This C program computes electronic transport coefficients for the 3D non-interacting one-band models of cuprates high temperature superconductors.
Installation only requires a C compiler (gcc). In the main directory, run:
$ make
This should produce the tbd
executable in the main directory. To copy this executable to your $HOME/bin/
directory (given you have one) run:
$ make install
You can increase the speed of the computation by adding flags -march=native
and -ffast-math
in the makfile. Be aware that using -ffast-math
break strict IEEE compliance.
You can go in the example
directory to test your installation:
$cd example
$ ../tbd
the program will run and output two files:
transport_vs_Mu.dat
transport_vs_T.dat
Both files contain the same data. The first contains blocks for each temperature values, in which the doping changes. The other contains blocks for each doping value, in which the temperature changes. This redundance is intended to make plotting with gnuplot easier. Since the loop on Mu is the outer loop, you must wait until the program fnishes to get transport_vs_Mu.dat
, whereas transport_vs_T.dat
is continuously updated during the computation.
The code reads all adjustable parameters of the model from the 'model.dat' file. These parameters are:
t
: first neighbour hopping, defines the energy scale (t=1)tp
: second neighbour hoppingtpp
: third neighbour hoppingtppp
: fourth neighbour hppingtz
: interplane hopping
For example, the parameters for Nd-LSCO are t=1, tp=-0.14, tpp=0.07, tppp=0, and tz=0.07.
ETA
: baseline, should always be non-zero, can be very small : [0.001,0.1]ETAell
: constant mean-free path, proportional to the velocity : [0,0.1]ETAdos
: density of states (DOS), proportional to 1/|v_k| : [0,0.1]ETAaFL
: Fermi-Liquid, proportional to (omega^2+pi^2*T^2) : [0,1]ETAbFL
: Fermi-Liquid and DOS, proportional to (omega^2+pi^2*T^2)/|v_k| : [0,1]ETAaPL
: Planckian Limit, proportional to T : [0,0.3]ETAk
: Large scattering at the antinode, proportional to cos(2*phi_k)^12 : [0,1]ETAw
: Linear in frequency, proportional to omega: [0,1]ETAkw
: Inverse in frequency, proportional to omega: [0,10]
WARNING: at the moment there are no safeguards that prevent the scattering to be negative at low energies (because of the linear dependence on omega). This might cause problems at high temperature.
muMin
: starting chemical potential [-6,6]muMax
: ending chemical potential [-6,6] larger than muMinnMu
: number of chemical potential values.nMu = 1
will makemuMin
the only value considered.Tmin
: lowest temperature in units of t, [0.005,0.1], For t=1 referencing 250 meV, T=0.005 is roughly equivalent to 6 KTmax
: highest temperature, [0.005,0.1], For t=1 referencing 250 meV, T=0.1 is roughly equivalent to 300 KnT
: number of temerature considered.nT = 1
will makeTmin
the only value considered.logT
: boolean. 0 will yield equally spaced values of T, 1 will produce equally spaced values on a log scale.
nK
: nb of kx and ky considered. 401nKz
: nb of kz considered in kz, if tz=0, it will be automatically set to 1nOmega
: nb of omega values evaluated. Ex: 31amplitudeCutoff
: The range of the omega integral is different for each temperature. It ends when the value of fermi function derivative reaches this cutoff, suggested value : 0.005.
The Gnuplot scripts provided in the plotting
and example
directory contain a few exemples of calculated units to compare results with experiments.