Differential scanning calorimetry (DSC) is a technique in thermal analysis that measures the difference in heat-flow rate into a sample and a reference while both undergo a controlled temperature program. The program may involve heating, cooling, constant-temperature holds, or combinations of these steps. DSC is used to investigate melting, crystallization, glass transitions, chemical reactions, and other processes that absorb or release heat. Its principal outputs are heat-flow curves from which transition temperatures, changes in enthalpy, and heat capacity can be determined.
Measurement principle
DSC compares the thermal response of a sample with that of a reference, commonly an empty pan similar to the sample pan. In measurements on solutions, particularly biological solutions, the reference may instead contain a matching buffer or solvent. The differential arrangement reduces the contribution of the apparatus and helps distinguish the sample’s behavior from the instrumental background.
When a sample is heated without undergoing a transition or reaction, the heat required depends mainly on its heat capacity. Melting or another endothermic process requires additional heat, whereas crystallization or an exothermic reaction releases heat. These changes produce features in a plot of differential heat-flow rate against temperature or time.
For an idealized experiment at constant pressure, the sample’s heat-flow rate can be expressed as
where is sample mass, is mass-specific heat capacity at constant pressure, is the scanning rate, and represents heat absorbed or released by transformations or reactions.
An actual instrument measures a differential signal affected by the sample, reference, containers, and thermal response of the apparatus. Quantitative interpretation therefore requires appropriate corrections and calibration.
Under suitable constant-pressure conditions, heat exchanged during a process corresponds to its enthalpy change. DSC is nevertheless not automatically an equilibrium measurement: a programmed temperature ramp can drive a material away from thermodynamic equilibrium, and the observed response may depend strongly on time and scanning rate.
Instrument designs
Heat-flux DSC
In a heat-flux instrument, the sample and reference occupy separate positions within a common furnace and are connected to it through defined thermal paths. Differences in their heat uptake produce a small temperature difference across the sensing arrangement.
A calibrated relationship converts this temperature difference into differential heat-flow rate. The conversion depends on the thermal properties and geometry of the instrument; modern instruments may apply corrections for thermal resistance, heat storage, and dynamic response.
Power-compensation DSC
In a power-compensation instrument, the sample and reference have separately controlled heaters. A control system supplies different amounts of power to keep their temperatures closely matched as they follow the programmed temperature.
The difference in supplied power, after correction and calibration, provides the calorimetric signal.
Both designs measure differential thermal behavior. Their practical performance depends on sensor construction, control systems, temperature range, and calibration rather than on design category alone.
Relationship to differential thermal analysis
Differential thermal analysis (DTA) records the temperature difference between a sample and a reference during a temperature program. DSC is designed to provide a quantitatively calibrated heat-flow measurement. Heat-flux DSC also detects temperature differences, so the distinction is not simply whether a temperature difference is measured, but how the instrument converts that measurement into calorimetric information.
Reading a DSC curve
A DSC curve usually displays heat-flow rate, often normalized by sample mass, against temperature or time. Instruments and software use different plotting conventions: an endothermic event may point upward or downward. The indicated sign convention must therefore be checked before interpreting a curve.
Melting and crystallization
Melting generally produces an endothermic peak associated with a phase transition from a crystalline solid to a liquid. Crystallization generally produces an exothermic peak as an ordered solid forms.
A transition can be characterized by several temperatures:
- Extrapolated onset temperature: commonly obtained from the intersection of an extrapolated pre-transition baseline and a tangent to the leading edge of the peak.
- Peak temperature: the temperature at the maximum deviation associated with the event.
- Endset temperature: an estimate of where the event finishes and the curve returns toward its post-transition baseline.
These quantities are not interchangeable. Peak temperature is particularly sensitive to scanning rate, sample size, heat transfer, and transformation kinetics. For a pure crystalline calibration material measured under suitable conditions, the extrapolated melting onset is commonly used for temperature calibration.
The enthalpy associated with a peak is obtained by integrating the baseline-corrected heat-flow signal:
where is the selected baseline. At a constant nonzero scanning rate,
If the signal is already normalized by mass, integration yields specific enthalpy rather than total enthalpy. Baseline selection can materially affect the result, particularly for broad or overlapping events.
Glass transition
The glass transition is a change in the molecular mobility of an amorphous material. In DSC it usually appears as a step in the baseline because the material’s heat capacity changes, rather than as a sharp latent-heat peak.
A reported glass-transition temperature, , is operationally defined: it may refer to the onset, midpoint, or another specified point on the step. It depends on scanning rate, prior cooling, aging, and sample composition. An enthalpy-recovery overshoot may accompany the step after physical aging.
The glass transition is a kinetic phenomenon and should not be interpreted as equivalent to equilibrium melting.
Reactions and other thermal events
A chemical reaction can generate an exothermic or endothermic feature. Examples include curing, oxidation, and some decomposition processes. Evaporation, loss of absorbed moisture, and solid-state rearrangements can also produce signals.
A peak’s shape or direction alone does not uniquely identify its cause. Different processes may overlap, and some may partly cancel one another.
Experimental procedure and calibration
Conventional DSC commonly uses milligram-scale samples in small metal or ceramic containers. Suitable sample size and container type depend on the instrument, material, and process being studied. Good thermal contact and reproducible sample placement help reduce thermal lag.
Important experimental variables include:
- Temperature program: heating and cooling rates, hold times, and previous cycles.
- Atmosphere: inert or reactive gas composition, flow rate, and pressure.
