CAChe Research
   Computer-Aided Chemistry & Biochemistry

                              A practical quantum chemistry tool for modeling
                                      biological systems and co-crystals

               • Improved handling for crystals & biological systems
               • Fast optimizations up to 15,000 atoms, e.g. proteins, crystals
               • New PM7 method improves intermolecular energies
               • New activation energy parameterization for transition states
               • All main group elements & transition metals (83)
               • FREE to academics

Crystals, co-crystals, condensed phase…..major improvements
MOPAC2016™ brings major improvements in the prediction of intermolecular interactions and hydrogen-bonding. This significantly improves geometries and energies of proteins, crystals, co-crystals, metal clusters, inorganics and other condensed phase systems.

NEW parameterization (PM7)

The most widely used semiempirical quantum chemistry package, MOPAC®, has been updated with a new and more accurate parameterization (PM7) for all the main group elements and transition metals.  The addition of Jurecka's dispersion function and integrated hydrogen-bond terms to PM7 significantly improves the prediction of intermolecular interactions.  Experimental and high-level ab initio data from over 9,000 compounds were used to parameterize the new method.  MOPAC2016™ with PM7 represents only the second major improvement in MOPAC® methodology, since PM3 was published in 1989; the first being PM6 in 2007. 

More accurate heats of formation from PM7 for simple organic molecules

Method           AUE       RMSE       LE
(all units = Kcal/mol)
PM7 (new)        4.01      5.89     -44.4
PM6              4.42      6.16     -42.2
B3LYP 6-31G(d)   5.14      7.36      36.1
PM3              6.23      9.44    -135.6
HF 6-31G(d)      7.34     10.64      72.1
AM1             10.00     14.65     200.4
AUE = Average Unsigned Error, RMSE = Root Mean Square Error, LE = Largest Error

Comparison of errors in heats of formation for a set of 1,366 compounds
containing only C, H, O, N, F, Cl, S, P and Br.

New activation energy parameterization for transition states: PM7-TS
Reaction barrier heights have been parameterized using published coupled cluster high-level ab initio calculations on the PM7 transition state and related ground state geometries (reactants and products).  The new method, PM7-TS reduces activation energy errors by a factor of approximately 3, relative to PM6 (Average Unsigned Error for PM7-TS: 3.8 Kcal/mol; PM6: 12.3 Kcal/mol)

is the most cited semiempirical program and has far more published accuracy data than any comparable program.  MOPAC has been used to check for and correct errors in the published experimental heats of formation tables on the NIST website.



                   MOPAC2009™ for CAChe

Users of the CAChe Worksystem, BioMedCAChe, and Quantum CAChe, etc. can now upgrade to MOPAC2009 with PM6.  MOPAC2009 handles many more elements than MOPAC2000; a total of 69 for PM6, 42 for AM1 and PM3, and 15 lanthanide sparkles, while PM6 provides significantly improved accuracy over the older methods.  Existing CAChe procedures for the old PM5 method in MOPAC2000 will automatically use the new PM6 method in MOPAC2009 (PM5 is replaced by PM6 in MOPAC2009).  The MOPAC2009 upgrade is available for both annual and permanent CAChe licenses, and is compatible with most versions of CAChe 6.x, 7.x, and later.  To test your CAChe installation with the MOPAC2009 upgrade, download the free demo version:

          View MOPAC2009 brochure (PDF)

Download MOPAC2009 for CAChe DEMO installer (EXE file)
Download MOPAC2009 for CAChe DEMO installer (ZIP file)
       DEMO version works with 1-12, 50-60, and 110-120 atoms only.  To restore your original
       version of MOPAC, follow the instructions in the installer to rename the original MOPAC
       compute engine before installation, or use your original CAChe CD to reinstall CAChe.
       "MOPAC2009 for CAChe" is produced by Stewart Computational Chemistry LLC, USA which
       has no relationship with Fujitsu Ltd, Japan, the owner of the CAChe products.

Sales information...                       or Email:



              High-accuracy custom parameterization

The methods in MOPAC® can achieve higher accuracy for thermodynamic properties, comparable to many high-level ab initio methods but much faster, by re-parameterizing for a subset of elements or compound classes.  PARAM automates and dramatically speeds-up the tedious and time-consuming parameterization process and, it comes complete with the extensive sets of experimental and DFT reference data used to parameterize PM6, PM3 and AM1.  Developed over many years by J.J.P. Stewart, PARAM has now been released as a supported software tool.

Custom re-parameterization is also available as a contract service.  Please contact for further information.


