Limonene shrinks foam to about a twentieth of its volume and a first plant was built in the 1990s. No study found measures what the recovered polystyrene smells of, how the solvent ages, or what the process costs, and the industrial solvent routes use other solvents.
Thermocol Dissolves in Orange-Peel Oil, as Researchers Showed in 1998. The Published Record Stops Where the Recycling Begins
Put a piece of thermocol in orange-peel oil and it collapses in seconds, leaving a small puddle of syrup. Videos of this are common, and Indian press coverage of an IIT Hyderabad project has presented it as a way to recycle foam waste. The chemistry is real. What the videos do not show is what happens to the syrup: whether it can be turned back into polystyrene, how often the solvent can be used again, and who would pay for the plant.
This piece reads the published record, from abstracts, patents and project pages, because the full texts were not available to this blog. The finding is that the first half of the story is well documented and old, and the second half has hardly been published at all.
What dissolves, and what you get back
Thermocol is expanded polystyrene, a plastic that is about 98% air by volume. Limonene, the main compound in citrus peel oil, is a solvent for polystyrene, and so the foam dissolves and the air escapes. What is left is not the “chemicals” of the foam taken out one by one but a solution of polystyrene chains. That solution has three possible uses: precipitate the polymer with an alcohol and pelletise it again; use the solution directly, for fibres, coatings or adhesives; or treat the dissolving step as a way to compact foam before something else. It does not give back styrene, the building block, which needs heat or a catalyst.
The dissolving is also quick and cold. In the Noguchi series of 1998, the first to describe a recycling system built on limonene, foam shrank to about one-twentieth of its volume and the recovered polystyrene showed almost no loss of molecular weight, which the authors credited to limonene protecting the polymer when heated. A later paper on extruded polystyrene reported solubility of about 0.26 g per mL of limonene at 25°C, which on this blog’s arithmetic is a solution of about a quarter polymer by weight. The Colombian study put the fall in volume at about 96%.
| What was claimed | By whom | How firm |
|---|---|---|
| Foam shrinks to about 1/20 of its volume in limonene; recovered polystyrene shows almost no loss of molecular weight | Noguchi and colleagues, Packaging Technology and Science, 1998 | Peer-reviewed; seen by abstract |
| A prototype plant made polystyrene with mechanical properties like new | The same series, part 2 | Via a review chapter |
| Limonene can be reused at least 10 times | A review chapter | Secondary; not found in the chapter’s full text |
| About 0.26 g of polystyrene per mL of limonene at 25°C; terpenes cause no chain degradation | García and co-authors, Waste Management, 2009 (extruded polystyrene) | Abstract and a summary; numbers not checked |
| Foam volume cut by about 96%; limonene mixes better than eucalyptus oil | A Colombian study of temperature and concentration | Regional journal |
| Fibres of 300–900 nm from waste foam in limonene | Shin and Chase, Polymer Bulletin, 2005 | Peer-reviewed; coarse, poor quality |
| Oil-absorbing fabric: 40.5 g of oil per gram, 77.3% retained after an hour | IIT Hyderabad, J Applied Polymer Science, 2019 | Peer-reviewed; peel extract, not pure limonene |
Limonene is not the best terpene
At 50°C, p-cymene, another compound found in plant oils, dissolved more polystyrene than limonene did, about 212 g per 100 g of solvent against 182 to 197 g. At that strength the solution is about two-thirds polymer. Checked 25 October 2026: these figures match Table 1 of the full paper (Kitami Institute of Technology repository copy): p-cymene 212.0, (R)-limonene 181.7, α-terpinene 197.1, terpinolene 191.4 and 2-p-cymenol 156.7 g per 100 g of solvent. An earlier study by the same group, using a different polystyrene film, measured d-limonene at 127 g and its isomers at 122–131 g at the same temperature (Table 2 of Hattori’s InTech review chapter), so solubility depends on the polystyrene as well as the solvent; p-cymene came out at 212 g in both. p-Cymene and the polymer could also be recovered almost completely by steam distillation. The two sets of figures are for different temperatures and polystyrene, so they cannot be set side by side. The point is that limonene is one of several workable solvents, and the industrial solvent processes that exist, CreaSolv and Polystyvert, use others. The European CreaSolv plant at Terneuzen in the Netherlands was built for 3,000 tonnes a year of insulation-foam waste. Its operator filed for insolvency in March 2022 and a German group of foam makers, German EPS Converters, bought it and restarted it. It was reported in regular operation in June 2023 at an initial capacity of 3,000 tonnes a year, with 8,000 as the target, after about €17 million of investment. Its current throughput was not found.
