*From*: jkl' at \`ccl.net*Subject*: Experimental geometries to Z-matrix*Date*: Fri, 14 Aug 1992 04:58:56 -0400

Dear Netters, I enclose 2 short programs which I found useful in converting experimental geometries to Z-matrix. They relate to my previous posting. You will have to draw a picture to follow my variable names. Cut them and use them if you need them, or change them back to C if you did not buy the overpriced Fortran Compiler for your workstation (or use a CRAY, it always has FORTRAN, -:) ) The two short programs are followed by the example. I will put them in Comp.Chem.List archives as methyl.f and amino.f ============= PROGRAM METHYL ============== C Jan Labanowski, Aug 13, 1992 C This "program" calculates geometrical parameters of the methyl group rotor C suitable for constructing Z-matrix C The experimental data from microwave spectroscopy for CH3--something C are frequently given as a C--H bond lengths and H--C--H angle for C an equilateral methyl piramid (idealized), and the deviation of the C the C--something bond from the rotation axis of the CH3 (so called C methyl group tilt). These data have to be converted to other parameters C to be used in construction of the Z-matrix for the molecule. C C Input: C d = C--H bond length and theta = angle H--C--H C C Output: C a = distance H....H (the side of the equilateral triangle formed by 3 H's) C h = the height of the equilateral triangle formed by 3 H atoms C v = height of the isosceles H--C--H C alpha = angle between the C--H bond and the height of the CH3 pyramid C Hp = the height of the pyramid formed by CH3 C beta = angle between height of the H--C--H isosceles (v) and the C height of pyramid (Hp) C Sorry for FORTRAN ugliness, but it is a translation of the C original PROGRAM METHYL DOUBLE PRECISION d, theta, alpha, h, Hp, a, v, deg, beta WRITE(*,*) 1 ' Enter C--H bond length and H--C--H angle (in deg):' READ(*,*)d,theta WRITE(*,*) deg = 3.1415926536D0/180.0D0 a = 2.0D0*d*dsin(0.5D0*theta*deg) v = d*dcos(0.5D0*theta*deg) h = 0.5D0*a*dsqrt(3.0D0) Hp = dsqrt(d*d - a*a/3.0D0) alpha = dacos(Hp/d)/deg beta = dacos(Hp/v)/deg 100 FORMAT(1X,A,F10.5) WRITE(*,100) 'C--H bond length = ', d WRITE(*,100) 'H--C--H angle = ', theta WRITE(*,100) 'Height of the H--C--H isosceles =', v WRITE(*,100) 'H....H distance = ', a WRITE(*,100) 'Height of pyramid base (h)= ', h h = 2.0D0*h/3.0D0 WRITE(*,100) '2/3*h = ',h WRITE(*,100) 'Angle between C--H bond and pyramid height = ', 1 alpha WRITE(*,100) 'Pyramid height = ', Hp WRITE(*,100) 'Angle between wall height and pyramid height =', 1 beta STOP END ================= end of methyl.f ============ =================beginning of amino.f ========= C Jan Labanowski, Aug 13, 1992 C This "program" calculates geometrical parameters of the amino group C suitable for constructing Z-matrix C The experimental data from microwave spectroscopy for something-C-NH2 C are frequently given as a N--H bond lenth, C--N--H angle and H--N--H C angle. Depending where you start, you need some dummy atoms to build C the molecule. This programs gives you a few parameters you might use. C C Input: C d = N--H bond length, thCNH angle C--N--H, thHNH angle H--N--H C C Output: C a = distance H....H (the base of the H--N--H isosceles) C v = the height of the H--N--H isosceles (i.e., v is the bisector of C the H--N--H angle). It starts at C and ends at point Y. Point Y C is located half way between H atoms. C l = distance N--X (X is the projection of Y on the line passing through C C--N). l is a projection of v on the C--N line. C h = is the distance from Y to X. (h and l are perpendicular, C h is a bisector of the H--X--H angle). C b = H--X distance, i.e. distance of the point half way between H atoms C and C--N line C alpha = C--N--Y angle (angle between C--N bond