MOPAC, technical information
          Brochure, MOPAC2016 (PDF)

          PM7 accuracy information 

          Graphical user interfaces for MOPAC

          Background information on MOPAC

          Technical information on PM6

          Current version number & release notes



For sales inquiries, price quotations and purchase orders, etc:
   Worldwide:  CAChe Research   Email:
   India:          Seascape              Email:
   Japan:        Ryoka Systems     Email:
   China:        Hongcam               Email:

Academics:   Free Downloads...

MOPAC2016™ is available as a standalone command-line driven application for Windows®, Linux and Macintosh. For versions of MOPAC2016™ bundled with graphical user interfaces, click here.
MOPAC2016 (Windows, Linux, Mac)       Acad*.    Govt.    Comm.

1 permanent license (1 node)         Free      $2,500   $5,000
1-year Support** & updates (1 node)  $1,000    $1,000   $1,000
5 permanent licenses (5 nodes)         -       $5,000  $10,000
1-year Support** & updates (5 nodes)   -       $2,000   $2,000
Annual site-license & support          -       $5,000  $10,000
Permanent site-license (executable)    -      $10,000  $20,000
Permanent site-license (source-code)   -      $10,000  $20,000
Permanent site-license (source & executable)  $12,500  $25,000
MOPAC2009 for CAChe, Windows         $1,000    $2,500   $5,000
    Permanent single node license     
MOPAC2009 for CAChe, Windows         $2,000    $5,000  $10,000
    Annual site-license
PARAM (1 node)                      $30,000   $40,000  $50,000
* The free Academic license is only for non-profit, non-proprietary use. 
  A new copy of the executable needs to be downloaded annually.

**Annual Support includes prompt responses to email questions, usually within two working days, and software version updates (including MOPAC2016™) when available.  Current MOPAC2012™ and MOPAC2016™ customers without a Support contract, may upgrade to the latest version for 50% of the current list price.  The first year of annual support is included free with product purchase.

Multi-core single-chip processors (e.g. dual core and quad-core) are considered as 1 node for MOPAC licensing purposes.

License Agreement
   MOPAC2016 uses a click-thru end-user license agreement



          Installation, FAQs, & troubleshooting

          Online MOPAC manual

          Online PARAM manual

          Web support request form
         Email support request

          Release notes & current version number

          Download latest version of MOPAC

          Download MAKPOL, a free utility for constructing extended solids from
          a unit cell for input into MOPAC

          Download BZ, a free utility for analysing Brouillin Zone band structure
          in the output of MOPAC calculations on solids

Bibliography for PM6, PM7 and MOPAC

Optimization of parameters for semiempirical methods V: Modification of NDDO approximations and application to 70 elements,  James J. P. Stewart, J Mol Model (2007) 13:1173–1213, 
DOI 10.1007/s00894-007-0233-4

Application of the PM6 method to modeling the solid state,  James J. P. Stewart, J Mol Model (2008), DOI 10.1007/s00894-008-0299-7

Calculation of Quantum-Mechanical Descriptors for QSPR at the DFT Level: Is It Necessary?  T Puzyn, N Suzuki, M Haranczyk, J Rak; J. Chem. Inf. Model., May (2008),  DOI 10.1021/ci800021p

Modeling of Triazolic Fungicide Inhibitors Binding with Iron-Containing Sterol 14a-Demethylase with the PM6 Semiempirical Method,  JG Rocha, CMR Sant’Anna, 4th Brazilian Symposium on Medicinal Chemistry, BrazMedChem 2008

Validation of semiempirical PM6 method for the prediction of molecular properties of polycyclic aromatic hydrocarbons and fullerenes, A Alparonea, V Librando, Z Minniti; Chemical Physics Letters, Vol 460, Issues 1-3, 20 July 2008, P. 151-154

Theoretical exploration of the cooperative effect in NMF–NMF–amino acid residue hydrogen bonding system (compares PM6 and ab initio methods), Xichen Li et al, Phys. Chem. Chem. Phys., 2008, 10, 5607 - 5615, DOI: 10.1039/b804291h

Application of the PM6 method to modeling proteins, James J. P. Stewart, J Mol Model, 10 Dec 2008, DOI 10.1007/s00894-008-0420-y

Google Scholar listing for PM6, PM7, MOPAC2007, MOPAC2009, MOPAC2012, MOPAC2016
NEW for 2016
MOPAC2016™ for large structures & condensed phase
NEW Release!

Home  |  SW Products  |  MOPAC  |  CEPOS  |  Consulting  |  Resources  |   Contact us  |  News

ACS/AIChE Talks  |  What happened to CAChe?  |  DAG Bio  |  CCL Posts

While every effort is made to maintain accuracy of information on this web site,
no liability is accepted for errors, omissions or out-of-date information.
Copyright © 2007 - 2019 CAChe Research