> For the complete documentation index, see [llms.txt](https://doc.duaer.com/zh/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://doc.duaer.com/zh/getting-started/mpdd.md).

# 在 Duaer 里查MPDD 材料描述符数据库

在 Duaer 里：按化学式或元素检索 Material-Property-Descriptor Database 中的结构，含空间群、晶系、密度与机器学习预测的形成能。一次成功查询用 1 额度。
## 能拿到什么

Duaer 数据市场里的「MPDD 材料描述符数据库」：按化学式或元素检索 Material-Property-Descriptor Database 中的结构，含空间群、晶系、密度与机器学习预测的形成能。查到结果时用 1 额度，结果交给数字组织的下一个节点。

从节点创建器选 Duaer 数据 → MPDD 材料描述符数据库，或打开 [数据市场](/zh/getting-started/data-market.md)。相关：[Alexandria](/zh/getting-started/alexandria.md)、[Materials Project](/zh/getting-started/materials-project.md)。

## 适合做什么

- 大批量初筛候选结构的形成能。
- 按对称性与密度比较同组分的结构。

## 在画布里检索

1. 打开节点创建器，选 Duaer 数据 → MPDD。
2. 填 Formula（如 Fe2O3）或 Elements（如 Fe,O）。
3. Limit 默认 10，最大 20。执行。

查到结果才扣 1 额度；没有结果、输入有误或上游失败都不扣。

## 用 API 调用

- GET https://api.duaer.com/v1/data/mpdd?formula=Fe2O3
- Authorization: Bearer <Duaer key>

下方是与数据市场「复制技能」相同的英文技能，含参数、示例与返回字段。

## 调用技能（英文）

与数据市场「复制技能」一致。

```
---
name: duaer-mpdd
description: >-
  Duaer MPDD. Structures in the Material-Property-Descriptor Database by formula or elements, with space group, crystal system, density, and machine-learned formation energy.
  One successful search uses 1 Duaer credit.
---

# Duaer MPDD

Duaer MPDD searches the Material-Property-Descriptor Database, millions of structures with symmetry, density, and a machine-learned (SIPFENN) formation energy for fast screening.

## When to use

- Screen many structures of a composition by predicted formation energy.
- Get space group and density for a formula.

## When not to use

- DFT-computed stability. Use https://skills.duaer.com/alexandria.md.
- Experimental structures. Use https://skills.duaer.com/cod.md.

## Call

`GET https://api.duaer.com/v1/data/mpdd?formula=Fe2O3`

Header: `Authorization: Bearer <Duaer key>`

Use an account key or a model API key.

Get a Duaer key: https://skills.duaer.com/keys.md

## Parameters

Provide `formula` or `elements`.

- `formula` — Formula such as Fe2O3. Matches the reduced formula.
- `elements` — Optional. Materials that contain all of these, such as Fe,O.
- `limit` — Optional. Rows to return, from 1 to 20. Default 10.

## Examples

- `GET https://api.duaer.com/v1/data/mpdd?formula=Fe2O3` — iron oxide structures with symmetry and density.
- `GET https://api.duaer.com/v1/data/mpdd?elements=Mg,Al,O&limit=5` — magnesium aluminium oxides.

## Result

The response is `{ "items": [...] }`. Each item has `source`, `title`, `url`, and `summary`, plus:

- `structureId` — provider id.
- `formula`, `formulaCell`, `elements`, `sites` — reduced formula, cell formula, elements, and atoms in the cell.
- `spaceGroup`, `spaceGroupNumber`, `crystalSystem` — symmetry.
- `densityGcm3`, `volumeA3` — density in g/cm³ and cell volume in ų.
- `predictedFormationEnergyEvPerAtom` — machine-learned formation energy, not DFT.

Fields without a value are left out.

## Credits

A search that returns at least one row uses 1 credit.
An empty search, a search that finds nothing, or a failed search uses 0.
Wrong input returns 400 with a message and uses 0.
No remaining credits returns 402 and does not search.
A missing key returns 401.

## Related

- https://skills.duaer.com/alexandria.md — Duaer Alexandria
- https://skills.duaer.com/materials-project.md — Duaer Materials Project

```
## Questions

### Duaer 的MPDD 材料描述符数据库检索扣多少额度？

查到结果时扣 1 额度；没有结果或出错不扣。

### Duaer MPDD 的形成能可靠吗？

Duaer MPDD 的形成能来自机器学习模型 SIPFENN，适合快速初筛；定论请再用 DFT 数据源核对。

## 相关

- [在 Duaer 里查Alexandria](https://doc.duaer.com/zh/getting-started/alexandria.md)
- [在 Duaer 里查 Materials Project](https://doc.duaer.com/zh/getting-started/materials-project.md)