Joel A. Haber

Joel A. Haber

Auto-generated profile, based on publicly available information

Joel A. Haber

Division of Engineering and Applied Science, California Institute of Technology, Pasadena, CA 91125, USA

Research Focus & Metrics

Academic Career Length

21Years (2006-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
24 May 2021 in Nature Catalysis
Verified
Journal Article
Closed access
PDF file
HLHao Li
Hao Li
SKSara Kelly
Sara Kelly
DGDan Guevarra
Dan Guevarra
ZWZhenbin Wang
Zhenbin Wang
YWYu Wang
Yu Wang
JNJens K. Nørskov
Jens K. Nørskov

13 January 2017 in ACS Catalysis
Verified
Journal Article
Closed access
Marco FavaroMarco Favaro
Marco Favaro
WDWalter S. Drisdell
Walter S. Drisdell
MMMatthew A. Marcus
Matthew A. Marcus
JGJohn M. Gregoire
John M. Gregoire
ECEthan Crumlin
Ethan Crumlin
JYJunko Yano
Junko Yano

29 January 2020 in Chemical Science
Verified
Journal Article
Open access
PDF file
BRBrian Rohr
Brian Rohr
HSHelge S. Stein
Helge S. Stein
DGDan Guevarra
Dan Guevarra
YWYu Wang
Yu Wang
JHJoel A. Haber
Joel A. Haber
JGJohn M. Gregoire
John M. Gregoire
Sequential learning (SL) strategies, i.e. iteratively updating a machine learning model to guide experiments, have been proposed to significantly accelerate materials discovery and research. Applications on computational datasets and a handful of optimization experiments have demonstrated the promise of SL, motivating a quantitative evaluation of its ability to accelerate materials discovery, specifically in the case of physical experiments. The benchmarking effort in the present work quantifies the performance of SL algorithms with respect to a breadth of research goals: discovery of any “good” material, discovery of all “good” materials, and discovery of a model that accurately predicts the performance of new materials. To benchmark the effectiveness of different machine learning models against these goals, we use datasets in which the performance of all materials in the search space is known from high-throughput synthesis and electrochemistry experiments. Each dataset contains all pseudo-quaternary metal oxide combinations from a set of six elements (chemical space), the performance metric chosen is the electrocatalytic activity (overpotential) for the oxygen evolution reaction (OER). A diverse set of SL schemes is tested on four chemical spaces, each containing 2121 catalysts. The presented work suggests that research can be accelerated by up to a factor of 20 compared to random acquisition in specific scenarios. The results also show that certain choices of SL models are ill-suited for a given research goal resulting in substantial deceleration compared to random acquisition methods. The results provide quantitative guidance on how to tune an SL strategy for a given research goal and demonstrate the need for a new generation of materials-aware SL algorithms to further accelerate materials discovery.

16 October 2018 in ACS Catalysis
Verified
Journal Article
Closed access
LZLan Zhou
Lan Zhou
ASAniketa Shinde
Aniketa Shinde
JMJoseph H. Montoya
Joseph H. Montoya
ASArunima Singh
Arunima Singh
SGSheraz Gul
Sheraz Gul
JGJohn M. Gregoire
John M. Gregoire

9 March 2023 in Nature Synthesis
Verified
Journal Article
Closed access
PDF file
JGJohn M. Gregoire
John M. Gregoire
LZLan Zhou
Lan Zhou
JHJoel A. Haber
Joel A. Haber

9 December 2015 in Energy & Environmental Science
Verified
Journal Article
Closed access
PDF file
DGD. Guevarra
D. Guevarra
ASA. Shinde
A. Shinde
SSS. K. Suram
S. K. Suram
ISI. D. Sharp
I. D. Sharp
FTF. M. Toma
F. M. Toma
JGJ. M. Gregoire
J. M. Gregoire
The development of an efficient photoanode remains the primary materials challenge in the establishment of a scalable technology for solar water splitting. The typical photoanode architecture consists of a semiconductor light absorber coated with a metal oxide that serves a combination of functions, including corrosion protection, electrocatalysis, light trapping, hole transport, and elimination of deleterious recombination sites. To date, such coatings have been mostly limited to simple materials such as TiO2 and Co-Pi, with extensive experimental and theoretical effort required to provide an understanding of the physics and chemistry of the semiconductor-coating interface. To provide a more efficient exploration of metal oxide coatings for a given light absorber, we introduce a high throughput methodology wherein a uniform BiVO4 thin film is coated with 858 unique metal oxides covering a range of metal oxide loadings and the full Ni–La–Co–Ce oxide quaternary composition space. Photoelectrochemical characterization of each photoanode reveals that approximately one third of the coatings lower the photoanode performance while select combinations of metal oxide composition and loading provide up to a 14-fold increase in the maximum photoelectrochemical power generation for oxygen evolution in pH 13 electrolyte. Particular Ce-rich coatings also exhibit an anti-reflection effect that further amplifies the performance, yielding a 20-fold enhancement in power conversion efficiency compared to bare BiVO4. By use of in situ optical spectroscopy and comparisons between the metal oxide coatings and their extrinsic optical and electrocatalytic properties, we present a suite of data-driven discoveries, including composition regions which form optimal interfaces with BiVO4 and photoanodes that are suitable for integration with a photocathode due to their excellent power conversion and solar transmission efficiencies. The high throughput experimentation and informatics provides a powerful platform for both identifying the pertinent interfaces for further study and discovering high performance photoanodes for incorporation into efficient water splitting devices.