Last updated June 6, 2024
Our modern economy is built around petroleum. A lot of people “know” this, but they don’t really know it. It is sometimes hard to grasp how intertwined modern life is with fossil fuels, and how our ability to leverage this million year old biomass in a myriad of ways is a fairly modern phenomenon. Most people are familiar with the use of fossil fuels to power our cars, planes, major industrial processes, and electric grid, and home heating/cooling systems. What often goes unnoticed is how, beginning in the 1930s, our increasing sophistication in crude oil refinement and processing gave birth to the chemicals industry, which is the foundation for much of the materials we experience in our everyday lives. All you have to do is look at this chart to understand that everything from semiconductors to your sweet new Hokas comes from the chemicals industry.
https://lh7-us.googleusercontent.com/e6D-r7gy5G64zuaFTr5-esGo3UfS2npYLLO0LkhVD8i2h8Zv6h2x_k1rUV1jnDzFX7i-SP5C5JT7_CgSyymcSTLBYM8nxTEvXSXyS4PCCh4DUtutDgNdYK7hJywqjDHOCHbhA0-9OUVjHNctmJvshpo
And a lot of what you see in this chart is relatively new. After all, high density polyethylene (HDPE) was introduced in 1954, and only began to find true commercial traction when it was used to create the hula hoop. Now this highly durable plastic is used in everything from bottles, to cutting boards to toothpaste containers, to industrial tubing. Indeed, even 80 years ago, the next big thing started off looking like a toy.
All of this is to say that in order to imagine a decarbonized world, we have to have an understanding of the carbonized one, and an appreciation for how vast and intertwined it really is with modern life. Banning fossil fuels or petrochemical manufacturing will not solve the problem, and if you spend some time with the chart above, it becomes clear that such a ban, if it were possible, would cause immense suffering.
https://lh7-us.googleusercontent.com/avhNHMw-zlxPi7t7W3NY-MKfIburrurNQJMQliI6InVQ3tFpxQGlp6W4Y2W4x3spvgYQEfObhOWuzUsLidIvRqJDw7YSvGCz4z-ty_S_iFQXL3wh4VKvBR5Fj9lAB7RrXBYFjNb4Jz8iQfzgAn9E-_Y
14% of all crude oil and 8% of all natural gas is used for the production of chemicals globally. Our efforts should focus on developing and commercializing the technologies to enable viable replacements of these feedstocks, and then implement these technologies at scale.
Fortunately, we are entering a period where biology, not petroleum, can become the foundation for our industrialized world.
We have operated for the last 200 years in the Industrial Age, and I think one of the most overlooked areas in the innovation landscape is our impending transition into the BioIndustrial Age.
Core to this Bioindustrial economy is the notion that biology is the most efficient, scalable, renewable, adaptable manufacturer in the world, and due to a number of different technological advancements, we are starting to deploy it for our own purposes at scale.
To understand this, let’s look at fossil fuels from first principles. Crude oil is a many millions year old biomass that has been compressed over time, resulting in an extremely carbon rich material. We use old, carbon rich plants and fossils to power our economy. Not only will our supply of these fuels deplete one day, but also our use of these materials emits molecules like carbon dioxide, methane, and nitrous oxide that, when out of balance in our atmosphere, drive global warming.
To use these fuels (“feedstock”) to make chemicals, we build massive manufacturing plants that use significant energy resources to place the material under immense heat and pressure. Our entire chemical supply chain is built around these manufacturing processes. Said another way, in software, most “supply chains” are built around demand and most software markets are demand side constrained. The economics of making the product is not the gating factor, but rather the scale and velocity of demand for the product.
In chemicals (and really any commodity or near-commodity product), the supply chain is built around supply side constraints. Chemical production facilities are massive. The largest can reach up to 2,000 acres (~1,500 football fields), and the average chemical plant in the US costs $1.6Bn to build. The economics of such a large facility necessitates that facilities distribute equally large shipments, regularly shipping out rail cars of chemicals. As a result, there is a vast distribution network in the chemicals industry that break down the product and resell it in smaller packages, sometimes selling it to other distributors who further repackage and mix it into new products and mark it up along the way.
