Metadata-Version: 2.3
Name: matopt
Version: 0.0.1
Summary: Materials Design via Mathematical Optimization
Project-URL: Homepage, https://github.com/xyin-anl/matopt
Project-URL: Bug Tracker, https://github.com/xyin-anl/matopt/issues
Author-email: Shawn Yin <shawn.pypi@gmail.com>
Classifier: Operating System :: OS Independent
Classifier: Programming Language :: Python :: 3
Requires-Python: >=3.9
Requires-Dist: numpy
Requires-Dist: pyomo
Description-Content-Type: text/markdown

> [!IMPORTANT]
> This is a standalone version adapted from the [original implementation](https://github.com/IDAES/idaes-pse/tree/main/idaes/apps/matopt)

<img src="assets/matopt.png" alt="MatOpt Logo" width="300">

MatOpt (Materials Design via Mathematical Optimization) is an Algebraic Modeling Language (AML) system specifically design to create [Pyomo](https://www.pyomo.org/) objects for optimization-based nanomaterials design. MatOpt can be used to model crystalline nanostructured materials, including but not limited to particles, surfaces, and periodic bulk structures.

The main goals of this package are as follows:

- To simplify the representation of nanostructured materials, streamlining the creation of materials optimization problems.
- To automate many of the necessary steps of materials optimization, speeding up the development of new models and accelerating new materials discovery.
- To provide a simple interface so that users do not need to handle the details of building efficient mathematical optimization models or the specific Pyomo syntax to do this.

If you are using MatOpt, please consider citing:

- Hanselman, C.L., Yin, X., Miller, D.C. and Gounaris, C.E., 2022. [MatOpt: A Python Package for Nanomaterials Design Using Discrete Optimization.](https://doi.org/10.1021/acs.jcim.1c00984) _Journal of Chemical Information and Modeling_, 62(2), pp.295-308.

## Basic Usage

There are two main sub-modules contained in the package serving two distinct purposes:

- The `matopt.materials` module contains objects and methods for efficiently representing and manipulating a nanomaterial and its design space.
- The `matopt.aml` module contains objects and methods for speeding up the casting of a `pyomo` model with simplified modeling syntax and automatic model formulation tools.

**Dependencies**

User access to the solver CPLEX through Pyomo is assumed. For users who do not have access to CPLEX, the use of [NEOS-CPLEX](https://neos-guide.org/neos-interfaces#pyomo) is suggested as an alternative (Users need to setup a `NEOS_EMAIL` environment variable).

**Define design canvas**

Several pieces of information about the material and design space need to be specified in order to formulate a materials optimization problem. To fulfill this need, the `matopt.materials` module defines generic and simple objects for describing the type of material to be designed and its design space, also referred to as a "canvas".

Some key objects are listed as follows:

**Build model via descriptors**

The material type and design space specified provide indices, sets, and parameters for the optimization model. Using simple syntax, inspired by materials-related terminology, MatOpt users define a `MatOptModel` object, which will be translated into a Pyomo `ConcreteModel` object automatically.

MatOpt uses `MaterialDescriptor` objects to represent variables, constraints, and objectives. A `MatOptModel` object holds lists of `MaterialDescriptor` objects. By default, several universal site descriptors are pre-defined in the model.

<table>
<col width="15%" />
<col width="84%" />
<thead>
<tr class="header">
<th align="left">Descriptor</th>
<th align="left">Explanation</th>
</tr>
</thead>
<tbody>
<tr class="odd">
<td align="left"><code>Yik</code></td>
<td align="left">Presence of a building block of type k at site i</td>
</tr>
<tr class="even">
<td align="left"><code>Yi</code></td>
<td align="left">Presence of any type of building block at site i</td>
</tr>
<tr class="odd">
<td align="left"><code>Xijkl</code></td>
<td align="left">Presence of a building block of type k at site i and a building block of type l at site j</td>
</tr>
<tr class="even">
<td align="left"><code>Xij</code></td>
<td align="left">Presence of any building block at site i and any building block at site j</td>
</tr>
<tr class="odd">
<td align="left"><code>Cikl</code></td>
<td align="left">Count of neighbors of type l next to a building block of type k at site i</td>
</tr>
<tr class="even">
<td align="left"><code>Ci</code></td>
<td align="left">Count of any type of neighbors next to a building block at site i</td>
</tr>
</tbody>
</table>

