Choosing a mercaptan chain transfer agent: why mercaptan sulfur is the assay that matters

In short

• TDM is not one molecule. It is a family of branched C₁₂ thiol isomers, which is why its purity figure reads 98.50% min. and its boiling range covers 24 degrees and why a purity figure alone tells you less here than on a single-compound product.
• Mercaptan sulfur is the real assay, because it measures the –SH group that does the work. Thiol that has oxidised to disulfide is still the same carbon skeleton and transfers nothing.
• The published window, 15.50–16.00%, brackets the theoretical value for a C₁₂ mercaptan almost exactly: 15.84%. That is a checkable number, and a lot that sits at the bottom of the window is telling you something.
• TDM and 1-dodecanethiol have the same molecular formula and the same sulfur content, and behave differently. Structure, not composition, is what separates them.
• Chain transfer is a per-molecule event, so dose by moles. At equal mass, 1-octanethiol delivers about 38% more thiol groups than a C₁₂.
• TDM has no single transfer constant. The published effective values for styrene emulsion polymerisation at 70 °C are 0.513, 0.317 and 1.509 — one compound, nearly five-fold apart, because in emulsion the figure measures transport as much as chemistry.

A chain transfer agent does one thing: it ends a growing polymer chain and starts a new one, so that the average molecular weight lands where you want it instead of where the kinetics would put it. Everything that follows — melt flow, mechanical properties, processability — is downstream of how many transfer events happen and how fast.

Which makes the buying question narrower than the product names suggest. You are not buying a liquid. You are buying –SH groups, at a certain reactivity, at a certain price per mole.

What the site publishes

And for TDM the full specification:

The dashes are not omissions, they are the state of the sheet. Four of the seven lines name the method they are measured by; mercaptan sulfur and refractive index do not, and mercaptan sulfur is the line this whole note says matters most. That is the first thing to ask a supplier for — this one included.

TDM is not one molecule, and that changes what "purity" means

Tertiary dodecyl mercaptan is made by adding hydrogen sulfide across a branched C₁₂ olefin — propylene tetramer or a similar oligomer — which is itself a mixture of isomers. The product inherits that: commercial TDM is a family of branched C₁₂ thiols, not a single compound with impurities in it.

The boiling range says the same thing out loud. 227–251 °C is a 24-degree spread. A pure compound boils at a point, not across a quarter of a hundred degrees; a 24 °C range is what a distribution of branched C₁₂ isomers looks like when you distil it. The specification sheet has been describing an isomer family all along, in a line most buyers skim past.

So Purity: 98.50% min. cannot mean "98.5% of one substance". It means 98.5% of the material is in the intended family. That is a perfectly good specification and it is not the same claim a purity figure makes on a single-compound product, and reading it as though it were leads to the wrong comparison — TDM at 98.5% against 1-dodecanethiol at 99%+ is not a quality gap, it is two different kinds of number.

This is exactly why mercaptan sulfur exists on the sheet. An isomer family cannot be assayed usefully by asking how much of it is one peak. It can be assayed by asking how much thiol functionality is present, across every isomer at once.

Mercaptan sulfur, and a number you can check

Mercaptan sulfur reports the sulfur present as –SH, as a mass percentage. For a C₁₂ mercaptan, C₁₂H₂₆S, the arithmetic is fixed:

molecular weight        202.40
sulfur                   32.06
32.06 / 202.40    =     15.84%

The published window is 15.50–16.00%. Theory sits at 15.84%, just above the middle. That is what a well-set specification looks like: tight enough that material outside it is not a C₁₂ mercaptan in the expected proportion, wide enough to allow real analysis.

It also gives you a reading. A lot at 15.9% is close to theoretical — nearly all of the mass is doing chain transfer. A lot at 15.5% has about 2% less thiol functionality per kilogram than one at 15.84%, and the difference has to be something: unreacted olefin, heavier oligomer, or thiol that is no longer thiol.

That last case is the one to watch. Thiols oxidise to disulfides on exposure to air. A disulfide made from two TDM molecules has the same carbon skeleton, similar volatility, and will very likely still pass a chromatographic purity test — and it does not transfer a chain. Mercaptan sulfur catches it because the S is no longer present as –SH. A purity figure may not.

The practical implication is about handling as much as buying: an opened drum with headspace, or a hot ISO tank standing for weeks, drifts in one direction only.