Other solvents that dissolve it
Beyond the terpenes, the record names three more groups. The table gives what was seen for each, from abstracts, patents and project pages. The studies used different foams, temperatures and loadings, so the figures cannot be ranked against each other.
| Solvent | What was seen | Recovery | Concern |
|---|---|---|---|
| Limonene | Dissolves foam fast; volume cut of 96–98% reported in a Colombian study and a student project | Ethanol anti-solvent; distillation | Flash point about 47°C, strong odour, much solvent per unit of foam |
| p-Cymene | Dissolves more than limonene at 50°C; 33% by weight at room temperature in a Polystyvert patent | Steam distillation; heptane anti-solvent | Scale-up cost not published |
| Essential oils, unpurified | Full dissolution at up to 1:1 by weight, at least four times faster than limonene (one repository paper) | Methanol precipitation, oil reused | Single source |
| Dimethyl carbonate | A foam cube dissolved in about 45 seconds at room temperature (patent example) | Butanol or other non-solvent, filter, extrude | Low boiling and flammable; patent figures are self-reported |
| Isoamyl acetate, 1,3-dioxolane | Claimed to need lower distillation temperatures than limonene (patent) | Distillation | Dioxolane boils near 76°C; patent only |
| Ethyl acetate, acetone | Acetone cut foam volume by up to 55% and the residue went into lightweight concrete; ethyl acetate used with alcohol or water anti-solvents (theses) | Anti-solvent; evaporation | Partial effect for acetone; very flammable |
| Toluene, benzene, THF, methylene chloride | Good solvents, the benchmark | Distillation | Toxic; limits food-contact reuse; benzene is carcinogenic |
| Ionic liquids | Dissolve polystyrene to short chains with little energy (2025 review) | Not shown | Cost, scale and toxicity unclear |
Two things stand out. First, every route needs a way to get the polymer and the solvent apart: heat (which can degrade the polymer), a second solvent such as ethanol or heptane (which must itself be recovered), or supercritical carbon dioxide. Second, the only solvent-based polystyrene plants found at industrial scale use p-cymene or the proprietary CreaSolv formulation, not limonene. A search for Cyrene, gamma-valerolactone and ethyl lactate with foam returned no studies, which is a gap in what was found, not evidence that they fail.
The loop, and where the data stop
- Dissolving is documented. The limit of the three-part mixture is not. The amount of polystyrene that a limonene and alcohol mixture can hold, which sets the size of the plant, was reported as missing from the literature until a 2024 paper and preprint.
- Residual limonene in the recovered polymer. Limonene boils at 176°C, which makes it hard to strip, and it has a strong smell. No source gave how much stays in pellets or whether it can be sold into food-contact or indoor uses.
- Solvent ageing. The claim that limonene can be reused at least ten times comes from a review. In air limonene oxidises to carvone and limonene oxide, and oxidised limonene is a skin sensitiser. No source measured the solvent after repeated cycles.
Two further gaps do not fit in the figure. Distilling a liquid with a flash point near 45°C is a fire risk that a plant has to be licensed for. And an independent cost or life-cycle study of a limonene foam process was not found; the two recent studies on foam recycling cover a mechanical compaction route in Colombia (viable over twelve years, with a 68% lower carbon footprint) and a chemical route to styrene.