User-specified descriptors are defined by `DescriptorRule` objects in conjunction with `Expr` expression objects. Available expressions include:

<table>
<col width="24%" />
<col width="75%" />
<thead>
<tr class="header">
<th align="left">Expression</th>
<th align="left">Explanation</th>
</tr>
</thead>
<tbody>
<tr class="odd">
<td align="left"><code>LinearExpr</code></td>
<td align="left">Multiplication and addition of coefficients to distinct descriptors</td>
</tr>
<tr class="even">
<td align="left"><code>SiteCombination</code></td>
<td align="left">Summation of site contributions from two sites</td>
</tr>
<tr class="odd">
<td align="left"><code>SumNeighborSites</code></td>
<td align="left">Summation of site contributions from all neighboring sites</td>
</tr>
<tr class="even">
<td align="left"><code>SumNeighborBonds</code></td>
<td align="left">Summation of bond contributions to all neighboring sites</td>
</tr>
<tr class="odd">
<td align="left"><code>SumSites</code></td>
<td align="left">Summation across sites</td>
</tr>
<tr class="even">
<td align="left"><code>SumBonds</code></td>
<td align="left">Summation across bonds</td>
</tr>
<tr class="odd">
<td align="left"><code>SumSiteTypes</code></td>
<td align="left">Summation across site types</td>
</tr>
<tr class="even">
<td align="left"><code>SumBondTypes</code></td>
<td align="left">Summation across bond types</td>
</tr>
<tr class="odd">
<td align="left"><code>SumSitesAndTypes</code></td>
<td align="left">Summation across sites and site types</td>
</tr>
<tr class="even">
<td align="left"><code>SumBondsAndTypes</code></td>
<td align="left">Summation across bonds and bond types</td>
</tr>
<tr class="odd">
<td align="left"><code>SumConfs</code></td>
<td align="left">Summation across conformation types</td>
</tr>
<tr class="even">
<td align="left"><code>SumSitesAndConfs</code></td>
<td align="left">Summation across sites and conformation types</td>
</tr>
</tbody>
</table>

Several types of `DescriptorRules` are available.

<table>
<col width="26%" />
<col width="73%" />
<thead>
<tr class="header">
<th align="left">Rule</th>
<th align="left">Explanation</th>
</tr>
</thead>
<tbody>
<tr class="odd">
<td align="left"><code>LessThan</code></td>
<td align="left">Descriptor less than or equal to an expression</td>
</tr>
<tr class="even">
<td align="left"><code>EqualTo</code></td>
<td align="left">Descriptor equal to an expression</td>
</tr>
<tr class="odd">
<td align="left"><code>GreaterThan</code></td>
<td align="left">Descriptor greater than or equal to an expression</td>
</tr>
<tr class="even">
<td align="left"><code>FixedTo</code></td>
<td align="left">Descriptor fixed to a scalar value</td>
</tr>
<tr class="odd">
<td align="left"><code>PiecewiseLinear</code></td>
<td align="left">Descriptor equal to the evaluation of a piecewise linear function</td>
</tr>
<tr class="even">
<td align="left"><code>Implies</code></td>
<td align="left">Indicator descriptor that imposes other constraints if equal to 1</td>
</tr>
<tr class="odd">
<td align="left"><code>NegImplies</code></td>
<td align="left">Indicator descriptor that imposes other constraints if equal to 0</td>
</tr>
<tr class="even">
<td align="left"><code>ImpliesSiteCombination</code></td>
<td align="left">Indicator bond-indexed descriptor that imposes constraints on the two sites</td>
</tr>
<tr class="odd">
<td align="left"><code>ImpliesNeighbors</code></td>
<td align="left">Indicator site-indexed descriptor that imposes constraints on neighboring sites</td>
</tr>
</tbody>
</table>

From the combination of the above pre-defined descriptors, expressions, and rules, a user can specify a wide variety of other descriptors, as necessary.

**Solve optimization model**

Once the model is fully specified, the user can optimize it in light of a chosen descriptor to serve as the objective to be maximized or minimized, as appropriate. Several functions are provided for users to choose from. The results of the optimization process will be loaded into `Design` objects automatically. Users can then save material design(s) into files for further analysis and visualization using suitable functions provided. MatOpt provides interfaces to several standard crystal structure file formats, including CFG, PDB, POSCAR, and XYZ.

