Obviousness — US Patent 9525084

To analyze the obviousness of US patent 9,525,084 under 35 U.S.C. § 103, we will identify combinations of prior art references that would render the claims obvious and explain the motivation for a person having ordinary skill in the art (POSITA) to combine them. Given that the provided patent text does not include a "References Cited" section with specific prior art patent or publication numbers, we will rely on the descriptions of "known" and "conventional" technologies presented within the patent itself as representative of the prior art.

Identified Prior Art References:

Prior Art A1 (Conventional Silicon Photodetector): A conventional silicon photodiode (PD) for 850 nm wavelength operation, with a data rate of 1.25 Gb/s, an epitaxial layer greater than 10 micrometers (µm) thickness, and a 53% quantum efficiency. This type of device is characterized by an absorption region whose thickness is too large for high bandwidth and is not easily compatible with CMOS processing. [cite: "a conventional silicon photodiode at 850 nm with a data rate of 1.25 Gb/s has an epitaxial layer greater than 10 ⁇ m thickness and a 53% quantum efficiency.", "However, in order to obtain adequate quantum efficiency, the thickness of the silicon “I” region becomes so large that the device's maximum bandwidth is too low for many current and future telecom and data center applications.", "the relatively thick layers of conventional silicon photodiodes are not easily compatible with CMOS processing."]
Prior Art A2 (Conventional Germanium Photodetector): A conventional germanium (Ge) photodetector, known to detect infrared out to a wavelength of 1700 nm, but with relatively high multiplication noise. [cite: "Germanium (Ge) detects infrared out to a wavelength of 1700 nm, but has relatively high multiplication noise."]
Prior Art B (General Knowledge of Microstructures for Absorption Enhancement): It is generally known that microstructures (e.g., pillars, voids, holes) are used to improve the bulk absorption constant or absorption of a material at wavelengths near the semiconductor material's band gap. This enhancement can be attributed to various optical effects such as resonance, coupled resonances, field enhancement, scattering, and sub-wavelength effects. [cite: "microstructures are used to improve bulk absorption constant or absorption of the material at wavelengths in the proximity of the semiconductor material's band gap. This allows for extending the operating optical wavelengths and/or operating spectrum of the photosensor.", "the microstructures can have effects such as resonance, coupled resonances, field enhancement, near field and sub wavelength effects, scattering, plasmonics, linear and non linear optical field, photonic crystal, absorption mode or lossy mode in high contrast grating at the near field regime, which are both linear and non linear effects that can effectively increase the effective absorption length resulting in a greater absorption of the photons for a given physical length and or the absorption coefficient can be enhanced to an effective coefficient."]
Prior Art C (Desire for Monolithic Integration): The art recognizes the desirability of monolithically integrating photodetectors with electronic circuits (e.g., CMOS ASICs) on a single silicon chip to reduce packaging costs, increase data transmission bandwidth, and reduce electrical power usage. However, this was challenging for conventional photodetectors due to their thick absorption layers being incompatible with standard CMOS processes. [cite: "photodiodes on Si based material as described in this patent specification, can be monolithically integrated with integrated electronic circuits on a single Si chip, thereby significantly reducing the cost of packaging.", "a top or bottom illuminated Si and Ge on Si PD/APD that can be integrated with Si is not known to be commercially available at data rates of 5 Gb/s or more at wavelengths of 850 nm and 1550 nm.", "the relatively thick layers of conventional silicon photodiodes are not easily compatible with CMOS processing."]

