Billions of Pieces, Billions of Questions
Eager to capitalize on the promise of nanotechnology, businesses are finding the barriers to large-scale nanomanufacturing are anything but small.
A TEM image of a 200-nm wide hybrid bundle of zinc oxide nanowires grown on carbon nanotubes (CNTs). This structure combines the attractive energy harvesting properties of zinc oxide nanowires with the high electrical conductivity of CNTs, making it a building block for nanostructured energy materials. Image: John Anastasio Hart, Univ. of Michigan
In 2004, in a laboratory in England, two researchers placed Scotch tape on a graphite crystal—the same sort of material used to make pencils—and peeled it off. With this comically simple idea, Andre Geim and Konstantin Novoselov managed to isolate, for the first time, a layer of graphene.
The production method may have been low technology, but the task of confirming that the layer had been transferred onto a silicon substrate for study was arduous and difficult because all activity was taking place on the atomic scale. For this, these researchers were recently presented with a 2010 Nobel Prize. The pattern of development in nanotechnology follows much the same model today. Making a layer of graphene isn’t terribly difficult; but making many of them, of the same size, and controlling how the properties are utilized in a product is a considerable challenge.
In September, the National Center for Manufacturing Sciences (NCMS), an Ann Arbor, Mich.-based consortium of 175 North American companies, released the third in its series of surveys measuring the momentum of the nanotechnology and deployment in the U.S. manufacturing industry.
The data showed that U.S. manufacturers are increasingly using nano-enhanced materials to improve their products, having identified nanotechnology as being vital to competing globally. But among the report’s revelations were daunting challenges that include a lack of awareness of nanotechnology’s potential advantages, a lack access to the required design tools and training, and a lack of access to capital to fund investment in new technologies.
Most experts agree that the U.S. nanotechnology industry is currently well positioned to compete globally. The government agency National Nanotechnology Initiative (NNI), Arlington, Va., pours billions of dollars a year into nanotechnology research grants and nanotechnology hubs that serve as both educators and incubators. Top U.S. tech companies are certainly investing, too. Foreign R&D investment is also increasing, and more than half of all companies that maintain internal R&D programs have activities in these areas.
Nanotechnology is becoming a general purpose technology, a result that was anticipated by the National Science Foundation (NSF) ten years ago. After a decade, though, nanomanufacturing remains under-developed.
Roadblocks to nanomanufacturing
The promise of a revolution in the quality and function of products has created a nanotech boom. According to a technical market research report from BCC Research, (“Nanotechnology: A Realistic Market Assessment”) the global market value for nanotechnology is an estimated U.S. $15.7 billion in 2010. It’s expected to increase to nearly $27 billion for 2015, reflecting a 5-year compound annual growth rate of 11.1%.
Nanotechnology is currently in use across a variety of industries at every stage of the innovation cycle, from pure R&D through prototyping and pilot production and up to full scale manufacturing, says Vincent Caprio, executive director of the NanoBusiness Alliance LLC, a Skokie, Ill. firm that provides support to and tracks the progress of nanotechnology companies.
The most widely quoted numbers are for “products incorporating nanotechnology”, he says, and 2015 predictions vary from $1 trillion to about $3 trillion. These estimates vary widely because of the uncertainty about what nanotechnology adds to the value of products. More important than these trillion-dollar numbers, says Caprio, are estimates of global revenues for products containing nanotechnology. All show growth by seven to 10 times over the last five years and are expected to sustain rapid growth.
With this sort of market future, it’s no wonder that businesses are as excited as researchers about the possibilities. Back in the laboratory, however, researchers are dismayed at a mismatch between new products they generate and the sort of process and product quality control that are typical of industrial operations. As a result, few new nanotechnology products are industry ready, and developers are looking for a key to help them advance.
According to John Anastasio Hart, an assistant professor of engineering at the Univ. of Michigan, Ann Arbor, who teaches a course in nanomanufacturing, the real puzzle is “maintaining adequate process control while adapting processes to large-scale manufacturing. There are many examples where ‘lab’ methods that gave initial demonstrations were not suitable for scale-up, so demonstrations that also show scalability are most exciting.”
