Electronic and Optoelectronic Grade MTHPA

Why Optical Parts Need a Different Conversation

MTHPA is the same molecule in every grade, so the specification that separates an optoelectronic grade from a general casting grade is not a different chemical – it is two numbers held much tighter, plus a set of numbers that are not on the standard certificate at all.

Two grades in the range exist specifically for this work: one at Pt-Co 20 with a 41.0% anhydride group minimum, and a custom grade at Pt-Co 10 with free acid at 0.1 or 0.2%. This page is about what those two numbers buy, where they are wasted, and what to ask for that the certificate does not show.

What “Optoelectronic Grade” Fixes

Item General casting grade Optoelectronic grade
Appearance Pale yellow, clear Colourless, clear
Colour, Pt-Co No number, described as pale yellow 20 max, or 10 max on the custom grade
Free acid 0.5% max 0.5% max on the Pt-Co 20 grade, 0.1 or 0.2% max on the custom grade
Anhydride group content 40.5% min 41.0% min
Viscosity at 25 deg C 30 to 50 mPa.s 30 to 50 mPa.s
Crystallisation point -15 or 0 deg C max 0 deg C max

The pattern is worth naming: the optical grade tightens the numbers that describe what the material contains – anhydride that is intact, colour that is low – and leaves the numbers that describe handling alone. Viscosity and crystallisation point are the same as the general grade, which is the point. An optical customer is not buying a thinner liquid; they are buying a cleaner one.

Colour: Why Pt-Co 20 and Not 50

Colour in the drum and colour in the cured part are related but not identical, and the relationship is what the Pt-Co limit is really buying.

Three mechanisms move cured colour, and only one of them is in the raw material:

  • Colour carried in. A yellow liquid gives a yellow casting. This is what Pt-Co measures directly.
  • Colour developed during cure. Iron, and the oxidation of the retained ring double bond, both darken the part during the cure cycle. A high-iron lot darkens more than its Pt-Co reading predicts.
  • Colour developed in service. Heat and UV continue to move the colour of an unsaturated anhydride system. MTHPA keeps one double bond; methylhexahydrophthalic anhydride does not.

That third mechanism is the reason a light-coloured MTHPA is not automatically the right choice for a long-life optical part. Where the part sits behind a lens in direct sunlight or runs hot for years, the hydrogenated anhydride family is the safer comparison. For the liquid pair, the trade between MTHPA and the hydrogenated product is set out in MTHPA vs MHHPA, and the hydrogenated product itself is on the MHHPA page; the solid end of the same family is compared in THPA, 3,4,5,6-THPA and HHPA. MTHPA’s optoelectronic grades are chosen where the colour at the time of casting, the free acid level and the processing window of a liquid system are the binding requirements.

Practical reading of the number: Pt-Co 50 is a light yellow and is visible in a thin section; Pt-Co 20 is nearly water-white and is invisible in most parts below a few millimetres; Pt-Co 10 is water-white and is bought when the part is the optical path itself. Between 20 and 10 the step is small in visual terms and large in yield terms, which is why the 10 grade is custom rather than catalogue.

Free Acid: The Hydrolysis Limit That Decides Two Things

Free acid matters twice over in electronic work, once chemically and once electrically.

Chemically, hydrolysed anhydride is anhydride that will not react. The arithmetic is simple: one kilogram of water per tonne – 0.1% – destroys about 9.2 kg of MTHPA and forms about 10.2 kg of the diacid, which reads as about 1.0 point of free acid. The same calculation for the smaller THPA molecule gives 0.94, which is the number usually quoted for the family; MTHPA is slightly harsher because its molecular weight is higher.

Electrically, the diacid is a mobile, ionic species in a system that is supposed to be an insulator. In a casting the effect is small; in a thin encapsulation over a fine-pitch device, ionic contamination is measured in micrograms and the tolerance for ionisable material is set by the device, not by the resin.

Free acid limit Equivalent water pick-up What has to be true at the customer site
0.5% max About 0.5 kg per tonne Closed drums, ordinary handling
0.2% max About 0.2 kg per tonne Closed drums, dry store, short openings, lined drums
0.1% max About 0.1 kg per tonne Blanket after partial use, small working containers, one owner for the drum

This is the single most common way an optoelectronic grade degrades in use. A 0.1% drum opened weekly in a humid plant will not test at 0.1% at year end; the certificate describes the drum on the day it was sealed. The conversion between free acid and acid value, and the titration methods, are in free acid, colour, iron and moisture.

Volatiles, Voids and the Vacuum Window

All grades in the range carry a volatile limit of 1.5% max, which is a casting limit rather than an optical one. In encapsulation the practical consequence is different: a volatile fraction that is harmless in a 20 mm casting becomes a visible bubble or a haze in a 0.5 mm layer.

Two things follow. First, for thin encapsulations the question is not only the volatile number but the test method behind it, because a result depends on temperature and time and two suppliers can both report 1.5% on different methods. Second, the vacuum and dispense window is part of the specification in practice: a liquid anhydride holds dissolved air, and a shorter degas cycle than the material needs produces voids that no colour specification will catch.

