A
Structure, process and circuit overlap
A single semiconductor invention may be claimed as a device structure, a fabrication method, a circuit or a system. Relevant earlier disclosures can sit in any of these framings, so the search follows each.
HBM · Chiplets · Packaging · TSV · Memory
Semiconductor Prior Art Search Services
Investigate earlier disclosures before you tape out, file or commit.
Semiconductor innovation now spans devices, processes, packaging, memory and interfaces. Hashi IP Solutions runs technology-focused prior art searches across patent and relevant non-patent literature, so semiconductor R&D and IP teams can review earlier disclosures with structured, traceable findings.
Industry challenge
Why Semiconductor Prior Art Is Hard to Find
The same idea can be claimed as a structure, a process or a circuit — and comparable disclosures are spread across patents, papers and standards-related documents.
B
Dense, evolving terminology
Interposers, bridges, redistribution layers, hybrid bonding and micro-bumps are described in many ways across companies, eras and regions. Searches expand terms, acronyms and classifications deliberately.
C
Heavy non-patent literature
Much semiconductor knowledge is published in conference papers, journals and standards-related documents, not only in patents. These can matter as much as patent publications.
D
Cross-layer inventions
Packaging, thermal, signal-integrity, memory-controller and interface ideas often interact. Comparable solutions may be filed under a neighbouring layer of the stack.
E
Decisions, not just documents
A structured search helps teams find relevant disclosures and inform patentability and product decisions. It does not guarantee patentability, freedom to operate, grant or the avoidance of infringement.
Coverage
Semiconductor Technologies We Search
01
High Bandwidth Memory (HBM)
Stacked DRAM architectures, base and logic dies, memory-stack thermal management, refresh and repair schemes, wide-I/O interfaces and controller–PHY integration.
02
Chiplets and Heterogeneous Integration
Die partitioning, die-to-die interconnects, chiplet interface concepts, known-good-die testing, disaggregated SoC architectures and multi-die system design.
03
Advanced Packaging
2.5D and 3D packaging, silicon and organic interposers, embedded bridges, fan-out wafer-level packaging, redistribution layers, hybrid bonding, warpage control and package-level thermal solutions.
Advanced packaging insight04
Through-Silicon Vias (TSV)
TSV formation, liners and barriers, fill materials, keep-out zones, stress mitigation, wafer thinning, reveal processes and TSV-based stacking.
05
Memory Technologies
DRAM, NAND and NOR flash, 3D NAND, emerging non-volatile memories, memory cells, sensing and programming schemes, error correction and memory-controller techniques.
06
Interfaces and Interconnects
High-speed serial and parallel interfaces, memory interfaces, SerDes, equalisation, clocking, signal and power integrity, and on-package and off-package interconnects.
07
Logic Devices and Process Technology
Transistor architectures such as FinFET and gate-all-around, lithography-related methods, deposition, etching, metallisation, back-side power delivery and process integration.
08
Power and Analog Semiconductors
Power devices including wide-bandgap materials, power management ICs, data converters, RF front-end circuits and mixed-signal design.
09
AI Accelerators and Compute Architectures
Processing-near-memory, compute-in-memory, accelerator microarchitectures, on-chip networks and memory-hierarchy innovations.
Software prior art search010
Test, Reliability and Yield
Design-for-test, built-in self-test, wafer and package test, failure analysis, reliability mechanisms and yield-improvement methods.
Sources & terminology
Searching Beyond One Vocabulary
Coverage depends on the agreed scope, source accessibility and publication dates. No search can promise exhaustive discovery.
Sources we may investigate
- Patent publications and patent-family information
- Technical disclosures in claims, descriptions, figures and embodiments
- Conference proceedings, journals and other technical publications
- Publicly accessible standards-related documents, such as JEDEC or IEEE publications, where relevant and accessible
- Public datasheets, product literature and other dated disclosures, where appropriate
How we expand the search
- Structural features (e.g. ‘bridge die’ → embedded interconnect, silicon bridge)
- Process alternatives (hybrid bonding, thermocompression, micro-bump reflow)
- Company- and era-specific terminology for equivalent structures
- Acronyms and synonyms (HBM, TSV, RDL, FOWLP, GAA, PHY)
- Patent classifications for devices, processes, packaging and circuits
JEDEC & IEEE
Standards-Related Research, Carefully Bounded
Standards as context, not conclusions
Standards-related documents from bodies such as JEDEC and IEEE can describe interface, memory and test concepts with dates attached. They can help identify earlier technical disclosures where they are publicly accessible and in scope.
Access and dating vary
Some standards documents are restricted, paywalled or revised over time. Publication and availability dates are recorded where they can be established, and gaps are noted rather than assumed.
No essentiality findings
A prior art search does not determine whether a patent is essential to a standard, nor does it provide licensing or legal conclusions. Those questions require separate analysis and qualified professional review.
Search objectives
Match the Search to the Decision
Novelty and patentability-oriented searches
Investigate earlier disclosures relevant to the technical features of a proposed semiconductor invention before or during drafting.
Patentability searchTechnical prior art investigation
Explore existing approaches to a defined device, process, packaging or interface problem.
R&D and roadmap support
Help engineering teams understand relevant earlier solutions before committing to a new architecture or process direction.
