Overview of ARPA-E
This chapter provides a descriptive overview of ARPA-E, starting with its origins and attributes. It then describes the agency’s organizational structure and presents sufficient detail on its operations and procedures to enable readers not familiar with the agency to better understand the assessment presented in Chapters 3 and 4.
ARPA-E was recommended to Congress in the 2005 National Academies report Rising above the Gathering Storm: Energizing and Employing America for a Bright Economic Future. (NAS et al., 2007). The purpose of that report was to recommend specific federal government actions with the potential to maintain and expand U.S. competitiveness and to detail concrete steps for acting on those recommendations. ARPA-E grew out of the following recommendation:
The federal government should create a DARPA [Defense Advanced Research Projects Agency]-like organization within the Department of Energy called the Advanced Research Projects Agency-Energy (ARPA-E) that reports to the under secretary for science and is charged with sponsoring specific R&D programs to meet the nation’s long-term energy challenges....ARPA-E would provide an opportunity for creative “out-of-the-box” transformational research that could lead to new ways of fueling the nation and its economy, as opposed to incremental research on ideas that have already been developed.…Like DARPA, ARPA-E would have a very small staff, would perform no R&D itself, would turn over its staff every 3 to 4 years, and would have the same personnel and contracting freedoms granted to DARPA (NAS et al., 2007, p. 152).
It was further recommended that ARPA-E be funded at an initial level of $300 million in its first year, rising to $1.0 billion per year over a 5- to 6-year timeframe.
In 2007, the 110th Congress enacted the America COMPETES Act (“the Act”), which implemented a number of the Gathering Storm’s recommendations, including the creation of ARPA-E (Gonzalez, 2015). The 2007 Act was subsequently superseded by the America COMPETES Reauthorization Act of 2010, which incorporated much of the original language of the 2007 Act but introduced some modifications, including the language authorizing ARPA-E.1 In 2009, ARPA-E received its first appropriated funds and commenced operations.2
The Act charters ARPA-E to look much like DARPA. It tasks ARPA-E with funding high-risk, potentially high-return research, to translate “scientific discoveries and cutting-edge inventions into technological innovations,”3 and leaves the agency free to support research into any type of technology or fuel that supports its mission. To enable the agency to be well positioned to identify and support this kind of research, the Act exempts it from many federal rules and regulations. The Act also deliberately attempts to structure the agency to be different from other Department of Energy (DOE) offices or programs,4 especially with regard to its level of flexibility, and to position it to be relatively autonomous within the department. For example, the Act provides funding separate from that for ongoing DOE programs and standing offices, gives the director considerable control over the agency, and seeks to make it separate from the rest of the DOE hierarchy. The director is to report directly to the secretary of energy rather than another office.5 ARPA-E’s budget request and congressional appropriations are to be made “separate and distinct from the rest” of the DOE budget, and without specification regarding how the appropriation should be allocated.6
1 P.L. 111-358.
2 The American Recovery and Reinvestment Act of 2009 provided an initial $400 million to fund the establishment of ARPA-E.
3 42 U.S.C. § 16538(c)(2)(B) (2017).
4 The term “program” has a very specific meaning for ARPA-E that is different from that in the rest of DOE. For ARPA-E, a program refers to a specific focused technology funding opportunity and the research projects funded under it. These always have a narrow, defined scope, limited duration, and no permanent staff. Programs in the rest of the department can refer to an ongoing activity of indefinite duration, with a scope that can be broad and may change over time, has long-term staff, and may or may not provide funding for research activities.
5 42 U.S.C. § 16538(d).
6 42 U.S.C. § 16538(n)(3).
To run ARPA-E and make investments in projects, the director is mandated to hire program directors. These individuals are to serve for a 3-year term that can be renewed, and are given a great deal of autonomy and flexibility in establishing and managing agency programs. The Act also empowers the program directors with significant discretion in recommending projects for funding (as opposed to strict reviewer scoring) and assigns to them the authority and obligation to recommend that the director terminate projects that fail to meet expectations. Preparing technologies for eventual transfer from laboratory to market is a key element of ARPA-E’s mission (see below), termed “technology-to-market,” or “T2M.” ARPA-E project teams prepare a Technology-to-Market Plan, with agency staff from the Technology-to-Market Program assisting in the plan’s development and implementation. Fairly or not, ARPA-E is sometimes compared and contrasted with DOE’s Office of Energy Efficiency and Renewable Energy (EERE), a much larger research organization with 700 employees, hundreds of contractors, and seven field offices (Ling, 2013). Reflecting the mandate extended to ARPA-E by Congress to enable it to fulfill its mission, the agency features several important structural characteristics that differ from those of most other federal research funding organizations. Figure 2-1 provides a high-level schematic of those characteristics and key operational features of ARPA-E. Each of those is described in this chapter.
Although ARPA-E is not the first federal organization tasked with supporting high-risk, potentially high-reward research or the first government entity intended to emulate DARPA, it enjoys statutory and resource advantages not available to any of its antecedents other than DARPA itself. For example, DOE operated a small organization dedicated to novel, high-risk energy research between 1977 and 2000, but this organization—the Advanced Energy Project Division (AEP)—was firmly embedded in the DOE structure, fully subject to federal civil service requirements, and ultimately unable to achieve “transformational” results (DOE, 1999).7 The Homeland Security Act of 2002 created the Homeland Security Advanced Research Projects Agency (HSARPA) to support “cutting-edge research” aimed at developing “revolutionary changes in technologies, new capabilities and threat and risk assessments” related to homeland security. HSARPA’s funding has been limited compared with that of DARPA (in fiscal year [FY] 2007, for example, $38 million compared with $3.1 billion for DARPA), and for most of its existence the Department of
7 In the mid-1990s, AEP was operating under annual budgets of $10–11 million. In 1995, it was absorbed into DOE’s Computational and Technology Research Program. DOE began ramping down AEP’s funding in fiscal year (FY) 1998, requesting only $3 million in its FY 1999 budget, directing the cancellation of several projects, and barring the initiation of “new novel energy-related concepts.”
