Cities of chemistry

Travel

September 1, 2026

For any career, a network is everything. For chemists, that network is highly specialized. You need laboratory space. You need coworkers and mentors with certain skill sets. And you need institutions. Depending on how you’d like your career to unfold, those institutions can include academic centers, big industrial companies, start-up accelerators and incubators, and venture capital firms.

These institutions are all around the US and all around the globe. As it turns out, in an increasingly digital world, geography still matters.

Different cities and regions play host to different kinds of chemistry ecosystems. Houston, long the stronghold of the petrochemical industry, is now establishing itself as the geographic cornerstone of the US energy transition. Boston may not house the headquarters of top US pharmaceutical companies, but it commands the most chemistry-related venture capital investment and the third-highest chemistry article output, reflecting its status as a life sciences hub spanning start-ups, academia, and research hospitals.

C&EN has evaluated the features that make a city, state, or country ripe for building a career in chemistry. Using a mix of public and private data, we’ve created interactive tools so you can see what regions are dominant as research centers, chemical industry hubs, and start-up havens—ultimately, whatever matters most to you.

There’s no right way to be a chemist. But some places are right for chemistry.

Research centers

Research output by metro area and biomedical funding by state

For research chemists—whether they are in academia, government, or industry—California, New York, and Massachusetts emerge as hotbeds of activity in the US. These are the states where scientists are churning out the highest numbers of chemistry-related papers in natural and health science journals as measured by the Nature Index.

When it comes to US National Institutes of Health (NIH) funding, which encompasses chemistry and other scientific projects, California and the Northeast commanded the highest dollar amounts in fiscal 2024, likely owing to their high concentrations of universities and research hospitals. California is also home to several of the nation’s best university chemistry programs.

According to US News and World Report, the top doctoral programs for chemistry in the US can be found in California, the Northeast, and Illinois. US News ranks doctoral programs using qualitative surveys from officials in each program. These academic centers also serve as engines for local chemistry ecosystems through spin-off businesses and research partnerships.

Case study: How UCSF chemist Michelle Arkin built her career in the San Francisco Bay Area

When Michelle Arkin finished her PhD in chemistry at the California Institute of Technology just outside Los Angeles, she wanted to focus on protein science. “Who runs a really good lab?” she asked colleagues.

The resounding answer was Jim Wells. So in 1997, Arkin packed up and headed north to the San Francisco Bay Area to join Wells at Genentech, where she became a postdoctoral fellow. In 1998, when Wells left to help found Sunesis Pharmaceuticals, Arkin went with him, kicking off a career that would solidify her as a chemistry force in Northern California.

“Being in a start-up company was really exciting,” Arkin says of her time at Sunesis. “I really enjoyed working with my colleagues and the rapid learning that happens at a good start-up.”

After 9 years, Arkin decided to make her way back to early-stage discovery—and to chemistry. She’d been working on biology teams and wanted to recapture her sense of self as a chemist. Arkin didn’t have to go far. Just a few miles north of Sunesis’s headquarters in South San Francisco, the University of California, San Francisco, was looking for someone to help start the Small Molecule Discovery Center. As luck would have it, Wells was the head.

Arkin joined as director in 2007, overseeing a team working on small-molecule screening tools, fragment discovery, and drug discovery technologies. She also became a professor at UCSF.

Eventually, Arkin was able to bridge the worlds of academia and industry. In the past 6 years, she has cofounded four start-ups: Elgia Therapeutics, which is developing drug candidates for inflammatory and fibrotic diseases; Ambagon Therapeutics, which is making molecular glues to combat neurodegeneration and cancer; ResNovas Therapeutics, a firm working on targeted protein degraders; and BNM Oncology, which aims to treat a rare cancer. Some of them were started with the help of UCSF’s tech transfer office, which Arkin describes as “very sophisticated.”

“In the Bay Area, if you have a lot of new technology crashing against traditional methodologies, maybe something really new and exciting can develop,” Arkin says.

Arkin herself will not be there much longer. She has accepted a role at the University of Toronto Mississauga as the Canada Impact+ research chair in chemical biology—part of a Canadian government program to attract more scientists—and will work in the university’s Centre for Medicinal Chemistry starting next year.

