Thermoanalytical Methods: TGA, DTA, DSC
CSIR NET: thermogravimetry and DTG, DTA and DSC curves, calcium oxalate and copper sulfate decomposition, polymer crystallinity, thermometric titration.
- 17 min
On this page · 15 sections
- Prerequisites
- The Hook: One Crystal, Three Steps
- 1. The Family of Thermal Methods
- 2. Thermogravimetric Analysis (TGA)
- 3. Differential Thermal Analysis (DTA)
- 4. Differential Scanning Calorimetry (DSC)
- 5. Thermometric (Enthalpimetric) Titrations
- 6. Comparing the Three Main Methods
- Memory Tricks
- Don’t Confuse
- Common Mistakes
- Practice
- Quick Revision Table
- Key Takeaways
- What’s Next
Prerequisites
- Electroanalytical Methods: like coulometry, thermogravimetry is a stoichiometric, calibration-light method; the arithmetic habits carry over.
- s-Block and Group 13 Compounds: s-block oxides and hydrides. Carbonates of the group 2 metals decompose at higher temperatures down the group, a trend TGA shows directly.
- Percentages, Ratios and Averages: every TGA question reduces to a percentage mass loss.
- Molar masses and simple stoichiometry; enthalpy and heat capacity at BSc level.
The Hook: One Crystal, Three Steps
1. The Family of Thermal Methods
Thermal analysis is a group of techniques in which a property of a sample is measured while its temperature is changed according to a controlled programme (usually a linear heating rate) in a controlled atmosphere.
| Technique | Measures | Plotted against T (or t) | Detects |
|---|---|---|---|
| TGA (thermogravimetric analysis) | mass m | m or % mass | anything that gains or loses mass: dehydration, decomposition, oxidation, desorption |
| DTG (derivative thermogravimetry) | dm/dT | rate of mass change | the same events, as peaks; separates overlapping steps |
| DTA (differential thermal analysis) | temperature difference ΔT between sample and inert reference | ΔT | any event with a heat effect, with or without mass change: melting, phase transitions, crystallisation, reactions |
| DSC (differential scanning calorimetry) | heat flow difference dH/dt (power) | heat flow | the same events as DTA, quantitatively: ΔH, heat capacity, glass transition |
| TMA / dilatometry | dimensions | length change | expansion, softening |
| EGA (evolved gas analysis) | gases released (by MS or IR) | ion current or absorbance | identity of what leaves at each step |
| Thermometric titration | temperature of a solution | volume of titrant | end point from the heat of reaction |
graph TD
A[Thermal event on heating] --> B{Mass change?}
B -->|Yes| C[Seen in TGA and DTG
and in DTA/DSC if heat is involved]
B -->|No| D[Invisible to TGA
seen only in DTA/DSC]
C --> C1[dehydration, decomposition,
oxidation, sublimation, evaporation]
D --> D1[melting, crystallisation,
polymorphic transition,
glass transition]2. Thermogravimetric Analysis (TGA)
Instrument and curve
A thermobalance (a sensitive microbalance with the sample pan inside a programmable furnace) records mass continuously. Purge gas is inert (N₂, Ar) or reactive (air, O₂, CO₂).
Reading a TGA curve:
- A horizontal plateau means a composition that is stable over that temperature range.
- A step down is a mass loss; its height, as a percentage of the initial mass, identifies what was lost.
- A step up is a mass gain, for example oxidation of a metal in air.
- The DTG curve (first derivative) shows each step as a peak; the peak temperature is the point of fastest mass loss, and overlapping steps that merge in TGA often separate in DTG.
Calcium oxalate monohydrate: the standard example
| Step | Approximate range (°C) | Reaction | Mass lost (% of CaC₂O₄·H₂O) | Residue (%) |
|---|---|---|---|---|
| 1 | about 100–250 | \( \text{CaC}_2\text{O}_4\cdot\text{H}_2\text{O} \rightarrow \text{CaC}_2\text{O}_4 + \text{H}_2\text{O} \) | 12.3 | 87.7 |
| 2 | about 400–500 | \( \text{CaC}_2\text{O}_4 \rightarrow \text{CaCO}_3 + \text{CO} \) | 19.2 | 68.5 |
| 3 | about 600–800 | \( \text{CaCO}_3 \rightarrow \text{CaO} + \text{CO}_2 \) | 30.1 | 38.4 |
(M = 146.11 g mol⁻¹ for CaC₂O₄·H₂O; temperature ranges shift with heating rate and atmosphere.) Because the residues are well-defined compounds stable over wide ranges, this compound is also used to check that a thermobalance is working correctly.
