Contract analysis

3. Declaration of Dr. Harold Whitfield (Whitfield Technical Consulting Group, Inc.). Dr. Whitfield will offer the opinion that a person of ordinary skill in the art ("POSITA") working in the field of integrated photonics as of the filing date of the '567 Patent would understand the term "adiabatically coupled" to refer to a coupling configuration in which optical mode transitions occur without significant power loss

4. Extrinsic evidence — technical dictionaries and photonics textbooks. Standard reference works in the field of photonics define the term "adiabatic" in the context of waveguide transitions as referring to gradual transitions that preserve the quantum or modal state of the light, without requiring continuity of any specific structural feature. See, e.g., B. E. A. Saleh & M. C. Teich, Fundamentals of Photonics (3d ed

3. Applicant's remarks in the prosecution history. In the applicant's response to the office action, the applicant stated that the claimed invention was distinguished from the Tanaka prior art because "the present invention employs an adiabatic coupling achieved through a continuous taper, unlike the segmented approach of Tanaka." Lumos contends that these remarks confirm that "adiabatically coupled" requires a conti

4. Chen & Liu, "Segmented Taper Coupler Design for Low-Loss Fiber-to-Chip Coupling," Journal of Photonic Integration, Vol. 14, No. 3, pp. 217–229 (2016). Lumos cites this extrinsic evidence to demonstrate that segmented taper designs were known in the art and were treated as distinct from adiabatic taper designs. The article distinguishes "segmented" coupler architectures from "adiabatic" coupler architectures, which

5. Declaration/testimony of Dr. Raymond Ng (Brannick Howell & Stein LLP, Of Counsel / Technical Consultant). Dr. Ng will offer the opinion that a POSITA would understand "adiabatically coupled" in the context of integrated waveguide couplers to require a gradual, continuous taper that prevents excitation of higher-order modes throughout the entire transition region. Dr. Ng will testify that the term carries a specifi

2. The claim language itself is clear and unambiguous and requires no construction beyond its plain and ordinary meaning. The claim recites "a tapered silicon waveguide having a tip width of less than 200 nanometers," which a POSITA would readily understand to mean a silicon waveguide that narrows (tapers) to a tip that is less than 200 nm wide. Lumos's proposed construction impermissibly imports limitations from the

3. Declaration of Dr. Harold Whitfield. Dr. Whitfield will testify that the term "tapered silicon waveguide having a tip width of less than 200 nanometers" is well understood by a POSITA and does not require construction. Dr. Whitfield will further testify that a POSITA would not read the term to require a "single, unitary" waveguide structure, as the claims do not impose such a limitation, and that the phrase "as fa

2. Prosecution history. During an examiner interview, the applicant described the tapered waveguide as a "single structure" that narrows from the input width to the tip. Lumos contends that these statements confirm that the applicant understood the claim to require a single, unitary waveguide rather than a segmented or composite structure.

3. Declaration/testimony of Dr. Raymond Ng. Dr. Ng will testify that a POSITA would understand the term "tapered silicon waveguide" to imply a single, continuously narrowing structure, and that the phrase "as fabricated" appropriately clarifies that the 200 nm tip width limitation is a structural feature of the fabricated device rather than a theoretical or simulated dimension. Term 3: "silicon nitride waveguide laye

3. Declaration of Dr. Harold Whitfield. Dr. Whitfield will testify that a POSITA would understand the term "silicon nitride waveguide layer" to encompass waveguide layers made of silicon nitride in its various stoichiometric and non-stoichiometric compositions, including SiNx films with varying nitrogen content. Dr. Whitfield will further testify that in the silicon photonics field, "silicon nitride" is commonly used

2. Declaration/testimony of Dr. Raymond Ng. Dr. Ng will testify that the term "silicon nitride waveguide" as used in industry standard practice refers to waveguides made of stoichiometric Si₃N₄, and that a POSITA would understand the claim language to require a stoichiometric composition deposited as a single continuous film. Dr. Ng will further testify that non-stoichiometric silicon nitride films (SiNx) have materi

3. Declaration of Dr. Harold Whitfield. Dr. Whitfield will offer the opinion that a POSITA in the field of semiconductor epitaxy and photodetector fabrication would understand the term "continuously varying" in the context of a graded buffer layer to mean that the germanium concentration changes across the thickness of the layer, encompassing both smooth gradients and fine step-grading that approximates a continuous