To make this concrete, Tony Maiorana at Polymerist describes it as such:
If we think about traditional industrial chemistry, it’s generally a collection of companies that form a supply chain where each company specializes at doing specific reactions. For instance, ExxonMobil could extract and refine crude oil and pipe some naphtha distillate over to INEOS who turns that distillate into molten phenol and loads it onto a railcar. Westlake Epoxy might take those railcars of molten phenol and react it with acetone to make bisphenol A and then do a reaction with epichlorohydrin (purchased from Olin) to convert into an epoxy resin they can sell via bulk tank truck. PPG might buy that base epoxy resin and formulate it into an anti-corrosion coating used on the Verrazano Bridge in order to protect the steel from corrosion.
As you can probably begin to tell, there a few key characteristics that make up our chemicals manufacturing industry:
In biology, cells are our factories (aided by catalysts called enzymes), and they react with organic materials to create a variety of molecules. If you’ve ever brewed beer, you know that yeast + sugar water at room temperature in a mostly sealed container creates beer + carbon dioxide.
What if we could design microbes to react with specific non-petroleum based feedstocks to produce the exact molecules we want? If we could do that, then we could move off of fossil fuels, and reduce the number of steps required to get from feedstock to final product, vastly simplifying the supply chain. Cells already do chemical conversions at ambient temperatures, and can operate at higher yields (higher output per input). If we could do all of this, we would have the foundation for a cheaper and cleaner chemicals industry.
This is the promise of Synthetic Biology and the BioIndustrial Economy. There is no single definition of synthetic biology, but I think of it in this context as the practice of engineering custom biological systems and catalysts to perform specific chemical transformations.
Synthetic biology has most recently become more well known for its application in pharmaceuticals and alternative meat, however it has a far greater number of industrial applications from chemicals, to biofuels, fragrances, textiles and food ingredients. Each vertical has its own set of technologies as well as infrastructure and economic hurdles. Here, I’ll be focusing on chemicals and chemical derived products.
Before jumping into the market dynamics of the BioIndustrial economy, let’s first talk about the technological changes enabling it.
Engineering microbes to create a more efficient, cleaner chemical reaction isn’t a new idea. The issue is that the process of engineering these microbes has historically been prohibitively expensive.
To begin, let’s start with our understanding of biological engineering in the first place. In 1962, James Watson, Francis Crick and Maurice Wilkins won the Nobel Prize in Physiology for their discovery of the molecular structure of DNA. Soon after in 1965, François Jacob, André Lwoff, and Jacques Monod won the same Nobel Prize for their groundbreaking discoveries concerning how genes regulate the production of enzymes and viruses in cells. Put simply, they discovered that genes have an "on" and "off" switch that controls when they are active or inactive and that certain genes, called regulator genes, act as the switches that turn other genes on or off. In essence, they discovered the first mechanism by which we might develop new expressions of existing genes. This laid the foundation for the birth of genetic engineering technologies developed in the following decades.
Great, the foundation for manipulating genetics was developed in the 60s, but what good is it if it is really expensive to deploy? In 1990, the US government launched a massive effort to sequence the entirety of the human genome. As a brief reminder, the human genome is the complete set of instructions encoded in our DNA across all 23 chromosome pairs. It contains around 3 billion base pairs, and sequencing the human genome means figuring out the exact order of those 3 billion base pairs for a person.
The first sequence took 10 years and $3Bn. Today, you can sequence genomes in a day for less than $1,000. The cost decline is absolutely staggering, and it is probably one of the least appreciated technological achievements of our modern world (along with the development of the Haber-Bosch process, in my opinion).
https://lh7-us.googleusercontent.com/HSfIv6qn_imWGI4H35qI4VrQ1rmdYoybS3be7Sh-XhsMNV8KRUKcC2y27AA8dr6w0neugMF-9XoTtYdRc2eCa0MlFu-pmHE76il0Lapx71osBEGvFiZGpQTUmNPgrM_PlVmFvrX0U7TIA9gFRO0brPA
As with any new technological discovery, being able to deploy it quickly and cheaply often leads to a cambrian explosion of new ideas and tinkerers pushing the science forward at an exponential pace. Fast and cheap genome mapping became the cornerstone of the technological shift we call synthetic biology today.
In 2012, as genome sequencing crossed the $10,000/genome mark, Jennifer Doudna and Emmanuelle Charpentier published the discovery of CRISPR in Science Magazine (and later won the Nobel Prize for it in 2020).
CRISPR is a gene editing tool that allows us to programmatically modify DNA sequences in living organisms. It's like writing instructions for molecular scissors that can find, cut and paste pieces of specific genetic code.
It’s difficult to overstate the potential of this technology. CRISPR's ease of use, versatility, speed, scalability, and cost-effectiveness are democratizing gene editing, making it accessible to a wide range of researchers, and increasing the surface area of genetic engineering applications.
I personally first encountered CRISPR at Bowery Farming when I led our indoor farming seed breeding partnership efforts. It is still in its early days, particularly due to regulatory considerations, but CRISPR has the potential to provide the backbone of the next Green Revolution (another post for another time).
What has been remarkable about CRISPR is the speed with which the technology has reached commercial scale. VC deployed over $1Bn into CRISPR startups in 2021. Yes that was in the ZIRP days, but the technology was an un-cited academic publication less than 10 years prior! Bringing us back to the chemicals world for context, polyethylene was invented in 1933 but the single use plastic bag for which the polymer is most famous didn’t even exist until [1965](https://www.unep.org/news-and-stories/story/birth-ban-history-plastic-shopping-bag#:~:text=1965 – The one-piece polyethylene,cloth and plastic in Europe.).
In 2018, Frances Arnold won the nobel prize in chemistry for her work on enzyme engineering and use in chemistry - work that she had begun in the late 1980s. Her research was centered on the idea that if she could “catalyse the chemical reactions that occur in the Earth’s organisms and, if she learned to design new enzymes, she could fundamentally change chemistry.”
Enzymes are nature’s catalyst. They function as machines in our cells that make our biological manufacturing processes accelerate, whether it’s digesting food or fighting infections. Arnold’s work resulted in the development of a variety of enzymes that could do everything from break down plastic to develop fuels from plant waste.
What I find fascinating about her work was that it began in the 1980s, and was economically untenable until the technologies referenced above, as well as others, were developed. Originally, Arnold had to manually introduce random mutations into enzyme genes. The advent of genome sequencing allowed her to outsource the synthesis of billions of mutated gene variants to commercial DNA synthesis companies at lower costs. Similarly, it became cheaper and faster to sequence the genes of the best performing enzymes.
Additionally, advances in computational modeling and machine learning enabled Arnold to guide which mutations to make and better predict which variants might have desired properties, further accelerating discovery timelines.
Although there is no mention of CRISPR in Arnold’s Nobel Prize winning work, her discoveries in enzymatic engineering and CRISPR technology now make for a wildly potent combination that we are just starting to see deployed commercially. After identifying beneficial enzyme mutations, CRISPR is now the cheapest, most efficient pathway to incorporating those mutations into production strains.
The sequencing of the above technological discoveries is why investing at the leading edge of technology is fascinating. Each subsequent innovation compounded the effect of the last and enabled the next, culminating in where we are today, at the confluence of all four technologies into one overarching trend and movement. Now is the time for biology to rewrite the chemicals industry.
First let’s establish why, at a high level, the bioeconomy has the potential to rewrite the way we think about chemicals. If we can scale the engineering of microbes to make molecules using abundant, organic feedstocks, we should be aiming for a cheaper, cleaner, safer, higher performing, simplified chemicals industry and supply chain.
There are a number of key levers to consider to understand the BioIndustrial Economy’s advantage against the PetroIndustrial Economy.
As discussed above, the entire chemicals industry is built around petroleum derived feedstocks, making the chemicals industry particularly sensitive to oil prices, and reliant on building capacity in oil rich areas (no surprise that Texas has the most chemical plants in the US).
Microbes, however, can be engineered to utilize renewable biomass feedstocks such as corn or abundant plant matter like cellulose and lignin, as well as excess materials such as waste methane or CO2.
Not only are these feedstocks more abundant than oil and thus decrease supply chain risk, but they are also cheaper and less susceptible to price fluctuations in the market.
Efficient microbe production is one of the most challenged areas of the bioindustrial economy. Although it has become significantly cheaper to design and manufacture microbes for specific reactions with specific feedstocks, doing so at scale continues to be a cost barrier.
Further, enzyme catalysis in chemical processes can be limited by the lack of enzyme stability at high temperatures where the feedstock is more soluble, making the production process less efficient.
As a result, enzyme production and enzyme efficiency is one of the greatest opportunities for innovation in the BioIndustrial Economy, and a critical lever in driving the economics of the industry at large. For cell-free (in-vitro) enzymatic synthesis, immobilization is one of the leading methods for enhancing the economics of enzymatic catalysis, effectively allowing for the re-use of enzymes (continuous fermentation), driving down enzyme cost.
Petroleum based chemistry requires significant (800 degrees F) heating for reactions, and this requires a significant amount of energy.
Engineered microbes can catalyze reactions under milder temperatures, pressures, and pH conditions compared to the harsh conditions often required for chemical catalysts. This reduces energy costs associated with heating, cooling, and maintaining extreme environments.
More specifically, cell-free (in-vitro) enzymatic synthesis has the potential to be less energy intensive than that of fermentation with live yeast (where enzymes are located in the yeast) because fermentation tanks require large air compressors/cooling units.
Yield is a key operating metric for any physical manufacturing operation, whether it’s chemicals or machine parts or farming. Yield is the amount of product you get out for a certain amount of resources you put in. A 50% yield means you get 1 widget for every 2 widget raw materials you put in.
For typical chemical manufacturing, yield is 60%. On top of this the other 40% is either waste or other products that require additional and separate processing and distribution costs.
Microbes, however, can be engineered to exhibit incredible selectivity, or the degree to which a molecule selectively reacts with a specific microbe. This increases product yields, reduces waste treatment costs, and simplifies downstream processing.
When evaluating a new precision fermentation technology and process, titer and rate are the key concepts in evaluating unit economics. Titer is the final concentration or amount of the desired target molecule or product produced per unit volume at the end of the fermentation process, measured in grams per liter (g/L) of fermentation broth.
A higher titer means more product per unit volume, which can significantly reduce downstream processing and purification costs. Increasing titer is a major focus area for optimization in precision fermentation, as it directly impacts capital expenditures (e.g. smaller fermentation tanks) and operating expenses like separation costs.
Synonym’s study of 246 precision fermentation facilities across 40 countries clearly points to titer’s outsized effect on precision fermentation economics. The graph below shows that titer has 1/x relationship to COGS ($/kg). Said another way, doubling titer cuts COGS in half.
https://lh7-us.googleusercontent.com/6Z9Btu5jPQza6JEFBUIJT41V4-TxObp7ONrA76Qy_HnYKEfinczN1jGA0urGxFRjb-1ajuDGD4jnBRB4WEqBbaeAKtTbRWYsPR7Yx0NW8kKktFpN5Q_ITwaDyr8kaCAUIYNE7S9ajG90powpTBXD2E4
To fully understand the overall productivity of a process, titer needs to be contextualized with the time it takes to actually create the molecule. Productivity = titer x fermentation time, which can be expressed as g/L/hr or g/L/day.
Strategies to boost productivity are technologically and process driven, including optimizing the microbial strain, fermentation conditions (pH, temperature, aeration, etc.), media composition, and feeding strategies.
As stated above, the Petrochemical Industry operates at an absolutely massive capital expenditure. The largest can reach up to 2,000 acres (~1,500 football fields), and the average chemical plant in the US costs $1.6Bn to build.
Remember, petrochemical production requires immense heat, which is generated via steam, which is often generated on-site through boilers. An enzymatic reaction does not need steam, and as a result does not need the capital expenditure required to generate massive amounts of heat on site. Further, due the selectivity of engineered microbes, capex for processing and waste treatment is reduced as well.
That said, biochemical processes still benefit from economies of scale. Synonym’s study of 246 precision fermentation facilities across 40 countries suggests that total production capacity has an inverse squared relationship with COGS where every doubling of total capacity reduces COGS by ~1.4x. Still, it is far less pronounced when compared to productivity (above), and technologically driven productivity is the ultimate driver of fermentation economics.
https://lh7-us.googleusercontent.com/Z76w90yN_jMuD1Sg_mqxT9KQM_B8Ihr_AAxGeYdaqHBDaLoSART6x8u_g_0_mFiQ5NOi5F1UVy2Y34GIcgfbaZ2Wdc9Y9PgUJkY4rM4K9G5n6A4Ymcn5lP5M_0ahJzigrvztjl96eaWmc01qL6xTvIQ
All biochemical production companies ultimately need to decide how much production, if any, they will do in house, and how much they will outsource to contract manufacturers (CMOs). As we’ll discuss later, there exists a dearth production capacity globally, both at the pilot phase and at the commercial phase. The contract manufacturing production base is developing right now, and as a result is another key area of opportunity and innovation and will have an outsized impact on the ultimate economics of the industry.
For those companies that choose to manufacture in house, it is important to have a clear view of what capital requirements exist to reach each major commercial milestone. In manufacturing, if you grow too quickly before being profitable at the unit economics level, it becomes almost impossible to earn your way out of that cash sink because it is highly unlikely that your old uneconomical unit (your last facility vintage) can reach the profitability you need and becomes a major drag on the business.
