VASP Density of States (DOS) INCAR Calculator
An interactive tool to understand and generate key INCAR parameters for accurate dos calculations using vasp.
Interactive Broadening & INCAR Generator
Determines the method for handling partial occupancies. ISMEAR = -5 is recommended for final DOS calculations.
The width of the smearing in electron-volts (eV). Smaller values give sharper peaks but may require more k-points.
Number of points to represent the DOS. Higher values lead to a smoother plot. Default is often 301.
Set to 11 to calculate the site- and lm-projected Density of States (PDOS).
Generated INCAR Snippet
Peak FWHM: 0.12 eV
# INCAR settings for DOS calculation ISMEAR = 0 SIGMA = 0.05 NEDOS = 2001 LORBIT = 11
What are DOS Calculations Using VASP?
In computational materials science, performing dos calculations using vasp (Vienna Ab initio Simulation Package) is a fundamental task. The Density of States (DOS) of a material describes the number of available electronic states at each energy level. It is a crucial property that helps scientists understand and predict a material’s electronic behavior, such as whether it is a metal, semiconductor, or insulator. VASP is a powerful software for performing quantum mechanical calculations, and it provides several parameters in its input file (named `INCAR`) to control the precision and type of DOS calculation. This calculator focuses on helping you understand the most critical of these parameters.
VASP INCAR Formula and Explanation
While VASP performs a complex quantum mechanical calculation, the user controls it via simple text-based tags in the `INCAR` file. For a DOS calculation, the most important tags are `ISMEAR`, `SIGMA`, and `NEDOS`. Our calculator also includes `LORBIT` for projected DOS.
The `SIGMA` parameter, when using Gaussian smearing (`ISMEAR = 0`), applies a broadening function to the discrete energy levels calculated by VASP. This turns the sharp energy “spikes” into smooth peaks, which are easier to visualize and are more representative of real-world measurements. The conceptual formula for a Gaussian function used for this broadening is:
g(E) = (1 / (σ * √(2π))) * exp(- (E – E₀)² / (2 * σ²))
This formula is not something you input directly, but it’s what VASP uses internally to create the smooth DOS curve based on your `SIGMA` value.
| Variable | Meaning in INCAR file | Unit | Typical Range |
|---|---|---|---|
ISMEAR |
Smearing method selector. `-5` (tetrahedron) is best for DOS, `0` (Gaussian) is good for metals. | Categorical | -5, 0, 1, 2… |
SIGMA |
The width of the electronic smearing function. | eV (electron Volts) | 0.01 – 0.2 |
NEDOS |
The number of grid points used to calculate the DOS data. | Integer | 301 – 10001 |
LORBIT |
Controls writing of projected, site- and orbital-decomposed DOS to the DOSCAR file. | Integer | 10, 11 are common |
Practical Examples
The right settings for dos calculations using vasp depend heavily on the type of material. Here are two common scenarios.
Example 1: DOS for a Metal (e.g., Copper)
For a metal, there are states at the Fermi level, which can make convergence difficult. Gaussian smearing is often used in the initial self-consistent run, followed by a more accurate non-self-consistent run with the tetrahedron method.
- Inputs: A converged charge density file (`CHGCAR`).
- INCAR Settings for final DOS run:
ISMEAR = -5,LORBIT = 11,NEDOS = 4001. - Result: A continuous DOS plot with no band gap at the Fermi energy (0 eV).
Example 2: DOS for a Semiconductor (e.g., Silicon)
For a semiconductor or insulator, there is a band gap. The tetrahedron method (`ISMEAR = -5`) is highly recommended for these systems as it provides a very accurate DOS without smearing states into the gap.
- Inputs: A converged charge density file (`CHGCAR`).
- INCAR Settings:
ISMEAR = -5,SIGMA = 0.05(SIGMA is ignored for ISMEAR=-5 but good practice to keep),NEDOS = 2001,LORBIT = 11. - Result: A DOS plot showing a clear energy gap (a region of zero states) around the Fermi energy. For more details on band structure, you might check out resources on Band Structure Calculation.
How to Use This DOS Calculator
This tool is designed to help you visualize the effect of the `SIGMA` broadening parameter and to generate the correct `INCAR` tags for your own dos calculations using vasp.
- Select Smearing Method: Choose the `ISMEAR` value appropriate for your system. For the final, high-quality DOS, `ISMEAR = -5` is usually the best choice.
- Adjust Broadening: Use the `SIGMA` slider or input field to see how different broadening values affect the sharpness of a spectral peak in the live chart.
- Set Resolution: Choose the number of grid points (`NEDOS`). More points give a smoother curve.