As for the Bay Area, Arkin knows the tides of investment—and which technical areas are of interest—can ebb and flow, especially given the region’s status as a software hub. But she expects yet more opportunities for young chemists. “I truly hope that we appreciate the value of a local medicinal chemist in biotech,” she says.

Where industry lives

Where leading US biopharma and chemical companies are headquartered

Broadly speaking, the US industrial chemical sector is composed of chemical makers and biopharmaceutical firms. For the former, the Houston metropolitan area emerges as the clear giant. Petrochemical firm ExxonMobil, polymer and fuel additive maker LyondellBasell Industries, and petrochemical major Chevron Phillips Chemical (CPChem) all plant their headquarters in Houston and its suburbs.

For the biopharmaceutical area, a few stalwarts dominate. Eli Lilly and Company is currently the largest pharmaceutical company in the world by market capitalization, and it had $65.2 billion in revenue last year, sending its home base of Indianapolis to the top of the chart. Merck & Co. similarly propels Rahway, New Jersey, high up in the rankings. Not all these firms’ research takes place at their headquarters, but where a company chooses to call home still affects the local chemistry economy.

By state, the most revenue among top biopharmaceutical and chemical companies in 2025 came from New Jersey–headquartered companies, anchored entirely by the biopharmaceutical sector: Merck, Johnson & Johnson, and Bristol Myers Squibb. Texas commanded about half as much; all $92.1 billion in collective revenue came from the chemical makers ExxonMobil Chemical, LyondellBasell Industries, CPChem, and Celanese.

It may be no surprise that biopharmaceutical companies are the biggest moneymakers of the chemical industry. Lilly sails past its competitors, owing mostly to the resounding financial success of the glucagon-like peptide 1 (GLP-1) receptor agonists Zepbound and Mounjaro.

Case study: Greentown Houston seizes on corporate powerhouses for next-generation innovation

At an old Fiesta Mart in Houston’s Midtown district, a group of innovators is making an ambitious bet: that Houston’s position as the energy capital of the US will also make it fertile ground for cleantech start-ups.

The start-up incubator Greentown Labs opened its Houston outpost in 2021. Greentown Houston is now home to a growing class of companies aiming to make the oil industry more efficient and provide alternatives to fossil fuels.

Resident start-up Capwell, for instance, has designed a system to capture methane emissions from low- and intermittent-flow events in oil and gas facilities. FlowCellutions has developed a sensor that can predict battery problems, with the goal of preventing power outages. GigaDAC has developed an airflow system designed to efficiently capture carbon dioxide. Today, Greentown Houston is home to about 130 companies and counting.

While Houston is clearly the energy capital of the world, it’s becoming more than that, says Greentown Labs’ Houston head, Kelsey Kearns, who grew her career in small-business strategy before joining Greentown last year. “Greentown’s mindset on this is that it’s really more than a transition. It’s a transformation.”

Setting up shop in Houston put Greentown near a bevy of corporate and academic partners. Houston-based CPChem and Shell USA back Greentown and, in doing so, gain access to resident start-ups through meetings, pitch days, and the open-innovation program Greentown Go. Greentown Houston’s landlord is Rice University; the college is in the process of transforming part of Midtown into the Ion District, billed as an innovation hub that connects entrepreneurs, academics, start-ups, and established companies.

“One of my big goals, personally, is for us to be really ingrained and embedded in the Houston scene here,” Kearns says. “How are we going to be able to collaborate with more companies and therefore help our start-ups even more?”

Innovation hubs

Following the money behind chemistry’s next breakthroughs

The ingredients for strong start-up ecosystems are somewhat different than those for larger companies and research institutions. To succeed, start-ups need, most crucially, money—specifically, venture capital on favorable terms.

C&EN used data that PitchBook collected on venture investment in the biopharmaceutical and energy start-up sectors as a proxy for chemistry innovation, since they are the most chemistry-intensive sectors PitchBook tracks. The data show that California dominates, cementing the state’s reputation as a haven for entrepreneurs. Massachusetts follows closely behind, likely owing to the high concentration of biotech start-ups in greater Boston. Texas lags a bit, in third place, although recent reports point to a surge of investment activity in the state.

