Elsevier Science Home
Computer Physics Communications Program Library
Full text online from Science Direct
Programs in Physics & Physical Chemistry
CPC Home

[Licence| Download | New Version Template] adiw_v1_0.gz(81 Kbytes)
Manuscript Title: AMYR 2: a new version of a computer program for pair potential calculation of molecular associations.
Authors: F. Torrens, M. Rubio, J. Sanchez-Marin
Program title: AMYR 2
Catalogue identifier: ADIW_v1_0
Distribution format: gz
Journal reference: Comput. Phys. Commun. 115(1998)87
Programming language: Fortran.
Computer: SGI Origin 2000.
Operating system: IRIX6.4, VM/CMS, AIX3.2, VMS, MacOS 7.1.
RAM: 4M words
Word size: 32
Peripherals: disc.
Keywords: Biology, Molecular association, Structure, Electrostatic model, Stationary point, Analysis, Hessian eigenvalues, Pair potential, 1/r expansion, Geometry optimization, Vectorization.
Classification: 3, 16.1.

Other versions:
Cat Id Title Reference
ACEO_v1_0 AMYR CPC 29(1983)351
ACBG_v1_0 AMYRVF CPC 66(1991)341

Nature of problem:
The program determines the optimum separation and relative orientation of two interacting molecular systems through a minimization of the interaction energy. The new version allows for two models to evaluate the electrostatic contribution: the one-centre-per atom and the three- centres-per atom models. The stationary point analysis of the intermolecular potential by means of the Hessian eigenvalues is also allowed.

Solution method:
The interaction energy is evaluated by a 1/R expansion (with 1/R, 1/R**4, 1/R**6 and 1/R**12 terms), parameterized on the basis of accurate SCF results [1-6]. Two electrostatic models can be used: the one-centre-per atom and the three-centres-per atom model by Hunter and Sanders [7]. The latter model is advisable for pi-delocalized systems. In other cases, it must be used with care, so that the one-centre-per atom model can be more convenient. The search for the minimum energy can be carried out by one of the following options: (1) simplex, (2) steepest-descent [3], (3) Davidon, and (4) Broyden-Fletcher-Goldfarb- Shanno optimization method [8-16]. For the final geometries the components of the 6 x 6 Hessian matrix can be calculated, at each stationary point so reached, by means of the expressions given in Ref. [17]. The Hessian eigenvalues allow for the characterization of the critical points. Classification of critical points is made by means of the index of critical point, lambda, which is the number of negative eigenvalues of the Hessian matrix. Final geometries can correspond to minima (lambda=0), first order transition states or saddle points (lambda=1) ... till fifth order transition states (lambda=5).

Reasons for new version:
The three-centres-per atom electrostatic model predicts better geometric parameters than the one-centre-per atom model [17]. Most of the final points of the optimization procedure are not true minima but transition states. An algorithm for the stationary point analysis of the intermolecular potential is required. The differences between both electrostatic models become more significant when the stationary points are characterized by means of the Hessian eigenvalues: only a few stationary points reported by the one-centre-per atom model are actual minima.

Summary of revisions:
The most important new capabilities since the previous version described in CPC are:
  1. The three-centres-per atom electrostatic model by Hunter and Sanders [7] has been implemented [17]. This model uses a set of three punctual charges for each aromatic atom; a positive charge is placed at the nucleus position, and two equal negative charges are placed just at the same distance above and below the positive charge, in the straight line perpendicular to the molecular plane.
  2. An algorithm [17] for the stationary point (minima and transition states) analysis of the intermolecular potential by means of the Hessian eigenvalues has been added.
  3. A fractional molecular charge is allowed [13].
  4. Input has been converted to free format.
  5. Output has been cleaned up and reformatted.
  6. The code has been made portable. It now runs on supercomputers, minicomputers, workstations as well as personal computers [15].

The maximum number of atoms can be changed by resetting the KAT parameter. Calculations should be restricted to molecules consisting of H, Li, C, N, O, Na, S, K and Ca atoms belonging to the 82 classes for which expansion coefficients are available. The maximum number of polarizability data per atom in order to interpolate polarizabilities for atomic charges can be changed by resetting the KINT parameter. The initial position of the second system with respect to the first one is defined by means of a six-fold vector (three translations and three rotations) about the Cartesian axes. Molecular associations involving only two molecular systems are allowed. A third molecule can be added if the remaining aggregate is frozen in a fixed geometry [16].

Unusual features:
The code is distributed with a documentation file containing instructions for input and other useful information for users.

Running time:
This depends strongly on the particular system studied, because the number of atom-atom intermolecular interactions is formally proportional to the square of the number of atoms in the molecules. For the test run distributed with the code, the computation time is 66 CPU seconds on the IBM RS/6000.

[1] S. Fraga, K.M.S. Saxena, M. Torres, Biomolecular Information Theory (Elsevier, Amsterdam, 1978) p. 83.
[2] S. Fraga, J. Comput. Chem. 3 (1982) 329.
[3] S. Fraga, Comput. Phys. Commun. 29 1983) 351.
[4] S.H.M. Nilar, S. Fraga, J. Comput. Chem. 5 (1984) 261.
[5] J.A. Sordo, S. Fraga, J. Comput. Chem. 7 (1986) 55.
[6] R. Daudel, Theorie Quantique de la Reactivite Chimique (Gauthier- Villar, Paris, 1967).
[7] C.A. Hunter, J.K.M. Sanders, J. Am. Chem. Soc. 112 (1990) 5525.
[8] F. Torrens, J. Sanchez-Marin, E. Orti, I. Nebot-Gil, J. Chem. Soc. Perkin Trans. 2 (1987) 943.
[9] F. Torrens, A.M. Sanchez de Meras, J. Sanchez-Marin, J. Mol. Struct. (Theochem.) 166 (1988) 135.
[10] F. Torrens, R. Montanana, J. Sanchez-Marin, Vectorizing pair- potential AMYR program for the study of molecular associations, in: High Performance Computing I, J.L. Delhaye, E. Gelenbe, eds. (North-Holland, Amsterdam, 1989) p. 299.
[11] F. Torrens, J. Sanchez-Marin, F. Tomas, J. Chem. Res. (S) (1990) 176.
[12] F. Torrens, E. Orti, J. Sanchez-Marin, Structural and charge effects on the phthalocyanine dimer, in: Physical Chemistry Modelling of Molecular Structures and Properties XLIV, J.L. Rivail, ed., Stud. Phys. Theor. Chem. 71 (Elsevier, Amsterdam, 1990) p. 221.
[13] F. Torrens, E. Orti, J. Sanchez-Marin, Electrically conductive phthalocyanine assemblies, Structural and non-integer oxidation number considerations, in: Lower-Dimensional Systems and Molecular Electronics, R.M. Metzger, P.Day, G.C. Papavassiliou, eds., NATO-ASI Ser. B 248 (Plenum Press, New York, 1991) p. 461.
[14] F. Torrens, E. Orti, J. Sanchez-Marin, Comput. Phys. Commun. 66 (1991) 341.
[15] F. Torrens, E. Orti, J. Sanchez-Marin, J. Mol. Graphics 9 (1991) 254.
[16] J. Rodriguez, J. Sanchez-Marin, F. Torrens, F. Ruette, J. Mol. Struct. (Theochem.) 254 (1992) 429.
[17] M. Rubio, F. Torrens, J. Sanchez-Marin, J. Comput. Chem. 14 (1993) 647.