The business plan
Turn back the clock six years to and it was an entirely different story. Powerlase was a fledgling firm newly spun out of London's Imperial College anxious to attract venture-capital funding to get its ideas off the ground. "Initially we were producing extreme ultraviolet (EUV) sources for semiconductor lithography, but it was always a long-distance market," Mike Mason, the company's vice-president of technology, explained. "We also identified that our lasers could be used for flat-panel display manufacture and that was an emerging market at the time. In order to make money in the short term, we concentrated on this application. EUV is still there and will hopefully come to fruition in the next few years."
If you want to learn more, please visit our website.
Today, after attracting substantial venture capital funding as well as volume orders for its products, Powerlase is operating out of a 20,000 ft2 facility, has a sales and service centre in Korea, and a workforce of around 50. "We have been careful not to explode too fast," said Mason. "When I started in , I was the 10th employee, so it has been organic growth."
So what does Powerlase see as the secret to its success? "The key thing, and this is less common than you might think, is that we are entirely customer driven," explained Mason. "We talk to our customers on a regular basis – particularly our Asian customers who value face-to-face contact very highly. We try to give the customer exactly what they want."
Being a relatively small company compared with its competitors, Powerlase also rates its flexibility as a key advantage. "Our customers guide us and we can provide them with what they want, when they want it and how they want it," added Henry. "It is easy to sell a clever piece of technology once, but to sell it in volume and to get repeat orders requires people to have confidence and believe in you. That is what we have achieved."
Scalable technology
Powerlase's expertise lies in developing Q-switched DPSS lasers that produce nanosecond pulses with high average powers and high peak powers. At the heart of the laser is a gain module, the precise design of which is a carefully guarded secret.
"We use diode lasers to side pump a YAG rod inside the gain module," explained Mason. "The key to our performance is in the detailed design. This module provides a scalable architecture that gives us the capability to produce a range of powers and frequency convert to different wavelengths. We have IP covering the cavity formed around the module and the way that we cope with the high powers produced via acousto-optic Q-switching, which is non-trivial."
The end result is a DPSS laser system that produces pulses with a duration of tens of nanoseconds at repetition rates of tens of kilohertz at , 532 and 355 nm. The company says that its infrared lasers can produce average powers ranging from 200 to W, peak powers in excess of 15 MW and pulse energies in excess of 300 mJ.
"The pulse durations are really quite short at low repetition rates," added Mason. "At 3–6 kHz they are in the range of 25–35 ns, which boosts the intensity of the process, reduces thermal impact on the substrate and improves quality."
Rapid laser patterning
PDP manufacture has been an incredibly successful market for Powerlase. In this application, the company uses its laser systems to selectively remove and pattern complex shapes into the thin layer of ITO that sits on the glass superstrate (the front panel on the plasma display).
The process involves collimating the beam, passing it through a homogenizer, imaging it onto a mask and re-imaging it onto the superstrate at a given magnification. Henry explains that the company typically uses orthogonal microlens arrays to homogenize the beam, which give a top-hat energy distribution rather than a Gaussian distribution. The end result is features that have an edge resolution of just 1 µm.
"We can pattern well over 1 mm2 using a single pulse and then you scan this in order to build up the overall structure," explained Henry. "If you are patterning 1 mm2 per pulse and you are processing at 6 kHz then you can scan at 6 m/s. Our customers typically try to turn out plasma displays at less than 30 s per unit on a production line, which requires very fast processing."
And to improve the speed and throughput even further, most systems have multiple lasers running in parallel. For example, a manufacturer may want to put eight 42-inch television panels, or six 50-inch panels, on a Gen8 glass substrate (approximately 5 m2 in size). This could require six or eight lasers, each one processing a different panel area on the substrate simultaneously to increase the productivity.
Displacing wet-etching
Given the price pressures that manufacturers of all variants of flat-panel displays are facing, it is easy to see why cost-effective processing techniques are a priority. Powerlase is keen to stress that its RLP scores on two points: it is both technologically viable and commercially convincing.
