Molpro Interface

ElemCo.MolproInterface — Module

This module provides an interface to Molpro to read and write orbitals and other data.

It includes functions to read Molpro matrop files to import overlap/density matrices and orbital coefficients.

A convenient way to use the interface is given below.

ElemCo.jl Molpro Interface: Seamless Julia-Molpro Integration

The ElemCo.jl package provides a streamlined workflow for integrating Julia calculations with Molpro quantum chemistry computations through a simple include mechanism and macro-based data exchange.

Setup

Add the following line to your Molpro input file:

include,elemcoil

This includes a separate configuration file elemcoil that handles all data exchange between Molpro and Julia.

Configuration File (elemcoil)

The elemcoil file defines the interface between Molpro and Julia:

$XML='data/mol.xml'
$ORBITALS='data/orbs.dat'
!$ECORBITALS='data/ecorbs.dat'
$ECVARIABLES='data/ecvars.dat'

system,'mkdir -p','data'

put,xml,$XML

{matrop
load ORB ORB
write,ORB,$ORBITALS,new,sci
}

system,'julia input.jl',' > elemcoil.log'

!{matrop
!read,ORB,file=$ECORBITALS
!save,ORB,3200.2,ORBITALS
!}

readvar,$ECVARIABLES

Key Components:

Variable definitions: Define file paths for XML output, orbital data, and variable exchange
Data export: Automatically exports Molpro XML and orbital matrices to specified files
Julia execution: Calls the Julia script input.jl and logs output to elemcoil.log
Data import: Reads back modified variables and orbitals (commented sections for optional use)

Julia Script (input.jl)

Use the convenient @molpro_input and @molpro_output macros for seamless data access:

using ElemCo

# Read Molpro data using the @molpro_input macro
@molpro_input

# Perform Julia calculations
result = @cc dcsd

# Export results back to Molpro using @molpro_output macro, with a prefix for variable names
@molpro_output result prefix="EC_"

This script reads the Molpro data defined in elemcoil, performs calculations using Julia's ElemCo.jl package, and exports results back to Molpro with a specified prefix.

The geometry and basis set information is automatically handled by the @molpro_input macro. Note that the XML file in Molpro stores only the AO basis set. The fitting basis set is not stored in the XML file. If you need a specific fitting basis set, you can define it before the @molpro_input macro in the usual way, e.g.:

geometry = Dict("mpfit" => "avtz-mpfit", "jkfit" => "vqz-jkfit")
@molpro_input

Workflow Benefits

Automatic data handling: No manual file path management - everything is configured in the elemcoil file
Clean separation: Molpro handles quantum chemistry, Julia handles specialized calculations
Bidirectional communication: Variables and data can flow both ways between Molpro and Julia
Logging: All Julia output is captured in elemcoil.log for debugging
Flexible integration: Commented sections allow for orbital modifications when needed

Use Cases

Post-processing: Analyze Molpro results with Julia's rich ecosystem
Method development: Implement new electronic structure methods in Julia
Data analysis: Use Julia's plotting and statistical capabilities on quantum chemistry data
Orbital manipulation: Modify or transform molecular orbitals using Julia algorithms
Property calculations: Compute additional molecular properties not available in Molpro

This interface makes it easy to leverage Julia's computational capabilities within existing Molpro workflows while maintaining clean, readable code.

Molpro Interface

Exported functions and types

Internal functions and types