Subject:
Date: Tue, 24 Aug 1999 21:01:50 +0200 (MET DST)
From: Montserrat Villard-Martin <villard@iap.fr>
To: rfosbury@eso.org, jvernet@eso.org

        Dear Bob,

        I have run a sequence of models that shows more clearly that
depletion cannot explain the correlation (see plots attached to this
e-mail).
I prefer that  you decide what information is important for the paper
and  the best way is to present it. If I can help to produce a table
(I guess this would be the best way?), or whatever you decide let me know.

        I have investigates the possibility
that, for total (gas+dust) abundance, depletion of C (and other metals) is
responsible for the observed correlations. I therefore fixed U (0.07),
PL=-1.5, etc and varied the dust/gas ratio.
The metal abundances are depleted  consistently. The dust/gas ratio
values are 0, 0.5, 1 relative to solar (see below for a definition of
this parameter).

        Two are the main reasons that exclude depletion to explain the
correlations:

        1) NV/HeII is insensitive to depletion and we cannot explain
the large values of this ratio

        2) Objects with large NV/CIV imply in this scenario the
highest depletion. Acording to the models, these objects should show
low CIV/HeII and CIII/HeII ratios. However, this is not the case.
Objects with large NV ratios, show large C/HeII ratios as well.

 I send you below the information you requested, for models
with [dust/gas]=0, 0.2, 0.5, 1, 1.5 x solar value.
Let me know if you have more questions. I am not sure if the
explanations are clear enough!

        Cheers

        Montse

Description of the models calculated to study the effects of
               depletion on the N/C line ratios
---------------------------------------------------------------------

The models assume SOLAR SYSTEM METAL ABUNDANCES
(Anders E. & Grevesse N., 1989, Geochim. Cosmochim. Acta, 53, 197)

These  abundances WITHOUT DEPLETION  RELATIVE TO H in number
[n(x)/n(H)] are:

    H  1.000E+00    He 1.000E-01    C  3.630E-04    N  1.120E-04
    O  8.510E-04    Ne 1.230E-04    Fe 4.680E-05    Mg 3.800E-05
    Si 3.550E-05    S  1.620E-05    Ca 2.290E-06    Ar 3.630E-06

        The abundances of metals in the gas phase are depleted according
to the dust-to-gas ratio, which  is specified in solar neighborhood
units by the quantity MU.

                                          [dust/gas]
                                  MU =  ----------------
                                        [dust/gas]_solar

(Detailed description of depletion calculations is in Binette L., Wang J.,
Villar-Martin M., Martin P.G.,  Gladis Magris C.,  1993, ApJ, 414, 535)

We define the total metallicity relative to solar as

   Ztot = Zgas/Zsun  + Zdust/Zsun = 1.

It is 1 because we are considering solar abundances

Every model gives us Zgas/Zsun and therefore,  indirectly we get
the dust metallicity  relative to the solar metallicity (Zdust/Zsun)
for each MU.

  On the other hand,  the relative total (i.e. gas and dust) number of
C to H is given by:

C/H = [Cgas + Cdust]/H = 3.630E-04 (from Anders  & Grevesse)
The models below  give Cgas/H for each MU value. Therefore, we
can calculate Cdust/H, i.e., how many atoms of C are in dust grains,
for every atom of H.

        MU=0.2
        ------

 Zgas/Zsun:  0.9079,   Ztot:  1.0000,  DUST-to-Gas/Loc.ISM:  0.2000 {ISM
Cur}
    H  1.000E+00    He 1.000E-01    C  3.588E-04    N  1.120E-04
    O  8.510E-04    Ne 1.230E-04    Fe 1.819E-05    Mg 3.197E-05
    Si 1.075E-05    S  1.620E-05    Ca 9.924E-08    Ar 3.630E-06

 Zdust/Zsun= 1 - 0.91 = 0.0921.

 Cgas/H = 3.630E-04  ==>  Cdust/H = 3.630E-04 - 3.588E-04 = 4.2E-6.
Depletion
is not important. Most C atoms are still in the gas phase

N/C=1 N/Csolar

        MU=0.5
        ------

 Zgas/Zsun:  0.7689,   Ztot:  1.0000,  DUST-to-Gas/Loc.ISM:  0.5000 {ISM
Cur}
    H  1.000E+00    He 1.000E-01    C  2.064E-04    N  1.120E-04
    O  8.090E-04    Ne 1.230E-04    Fe 4.589E-06    Mg 1.598E-05
    Si 3.236E-06    S  1.618E-05    Ca 7.883E-09    Ar 3.630E-06

 Zdust/Zsun= 0.23

Cgas/H = 3.630E-04  ==>  Cdust/H = 3.630E-04 - 2.064E-04 = 1.6E-04.
Depletion
of C starts to be important.

N/C=1.73 N/Csolar

        MU=1 ([dust/gas]= [dust/gas]_solar)
        ----

 Zgas/Zsun:  0.5350,   Ztot:  1.0000,  DUST-to-Gas/Loc.ISM:  1.0000 {ISM
Cur}
    H  1.000E+00    He 1.000E-01    C  1.148E-04    N  8.896E-05
    O  5.369E-04    Ne 1.230E-04    Fe 9.338E-07    Mg 7.582E-06
    Si 8.917E-07    S  1.022E-05    Ca 4.569E-10    Ar 3.630E-06

Cgas/H = 1.148E-04   ==>  Cdust/H = 2.48e-4

N/C=2.5 N/Csolar

        MU=1.5
        ------

 Zgas/Zsun:  0.2982,   Ztot:  1.0000,  DUST-to-Gas/Loc.ISM:  1.5000 {ISM
Cur}
    H  1.000E+00    He 1.000E-01    C  4.702E-05    N  4.182E-05
    O  2.492E-04    Ne 1.230E-04    Fe 1.702E-07    Mg 2.750E-06
    Si 2.162E-07    S  4.570E-06    Ca 3.256E-11    Ar 3.630E-06

Cgas/H = 1.148E-04   ==>  Cdust/H = 2.89e-4

N/C= 2.8 N/Csolar  The relative abundance of N/C is 2.8 times larger
than in the non-dusty case (solar value)

    ---------------------------------------------------------------------
             Name: dep.ps
   dep.ps    Type: APPLICATION/PostScript
         Encoding: BASE64