Same formula, different molecule

TDM and 1-dodecanethiol are both C₁₂H₂₆S. Same molecular weight, same theoretical mercaptan sulfur, same sulfur per kilogram. On any composition-based comparison they are interchangeable.

They are not interchangeable.

The difference is where the –SH sits. In TDM it is on a tertiary carbon, crowded by branches; in 1-dodecanethiol it is a primary thiol at the end of an unbranched chain. Hydrogen abstraction from those two environments does not proceed at the same rate, and a chain transfer agent is nothing but a controlled hydrogen abstraction. Two agents with identical sulfur content therefore give different molecular weight distributions at the same loading.

The site's own application split reflects this rather than contradicting it: TDM is described for ABS and SB rubber, 1-dodecanethiol for SB latex, adhesives and coatings. Substituting one for the other because the analysis matches is the most expensive mistake available on this product family, and the analysis will not warn you, because on paper they agree.

Dose by moles, not by mass

Chain transfer happens one molecule at a time. What controls the outcome is how many thiol groups you added, not how many kilograms.

For the two C₁₂ agents that distinction does not bite, because they weigh the same per molecule. Bring 1-octanethiol into the comparison and it does:

C12H26S   202.40 g/mol   →   4.94 mol of –SH per kg
C8H18S    146.29 g/mol   →   6.84 mol of –SH per kg
                              6.84 / 4.94  =  1.38

At equal mass, octanethiol supplies about 38% more thiol groups. Swapping a C₁₂ for the C₈ at the same weight loading does not fine-tune a molecular weight; it overshoots by nearly forty per cent before the difference in reactivity is even considered.

That is arithmetic on two molecular formulas, and it is worth doing on paper before it is done in a reactor.

The transfer constant, and why asking for one number is the wrong question

"Chain transfer constant for dodecyl mercaptan in styrene" is a question people type into a search engine, and the honest answer is that in an emulsion process it does not have a single value.

The constant is a ratio, C_X = k_fX/k_p — how fast a growing radical abstracts the thiol hydrogen against how fast it adds another monomer. In bulk or solution that is a chemical property of two molecules. In emulsion it is not, because the mercaptan has to get from where it is to where the radical is, and the published figures are effective constants with the transport folded in. Minari and co-workers state the dependency plainly: the mass transfer of a chain transfer agent between phases "is affected by the stirring rate, the monomer-droplets size, the particle size, and the CTA diffusivity."

Their Table I collects what the literature reports for the emulsion polymerisation of styrene at 70 °C:

TDM's row is three numbers. Same compound, same monomer, same temperature, three published measurements spanning nearly five-fold. That is the most useful line on this page: anyone who quotes you a single transfer constant for TDM is quoting one laboratory's recipe, and a dosing calculation built on it inherits that laboratory's stirrer.

And two entries in the table are not about reactivity at all. n-Decyl reads 4.409 and n-dodecyl 0.699 — both straight-chain primary thiols, two carbons apart, differing six-fold. A primary –SH does not lose six-sevenths of its reactivity because the tail grew. What changes is where the molecule sits. Dietrich, Pryor and Wu measured the same trend in 1988 and tied it to exactly that: the apparent constants fell as the carbon chain lengthened, following the distribution coefficients of those mercaptans between the water phase and the monomer droplets.

What it means for a batch

Minari and co-workers give the consequence in one line: an effective C_X near unity holds the monomer-to-CTA ratio roughly constant through the run, and values above unity make molecular weight climb as the batch proceeds.

Everything in that table from C11 up sits at or below unity, TDM's 1.509 outlier aside, and everything shorter sits above it. A transfer agent consumed more slowly than the monomer becomes relatively more concentrated as the run goes on, so the chains formed late are shorter than the ones formed early and the distribution broadens. If a flat molecular weight through the batch is what the product needs, the lever is the feed profile — which is the same conclusion as "dose by moles", reached from the other direction.

What this means for buying it

We do not print a transfer constant on the product page, and we will not quote one on request as though it were part of the specification. It is not a property of the lot in the drum; it is a property of your reactor and our material together, and publishing a number for it would be inventing a commitment nobody can test on receipt.

What is worth asking a supplier, and what we answer for, is consistency: the same isomer family from the same producer, lot to lot. The table above is the argument for it. If swapping between two nominally identical C₁₂ mercaptans can move the effective constant further than your dose adjustment does, then the thing that keeps a process where you set it is not the number on a data sheet — it is whether the next drum is the same material as the last one.