Obviousness Analysis of Independent Claims:

1. Independent Claim 25: Silicon Photodetector with Microstructure-Enhanced Absorption

Claim Summary: A photodetector with a silicon photon absorbing region including a plurality of holes, each having a cross-section parallel to the upper substrate surface with a maximum dimension between 400 nm and 2500 nm, and each hole's center spaced apart by less than 3500 nm from a nearest adjacent hole's center.
Combination of Prior Art References: Prior Art A1 (Conventional Silicon Photodetector) + Prior Art B (General Knowledge of Microstructures for Absorption Enhancement).
Motivation to Combine: A POSITA would recognize that the conventional silicon photodiode (Prior Art A1) at 850 nm, while functional, suffers from limited bandwidth due to its thick (>10 µm) absorption layer. To achieve higher data rates (e.g., >5 Gb/s for current and future applications), a thinner absorption region is necessary to reduce carrier transit time. However, simply reducing the thickness of a conventional silicon absorption layer would drastically decrease its quantum efficiency. A POSITA, aware of the general knowledge that microstructures can enhance light absorption (Prior Art B), would be motivated to introduce such microstructures (e.g., holes) into the silicon absorption region of the conventional photodetector. This would enable a reduction in the physical thickness of the silicon layer while maintaining or improving the quantum efficiency, thereby overcoming the trade-off between bandwidth and absorption. [cite: "However, in order to obtain adequate quantum efficiency, the thickness of the silicon “I” region becomes so large that the device's maximum bandwidth is too low for many current and future telecom and data center applications.", "microstructures are used to improve bulk absorption constant or absorption of the material at wavelengths in the proximity of the semiconductor material's band gap. This allows for extending the operating optical wavelengths and/or operating spectrum of the photosensor."]
Reasonable Expectation of Success: The application of known optical effects (resonance, scattering, slow-wave, etc.) associated with microstructures (as explained in Prior Art B) to a semiconductor material like silicon is a recognized approach in photonics. Optimizing the specific dimensions (hole diameter and spacing) as claimed (400-2500 nm maximum dimension, <3500 nm spacing) would be considered a matter of routine design and optimization for a POSITA, utilizing established simulation techniques (like FDTD, which the patent frequently references) to achieve desired absorption characteristics at specific wavelengths (e.g., 850 nm). The patent itself demonstrates through numerous simulations and examples that such microstructure designs achieve significant absorption enhancement.

2. Independent Claim 35: Germanium-Based Photodetector with Microstructure-Enhanced Absorption

Claim Summary: A photodetector with a germanium-based photon absorbing region including a plurality of holes, each having a cross-section parallel to the upper substrate surface with a maximum dimension between 750 nm and 3000 nm, and each hole's center spaced apart by less than 5000 nm from a nearest adjacent hole's center.
Combination of Prior Art References: Prior Art A2 (Conventional Germanium Photodetector) + Prior Art B (General Knowledge of Microstructures for Absorption Enhancement).
Motivation to Combine: A POSITA would know that conventional germanium detectors (Prior Art A2) can detect infrared light out to 1700 nm but are associated with relatively high multiplication noise. Similar to silicon, enhancing absorption in a thinner germanium layer could lead to improved bandwidth and potentially better noise performance due to shorter transit times. Thus, a POSITA, motivated to improve the performance of germanium photodetectors, would apply the known principle of using microstructures to enhance absorption (Prior Art B) to the germanium-based absorbing region. This would allow for a thinner Ge layer, enabling higher speed operation and potentially mitigating some noise issues associated with thicker devices, particularly at wavelengths greater than 1200 nm or 1400 nm. [cite: "Germanium (Ge) detects infrared out to a wavelength of 1700 nm, but has relatively high multiplication noise.", "microstructures are used to improve bulk absorption constant or absorption of the material at wavelengths in the proximity of the semiconductor material's band gap. This allows for extending the operating optical wavelengths and/or operating spectrum of the photosensor."]
Reasonable Expectation of Success: The principles of light-matter interaction and absorption enhancement through microstructures (Prior Art B) are universally applicable across different semiconductor materials. A POSITA would routinely apply these principles to germanium-based materials, adjusting the specific dimensions (hole diameter 750-3000 nm, spacing <5000 nm) to optimize performance for the target longer wavelengths, using standard simulation and fabrication techniques available in the art. The patent presents extensive examples and simulation results for Ge on Si microstructured photodetectors, confirming the feasibility of such structures. [cite: "FIGS. 82A-82C relate to a simulation using finite difference time domain (FDTD) on a Ge on Si microstructured photodiode similar to the structure depicted in FIG. 79", "FIGS. 83A-83C relate to a simulation using finite difference time domain (FDTD) on a Ge on Si microstructured photodiode similar to the structure in FIG. 80"]