## Examples

Below is a basic example of designing stable monometallic Pt nanoclusters given the size constraint. User can specify the problem using intuitive syntax and solve it using free NEOS-CPLEX solver (need to specify an email using `NEOS_EMAIL` environment variable)

```python
import numpy as np
from math import sqrt
from matopt.materials.atom import Atom
from matopt.materials.lattices.fcc_lattice import FCCLattice
from matopt.materials.canvas import Canvas
from matopt.aml.expr import SumSites
from matopt.aml.rule import PiecewiseLinear, EqualTo
from matopt.aml.model import MatOptModel

Lat = FCCLattice(IAD=2.7704443686888935)
Canv = Canvas()
Canv.addLocation(np.array([0, 0, 0], dtype=float))
Canv.addShells(2, Lat.getNeighbors)
Canv.setNeighborsFromFunc(Lat.getNeighbors)
Atoms = [Atom("Pt")]

N = 22
m = MatOptModel(Canv, Atoms)
m.addSitesDescriptor(
    "CNRi",
    bounds=(0, sqrt(12)),
    integer=False,
    rules=PiecewiseLinear(
        values=[sqrt(CN) for CN in range(13)],
        breakpoints=[CN for CN in range(13)],
        input_desc=m.Ci,
        con_type="UB",
    ),
)
m.addGlobalDescriptor(
    "Ecoh", rules=EqualTo(SumSites(desc=m.CNRi, coefs=(1.0 / sqrt(12) * 1.0 / N)))
)
m.addGlobalDescriptor("Size", bounds=(N, N), rules=EqualTo(SumSites(desc=m.Yi)))

D = None
try:
    D = m.maximize(m.Ecoh, tilim=100, solver="neos-cplex")
except:
    print("MaOpt can not find usable solver (CPLEX or NEOS-CPLEX)")
if D is not None:
    D.toPDB("result.pdb")
```

For more usage cases and detailed explanation and walkthrough, please check the [original repository](https://github.com/IDAES/idaes-pse/tree/main/idaes/apps/matopt). In each case, a Jupyter notebook with explanations as well as an equivalent Python script is provided.

## Applications

- Hanselman, C.L. and Gounaris, C.E., 2016. [A mathematical optimization framework for the design of nanopatterned surfaces.](https://aiche.onlinelibrary.wiley.com/doi/full/10.1002/aic.15359) _AIChE Journal_, 62(9), pp.3250-3263.
- Hanselman, C.L., Alfonso, D.R., Lekse, J.W., Matranga, C., Miller, D.C. and Gounaris, C.E., 2019. [A framework for optimizing oxygen vacancy formation in doped perovskites.](https://www.sciencedirect.com/science/article/pii/S0098135418310998) _Computers & Chemical Engineering_, 126, pp.168-177.
- Hanselman, C.L., Zhong, W., Tran, K., Ulissi, Z.W. and Gounaris, C.E., 2019. [Optimization-based design of active and stable nanostructured surfaces.](https://pubs.acs.org/doi/abs/10.1021/acs.jpcc.9b08431) _The Journal of Physical Chemistry C_, 123(48), pp.29209-29218.
- Isenberg, N.M., Taylor, M.G., Yan, Z., Hanselman, C.L., Mpourmpakis, G. and Gounaris, C.E., 2020. [Identification of optimally stable nanocluster geometries via mathematical optimization and density-functional theory.](https://pubs.rsc.org/en/content/articlelanding/2019/me/c9me00108e#!divAbstract) _Molecular Systems Design & Engineering_.
- Yin, X., Isenberg, N.M., Hanselman, C.L., Dean, J.R., Mpourmpakis, G. and Gounaris, C.E., 2021. [Designing stable bimetallic nanoclusters via an iterative two-step optimization approach.](https://pubs.rsc.org/en/content/articlelanding/2021/ME/D1ME00027F) _Molecular Systems Design & Engineering_, 6(7), pp.545-557.
- Yin, X. and Gounaris, C.E., 2022. [Search methods for inorganic materials crystal structure prediction.](https://doi.org/10.1016/j.coche.2021.100726) _Current Opinion in Chemical Engineering_, 35, p.100726.
- Hanselman, C.L., Yin, X., Miller, D.C. and Gounaris, C.E., 2022. [MatOpt: A Python Package for Nanomaterials Design Using Discrete Optimization.](https://doi.org/10.1021/acs.jcim.1c00984) _Journal of Chemical Information and Modeling_, 62(2), pp.295-308.