Among the early beneficiaries of high-throughput methods were colloidal nanoparticles, which are now a leading nanotechnology material used in products. Carbon nanotubes, on the other hand, illustrate Hart’s point. Only recently, nearly 20 years after their discovery, have materials experts been able to synthesize, or grow, carbon nanotubes of an adequate purity and alignment (and cost) to serve as raw material for product development. For single-walled carbon nanotubes, which are difficult to make and amount to only a few dozen kilograms per year, production stills lags far behind demand.
J. Alexander Liddle, group leader for the Nanofabrication Research Group at the Center for Nanoscale Science and Technology at the National Institute of Standards and Technology, Gaithersburg, Md., adds that finding the right route to fabrication or synthesis is always a major challenge.
“In the research arena, there are typically many ways to skin a given cat, but identifying a truly scalable process is hard work,” he says. “Not only does the fundamental physics have to be right, but there are also considerations of cost, safety, and sustainability, all of which must be satisfied to yield an economically viable process.”
Liddle points out that several decades of research in methods to grow ultra-pure, single crystal semiconductors preceded the invention of the transistor. Now, 300-mm dia, several hundred kilogram Si single crystals are standard in manufacturing. But it didn’t, he says, happen overnight. The semiconductor industry is one of the only sectors to have successfully adapted to the nanoscale features. They were able to do this largely by concentrating a limited selection of materials, primarily silicon, and developing process tools in a gradual way. For materials such as metals or polymers, this sort of infrastructure doesn’t exist.
“For most nanotechnology companies, including nanostructured materials manufacturers, the challenges are not unlike those faced by champions of other emerging technologies,” Caprio adds. Principal among these are developing a strategic vision for translating technical achievements into desirable products that address market needs; overcoming any barriers that inhibit entry to those markets; and, perhaps most importantly today, he says, gaining access to sufficient capital to kick off that strategy and carry its initial stages.
Scale-up is not the only problem facing nanotechnology companies. New manufacturing processes, as well as new characterization and monitoring tools, may need to be developed and adapted for compatibility with existing manufacturing lines. Workers need to be trained. In some cases, expensive capital equipment will be required. Customers must be convinced that the enhanced performance promised by nanomaterials is worth the risk of modifying their tried and true processes.
In regulated industries, companies and regulators have to figure out which existing standards and protocols are applicable to nanomaterials and develop new ones where needed. The perception of regulatory uncertainty raises a red flag for investors in nanotechnology.
Measurement for nanomanufacturing
A major question mark facing companies that intend to produce a product that uses nanotechnology is that process control measurements hardly yet exist for many promising areas of industrial application. The shortcomings in particular are highlighted by the lack of knowledge about the health effects of certain nanomaterials used in biotech or food-related industries. But even areas where health is not a problem, nanoscale measurements are typically difficult outside the rigidly controlled environment of the laboratory.
For researchers, there are a wide variety of imaging and measurement tools available. Atomic force microscopy and electron microscopy are particularly well-known, but confocal and Raman optical instruments also contribute. But these instruments are designed primarily for researchers. None are process-oriented.
Craig Schwandt, a scientist with the electron-optics group at McCrone Associates, a Westmont, Ill.-based microscopy consultancy firm, says that techniques for observing materials at the nanoscale are improving quickly to keep up with demand. Electron microprobes can now be used for quantitative analysis rather than just imaging, and energy-dispersive x-rays can do micro- and nano-analysis relatively quickly.
This pattern of carbon nanotube “forests” on a silicon substrate, imaged by an SEM, was made by reacting a hydrocarbon gas mixture on a pattern of metal catalyst nanoparticles. The arrangement of the particles determines where the CNTs grow. Such processes will need to be scaled up considerably to meet the needs of product developers. Image: John Anastasio Hart, Univ. of Michigan
“At the nanoscale, the physical properties of materials are different. They tend to want to adhere in clumps. Isolation is difficult, so coming up with tricks for particles to disperse more is still a challenge,” says Schwandt. Simply watching what particles are doing often inadvertently changes the way they’ll behave in, say, a human body. Sample preparation methods allow researchers to move from around 100 nm to the single nanometer scale, but these typically require embedding the sample or capturing in different types of media. The price of such efforts is the need to analyze the media itself—whether it’s a crystal or liquid, and account for its presence when obtaining data.
