Index Home About Blog
From: (Bruce Hamilton)
Subject: Re: fHuman vs. natural influences on the environment
Date: Sat, 07 Sep 1996 18:50:06 GMT (charliew) wrote:

>In article <50ooc8$>,
>   jbh@ILP.Physik.Uni-Essen.DE (Joshua B. Halpern) wrote:

>>Well, this is the old, just because we don't know everything,
>>we don't know anything.  Charlie, in that chemistry set you
>>work with, do you, or anybody know everything about any of
>>the reactors and the reactions that are going on in that
>>foul mixture you call oil.  Do you even know the EXACT
>>composition of today's deliveries?  But somehow, you do
>>make gas.

>In the refinery, we apparently have a luxury that you do not
>have.  We can perform an "experiment" numerous times to
>validate hypotheses.  Since crude refining has existed for
>about a century, there has been ample time to formulate a
>practical approach to refining crude, even though we do not
>know everything about the process.  Thus, we don't need to
>know everything in this case, because there is enough
>feedback with the "real world" to guide us in the correct

The petroleum industry has always been dominated by the
performance standards of their products, and many of
the developments were hit and miss - consider the huge
trials to discover effective anti-knock compounds - they
tried every chemical they could think of - including melted
butter, and after many years, Midgley evaluated the
known results against the periodic table to come up with
the alkyl leads. Even today, we still don't know all the
mechanisms of hydrocarbon reactions that produce knock.
Not a good history of rigid scientific methodology  :-), and
Midgley is one of the chemists I most respect!.

The industry used performance standards, not compositional
standards.  Even today, with my hundred metre capillary
column that sees 500+ compounds in gasoline - it still
does not reflect the complete knowledge of the fuel's
composition. All gasolines are sold on the basis of a set of
clearly defined performance properties that, until very
recently, had virtually no significant hydrocarbon compositional
constraints. As long as the fuel met those performance
criteria - Research Octane Number, Motor Octane Number,
( specific engine tests are the final arbitor of product quality
in almost all countries, although refineries may use other
 on-line octane measuring systems for process control ),
Volatility according to season, stability ( no gum-forming ),
etc. etc., the fuel was considered a marketable product.

Only since about 1970, when gas chromatography was
 widely available, and we could easily monitor individual
HCs in the refinery streams, products and environmental
samples have we started to place significant constaints
on composition of the fuel. You can review any refining
text for the first 70 years of your 100 years, and there
will be minimal compositional information. In other words,
the experiments were performed and monitored to
determine the effects on the specified performance
properties - very specific information - used for process
development and optimisation. That information is
virtually irrelevent for today's gasoline.

The petroleum industry has always been conservative in
it's approach to refining and product research, if the product
is marketable, sell it. Almost all major changes to gasoline
have been induced by regulators or the automotive industry,
often in the face of aggressive rearguard actions by the
industry.   Consider the huge battles about the toxicity
of alkyl leads in the 1920s - the government public health
office was right, but the economic imperative was even
more important, and the industry managed to hang onto
lead until the 1970s - in the face of sound scientific data
that had been accumulating since the 1920s. Where were
the experiments being performed? - not by the industry.

Not surprisingly, the industry has been very reluctant
to perform experiments that would adversely affect the
profitability or viability of their operation - even in the
face of significant scientific evidence. Given that
history, surely it is better to constrain the experiment
until we understand the effects of continuing. Note also
that the API has a strong presence in the anti-IPCC camp.

>Was that a big enough "cut" to satisfy you?

Yes, now about the automatic word-wrapping feature :-)

          Bruce Hamilton

Index Home About Blog