I’ve been away from the blog for a while (too long) while I’ve been writing and revising journal papers. But as this is such a useful way to compile thoughts, references and discourses, here I am again with another review of additive manufacturing studies.
I revisited this 2009 report on AM:
It’s certainly geared to distributed manufacturing rather than personal fabrication, but in this ‘New Industrial Revolution’ we seem bent upon achieving, a lot of these impacts and benefits apply to both scales.
Chapter 8 in the report is called Energy and Sustainability:
“There are a number of clear, potential benefits to the adoption of AM for part production, which could be driven by the sustainability agenda. These include:
- More efficient use of raw materials in powder/liquid form by displacing machining which uses solid billets
- Displacing of energy-inefficient manufacturing processes such as casting and CNC machining with eradication of cutting fluids and chips
- Ability to eliminate fixed asset tooling, allowing for manufacture at any geographic location such as next to the customer, reducing transportation costs within the supply chain and associated carbon emissions
- Lighter weight parts, which when used in transport products such as aircraft increase fuel efficiency and reduce carbon emissions
- Ability to manufacture optimally designed components that are in themselves more efficient than conventionally manufactured components by incorporating conformal cooling and heating channels, gas flow paths, etc.” (page 28)
They also write:
“In principle, some AM processes (such as DMLS, SLM and possibly EBM) use less energy per unit volume of material in the final part than alternative manufacturing processes such as die casting or CNC machining. This appears to have a number of economical and environmental (coupled) benefits. However, very little is known about the waste streams associated with different AM processes. It is known that some polymeric AM processes have very high waste streams (e.g., SLS – powder refresh, FDM/OBJET/SLA – support structure materials). We also know that many metallic processes require significant levels of post-process heat treatment to reduce residual stresses, in addition to considerable energy loss from highly inefficient laser systems and optical tracks. These are waste streams, as they add nothing to the part. Moreover, AM machines are not designed to be efficient. Thermal management is often poor and energy loss is considerable.” (page 29)
A critical issue that “AM can greatly contribute to addressing is the reuse or remanufacturing of parts or products. (page 30)
But – as has been said here again and again – these authors say there needs to be more research and better models for analysis, development of sustainable materials, development of science-based sustainable product design principles.
“Next-generation AM processes must fully demonstrate their incorporation of sustainability principles including energy efficiency and the following major sustainability targets/goals:
- Reduced manufacturing costs, material and energy use, industrial waste, toxic and hazardous materials and adverse environmental effects;
- Improved personnel health, safety and security in AM processes and use of products made by AM; and
- Demonstrated reparability, reusability, recoverability, recyclability and disposability of products produced from AM.” (page 30)
I sense the third one will experience the most roadbumps in this roadmap.
(As an aside, this report was produced at the University of Texas at Austin, who hosts the Solid Freeform Fabrication Symposium, whose Proceedings in turn are much cited.)
Faludi, J., Bayley, C., Bhogal, S., Iribarne, M., 2015. Comparing environmental impacts of additive manufacturing vs traditional machining via life-cycle assessment. Rapid Prototyping Journal 21 (1), 14–33.
This brand-spanking new study came to my attention because the first author is a buddy in the o2 global network. In this paper, the team was studying “the environmental impacts of two additive manufacturing machines to a traditional computer numerical control (CNC) milling machine to determine which method is the most sustainable”.
I like how they word the target and target audience of the paper:
“The goal of this research was to conduct a comprehensive comparison across all major sources of ecological impacts (energy use, waste, manufacturing of the tools themselves, etc.) and all major types of impacts (climate change, toxicity, land use, etc.) so that prototypers and job shop owners can make an informed decision about which technology to purchase or use, and so the makers of 3D printers can understand their priorities for improving environmental impacts.” (By 3D printers here we are talking about FDM and inkjet machines printing in plastic.)
Here the utilization of the machines stood out as most important: if the printers were used constantly or used only occasionally and sitting around on stand-by for long periods of time. “Higher utilization both reduces idling energy use and amortizes the embodied impacts of each machine.” The authors suggest that “the best strategy for sustainable prototyping is to share tools, to have the fewest number of machines running the most jobs each.”
