By Charles A. Bishop, Ph.D., principal, C.A.Bishop Consulting, Ltd.
Introduction
Vacuum metallizing is not a perfect process. It has been refined over many years, but there is still room for improvement. Currently, there is increased interest in energy efficiency and waste reduction. Packaging materials are under the spotlight and, as metallized film is widely used in packaging, there is interest in the manufacturing process and whether there are opportunities to further reduce the energy used and waste generated. There is a lot of energy used in the mining and production of aluminum and so inefficient use of aluminum squanders energy and is costly.
Aluminum waste can be produced in two ways. There is the regular waste from metallized film which includes the edge trim from the shielded, unmetallized edge of the film as well as the lead-in and tail-end of each roll, which is not metallized or is unacceptable film as the process stabilizes to the desired coating thickness. Similarly, there is the use of aluminum that includes the collection efficiency of the aluminum onto the film, determined by the deposition drum size, source-to-substrate distance and shield geometry. This also should include the discarded end-of-reel wire. The second source of waste would be any of the coated films that fail to meet specification. These may not be totally wasted as they may be sold at lower cost to be used in alternative applications.
To minimize aluminum waste requires changes to the process to improve the collection efficiency but also to improve the coating quality so that fewer rolls fail to meet specification. As with any system or process change there can be improvements in some areas but at the expense of something detrimental to another aspect of the process. In recent years, there has been a renewed interest in the use of induction-heated evaporation of aluminum; this has some easily demonstrable advantages over resistance-heated evaporation of aluminum but also has some disadvantages. This is why I refer to induction-heated aluminum evaporation as getting only “halfway to paradise.”
Downsides of resistance heating
Resistance-heated evaporation boats have a problem that the molten aluminum attacks the binder within the powder ceramic boats. There are flow currents in the molten pool and, as the binder softens, the ceramic particles can be scoured away. The highest erosion is either side of the wire contact point, and the flow takes the particles toward the ends of the molten pool. The particles accumulate around the edges of the pool, particularly at the ends, and these particles can contribute to spitting that is seen as pinholes in the deposited coating [1].
Spitting can be made worse if the molten-pool size changes with any variations in boat temperature or wire-feed variation. This erosion process also means that the boat is a consumable (its lifetime depending on the how hard the boat is used). If the boat is run hard at high temperature and with a corresponding high wire-feed rate, the softening of the binder and erosion will be faster, and so the boat lifetime will be reduced [2].
Induction heating explained
The induction-heated source is a larger circular source that is filled with sufficient aluminum to be able to coat a roll of film. This means there are significant differences to the operation of the sources compared to resistance-heated evaporation. With resistance-heat, the source is heated and the aluminum is fed onto the top of the source, and the aluminum wets the hot surface of the boat. The molten pool is controlled by the wire feed which requires precision control to minimize pool-size variations. In comparison, the induction source heats up the aluminum directly, and the crucible merely confines the molten aluminum; there is no feed system to control and no possibility of the pool size varying.
This confinement and stability of the molten pool means a source of spitting is eliminated [3,4]. This is not to say that it stops all pinholes; it makes no difference to the pinholes caused by particulates carried by the substrate, but it does reduce those associated by the process, which is helpful.
Figure 1 shows a set of induction-heated sources being charged with pellets of aluminum. This is shown with a new set of crucibles and insulation material wrapped around the induction coils. The white insulation fabric becomes blackened with backscattered aluminum during deposition.
FIGURE 1. Resistance-heated sources (left) and induction-heated sources (right). Courtesy of BOBST Manchester, Ltd.
Wired or wireless?
Once deposition is completed, there is another difference between the two source types. With resistance-heated boat sources, the wire feed is stopped, and the aluminum pool allowed to evaporate so the boats are clean of aluminum when the system is returned to atmospheric pressure. There is usually a short delay between switching off the power to the boats and venting the system. This is to allow the boats to cool and so reduce possible thermal-shock cracking of the boats. After venting, there are two different ways the boats may be treated between deposition runs. One approach is to not touch the boats at all between runs unless it is time to replace the boats or to clear out one of the guide tubes for the feed wire. The other approach is to scrape and vacuum the boat surface to clean off any residual particles so as to minimize their possible contribution to spitting in the next deposition run.
