EUTECNIQ Sciences

See every page on one mapArrow keys move through results · Enter opens · Esc closes

Technologies · Extraction · Solvent choice

Pick an extraction solvent without hiding the trade-offs

A solvent is judged on several things at once: how much of the wanted compound it dissolves, how much of an unwanted one comes with it, the heat needed to boil it off and how easily it burns. Instead of folding these into one score, this screen keeps every option that no other option beats on all of them at once. That set is the Pareto front. Every solubility on this page is a published measurement.

Published dataSolubilities from two peer-reviewed studies, obtained through NIST ThermoML; solvent properties and rules from the records cited below. The screening method is an EUTECNIQ proposal. EUTECNIQ has run no extractions and owns no equipment.

Interactive screen

Choose what matters; the front shows what survives

Pick a case, a target and the objectives. Options on the Pareto front are drawn solid with an outline; options that another one beats on every chosen objective fade. Switch to a single weighted score to see which front options a score would drop.

Interactive calculator Change the inputs and the result updates. All solubilities are published measurements; the screen only combines them.

Case
View
On the Pareto front–
Represented on the front–
All options in a table

The screen needs JavaScript. The measured values are listed in the sources below.

What the measurements show

Four results from the data

Each number below is computed on this page from the published measurements, with the screen’s default settings unless stated.

240×

Ethyl acetate separates caffeine best

At 25 °C, ethyl acetate dissolves about 240 times more caffeine than theobromine by mass, against 37 times in water, the highest ratio of the six solvents. U.S. rules also allow it as a solvent for decaffeinating coffee and tea (21 CFR 173.228).

0.034

No solvent here dissolves theobromine selectively

Theobromine is less soluble than caffeine in all six solvents at every temperature measured; the best ratio is 0.034, in water at 55 °C. On solubility alone, the data favor dissolving the caffeine away and leaving theobromine behind.

13 of 30

The best-scoring solvent may not be allowed

Dissolving caffeine with the four default objectives, 13 of 30 options from 5 solvents stay on the front. 1-Propanol is at least as good as ethanol on all four at every temperature and clearly better on three, so ethanol never reaches the front. But 1-propanol has no entry in 21 CFR 173 Subpart C, and EU rules list it only for flavorings. Switch it off and ethanol joins the front: a rule is a switch, not a weight.

1.7–2.9

Some differences are inside the error bars

Rutaecarpine dissolves 1.7 to 2.9 times more than evodiamine across the ethanol–water grid, but each published value carries 4 to 36% uncertainty. Counting those differences as ties, the ratio stops separating the mixtures: on amount, ratio and heat only the most ethanol-rich mixture tested (91% ethanol by mass) stays on the front, and adding “less ethanol” as an objective brings all six mixtures back.

Method

How the screen works

An option is one solvent at one temperature. It is on the Pareto front if no other option is at least as good on every chosen objective and better on at least one.

  • Target dissolved. From the measured mole-fraction solubility x: w = x/(1−x) × Mtarget/Msolvent × 1000, in grams per kilogram of solvent. For ethanol–water, Msolvent is the mixture’s mean molar mass.
  • Ratio. Grams of target divided by grams of the other compound, each at saturation in the same solvent and temperature.
  • Heat to boil off. Q = L / w × 1000, in megajoules per kilogram of target dissolved: the solvent’s latent heat at its normal boiling point, once, with no heating, losses or recovery. For ethanol–water, L is mass-weighted, which ignores the heat of mixing.
  • Flash point of the pure solvent (NIOSH). Ethanol content by mass, for the mixtures. Temperature as measured.
  • Ties. With the box ticked, a difference counts only when it exceeds the combined expanded uncertainty the authors reported: about 3.7 to 4.3% per value for the methylxanthines and 4 to 36% for the Evodia alkaloids. Amounts, ratios and heat are compared on a log scale.
  • Temperatures. Methylxanthine values are interpolated (log solubility against 1/T) to 15, 25, 35, 45 and 55 °C from measurements within 0.6 K of each; the Evodia values were measured at those temperatures.
  • Single score. Each chosen objective is scaled from 0 to 1 across the options shown (log scale for amounts, ratios and heat; flash point capped at 60 °C, with water counted at the cap), weighted and summed. The score keeps as many options as the front holds, so the two can be compared.

The approach follows a 2026 study that searched for ionic liquids with a Pareto archive and compared it with a single score: in that study, the score’s top candidates came from one anion family while the Pareto set spanned three (Park, Park and Lee, 2026, computed predictions). This page applies the same comparison to measured data for botanical compounds.

Limits

What the screen does not tell you

  • Each solubility was measured for a pure compound alone. In a plant, the result also depends on how much of each compound is there, on the plant matrix, on contact time and on how dissolved compounds affect each other.
  • The ratio applies where solubility limits the process, as in concentrating an extract or recrystallizing it. A dilute extraction may never reach saturation.
  • In water, caffeine and theophylline crystallized as hydrates in the measurements, so their solid form there differs from the other solvents.
  • The heat figure is latent heat only. A process also heats the solvent and the plant, loses solvent and recovers it.
  • Flash points are for pure solvents. Rules are U.S. and EU food extraction-solvent rules; a dietary supplement maker also sets its own residual-solvent limits in its specifications.
  • This is an engineering screen, not advice about any product, and it says nothing about health effects.

