1 .TH g_tcaf 1 "Thu 16 Oct 2008"
3 g_tcaf - calculates viscosities of liquids
10 .BI "-n" " index.ndx "
11 .BI "-ot" " transcur.xvg "
12 .BI "-oa" " tcaf_all.xvg "
14 .BI "-of" " tcaf_fit.xvg "
15 .BI "-oc" " tcaf_cub.xvg "
16 .BI "-ov" " visc_k.xvg "
28 .BI "-[no]normalize" ""
32 .BI "-beginfit" " real "
33 .BI "-endfit" " real "
35 g_tcaf computes tranverse current autocorrelations.
36 These are used to estimate the shear viscosity eta.
37 For details see: Palmer, JCP 49 (1994) pp 359-366.
40 Transverse currents are calculated using the
41 k-vectors (1,0,0) and (2,0,0) each also in the y- and z-direction,
42 (1,1,0) and (1,-1,0) each also in the 2 other plains (these vectors
43 are not independent) and (1,1,1) and the 3 other box diagonals (also
44 not independent). For each k-vector the sine and cosine are used, in
45 combination with the velocity in 2 perpendicular directions. This gives
46 a total of 16*2*2=64 transverse currents. One autocorrelation is
47 calculated fitted for each k-vector, which gives 16 tcaf's. Each of
48 these tcaf's is fitted to f(t) = exp(-v)(cosh(Wv) + 1/W sinh(Wv)),
49 v = -t/(2 tau), W = sqrt(1 - 4 tau eta/rho k2), which gives 16 tau's
50 and eta's. The fit weights decay with time as exp(-t/wt), the tcaf and
51 fit are calculated up to time 5*wt.
52 The eta's should be fitted to 1 - a eta(k) k2, from which
53 one can estimate the shear viscosity at k=0.
56 When the box is cubic, one can use the option
59 averages the tcaf's over all k-vectors with the same length.
60 This results in more accurate tcaf's.
61 Both the cubic tcaf's and fits are written to
64 The cubic eta estimates are also written to
71 the transverse current is determined of
72 molecules instead of atoms. In this case the index group should
73 consist of molecule numbers instead of atom numbers.
76 The k-dependent viscosities in the
79 fitted to eta(k) = eta0 (1 - a k2) to obtain the viscosity at
83 NOTE: make sure you write coordinates and velocities often enough.
84 The initial, non-exponential, part of the autocorrelation function
85 is very important for obtaining a good fit.
89 Full precision trajectory: trr trj cpt
93 Structure+mass(db): tpr tpb tpa gro g96 pdb
99 .BI "-ot" " transcur.xvg"
103 .BI "-oa" " tcaf_all.xvg"
111 .BI "-of" " tcaf_fit.xvg"
115 .BI "-oc" " tcaf_cub.xvg"
119 .BI "-ov" " visc_k.xvg"
125 Print help info and quit
127 .BI "-nice" " int" " 19"
130 .BI "-b" " time" " 0 "
131 First frame (ps) to read from trajectory
133 .BI "-e" " time" " 0 "
134 Last frame (ps) to read from trajectory
136 .BI "-dt" " time" " 0 "
137 Only use frame when t MOD dt = first time (ps)
140 View output xvg, xpm, eps and pdb files
142 .BI "-[no]xvgr" "yes "
143 Add specific codes (legends etc.) in the output xvg files for the xmgrace program
146 Calculate tcaf of molecules
149 Also use k=(3,0,0) and k=(4,0,0)
151 .BI "-wt" " real" " 5 "
152 Exponential decay time for the TCAF fit weights
154 .BI "-acflen" " int" " -1"
155 Length of the ACF, default is half the number of frames
157 .BI "-[no]normalize" "yes "
160 .BI "-P" " enum" " 0"
161 Order of Legendre polynomial for ACF (0 indicates none):
171 .BI "-fitfn" " enum" " none"
190 .BI "-ncskip" " int" " 0"
191 Skip N points in the output file of correlation functions
193 .BI "-beginfit" " real" " 0 "
194 Time where to begin the exponential fit of the correlation function
196 .BI "-endfit" " real" " -1 "
197 Time where to end the exponential fit of the correlation function, -1 is till the end