The Three Numbers That Are Not on the Certificate

Ionic content is not a standard line on an MTHPA certificate of analysis. For optoelectronic work it is often the number that decides the part, so it has to be asked for by name:

  • Chloride, and hydrolyzable chloride in particular. The relevant species for metal corrosion and for ionic conductivity, and normally reported separately from total halide. It is not a line on the routine certificate, but it is measured on request, and the figure for the current production is below 2 ppm. Treat that as a per-lot result rather than a catalogue value – a device customer still needs incoming data for their own qualification – but at 2 ppm chloride is unlikely to be the limiting ion in the system.
  • Sodium and potassium. Mobile cations; relevant where the encapsulation sits over an aluminium or gold metallisation.
  • Total ionisable content or conductivity extract. The aggregate number some device customers specify instead of individual ions.

None of these are published as catalogue values, and none of them can be inferred from the colour. Where a figure is available – chloride, at below 2 ppm on the current production – it can be quoted; where it is not, the honest answer is that the material is not qualified for the device rather than that it is.

Formulation Notes for Optical Encapsulation

The MTHPA side of an optical formulation is the easy half; the resin and the cure schedule decide the rest. Typical systems pair a liquid cycloaliphatic or hydrogenated bisphenol epoxy with MTHPA and a latent or amine accelerator, cured in the 120 to 160 deg C band with a post-cure.

Three formulation points recur:

  • Dose from the certificate, not the family. At 41.0% anhydride group content the effective equivalent weight is 175.7, not 166.18, so a dose calculated from the theoretical figure is about 5% light. The calculation and worked examples are in anhydride curing agent dosage.
  • The accelerator can cost more colour than the anhydride. Amine accelerators yellow under heat; a system optimised for colour often runs a lower accelerator level and a longer cure rather than a fast cycle.
  • Refractive index matching is a formulation issue, not a grade issue. No grade of MTHPA is sold against a refractive index specification, and the cured index is set by the whole formulation.

Where the Grade Does Not Matter

Buying Pt-Co 10 for a part nobody sees is a common and expensive habit. The grade choice should follow the failure mode that the application actually has:

  • Hidden potting with no optical path: a general pale yellow grade is correct, and the free acid limit is the number that matters.
  • Visible but not optical: Pt-Co 50 is the first grade with a real colour number, and it usually ends the argument.
  • Optical path, short life, or a diffuser: Pt-Co 20.
  • Optical path with a tight hydrolysis limit or a long qualification: Pt-Co 10 with free acid at 0.1 or 0.2%.

The middle two cases are where most money is spent and where most of the savings are, because the step from 50 to 20 costs less than the step from 20 to 10. The grade-by-grade decision table is in MTHPA grades.

Incoming Inspection for an Optoelectronic Lot

  1. Open the drum and look before stirring. A clear liquid with no sediment or haze at the bottom; anything else goes back to the supplier before a sample is taken.
  2. Check the drum temperature and the season. A 0 deg C grade that has travelled through a cold night may be partly crystallised; melt and homogenise before sampling, and record that the sample is from a homogenised drum.
  3. Colour, neat, against the certificate. A Pt-Co cell is quick and it catches the drum that picked up iron in a steel pump.
  4. Free acid, if the lot is a 0.1 or 0.2% grade. This is the number that moves first in storage, and the number that a previous lot’s certificate cannot predict.
  5. Anhydride group content or acid value, once per lot. Confirms the lot is fully anhydride rather than partly hydrolysed in transit.
  6. Ask for the ion data of the lot, not of the product. Device qualifications are lot-based, and a product-level statement is not a substitute.

FAQ

Is the optoelectronic grade a different chemical?

No. It is the same CAS 11070-44-3 and the same formula C9H10O3, held to a tighter colour, a tighter free acid and a higher anhydride group minimum. The difference is in the specification and in the packing, not in the molecule.

Can I use the Pt-Co 20 grade for a general casting?

Yes, and it works, but it is not a cost-effective choice unless the colour is needed. The free acid difference between the general and the Pt-Co 20 grade is zero – both are 0.5% max – so a general casting gains nothing from the optical grade except the colour.

Why does my free acid rise even though the drum was closed?

Because the headspace in a partly used drum is air, and the drum breathes with every temperature cycle. A drum that is half full and closed still exchanges moisture with the headspace, and the headspace with the liquid surface. Blanketing with dry nitrogen after partial use is the only reliable answer.

Does a light-coloured grade mean a lower iron content?

Not necessarily. Colour is a visual result; iron is a contaminant that may be present in a form that does not yet show as colour. For a device application, ask for iron separately and specify lined drums.

What is the main risk in switching from MTHPA to MHHPA for an optical part?

The other direction of the same trade. MHHPA has no ring double bond and better colour stability in service, with a different viscosity and crystallisation behaviour, so the process has to be re-qualified even though the chemistry is close. The dose also changes, because the equivalent weight changes.

Related Reading

Scroll to Top