Henry explains that there are a number of factors to consider when comparing RLP with wet-etch lithography. For example, the chemical costs in wet-etch lithography are high and the acids become contaminated with the removed materials. In comparison, RLP does not use any chemicals, only de-ionized water to rinse the processed substrate.
"It is also difficult to get a chemical process to be consistent over a large area and there is significant yield loss," added Henry. "Our RLP PDP yields are greater than 99%, whereas wet-etch can be anywhere between 80 and 85%. RLP is less sensitive to variations in the thickness of the films or the topography of the glass."
The number of steps in the process is also a consideration. While RLP requires just two (laser processing followed by rinsing), wet-etching can involve between six and 10 steps. "It is important to remember that flat-panel display manufacturing takes place in cleanrooms," added Henry. "If you have to build a vast cleanroom to accommodate up to 10 processing stations, then that is much more expensive. All of our lasers are fibre-delivered so it is common for our customers to have the laser outside the cleanroom so that they can save footprint and streamline their requirements."
Emerging applications
Powerlase was undoubtedly in the right place at the right time to push its technology into the PDP market. Today, many other markets involving thin films deposited on plastics, glass and a range of other materials are lining up as fresh opportunities. One that Powerlase is particularly excited about is solar cells, as manufacturers are continually looking for ways to lower production costs.
"A large variety of thin-film solar cells need to be patterned and the industry would like to move away from wet-etch," said Henry. "We are patterning layers to achieve all sorts of effects. It is also possible to process bulk silicon solar cells. It is a huge market for us and very interesting. The technology that we have perfected for PDP transfers directly to solar cell."
Other applications include mobile displays, OLEDs and even automotive, where in future the radio aerial could be incorporated into the car's windscreen. Powerlase also expects to be returning to its roots as the semiconductor industry begins to approach the processing nodes that require EUV lithography.
"People are now starting to take the approach of producing EUV using multiple solid-state lasers seriously for the first time," concluded Mason. "We are seeing a lot of traction and the field is accelerating. There will come a point when the industry cannot achieve the next node and must have a technology to move forward. EUV is the front runner and we have to be there at the right time."
In the previous article, I talked about how to choose the right partner to support your process and integration needs at the very early stage of discovering laser as a potential solution to your product development or manufacturing needs.
In this article, let's talk about one of the most common and important questions that we get asked all the time, which is “what's the right type of laser for my application?” And this question is asked in many different ways:
what type of laser is best to be used?
For more information, please visit Microtreat.
What type of laser is best to be used?ser?
Should we be using CO2 or UV or infrared laser?
Should we be using a fiber laser or a solid state laser?
Can it be done with the laser?
One of the biggest challenges I see when people are trying to address this question is knowing the best approach to this question. In our methodology at TLS, we refer to this early laser feasibility stage as Phase One. There are steps on how to approach this methodically, but in this article, I will first focus on the three common mistakes that people make time and time and again. In my observation, these missteps, though unintentional, can cost companies a lot of money and waste a lot of resources, which ultimately lead to major delays as well into product development and/or manufacturing -- sometimes by years if this isn't approached in the right way.
While it’s important to protect your IP, it’s important for the laser expert to have a full understanding of your application, even at a high level. Laser experts or engineers can best solve your production problem, your manufacturing problem, or your product development problem when they understand clearly how the problem fits into your product, the reason for the process, the purpose of the feature and how everything affects your product – the big picture. A good laser applications engineer should then be able to translate these high-level requirements and information you provide into a process specification, and that in turn will translate to developing the actual process that serves your needs.
At this early stage, it's important to be open minded and creative in your thinking of a solution. It is also critical for both parties — the laser applications engineer and the customer/end-user -- to take the time to educate each other. Doing this can save a lot of time from boxing yourself into an inefficient solution right from the start or coming to the wrong conclusion based on incorrect assumptions.