Homeland Security has ensured that its projects involve conventional research and development (R&D) with moderate risk.8
Agency Mission and Goals
Congress statutorily defined ARPA-E’s mission as to “overcome the long-term and high-risk technology barriers in the development of energy technologies.”9 The agency’s basic statutory goals are to develop technologies that reduce imports of fossil fuel, reduce energy-related emissions, improve energy efficiency in all economic sectors, and ensure that the United States maintains a technological lead in the development and deployment of advanced energy technologies.10 Additionally, the statute directs ARPA-E to pursue these objectives through particular means:
- identifying and promoting revolutionary advances in fundamental and applied sciences;
- translating scientific discoveries and cutting-edge inventions into technological innovations; and
- accelerating transformational technological advances in areas industry is unlikely to undertake.11
The director of ARPA-E is responsible for achieving the agency’s statutory goals through targeted acceleration of “novel early-stage energy research with possible technology applications” and “advanced manufacturing processes for the domestic manufacturing of novel energy technologies.”12 The Act emphasizes that ARPA-E is expected to pursue those research technology themes and specific ideas that industry is not likely to undertake by itself.13
8 A 2007 analysis found that while Congress had provided HSARPA with a strong and flexible mandate modeled on DARPA, “HSARPA has never been adequately utilized or implemented....HSARPA was never allowed autonomy and flexibility and instead was closely controlled by a budget and policy bureaucracy within the S&T Directorate that limited HSARPA’s funding and effectively made all R&D investment and award decisions.” Testimony of William B. Bonvillian, U.S. Congress, Subcommittee on Energy and Environment, Committee on Science and Technology, Establishing the Advanced Research Projects Agency—Energy, Hearing, 110th Cong., 2d sess. (April 26, 2007).
9 42 U.S.C. § 16538(b).
10 42 U.S.C. § 16538(c).
11 42 U.S.C. § 16538(c)(2).
12 42 U.S.C. § 16538(e)(3)(A) and (C).
13 42 U.S.C. § 16538(c)(2)(C).
ARPA-E’s director is authorized to hire up to 120 “scientific, engineering, and professional personnel without regard to the civil service laws.”14 The director may set pay levels up to Level II of the Office of Personnel Management’s Executive Schedule (EX-II), again without regard to the civil service laws.15 Additional payments may be made within limitations.16 This authority is meant to empower ARPA-E’s director to hire scientists and engineers to serve as program directors quickly and to attract talented individuals. As noted above, use of this hiring authority for program directors is limited to a 3-year appointment, which may be renewed. This provision ensures that program directors pass through ARPA-E with the intent of executing technically challenging projects during their tenure, and that new talent and ideas circulate through the agency (H. Rep. 110-289, 2007).
Additionally, the secretary of energy, or the director of ARPA-E “serving as agent of the Secretary,” may contract with private recruiting firms to hire “qualified technical staff.”17 The director is authorized to use “all authorities in existence on August 9, 2007” provided to the secretary of energy to hire administrative, financial, and clerical staff.18
Powers Vested in Program Directors
Similar to DARPA program managers, ARPA-E program directors are accorded wide latitude in identifying research themes; creating new programs; making funding recommendations to the director; supervising projects; identifying commercial opportunities; and, when necessary, terminating projects.19 Program directors are expected to identify “innovative cost-sharing arrangements” for ARPA-E projects (albeit subject to the limitations established by federal cost-sharing regulations) and to help performers identify “mechanisms for commercial applications” of successful energy technology development projects, including partnerships between awardees and commercial entities.20
14 42 U.S.C. § 16538(g)(3)(A)(i).
15 42 U.S.C. § 16538(g)(3)(A)(ii).
16 The extra payment cannot exceed the least of three amounts: (1) $25,000, (2) 25 percent of the employee’s base rate of pay, and (3) the limitation amount applicable in a given calendar year pursuant to 5 U.S.C. § 5307(a)(1). That section prohibits payments that, when added to the base amount, would exceed the Office of Personnel Management’s Executive Schedule Level I (EX-1), which in 2016 was $205,700. 42 U.S.C. § 16538(g)(3)(A)(iii).
17 42 U.S.C. § 16538(g)(3)(D).
18 42 U.S.C. § 16538(g)(3)(C).
19 42 U.S.C. § 16538(g)(2)(B).
20 42 U.S.C. § 16538(g)(2)(v)-(vii). Federal cost-sharing requirements for research contracts are set forth in 2 C.F.R. § 200.306.
ARPA-E’s institutional autonomy is one of the agency’s most notable attributes. While the Gathering Storm recommendation suggests housing ARPA-E within what was then DOE’s Office of Science, Congress decided to make it an autonomous agency within DOE, outside of the departmental bureaucracy (H. Rep. 110-289, 2007). The director is appointed by the President, with the advice and consent of the Senate, and reports directly to the secretary of energy.21 No other programs within DOE are to report to the director of ARPA-E, a provision that prevents the director from simultaneously heading another office or program within the department (Stine, 2009).22 DOE’s budget request for ARPA-E is to be “separate and distinct” from the rest of its budget, and congressional appropriations for the Energy Transformation Acceleration Fund administered by ARPA-E are likewise to be “separate and distinct” from the remainder of the department’s budget.23 As of this writing, ARPA-E is even physically separate from the rest of DOE, with offices in a different building in Washington, DC.
