thermohipy 0.3.3

A thermogravimetric analysis helper tool by Hikari.

ThermoHi

A lightweight Python toolkit for thermal kinetics data analysis and visualization.

[]

---

📖 Introduction

ThermoHi(thermohipy) is a small, research-oriented Python package for thermal kinetic analysis
and data visualization, designed for TG/DTG analysis in pyrolysis. It supports apparent activation energy calculation using model-free methods such as FWO, KAS, Starink, Friedman, and Vyazovkin method with clean APIs and ready-to-plot results. (You can also export the data and use your own software to create the plots.)

An example file (example.py) is included to illustrate the usage of ThermoHi's main functionalities. Supports temperature input in Celsius (°C), Fahrenheit (°F), and Kelvin (K).All values are internally converted to Kelvin.

🔍 Notes

There is no limit on the number of heating runs or conversion values; the examples include data with 3 heating rates and 5 conversion values. Data can be exported for plotting. Note: Vyazovkin method may generate a very large amount of data.

ThermoHi performs Arrhenius-type parameter extraction for non-isothermal data under linear heating conditions. The method is applicable to various thermo-analytical datasets provided that a well-defined characteristic temperature can be identified,(e.g., DSC, DMA). For non-chemical transitions (e.g., glass transition or relaxation processes), the extracted value should be interpreted as an apparent activation energy. Users are responsible for evaluating the physical validity of the underlying model.

---

Why ThermoHi?

ThermoHi aims to automate repetitive thermal analysis tasks (e.g., activation energy estimation), allowing researchers to spend less time clicking buttons — and more time thinking, drinking coffee ☕, saving the world, or simply focusing on what truly matters.

If ThermoHi contributes to your research, please consider citing the associated publication(s).

🔗 The paper's link is pending update.

---

Workflow

The core workflow of ThermoHi is:

data_object(DataList) → KineticAnalysis(alpha, data_object) → Result → plotting(export results)

where:

DataList（<class 'list'>）

• A list containing:
  • Temperature, T (list)
  • Heating rate, β (list)
  • dα/dT (list)

data_object(<class 'list'>)

• a list used for calculating multiple sets of data,

  e.g., [DataList_1, DataList_2, ..., Datalist_n]

Note: len(alpha) = len(DataList)

alpha(<class 'list'>)

• a list of conversion values(α, 0 < α < 1), e.g., [0.2, 0.4, 0.6, 0.8]

KineticAnalysis(alpha, Datalist)

• calculating activation energy, 5 mentioned model free methods were provided,

  e.g., thermohipy.KineticAnalysis(alpha, Datalist).fwo_ea(return_data = False)

FittingPlot(alpha, Datalist, KineticAnalysis)

• optional, generate fitting plots withmatplotlib for reference,

  e.g., thermohipy.FittingPlot(alpha, data_object, analysis).kasplot()

Result (<class 'dict'>)

• A dictionary containing:
  • α : conversion value
  • Eα : activation energy
  • k, b : slope and intercept of fitting curve
  • R² : coefficient of determination

---

Saving examples

If save_excel=True, the analytical results will be saved as an .xlsx file in the current working directory. The absolute file path will be printed in the terminal.

The structure kinetic results of FWO, KAS, Starink and Friedman method is almost the same:

and the plotting data(scatter points)

The fitting curves of Vyazovkin method:

Plotting examples

FWO method

$$\ln\left(\beta\right) = Const. -\frac{E_{\alpha}}{RT}$$

KAS method

$$\ln\left(\frac{\beta}{T^2}\right) = Const. -\frac{E_\alpha}{RT}$$

Starink method

$$\ln\left(\frac{\beta}{T^{1.92}}\right) = Const. -1.0008\frac{E_{\alpha}}{RT}$$

Friedman method

$$\ln\left(\frac{\text{d}\alpha}{\text{d}t}\right) =\ln[Af(\alpha)]- \frac{E_{\alpha}}{RT}$$

Vyazovkin method

$$\Phi(E_\alpha) = \min \sum_{i=1}^{n} \sum_{\substack{j=1 \ j \neq i}}^{n} \frac{I(E_\alpha, T_{\alpha,i}) \cdot \beta_j}{I(E_\alpha, T_{\alpha,j}) \cdot \beta_i}$$

here we use Cai Approximation (AIChE Journal 2006;52:1554–7) to calculated $I(E_\alpha, T_\alpha)$:

$$I(E_\alpha, T_\alpha) \approx \frac{R T_\alpha^2 E_\alpha + 0.66691 R T_\alpha}{E_\alpha \left(E_\alpha + 2.64943R T_\alpha\right)}\exp\left(-\frac{E_\alpha}{R T_\alpha}\right)$$

🔬 Planned Features

Future versions of ThermoHi will include:

• Support for importing experimental data directly from .csv or .xlsx files (via pandas)
• Unified plotting style for publication-ready figures (Matplotlib themes)
• Exporting results for analysis as .xlsx
• Directly read TG/DTG experimental data and automatically extract corresponding (T, β, dα/dT) values based on user-selected α

🚀 Quick Start

Note: the data is solely for demonstration.

import thermohipy as th
alpha = [0.2, 0.4, 0.6]
beta = [1, 5, 10, 20]
t1, t2, t3 = [150, 175, 200, 210], [165, 180, 210, 220], [400, 420, 440, 445]
dadT1 = [0.004, 0.005, 0.006, 0.007]
dadT2 = [0.005, 0.006, 0.007, 0.008]
dadT3 = [0.006, 0.007, 0.008, 0.009]
data = [
    th.DataList(t1, beta, dadT1),
    th.DataList(t2, beta, dadT2, unit='c'),
    th.DataList(t3, beta, dadT3, unit='f')
]

analysis = th.KineticAnalysis(alpha, data)
result1 = analysis.fwo_ea()
print(result1)

pic1 = th.FittingPlot(alpha, data_object, analysis).kasplot()

result2 = th.ExportData(alpha, data_object, analysis).export_fwo(save_excel=True)
print(result2)
result3 = th.ExportData(alpha, data_object, analysis).export_vyazovkin(save_excel=False)
print(result3)

---

📄 License

Distributed under the MIT License.
See LICENSE for details.

Author

Hikari Quicklime, Ph.D. Forestry Industry Researcher & Independent Developer
gitHub: QuicklimeHikari

⚙️ Installation

pip install thermohipy

Update 2026.5.2

The output data arrangement has been modified to facilitate plotting. before: the arrangement is x1 x2 x3 ... xn; y1 y2 y3 ... yn after: the arrangement now is x1 y1; x2 y2; x3 y3; ... xn yn