What actually gets used
Where limonene-dissolved foam has been put to work, it has been used as a solution, not as a source of pellets. Shin and Chase spun waste foam in limonene into fibres of 300 to 900 nanometres in 2005, but they were coarse and of poor quality, and later work mixes in a second solvent. The IIT Hyderabad group of Chandra Shekhar Sharma, working with a grant of about ₹35.7 lakh under the Department of Science and Technology’s waste management programme, made a non-woven fabric from foam and citrus peel extract that soaks up oil. DST says a pilot machine is ready and technology transfer is under way through a company in the IIT Hyderabad incubator. These are peel extracts, not purified limonene, and the published descriptions do not say how the solvent is recovered.
The additives matter for what is recovered. Foam waste typically carries 0.1% to 1.5% of the pentane used to expand it. The flame retardant HBCD, a persistent pollutant in insulation foam, stays in the liquid when polystyrene is precipitated, which is the principle behind CreaSolv, with a target of under 100 parts per million in the polymer. Typical packaging foam is not the HBCD-treated kind. Nothing found covered food contamination or dye removal.
The Indian setting
Thermocol has been on the regulatory map for some years. Maharashtra banned thermocol products in March 2018 and has since relaxed parts of its plastics rules. The national ban on identified single-use plastic items that took effect on 1 July 2022 names polystyrene and expanded polystyrene, including thermocol for decoration. Packaging foam is not covered by the ban, and falls under extended producer responsibility for producers and brand owners; whether and how foam is counted separately in the EPR data is not clear from what was found.
Foam recycling already exists in India by mechanical routes. A Pune recycler has processed it since the early 1970s, according to a profile, and a Ghaziabad firm lists about 720 tonnes a year, though the only throughput figure found is self-reported in a directory. Small compaction machines are advertised at about ₹2.65 lakh to ₹3.5 lakh. No resale price per kilogram or national collection volume turned up. Against that, a solvent route would have to beat compaction on cost, and there is no cost data to say whether it can.
The solvent side is better placed than it looks but is unmeasured. India grows about ten million tonnes of oranges a year, and a preprint on Nagpur orange waste found 91.9% limonene in the extracted oil. But there were no figures for how much peel oil is produced, no trade data for citrus oils, and world supply of technical-grade limonene is about 30,000 tonnes a year according to one supplier, with other estimates up to 70,000 tonnes. A foam-recycling industry would be competing for a by-product that cleaning, fragrance and coatings makers already buy.
What would fix this
- The Department of Science and Technology should fund an independent pilot of a limonene or peel-oil foam loop with a published mass balance: polystyrene recovered, solvent lost per cycle, residual solvent in the polymer, and solvent quality after ten cycles.
- The Central Pollution Control Board should publish expanded polystyrene collection and recycling tonnage separately from the EPR filings, so the size of the problem, and the share handled by compaction recyclers, can be seen.
- The Bureau of Indian Standards should set a specification for recycled polystyrene from solvent routes, with a limit for residual solvent and a statement of which uses, such as food contact, are excluded.
- The Ministry of Food Processing Industries and APEDA should publish citrus peel oil output by processor and region, and the Directorate General of Commercial Intelligence and Statistics should report citrus oils at a finer tariff line, so supply can be judged.
- The Petroleum and Explosives Safety Organisation should say how a plant that distils limonene would be licensed, given its flash point of 45–48°C, so that safety requirements are known before a pilot is built.
These are proposals from this blog, not recommendations any agency has adopted.
Corrected 25 October 2026. The limonene figure of 182–197 g per 100 g at 50°C was checked against the full 2010 paper and is unchanged; a note now explains why an earlier study by the same group gives 127 g. The “reused at least 10 times” claim was not found in the review chapter cited for it. Sources: Hattori et al., Journal of Wood Science 2010, Table 1, and Hattori’s InTech review chapter; Ozeki et al., Polymer Journal.