- Container: open, covered, hermetically sealed, or pressure-rated.
- Sample condition: mass, particle size, moisture, composition, and previous thermal treatment.
- Reference and baseline: container matching, reference composition, and instrumental background.
Container choice can alter the process itself. An open pan permits volatile material to escape, whereas a sealed pan can retain it and allow pressure to develop. Heating volatile or reactive materials in sealed containers also requires attention to instrument and container pressure limits.
Temperature calibration commonly uses materials with well-characterized melting transitions, such as indium. Heat-flow calibration uses known transition enthalpies or other appropriate standards. Heat-capacity measurements often use sapphire as a reference material.
Calibration conditions should be relevant to the measurement conditions because scanning rate, atmosphere, container configuration, and temperature range can influence the response. Measurement uncertainty includes contributions from mass measurement, calibration, repeatability, baseline selection, and data processing.
A first heating scan records both material properties and aspects of the specimen’s prior processing or storage. A subsequent cooling and reheating cycle can impose a more controlled thermal history. However, reheating does not recreate the original material if the first cycle caused irreversible reaction, degradation, or loss of components.
Applications
Polymers
DSC is widely used to characterize polymers. It can reveal glass-transition temperatures, melting ranges, crystallization behavior, cold crystallization, and curing reactions. These measurements help distinguish effects of composition, processing, and thermal history.
For a semicrystalline polymer that undergoes cold crystallization during heating, a commonly used estimate of the initial crystalline mass fraction is
where is the measured specific melting enthalpy, is the positive magnitude of the specific cold-crystallization enthalpy, is the polymer mass fraction in the specimen, and is an appropriate reference melting enthalpy for fully crystalline polymer.
This estimate assumes compatible reference data, correct mass normalization, and adequate separation of relevant events. Crystal reorganization during heating, multiple crystalline forms, fillers, and overlapping transitions can complicate the calculation.
Biological macromolecules
High-sensitivity DSC measures thermally induced changes in proteins, DNA, and other biological macromolecules in solution. For proteins, an excess heat-capacity peak can reflect disruption of the folded structure studied in protein folding.
After a suitable baseline correction, integration gives a calorimetric transition enthalpy. A characteristic transition temperature may describe the thermal unfolding process, but its interpretation depends on whether unfolding is reversible and whether equilibrium is approached. Aggregation and irreversible reactions can make an apparent transition temperature dependent on scanning rate and concentration.
Consequently, thermal transition temperature is not, by itself, a complete measure of molecular stability or biological function.
Pharmaceuticals, foods, and other materials
In pharmaceutical materials research, DSC is used to investigate crystalline forms, amorphous content, melting, hydration changes, and interactions among formulation components. A shifted or broadened peak alone does not establish chemical incompatibility, because physical mixing, dissolution, and changes in crystal form may also alter the signal.
Food applications include the study of fat melting, freezing behavior, starch gelatinization, and protein denaturation. In metals and other inorganic materials, DSC can investigate melting, solid-state transformations, precipitation, and reactions.
Measurements at several scanning rates can also support studies of chemical kinetics. Derived kinetic parameters depend on the analysis method and assumptions; a single scan generally cannot establish a unique reaction mechanism.
Modulated and specialized methods
In temperature-modulated DSC, a small periodic temperature variation is superimposed on an underlying heating or cooling program. Analysis of the response helps distinguish heat-capacity-related behavior from processes with different time dependence.
The resulting “reversing” and “nonreversing” signals are operational components of a dynamic measurement. They are not universal labels for thermodynamically reversible and irreversible processes. Their interpretation depends on modulation period, amplitude, underlying scanning rate, and sample response.
Fast-scanning calorimetry uses low-thermal-mass sensors and very small specimens to achieve much higher scanning rates than conventional DSC. It can probe rapid crystallization and kinetically trapped states, although thermal contact and sensor calibration become especially important.
Other specialized approaches include high-pressure DSC, photocalorimetry for light-initiated reactions, and high-sensitivity solution calorimeters for dilute macromolecular samples.
Limitations and complementary techniques
DSC measures thermal effects rather than chemical identity or molecular structure. Important limitations include:
- Thermal lag: the actual sample temperature can differ from the reported sensor temperature.
- Rate dependence: transformations may shift or change shape as the scanning rate changes.
- Overlapping events: simultaneous processes can obscure one another.
- Baseline ambiguity: broad transitions and heat-capacity changes can make peak integration uncertain.
- Sample heterogeneity: a small specimen may not represent a bulk material.
- Irreversible changes: reactions, evaporation, or degradation can prevent repeatable cycling.
Thermogravimetric analysis complements DSC by measuring mass changes, helping distinguish events such as volatilization or decomposition from transitions without mass loss. X-ray diffraction can identify crystalline phases, while infrared spectroscopy can provide information about chemical structure. Coupled measurements can relate thermal events to evolved gases or structural changes, although linking a calorimetric feature to a specific mechanism still requires appropriate experimental evidence.
Historical development
DSC developed from earlier calorimetry and differential thermal methods. Commercial instruments appeared in the 1960s, and both power-compensation and heat-flux designs became established. Subsequent developments included improved sensors, automated cooling, digital control and analysis, high-sensitivity solution instruments, temperature modulation, and fast-scanning methods.
Standardized terminology and measurement procedures have become important because reported transition temperatures and enthalpies depend not only on the specimen but also on the temperature program, calibration, baseline treatment, and reporting convention.