Monday, July 25, 2011
Less is more, part 3
Where does the material go?
California is used to the concept of getting more from less. We only have to recall now Governor Jerry Brown's comments when he was governor the first time from 1975 to 1983 and declared that University employees shouldn’t complain about low pay because, as academics, they were getting “psychic rewards.” As a UC faculty member we are now seeing dramatic reductions on state support of education as the campuses, specially Berkeley, charge forward to keep our programs strong. Psychic indeed!
We are not talking about psychic rewards here or starving critical institutions!
We've been talking about making better use of what we start out with (the yield or "buy to fly ratio" approach) as well as process and product design for better results with lower impact.
I have been making good use of Professor Julian Allwood's research at Cambridge University to make the first point. His WellMet 2050 study is inspirational. We'll see more from that below.
Others, like the Air Force SAMI as well as corporate programs are making inroads on this as we discussed in the last posting. TMS (The Minerals, Metals and Materials Society) has produced, with support from the DOE and a host of others, a report in January 2011 titled "Linking Transformational Materials and Processing for an Energy-Efficient and Low-Carbon Economy: Creating the Vision and Accelerating Realization." You can download this report from TMS.
The report presents a prioritized set of new products and technologies prepared by TMS working groups focussed on the following themes:
- Functional Surface Technology
- Higher Performance Materials for Extreme Environments
- Multi-Materials Integration in Energy Systems
- Sustainable Manufacturing of Materials
It is a comprehensive forward-looking review of technology.
There are some obvious (to me) gaps however. For example, in the focus on sustainable manufacturing of materials the group highlighted:
- Net-Shape Processing of Structural Metals (that means making things to a final or near final shape without removing material - such as forging)
- Additive Manufacturing of Components and Systems (combining process or materials in reduced number of operations; but not necessarily rapid prototyping - the term usually associated with additive manufacturing)
- Low Cost Processing and Energy Reduction Technology for Metals (reducing the energy requirements for primary processing of metals like titanium, aluminum and magnesium)
- Separation of Materials for Recycling (promoting increased recycling rates), and
- Real-Time Sensor Technology for Gases and Molten Metals (feedback for process optimization and control).
I did not see much reference to increasing yield (except in the net-shape area) but certainly not at the level of importance to address the tremendous losses pointed out by Allwood.
So, let's pick there from the part 2.
The last posting presented a graphical representation of the cumulative yield (output over input) through several process steps and the accompanying cumulative process energy (energy/ton of material input). During the process steps, typically, yield is reduced (meaning material ends up on the shop floor) and, due to processing and material loss, the cumulative energy increases. As noted in the WellMet 2050 report "Going on a Metal Diet" that is the basis of this discussion "…these graphs will show that the (already energy efficient) process of liquid metal production dominates the cumulative energy build-up but yield losses in the downstream supply chain can increase the embodied energy in the final component by a factor of up to 10." Up to 10x increase due to downstream yield losses!
I mentioned that Allwood's study had looked at four case studies. I don't want to repeat the report here (and encourage you read the whole report) but let's look at the cases for aluminum. The figure below shows the cumulative energy (reference to the original liquid aluminum) as a function of cumulative yield (actual product to input liquid metal) for three of the case study products (and click on the
image for better detail) investigated - car door panel, aircraft wingskin panel and beverage can. The various steps in production, from liquid, are shown by the open circles on the graph.
Lines connecting the circles going vertically (or more vertical) indicate processes that preserve yield (that is, less wasted material). Lines going horizontally, (or more horizontal) indicate processes that reduce yield (that is, waste material). This is not necessarily to imply that the material is wasted gratuitously but that the inherent aspects of the process are not able to make efficient use of the material.
For instance, the door panel example, indicates that from cast ingot to stamped panel there is a tremendous loss of material (yield from 1 down to 0.4 meaning 60% of the material not ending up in the product) The actual "buy to fly ratio" would be better than the 0.4 shown on the graph for door panels since the auto manufacturer is unlikely to by aluminum in liquid form. More likely the material enters production as cold rolled coil (at approximately .7 yield) and then is converted to the panel. So, buy-to-fly is closer to 50% form the auto manufacturers perspective.
But you see how this does not tell the complete story - specially with respect to the cumulative energy - since most of that is in the liquid to cold coil processing.
Beverage cans are a similar story. Most material is "lost" from the ingot to cup stage. The can producer likely gets the stock as cold rolled coil. From there, the losses due to can production take the yield from approximately 0.7 to 0.55. From the perspective of the can maker, perhaps, this is a reasonable buy-to-fly ratio.
The figure below, from Kalpakjian and Schmidt's manufacturing text (presented on line), shows a schematic of can making from the original blank through the drawing process and the addition of the cap.
There are two major sources of process related material loss - the blanking of the disks used to start the forming process (think cutting circles out of square sheets) and the disk to first cup process due to the requirement to be able to hold on to the end of the disk during this first stage. There is some trimming at the end also. A similar process is required for the lid which, although not as deeply "drawn" still starts as a circle from a square sheet.
The least efficient from a materials efficiency point of view is the wing skin panel (and recall our earlier comments about aerospace buy-to-fly ratios). This ends up with an overall yield (from melt) of less than 10%. Assuming the manufacturer gets the material as rectangular plate (at about 45% yield) their part of the process yields a buy-to-fly of around 25%.
Recycling, oft mentioned with aluminum and other metals, will help, some. The problem is that with "low yield products" a lot of the material "going back into the pot" will not be post consumer waste but production waste. In the case of aerospace components most of that waste is in the form of metal chips removed to get the desired shape. Granted, aerospace is a special case due to the requirements of the product but this is a lot of material to leave on the shop floor. Composite materials will try to address this but they have material efficiency issues as well.
Next time we'll start the discussion about getting more from the material and, specifically, analysis tools to help us do that.