Sources and caveats
Noguchi, Miyashita, Inagaki and Watanabe, “A new recycling system for expanded polystyrene using a natural solvent”, Packaging Technology and Science 11, 1998 (part 1, abstract; part 2 not seen; part 3 is a life-cycle assessment by the same authors, not seen, so the “independent” study count above excludes it); a review chapter, “Recycling of expanded polystyrene using natural solvents” (InTech, doi 10.5772/59156, read in full), for the semi-industrial scale of limonene recycling, solubility tables and steam-distillation recovery; the “reused at least 10 times” statement was not found in that chapter on re-reading, so its source is unconfirmed; US patents 5629352 and 5863002 for limonene’s odour, flash point and instability; García et al., Waste Management 29, 2009 (abstract and a summary; the 0.26 g/mL figure is from a summary); Hattori et al., Journal of Wood Science 2010 (full text, Kitami Institute of Technology repository, Table 1, for p-cymene, 2-p-cymenol, limonene and its isomers at 50°C; an earlier study by the group with a different polystyrene film gives d-limonene 127 g per 100 g, InTech chapter Table 2); Shin and Chase, Polymer Bulletin 2005 (repository record and a patent citing it); a Colombian study of EPS dissolution in natural solvents (regional journal); Ozeki, Ida and Osa, Polymer Journal 2024 and 2025 (abstract read on nature.com: d-limonene is a medium solvent for atactic polystyrene, between toluene, a good solvent, and cyclohexane, a poor one); a 2024 preprint and paper on the solubility limit of the polystyrene, limonene and alcohol system; Yadav and Sharma, Polymer Bulletin 2018 and Journal of Applied Polymer Science 2019, a DST release and the India Science and Technology portal for the IIT Hyderabad work and its ₹35.7 lakh grant. CreaSolv and PolyStyreneLoop: the EU LIFE project page and Fraunhofer IVV (HBCD removal above 99.7%); Plasteurope and EUWID reports (insolvency, the German EPS Converters takeover, restart in 2022, regular operation in June 2023 at 3,000 t a year, €17 million invested; the insolvency year is inferred from the dates of those reports). Other solvents: US patents 7728047 and 7745503 (dimethyl carbonate), US 12162991 (isoamyl acetate, dioxolane) and a Polystyvert patent (p-cymene and heptane); a 2025 Springer review of green solvents (ionic liquids); student theses on ethyl acetate and acetone, a repository paper on essential oils. All are abstracts or search summaries. A 2025 gate-to-gate life-cycle and techno-economic study of densified polystyrene (Polymers) and a ScienceDirect study of chemical recycling of foam to styrene. Safety: the EU risk assessment committee opinion, safety data sheets and the Japanese NITE classification (flash point 45–48°C; air-oxidised limonene is a skin sensitiser). Supply: an IARC monograph (about 45,000 tonnes a year, early 1990s), a 2014 RSC paper (above 70,000 tonnes) and a supplier statement (about 30,000 tonnes of technical grade); these disagree and are of different grades and years. Indian policy: PIB and CPCB texts of the single-use plastics ban and the Maharashtra notification of 23 March 2018, as described in secondary guidance; the 2026 EPR timing was not confirmed. Indian recycling: a Business India profile and a trade directory listing. The 91.9% limonene in Nagpur peel oil is from a Research Square preprint. All figures were seen in abstracts, patents or search summaries, not in the papers themselves. OpenAlex was later reachable for metadata checks, but no citation counts or systematic search were run. The 1/4 polymer-by-weight figure assumes limonene density of 0.84 g/mL. No safety, process or investment advice is given.
About this article: Researched, written and edited by Umashankar Triplicane Dwarakanathan, with AI research assistance; every figure is meant to trace to the primary source cited. See the Editorial Policy for how sourcing, AI use and corrections work.