Copper sulfate pentahydrate: stepwise dehydration
CuSO₄·5H₂O (M = 249.68 g mol⁻¹) loses its water in three steps, 2 + 2 + 1, each water being 7.22% of the initial mass:
| Step | Product | Residue (% of initial) |
|---|---|---|
| lose 2 H₂O | CuSO₄·3H₂O | 85.6 |
| lose 2 H₂O | CuSO₄·H₂O | 71.1 |
| lose last H₂O (at a markedly higher temperature, around 250 °C) | CuSO₄ | 63.9 |
| decomposition at high temperature | CuO (+ SO₃, or SO₂ + O₂) | 31.9 |
The steps reflect different binding: in the pentahydrate four water molecules are coordinated to Cu²⁺ and the fifth is held by hydrogen bonds between sulfate and coordinated water; the last water to go is the most firmly held. TGA thus distinguishes water of different structural roles, which simple drying cannot.
Factors that shift a TGA curve
| Factor | Effect |
|---|---|
| Heating rate | faster heating shifts steps to higher apparent temperatures and can merge adjacent steps (thermal lag); slow heating gives better resolution |
| Sample mass and packing | large or tightly packed samples have temperature gradients and trap evolved gas: steps broaden and shift up |
| Atmosphere | a reversible decomposition is pushed to higher temperature by its own gaseous product: CaCO₃ decomposes at a higher temperature in CO₂ than in N₂; reactive gases (air) add oxidation steps |
| Particle size | finer powders react earlier |
| Buoyancy | the density of the hot gas falls as T rises, giving an apparent mass gain; corrected by a blank run |
| Crucible | material can react with the sample or catalyse decomposition |
So TGA temperatures are procedural, not thermodynamic constants; the mass losses are the reliable data.
Applications
- Composition of hydrates and mixtures, purity, and moisture or volatile content.
- Drying and ignition temperatures for gravimetric precipitates: the plateau shows the temperature range in which a weighing form is stable.
- Thermal stability of polymers, coordination compounds, pharmaceuticals; filler and ash content of a polymer (residue after burning off the organic matter in air).
- Kinetics of decomposition from curves at several heating rates.
- With evolved-gas analysis (TG–MS, TG–FTIR), identity of every evolved species.
3. Differential Thermal Analysis (DTA)
Sample and an inert reference (calcined α-alumina is common), in identical holders, are heated in the same furnace. Thermocouples record the temperature difference \( \Delta T = T_{\text{sample}} - T_{\text{reference}} \).
- With no thermal event, ΔT stays near zero (the baseline).
- An endothermic event makes the sample lag behind the reference: ΔT negative.
- An exothermic event makes the sample run ahead: ΔT positive.
By the common convention endotherms are drawn downward and exotherms upward (but always check the axis label, because instruments differ).
| Usually endothermic on heating | Usually exothermic on heating |
|---|---|
| melting (fusion) | crystallisation (including cold crystallisation of a polymer) |
| vaporisation, sublimation | oxidation, combustion |
| dehydration, desolvation | curing, cross-linking, polymerisation |
| most decompositions (CaCO₃ → CaO + CO₂) | some decompositions (explosives, peroxides, ammonium nitrate) |
| many solid–solid phase transitions on heating | chemisorption |
| desorption | transition of an amorphous or metastable phase to a stable one |
The peak area is proportional to the enthalpy change and the sample mass, but the proportionality constant depends on the instrument, sample geometry and thermal conductivity, so DTA is semi-quantitative unless carefully calibrated. Its strength is detecting transitions that involve no mass change at all.