The FDM printers that sit in most Fab Labs seem relatively benign according to this study, especially if they are TURNED OFF when not in use. We’ll see below that the materials they use also seem relatively good, especially in comparison to other AM materials for other processes. But then we see the quote above about material efficiency, where a key objective in the AM roadmap would be “more efficient use of raw materials in powder/liquid form by displacing machining which uses solid billets”. This seems to be the trend – more desktop 3D printers being developed are more about powders than filaments. But keep in mind what Faludi et al. report in this study, that the claims about waste reduction and material efficiency when comparing AM to conventional manufacturing are often overblown because the gains are outweighed by the impacts related to energy (including embodied energy). In short, grab those environmental benefits where you can, but really put your efforts to where the real impacts are. Here, don’t get a printer in-house until you have enough work to keep it running efficiently.
Short, D.B., Sirinterlikci, A., Badger, P., Artieri, B., 2015. Environmental, health, and safety issues in rapid prototyping. Rapid Prototyping Journal 21 (1), 105–110.
This article is in the same journal issue as the one above; it focuses especially on health and safety issues in rapid prototyping. One gets the picture that a rapid prototyping facility in an industrial context would probably have a lot of these issues in hand, such as the ventilation called for, but even this is not certain. In Fab Labs and makerspaces this is rare.
According to the authors,
“The modeling materials for the FDM systems are all inert, nontoxic materials developed from a range of commercially available thermoplastics and waxes. However, it is important not to exceed melting temperature recommendations to avoid the fumes produced during processing. They may cause eye, skin and respiratory tract irritation. Moreover post-processing operations such as grinding, sanding or sawing can produce dust, which may present an explosion or respiratory hazard.” So not a giant worry, but SLA machines and other technologies are beginning to enter Fab Labs at a rapid pace and that is another potential can of worms.
And then there are the waste management problems – especially through to the end of the life cycle when the product / material is downstream. What happens then? Seems it’s time these kinds of issues were taken up in the maker community. It wouldn’t take much to start compiling and distributing Health, Safety and Environment watchlists for these small-scale prototyping environments. Put a few posters up. Distribute the MSDSs (Material Safety Data Sheets). Open a window. That kind of thing.
Kellens, K., Renaldi, R., Dewulf, W., Kruth, J., Duflou, J.R., 2014. Environmental impact modeling of selective laser sintering processes. Rapid Prototyping Journal 20 (6), 459–470.
In this paper, based on LCI data, “parametric process models are developed allowing to estimate the environmental impact of the manufacturing stage of SLS parts”. The hope is that such work can improve future design-for-SLS processes, especially with regards to reducing the environmental impacts of waste materials and electricity consumption – but not just the design of 3D printed products, also the design of the equipment itself.
The importance of considering environmental impact in the design stage is also considered in this article:
Le Bourhis, F., Kerbrat, O., Dembinski, L., Hascoet, J., Mognol, P., 2014. Predictive model for environmental assessment in additive manufacturing process. Procedia CIRP 15, 26–31.
These authors emphasize how Design for Additive Manufacturing can optimize for the specifics of AM processes (such as the ability to produce complex shapes), and – at least in their analysis and system boundaries – electricity consumption is not always the most impactful factor compared to other flows (powders and fluids, in the metal deposition process they studied).
Yes, on the surface that might seem to conflict with what Faludi et al. conclude above, but in these studies the system boundaries are much smaller and they are concentrating only on the manufacturing stage in order to inform part and process design.
Most of the same authors above also published this article:
Le Bourhis, F., Kerbrat, O., Hascoet, J., Mognol, P., 2013. Sustainable manufacturing: evaluation and modeling of environmental impacts in additive manufacturing. The International Journal of Advanced Manufacturing Technology 69 (9-12), 1927–1939.
Here they tried out a couple of different ways to produce a part while doing the environmental evaluation (also considering energy consumption and material flows).
Let’s also put this article in the ‘optimizing design’ section:
Ratnadeep, P., Anand, S., 2012. Process energy analysis and optimization in selective laser sintering. Journal of Manufacturing Systems 31 (4), 429-437.