Volatility and odour are a plant question, not a nuisance

The shorter chain that gives octanethiol its higher thiol density also gives it a lower boiling point, which the site describes as high volatility and reactivity. Two consequences follow, and both cost money rather than quality.

Losses. A more volatile agent leaves with the vapour — during charging, during any stripping or devolatilisation step, and through the vent. Material that leaves is material that did not transfer a chain, so the dose that reaches the polymer is lower than the dose that was weighed.

Odour. Mercaptans are detectable at extraordinarily low concentrations; that property is why related compounds are used deliberately as gas odorants. A more volatile mercaptan puts more of itself into the air above the vessel. Whether that is manageable depends on the plant's vapour handling, not on the material, and it is a question to settle before a trial rather than during one.

Colour, density and boiling range at goods-in

Four lines on the TDM sheet are checkable on site in minutes.

Colour, Pt/Co ≤15 is a tight limit for a hydrocarbon liquid, and colour developing beyond it is usually oxidation — the same process that converts thiol to disulfide. Colour is the free early warning for the mercaptan sulfur number.

Specific gravity 0.860 ± 0.010 (GB/T 4472-2011) and refractive index 1.4670 ± 0.0020 will not resolve 98.5% from 98.9%, but they will catch a wrong drum, a mislabel or dilution, which are the failures that actually occur.

The boiling range, 227–251 °C (GB/T 615-2006), is the line that tests the isomer distribution most directly, and it is the one buyers check least. Branching lowers a boiling point, so a straight-chain C₁₂ thiol boils above this window rather than inside it. More usefully day to day: a lot whose curve has drifted at either end is telling you the isomer mix has moved, and that is exactly what a single-point density or refractive-index reading averages away.

What to ask for on the certificate

Mercaptan sulfur as a measured number, not "conforms". The whole reading above depends on where in the 15.50–16.00% window the lot sits.

The method for mercaptan sulfur. The purity method is on the sheet above — GB/T 9722-2006 — and on an isomer family it is still worth asking what that figure is integrating. Mercaptan sulfur names no method at all, here or on most sheets in this trade, and it is the number the rest of this note rests on.

Disulfide content, if the producer measures it. Not everyone does. A producer who does is telling you they understand what ages in this product.

Date of manufacture and storage conditions, because this is a material whose specification drifts in a known direction with air and time.

Related

• Mercaptan chain transfer agents — the family overview
• tert-Dodecyl mercaptan (TDM), CAS 25103-58-6 — full specification
• Tertiary Dodecyl Mercaptan (TDM) — packaging and supply

Sources. CAS numbers, the three products' names and described applications, and TDM's full specification — appearance, purity 98.50% min. (GB/T 9722-2006), mercaptan sulfur 15.50–16.00%, colour Pt/Co 15 max. (GB/T 605-2006), boiling range 227–251 °C (GB/T 615-2006), specific gravity 0.860 ± 0.010 (GB/T 4472-2011) and refractive index 1.4670 ± 0.0020 — are quoted from the product pages linked above, which carry the specification the producer supplies. Mercaptan sulfur and refractive index are published without a method; the dashes in the table are that fact, not an abbreviation. The molecular weights, the 15.84% theoretical sulfur figure and the 38% mole comparison are arithmetic on the formulas the compound names define (C₁₂H₂₆S and C₈H₁₈S). The production route from a branched C₁₂ olefin, thiol oxidation to disulfide, and the difference between primary and tertiary hydrogen abstraction describe general chemistry, not measurements on any particular lot.

The transfer-constant table and the two quoted statements about it are from R. J. Minari, J. R. Vega, M. González-Sierra, G. R. Meira and L. M. Gugliotta, "Emulsion Polymerization of Styrene with Iso-Octyl-3-Mercaptopropionate as Chain Transfer Agent", Journal of Applied Polymer Science 109, 3944–3952 (2008), doi:10.1002/app.28490 — its Table I, which collects values from the earlier literature rather than measuring them. The 1988 measurement of the chain-length trend is B. K. Dietrich, W. A. Pryor and S. J. Wu, "Chain transfer constants of mercaptans in the emulsion polymerization of styrene", Journal of Applied Polymer Science 36(5), 1129–1141 (1988), doi:10.1002/app.1988.070360514, quoted from its abstract. No transfer constant on this page was measured by us, and none of them describes any lot we supply.