In contrast, induction-heated sources have no feed wire to be stopped. Instead, the crucibles will still have some aluminum left in the bottom of the crucible, and the combined thermal mass is higher which means the crucibles cool down much more slowly than evaporation boats. This leads to a different cycle time for the two systems. It is possible for the system to be vented and the crucible recharged with aluminum pellets while the crucibles are still hot.
Figure 2 shows induction-heated sources while they are still red hot (~650° C.). Operators must be trained and have additional personal protective equipment to enable them to replenish the crucibles safely. A comparable process that had a similarly slow cooling time was modified to include a forced-gas cooling cycle to speed up the cooling process [5]. This reduced the hazard of replenishing a very hot source, but it did add to the source complexity and slowed down getting access to the crucibles, thus also reducing some of the advantage of the faster cooling time.
FIGURE 2. Removing a red-hot crucible from the induction coil (left), and A set of used sources showing the coated insulation fabric. This fabric helps reduce heat load to the substrate and heat loss from the source.
There have been techniques developed whereby the deposition sources can be replenished automatically during the process. This includes using a solid rod fed up through the base of the crucible [6,7] or the molten liquid is gravity-fed into the crucible through a pipe connected through the base of the crucible [8,9].
Variations in collection efficiency
The collection efficiency is related to the source-to-substrate distance and the deposition-drum diameter. Ideally the vapor flux should be deposited normal to the substrate surface, but this would make the deposition zone very narrow. With a deposition drum, the theoretical maximum width of the deposition zone would be to allow coating all the way up to the flux, reaching the tangent of the deposition drum. Atoms depositing onto the substrate at grazing incidence have been shown to have poor adhesion and produce porous coatings [10]. This means that in practice there has to be a compromise between the maximum collection efficiency and the adhesion and structure of the coating.
One way to increase the collection efficiency is to increase the diameter of the deposition drum [11]. As the drum diameter increases, if the deposition shield is placed where the depositing atoms meet the substrate at the same angle, then the deposition zone also increases (see Figure 3). The maximum that could be achieved is when the drum diameter is infinitely large, in effect when the substrate is flat (this is the case with free-span deposition) [12].
Free-span deposition has the disadvantage of no substrate cooling, which could limit the process speed, and so using a cooled deposition drum is generally preferable. The choice of drum diameter also is a compromise as increasing the size of the deposition drum can require a larger vacuum chamber and could require a larger pumping set – all adding to the vacuum metallizer’s cost.
FIGURE 3. Schematic showing how increasing the drum diameter can increase material-collection efficiency
In the strip-steel industry, they have taken a slightly different approach that has enabled them to achieve a collection efficiency >95% [13]. They have opted to use a slot source that is positioned close to the substrate so that there is little or no deposition onto any surface other than the substrate. Steel strip has the advantage that it can withstand the heat from the deposition more easily, and so it is coated in free-span. The same source cannot be simply transferred into an aluminum metallizer because the coating applied in such a short distance on the deposition drum would overheat the substrate. What has been done is to reduce the deposition rate from each source but then use a number of sources distributed around the deposition drum (see Figure 4) so that the heat load is spread around the deposition-drum circumference [5].
FIGURE 4. Schematic of multiple slot sources distributed around a deposition drum (left), and a couple of sources showing source chimneys offset for each slot source to improve overall deposition efficiency (right)
New air-to-air coating process
As for most metallizing, the rate limiting factor is the heat-transfer coefficient between the web and the deposition drum. By distributing the heat load over a greater surface area, winding speed can be increased for the same coating thickness. By increasing the material efficiency to >95%, the need to clean the surrounding shields is considerably reduced. These two factors are part of the trade-off for the longer time required for the source cooling.
The strip-steel industry also has adopted air-to-air processing, which allows longer lengths of strip to be coated than could be done if the strip had to be wound into rolls [14]. Historically, air-to-air metallizers were aimed at being used continuously on the end of a polymer-film line, but they failed to work as designed. The problem was that the poor deposition efficiency meant that stray coating coated the deposition shields too rapidly and encroached on the deposition zone, making coating thickness vary with time. The other problem was that when the coater was vented the very thick coating on the shields took much longer to clean. With the higher-efficiency induction-heated source, coating onto the shields is reduced and makes the air-to-air system feasible. Polymer films can be automatically spliced to enable continuous coating for extended periods of time which makes better use of induction-heated sources and further minimizes the impact of slower source cooling.
