Design Intelligence Fostering Formidable Energy Reduction and Enabling Novel Totally Impactful Advanced Technology Enhancements (DIFFERENTIATE) SBIR/STTR | DE FOA 0002108

Opportunity Information: Apply for DE FOA 0002108

The Design Intelligence Fostering Formidable Energy Reduction and Enabling Novel Totally Impactful Advanced Technology Enhancements (DIFFERENTIATE) SBIR/STTR opportunity is a Department of Energy ARPA-E funding call aimed at speeding up energy technology innovation by embedding modern machine learning methods directly into the engineering design workflow. ARPA-E exists to push high-risk, high-reward energy ideas that are still too early for most private investors, with an emphasis on technologies that can materially improve US energy and economic security. In practical terms, ARPA-E looks for work that can move the needle within a defined, relatively short development window, helping translate scientific advances into early-stage technologies that later attract follow-on funding and commercialization.

This particular FOA is issued under ARPA-E's joint Small Business Innovation Research (SBIR) and Small Business Technology Transfer (STTR) programs, which are government-wide mechanisms created to stimulate private-sector innovation, meet federal R&D needs through small businesses, and increase the odds that federally funded breakthroughs turn into real products and companies. SBIR also explicitly encourages participation by disadvantaged and women-owned small businesses, while STTR is structured to promote tighter collaborations between small businesses and research institutions. For this DIFFERENTIATE solicitation, ARPA-E indicates funding will be provided for Phase I and Phase II only (not Phase IIs), meaning the program is focused on early feasibility plus a follow-on development phase aimed at a more mature prototype or validated approach.

At the core of DIFFERENTIATE is the idea that many energy engineering efforts share a common design pattern: you start with a hard performance or cost target, propose candidate concepts, test and refine those concepts using models and simulations of increasing fidelity, and then iterate based on what you learn until the solution meets requirements or proves unworkable. The FOA describes this as a four-step loop. First, a concrete challenge is posed, such as achieving very high efficiency electricity generation from natural gas. Second, a solution is hypothesized and initially refined using reduced order models (simplified models that are faster to run and useful for exploring design space). Third, the concept is tested with high-fidelity tools such as partial differential equation based solvers (for example CFD or finite element analysis) and/or experiments and demonstrations. Fourth, the concept is updated based on insights from those higher-fidelity evaluations, and the loop repeats. The bottleneck DIFFERENTIATE is trying to address is that this cycle can be slow, expensive, and heavily dependent on expert intuition, which limits how many ideas can be explored and how quickly designs can converge.

The program objective is to make that design loop dramatically more productive by developing machine-learning-enhanced tools that help engineers generate, evaluate, and optimize designs faster and more creatively. ARPA-E highlights three target capability areas. The first is hypothesis generation tools for conceptual design, meaning methods that help propose promising new system concepts, architectures, or configurations rather than relying only on manual brainstorming and incremental tweaks. The second is high-fidelity hypothesis evaluation tools for detailed design, meaning approaches that integrate machine learning with expensive simulations and experiments to reduce time-to-answer, guide where to spend compute or lab effort, and extract more value from each run or test. The third is inverse design tools, which focus on working backward from desired performance to automatically identify design parameters, structures, or operating strategies that meet those goals, often framed as optimization under constraints.

The FOA is also notable for the kind of teams it tries to attract. It explicitly encourages cross-disciplinary collaborations that combine mathematical optimization and operations research, computer science and machine learning, and domain energy engineering expertise. That emphasis reflects a key assumption of the program: meaningful acceleration will come from pairing strong ML methods with an intimate understanding of the physical systems, constraints, and validation pathways that real energy technologies require, rather than building generic ML software disconnected from engineering reality.

Administratively, the opportunity is a discretionary science and technology R&D funding call using a cooperative agreement, which typically means ARPA-E expects more active involvement than a standard grant (for example, technical milestones, regular engagement, and programmatic guidance). The funding opportunity number is DE-FOA-0002108 under CFDA 81.135. Eligibility is listed as unrestricted in the summary (open to any entity type), with the usual caveat that applicants must still comply with SBIR/STTR rules and any additional eligibility clarifications in the FOA text. The posting indicates an anticipated small number of awards (expected awards: 3) with an award ceiling listed at $1,933,010. The notice also references a concept paper due date of May 20, 2019, and encourages applicants to submit at least 48 hours before the deadline to avoid last-minute submission issues.

Overall, DIFFERENTIATE is best read as ARPA-E placing a strategic bet on machine learning as a force multiplier for engineering design in energy, not by funding a single energy device, but by funding enabling design methods that can be reused across many technologies. The end goal is faster iteration, better exploration of complex design spaces, and quicker progression from an idea to a credible, high-performing engineered solution that can ultimately reduce energy imports, cut emissions, improve efficiency across sectors, and keep the United States at the leading edge of advanced energy technology.

• The Department of Energy, Advanced Research Projects Agency Energy in the science and technology and other research and development sector is offering a public funding opportunity titled "Design Intelligence Fostering Formidable Energy Reduction and Enabling Novel Totally Impactful Advanced Technology Enhancements (DIFFERENTIATE) SBIR/STTR" and is now available to receive applicants.
• Interested and eligible applicants and submit their applications by referencing the CFDA number(s): 81.135.
• This funding opportunity was created on Apr 05, 2019.
• Applicants must submit their applications by May 20, 2019 The Concept Paper Due Date is 05/20/2019. Applicants are strongly encouraged to submit their applications at least 48 hours in advance of the submission deadline.. (Agency may still review applications by suitable applicants for the remaining/unused allocated funding in 2026.)
• Each selected applicant is eligible to receive up to $1,933,010.00 in funding.
• The number of recipients for this funding is limited to 3 candidate(s).
• Eligible applicants include: Unrestricted (i.e., open to any type of entity above), subject to any clarification in text field entitled Additional Information on Eligibility.

Apply for DE FOA 0002108

[Watch] Creating a grant proposal using the step-by-step wizard inside the applicant portal: