chemistry · manufacturing · family: the solution exists but nobody can afford it
biology can't beatpetroleum on price
Biological production of commodity molecules carries an inherent cost floor that chemical synthesis does not
Problem statement
Synthetic biology companies that engineer microorganisms to produce chemicals via fermentation consistently fail to achieve cost parity with petrochemical or chemical synthesis routes for commodity-scale molecules. Zymergen ($1.5 billion raised, sold to Ginkgo Bioworks for $300 million), Amyris (bankrupt 2023, never profitable), and Ginkgo Bioworks (6 of 219 programs commercialized in 15 years) all demonstrate the same pattern: engineered organisms can produce target molecules in the lab, but the overhead of keeping organisms alive, and then extracting the compound from them, creates a cost floor above chemical synthesis for molecules that are not complex enough to justify the premium.
Why this matters
McKinsey projects that new bio-based materials alone could become a $300 billion market by 2040, swelling into the trillions once biofuels, pharma, cosmetics, and other personal care products are added in. Investor cash poured into synthetic biology companies — Zymergen alone raised $1.5 billion, and the three flagship firms each once had multibillion-dollar market capitalizations — predicated on the premise of greener, more sustainable manufacturing that could replace petrochemical routes. The systematic failure of this thesis for commodity molecules — biofuels, commodity fragrances, bioplastics precursors — has destroyed billions in capital. Understanding the boundary between molecules where biology wins (structurally complex, stereochemically specific, high-value) and molecules where chemistry wins (simple, commodity-scale, thermochemically accessible) would redirect investment and R&D toward problems where biological production has genuine advantage.
What’s been tried and why it hasn’t worked
Amyris spent 20 years (founded 2002, bankrupt 2023) attempting to make molecules via engineered yeast fermentation — first the antimalarial artemisinin, then biofuel, then personal care and beauty ingredients. Its biodiesel performed well in tests, but production costs were far too high to compete with conventional diesel: in its first large contract it sold fuel at a loss even with Brazil's public subsidy. Yields were poor, the yeast kept dying, bacterial contamination was a recurring problem, and Amyris struggled to meet the 40–50 million-liter production quotas that are standard in the fuel industry. Zymergen's flagship product (Hyaline, a film for foldable phone screens) was not even being produced by fermentation at the time of its 2021 prospectus — it was "a non-fermentation produced biomolecule sourced from a third party" that the company planned to "convert to a fermentation-produced molecule" in 2022. Ginkgo Bioworks' platform approach — highly automated labs for testing many genetic modifications in parallel — has not produced a blockbuster product, and its customers complained of high costs and delayed deliveries; the lesson Constantin draws across all three companies is that it does not make sense to specialize exclusively in microbial strain engineering and not the rest of the manufacturing process, when the new products will necessarily require manufacturing process development.
What would unlock progress
Clear techno-economic frameworks that identify which molecular targets genuinely benefit from biological production (molecules requiring stereospecific construction, multi-step transformations, or bio-derived precursors not available from petrochemistry) versus targets where chemical synthesis will always win on cost. Advances in continuous fermentation (reducing batch costs), cell-free biosynthesis (eliminating organism overhead), and consolidated bioprocessing (reducing downstream purification steps) could lower the biological cost floor. Machine learning models trained on failed and successful bio-production campaigns could predict commercial viability earlier, preventing billion-dollar misallocations.
Entry points for student teams
A team could develop a techno-economic decision framework comparing biological and chemical synthesis routes for a set of target molecules, using published cost data from fermentation companies such as Amyris and commodity chemical benchmarks. A more technically oriented team could prototype a cell-free biosynthesis system for a target molecule, measuring whether removing the organism eliminates the cost floor. Relevant disciplines: chemical engineering, biochemistry, economics, data science.
Genes
Structural cousins: stuck the same way, in other fields
Sources
Sarah Constantin, "What's Behind the SynBio Bust?", Rough Diamonds (Substack), January 30, 2025, KdT Ventures, "Zymergen: A Cautionary Tale," Medium, 2021, Amyris Chapter 11 filing, SEC, August 2023 (Amyris, Inc. Form 8-K filed August 10, 2023, EDGAR accession 0001365916-23-000094, reporting the August 9, 2023 petition); Robert F. Service, "Synthetic biology, once hailed as a moneymaker, meets tough times," Science (ScienceInsider), 22 August 2024, accessed 2026-02-20 go to source 1 ↗ go to source 2 ↗ go to source 3 ↗
Working notes
Notes made while writing and checking this brief, shown as written. They use working shorthand.
The synbio bust of 2021–2024 (Constantin dates the stock-price collapses to 2021–2023; Science, writing in August 2024, says the three flagship firms faltered over the past year: Amyris bankrupt in 2023, Zymergen sold to Ginkgo Bioworks for $300 million after raising $1.5 billion and liquidated in 2024, Ginkgo's stock price "plummeted almost to nothing" in 2021 and layoffs of more than one-third of its workers in 2024) provides the richest failure data in the collection for this problem. Distinct from manufacturing-biomanufacturing-scaleup-prediction, which covers the technical prediction of scale-up failure; this brief covers the fundamental economic barrier — even when scale-up succeeds, the cost floor of biological production is above chemical synthesis for simple molecules. Constantin's rule for where biology wins: biomanufacturing is used in practice only for highly complex, expensive compounds that can't be made thermochemically at all (her examples of routine biomanufacturing include industrial enzymes, citric acid, xanthan gum, and protein drugs such as insulin). Related: circular-pha-bioplastic-fermentation-economics (same cost-floor challenge in bioplastics).
Reconciliation 2026-09-30: Fact repair after reading the whole write-up against every source it names. All four were fetched on 2026-09-30. Read in full: Sarah Constantin's post (directly from https://sarahconstantin.substack.com/p/whats-behind-the-synbio-bust; the post is dated January 30, 2025); the KdT Ventures post (the Medium page blocks automated readers, so the Internet Archive copy of 2025-10-14 was read; it is a two-minute post dated August 5, 2021); the Science news article (the publisher page and the newest archive copy both block, so the Internet Archive capture of 2024-08-23 was read; the article is by Robert F. Service, dated 22 August 2024, doi 10.1126/science.zfs12p8); and the Amyris Form 8-K filed with the SEC on August 10, 2023 (EDGAR accession 0001365916-23-000094), which reports the Chapter 11 petition of August 9, 2023 and carries no financial history. (1) The Source line gave the Constantin post as 2024 and the Science article as 2023; they are dated January 30, 2025 and 22 August 2024, and the line now carries those dates, the Science author, and the 8-K identifier. (2) "Zymergen ($1.57B raised, acquired at 94% loss)": Science says Zymergen "had previously raised $1.5 billion from investors" and "was sold to Ginkgo Bioworks for $300 million"; no source gives 94%. The text now says $1.5 billion raised and sold for $300 million, here and in the notes above. (3) "Amyris ($1.67B raised ... $1.9B accumulated deficit)" and, below, "20 years and $1.9 billion": none of the four sources gives either dollar figure; Constantin says Amyris was founded in 2002, reached $153M in revenue in 2019, "never managed to reach profitability" and filed for bankruptcy in 2023. The text now says bankrupt 2023, never profitable, and keeps "20 years," which the 8-K's press release exhibit also gives ("Since its founding 20 years ago"). (4) The list of overhead items (sterility maintenance, metabolic waste management, oxygen transfer, substrate feeding, temperature control) is not in any source; Constantin says a biological route "involves all the 'overhead' of keeping the organism alive" and that extracting the compound from the organism is a further cost. The sentence now says that and no more. "Fully commercialized" became "commercialized," the word Constantin uses. (5) Removed because no source contains them: the bioeconomy "valued at over $1 trillion"; the 2022 US Executive Order; and "over $10 billion in venture capital ... between 2020–2023." The paragraph now uses what Science reports (McKinsey's $300 billion bio-based materials projection for 2040, "swell into the trillions" with other sectors added; the three flagship firms each once had multibillion-dollar market capitalizations) and what KdT reports (companies raised capital "on the premise of more green, sustainable manufacturing"). (6) "Isoprenoids (initially biofuels, then fragrances)": no source names isoprenoids; Constantin gives the sequence artemisinin, then biofuel (2006), then personal care and beauty ingredients (2012). The sentence now follows her. (7) The metabolic-engineering sentence ("modified central carbon metabolism, genetic switches, enzyme optimization ... too much sugar and oxygen per mole") is not in any source and was replaced by what Constantin reports: poor yields, the yeast kept dying, recurring bacterial contamination, production costs far too high, fuel sold at a loss under its first large contract even with Brazil's subsidy. (8) "Equipment malfunctions (each costing ~$1.35M)" is in no source and was removed; "yields at 40–50 million liters never matched projections" now says what Constantin says, that Amyris struggled to meet the 40–50 million-liter production quotas standard in the fuel industry. (9) "Hyaline polyimide film ... chemically sourced ... at IPO": no source says polyimide or chemically sourced; Constantin quotes the 2021 prospectus calling Hyaline "a non-fermentation produced biomolecule sourced from a third party" to be converted "to a fermentation-produced molecule" in 2022, and describes it as a film for foldable phone screens. The sentence now uses those words. (10) The Ginkgo sentence ("downstream royalties," strain engineering "perhaps only 10–20% of the challenge," "cannot be horizontally platformed") had no source for the royalties or the percentage. It now says what Constantin reports about Ginkgo (automated parallel labs, no blockbuster product, customer complaints of high costs and delayed deliveries) and her stated lesson that specializing exclusively in strain engineering and not the rest of the manufacturing process does not make sense when the products require manufacturing process development. (11) The entry point named Amyris farnesene and Genomatica 1,4-butanediol cost data, and the notes named Genomatica's 1,4-BDO and DSM's vitamins as successes; none of these products or companies appears in any source. The entry point now points to fermentation companies such as Amyris, and the notes now give Constantin's rule and examples for where biomanufacturing is used. Both entry points remain, and the techno-economic framework needs no special facility. (12) "The synbio bust of 2023–2024 ... Ginkgo down 98% from peak": no source gives 98%. Constantin dates the stock-price collapses to 2021–2023 and says Ginkgo's stock price "plummeted almost to nothing in 2021"; Science, dated 22 August 2024, says the three firms "have faltered over the past year" (Amyris bankrupt in 2023, Zymergen liquidated in February 2024, Ginkgo layoffs of more than one-third of its workers in June 2024), so the original 2023–2024 had support too. The notes now give the range 2021–2024 with each source's dating. Confirmed in the sources and left as written: the cost floor thesis of the long title and display title (Constantin: simple molecules produced thermochemically are "invariably cheaper than biomanufactured molecules"); 6 of 219 Ginkgo programs commercialized in 15 years; Amyris bankrupt in 2023 (8-K, Science, Constantin); Zymergen's cautionary tale about losing sight of the customer (KdT); biofuels, bioplastics precursors and aromachemicals as cases where chemistry wins. Not checked, because they are proposals rather than sourced claims: the "What Would Unlock Progress" section. No title changed. Genome tags untouched. A second reader went over every repair on the same day against the same four sources (plus the 8-K's press-release exhibit and EDGAR filing index) and made the two adjustments recorded in (3) and (12); all other repairs stand as written.