When parsing these reactions

H+CH2+M=CH3+M			 6.E14		 0.0		0.0
H+CH2(+M)<=>CH3(+M)                      6.0E+14     .000        .00
     LOW  /  1.040E+26   -2.760   1600.00/
     TROE/   .5620  91.00  5836.00  8552.00/

it seems that ck2yaml will interpret them both as pressure dependent and set pressure_dependent=True on both, but will only set the third_body attribute to M in the second reaction (the first reaction will have third_body=None).

When running through check_duplicate_reactions it will then reach the lines

                elif (r1.third_body.upper() == 'M' and
                      r1.kinetics.efficiencies.get(r2.third_body) == 0):

and give an AttributeError: 'NoneType' object has no attribute 'upper' because r1.third_body is None.

Presumably the fix is for the parsing of the first reaction to set third_body='M' ??

But putting

                if kind == 'third-body':
                    third_body = True
                    falloff3b = 'M'  # <<<< new line

in the parse_expression method means we're setting a variable named "fall-off third body" for something that's technically not a "fall off" reaction. And may have other issues.

Or is the answer to change how check_duplicate_reactions works?

Should

H+CH2+M=CH3+M			6.E14		 0.0		0.0
H+CH2(+M)<=>CH3(+M)                      6.000E+14     .000        .00
     LOW  /  1.040E+26   -2.760   1600.00/
     TROE/   .5620  91.00  5836.00  8552.00/

be considered duplicates anyway?

I know that Cantera does not consider these to be duplicates:

H+CH2=CH3			6.E14		 0.0		0.0
H+CH2(+M)<=>CH3(+M)                      6.000E+14     .000        .00

System information

• Cantera version: 2.4
• OS: Mac OS X
• Python: Python 3.7

Attachments
mechanism.txt
thermo.txt