WeL ¼

L Fr L

r gde

After some substitutions, we arrive at the following alternate  relationship  for am/ad

where

AM  ¼ AMAL (46)

 am

。qV。a—v。lV。bþv

Re—b—v—e

Fr/=2—a—b=2þv=2—e=2

A0 A  =A

(47)

ad  / qL

lL ð

L Þð  L Þ

V  ¼   V L

× ðWeaþbþe v

Our final equation for am/ad     is

L ÞðReVÞ ð41Þ

Therefore, it is not generally possible to deconvolute experimentally measured hHETPi data from binary distilla- tion experiments into unique correlating expressions for ky, kx,  and  am  from  the  hHETPi dataset  alone。  To  resolve the

A B

 am

A   。qV。  。lV。

ReX

FrD

WeE

ReU

(42)

‘‘Eq。  45’’  dilemma,  the  absolute  magnitude  of  any  one   of

ad  ¼   M     q

lL ð

L Þð

L Þð

LÞð VÞ

the  front  factors—AV,  AL,  and  AM—appearing  in  Eq。      44

must be established。

Equation  44  contains 13  fitting  coefficients。 Ten  of  these

Note  the  appearance  of  the  groups  (qV/qL)  and  (lV/lL)  in

our equation for am/ad。 These groups have either not been con- sidered in the development of other correlations for the mass- transfer area or the exponents A and B have been assumed to be zero a priori。3–5  We choose to retain these two groups。

For sheet metal structured  packings,  we adjust Eq。  42  in an ad hoc way to allow it to account for the corrugation in- clination angle

appear as exponents on dimensionless groupings。 Trying to fit data to an equation in which the same physical quantity simul- taneously contributes to one or more dimensionless groups in the equation presents statistical difficulties。26 Meaningless cor- relations have been known to result when  variations  in the data to be fit and in the physical parameters making up the dimensionless groupings are sufficiently large and random。27

To try to avoid the majority of these statistical difficulties, we shall fix the values of several of the power-law    exponents

 am

。q  。A。l  。B

A

ReX

FrD

WeE

。  cos 。t

ReU

in Eq。 44 by appealing to verified results from other types of

ad  ¼   M     q

lL ð

L Þð

L Þð

LÞð