4. Differential Scanning Calorimetry (DSC)
DSC measures heat flow rather than a temperature difference, so peak areas give enthalpies directly.
| Design | How it works |
|---|---|
| Power-compensated DSC | sample and reference in separate small furnaces; extra electrical power is supplied to whichever needs it to keep both at the programmed temperature; the power difference is the signal |
| Heat-flux DSC | one furnace, like DTA, but with a calibrated heat path, so ΔT is converted into heat flow |
What a DSC trace gives:
- Enthalpy of a transition: area under the peak = ΔH (J, or J g⁻¹ after dividing by mass).
- Heat capacity: the displacement of the baseline from zero is proportional to \( C_p \times \) heating rate.
- Glass transition \( T_g \): a step in the baseline (a change in heat capacity), not a peak. No latent heat is involved.
- Melting point and purity: impurities lower and broaden the melting peak (van ’t Hoff melting-point depression); analysis of the peak shape estimates purity.
- Polymorphism: different crystal forms of a drug melt at different temperatures and may interconvert.
- Curing, oxidation stability, and reaction kinetics.
Temperature and enthalpy are calibrated with pure metals of known melting point and enthalpy of fusion; indium (mp 156.6 °C) is the usual standard.
A semicrystalline polymer on one DSC run
Heating a quenched (partly amorphous) polymer typically shows, in order of rising temperature:
- a step at \( T_g \) (amorphous regions become rubbery);
- an exothermic peak of cold crystallisation (mobile chains now crystallise);
- an endothermic peak of melting at \( T_m \);
- at high temperature, decomposition (often seen also in TGA).
The degree of crystallinity of the sample as it was before heating:
$$\boxed{X_c = \frac{\Delta H_m - \Delta H_{cc}}{\Delta H_m^{\circ}}}$$where \( \Delta H_m \) is the measured melting enthalpy, \( \Delta H_{cc} \) the cold-crystallisation enthalpy (subtracted because those crystals formed during the scan), and \( \Delta H_m^{\circ} \) the melting enthalpy of a hypothetical 100% crystalline polymer (a literature value supplied with the question).
5. Thermometric (Enthalpimetric) Titrations
The titrant is added to the analyte in an insulated vessel and the temperature is recorded against volume. Until the end point, each portion reacts and releases (or absorbs) heat; after it, only the small effect of dilution remains. The end point is the break between two nearly straight lines.
The signal depends on ΔH, not on the equilibrium constant directly, so reactions with an adequate enthalpy but a small K can still give usable breaks where pH or potential changes are too gradual. Very weak acids such as boric acid are the textbook illustration. The method also works in coloured, turbid and non-aqueous media.
6. Comparing the Three Main Methods
| TGA | DTA | DSC | |
|---|---|---|---|
| Measured signal | mass | ΔT | heat flow |
| Melting of a pure solid | not seen (no mass change) | endothermic peak | endothermic peak, ΔH from area |
| Dehydration | step | endotherm | endotherm |
| Glass transition | not seen | small baseline shift | step in baseline (ΔCp) |
| Quantitative output | stoichiometry from mass loss | semi-quantitative | enthalpy, heat capacity |
| Upper temperature | high (1000 °C and above common) | high | usually lower (several hundred °C in common instruments) |
Memory Tricks
Don’t Confuse
| This | vs | That |
|---|---|---|
| TGA: mass change only | DTA/DSC: heat effects, including events with no mass change | |
| DTA: temperature difference (semi-quantitative) | DSC: heat flow (ΔH and C_p directly) | |
| DTG peak: maximum rate of mass loss | DTA peak: maximum heat effect | |
| Glass transition: baseline step (second-order-like) | melting: endothermic peak (first-order) | |
| Cold crystallisation: exotherm on heating | melting: endotherm on heating | |
| CO loss from CaC₂O₄ in N₂: endotherm | in air: exotherm (CO burns) | |
| Procedural decomposition temperature (depends on rate, atmosphere) | thermodynamic transition temperature (melting point of indium) | |
| Buoyancy: apparent mass gain with T | oxidation: real mass gain |
Common Mistakes
Practice
All questions below are original practice questions written for this page. They are not from any CSIR NET paper.