To maximize the benefits of air-to-air coating, the source needs to be replenished. This can be done by the rod- or liquid-feed systems previously described, but this leads to another problem, namely, the composition of the liquid in the crucible will change with time [14]. In coating steel strip, the sources have been successfully operated continuously for 2-4 weeks. Impurities in the feed material, although in tiny amounts, will accumulate in the molten liquid as it does not get evaporated. Over time, this accumulation can become a problem either as the impurities build up on the surface of the liquid or may, as the concentration increases, alloy with the aluminum and change the composition of the deposited coating.
Conclusion
All the components are in place to make the next step in more efficient metallizing. High-efficiency sources have been used in production, distributed slot sources spread the heat load around the deposition drum, replenishing enclosed slot or jet sources using rod or liquid feeding has been used, air-to-air vacuum coating has been employed in production for metal foil, and polymer-web splicing is routinely done by converters.
Hence, my original statement that we are “halfway to paradise.” Now all we have to do is combine these new technologies to produce the first successful, air-to-air metallizer with a distributed high-efficiency slot or jet source.
References
1. Yappi R.N. “A New Generation of Corrosion-Resistant Evaporator Boats” Proceedings of AIMCAL Technical Conference Europe, Prague, Czech Republic. June 2012.
2. Section 5: Metal evaporation and heated boat data. In “Metallizing Technical Reference,” Editors Bishop, C.A., and Mount, E.M. Publisher, AIMCAL www.aimcal.org
3. Hilberg R. & Decker W. “Metallizer with new inductively heated evaporation sources.” Proceedings of AIMCAL Fall Technical Conference, Oct. 2012.
4. Higashide K. “Evaporation system based on efficient induction-heated crucible – EWA series.” Proceedings of AIMCAL R2R Conference Asia, Daejeon, South Korea. May 2019.
5. Fonseca J.J. & Gorton W. \”A Novel Roll Coater for Double-Sided Heavy Deposition of Magnesium onto PTFE Films.” Society of Vacuum Coaters 38th Annual Technical Conference Proceedings (1995), pp 64-70.
6. Bancroft G.H. “Improvements in processes for coating with metals by thermal evaporation in vacuo and apparatus therefor.” GB701790 (A) June 1, 1954.
7. Smith P.A.J. & Chang P. “Rod-fed electron-beam evaporation system.” European Patent EP0785291, published July 23, 1997.
8. Swisher R, Yadin E. & Pipkevich G. “Web Coating with Lithium: Technical Solution for Metal Anode Structures in Li Batteries.” Society of Vacuum Coaters 45th Annual Technical Conference Proceedings (2002) pp 535-538.
9. Yadin E. & Andreev Y. “Zinc and Magnesium Vapor Generators in a Steel-Strip Coating System.” Society of Vacuum Coaters 42nd Annual Technical Conference Proceedings (1999), pp 39-42.
10. Spencer A.G. & Howson R.P. “The properties of magnetron-sputtered CoNi thin films.” Vacuum. Vol. 36, Issues 1-3, January-March 1986, pp 103-105.
11. “BOBST Unveils New-Generation Metallizer.” Paper, Film and Foil Converter, Nov. 3, 2016. https://pffc-online.com/coat-lam/metallizing/14066-bobst- unveils-new-generation-metallizer
12. Casey F, Ahmed N.A.G. & Ellis G. “Properties of Metallized Film in a Free-Span Web Metallizer.” Society of Vacuum Coaters 33th Annual Technical Conference Proceedings (1993), pp 213-219.
13. Schmitz B., Choquet P., Deweer B. & Chaleix D. “Integration of Jet Vapor Deposition (JVD) in Steel Industry.” Conference proceedings at Vacmess 2005.
14. Zoestbergen E., Maalman T.J.F., Commandeur C. & Goodenough M. “Influence of contamination on the thermal evaporation of a zinc melt.” Surface and Coatings Technology, Vol. 218, March 15, 2013, pp 108-113.
Charles A. Bishop, principal at UK-based C.A.Bishop Consulting, Ltd., holds a Bachelor’s degree in Materials Engineering with a Diploma in Industrial Studies. His research led to developing a process for manufacturing titanium-based bone implants for tendon location. He went on to obtain a Master’s degree and Ph.D. following further research into vacuum-deposition processes. Charles has 40 years of experience in vacuum deposition, mainly onto flexible webs. He writes the “Vacuum Verbiage” Q&A technical column for this publication, and moderates the online “Vacuum Web Coating” Technical Channel. Charles can be reached at +44-1509-502076, email: cabuk8@btinternet.com
Comments