Since ARPA-E’s core mission is to fund “high-risk” projects, the drafters of its enabling legislation clearly contemplated that some of the projects it funded would prove unable to meet their targets. Consequently, ARPA-E’s director is responsible for “terminating programs carried out under this section that are not achieving the goals of the programs.”24 Program directors are responsible for “recommending program restructure or termination of partnerships between awardees and commercial entities.”25 The Conference Report on the 2007 America COMPETES Act states:
Similar to the Defense Advanced Research Projects Agency the Director is to establish and monitor milestones, initiate research projects quickly, and just as quickly terminate or restructure projects if such milestones are not achieved (H. Rep. 110-289, 2007).
21 42 U.S.C. § 16538(d)(1) and (3).
22 42 U.S.C. § 16538(d)(4).
23 42 U.S.C. § 16538(n)(3). ARPA-E’s director “to the maximum extent practicable” is to ensure that the agency’s programs are coordinated with, and do not duplicate, efforts of programs and laboratories within DOE and other relevant research agencies. The director of ARPA-E may, “to the extent appropriate,” coordinate technology transfer efforts with DOE’s Technology Transfer Coordinator. 42 U.S.C. §16538(i)(1) and (2). DOE’s technology transfer coordinator advises the secretary “on all matters relating to technology transfer and commercialization.” 42 U.S.C. § 16391.
24 42 U.S.C. § 16538(e)(4).
25 42 U.S.C. § 16538(g)(2)(B)(viii).
Federal Funding of ARPA-E
As indicated in Table 2-1, ARPA-E has never been funded at the scale or level of resources envisioned by the committee that produced the Gathering Storm report. Nevertheless, it has funded more than 400 projects that it has regarded as potentially transformational. As noted above, ARPA-E began operations in 2009, when the Omnibus Appropriations Act of 2009 provided $15 million in funding, augmented in that year by $400 million in special funding appropriated from the American Recovery and Reinvestment Act of 2009 (ARRA, also commonly known as “the Stimulus”)26 (ARPA-E, 2011b). Thereafter, ARPA-E’s annual appropriations were provided under the annual Energy and Water Development and Related Agencies Appropriations Act, generally at the level of $280 million per fiscal year27 (Gonzalez, 2015).
TABLE 2-1 Recommended Funding Levels (in millions of dollars) for ARPA-E versus Authorizations and Appropriations, 2007–2017
|Fiscal Year||NAS et al. (2007) Recommendation||America COMPETES 2007||America COMPETES Reauthorization 2010||Appropriations|
* $15 million in regular appropriations and $400 million in one-time funding pursuant to the ARRA.
26 The Omnibus Appropriations Act of 2009 provided $15 million for ARPA-E in its Science appropriation. The Energy and Water Development and Related Agencies Appropriations Act of 2010 transferred the $15 million to ARPA-E.
27 In 2013, the budget exercise known as “sequestration” and enacted recessions reduced the appropriated amount to $251 million.
STRUCTURE AND OPERATIONS
Although ARPA-E was authorized by the America COMPETES Act in 2007, it was not funded until 2009. DOE staff estimated it would take approximately 1 year for ARPA-E to become operational. DOE leadership pushed for a more aggressive timetable that included simultaneously undertaking a number of tasks required to establish the agency that would otherwise have been done serially. These tasks included hiring a director and key personnel, and drafting solicitations for industry and university R&D proposals (Leopold, 2009).
The initial hires had experience with energy technology start-ups and venture capital, and were chiefly responsible for building the agency and commencing operations until the Senate confirmed the agency’s first director, Arun Majumdar, in October 2009. Early agency personnel said that building the agency from scratch under the aegis of the America COMPETES Act’s special authorities allowed it to exist outside of the normal “layers of bureaucracy.” Although modeling the new agency on DARPA was not taken as a given, they eventually decided to look to DARPA as a model for many of ARPA-E’s structural features and operational practices. Zan Alexander, a former DARPA deputy director, served as a consultant to ARPA-E, and her experience facilitated the adoption of a number of DARPA practices (Marks, 2014).
A number of key characteristics of ARPA-E have been noted as guiding the agency’s creation and as essential to its carrying out its mission:
- The agency must be structured to support risk taking as opposed to the risk avoidance that commonly characterizes federal agencies.
- The agency must be able to recruit world-class talent, which in turn requires empowerment of program directors to initiate and run programs of their own creation.
- ARPA-E personnel should normally leave the agency after 3 years.
- The agency’s culture should be characterized by “constructive challenge,” with each program director’s program being subject to scrutiny by other program directors.
- The agency’s budget request should remain at a general level, rather than being broken down into multiple line items, thus preserving ARPA-E’s discretion in seeking to fund potentially transformational technologies28 (Danielson, 2015; Orr, 2015).
28 Under Secretary of Energy for Science and Energy Franklin Orr observes that outside of ARPA-E, DOE undergoes “deep scrutiny” of its budgets by Congress and the Office of Management and Budget (OMB), which probe “very far down into budget line items.”
ARPA-E made its first Funding Opportunity Announcement (FOA) in May 2009. In response, the new agency received 3,700 concept papers. From these, it encouraged 334 full applications, which were reviewed by two sets of panels of external reviewers. Based on these reviews and a rigorous selection process, ARPA-E selected 37 projects for funding. The first awards were made in October 2009, 6 months after the agency had received its initial appropriated funds (Lawrence and Abramson, 2014). Awards were made by January 15, 2010, for 35 of the 37 selected projects, a speed that “has now set records within the DOE” (Majumdar, 2010).
Arun Majumdar stated that his initial challenges were threefold: developing the external relationships crucial to the agency’s success, creating a new government organization, and launching the right projects and demonstrating results (Lawrence and Abramson, 2014). Meeting these challenges entailed three tasks:
- Outreach—Majumdar “reached out to key people in the Department of Energy,” established an advisory committee of energy experts, established relationships with other government agencies to identify potential synergies, and convened a meeting of stakeholders in Washington, DC.
