About

John F. Hartwig is the Henry Rapoport Chair in Organic Chemistry and a Professor of Chemistry at the University of California, Berkeley. Born in 1964, he holds a B.A. from Princeton University and a Ph.D. in Chemistry from UC Berkeley. His postdoctoral work was conducted as an American Cancer Society Postdoctoral Associate at the Massachusetts Institute of Technology. Hartwig has held faculty positions at Yale University, where he served as Assistant, Associate, and then Irénée DuPont Professor of Chemistry, and at the University of Illinois, Urbana-Champaign, as Kenneth L. Rinehart Jr. Professor. Since 2011, he has been a faculty member at UC Berkeley, where he also holds the title of The Dow Chair in Sustainable Chemistry. His research focuses on the discovery and understanding of new reactions of organic compounds catalyzed by transition metal complexes and artificial metalloenzymes. His group investigates small-molecule catalysts and artificial metalloenzymes for selective reactions of organic molecules, including catalytic functionalization of alkanes and arenes, cross-couplings, fluorination, addition to alkenes, and hydrocarbyl functionalization. His work combines organic synthesis, organometallic synthesis, protein design, and mechanistic analysis, leading to the discovery of new classes of organometallic reactions. Hartwig has authored a leading textbook in organometallic chemistry titled 'Organotransition Metal Chemistry: From Bonding to Catalysis' and has received numerous awards, including the Wolf Prize in Chemistry, the John Gamble Kirkwood Award, and election as a Fellow of the Royal Society of Chemistry and the American Academy of Arts & Sciences.

Research topics

• Chemistry
• Organic chemistry
• Materials science
• Combinatorial chemistry
• Stereochemistry
• Polymer chemistry

Selected publications

• Landscaper: Understanding Loss Landscapes Through Multi-Dimensional Topological Analysis

  Open MIND · 2026-02-06

  preprint

  Loss landscapes are a powerful tool for understanding neural network optimization and generalization, yet traditional low-dimensional analyses often miss complex topological features. We present Landscaper, an open-source Python package for arbitrary-dimensional loss landscape analysis. Landscaper combines Hessian-based subspace construction with topological data analysis to reveal geometric structures such as basin hierarchy and connectivity. A key component is the Saddle-Minimum Average Distance (SMAD) for quantifying landscape smoothness. We demonstrate Landscaper's effectiveness across various architectures and tasks, including those involving pre-trained language models, showing that SMAD captures training transitions, such as landscape simplification, that conventional metrics miss. We also illustrate Landscaper's performance in challenging chemical property prediction tasks, where SMAD can serve as a metric for out-of-distribution generalization, offering valuable insights for model diagnostics and architecture design in data-scarce scientific machine learning scenarios.

  DOI

• Mapping the Undirected Borylation of C(sp ³ )–H Bonds in Strained Rings

  Journal of the American Chemical Society · 2026-01-02 · 2 citations

  articleOpen accessSenior authorCorresponding

  Aliphatic small saturated carbocycles and azacycles are increasingly used as bioisosteres and structural cores in medicinally active compounds due to the beneficial pharmacological and physicochemical properties they can impart. Therefore, a need exists to modify these motifs and to install groups that enable their incorporation into organic structures; these goals can be accomplished by introducing functional groups at the position of the C–H bonds on the rings. However, functionalization of secondary C–H bonds in strained rings, such as cyclopropanes and cyclobutanes, confronts several challenges, including the greater strength of these bonds than those in unstrained rings. Although catalytic, undirected borylation has been reported to functionalize the C–H bonds of selected strained rings, the examples of such reactions in earlier studies are limited in scope, principally involving rings with a small number and size of substituents. We report the borylation of fused, spirocyclic, and polysubstituted cyclopropanes, cyclobutanes, azetidines, and β-lactams with high diastereoselectivity with the most recent generation of catalysts for the borylation of alkyl C–H bonds. Important for the formation of more complex structures, reactions occur, unlike reactions of larger rings, with steric hindrance on adjacent carbon atoms. The stoichiometry of the diboron reagent, the s-character of the C–H bond, and the disposition of substituents influence reactivity and product distribution in ways charted by our results. With these advances, the borylation of C–H bonds becomes a viable approach for the modification and incorporation of these ring systems into pharmaceutically active structures.

  PublisherOA PDFDOI

• Landscaper: Understanding Loss Landscapes Through Multi-Dimensional Topological Analysis

  ArXiv.org · 2026-02-06

  articleOpen access

  Loss landscapes are a powerful tool for understanding neural network optimization and generalization, yet traditional low-dimensional analyses often miss complex topological features. We present Landscaper, an open-source Python package for arbitrary-dimensional loss landscape analysis. Landscaper combines Hessian-based subspace construction with topological data analysis to reveal geometric structures such as basin hierarchy and connectivity. A key component is the Saddle-Minimum Average Distance (SMAD) for quantifying landscape smoothness. We demonstrate Landscaper's effectiveness across various architectures and tasks, including those involving pre-trained language models, showing that SMAD captures training transitions, such as landscape simplification, that conventional metrics miss. We also illustrate Landscaper's performance in challenging chemical property prediction tasks, where SMAD can serve as a metric for out-of-distribution generalization, offering valuable insights for model diagnostics and architecture design in data-scarce scientific machine learning scenarios.