Level 1: Foundation
- Which of TGA, DTA and DSC can detect the melting of a pure organic solid, and which gives its enthalpy of fusion directly?
- What does a peak in a DTG curve represent?
- How does the glass transition appear in a DSC trace?
Answers
- DTA and DSC both detect melting (endotherm); TGA does not, since there is no mass change. DSC gives ΔH_fus directly from the peak area.
- The temperature of the maximum rate of mass loss for a step.
- As a step change in the baseline (a change in heat capacity), not as a peak.
Level 2: Part B Style
Practice Q1. A hydrate BaCl₂·xH₂O loses 14.75% of its mass on heating to constant mass at 200 °C. Taking Ba = 137.33, Cl = 35.45, H₂O = 18.02, x is:
(a) 1 (b) 2 (c) 3 (d) 4
Answer: (b) 2
Anhydrous BaCl₂ = 208.23. Fraction lost = 18.02x/(208.23 + 18.02x) = 0.1475. Solving: 18.02x(1 − 0.1475) = 0.1475 × 208.23, so 15.36x = 30.71 and x = 2.00. Check: 36.03/244.26 = 14.75%.
Practice Q2. In the TGA of CaC₂O₄·H₂O (M = 146.11), the mass loss in the second step, as a percentage of the original sample, is closest to:
(a) 12.3% (b) 19.2% (c) 21.9% (d) 30.1%
Answer: (b) 19.2%
Step 2 is loss of CO: 28.01/146.11 = 19.2%. (a) is the water step, (d) the CO₂ step, and (c) expresses the CO loss as a percentage of anhydrous CaC₂O₄ (28.01/128.10), not of the original sample.
Practice Q3. Which event on heating gives an exothermic peak in DSC?
(a) melting of indium (b) dehydration of a hydrate (c) cold crystallisation of an amorphous polymer (d) glass transition
Answer: (c)
Crystallisation releases the lattice enthalpy: exothermic. Melting and dehydration are endothermic; the glass transition is a baseline step with no latent heat.
Practice Q4. The decomposition step of CaCO₃ in a TGA run moves to a higher temperature when:
(a) the heating rate is reduced and the purge gas is N₂ (b) the purge gas is changed from N₂ to CO₂ (c) a smaller, finely powdered sample is used (d) the buoyancy correction is applied
Answer: (b)
CaCO₃ ⇌ CaO + CO₂ is reversible; CO₂ in the purge gas raises the temperature at which the CO₂ pressure over the solid exceeds that of its surroundings. (a) and (c) shift the step to lower temperature; (d) corrects mass, not temperature.
Level 3: Part C Style
Practice Q5. A mixed precipitate of calcium and magnesium oxalates is heated in a thermobalance. Assume that at 500 °C the residue is CaCO₃ + MgO, and at 900 °C it is CaO + MgO. Residue masses: 0.3600 g at 500 °C and 0.2720 g at 900 °C (CaCO₃ 100.09, MgO 40.30, CO₂ 44.01, Ca 40.08, Mg 24.31). Which statements are correct?
- The sample contained 2.00 mmol of calcium.
- The mass of MgO in the residue is about 0.160 g.
- The sample contained about 3.97 mmol of magnesium.
- The mass of calcium is greater than the mass of magnesium.
(a) 1, 2 and 3 only (b) 1 and 4 only (c) 2 and 3 only (d) 1, 2, 3 and 4
Answer: (a) 1, 2 and 3 only
Only CaCO₃ loses mass between 500 and 900 °C: 0.3600 − 0.2720 = 0.0880 g CO₂ = 0.0880/44.01 = 2.00 mmol, so 2.00 mmol Ca: statement 1 true.
CaCO₃ at 500 °C: 2.00 mmol × 100.09 = 0.2001 g; MgO = 0.3600 − 0.2001 = 0.1599 g: statement 2 true.
Mg = 0.1599/40.30 = 3.97 mmol: statement 3 true.
Mass of Ca = 2.00 × 40.08 = 80.1 mg; mass of Mg = 3.97 × 24.31 = 96.5 mg. Magnesium is heavier: statement 4 false.