- Organization building—Majumdar prioritized hiring the most talented individuals and establishing a culture in which they could thrive. Emphasis was placed on engaging in open dialogue, challenging ideas and proposals, and empowering individuals while also holding them accountable.
- Project selection—ARPA-E’s early funded projects focused on themes that included extremely efficient photovoltaic solar collectors, batteries for transportation, wind turbines, and geothermal energy (Lawrence and Abramson, 2014).
Majumdar has commented on the ease of creating new processes and establishing culture at a new agency relative to attempting to change an existing organization. This process was undertaken at a rapid pace that felt like “building the plane while we are flying it.” He also believes this rapid pace helped establish a culture featuring an attitude of accomplishing great things and a sense of urgency (Lawrence and Abramson, 2014). Early program directors corroborate this view, stating that they entered an environment with no existing programs and thus no baseline, and that early leadership was focused on hiring the best and the brightest talent willing to join ARPA-E and then providing them with autonomy and empowerment to identify the most critical and important problems to solve.
He comments that “ARPA-E avoids all of that,” and that Congress and OMB are “not normally willing to do that.”
The first generation of ARPA-E programs reflected the “blue sky” character of the agency’s early days. Over time, as successive programs came into existence, a baseline was created, and subsequent program directors were hired with consideration of their specific expertise in broad thematic areas the agency had identified as strategic.29
Organizational Framework and Staffing
ARPA-E’s structure and operational practices generally resemble those of DARPA, albeit on a smaller scale30 (Bonvillian and van Atta, 2011). Like DARPA, ARPA-E has a “flat,” three-level organizational structure, as shown in Figure 2-2. Its current staff consists of a director, 2 advisors to the director, 3 deputy directors, and 16 program directors, plus 4 fellows; 15 acquisitions, budget, external affairs, and legal staff; and 13 T2M advisors who work with performers to develop and implement commercialization strategies. A number of contractors provide back-office support in the areas of communications, information technology (IT), administration, and contracting support. ARPA-E also utilizes systems engineering and technical assistance (SETA) contractors, who play a large role in assisting program directors with the active management of awards.
ARPA-E has adopted many DARPA practices, including empowerment of program directors, a streamlined project approval process, hands-on project
29 Personal consultation with Ilan Gur, former ARPA-E Program Director, founder and director of Cyclotron Road at Lawrence Berkeley National Laboratory (September 22, 2015).
30 ARPA-E’s original staff included a former DARPA program manager and an experienced DARPA advisor and performer.
management, use of special hiring authority to build a high-quality human infrastructure, and maintenance of institutional autonomy. Since 2014 it has administered the government contracting process itself using its own contracting staff. This approach enables the agency to hand-pick its procurement staff; act more quickly in such matters as creating contracts and cooperative agreements; and, when appropriate, act flexibly and take strategic risks. ARPA-E has introduced new institutional arrangements, such as the creation of its internal T2M program (reflecting the reality that, unlike DARPA awardees, ARPA-E performers do not sell products primarily into a single-stream procurement system) and a research fellows program to enhance its in-house scientific and technical expertise. Figure 2-3 illustrates how ARPA-E has borrowed from and how it departs from the DARPA model.
As of the time of this study, new programs appear to arise almost solely because a prospective program director pitches the idea as part of his or her interview for the position. Once a program and its program director are accepted, ARPA-E begins a new program with a “deep dive”—a comprehensive exploration of an energy-related theme aimed at identifying individual topics that represent potential projects for the development of technologies that would
enable the agency to achieve one or more of its goals. The objective of the exercise is to identify technology “white spaces” not likely to be addressed by the private sector or by other federal research programs. These white spaces can be of at least two kinds, as shown in Figure 2-4. First, a white space can be a gap in the funding pipeline of all government and private R&D funding (e.g., no funding available for a specific technology at technology readiness level [TRL] 3). ARPA-E funding can move in to address this gap, filling the white space. Second, white space can be a capability just beyond the frontier of what is currently technically possible, which ARPA-E identifies as a strategic need. ARPA-E can attempt to push this frontier by giving directed funding for projects at the frontier.
ARPA-E technical staff review the scientific literature, conduct site visits (universities, companies), commission external studies, attend conferences, and participate in webinars to identify such areas, examine the current state of the art, and identify the major players and the principal technological hurdles.31 One of the agency’s original program directors, for example, spent 6 months canvassing the energy community to identify white spaces in research on power system grids (Heidel and Gould, 2015). ARPA-E solicits additional thematic information by issuing requests for information (RFIs) on topics of interest, which include technical questions for which the answers can assist in assessing potential FOAs (ARPA-E, 2012a).
One ARPA-E program director stated that program directors often draw on the knowledge base at DOE when formulating new projects, primarily
31 Response by ARPA-E to question from Dr. Paul Broun, Chairman, U.S. Congress, House Subcommittee on Investigations and Oversight, Committee on Science, Space and Technology, Hearing, 112th Cong., 2d sess. (January 24, 2012), 73.
through informal processes. They meet with DOE research managers and directors to learn what they are working on and solicit their comments on draft project ideas. Importantly, the program directors ask their DOE colleagues whether they are aware of other places where research related to a draft project theme is ongoing, and “you know you’re on the right track if you get blank stares or they say ‘that will never work’” (Heidel and Gould, 2015).
