and/or configurations will depend upon the specific application or applications for which the inventive teachings is/are used. Those skilled in the art will recognize or be able to ascertain, using no more than routine experimentation, many equivalents to the specific inventive embodiments described herein. It is, therefore, to be understood that the foregoing embodiments are presented by way of example only and that, within the scope of the appended claims and equivalents thereto, inventive embodiments may be practiced otherwise than as specifically described and claimed. Inventive embodiments of the present disclosure are directed to each individual feature, system, article, material, kit, and/or method described herein. In addition, any combination of two or more such features, systems, articles, materials, kits, and/or methods, if such features, systems, articles, materials, kits, and/or methods are not mutually inconsistent, is included within the inventive scope of the present disclosure.

[00124] Also, various inventive concepts may be embodied as one or more methods, of which an example has been provided. The acts performed as part of the method may be ordered in any suitable way. Accordingly, embodiments may be constructed in which acts are performed in an order different than illustrated, which may include performing some acts simultaneously, even though shown as sequential acts in illustrative embodiments.

[00125] All definitions, as defined and used herein, should be understood to control over dictionary definitions, definitions in documents incorporated by reference, and/or ordinary meanings of the defined terms.

[00126] The indefinite articles“a” and“an,” as used herein in the specification and in the claims, unless clearly indicated to the contrary, should be understood to mean“at least one.” [00127] The phrase“and/or,” as used herein in the specification and in the claims, should be understood to mean“either or both” of the elements so conjoined, i.e., elements that are conjunctively present in some cases and disjunctively present in other cases. Multiple elements listed with“and/or” should be construed in the same fashion, i.e.,“one or more” of the elements so conjoined. Other elements may optionally be present other than the elements specifically identified by the“and/or” clause, whether related or unrelated to those elements specifically identified. Thus, as a non-limiting example, a reference to“A and/or B”, when used in conjunction with open-ended language such as“comprising” can refer, in one embodiment, to A only (optionally including elements other than B); in another embodiment, to B only (optionally including elements other than A); in yet another embodiment, to both A and B (optionally including other elements); etc.

[00128] As used herein in the specification and in the claims,“or” should be understood to have the same meaning as“and/or” as defined above. For example, when separating items in a list, “or” or“and/or” shall be interpreted as being inclusive, i.e., the inclusion of at least one, but also including more than one, of a number or list of elements, and, optionally, additional unlisted items. Only terms clearly indicated to the contrary, such as“only one of’ or“exactly one of,” or, when used in the claims,“consisting of,” will refer to the inclusion of exactly one element of a number or list of elements. In general, the term“or” as used herein shall only be interpreted as indicating exclusive alternatives (i.e.“one or the other but not both”) when preceded by terms of exclusivity, such as“either,”“one of,”“only one of,” or“exactly one of”“Consisting essentially of,” when used in the claims, shall have its ordinary meaning as used in the field of patent law.

[00129] As used herein in the specification and in the claims, the phrase“at least one,” in reference to a list of one or more elements, should be understood to mean at least one element selected from any one or more of the elements in the list of elements, but not necessarily including at least one of each and every element specifically listed within the list of elements and not excluding any combinations of elements in the list of elements. This definition also allows that elements may optionally be present other than the elements specifically identified within the list of elements to which the phrase“at least one” refers, whether related or unrelated to those elements specifically identified. Thus, as a non-limiting example,“at least one of A and B” (or, equivalently, “at least one of A or B,” or, equivalently“at least one of A and/or B”) can refer, in one embodiment, to at least one, optionally including more than one, A, with no B present (and optionally including elements other than B); in another embodiment, to at least one, optionally including more than one, B, with no A present (and optionally including elements other than A); in yet another embodiment, to at least one, optionally including more than one, A, and at least one, optionally including more than one, B (and optionally including other elements); etc.

[00130] In the claims, as well as in the specification above, all transitional phrases such as “comprising,” “including,” “carrying,” “having,” “containing,” “involving,” “holding,” “composed of,” and the like are to be understood to be open-ended, i.e., to mean including but not limited to. Only the transitional phrases“consisting of’ and“consisting essentially of’ shall be closed or semi-closed transitional phrases, respectively, as set forth in the United States Patent Office Manual of Patent Examining Procedures, Section 2111.03.