- Choose Projection: Select `LORBIT=11` if you need to analyze the contributions from specific atoms or orbitals (pDOS).
- Interpret Results: The chart shows how a single energy state is “smeared” by the Gaussian function. The primary result shows the Full Width at Half Maximum (FWHM) of this peak, which is a direct consequence of your `SIGMA`.
- Copy Snippet: Use the “Copy INCAR Snippet” button to get a correctly formatted block of text to paste directly into your `INCAR` file.
Key Factors That Affect DOS calculations using VASP
- K-Point Density: The number of k-points used to sample the Brillouin zone is critical. A denser k-point mesh leads to a smoother and more accurate DOS. This is often more important than `NEDOS`.
- Energy Cutoff (ENCUT): This parameter in `INCAR` determines the plane-wave basis set size. It must be sufficiently converged for the system and pseudopotentials used.
- Self-Consistent Field (SCF) Convergence: The underlying electronic structure must be fully converged before running a non-self-consistent calculation for the DOS. See our guide on VASP SCF Convergence for help.
- Choice of ISMEAR: As discussed, using `ISMEAR = -5` is crucial for insulators and semiconductors to avoid artificial states in the band gap. For metals, `ISMEAR > 0` can be necessary for convergence.
- Value of SIGMA: When using smearing methods (`ISMEAR >= 0`), the value of `SIGMA` is a compromise. Too small, and the DOS is spiky and needs many k-points. Too large, and important features are washed out.
- Calculation Type (SCF vs. Non-SCF): For a high-quality DOS, it’s standard practice to first perform a fully converged self-consistent calculation, and then a second, non-self-consistent calculation (`ICHARG = 11`) with a denser k-point mesh and `ISMEAR = -5`.
Frequently Asked Questions (FAQ)
1. What is the best ISMEAR setting for DOS calculations?
For the final density of states, especially for semiconductors and insulators, ISMEAR = -5 (the tetrahedron method with Blöchl corrections) is highly recommended. It provides the most accurate description of the electronic structure without artificial smearing.
2. Why is my DOS plot so “noisy” or “spiky”?
This is almost always due to an insufficient number of k-points in your calculation. To get a smooth DOS, you need to use a much denser k-point mesh than for a simple energy calculation. For help, read about Choosing VASP K-Points.
3. Does SIGMA matter if I use ISMEAR = -5?
No, when using the tetrahedron method (`ISMEAR = -4` or `-5`), the `SIGMA` tag is ignored by VASP. The broadening is handled geometrically by the method itself.
4. What is the difference between DOS and projected DOS (pDOS)?
The total DOS sums up all electronic states from all atoms and orbitals. The projected DOS (pDOS), enabled by `LORBIT`, breaks down the total DOS into contributions from individual atoms and their specific orbitals (s, p, d, f). This is essential for understanding chemical bonding. Learn more at our Projected DOS Analysis guide.
5. What unit is the energy axis in a VASP DOS plot?
The energy is in electron Volts (eV). By default, VASP and most plotting tools will shift the energy scale so that the Fermi energy is at 0 eV.
6. What does a non-zero DOS at the Fermi level mean?
A significant, non-zero density of states at the Fermi energy (0 eV) is the defining characteristic of a metal. It indicates that there are available electronic states at the highest occupied energy level, allowing for electrical conduction.
7. Should I run the DOS calculation in the same step as my geometry optimization?
No. The best practice is a three-step process: 1) Fully relax the geometry. 2) Perform an accurate self-consistent calculation on the relaxed structure. 3) Perform a final non-self-consistent (`ICHARG=11`) calculation using the charge density from step 2, but with a denser k-point mesh and `ISMEAR=-5` for the high-quality DOS.
8. How do I choose EMIN and EMAX?
These `INCAR` tags set the energy range for the DOSCAR file. If you leave them out, VASP will automatically determine the range. It’s often best to run a quick calculation first, see the energy range in the `OUTCAR` or `vasprun.xml` file, and then re-run the DOS calculation with a specific `EMIN` and `EMAX` to focus on the region of interest (e.g., around the band gap).
Related Tools and Internal Resources
Enhance your understanding of dos calculations using vasp with our other expert guides and tools:
- Band Structure Calculation: Learn to plot the E-k diagram, a perfect companion to a DOS plot.
- VASP SCF Convergence: A detailed guide on achieving convergence for your VASP calculations.
- Choosing VASP K-Points: Master the art of selecting the right k-point mesh for any calculation.
- Projected DOS Analysis guide: A deep dive into interpreting orbital contributions.
- VASP Pseudopotentials Explained: Understand how to choose the right POTCAR files for your elements.
- Understanding VASP INCAR Tags: A comprehensive reference for all major INCAR settings.