There is a crucial difference between the number of investments that flow to start-ups in a state and the overall value of those investments. Start-ups in Texas and New York, for instance, received a similar number of investments in 2025, but the Texan firms collectively got more than three times as much money as the New York start-ups did.

Many start-up founders also seek the support of accelerator programs, like Y Combinator, and incubators to help get their businesses off the ground. Accelerators often come with investments or cash prizes. San Francisco is home to some of the most active start-up accelerators in the US by number of investments.

Chemistry start-ups also often need laboratories. The Boston area dominates when it comes to commercially available lab space, driven in part by a pandemic-era building boom. The San Francisco Bay Area and San Diego closely follow.

Case study: Tevard Biosciences leverages Lilly to make RNA drugs in Boston

Daniel Fischer and Elisabeth Gardiner were working out of LabCentral, a biotech incubator in Cambridge, Massachusetts’ Kendall Square, when representatives from Eli Lilly and Company stopped by. They were looking for entrepreneurs who might be interested in leasing space in Lilly’s newest building: a glistening, 12-story mix of lab and office space overlooking the water in Boston’s Fort Point.

Much of that building, dubbed the Lilly Seaport Innovation Center, would host Lilly’s own research efforts in RNA- and DNA-based drug candidates. But part of it was intended to house the Boston outpost of Lilly Gateway Labs, an incubator for start-ups aligned with Lilly’s interests.

Fischer wasn’t keen at first. As a cofounder and the CEO of Tevard Biosciences, he liked the dynamism of Kendall Square, home to perhaps the nation’s highest concentration of biotechs and pharma outposts. But the new Lilly Gateway Labs wasn’t far—only 20 min by subway—and it would put the eyes of Lilly researchers on Tevard’s transfer RNA (tRNA) drug candidates.

“When we started thinking about all the potential benefits of being close to Lilly, we made a decision to come here,” Fischer says. So in 2024, Tevard moved in.

Today, Tevard has about 20 employees at Lilly Gateway Labs. The company is developing tRNAs designed to restore full-length proteins in diseases in which premature stop codons prevent them from forming correctly. Its lead drug candidates are meant to address Duchenne muscular dystrophy and titin-related (TTN) cardiomyopathy. The company is currently running studies in monkeys. Meanwhile, leaders say they’ve benefited from the atmosphere at Lilly Gateway Labs. Twice a year, Lilly hosts poster sessions where resident start-ups can present, and in September, Gateway Labs will host its annual portfolio day in Boston.

“It’s really a lot of fun, and it’s an opportunity to see just everything,” says Gardiner, who is Tevard’s chief scientific officer. “There’s potential collaboration with another Lilly Gateway company or with Lilly themselves.”

About 12 other start-ups work at Lilly Gateway Labs in Boston. As for Tevard, Fischer recently renewed its lease, so the firm will be there for at least another year.

“What makes the Boston site really special is that of all of our sites across the US and outside the US, it’s the only site where we have embedded Gateway into a Lilly research building,” says Julie Gilmore, Lilly Gateway Labs’ vice president and global head. “The whole point of a model like this is how we can engage to accomplish something together that neither of us is going to do well, or as fast, alone.”

International powerhouses

Global chemistry output is concentrated heavily in Asia

The US may have strong chemistry chops, but it is not the center of the chemistry universe.

When it comes to publishing chemistry-related papers, only the New York, San Francisco, and Boston metropolitan areas hold their own on the world stage. Chemistry paper output is highest in Asia—specifically, China, where Beijing and Shanghai far outpace every other city in the world. Article count is only one way to measure chemistry innovation, but it reflects a clear reality for the global chemistry enterprise: China is a force to be reckoned with.

China is also home to more than half the world’s top 20 universities for chemistry, according to US News and World Report. The US has 5. Singapore, Saudi Arabia, and the UK also play host to world-renowned chemistry schools, but not to the same degree. US News’s subject-specific rankings evaluate academic research performance, taking into account global and regional reputations alongside bibliometric data, including research publications and citation counts.

The complete picture

A US-specific slider model for chemistry innovation priorities

Where you choose to build your career will ultimately depend on different factors.

Start-up founders, for instance, are likely to seek lab space and venture capital investment for their budding companies. They may also value collaborations with nonprofit research institutions, like universities and hospitals, or they may seek mentorship and camaraderie with other chemists through local chapters of professional societies.

