**IMPORTANT**: We do not recommend to analyze data, where measured reaction is not complete. We recommend to perform measurements where both begin and end of reaction are well measured.

However, it is not always possible. For example, for heating measurements of curing process the decomposition reaction sometimes starts before curing is finished. Here is example of rough evaluation of incomplete data. The kinetic results are enabled only in the temperature range where reaction was measured.

## II. Model Free Analysis

If the data are not measured completely then the total effect is unknown. For DSC data the total effect is the enthalpy of thermal effect which is calculated as the peak area. If the enthalpy of the total effect is unknown then the conversion can not be calculated correctly, and the Model-free methods based on the defined conversion values, may not be used.

Methods of:

- Friedman,
- Ozawa-Flynn-Wall,
- Kissinger-Akahira-Sunose,
- ASTM 1641 ,
- numerical method

**may not be used** for incomplete data.

Only two Model-free methods **may be used**:

- ASTM E698,
- ASTM E2890

because they use points of maximum rate and do not use points with defined conversion.

The maximum point from each measurement is shown on this graph. Activation energy and pre-exponential factor are calculated from the slope and intersect of the straight line by Ozawa formula. This method can evaluate only activation energy for the main peak. This method is not suitable for the complex reactions with several maxima points.

The maximum point from each measurement is shown on this graph. Activation energy and pre-exponential factor are calculated from the slope and intersect of the straight line by Kissinger Method. This method can evaluate only activation energy for the main peak. This method is not suitable for the complex reactions with several maxima points.

## III. Model Based Analysis

Model-based analysis can be helpful for the measurements which are incomplete. But the **model will work** only in the temperature range, **where experimental data are present**. **Outside **of this range the predictions can be **incorrect **because experimental data do not contain information about any reaction steps outside this range.