Merge pull request #90 from LLNL/holm10-tg_rewrite-H2CRMcompatible

Fix dissociation into different species

.gitattributes

@@ -3,3 +3,4 @@ src/uedge/VERSION merge=ours
 *.m linguist-language=Fortran
 *.v linguist-language=Fortran
 doc/* linguist-documentation
+

.github/workflows/build-test-wheel.yml

@@ -21,7 +21,7 @@ jobs:
     - name: Install pip dependencies
       run: |
         python -m pip install --upgrade pip
-        python -m pip install forthon>=0.10.7 numpy==1.26.4 h5py
+        python -m pip install "forthon>=0.10.7" numpy h5py
         python -m pip install flake8
         python -m pip install pytest-isolate pytest-xdist
     - name: Build UEDGE
@@ -30,7 +30,7 @@ jobs:
     - name: Return to pytests branch
       uses: actions/checkout@v4
       with:
-        ref: pytests-develop
+        ref: pytests
         repository: LLNL/UEDGE 
     - name: Test with pytest
       run: |
@@ -75,8 +75,8 @@ jobs:
         uses: pypa/[email protected]
         env:
             CIBW_BUILD: ${{matrix.pyver}}-*
-            CIBW_SKIP: "pp37-* cp37-* *-win32 *-manylinux_i686"
             CIBW_ARCHS_LINUX: auto
+            CIBW_SKIP: "pp37-* cp37-* *-win32 *-manylinux_i686"
             CIBW_BEFORE_BUILD: pip install numpy 'forthon>=0.10.7' 'build<0.10.0'
 
       - name: Upload Artifacts
@@ -153,8 +153,8 @@ jobs:
         uses: pypa/[email protected]
         env:
             CIBW_BUILD: ${{matrix.pyver}}-*
-            IBW_SKIP: cp37-*
             CIBW_ARCHS: ${{matrix.arch}}
+            CIBW_SKIP: "pp37-* cp37-* *-win32 *-manylinux_i686"
             CIBW_BEFORE_BUILD: pip install numpy 'forthon>=0.10.7' 'build<0.10.0'
 
       - name: Upload Artifacts

aph/aph.v

@@ -11,6 +11,8 @@ sgvcxc   real [m^2/s] /2.e-14/# const. value of sigv_cx for issgvcxc=1;
                               # const. sig for issgvcxc=2, sig-v=sig*sqrt(T/mi)
 			      # with units now [m^2]
 isaphdir integer       /1/    #=1 uses aphdir; =0 uses explicit rate file names
+crmnfname character*120 /"crumpet_nrates.dat"/ # Name of CRM rate data file
+crmefname character*120 /"crumpet_Erates.dat"/ # Name of CRM E-sink data file
 aphdir	character*120 # name of directory containing data files
 data_directory	character*120 # another dirname containing data files. This is to be be passed in
 
@@ -38,6 +40,34 @@ htlcx(0:htnt,0:htnn,0:htns-1)	_real
 htlqa(0:htnt,0:htnn,0:htns-1)	_real	
 	# log(htqa) where htqa is ADPAK rate parameter data for energy loss
 