If you are interested in “a methodology to calculate the laser energy of a part manufactured in the SLS process and to correlate the energy to the part geometry, slice thickness and part orientation”, then check out this article. The lit review section also has heaps more citations to energy studies in additive manufacturing.
And another well-cited energy study here:
Sreenivasan, R., Goel, A., Bourell, D.A., 2010. Sustainability issues in laser-based additive manufacturing. Physics Procedia 5, 81–90.
Let’s let the authors tell us what they were doing:
“The goal is to reduce energy consumption in SLS of non-polymeric materials. The approach was to mix a transient binder with the material, to create an SLS green part, to convert the binder, and then to remove the open, connected porosity and to densify the part by chemical deposition at room temperature within the pore network.”
I’m just going to skip over that level of detail. Suffice to say that – given how many researchers use the Eco-indicators – let’s be happy that so much work is *also* done developing these evaluation tools and metrics.
ADDED Feb 2015:
Almost forgot this. Must be because it was sitting right in front of me on my desk.
Baumers, M., Tuck, C., Wildman, R., Ashcroft, I., Rosamond, E., Hague, R., 2013. Transparency Built-In. Journal of Industrial Ecology 17, 418–431.
This nicely dramatic title for an academic paper comes from the authors’ description of AM as inherently transparent: it’s a “one-stop” manufacturing process, so even for a complex design there’s no need for additional steps like making moulds or dies or other tooling. Sometimes just some finishing steps. This makes measuring the energy flows in production a lot easier, and in fact, it seems considering cost efficiency when planning AM builds and production processes “is likely to lead to the secondary effect of minimizing process energy consumption”. This doesn’t necessarily happen in conventional manufacturing, so immediately we see sustainability opportunities. In this study the authors present a methodology for “design for energy minimization”: a tool to estimate process energy flows as well as costs, using Direct Metal Laser Sintering experiments to test it.
Then there is the JM Pearce gang, who are quite prolific. Here are four articles, but there are more out there.
Krieger, M.A., Mulder, M.L., Glover, A.G., Pearce, J.M., 2014. Life cycle analysis of distributed recycling of post-consumer high density polyethylene for 3-D printing filament. Journal of Cleaner Production 70, 90–96.
This study promotes not only distributed production but distributed recycling: the authors claim that there are benefits to actors producing their own printer filaments from post-consumer plastics with their own low-cost (and open source, of course) shredding-extruding systems compared to a centralized recycling system. This is especially in areas where the population is not so dense, since recycling collecting and transport is impactful. I’m not going to dig into their LCA procedures to find holes at this point; someone else can do that. I’m more interested in what these researchers want to promote.
Last year I was talking to a Fab Lab manager who also works with an industrial filament manufacturer, and she was sceptical about these homegrown ‘recycle-bots’. She said it’s challenging enough to make consistent-quality filament that works without glitch in your printer at the commercial scale – how is that possible with these grassroots systems? Seems to me it would take a level of expertise that is itself not widely distributed.
Anyway, the paper presents some interesting scenarios and is quite a new take on this New Industrial Revolution the maker movement is supposed to represent – where a cottage industry could develop around the collection and reprocessing of plastic waste into, for example, spare parts and other Useful Things. I seem to remember a scene like that in Ian McDonald’s Brasyl, and these authors do mention some initiatives in the global South, but they also intend it to develop and benefit regions in the North.
Baechler, C., DeVuono, J., Pearce, J.M., 2013. Distributed recycling of waste polymer into RepRap feedstock. Rapid Prototyping Journal 19 (2), 118-125.
In this earlier paper, a Pearce crew report on the filament quality they made in the RecycleBot: “Filament was successfully extruded at an average rate of 90 mm/min and used to print parts. The filament averaged 2.805 mm diameter with 87 per cent of samples between 2.540 mm and 3.081 mm.” The problems are quite well documented too, as well as the design of the device itself. You could get your hands on a windshield wiper motor and the other components and make your own.
Kreiger, M., Pearce, J.M., 2013. Environmental Life Cycle Analysis of Distributed Three-Dimensional Printing and Conventional Manufacturing of Polymer Products. ACS Sustainable Chemistry & Engineering 1, 1511–1519.