Below, we have put together an interactive tool that allows you to weigh which metrics are most important to you and, in turn, learn which US cities come out on top.

Build your own city ranking

Choose a preset or adjust the weights to see how different priorities change the ranking across 25 selected US cities and regions. Presets are designed to serve as examples, not definitive estimates of importance. When one weight changes, the others adjust proportionally to keep the total at 100%. A dimension with no data contributes zero points but does not mean no activity.

No single winner

Chemistry is one of the few professions for which geography still matters. But there’s no one right way to build a career in chemistry—and there’s no single city, in the US or worldwide, that emerges as the center of chemical innovation.

New York City has the highest article output in the US, but it has comparatively little lab space. It might be the right place for someone seeking to join a New York–based university or research hospital, but not for someone setting up a commercial laboratory. Likewise, Illinois-based cities may not top the charts for any one metric, but the state is home to biopharma heavyweight AbbVie and several of the country’s top chemistry schools, showing a clear career path for a subset of medicinal chemists.

These data provide only snapshots of the chemistry enterprise. Even now, tides are shifting. Texas is capturing more and more venture capital. Plenty of commercial lab space in Boston is sitting empty, since much of it was built on spec—without tenants in place—in the last few years. NIH funding has dropped overall since these numbers were released, and it’s not clear when or where it will increase again.

Only time will tell what the future cities of chemistry will be.

Methodology and data sources

Sections 01–04 present selected indicators of chemistry activity. All data except the slider model preserve the geographies defined by each source. These may include cities, metro areas, markets, and states. C&EN’s analyses use the latest complete, suitable data that were available from each source; periods range from 2024 through 2026. Because US National Institutes of Health (NIH) funding and Nature Index research output appear together in section 01, C&EN used fiscal 2024 NIH data to match the latest city-level chemistry data from Nature Index, which cover 2024. The source list below identifies each dataset.

The slider model is a modeled comparison of 25 selected US chemistry locations: Ann Arbor; Atlanta; Austin, Texas; Baltimore; Boston; Boulder, Colorado; Chicago; Columbus, Ohio; Dallas; Denver; Houston; Indianapolis; Los Angeles; Madison, Wisconsin; New Haven, Connecticut; New York City; Philadelphia; Pittsburgh; Princeton, New Jersey; the Raleigh-Durham area, North Carolina; San Diego; San Francisco; San Jose, California; Seattle; and Washington, DC. It is not an estimate of every US city. C&EN selected locations that recur in the source data or contribute substantially to at least 1 of 10 dimensions: chemistry-related venture capital, commercial lab inventory, higher education, industry presence, institutional support, NIH funding, patent publication activity, research output, start-up support, and talent pool.

The slider model maps records from sources to the common set of 25 locations. When a source reports only a state total, the model distributes it by commercial lab inventory reported by the firm Jones Lang LaSalle (JLL) if every represented location in the state has a matching JLL market; otherwise, it divides the total evenly. Either method preserves the reported state total. These are modeled allocations, not source-reported city values. JLL’s Bay Area and greater DC markets map to San Francisco and Washington, DC, respectively; these are regional market-to-target mappings.

Text that appears when hovering over each bar identifies each value as direct, mapped, allocated, proxy, or no data. A direct value uses a source geography that matches the target, such as a Chicago record used for Chicago. A mapped value assigns the full source value to a corresponding target when the source and target use different geographic names or boundaries. For example, the JLL Bay Area market maps to San Francisco. An allocated value divides a broader source total among multiple targets using the distribution method described above. A state total may be distributed among the represented locations according to their shares of JLL commercial lab inventory; Boston receives all Massachusetts funding because it is the model’s only Massachusetts location, for instance. The resulting values preserve the reported state total. A proxy value uses a broader geography to represent a target when target-specific data are unavailable. A proxy value is not divided and should be interpreted as an estimate.

Each dimension is scaled from 0 to 100 relative to the highest value among the 25 locations. To limit the influence of unusually large values, we used a logarithmic scale before normalization for venture capital, start-up support, patent publication activity, and industry presence.