The chemicals industry is massive, and is an input to almost every physical item we encounter in daily life. It should come as no surprise that 1/4 of total US GDP comes from the chemicals industry. Globally, this is a $5 trillion industry, and the global top 50 chemicals companies do $800Bn in revenue annually.
More specifically, the market for specialty, higher margin chemicals, was a $723Bn market as of 2021, and is expected to reach $1.1T in by 2030. Of the large industrial and global commodity markets, chemicals is one of the fastest growing, expanding at an almost 6% CAGR.
That said, a common pitfall in the Industrial Tech world is to point to a large TAM as evidence of a promising opportunity without being realistic about what technology is required to be cost competitive with each pricing tranche within the TAM. Most Industrial Tech companies start at a price premium with the plan to drive cost down over time, and as a result, their actual serviceable market is materially smaller than it might originally appear.
The question of scalability is not just about the total possible market size, but also about what economics are required to succeed at each phase of commercialization, and what capital partners can support each stage. For example, what capital is required to reach $1M in revenue? How about $10M?
This is important because it also impacts your serviceable market size. If you need $200M in capital to reach $10M in revenue, that means that it’s likely that the ASP required to sustain a viable business is pretty low, which is going to put a lot of pressure on the economics before you reach a massive scale.
This may be the least mature economic consideration of the unit economics of the BioIndustrial Economy. Given very few BioIndustrial chemical / chemical products companies have reached large commercial scale yet, it is unclear where TAM will begin to be inhibited by unit economic profitability.
The above economic considerations are a simplification and generalization for the myriad of companies that are building in the BioIndustrial Economy, but I find it to be a helpful guide for evaluating different technologies and identifying where in this new supply chain there might be opportunity.
Separately, I use these categories as a checklist of how a new technology might match up against the incumbent petroleum based product. Ultimately, I have the greatest conviction that enduring value will be created by technology companies that fundamentally change the underlying cost structure and carbon footprint of the commodities at the foundation of the industrial world. Cost parity with petrochemicals should be the north star for all of these companies, as I do not believe this market will sustain a green premium for very long. Climate is a prompt or a lens through which to apply technology, but it is cost, quality and speed first, carbon impact second.
Although the BioIndustrial Economy continues to evolve, I see four core categories defining the space, each with further subcategories.
https://lh7-us.googleusercontent.com/3Im_BlQlzUA0qD0tOatYFbswsjlZhQBqf4QsQQVkJpMeD-jQ41cAke4_gMYyQ_gpPTvOTxAwIpQCOON4i-r_a63j-m0o4R2qVt5Bjln-Fm8GBSAcuTUuCLkNHAbp0to_XBX7MH4dOQZu5Gwx3o4hx2Y
In the world of chemicals, the most salient companies of the BioIndustrial Economy are those that are producing and selling chemicals or chemicals derived products themselves. They will often engineer and potentially produce microbes themselves, manage the production process themselves, and sell the final product, which may be a biochemical, like Solugen’s hydrogen peroxide, or a biochemical based product, like Mango Materials’ bioplastic.
Some have existed for a long time relative to the technological trends driving the industry. For example, Natureworks released it’s biobased polylactic acid made from corn starch back in 2003, and reached the 1 billion lbs sold in 2014. However, the product was far more expensive than its petroleum based counterpart, and found a comfortable, sustainable niche supplying makers of bone screws and stents and other medical materials that required biodegradability.
The companies today that are operating in this area can be broken out into two core categories defined by the technologies they employ to produce their end products.
Precision Fermentation refers to the use of genetically engineered microorganisms, like yeast or bacteria, to produce specific target molecules.
Enzymatic catalysis involves the use of isolated enzymes, which are biological catalysts, to facilitate and accelerate chemical reactions.
The key difference is that precision fermentation relies on the use of engineered microorganisms to produce the desired molecules, while enzymatic catalysis focuses on the use of isolated enzymes as catalysts, which can be applied in both cell-based and cell-free processes. However, the two approaches are often complementary, as precision fermentation can be used to produce enzymes that are then leveraged for enzymatic catalysis. The reason I separate the two is that as of late, they seem to be developing separate ecosystems and slightly different supply chain considerations.
To give you a sense of the differences, here are some of the leading companies as well as newcomers in each category