Invalidity-oriented research
Look for earlier disclosures relevant to the claims of an identified patent, within an agreed scope.
Invalidity searchLandscape and trend analysis
When the question is where a technology area is moving rather than what came before one invention, a landscape study fits better.
Patent landscapeWorkflow
From Semiconductor Invention to Structured Evidence
01
Understand the invention
Identify the device, process or system level, the technical problem, the proposed solution, key features and the search objective.
02
Define the search scope
Establish concepts, synonyms, classifications, jurisdictions, date constraints and relevant non-patent sources.
03
Search patent literature
Investigate relevant patent publications and available family information with a strategy tailored to the invention.
04
Investigate technical literature
Review appropriate conference, journal, standards-related and product literature where relevant and accessible.
05
Analyse and report findings
Organise the most relevant results, explain technical similarities and differences, and provide traceable references for review.
Deliverables
What Your Team Receives
Subject to the agreed scope. Related services: patent search services, novelty search, freedom-to-operate search, invalidity search and patent analytics.
- Executive summary of the principal findings
- Search scope, assumptions and methodology
- Relevant patent publication numbers and links, where available
- Publication dates and patent-family details, where available
- Relevant non-patent and standards-related citations
- Technical summaries of selected references
- Mapping of disclosed features against the invention features under review
- Relevant similarities, differences and limitations
- Observations and follow-up search recommendations, where appropriate
Who it's for
Built for Semiconductor Innovators and IP Teams
Semiconductor companies and fabless teams
Investigate earlier disclosures before advancing device, circuit or architecture innovations into filings.
Memory developers
Review prior disclosures on DRAM, HBM, NAND, emerging memories and controller techniques.
Packaging and OSAT teams
Explore earlier packaging, interposer, bonding and thermal disclosures.
Foundry and process engineers
Investigate process, materials and integration approaches.
Chip architects and AI hardware teams
Assess chiplet, accelerator and memory-hierarchy concepts against earlier work.
Equipment and materials suppliers
Review disclosures relevant to tools, materials and process steps you supply.
Corporate patent teams and IP law firms
Obtain structured technical findings to inform further professional review.
Why Hashi IP
A Technology-First Approach
Technology-first scoping
Scope starts from the actual device, process or system problem and the invention's features.
Cross-layer perspective
Searches consider device, process, packaging, circuit and system framings of the same idea.
Patent and technical literature
Patent publications and appropriate non-patent sources are considered within the agreed scope.
Structured, traceable reporting
Findings are organised so engineering and IP stakeholders can review them quickly.
Connected IP intelligence
Findings can lead into patentability, invalidity, landscape or analytics work where useful.
Start a search
Developing a New Semiconductor Technology? Start With the Earlier Disclosures.
Commission a focused prior art search for your HBM, chiplet, packaging, TSV, memory or interface invention. Findings inform — and do not replace — review by qualified patent professionals.
Please don't share confidential invention details when booking. A short, non-confidential summary is enough to scope the work. See our privacy policy.
FAQ
Questions From Semiconductor R&D and IP Teams
What is a semiconductor prior art search?
A structured investigation of patent and relevant non-patent literature for earlier disclosures related to the technical features of a semiconductor invention, such as a device structure, process, package or interface.
Can you search HBM and stacked-memory inventions?
Yes. Searches can cover stacked DRAM, base dies, TSV-based stacking, thermal management, repair schemes and memory interfaces, using expanded terminology and classifications.
Do you cover chiplets and advanced packaging?
Yes — including die-to-die interconnects, interposers, embedded bridges, fan-out packaging, redistribution layers and hybrid bonding, framed as structures, processes or systems as appropriate.
How are TSV-related inventions searched?
The invention is broken into formation, fill, stress, thinning and stacking aspects, each searched with technical synonyms and classifications across patent and technical literature.
Do searches include JEDEC or IEEE documents?
They can, where documents are publicly accessible and in scope. Some standards documents are restricted or revised over time, so availability and dates are recorded where they can be established.
Does a prior art search determine standard essentiality?
No. Essentiality and licensing questions require separate analysis and qualified professional review. A prior art search identifies earlier disclosures only.
Why does non-patent literature matter in semiconductors?
Many semiconductor concepts first appear in conference papers, journals and technical documents. Excluding them can leave relevant earlier disclosures unexamined.
Can a search cover both process and device claims?
Yes. Because the same idea may be claimed as a structure, a method or a circuit, the search considers each framing relevant to the invention.
Is a prior art search the same as a freedom-to-operate search?
No. A prior art search looks at earlier disclosures. A freedom-to-operate search examines in-force claims a product might fall within, and any infringement opinion remains with qualified patent counsel.
Can a search guarantee that nothing relevant exists?
No search can promise exhaustive discovery. Results depend on the agreed scope, source accessibility, publication dates and terminology used in earlier documents.
What should we provide to scope a search?
A short, non-confidential outline is enough to start: the technology area, the problem addressed, the general approach, target jurisdictions and the decision the search should support. Detailed disclosures follow once confidentiality terms are in place.
When should a semiconductor team consider a prior art search?
Typically before drafting an application, before committing to a new architecture or process direction, or when assessing an identified patent. Timing and scope depend on the invention and the objective.