“This study evaluates the potential of using a distributed network of 3D printers to produce three types of plastic components and products. A preliminary life cycle analysis (LCA) of energy consumption and greenhouse gas (GHG) emissions is performed for distributed manufacturing using low-cost open-source 3D printers and compared to conventional manufacturing overseas with shipping.” The researchers used a RepRap printer, calculations for both PLA and ABS, as well as for conventional electricity and power from a solar photovoltaic source. The objects were a toy (a polymer block fabbed locally vs a wooden block made in and shipped from Switzerland); a water spout (a locally fabbed spout that is intended to fit onto an existed, reused, 2L bottle vs an entire watering can made in China); and a citrus juicer.
The authors have a number of ‘tips’ for making distributed manufacturing of this type even more sustainable, such as using solar PV systems, controlling temperatures during printing to enhance energy efficiency and taking recycled filaments more prominently into use. PLA is seen to have benefits over ABS, being a bio-based polymer and needing lower temperatures in printing, hence affecting energy consumption. And using a local 3D printer means you can control the design (and fill) of the product, optimizing the use of material.
Nevertheless, some of us have discussed the article and agree that the choice of objects is a bit odd and we wonder about the comparability of the mass manufactured choice vs the fabbed object.
Anyway here again we have the clear promotion of open hardware, which is not so common as a meta-level agenda in AM studies.
Wittbrodt, B. T., Glover, A. G., Laureto, J., Anzalone, G. C., Oppliger, D., Irwin, J. L., Pearce, J. M., 2013. Life-cycle economic analysis of distributed manufacturing with open-source 3-D printers. Mechatronics 23 (6), 713−726.
And there we have it right in the title: open source 3D printers. This is published as a Technical Note in this journal and is described as a “life-cycle economic analysis (LCEA) of RepRap technology for an average US household”. They took 20 designs from Thingiverse and after some numerical wizardry concluded that the household in question could save hundreds to thousands of dollars a year if they printed this stuff (a razor, a spoon rest, a phone dock, a phone case, shower curtain rings etc etc) instead of buying it. Again it is interesting to read for the plethora of positive scenarios they spin about distributed open source 3D printers, if not the results of the study itself.
Tabone, M. D., Cregg, J. J., Beckman, E. J., Landis, A. E., 2010. Sustainability metrics: life cycle assessment and green design in polymers. Environmental Science & Technology 44 (21), 8264−8269.
This is not about Additive Manufacturing per se, rather polymers, but it’s worth a check to see a summary of 12 polymers and the authors’ summaries of them – regarding their environmental impact (via LCA) and compliance with “green design principles” (12 Principles of Green Chemistry and 12 Principles of Green Engineering).
For instance, for the biopolymers they studied, the materials’ production resulted “in the highest impact in 5 of the 10 categories: ozone depletion, acidification, eutrophication, carcinogens, and ecotoxicity”. The biopolymers also “adhere well to several green design principles: the use of renewable and regional resources, low emissions of carcinogens, and low emissions of particulates”. However some of the fossil-fuel-feedstock polymers fared surprisingly well compared to the bio-based materials: “Polyolefins (PP, LDPE, HDPE) rank 1, 2, and 3 in the LCA rankings. Complex polymers, such as PET, PVC, and PC place at the bottom of both ranking systems.” It is therefore not a foregone conclusion that using PLA in your printer is clearly the environmental choice, due to the problems with how it’s produced.
As we saw above with the study on “distributed recycling”, maybe makers should also get involved in the sustainable ‘growing’ and production of their own biobased plastics, avoiding petroleum fertilizers. Could give whole new meaning to being “off-grid”. We could set up a village network. I’ll grow the potatoes for people food and use the waste to make PLA, and I’ll trade you a bundle of filaments for some cloth that someone has woven from linen – derived from the flax field next door. I guess these fields will be on the roofs of our blocks of flats / apartment buildings. And, depending on how much the sea level has risen by then, it’s possible that I have to transport that filament to you by boat. Luckily I live on the third floor of my building.