Small tweaks to the example code I posted earlier accommodates the "real" file format:

use strict;
use warnings;

my $fileStr = <<STR;
Property  job 1 : Activity coefficients ln(gamma) ;
 Settings  job 1 : T= 298.15 K ; x(6)= 1.0000 ;
 Units     job 1 : Concentrations x : mole fraction ;

  Nr Compound                                 ln(gamma)
   1 F002                                    4.66656083
   2 F011                                   26.13597035
   3 F101                                   32.47411476
   4 F11-1                                  29.58963453
   5 F111                                   30.24092207
   6 h2o                                     0.00000000
   7 acetonitrile                            2.14102090
   8 chlorobenzene                           8.72282917
   9 chcl3                                   6.98143674
  10 cyclohexane                            10.20251798
  11 1,2-dichloroethane                      6.32324557
  12 ch2cl2                                  5.50767091
  13 1,2-dimethoxyethane                     2.56706253
  14 n,n-dimethylacetamide                  -1.64673734

 Property  job 2 : Activity coefficients ln(gamma) ;
 Settings  job 2 : T= 298.15 K ; x(7)= 1.0000 ;
 Units     job 2 : Concentrations x : mole fraction ;

  Nr Compound                                 ln(gamma)
   1 F002                                    1.69945785
   2 F011                                    0.74578421
   3 F101                                    2.67268035
   4 F11-1                                   1.64808218
   5 F111                                    1.95840198
   6 h2o                                     2.08530828
   7 acetonitrile                            0.00000000
   8 chlorobenzene                           1.08379112
   9 chcl3                                   0.46576330
  10 cyclohexane                             3.71606919
  11 1,2-dichloroethane                     -0.02354847
  12 ch2cl2                                 -0.23798262
  13 1,2-dimethoxyethane                     1.22044280
  14 n,n-dimethylacetamide                   0.44524110
STR

open my $fIn, '<', \$fileStr or die "Couldn't open \$fileStr: $!\n";

# Look for the empty line between records
local  $/ = "Nr Compound";

while (defined (my $record = <$fIn>)) {
    my @lines = grep {/^\s*\d+\s+\S+\s+-?\d+\.\d+/} split "\n", $recor
+d;

    next if !@lines;

    my %recordData = map{/\d+\s+(\S+)\s+(\S+)/; ($1, $2)} @lines;
    my @solvents = grep {!/^F\d+/} keys %recordData;
    my @fractions = grep {/^F\d+/} keys %recordData;
    my ($zeroSolvent) = grep {0.0 == $recordData{$_}} @solvents;

    print "${zeroSolvent}_$_ => $recordData{$_}\n" for @fractions;
}
[download]

Prints:

h2o_F101 => 32.47411476
h2o_F011 => 26.13597035
h2o_F002 => 4.66656083
h2o_F11-1 => 29.58963453
h2o_F111 => 30.24092207
acetonitrile_F111 => 1.95840198
acetonitrile_F101 => 2.67268035
acetonitrile_F11-1 => 1.64808218
acetonitrile_F002 => 1.69945785
acetonitrile_F011 => 0.74578421
[download]

The key differences are choosing a different string to recognize records and only keeping interesting lines for processing from each record.