3. Independent Claim 1 (Integrated Detector/Processor Circuit) and Independent Claim 39 (Optical/Electronic System)

Claim Summary: These claims describe a single semiconductor chip integrating a microstructure-enhanced photodetector with an electronic processor/active electronic circuit. A key feature is that the photon absorbing region of the photodetector and the electronic processor have thicknesses of the same order of magnitude. Claim 39 further specifies a laser modulated at Gb/s rates and an optical fiber for signal transport.
Combination of Prior Art References: Prior Art A1 (Conventional Silicon Photodetector) + Prior Art B (General Knowledge of Microstructures for Absorption Enhancement) + Prior Art C (Desire for Monolithic Integration).
Motivation to Combine: A POSITA is strongly motivated to monolithically integrate photodetectors with electronic processing circuits (e.g., CMOS, BiCMOS, bipolar devices) on a single silicon chip (Prior Art C) to reduce packaging costs and improve overall system performance, particularly for high-speed data communication. However, a significant barrier to this integration is that conventional silicon photodetectors (Prior Art A1) require thick (>10 µm) absorption layers for adequate quantum efficiency, making them "not easily compatible with CMOS processing," which typically utilizes much thinner active layers. [cite: "photodiodes on Si based material as described in this patent specification, can be monolithically integrated with integrated electronic circuits on a single Si chip, thereby significantly reducing the cost of packaging.", "the relatively thick layers of conventional silicon photodiodes are not easily compatible with CMOS processing."]

To overcome this incompatibility, a POSITA would turn to the known principle of using microstructures to enhance absorption (Prior Art B). By incorporating microstructures (e.g., holes) into the photodetector's absorbing region, its effective absorption coefficient can be significantly increased, allowing for a drastic reduction in the physical thickness of the absorbing layer (e.g., to 0.5-5 µm, as claimed in Claim 13). This reduction in thickness makes the photodetector's absorbing region compatible with the typical thicknesses of electronic circuits like CMOS, bringing their respective thicknesses to "the same order of magnitude" as required by the claims. This compatibility directly enables the desired monolithic integration on a single silicon substrate, thereby achieving the goals of reduced cost and enhanced data transmission capabilities. The patent explicitly states that "the MS-PD/APD epitaxial layer thickness is compatible with the epitaxial layer thickness and structure of CMOS (Complementary metal-oxide-semiconductor) processes and therefore can be integrated with a CMOS ASIC." [cite: "the MS-PD/APD epitaxial layer thickness is compatible with the epitaxial layer thickness and structure of CMOS (Complementary metal-oxide-semiconductor) processes and therefore can be integrated with a CMOS ASIC."]
Reasonable Expectation of Success: Given the clear motivation for monolithic integration and the known mechanism of absorption enhancement via microstructures, a POSITA would have a reasonable expectation of success in combining these elements. The technical challenge would lie in optimizing the microstructure design and fabrication processes to achieve both high quantum efficiency at reduced thickness and compatibility with existing electronic circuit fabrication processes. However, such optimization is well within the capabilities of a POSITA in semiconductor device engineering, supported by simulation tools and extensive knowledge of material science and optical physics. The patent details how this integration can be achieved (e.g., FIGS. 44-52 illustrating integration with TIAs and ASICs).

In summary, the independent claims of US9525084 would likely be considered obvious to a person having ordinary skill in the art when combining conventional photodetector designs with the known principles of microstructure-enhanced absorption, driven by a clear motivation to improve bandwidth, reduce packaging costs, and enable monolithic integration with electronic circuits. The specific dimensions and performance metrics claimed would be considered parameters optimized through routine experimentation and simulation.