and v). C beta = half of H--X--H angle, i.e. half of the dihedral angle between C two CNH planes, i.e., H--X--Y angle. PROGRAM AMINO DOUBLE PRECISION d, thCNH, thHNH, alpha, h, l, a, v, deg, beta WRITE(*,*) 1 ' Enter N--H bond length, C--N--H angle, H--N--H angle (in deg):' READ(*,*)d,thCNH,thHNH WRITE(*,*) deg = 3.1415926536D0/180.0D0 a = 2.0D0*d*dsin(0.5D0*thHNH*deg) C v = d*dcos(0.5D0*thHNH*deg) v = dsqrt(d*d - a*a/4.0D0) b = d*dsin((180.0D0-thCNH)*deg) h = dsqrt(b*b - a*a/4.0D0) l = dsqrt(v*v - h*h) alpha = 180.0D0 - dasin(h/v)/deg beta = acos(h/b)/deg 100 FORMAT(1X,A,F10.5) WRITE(*,100) 'N--H bond length = ', d WRITE(*,100) 'C--N--H angle = ', thCNH WRITE(*,100) 'H--N--H angle = ', thHNH WRITE(*,100) 'Height of the H--N--H isosceles =', v WRITE(*,100) 'H...H distance = ', a WRITE(*,100) 1 'Projection of H--N--H bisector on the C--N line = ', l WRITE(*,100) 1 'Distance between H...H midpoint and C--N line = ', h WRITE(*,100) 1 'Angle between C--N bond and bisector of H--N--H = ', alpha WRITE(*,100) 1 'Half of dihedral angle between C--N--H planes = ',beta WRITE(*,100) 1 'Distance between H atom and C--N line = ', b STOP END ====================== end of amino.f ========= Now I ran them for experimental data for CH3NH2 from Landolt-Boernstein, Group II, Vol. 7, Neue Serie. They were: dNH=1.0096 dCN=1.4714 dCH=1.0987 aHCN=108.0 aHCN=110.3 aHNH=107.1 Methyl group tilt was 2.9 The results were of methyl were: --------------------- Enter C--H bond length and H--C--H angle (in deg): 1.0987 108.0 C--H bond length = 1.09870 H--C--H angle = 108.00000 Height of the H--C--H isosceles = 0.64580 H....H distance = 1.77773 Height of pyramid base (h)= 1.53956 2/3*h = 1.02638 Angle between C--H bond and pyramid height = 69.09484 Pyramid height = 0.39204 Angle between wall height and pyramid height = 52.62263 Results of amino were: ---------------- Enter N--H bond length, C--N--H angle, H--N--H angle (in deg): 1.0096 110.3 107.1 N--H bond length = 1.00960 C--N--H angle = 110.30000 H--N--H angle = 107.10000 Height of the H--N--H isosceles = 0.59982 H...H distance = 1.62420 Projection of H--N--H bisector on the C--N line = 0.35027 Distance between H...H midpoint and C--N line = 0.48693 Angle between C--N bond and bisector of H--N--H = 125.72869 Half of dihedral angle between C--N--H planes = 59.05317 Distance between H atom and C--N line = 0.94689 =============================================================== Based on these, I made a Z-matrix for CH3NH2 for G9X. It is not meant to be used for geometry optimization. I needed it to measure angles and such. You could probably make a better Z-matrix for CH3NH2 ---------- %chk=something # HF/STO-3G Methylamine Geometry from Landolt-Bornstein 0 1 X C 1 dCX H 2 dCH 1 aHCX H 2 dCH 1 aHCX 3 t120 H 2 dCH 1 aHCX 4 t120 X 2 dCV 1 aVCX 3 t180 N 2 dCN 1 aNCH 3 t180 X 7 dNY 2 aCNY 3 t0 H 7 dNH 2 aCNH 8 tYXH H 7 dNH 2 aCNH 8 tmYXH dCX=0.39204 dCH=1.09870 dCN=1.4714 dCV=0.64580 dNH=1.00960 dNY=0.59982 aHCX=69.09484 aVCX=52.62263 aNCH=177.1 aCNH=110.3 aCNY=125.72869 t120=120.0 t180=180.0 t0=0.0 tYXH=59.05317 tmYXH=-59.05317 =================== I converted it to cartesians using xmol program which was announced on the list on 92/06/03 (it is a really nice piece of software, use it, when it is still available...). 10 Methylamine Geometry from Landolt-Bornstein X 0.000000 0.000000 0.000000 C 0.392040 0.000000 0.000000 H -0.000000 1.026375 0.000000 H -0.000000 -0.513188 0.888867 H -0.000000 -0.513188 -0.888867 X -0.000001 -0.513188 -0.000000 N 1.861556 -0.074442 -0.000000 X 2.236006 0.394142 -0.000000 H 2.236009 0.394145 0.812098 H 2.236009 0.394145 -0.812098 Sorry for the long message, but if someone have sent me this a week ago, I would be younger... Jan Labanowski Ohio Supercomputer Center jkl' at \`ccl.net