If a lab working with nanoclay can afford one analytical instrument, the case for a thermogravimetric analyser is strong. A single twenty-minute run on a few milligrams of powder tells you the moisture content, the organic modifier content, the thermal stability of that modifier, and — indirectly — whether the lot has been thermally damaged before it reached you.
That is a lot of decision-relevant information from one measurement, and it costs almost nothing per sample once the instrument exists.
What the instrument does
A TGA heats a small sample on a sensitive balance under a controlled atmosphere and records mass as a function of temperature. A typical run for clay: 5–15 mg, heated from ambient to 800–1000 °C at 10 °C/min, under nitrogen or air.
The output is a mass-versus-temperature curve. The derivative curve (DTG) is where the interpretation actually happens, because it turns gradual inflections into resolvable peaks.
ASTM E1131 provides the standard framework for compositional analysis by thermogravimetry, covering the determination of highly volatile matter, medium volatile matter, combustible material, and ash content. It is applicable from room temperature to about 1000 °C and can quantify components from 1% to 100% by weight. ISO 11358 is the related international standard.
The four regions in an organoclay trace
An organoclay gives a characteristic four-step trace. Learning to read it is most of the value.
Region 1: ambient to ~150 °C — free and adsorbed water. Mass loss here is surface moisture and loosely held interlayer water. For a well-stored organoclay this is typically 1–3%. A value above about 4% suggests the material has picked up moisture in storage and needs drying before use — see Moisture in Nanoclay: Drying Protocols and Hydrolytic Degradation. This region is your moisture measurement, and it is more informative than a simple loss-on-drying because you can see the shape of the loss rather than a single endpoint.
Region 2: ~150–500 °C — organic modifier decomposition. This is the surfactant burning off (see how organoclays are produced), and it is usually the largest single step in an organoclay, typically 25–40% depending on grade and cation exchange capacity. The onset temperature is the thermal stability figure that matters for processing.
Xie and colleagues characterised this region in detail for a range of alkylammonium montmorillonites, placing decomposition onset at approximately 155 °C by TGA and detecting evolved products from around 180 °C by TGA-MS, with Hofmann elimination as the mechanism (Chemistry of Materials, 2001, 13(9), 2979–2990). They also found that alkyl chain length, the number of alkyl substituents, and the exchange ratio did not shift the onset significantly — a result worth knowing, because it means selecting a longer-chain quaternary ammonium grade will not buy the thermal headroom that intuition suggests.
The DTG curve in this region often shows two or three overlapping peaks rather than one: surfactant physically adsorbed on the external surface decomposes first, intercalated surfactant later, and strongly bound species last.
Region 3: ~500–700 °C — dehydroxylation. Structural hydroxyl groups leave the octahedral sheet as water. For montmorillonite this is typically a 4–5% loss and it is a property of the mineral, not of the modification. It is a useful internal reference: it should be essentially identical between lots of the same clay, so a shift here points to a change in the mineral itself rather than in the treatment.
Region 4: above ~700 °C — residual. Remaining mass is the inorganic ash. This is your inorganic content and the number you need for calculating true volume fraction in a composite.
Reading a lot for thermal damage
This is the highest-value routine application and the reason to run TGA on incoming material.
An organoclay that has been over-dried by the supplier, or dried too aggressively in-house, shows two changes: reduced mass loss in Region 2 (less surfactant remaining) and a lower or broadened onset temperature (the surviving surfactant is partly decomposed).
Comparing each incoming lot against a retained reference sample makes this immediately visible. A lot showing 28% organic content where the reference showed 34% has lost a fifth of its modifier, and it will disperse worse and yellow more. Discovering this on arrival is much cheaper than discovering it after compounding forty tonnes.
Atmosphere: nitrogen or air
Both, ideally, and they answer different questions.
Nitrogen gives the thermal decomposition behaviour without oxidation, which is closer to conditions inside an extruder barrel. Use nitrogen for the stability onset figure you will use for process design.
Air gives complete combustion of organics and therefore a cleaner ash figure. Use air when the number you want is inorganic content.
A common compromise is a switched atmosphere: nitrogen to about 600 °C, then air to complete combustion of any char. This gives a good decomposition profile and a reliable ash value in one run.
Heating rate
10 °C/min is the conventional default and is a reasonable choice. Two points to keep in mind:
Faster rates shift apparent onset temperatures higher, because decomposition is kinetic and there is less time at any given temperature. A 20 °C/min run will report a higher onset than a 10 °C/min run on identical material. Comparisons are only valid at matched heating rates, so fix the rate in your method and record it.
Slower rates give better resolution between overlapping steps but take longer. For separating multiple surfactant environments in Region 2, 2–5 °C/min is worthwhile.
Applying TGA to compounds, not just powder
TGA on the finished compound answers two questions that nothing else answers as directly.
Actual clay loading. Ash content after full combustion gives the inorganic loading in the part. This catches feeder errors, letdown ratio errors, and unexpected losses. It is also the correct input for any volume-fraction calculation, since the nominal formulation figure includes surfactant that is no longer there.
Surfactant survival. Comparing the Region 2 organic content of the compound (corrected for clay loading) against the incoming clay quantifies how much modifier was lost during processing. This turns a qualitative argument about whether the extruder is too hot into a number you can track against process changes.
The second measurement is underused. If yellowing or odour is a live issue, running TGA on compound produced at three different melt temperatures gives a direct, quantitative dose–response curve in an afternoon.
Practical cautions
Sample mass. Too much sample gives poor heat transfer and smeared transitions; too little gives noise. 5–15 mg is the usual window for clay.
Pan material. Alumina or platinum. Aluminium pans melt well below the temperatures needed here.
Buoyancy correction. Apparent mass changes with gas density as temperature rises. Run a blank and subtract it, particularly if you care about small mass changes.
Baseline drift. Instruments drift. Run a calibration standard periodically rather than assuming the last calibration holds.
Powder losses. Fine, low-density powder can be blown out of the pan by the purge gas at start-up. Ramp the gas flow gently and check the pan afterwards if a result looks implausibly low.
TGA pairs naturally with XRD d-spacing measurement and FTIR analysis of the surface modification: XRD tells you whether the surfactant is in the gallery, FTIR tells you what it is, and TGA tells you how much of it is there and how much heat it will survive.