Inputs

Solvent facts used

Properties and rules for the six solvents
SolventLatent heat, MJ/kgBoiling point, °CFlash point, °CICH Q3CUnited States (21 CFR)European Union (Directive 2009/32/EC)
Water2.257100.0not flammablenot listedno entry in Part 173 Subpart C (no extraction-solvent tolerance)authorized, Article 2(3)
Methanol1.10164.511Class 2, 3,000 ppm§ 173.250: 50 ppm in spice oleoresins; 2.2% in hops extractAnnex I Part II: 10 mg/kg, all uses
Ethanol0.8578.413Class 3not in Part 173 Subpart C; § 184.1293 (GRAS for a named use)Annex I Part I: all uses under GMP
1-Propanol0.6997.222Class 3not in Part 173 Subpart CAnnex I Part III only: flavorings, 1 mg/kg
Ethyl acetate0.36277.1−4Class 3§ 173.228: solvent for decaffeinating coffee and teaAnnex I Part I: all uses under GMP
Acetone0.50156.1−18Class 3§ 173.210: 30 ppm in spice oleoresinsAnnex I Part I: all uses under GMP (not olive-pomace oil refining)

Latent heats: CoolProp 8.0.0 (water, methanol, ethanol, acetone); NIST Chemistry WebBook, Majer and Svoboda 1985 (1-propanol 41.44 kJ/mol at 370.3 K; ethyl acetate 31.94 kJ/mol at 350.3 K). Flash points: NIOSH Pocket Guide. ICH Q3C(R9) classes as on Extraction & purification.

Sources

Sources

  1. Zhong J, Tang N, Asadzadeh B, Yan W. Measurement and correlation of solubility of theobromine, theophylline, and caffeine in water and organic solvents at various temperatures. J Chem Eng Data. 2017;62:2570–2577. doi.org/10.1021/acs.jced.7b00065. Data from the NIST ThermoML file (SHA-256 ffc47d44…).
  2. Fan J-P, Yang X-M, Xu X-K, Xie Y-L, Zhang X-H. Solubility of rutaecarpine and evodiamine in (ethanol + water) mixed solvents at temperatures from (288.2 to 328.2) K. J Chem Thermodyn. 2015;83:85–89. doi.org/10.1016/j.jct.2014.12.004. Data from the NIST ThermoML file (SHA-256 b78efb82…).
  3. NIST Thermodynamics Research Center, ThermoML archive of published experimental data (NIST open license). trc.nist.gov/ThermoML.html. Retrieved November 13, 2026.
  4. Park J, Park J, Lee Y. Pareto-archive multiplayer Monte Carlo tree search for multi-objective inverse design of ionic liquids for CO₂ capture. Ind Eng Chem Res. 2026. doi.org/10.1021/acs.iecr.6c03910 (abstract).
  5. NIST Chemistry WebBook, SRD 69: enthalpies of vaporization of 1-propanol and ethyl acetate (Majer V, Svoboda V. Enthalpies of Vaporization of Organic Compounds. Blackwell; 1985). webbook.nist.gov. Retrieved November 13, 2026.
  6. Bell IH, Wronski J, Quoilin S, Lemort V. Pure and pseudo-pure fluid thermophysical property evaluation and the open-source thermophysical property library CoolProp. Ind Eng Chem Res. 2014;53(6):2498–2508. doi.org/10.1021/ie4033999 (CoolProp 8.0.0).
  7. NIOSH Pocket Guide to Chemical Hazards: n-propyl alcohol (flash point 72 °F), retrieved November 13, 2026; other solvents as cited on Extraction & purification.
  8. Electronic Code of Federal Regulations. 21 CFR Part 173 Subpart C (§§ 173.210, 173.228, 173.250) and § 184.1293, up to date as of October 15, 2026. ecfr.gov.
  9. Directive 2009/32/EC on extraction solvents used in foodstuffs, Annex I, text as it stood on December 31, 2020 (legislation.gov.uk); later EU amendments were not reviewed for this page. legislation.gov.uk.
  10. ICH Q3C(R9) Impurities: Guideline for Residual Solvents, as cited on Extraction & purification.

Computation: data76/prep.py and pareto_ref.py reproduce every number on this page from the two ThermoML files.

Health information notice. These statements have not been evaluated by the Food and Drug Administration. Nothing on this site is intended to diagnose, treat, cure or prevent any disease, and nothing here is medical advice.

Read the full notice

The information is for general education only. It does not replace advice, diagnosis or treatment from a qualified health professional, or any medication a clinician has prescribed. Do not start, stop or change any medicine, supplement or herbal product because of something you read here. Talk to your doctor or pharmacist first, especially if you are pregnant or breastfeeding, have a medical condition, take any medicines, are preparing for surgery, or are caring for a child. Research findings, traditional uses and safety data are reported as their sources state them; they are not recommendations or claims about any product. If you think you may have a medical emergency, call your doctor or 911 at once.

EUTECNIQ Sciences sells no supplements or health products today and receives nothing from any product or ingredient mentioned on this site; any commercial interest it later has in an ingredient will be stated on that ingredient’s page.

Ready to start a development project?

Tell us about an ingredient you want to specify, verify or license. Questions? Email [email protected]