Dimensions glossary

Chemistry-related venture capital uses PitchBook data for the pharmaceutical and biotechnology sector as well as the energy sector because those are the most chemistry-intensive sectors PitchBook tracks.

Commercial lab inventory draws from JLL’s most recent U.S. Lab Property Report. These data are merely a snapshot: they represent only the first quarter of 2026, and because they are limited to available inventory, they do not take into account factors such as price. JLL reports New Jersey as a regional market, but according to the firm, most of New Jersey’s lab inventory is in Princeton; all of New Jersey’s commercial lab inventory is thus assigned to Princeton in the slider model.

Higher education uses the current US News and World Report US chemistry graduate program scores and maps each institution to a selected location. The separate international display in section 04 uses US News’s global chemistry ranking, which follows a different methodology.

The industry presence dimension in the slider model, as well as section 02, “Where industry lives,” is based on headquarters, not local facilities or employment, and combines reported revenue from the top 10 biopharmaceutical firms and top 10 chemical makers in the US. Revenue reports come from US Securities and Exchange Commission filings with the exception of Chevron Phillips Chemical, whose revenue report is published separately on its website.

Institutional support draws on the American Chemical Society’s list of local chapters. Each chapter is counted once toward its nearest metro area; for example, the ACS Central Texas local section is counted toward Austin, Texas. ACS publishes C&EN but is not involved in editorial decisions.

NIH funding uses all fiscal 2024 awards reported for each state, including awards outside chemistry.

Patent publications are divided equally among their unique inventor locations before US state totals are calculated. Each publication has a total value of 1, divided equally among its distinct locations; repeated inventors at the same location do not increase its share. The location fractions are then summed by state. For example, one New Jersey location and one New York location receive 0.5 each. Two distinct New Jersey locations and one New York location receive one-third each, producing state totals of two-thirds for New Jersey and one-third for New York.

Research output uses the Nature Index’s measurements by count and share. The count approach credits a location when at least one author of an article is affiliated there, so one article can count for multiple locations. The share approach fractionally apportions each article among authors and their institutions, and each article has a total share of 1 available. The research output dimension in the slider model separately normalizes article count and article share to 0–100, then averages the two scores with equal weight.

Start-up support reflects the number of accelerator and incubator programs from Failory’s top 100 list that have headquarters mapped to the 25 locations on our list. Failory’s list is a broad, industry-agnostic snapshot, not a complete or chemistry-specific measure of start-up activity.

The talent pool dimension combines May 2025 state employment estimates for chemists, biochemists, and biophysicists from the US Bureau of Labor Statistics.

A dimension with no usable source record scores zero and appears as “No data.” This designation can lower a location’s score, but it means only that C&EN did not find usable data for that dimension, not that no underlying activity exists. Locations with data for fewer than 8 of the 10 dimensions are noted because their scores may reflect gaps in data availability as much as measured chemistry activity. C&EN chose 8 of 10 as a threshold to flag locations missing at least 2 dimensions.

The balanced preset gives every dimension a 10% weight. Other presets provide starting points that illustrate possible priorities and are not intended as definitive statements about which dimensions matter most for a particular focus. Locations are ranked by weighted score; exact ties share a rank and appear alphabetically. Results depend on the selected weights, data availability, geography mapping, and missing-data rule.

Limitations

The analysis is limited to 25 selected locations and the records C&EN could identify in the included sources. Source coverage varies by dimension and location. Because a dimension with no usable record scores zero, a location with limited source coverage may rank below a location with more complete coverage even when chemistry activity exists in both places.

The slider model combines records reported for cities, metro areas, commercial markets, and states. Mappings, state allocations, and proxies make those records comparable, but they are modeling choices rather than source-reported city values. A different mapping or allocation method could change a location’s score and rank. Because each dimension is normalized against the highest value among the 25 locations, the scores are relative to this group and should not be interpreted as absolute measures of chemistry activity.

The sources also measure different aspects of the chemistry enterprise and do not provide identical coverage. JLL reports selected commercial lab markets, and Failory’s broad directory is neither complete nor chemistry specific. Industry presence combines biopharma company revenue with chemical sales, which are not equivalent financial measures, and assigns those values by headquarters. The source periods range from 2024 through 2026, so the results are a snapshot assembled from the latest suitable data available rather than a single-year census.