Once ARPA-E has identified technological white spaces, program directors convene technical workshops to consult leading experts from relevant scientific, engineering, and commercial communities, an exercise that informs the project development process and helps break down silos between disciplines. ARPA-E staff concurrently engage in discussions with other DOE personnel to identify gaps in existing research, determine the current state of the art, and draw on the department’s internal expertise (ARPA-E, 2011b). The workshops serve as another means of determining the state of the art in a given field while providing a forum in which to discuss solutions to the critical challenges identified and determine performance targets the technical community thinks are aggressive yet reachable. While not all workshops lead to ARPA-E projects, they do inform the direction of projects in particular energy technology areas. At the close of each workshop, the program director may propose a new project, but must answer the questions shown in Box 2-1 satisfactorily before it can go forward.
The proposed project is then subjected to “constructive confrontation” and debate with the participation of all ARPA-E program directors. On the basis of this dialogue, the program director proposing the project refines the proposal, incorporates feedback, and seeks approval from ARPA-E’s director. If the director approves, a new project is created, and an FOA is released to the public, inviting proposals (Rohlfing, 2015).
Funding Opportunity Announcements
Throughout the federal research funding system, FOAs solicit research proposals through an open and transparent system. ARPA-E uses this system to request and receive research proposals, although it has implemented improvements to the system that have not been duplicated by other agencies. The FOA provides technical background and comprehensive information for potential applicants about the ARPA-E award process, performance expectations, funding limitations, and duration. ARPA-E FOAs fall into two broad categories:
- Focused—These FOAs target specific areas of energy technology identified by ARPA-E as having a significant potential impact on one
- OPEN—These FOAs are solicitations of proposals for potentially disruptive technologies “across the complete spectrum of energy applications” and are intended to “ensure that the agency does not
of its mission areas32 (ARPA-E, 2014a). These FOAs require that program directors and other ARPA-E staff delineate clear and well-defined project goals.
32 For example, a focused ARPA-E FOA released on December 8, 2014, invited proposals on the theme of micro-scaled optimized solar cell arrays with integrated concentration (MOSAIC). The technical target of this FOA is to foster a 50 percent improvement in solar-to-electrical power conversion over the performance of conventional “1-sun” flat plate photovoltaic modules (ARPA-E, 2015b).
miss opportunities to support innovative energy R&D that falls outside of the topics of the various focused technology programs that develop after focused solicitations have closed (ARPA-E, 2015a). OPEN FOAs were released in 2009, 2012, and 2015. They account for about one-third of ARPA-E’s funding support (Williams, 2015a).
Both focused and OPEN FOAs feature a number of “technical categories” and “technical subcategories” of solicited themes for research, together with specific performance targets with respect to those themes. In a 2014 focused FOA for solar arrays, for example, “Category 1: System Solutions for High-DNI [“direct normal incidence,” or high direct sunlight] Regions” comprised two subcategories with ARPA-E performance targets set as shown in Table 2-2. Submissions by applicants in response to FOAs must identify the technical subcategory or subcategories for proposed new technologies.
Entities eligible for ARPA-E awards include universities, national laboratories, large and small private companies, nonprofits, “project teams,” and consortia comprising such entities.33 Foreign entities may apply for funding, but all work must be performed by subsidiaries or affiliates incorporated in the United States. Application for ARPA-E funding follows a three-step process.
TABLE 2-2 Example of Focused Program Technical Targets from the MOSAIC Program Announced in 2014
|Subcategory 1A: High-DNI System with Macro-Tracking|
|1A.1||Solar energy harvesting efficiency||> 30% at module output|
|1A.2||Production cost||< $125/m2|
|1A.3||Array height||< 2.5 cm|
|1A.4||Projected system degradation||< 1%/year|
|Subcategory 1B: High-DNI System with Embedded Micro-Tracking|
|1B.1||Solar energy harvesting efficiency||> 30% at module output|
|1B.2||Production cost||< $150/m2|
|1B.3||Array height||< 2.5 cm|
|1B.4||Projected system degradation||< 1%/year|
SOURCE: ARPA-E, 2014b, p. 17.
33 42 U.S.C. § 16538(e)(3). “Project teams” refers to any entity with multiple collaborating players, including existing organizations and ad hoc teams.
Step 1. Concept Papers: Review Process
Applicants submit a five- to seven-page concept paper. External experts review an abstract of the proposal. This initial stage is intended to save applicants the time and cost of preparing applications with little likelihood of success. Concept papers were not widely used within DOE at the inception of ARPA-E. This step is widely viewed as a positive addition to the process and is critical in allowing ARPA-E program directors to shape projects. The reviewers assign each concept paper numerical scores according to two broad sets of criteria:
- Impact of the proposed technology (50 percent)—Reviewers consider (1) the extent to which the idea has potential for transformative and disruptive, rather than incremental, advancement over current technologies; (2) the extent to which the idea will positively impact at least one of the mission areas set forth by ARPA-E in the relevant FOA; and (3) the extent to which the applicant demonstrates awareness of competing commercial and emerging technologies and how the idea could result in significant improvements.
- Overall scientific and technical merit (50 percent)—Reviewers consider (1) the feasibility of the proposed work, (2) the extent to which the applicant proposes a “sound technical approach” superior to alternatives, (3) the extent to which outcomes and deliverables are clearly defined, (4) the extent to which the applicant identifies relevant techno-economic challenges, and (5) the demonstrated capabilities of the performers (ARPA-E, 2015a).
Following this review, the program director considers the evaluations and compiles a tentative list of concept papers to encourage to submit full applications based on a number of factors, including the numerical scores and comments received. Each FOA is accompanied by a Merit Review Plan, which is executed by an ARPA-E Merit Review Board. In carrying out the Merit Review Plan, a Merit Review Board, usually chaired by the program director who proposed the project and will oversee it, reviews and discusses the lists of papers with the program director, and the list is finalized through consensus. The Merit Review Board is not required to base its decisions solely on the numerical scores of the reviewers34 (Orr, 2015). The ARPA-E director reviews the Board’s recommendations and makes the final decision as to which applicants should be encouraged to submit a full application and which should be discouraged from doing so.