+
+***** Rtcrumpet:
+# data from CRUMPET rate tables
+cmpe	integer	/60/		# number of energy points for crm params
+cmpd	integer /15/		# number of density points for crm params
+crmdiss(cmpe,cmpd) _real [m**3/s] # Mol diss rate vs. temp and dens
+crmarate(cmpe,cmpd) _real [m**3/s] # a-create rate from mol vs tem, dens
+crmselm(cmpe,cmpd) _real [W m**3] # el e-change due to mol diss vs temp, dens
+crmsiam(cmpe,cmpd) _real [W m**3] # ia e-change due to mol diss vs temp, dens
+crmspotm(cmpe,cmpd) _real [W m**3] # Epot change due to mol diss vs temp, dens
+crmsrada(cmpe,cmpd) _real [W m**3] # Radiation due to mol diss vs temp, dens
+crmsradm(cmpe,cmpd) _real [W m**3] # Radiation due to mol diss vs temp, dens
+cekpt(cmpe)	_real		# natural log of temperature (eV)
+crlemin		real		# minimum of ekpt
+crlemax		real		# maximum of ekpt
+cerefmin	real	[eV]	# minimum temperature in table data
+cerefmax	real	[eV]	# maximum temperature in table data
+cdelekpt	real		# interval size for ekpt in table data
+cdkpt(cmpd)	_real		# log10 of density (/m**3)
+crldmin		real		# minimum of dkpt
+crldmax		real		# maximum of dkpt
+cdrefmin	real	[/m**3]	# minimum density in table data
+cdrefmax	real	[/m**3]	# maximum density in table data
+cdeldkpt	real		# dkpt interval in atomic data tables
+ctaumin		real		# minimum tau in rtau opt-dep table data
+ctaumax		real		# maximum tau in trau opt-dep table data
+cdeltau	    real		# log int. size of rtau in opt-dep table dat
+
 ***** Rtdegas:
 # data from DEGAS rate tables eh.dat, ehr1.dat, nwfits and atmc.dat
 mpe	integer	/48/		# number of energy points
@@ -142,6 +172,23 @@ readehr1(fname:string)	subroutine
 readehr2(fname:string)	subroutine
 	# read 2004 data (n=2-9) files thin.dat, thickLyA.dat, thickAllLy.dat
 	# in    fname  filename to read
+readcrumpete(fname:string)  subroutine
+    # read CRUMPET energy sink/source file
+    # in    fname  filename to read
+readcrumpetn(fname:string)  subroutine
+    # read CRUMPET rate file
+    # in    fname  filename to read
+sv_crumpet(te:real,ne:real,rate:integer) real function [X m**3/s]
+    # calculate CRUMPET atomic creation form molecules
+    # in    te=electron temperature [J]
+    # in    ne=electron density [/m**3]
+    # in    rate=data table to be read
+    #            10: mol diss rate, [X]=1
+    #            11: mol->atom rate, [X]=1
+    #            20: el E change due to mol interaction, [X]=J
+    #            21: i/a E change due to mol interaction, [X]=J
+    #            22: Epot change due to mol interaction, [X]=J
+    #            23: radiation due to mol interaction, [X]=J
 
 ***** Aphwrk:
 # working arrays for 2-d spline interpolation
@@ -166,5 +213,6 @@ rqacoef(1:nxdata_aph,1:nydata_aph)  _real # spline coeff's for line emission
 ***** Subs:
 # Subroutines that can be called from the parser
 aphread                                     subroutine
+crumpetread                                 subroutine
 
 