4. Prosecution history. The '234 Patent was allowed on the first office action following an examiner interview, without any narrowing amendments to the claims. The applicant did not disclaim step-graded buffer layer embodiments during prosecution, and no statements were made during prosecution that would support Lumos's narrow construction requiring a monotonically increasing gradient "without step changes." Lumos's

1. Claim language itself. Lumos contends that the term "continuously varying" is distinct from "step-graded" or merely "varying," and that the word "continuously" has a specific mathematical and physical meaning: without discontinuities. Lumos argues that by using the modifier "continuously," the patentee deliberately chose language that distinguishes a smooth, continuous gradient from a step-graded profile, and that

3. EP 3,234,567 A1 (Nordlicht Photonische GmbH, published August 2, 2017). Lumos cites this European patent application, which was identified as § 102(a)(2) prior art during prosecution of the '234 Patent, as disclosing step-graded germanium buffer layers for photodetector applications. Lumos argues that because the prior art disclosed step-graded buffer layers, the '234 Patent's use of the term "continuously varying

4. Declaration/testimony of Dr. Raymond Ng. Dr. Ng will testify that a POSITA in the field of semiconductor epitaxy would understand "continuously varying" to mean a smooth, monotonic gradient without step changes, consistent with standard terminology in the field. Dr. Ng will further testify that "continuous" and "step-graded" are well-recognized and distinct categories of compositional grading in semiconductor buff

2. Declaration of Dr. Harold Whitfield. Dr. Whitfield will testify that a POSITA would understand "germanium absorber layer" to encompass layers of germanium or germanium-rich material (including, for example, SiGe alloys with high germanium content) that serve as the photon absorption region of a photodetector. Dr. Whitfield will further testify that there is no basis in the claim language or specification to read a

2. Declaration/testimony of Dr. Raymond Ng. Dr. Ng will testify that in industry convention, a "germanium absorber" in the context of a germanium-on-silicon photodetector refers to an absorber layer composed of high-purity germanium, and that a POSITA would understand the term to imply a germanium atomic fraction of at least 98%. Dr. Ng will further testify that absorber layers with lower germanium content (e.g., SiG

1. Preamble is not limiting. The preamble of Claim 1 states "A germanium-on-silicon photodetector with enhanced responsivity." The body of Claim 1 recites all of the structural and functional limitations of the claimed invention, including the silicon substrate, graded buffer layer, germanium absorber layer, and their spatial and structural relationships. The phrase "enhanced responsivity" in the preamble merely desc

3. Declaration of Dr. Harold Whitfield. Dr. Whitfield will testify that a POSITA would not read a specific numerical responsivity threshold from the preamble language "enhanced responsivity," and that the term as used in the preamble merely conveys the general concept that the claimed photodetector structure achieves improved responsivity relative to prior art designs that lack the claimed graded buffer layer. Lumos'

2. Preamble gives life and meaning to the claim. Lumos argues that the preamble phrase "with enhanced responsivity" is essential to understanding the claimed invention and should be treated as a limiting claim element. Lumos contends that the preamble gives "life, meaning, and vitality" to the claim because it defines the purpose and scope of the invention — a photodetector that achieves enhanced responsivity as a di

3. Declaration of Dr. Harold Whitfield. Dr. Whitfield will testify that a POSITA would understand "monolithically formed on a single silicon die" to mean that the WDM component is fabricated on the same die during the same semiconductor fabrication process flow, encompassing both FEOL processing (e.g., waveguide patterning and doping) and BEOL processing (e.g., metallization and interconnect formation). Dr. Whitfield

4. Prosecution history. While the term "monolithically formed" was added by amendment during prosecution of the '678 Patent, the applicant's remarks in support of the amendment distinguished the claimed invention from heterogeneous bonding of separately fabricated optical components onto the silicon die. Specifically, the applicant stated: "The present invention is distinguished from heterogeneous integration approac

3. Declaration/testimony of Dr. Raymond Ng. Dr. Ng will testify that a POSITA would understand "monolithically formed" in the context of silicon photonics to require that the WDM element be fabricated entirely within the same substrate using standard semiconductor processing, excluding any heterogeneous bonding, flip-chip attachment, or transfer printing of separately fabricated optical components. Dr. Ng will testif

4. Lumos's accused LumosBridge 800G platform. Lumos notes that its accused LumosBridge 800G transceiver platform uses micro-transfer-printed III-V gain elements that are bonded to the silicon photonic integrated circuit ("PIC"). Lumos contends that the scope of this claim term is directly relevant to the infringement dispute, as OWS's broader construction would encompass Lumos's heterogeneous integration approach, wh

1. Claim language. The claim language recites "multi-channel optical transceiver" without specifying a minimum number of channels or requiring that the transceiver simultaneously transmit and receive. The plain meaning of "multi-channel" is "having more than one channel," and the plain meaning of "optical transceiver" is a device capable of transmitting and/or receiving optical signals. The claim does not recite "at

3. Declaration of Dr. Harold Whitfield. Dr. Whitfield will testify that a POSITA would understand "multi-channel optical transceiver" in accordance with its plain and ordinary meaning — an optical transceiver operating on multiple (i.e., more than one) wavelength channels — and that there is no basis to read in a requirement of "at least four" channels, simultaneous transmit/receive, or a "single optical interface."