34 Under Secretary of Energy for Science and Energy Franklin Orr, noting that ARPA-E’s selection decisions are not bound by the conclusions of the external reviewers, comments that “normal peer review kills many promising ideas” (Orr, 2015).
Step 2. Full Applications: Selection Criteria and Process
Following review of concept papers, ARPA-E encourages some applicants to submit full applications providing information about prior, current, and pending sources of funding (both private and public) and an explanation of why other funding sources will not support the project. Full applications are reviewed by ARPA-E program directors and other personnel, including the director, along with leading external experts in the field. ARPA-E states that these comments are the most valuable component of the external review of an application. The reviewers must disclose any actual or apparent conflicts of interest. Reviewers include “world-class scientists, engineers, and leaders from the technical community” who convene in person for a review panel (ARPA-E, 2010a). Reviewers are asked to evaluate the applications and assign numerical scores, and provide comments according to four broad sets of criteria listed below, two of which—impact of the proposed technology and scientific/technical merit—are carried forward from the concept paper stage:
- Impact of the proposed technology relative to the current state of the art (30 percent)—Applicants are asked to demonstrate (1) the extent to which the idea is potentially transformative and disruptive (not incremental); (2) “profound understanding of the current state of the art” and an innovative approach to significantly improving it; (3) awareness of competing commercial and emerging technologies and how the idea would yield significant improvements over these technologies; and (4) a “reasonable and effective strategy” for transitioning the new technology from the laboratory to commercial deployment.
- Overall scientific and technical merit (30 percent)—Applicants must demonstrate (1) the extent to which the idea is “unique and innovative”; (2) clearly defined outcomes and deliverables; (3) the feasibility of the proposed work based on preliminary data, background information, and/or sound scientific and engineering practices and principles; (4) a sound technical approach, including “appropriately defined technical tasks”; and (5) identification of major technical R&D risks and feasible, effective risk mitigation strategies.
- Qualifications, experience, and capability (30 percent)—Applicants must demonstrate (1) the skill and expertise necessary to execute their project plan, based on prior experience with R&D of similar risk and complexity; and (2) access to the equipment and facilities necessary to execute the plan.
- Management plan (10 percent)—Applicants must demonstrate (1) a plausible plan for managing people and resources; (2) allocation of appropriate levels of effort and resources to tasks; (3) a reasonable
project schedule, including major milestones; and (4) the appropriateness of the budget relative to the task (ARPA-E, 2015a).
Once external evaluations have been completed, the Merit Review Board, usually chaired by the program director who initiated the relevant FOA, reviews the evaluations, generates “discuss” lists, and meets in person with the review panel to discuss applications. Applicants are allowed to read reviewers’ comments and submit rebuttal comments before the Merit Review Board makes funding recommendations. The Merit Review Board then considers the evaluations, any rebuttal comments from applicants, and the technical merit review criteria and program policy considerations, and makes its recommendation to the ARPA-E “selection official” (the ARPA-E director). The selection official decides which full applications will be selected to negotiate terms of an award. Again, there is no obligation to use the scoring by the reviewers as the sole factor in determining which projects are selected, or to rank applications according their reviewer scores.
In a 2012 audit, ARPA-E program directors told the Government Accountability Office (GAO) that, in addition to the above four criteria, they were allowed to consider other factors, one of which was funding of a broad range of potentially transformational technological innovations with varying levels of risk to address a given technological challenge. This approach was seen as increasing “the likelihood that at least one of them would work.” The program directors also indicted that during the selection phase, they sometimes negotiated with applicants to adjust the proposed budget, the project scope and timeframe, and milestones to be achieved (GAO, 2012). This approach also demonstrates how ARPA-E presently manages the inherent trade-off between “bias” and “information” during the selection process and how it tries to aggregate information from its reviewer cohorts to maximize value during project selection (Boudreau et al., 2016; Hoffman et al., 2015; Li and Agha, 2015).
Step 3. Reply to Reviewer Comments
Applicants are given an opportunity to reply to reviewer comments (ARPA-E, 2010a). In contrast to DARPA, which does not share reviews with applicants, ARPA-E sends reviews to applicants prior to a final determination and provides the opportunity to prepare a reply to reviewer comments. Typically the agency limits the length of replies, often to four pages or less (Rohlfing, 2015).
Project Selection and Implementation
ARPA-E supports two categories of projects:
- Proof-of-concept—These projects are intended to provide the first preliminary data to prove or disprove new technology concepts. They are limited in cost (under $1 million) and duration (6–18 months).
- Technology development—These projects are intended to develop technology from ideas to laboratory-scale prototypes operating at agreed-upon specifications. Their duration is typically 36 months and their cost $2–10 million.
Both are solicited, selected, and implemented following the same process just described.
Most of ARPA-E’s projects are based on focused, thesis-driven programs centered on a theory of change. However, the agency’s initial personnel decided that focused programs should be augmented by the occasional open solicitations discussed earlier, intended to cast a broad net and encourage novel and creative ideas that staff had not yet conceived of and that had the potential to inspire whole new programs (Danielson, 2015).
Negotiations with Selected Applicants
After ARPA-E selects applicants for an award, the agency engages in negotiations with the potential recipients for a final award. This process is expected to take about 90 days. Program directors work with recipients to establish a work plan with technological milestones that include specific go/no go decision points over the course of the project. ARPA-E reviews awardees’ budget documents, and may require additional information and documentation. The parties must reach agreement on how patents and other intellectual property rights will be handled; potential conflicts of interest identified by awardees are reviewed; and ARPA-E evaluates the project’s environmental impact.