aph/aphrates.m

@@ -1336,3 +1336,155 @@ real function svdiss (te)
       end
 
 c----------------------------------------------------------------------c
+      real function sv_crumpet (te, ne, rate)
+      implicit none
+      real te, ne
+      integer rate
+
+Use(Dim)
+Use(UEpar)
+Use(Share)                # istabon
+Use(Physical_constants)   # ev
+Use(Data_input)
+Use(Rtdata)
+Use(Rtcrumpet)
+
+c     local variables --
+      integer je,jd
+      real zloge,zlogd,rle,rld,fje,fjd,mins,maxs,epsilon,
+     .     svd_crm11,svd_crm12,svd_crm21,svd_crm22,svd_crm1,svd_crm2
+
+
+c     Compute molecular dissociation rate - A. Holm's CRUMPET
+c     te [J]          = electron temperature
+c     ne [/m**3]      = electron density
+c     rate            = rate to access
+c                       10: mol. diss, X=[]
+c                       11: atom source from mol dis, X=[]
+c                       20: el. energy change from mol interactions, X=[J]
+c                       21: i/a energy change from mol interactions, X=[J]
+c                       22: Epot (binding E) change from mol interactions, X=[J]
+c                       23: Atom radiation source from mol interactionsm X=[J]
+c                       24: Mol. radiation source from mol interactionsm X=[J]
+c     svma_crumpet [X/s]    = radiation rate
+
+c----------------------------------------------------------------------c
+c     Logarithmic interpolation as in Stotler-95
+
+
+
+      if (ishymol .eq. 0) then
+            sv_crumpet=0
+      elseif (ismolcrm .eq. 0) then
+            sv_crumpet=0
+      else
+c     compute abscissae --
+         zloge=log(te/ev)
+         rle=max(crlemin, min(zloge,crlemax))
+         zlogd=log10(ne)
+         rld=max(crldmin, min(zlogd,crldmax))
+c     table indicies for interpolation --
+         je=int((rle-crlemin)/cdelekpt) + 1
+         je=min(je,cmpe-1)
+         jd=int((rld-crldmin)/cdeldkpt) + 1
+         jd=min(jd,cmpd-1)
+c     fractional parts of intervals (je,je+1) and (jd,jd+1) --
+         fje=(rle-cekpt(je))/(cekpt(je+1)-cekpt(je))
+         fjd=(rld-cdkpt(jd))/(cdkpt(jd+1)-cdkpt(jd))
+
+c     pick the appropriate matrix to interpolate from
+         if (rate .eq. 10) then
+              svd_crm11=crmdiss(je,jd) 
+              svd_crm12=crmdiss(je,jd+1)
+              svd_crm21=crmdiss(je+1,jd) 
+              svd_crm22=crmdiss(je+1,jd+1)
+         elseif (rate .eq. 11) then
+              svd_crm11=crmarate(je,jd) 
+              svd_crm12=crmarate(je,jd+1)
+              svd_crm21=crmarate(je+1,jd) 
+              svd_crm22=crmarate(je+1,jd+1)
+         elseif (rate .eq. 20) then
+              svd_crm11=crmselm(je,jd) 
+              svd_crm12=crmselm(je,jd+1)
+              svd_crm21=crmselm(je+1,jd) 
+              svd_crm22=crmselm(je+1,jd+1)
+         elseif (rate .eq. 21) then
+              svd_crm11=crmsiam(je,jd) 
+              svd_crm12=crmsiam(je,jd+1)
+              svd_crm21=crmsiam(je+1,jd) 
+              svd_crm22=crmsiam(je+1,jd+1)
+         elseif (rate .eq. 22) then
+              svd_crm11=crmspotm(je,jd) 
+              svd_crm12=crmspotm(je,jd+1)
+              svd_crm21=crmspotm(je+1,jd) 
+              svd_crm22=crmspotm(je+1,jd+1)
+         elseif (rate .eq. 23) then
+              svd_crm11=crmsrada(je,jd) 
+              svd_crm12=crmsrada(je,jd+1)
+              svd_crm21=crmsrada(je+1,jd) 
+              svd_crm22=crmsrada(je+1,jd+1)
+         elseif (rate .eq. 24) then
+              svd_crm11=crmsradm(je,jd) 
+              svd_crm12=crmsradm(je,jd+1)
+              svd_crm21=crmsradm(je+1,jd) 
+              svd_crm22=crmsradm(je+1,jd+1)
+         else    
+              call remark("Rate option not recognized!")
+         endif
+
+
+c ... Check whether we have all positive, all negative, or zero crossing
+            mins=min(svd_crm11,svd_crm12,svd_crm21,svd_crm22)
+            maxs=max(svd_crm11,svd_crm12,svd_crm21,svd_crm22)
+            if (mins .gt. 0) then
+c ... Minimum is positive, normal interpolation
+                svd_crm11=log(svd_crm11)
+                svd_crm12=log(svd_crm12)
+                svd_crm21=log(svd_crm21)
+                svd_crm22=log(svd_crm22)
+                svd_crm1=svd_crm11+fjd*(svd_crm12-svd_crm11)
+                svd_crm2=svd_crm21+fjd*(svd_crm22-svd_crm21)
+                sv_crumpet = exp( svd_crm1 + fje*(svd_crm2-svd_crm1) )
+
+            elseif (maxs .lt. 0) then
+c ... Maximum is negative, interpolate of negative values
+                svd_crm11=log(-svd_crm11)
+                svd_crm12=log(-svd_crm12)
+                svd_crm21=log(-svd_crm21)
+                svd_crm22=log(-svd_crm22)
+                svd_crm1=svd_crm11+fjd*(svd_crm12-svd_crm11)
+                svd_crm2=svd_crm21+fjd*(svd_crm22-svd_crm21)
+                sv_crumpet = -exp( svd_crm1 + fje*(svd_crm2-svd_crm1) )
+
+            else
+c ... Zero-crossing – linear interpolation   
+                jd=jd-1
+                je=je-1
+                fjd=((ne/1e6)-(10**(10+0.5*jd)))/ (10**(10+0.5*jd)*(10**0.5 - 1) )
+                fje=(te/ev - (10**(-1.2+0.1*je)))/( (10**(-1.2+0.1*je))*(10**0.1-1))
+
+                svd_crm1=svd_crm11+fjd*(svd_crm12-svd_crm11)
+                svd_crm2=svd_crm21+fjd*(svd_crm22-svd_crm21)
+                sv_crumpet = svd_crm1 + fje*(svd_crm2-svd_crm1)
+
+
+            endif
+
+        endif
+
+
+c----------------------------------------------------------------------c
+      return
+      end
+
+
+
+
+
+
+
+
+
+
+
+