  PublisherOA PDF

• Microstructure of amide-functionalized polyethylenes determined by NMR relaxometry

  Chemical Science · 2026-01-01

  articleOpen accessCorresponding

  Amidation of polyethylenes creates a range of amide-containing materials with enhanced properties, but the effect of these functional groups on the microstructure of these new materials is not known. Here we employ solid-state nuclear magnetic resonance (NMR) techniques to analyze the microstructure of amide-modified polyethylenes. While a decrease in crystallinity was observed with increasing amounts of functionalization, we found by measuring the chain mobility of the crystalline, amorphous, and interphasial regions of the polyethylenes with NMR relaxation techniques that the grafted amidyl groups partition into the rigid amorphous fraction (RAF) between the crystalline and amorphous regions. The chemical specificity of these NMR experiments creates precise assessments of the location of functional groups within the materials. Together, these insights into the microstructure and morphology of amide-containing polyethylenes lay a foundation for a deeper understanding of the structure and properties of functional polyolefins.

  PublisherOA PDFDOI

• A 3D, Structure-Based, Deep Learning Approach for Predicting the Regioselectivity of Transition-Metal Catalysis

  ChemRxiv · 2026-01-28

  articleOpen accessSenior author

  Rational development of transition-metal catalysts, even when guided by theory and mechanistic knowledge, involves significant trial and error. Although ML offers the potential to accelerate catalyst discovery and optimization, accurately modeling the complex structures of catalysts and the multistep mechanisms by which they react remains challenging with the limited sets of data available. Olefin hydroformylation is a quintessential example of this challenge: its catalytic cycle involves many, often reversible, steps, and decades of study have not yielded reliable structure-selectivity relationships. We report Libra-ML, a 3D structure-based deep learning approach for predicting experimental outcomes of transition-metal catalyzed reactions. To demonstrate the ability of Libra-ML to model the outcomes of complex catalytic reactions, we predicted the regioselectivity of hydroformylation with terminal olefins catalyzed by rhodium complexes. Comparisons to existing methods demonstrate the state-of-the-art performance of Libra-ML and illustrate the importance of capturing 3D structure to predict experimental outcomes with molecular catalysts.

  PublisherDOI

• Unleashing the Power of Potassium 2-Ethylhexanoate as a Mild and Soluble Base for Pd-Catalyzed C–N Cross-Coupling

  Journal of the American Chemical Society · 2025-12-02 · 1 citations

  articleSenior authorCorresponding

  The formation of C–N bonds by Pd-catalyzed cross-coupling is one of the most widely practiced reactions in chemical synthesis. Typical reaction conditions involve either a strong base, which limits the scope of substrates, or an insoluble, inorganic base, which complicates running reactions on a large scale. Reaction conditions for C–N couplings with a base that is both mild and soluble are needed. We report the discovery of a combination of a phosphorinane ligand (L147) and a soluble carboxylate base, potassium 2-ethylhexanoate (K-2-EH), which leads to the coupling of a wide range of base-sensitive coupling partners. To explore the enhanced substrate scope of the reaction with this base and catalyst, we evaluated the scope using representative reactants selected from published partners, using chemical descriptors and clustering to ensure their chemical diversity. These results show that the combination of this phosphorinane ligand and K-2-EH can couple primary aliphatic amines, amides, sulfonamides, and heteroaromatic nucleophiles as well as acidic secondary nitrogen nucleophiles, such as arylamines, heteroarylamines, and amides, with a range of electrophiles. A side-by-side comparison to form selected coupling products in the presence of a range of previously reported bases and ligands showed that the products that decomposed under standard reaction conditions were stable with K-2-EH as a base. Finally, models of quantitative structure–reactivity relationships, trained on ligand screening data, were developed to help reveal the structural features that engender reactivity.

  PublisherDOI

• Backbone editing and deconstruction of polyethylene by Beckmann rearrangement and hydrogenolysis

  Chemical Science · 2025-01-01 · 16 citations

  articleOpen accessSenior authorCorresponding

  Polyethylene is the most widely produced commodity plastic and is used in many applications, including packaging, insulation, and medical devices. However, the inertness of polyethylene makes chemical recycling inefficient and challenging. We report the conversion of oxidized high-density and low-density polyethylene, formed by direct, catalytic oxidation, to polyamides by Beckmann rearrangement of the corresponding oximes. These polyamides have enhanced surface properties over those of unmodified polyethylene, while maintaining the same, favorable mechanical profiles. The amide sites were reductively cleaved by hydrogenolysis with ruthenium-based catalysts to furnish alcohol- and amine-terminated fragments, which were used for the synthesis of polyurea-urethanes with poly(tetrahydrofuran) and methylene diphenyl diisocyanate. These experiments show how to install cleavable moieties into the backbone of polyethylene to facilitate deconstruction and the generation of new materials to affect greater sustainability in polyolefins.

