CERN home page CERN home page The Consult page The Consult page Writeups at CERN Writeups at CERN Magboltz Magboltz This page This page

Using Magboltz

Introduction

Calculates drift, diffusion, gain and attachment of electrons in gases with applied electric and magnetic fields at arbitrary angles. The common block structure of the results is outlined below.

LINUX PC version uses DRAND48 double precision random no generator (can also use RNDM2 from CERNLIB with the same precision).

The program allows anisotropic elastic and inelastic scattering, reference: NIM A 421 (1999) 234-240 The gas data base list below shows those x-sections which contain anisotropic scattering data.

PostScript plots of the database x-sections can be obtained on:- http://cern.ch/magboltz/cross.

This version allows spatial gradients to be included in the solution for the Townsend gain and attachment coeficients. The program automatically gives a solution with spatial gradients for both time of flight (TOF), pulsed townsend (PT) and steady state townsend (SST) parameters. The nomenclature is similar to Sakai et al. J. Phys. D10 (1977) 1035. The simulation of avalanche gain detectors at high field requires the use of SST Townsend parameters. The program automatically updates the common blocks /CTOWNS/ and /CTWNER/ with the SST parameters if the spatial gradients are greater than: | alpha - attachment | = 60/cm at NTP. where alpha is the Townsend coefficient and attachment the attachment coefficient. (For smaller values of | alpha - attachment | < 60/cm, the change of the gain or attachment is typically less than 3 % for the solution without spatial gradients).

Estimates of the Penning effect at high field can be obtained by inspection of the detailed collision frequencies for excited states in the gas mixtures. Penning effects can occur between excited states in the gas mixture which are higher in energy than the lowest ionisation potential in the mixture. Previous results in Argon hydrocarbon mixtures show a transfer efficiency of about 25 % from excited states in Argon to ionisation of hydrocarbons.

The program should always converge to a solution. the error on the integration should scale with the square root of the number of collisions (parameter NMAX). The output should always be checked to ensure that the number of collisions in the last energy bin is small, any value less than 500 should give reasonable systematic errors if the number of collisions is greater than 500 then the integration energy range should be increased.

The program is limited in precision by the statistical accuracy of the results. It is possible to obtain a statistical accuracy of better than 0.1 % on the drift velocity and 1 % on the diffusion coeficients in most counting gas mixtures in about 30 seconds of computing time on a PC, Alpha or workstation. At high field when the Townsend coeficient is included in the spatial gradient the computation time may be required to increase to a few minutes. When velocity vectors are small such as the case with small Lorentz angles the parameter NMAX will need to be increased to 20 or more.

Geometry

The electric field is taken along the z-axis and the magnetic field is taken in the z-x plane at an angle, BTHETA, to the electric field.

The results of the calculation are loaded into common blocks:

COMMON/VEL/WX,WY,WZ
COMMON/VELERR/DWX,DWY,DWZ
COMMON/DIFLAB/DIFXX,DIFYY,DIFZZ,DIFYZ,DIFXY,DIFXZ
COMMON/DIFERB/DXXER,DYYER,DZZER,DYZER,DXYER,DXZER
COMMON/DIFVEL/DIFLN,DIFTR  
COMMON/DIFERL/DFLER,DFTER
COMMON/CTOWNS/ALPHA,ATT    
COMMON/CTWNER/ALPER,ATTER
Where Output units in common blocks: NB: The observed ALPHA is ALPHA-ATT. For magnetic fields ALPHA and ATT are defined parallel to the electric field.

INPUT CARDS

First card

Format: 2I10,F10.5, variables: NGAS,NMAX,EFINAL

Second card

Format: 6I5, variables: NGAS1,NGAS2,NGAS3,NGAS4,NGAS5,NGAS6

Third card

Format: 8F10.4, variables: FRAC1,FRAC2,FRAC3,FRAC4,FRAC5,FRAC6,TEMP,TORR

Fourth card

Format: 6F10.3, variables: EFIELD,BMAG,BTHETA

Card 4*N+1

Use NGAS=0 to terminate correctly.

Gas numbers

Routine Gas Last update Notes Star rating
GAS1 CF4 2001anisotropic scattering 5*
GAS2 Argon 1997  5*
GAS3 Helium 4 1997  5*
GAS4 Helium 3 1992  5*
GAS5 Neon 1992  5*
GAS6 Krypton 2001  4*
GAS7 Xenon 2001  4*
GAS8 methane 1994  5*
GAS9 ethane 1999  5*
GAS10propane 1999  4*
GAS11isobutane 1999  3*
GAS12CO2 2001  5*
GAS13C(CH3)4 neo-pentane1995  3*
GAS14H20 1998  3*
GAS15Oxygen 19903-body attachment included 4*
GAS16Nitrogen 1985Pitchford and Phelps 4*
GAS17nitric oxide 1995attaching gas 4*
GAS18nitrous oxide 1995attaching gas 4*
GAS19C2H4 ethene1999  4*
GAS20C2H2 acetylene1992  3*
GAS21Hydrogen 2001  5*
GAS22Deuterium 1998  5*
GAS23Carbon monoxide 1998  5*
GAS24methylal 1988  2*
GAS25DME 1998  4*
GAS26Reid step model ? anisotropic version -
GAS27Maxwell model ?   -
GAS28Reid ramp model ?   -
GAS29C2F61999anisotropic 4*
GAS30SF6 ? do not use high percentage 3*
GAS31NH3 ammonia1999  3*
GAS32C3H6 propene1999  4*
GAS33C3H6 cyclopropane1999  4*
GAS34CH3OH methanol1999  2*
GAS35C2H5OH ethanol1999  3*
GAS36C3H7OH isopropanol1999  2*
GAS37Cæsium 2001no dimers 2*
GAS38Fluorine ? Morgan 2*
GAS39CS2 2001ion drift, dark matter 2*
GAS40COS 2001  2*
GAS41CD4 2001TPCs in neutron background environment3*
GAS42BF3 Boron trifluoride2001anisotropic 3*
GAS43C2HF5 or C2H2F4 ? estimated no data, anisotropic1*
GAS44Helium 3 2002anisotropic 5*
GAS45Helium 4 2002anisotropic 5*
GAS46Neon 2002anisotropic 5*
GAS47Argon 2002anisotropic 5*
GAS48Krypton 2002anisotropic 4*
GAS49Xenon 2002anisotropic 4*
GAS50methane 2002anisotropic 5*
GAS52-80Dummy routines ?   -

Author

Magboltz has been written by Stephen Biagi, stephen.biagi@hotmail.co.uk.
Last updated on 7/9/2006.