When we receive very specific process specifications from customers – let’s say a specific feature with “this much” tolerance or it has to be “this exact” throughput etc.-- often what we do is challenge this and ask for the big picture because we need to make sure there's an alignment between the process specification and the intention of the process.
An example would be a frequent request for chamfers to remove edge defects for the purpose of improving the mechanical strength of a CNC machined part. However, in laser processing, sometimes (not always) a chamfer can be skipped because laser cutting can have a quality that’s good enough to not require post-processing. In this example, the education between the two parties can help determine if a chamfer is necessary. There are many other examples of technical specifications, such as tapering of certain features. Typically, laser cutting, ablation, and drilling methods have a sort of taper angle to them. With some education to the customer early on, a laser application engineer can help them understand different types of limitations to the various manufacturing processes. The customer can in turn educate the laser expert on the true needs for the application. This production exchange can then help determine the degree of tapering appropriate for the application.
Another common mistake we see is trying out different lasers randomly without any deep thought or understanding about what laser was chosen, how it was used and what you learned from initial testing.
Good example here is you go to Shop A, B and C testing three different types of lasers. You may not know exactly what type of laser it is or how the lasers are set up. And you get:
bad result from Shop A
bad result from Shop B
OK result from Shop C
At that point, you have not gained a clear understanding on what works and what doesn’t, and why bad results vs OK results, or even how to get better results. All you know is that in Shop C, they got an OK result, but you don’t know how it’s going to translate.
I've developed some really powerful laser process solutions for companies over my career and the more applications I've worked on, the more I realize that process development takes time. The main thing to remember here is that just because one test, one supplier, or one laser may be a complete failure – or even sets your part on fire, it doesn’t mean there is no hope for a different laser out there to be the perfect tool for your application.
A frequent mistake that is not often talked about is the urge to draw a conclusion before enough work is done.
Customers often want simple answers asking, “is this the right laser for the job?” and an expert will reply, “it depends.” Laser process development is complex and it’s important to refrain from jumping to conclusion on whether a laser is the right laser without understanding the difference between the fundamental laser-material interaction, which is typically related to the laser type and the laser process, which is related to how that laser is being utilized. Often a right laser doesn’t seem to be doing the right job until the correct parameters are dialed in, and determining the right parameters requires an expert.
For example, you can have a super expensive $1M state-of-the-art ultrafast laser machine and your material could still catch on fire, even though you've invested in this very fancy and advanced laser tool. You may then want to determine that this is the wrong laser for the job, but it’s important to refrain from judgement, as the devil is in the details. Laser systems, especially micromachining laser systems with pulsed lasers, have a lot of adjustment capabilities. There are so many knobs, dials and parameters that you can adjust that the process space is actually extremely vast.
To help illustrate this better, I've spent 100 hours on an application where I've been dialing these knobs, trying out different configurations, and every single part came out looking absolutely horrendous. And it wasn't until hour number 101 where I started to get a result that looked like the solution was viable. Sometimes we can get contrastingly different results just by adjusting these parameters in finding the right process regime. It’s not always that long and tedious of course. Sometimes we run applications where after 30 minutes we've hit bull's eye. So it really depends on the application and the complexities of the process and the materials.
In summary, to avoid these common mistakes, remember that good process development requires expertise in the field, deliberation, and persistence to be successful: tests should be well-designed and purposeful to fit the big picture, so that findings can be meaningful as learnings, and a process to be developed methodically with persistence for the right result.
Of course, avoiding mistakes is just the start. In my next article, I will list steps that I use in working out the right type of laser. Please follow us onLinkedIn for our next blog update or subscribe to this blog. For immediate answers about laser processing and applications, us at .To see Turner Laser Systems or laser processes videos, visit our YouTube channel.
About Mark Turner
The company is the world’s best laser patterning system supplier. We are your one-stop shop for all needs. Our staff are highly-specialized and will help you find the product you need.