Practice Q6. A 25.0 mg sample of CuSO₄·5H₂O (M = 249.68) is heated in N₂. Plateaus are found at 21.39 mg, 17.78 mg and 15.98 mg, and at high temperature a final residue of 7.96 mg. Which statements are correct?
- The first two steps each correspond to loss of two water molecules.
- The plateau at 15.98 mg is anhydrous CuSO₄.
- The final residue is CuO.
- All five water molecules are bound identically, since each step has the same mass loss per water.
(a) 1, 2 and 3 only (b) 1 and 4 only (c) 2 and 3 only (d) 1, 2, 3 and 4
Answer: (a) 1, 2 and 3 only
One H₂O is 18.02/249.68 × 25.0 = 1.80 mg.
- 25.00 → 21.39: 3.61 mg = 2 H₂O; 21.39 → 17.78: 3.61 mg = 2 H₂O: statement 1 true.
- 17.78 → 15.98: 1.80 mg = 1 H₂O, leaving CuSO₄ (159.61/249.68 × 25.0 = 15.98 mg): statement 2 true.
- CuO: 79.55/249.68 × 25.0 = 7.96 mg: statement 3 true.
- The mass per water is of course the same; what differs is the temperature at which each group leaves, which shows different binding. The last water is held most firmly: statement 4 false.
Practice Q7. A DSC scan at 10 K min⁻¹ of a quenched polymer film shows: a baseline step at 75 °C, an exothermic peak of 25 J g⁻¹ at 130 °C, and an endothermic peak of 60 J g⁻¹ at 250 °C. Take the melting enthalpy of the 100% crystalline polymer as 140 J g⁻¹. Which statements are correct?
- The glass transition temperature is about 75 °C.
- The crystallinity of the film as received is 25%.
- The crystallinity just before melting is about 43%.
- A TGA run on the same film would show mass losses at 75 °C, 130 °C and 250 °C.
(a) 1, 2 and 3 only (b) 1 and 2 only (c) 2 and 4 only (d) 1, 2, 3 and 4
Answer: (a) 1, 2 and 3 only
- The baseline step is the glass transition: true.
- As received: (60 − 25)/140 = 0.25, i.e. 25%: true.
- Just before melting, the crystals formed by cold crystallisation are present too: 60/140 = 0.43, about 43%: true.
- None of these three events changes mass, so TGA would be flat until decomposition: false.
Quick Revision Table
| Idea | Fact |
|---|---|
| TGA | mass vs T; plateaus = stable compositions; steps = mass loss |
| DTG | dm/dT; peak at fastest loss; separates overlapping steps |
| DTA | ΔT (sample − reference); endo down, exo up by usual convention; semi-quantitative |
| DSC | heat flow; area = ΔH; baseline ∝ C_p; T_g as a step |
| CaC₂O₄·H₂O | −H₂O 12.3%, −CO 19.2%, −CO₂ 30.1%; CaO 38.4% remains |
| CO step in air | exothermic in DTA (CO burns); endothermic in N₂ |
| CuSO₄·5H₂O | loses 2 + 2 + 1 H₂O (7.22% each); CuSO₄ 63.9%; CuO 31.9% |
| Shifts in TGA | faster heating, larger sample, product gas in purge → higher T |
| Buoyancy | apparent mass gain; blank correction |
| Endothermic | melting, vaporisation, dehydration, most decompositions |
| Exothermic | crystallisation, oxidation, curing, energetic decompositions |
| Crystallinity | X_c = (ΔH_m − ΔH_cc)/ΔH_m° |
| Calibration | indium mp 156.6 °C for DSC; CaC₂O₄·H₂O for TGA checks |
| Thermometric titration | T vs V; works on ΔH, so weak acids like boric acid give breaks |
Key Takeaways
What’s Next
- The whole chapter on one page: Analytical Chemistry: Quick Revision.
- Revisit Separation Methods for precipitation with organic reagents and masking, the separations that precede gravimetric weighing forms.
- The s-block compounds behind several decomposition examples (CaCO₃ → CaO): s-Block and Group 13 Compounds.
- Chapter overview: Analytical Chemistry.