Most ARPA-E awards are made using cooperative agreements between the agency and successful applicants. The agency also has statutory authority to provide awards in the form of cash prizes, grants, contracts, and “other transactions.”35 In cooperative agreements, “substantial involvement is expected between the executive agency and the…recipient when carrying out the activity contemplated in the agreement.”36 Cooperative agreements serve as the legal
35 42 U.S.C. § 16538(f).
36 31 U.S.C. § 6305.
foundation for ARPA-E’s active and continuous engagement in a project, and provide that
- ARPA-E may intervene in the project at any time;
- ARPA-E’s role is not limited to administration of the award but includes substantial involvement in the direction and redirection of the technical aspects of the project as a whole; and
- ARPA-E may terminate or redirect projects that fail to achieve predetermined go/no go decision points or technical milestones/deliverables.37
Prime recipients of an award are required to provide at least 20 percent of the total project cost.38 The prime recipient’s share may consist of cash or in-kind contributions from the prime recipient or subrecipients39 (ARPA-E, 2015a). Large businesses are “strongly encouraged” to provide more than 20 percent of the total project cost; conversely, ARPA-E has reduced minimum cost-share requirements for small businesses (10 percent with a 1-year grace period) and nonprofits and universities (5 percent) (ARPA-E, 2015a). Under the America COMPETES Act, program directors are responsible for “identifying innovative cost-sharing arrangements for ARPA-E projects.”40 However, the agency’s awards remain governed by general federal cost-sharing rules and requirements, which limit innovative arrangements (ARPA-E, 2015a).
ARPA-E works with awardees to develop acceptable project milestones by starting from the hoped-for deliverable and working backward along the development path to establish critical binary go/no go milestones, typically beginning early in the second year of the project. Awardees tend to propose milestones they believe they can achieve, whereas the agency encourages them to be aggressive and push the milestones “to the edge, but not over.” Milestones are expected to be specific, unambiguous, and quantifiable. They may be revisited and renegotiated “if there is still a clear path to the final objective” (Rohlfing, 2015). As a result of go/no go reviews, the terms of cooperative agreements provide that ARPA-E may
37 ARPA-E Model Cooperative Agreement, Special Terms and Conditions, Clause 7.a.
38 42 USC §16352 (2017).
39 In-kind contributions may include personnel costs; facilities; administrative costs; indirect costs; the rental value of buildings and equipment; and the value of a service, other resource, or third-party in-kind contribution. 2 CFR 910.130 (2017).
40 42 U.S.C. § 16538(g)(2)(B)(v) (2017).
- authorize continuation of the project,
- redirect the project’s work,
- place the project on hold pending further supporting data,
- suspend or terminate the project per the agreement, or
- take “other appropriate actions” (ARPA-E, n.d.-b, p. 6).
Before receiving award funds, awardees are required to develop a T2M plan in close coordination with ARPA-E’s T2M advisors. Commercialization strategies developed to meet this requirement include training and the development of the business information necessary to understand market needs and tailor technology development to address those needs. ARPA-E also helps awardees develop relationships with relevant government agencies, technology transfer offices, companies, investors, and other organizations to facilitate transition to the commercial phase (ARPA-E, 2016g).
Project funding is set aside in its entirety at the beginning of a project to guarantee availability of the funds required to continue funding projects that are successfully meeting their milestones regardless of future appropriation levels.41 The average award size is $2.7 million. Award recipients invoice ARPA-E periodically for costs incurred. Prior to payment, the agency must determine that each invoiced cost is “reasonable, allowable and allocable.”42 ARPA-E’s goal is to pay invoices within 30 days.
One important point of differentiation from other DOE offices has to do with funding. ARPA-E’s goal for successful projects is for them to attract funding from other sources to continue the work and carry the technology farther along the development path. If the idea shows merit during its time as an ARPA-E project, the agency may try to extend additional funding while the project team seeks funding from other sources, but ultimately, other funders must step in to sustain the work. By contrast, other DOE offices often have
41 “One of the great things about ARPA-E and the way we’re structured is that we fully fund our projects right up front. So when we say we’ve awarded $900,000 to a specific project, that money is already set aside from our current year’s budget. So even if that project takes 2–3 years to complete, those scientists don’t have to worry about their budget being cut mid-project. That makes a huge difference. Our researchers can stay focused on the project at hand and hopefully knock down these technical challenges to get these technologies moving and change our energy future.” Personal consultation with ARPA-E Deputy Director Cheryl Martin, The Energy Collective (August 8, 2013).
42 2 C.F.R. 200, Uniform Administrative Requirements, Cost Principles and Audit Requirements for Federal Awards.
technology roadmaps and seek to sustain particular technologies much farther through the stages of development.
Monitoring and Continuous Engagement
ARPA-E remains closely engaged with awardees from the inception of a project through its conclusion. An initial program kickoff meeting is held to convene all of the performers from a given focused technology program, provide thumbnail overviews of all of the projects in the program, and begin providing awardees with information on such subjects as technical challenges and scientific ideas by means of tutorials and invited talks. Program directors subsequently engage in “active project management,” which includes review of quarterly performer reports, regular site visits, meetings, conference calls, and written feedback on results and reported quarterly progress. The subjects of these contacts can include general reviews of progress relative to milestones, specific responses to a new challenge, and identification of anticipated problems and potential solutions. Site visits enable visual confirmation of progress toward milestones and frank exchanges about nonperformance. During each subsequent year of a focused technology program, ARPA-E holds an annual meeting convening all performers of projects within the program, who provide updates on progress and discoveries to the entire group of performers.
When ARPA-E finds that a project is not meeting its milestones, it acts to address the situation. The first line of defense is the frequent contact between the ARPA-E program director and her or his staff and the performer team, which helps avoid either party being surprised to learn of the difficulty meeting milestones. In more challenging cases, contact by telephone or meetings may increase to weekly frequency. Actions taken include complete termination of the project, limited extensions of time to allow projects to meet a given milestone, and revision of milestones and objectives. ARPA-E indicated that as of May 2015, the agency had terminated a total of 21 projects prior to the end of their cooperative agreements’ end dates because of failure to achieve stipulated milestones (Williams, 2015a).