  PublisherOA PDFDOI

• Electronic Activation Enables the Borylation of Alkyl C–H Bonds in Saturated Nitrogen Heterocycles

  Journal of the American Chemical Society · 2025-11-11 · 5 citations

  articleSenior authorCorresponding

  Catalytic functionalization of C-H bonds is a valuable strategy for the synthesis and diversification of organic compounds. The catalytic borylation of heteroaromatic C-H bonds is well established, but the analogous borylation of alkyl C-H bonds in saturated heterocycles remains underdeveloped. Reactions with improved catalysts recently reported for the borylation of alkyl C-H bonds occurred with a narrow range of such nitrogen heterocycles. Here, we describe a broadly applicable approach to activate C-H bonds in saturated nitrogen heterocycles for iridium-catalyzed borylation by installing a group on nitrogen that increases the reactivity of specific C-H bonds. Mechanistic studies reveal that the electron withdrawing groups accelerate the rate of borylation by enhancing the energetics of multiple steps within the catalytic cycle.

  PublisherDOI

• Circularity in polydiketoenamine thermoplastics via control over reactive chain conformation

  Science Advances · 2025-01-22 · 4 citations

  articleOpen access

  Controlling the reactivity of bonds along polymer chains enables both functionalization and deconstruction with relevance to chemical recycling and circularity. Because the substrate is a macromolecule, however, understanding the effects of chain conformation on the reactivity of polymer bonds emerges as important yet underexplored. Here, we show how oxy-functionalization of chemically recyclable condensation polymers affects acidolysis to monomers through control over distortion and interaction energies in the rate-limiting transition states. Oxy-functionalization of polydiketoenamines at specific sites on either the amine or triketone monomer segments increased acidolysis rates by more than three orders of magnitude, opening the door to efficient deconstruction of linear chain architectures. These insights substantially broaden the scope of applications for polydiketoenamines in a circular manufacturing economy, including chemically recyclable adhesives for a diverse range of surfaces.

  PublisherDOI

• Efficient Aminations of Aryl Halides by a Cu(II) Catalyst

  Journal of the American Chemical Society · 2025-06-09 · 9 citations

  articleSenior authorCorresponding

  Aminations of aryl halides catalyzed by copper complexes with ancillary ligands have become valuable for the formation of anilines, and the mechanisms for these reactions have been shown to occur by a Cu(I)/(III) cycle. We show that the coupling of aryl and heteroaryl bromides with a range of nitrogen nucleophiles, including hydrazine hydrate and complex amines, occurs with copper and a simple dianionic, dimethylpyrrole-based oxalohydrazido ligand by a cycle in which Cu(II) complexes of this ligand are the resting state and an active, low-valent, catalytic intermediate. These couplings involving Cu(II) occur in many cases with just 0.1-0.2 mol % catalyst and take place under air, due to the absence of Cu(I) that is less stable to disproportionation and to air. Kinetic profiles and EPR spectroscopy of reactions initiated with Cu(I) and Cu(II) precursors provide strong evidence that both systems react through an active Cu(II) complex, thereby indicating that a recently uncovered mechanistic manifold for copper-catalyzed couplings of phenoxides with oxalamide ligands is also followed for the coupling of nitrogen nucleophiles catalyzed by copper complexes of oxalohydrazido ligands.

  PublisherDOI

Recent grants

• Catalytic Enantioselective Allyic Amination and Etherification

  NSF · $202k · 2006–2008

• NIH Grant R01GM058108

  NIH · $5.5M · 2016

• Catalytic Functionalization of C-H Bonds with Main Group Reagents

  NIH · $979k · 2015–2020

• Acquisition of a Cryoprobe and Related Accessories for a 600 MHz NMR Spectrometer to Support Health Research Activities at the University of California, Berkeley

  NIH · $265k · 2018–2020

• Catalytic Regioselective Functionalization of Alkane and Arenes

  NSF · $516k · 2009–2012

Frequent coauthors

• Douglas S. Clark

  Lawrence Berkeley National Laboratory

  63 shared

• Christopher D. Incarvito

  53 shared

• James P. Stambuli

  AbbVie (United States)

  43 shared

• Charles Edwin Webster

  Mississippi State University

  40 shared

• Jay D. Keasling

  Joint BioEnergy Institute

  39 shared

• Jaclyn M. Murphy

  39 shared

• Marko Hapke

  Johannes Kepler University of Linz

  35 shared

• Christo S. Sevov

  34 shared

Labs

• Hartwig GroupPI

  Mechanistically driven discovery of catalytic reactions

Awards & honors

• Arthur C. Cope Award (2021)
• Clarivate Citation Laureate (2020)
• John Gamble Kirkwood Award (2020)
• Wolf Prize in Chemistry (2019)
• Tetrahedron Prize for Creativity in Organic Chemistry (2018)