Page 40 - Materials Chemistry, Second Edition

AN OVERVIEW OF THE LIFE CYCLE ASSESSMENT METHOD 21

• For some products, upstream production processes or down-
stream disposal processes may be difficult to quantify;
• For some unit processes, the balance equations become impossible
due to the fact that these processes produce not just one product
but several co-products.

The first issue can be solved by a procedure known as cut-off, the second
one by allocation.
Cut-off is a solution to the problem that the system is theoretically
infinitely large. To produce a TV, we need machines, and these machines
are produced by machines, and these machines in turn need machines, etc.
But of course we have an intuitive idea that some very distant upstream
processes will be quite unimportant to the study. This means that we will
cut-off certain inputs: although we know that something is needed, and we
sometimes even know how much is needed, we do not go into the trou-
ble of specifying how these inputs are produced. It turns out to be diffi-
cult to specify reliable criteria when cut-off is allowed, or to estimate how
large the error is when a cut-off is made. Criteria on the basis of negligi-
ble contribution to mass or cost (e.g., smaller than 1%) often work pretty
well, but occasionally have been shown to yield large errors. Alternatively,
estimates of missing parts by means of similar processes (e.g., estimating
the production of a freezer by using production data for a refrigerator), or
by economic input-output tables may be helpful. Another approach is to
conduct a difference analysis: in comparing a cathode ray tube (CRT) with
liquid crystal display (LCD) television we may leave out the broadcasting
processes.
The second problem, co-product allocation, has given rise to one of the
biggest controversies in LCA theory. The example problem can be demon-
strated thus: If a transportation process needs gasoline, the upstream unit pro-
cess is a refinery that produced not only gasoline, but also diesel, kerosene,
heavy oils, and many other products. The direct impacts (from pollutants like
C0 2), but also the flows to and from other processes that may lead to impacts
(e.g., from oil drilling) may be argued not be attributable to gasoline only, but
need to be distributed over gasoline, diesel, and all the other co-products. This
is hardly contested, but the debate focuses now on how to do this. To make
the issue more concrete, the question at hand can be stated as: How much of
the C0 2 from a refinery is allocated to the gasoline? Different schools have
provided different arguments, none of which have been completely compel-
ling so far. Some solutions lead to strange results, while other solutions may be
very difficult to carry out (e.g., for lack of data or appropriate software). Still
others are rejected outright by many experts. To complicate the issue, the prob-
lem does not only occur in unit process that produce several co-products, but
also in unit processes that treat more than one type of waste, or where waste
is recycled into a useable good. It is not even agreed upon if the multi-output
case, the multi-input case, and the recycling case must be treated in the same
way or not.