ARPA-E Energy Innovation Summit
ARPA-E hosts annual Energy Innovation Summits to convene national leaders in energy from industry, academia, and government. The agency requires that all performers with active awards attend, and provides each a booth to showcase their research. One major goal of the summit is to connect project performer teams with potential sources of future support for their research and eventual technology development. Thus, ARPA-E also works to ensure that potential private and public follow-on funding sources—such as venture firms, banks, established technology development firms, other offices of DOE, and other federal funding sources—attend in addition to the performers. A summit
typically draws about 2,000 attendees (Reitenbach, 2015). The summits “have become major technology showcase events in Washington, attracting large attendance and featuring prominent business, executive branch and bipartisan Congressional leaders in speaking roles” (Bonvillian and van Atta, 2011). Dr. Cheryl Martin, former deputy director for commercialization at ARPA-E, offered a perspective on the 2015 summit in an interview:
…When ARPA-E started not even six years ago…it was a startup agency, and the first summit was formed to basically talk about what the agency could be and to show that it was supported very uniformly, you know, across Congress and business and academics, but the first conference night we didn’t have any projects. We had potential. And as we’ve evolved as an agency, we have $1.1 billion, 400 projects that we’ve funded, and down on the floor today, we have 250 booths of technologies where, some of them are actually real products now. And so you see that movement from idea and growth and new certainly not…maturity, but it’s an evolution that’s real, and I think that the buzz at the summit now reflects that (E&ETV, 2015).
The summits feature a “technology showcase” where breakthrough energy developments from ARPA-E performers are featured. The showcase also features other promising innovations in the energy field, regardless of whether they have been supported by ARPA-E,43 highlighting applicants that did not receive awards but that, in the agency’s view, deserve attention in the energy research community.
Summary of ARPA-E Programs
As of October 2016, ARPA-E had invested over $1 billion in more than 500 projects. It had launched 38 focused programs and conducted three open funding solicitations (in 2009, 2012, and 2015). Table 2-3 provides the names of all these programs, along with the number of projects receiving funding under each and the total amount of funding made available for each.
43 Statement of ARPA-E Director Arun Majumdar, U.S. Congress, House Subcommittee on Energy and Water Development and Related Agencies, Appropriations Committee, Budget Hearing, March 28, 2012.
TABLE 2-3 ARPA-E Programs as of October 2016
|Program Name||Number of Projects||Authorized Funding Amount ($ Million)|
Batteries for Electrical Energy Storage in Transportation (BEEST)
Innovative Materials and Processes for Advanced Carbon Capture Technologies (IMPACCT)
Agile Delivery of Electrical Power Technology (ADEPT)
Building Energy Efficiency Through Innovative Thermodevices (BEETIT)
Grid-Scale Rampable Intermittent Dispatchable Storage (GRIDS)
Plants Engineered to Replace Oil (PETRO)
High Energy Advanced Thermal Storage (HEATS)
Rare Earth Alternatives in Critical Technologies (REACT)
Green Electricity Network Integration (GENI)
Solar Agile Delivery of Electrical Power Technology (Solar ADEPT)
Methane Opportunities for Vehicular Energy (MOVE)
Advanced Management and Protection of Energy Storage Devices (AMPED)
Small Business Innovation Research/Small Business Technology Transfer (SBIR/STTR)
Innovative Development in Energy-related Applied Science (IDEAS)
Robust Affordable Next Generation Energy Storage Systems (RANGE)
Reducing Emissions using Methanotrophic Organisms for Transportation Energy (REMOTE)
Modern Electro/Thermochemical Advancements for Light-metal Systems (METALS)
Full-Spectrum Optimized Conversion and Utilization of Sunlight (FOCUS)
Strategies for Wide Bandgap, Inexpensive Transistors for Controlling High Efficiency Systems (SWITCHES) & SBIR/STTR
Reliable Electricity Based on Electrochemical Systems (REBELS)
Delivering Efficient Local Thermal Amenities (DELTA)
Methane Observation Networks with Innovative Technology to Obtain Reductions (MONITOR)
|Program Name||Number of Projects||Authorized Funding Amount ($ Million)|
Accelerating Low-cost Plasma Heating and Assembly (ALPHA)
Cycling Hardware to Analyze and Ready Grid-Scale Electricity Storage (CHARGES)
Advanced Research In Dry cooling (ARID)
Transportation Energy Resources from Renewable Agriculture (TERRA)
Generators for Small Electrical and Thermal Systems (GENSETS)
Traveler Response Architecture using Novel Signaling for Network Efficiency in Transportation (TRANSNET)
Micro-scale Optimized Solar-cell Arrays with Integrated Concentration (MOSAIC)
Network Optimized Distributed Energy Systems (NODES)
Generating Realistic Information for the Development of Distribution and Transmission Algorithms (GRID DATA)
Single-Pane Highly Insulating Efficient Lucid Designs (SHIELD)
Integration and Optimization of Novel Ion-Conducting Solids (IONICS)
ENergy-efficient Light-wave Integrated Technology Enabling Networks that Enhance Datacenters (ENLITENED)
Renewable Energy to Fuels through Utilization of Energy-dense Liquids (REFUEL)
NEXT-Generation Energy Technologies for Connected and Automated on-Road vehicles (NEXTCAR)
Rhizosphere Observations Optimizing Terrestrial Sequestration (ROOTS)
NOTE: Funding amount for the CHARGES program redacted due to the small number of awardees.
SOURCE: DOE, 2015.