Tapioca binder in coconut briquettes: function, ratio, and what it means for burning

By Mohamad Sinno, Charcoal Expert · 27 August 2026 · Reviewed by Budi Dharmawan, Head of QC, 4 September 2026

Every page that answers this question gives you a percentage. Almost none of them tell you what the percentage is a percentage of — and once you ask that, the numbers stop agreeing with each other in an interesting way.

The short answer

Tapioca starch is the binder. Carbonised coconut shell arrives from the kiln as powder and irregular pieces with no cohesion of its own, and a shaped briquette needs something to hold those particles together. Starch mixed with water gelatinises; as the briquette dries it retrogrades and sets into a rigid structure that locks the particles where they are. That mechanism is well understood and is not in dispute anywhere.

Our typical binder share is 1.5–3% of the mix — the figure the factory confirms, published as a range and with the word typical, which is how it was given to us. It is lower than the number on most pages about binders, and the rest of this page is about why that comparison is harder than it looks rather than about which number wins.

Ratio, cohesion and burn effects

Binder is added at mixing and its consequences are checked twice on the way out. The factory's checkpoint list puts viscosity, density, temperature and moisture at extrusion, then viscosity, density and moisture again at shape forming — the two points where a mix that is too wet, too dry, too thin or too stiff is caught before it has become a briquette.

Be precise about the forming step, because the published conventions are not describing our machine. Our line extrudes; most briquetting literature and most supplier explainers describe compression in a mold — a different mechanism with a different relationship between binder dose and cohesion. A source specifying a pressure in tons per square inch is telling you which of the two it means.

On burning, we publish nothing of our own — no burn time, no calorific value, no drop or shatter figure — because we hold no measurement of ours to publish, and a number we cannot stand behind is worse than a gap. What we can describe is what gets checked. At packing a batch faces a burn-behaviour check, a drop test, a crack test, a density check and an odour check, with a smoke test on a shisha head; a random carton audit then repeats weight, moisture, ash, cracks, drop and density. Those are gates, not results — a batch passes or it does not ship, and neither outcome produces a figure for a specification sheet.

Published convention does connect dose to burning. One Indonesian supplier explainer, published 6 January 2026, puts binder above roughly ten per cent as creating excessive ash and reducing calorific value without meaningful strength improvement, and three to four per cent as producing weak briquettes prone to breakage. A dated statement of where one practitioner puts the edges — not our measurement, and not something we have tested.

Why our number and the published numbers do not line up

Set our 1.5–3% next to a published 5–8% and the obvious reading is that somebody is wrong. The more useful reading is that the two figures are built on different foundations, and that none of the published ones say so. Here are the axes, side by side.

Right-hand column: supplier binder-science explainer, published 6 January 2026, updated 31 January 2026, viewed 27 August 2026. Left-hand column: our own confirmed production values.

Three of those rows make the comparison unsafe on their own. A percentage of the mix and a percentage of total briquette mass have different denominators, and no published source we found offers a conversion. Native and modified starches are dosed differently by the same author on the same page. And a dose calibrated for compression in a mold has no reason to transfer to an extruder.

One detail should make anyone cautious about the whole genre: that single source gives the standard as 5–8% in one section and 6–7% in another, a few hundred words apart, without noticing. If one careful page cannot hold one number steady, a figure averaged from a dozen pages is not a consensus — it is a pile.

So we are not going to reconcile them. Ours is ours; theirs are dated conventions with their own stated bases. Nor are we claiming the lower figure is better. Less binder is not automatically a virtue and more is not automatically padding. What we can say is that our figure comes with its basis attached, which is the part that was missing.

Reading binder on a spec sheet

Here is the thing no page in this space seems to say, and it is the practical one. Our laboratory reports are run on the finished briquette, which includes the binder — not on the carbonised shell alone. It follows directly that part of the measured volatile matter comes from the binder rather than from the shell.

That sentence changes how a certificate reads. Volatile matter is one of the four figures buyers compare hardest, and a briquette carrying more binder will, all else equal, carry more volatile matter for a reason that has nothing to do with how well the shell was carbonised. Rank two suppliers on a two-point volatile-matter difference and you may be ranking their binder recipes. Ask what the report was run on — of us or of anyone else.

What it does not give you is a binder percentage. Volatile matter is not a binder assay and cannot be back-solved into one without knowing the shell's own volatile contribution, which the certificate does not report. Binder share is readable as a direction, not as a number. How the four measured figures actually behave is set out at fixed carbon and volatile matter; the certificate is walked line by line in how to read a COA; and the specifications page carries the ranges those reports produce.

Binder is one stage of eight, and the two either side of it are now written up as well. What the shell went through before it reached the mixer is at carbonization, and what happens to the water the binder brought with it is at drying and moisture control. The stage-by-stage view of all eight is on the production process page.

Binder and odour: what we do not claim

Binder and odour are often mentioned in the same breath, so plainly: the factory runs an odour check at packing alongside the smoke test, and a batch that fails does not ship. That is a gate we can describe. What we will not do is attribute odour to binder in either direction — we have no measurement linking the two, and found no published mechanism worth reporting as convention. No claim that our dose improves odour, none that anyone else's causes it. An odour check exists; a binder–odour finding does not.

Questions buyers ask