3. Declaration of Dr. Harold Whitfield. Dr. Whitfield will testify that a POSITA would understand "electrically co-integrated with the modulator on a common substrate" to mean that the equalization circuit is fabricated on or placed on the same substrate as the modulator, such that the two are electrically connected via on-chip interconnects rather than off-chip connections. Dr. Whitfield will further testify that th

4. Prosecution history. During prosecution of the '890 Patent, the applicant added the "co-integrated…on a common substrate" limitation to distinguish prior art describing driver circuits that were "separately fabricated and bonded" to the modulator die. The applicant's remarks during prosecution focused on the distinction between separate fabrication (followed by bonding) and co-integration on a common substrate. Th

2. Applicant's remarks during prosecution. During prosecution, the applicant stated: "The present invention is distinguished from approaches in which the driver circuit is separately fabricated and subsequently bonded or attached to the modulator die. In the claimed invention, the equalization stage is co-integrated with the modulator on a common substrate, eliminating the need for bond wires, flip-chip connections,

4. Declaration/testimony of Dr. Raymond Ng. Dr. Ng will testify that in the field of silicon photonics, "co-integrated" is a specific term of art meaning monolithic fabrication in a shared process flow, and that a POSITA would understand the term to require that the driver equalization circuit and the modulator be fabricated in the same semiconductor process, sharing the same doping profiles, process layers, and meta

5. Lumos's accused product. Lumos notes that its accused product uses a separate CMOS driver die that is flip-chip bonded to the silicon photonic modulator die via copper pillar micro-bumps. Lumos contends that this product architecture falls outside the scope of the claim under Lumos's proposed construction, as the driver and modulator are separately fabricated and connected via flip-chip bonding rather than being c

2. Declaration of Dr. Harold Whitfield. Dr. Whitfield will testify that a POSITA would understand "pre-emphasis equalization stage" to encompass any circuit stage that applies pre-emphasis to compensate for frequency-dependent signal loss in the modulator, and would not limit the term to FIR filters or any specific number of filter taps. Dr. Whitfield will further testify that "pre-emphasis" is a general term in the

2. Declaration/testimony of Dr. Raymond Ng. Dr. Ng will testify that "pre-emphasis equalization" in the context of high-speed Mach-Zehnder modulator drivers conventionally refers to FIR-based feed-forward equalization, and that a POSITA would understand the term as used in the '890 Patent to require a FIR filter with at least two taps. Dr. Ng will further testify that other equalization techniques such as CTLE are ty

1. Claim language. The claim recites "common substrate," which plainly means a shared substrate. The claim language does not include the exclusionary language Lumos seeks to import — it does not exclude multi-chip modules, interposers, silicon bridges, or any other substrate configuration. Lumos's proposed construction improperly reads a series of negative limitations into the claim that are absent from the claim lan

3. Declaration of Dr. Harold Whitfield. Dr. Whitfield will testify that a POSITA would understand "common substrate" in the context of the claim to mean a shared substrate on which both the modulator and the driver circuit are disposed, without importing the extensive exclusions Lumos proposes. Dr. Whitfield will note that Lumos's proposed construction effectively converts a positive claim limitation ("on a common su

2. Prosecution history context. Lumos argues that the phrase "co-integrated…on a common substrate" was added as a unitary phrase during prosecution to distinguish multi-die approaches in which the driver and modulator are fabricated on separate substrates and subsequently connected. Lumos contends that "common substrate" must be read in conjunction with "co-integrated" as a single, integrated limitation, and that the

3. Declaration/testimony of Dr. Raymond Ng. Dr. Ng will testify that "common substrate" in the context of "co-integrated…on a common substrate" means a single monolithic semiconductor substrate, and that a POSITA would understand the term to exclude multi-chip modules, interposers, silicon bridges, and substrates formed by bonding or attaching separately fabricated semiconductor dies. Dr. Ng will testify that this un