Yet, for all of this difficulty, analytical instruments and metrology techniques play a significant part in developing nanotechnology, and Liddle certainly can’t do without them for his work at NIST.
Measurements play two critical roles. First, they give us a way to gain a deep understanding of the way materials behave at the nanoscale, as well as the way the processes used to make them work, sometimes down to the atomic scale,” says Liddle.
“Once we have that kind of information, then we’re in a position to design the right nanostructure for a particular application, as well as a robust process for its manufacture.” Second, measurements are required during manufacturing to control the process to ensure that manufacturers are making exactly what they planned to make. This is process control, and while these measurements are nothing new, the unique challenge in nanomanufacturing is the combination of scale and speed. For any given nanomanufacturing process, says Liddle, the default starting estimate is billions of components, all of which (within some set of carefully defined tolerances) must have the same size, shape, and composition.
To close the process loop, Liddle continues, manufacturers must have measurements that can determine those characteristics, at least in terms of an average and standard deviation, at speed. Imagine, for example, a roll-to-roll process like letterpress printing that runs at 10 m2/s), in line and affordably, generating nanopatterned circuitry at high volume.
Research for process like these are either recently underway or just beginning.
Although there will be some commonalities, new measurement methods will likely have to be developed to suit each set of new materials and devices, as well as the particular processes used in their manufacture,” says Liddle.
Part of the problem, according to Schwandt, is that some developers are unaware of the availability, or appropriate application of the tools at hand. He often sees people go to transmission electron microscopes to perform quality control tasks. They don’t necessarily need to jump through the hoops of sample preparation to the get the data they need. A field-emission scanning electron microscope is typically enough to sample particles just a few nanometers in diameter.
“Recent advances in field emission electron microscopy is something nanotechnologists might not be aware of. This technology has only been available for the last three years, so I don’t think that many fabricators of nanoscale materials understand that these instruments are available for their use, whether they go buy them or buy the service,” says Schwandt, who is accustomed to helping clients perform these sorts of analyses.
Another segment of the microscopy market, scanning probes and atomic force microscopes, are enjoying a renaissance because their ability to provide quantitative data on the nanoscale. Unfortunately for those who want a process-oriented tool, these are typically more comfortably employed in the research lab than the factory floor.
“My feeling is that there is very little [manufacturing] as far as the SPM (scanning probe microscopy) business is concerned, says Sergei V. Kalinin, a researcher at Oak Ridge National Laboratory, Oak Ridge, Tenn., who is a frequent user of nanoscale imaging tools as co-theme leader for functional imaging on the nanoscale at The Center for Nanophase Materials Sciences and the Oak Ridge Lab’s Materials Sciences and Technology Division. “Most companies,” he says, “primarily sell to R&D in academia, labs or industry, not manufacturing.”
He believes that analytical instruments are a key and enabling component for the successful joining of nanotechnology and manufacturing, but widespread of adoption and use of these tools has not caught up with the demand of commercialization.
“The fact that people started to talk about nanotechnology at all is largely driven by the fact that SPM measurements have become available, and with them the relatively easy to use, low-cost benchtop instrumentation for probing and manipulating materials locally. All of these characteristics are important,” says Kalinin. He anticipates the SPM will continue to a great bearing on how fast nanotechnology develops, more so than electron microscopy in large part because of the cost.
Top-down, or bottom-up?
Patrick Dessert, an engineering professor at Oakland Univ., presented a paper earlier this year at a Society of Manufacturing Engineers conference that exemplifies the fundamental problems faced by nanomanufacturers: does a bottom-up, “self-assembled” nanomanufacturing approach make more sense than using more traditional additive manufacturing techniques? His conclusion was that while bottom-up approaches are useful in the creation of small parts, additive manufacturing techniques currently assigned to rapid prototyping are the key to unlocking the potential of nanotechnology and creating the forecasted “revolution of manufacturing” at a macro scale. This is the sort of finding that engineers will need to guide the development of efficient nanofabrication processes.
In September, the National Science Foundation held a conference marking the 10-year anniversary of NNI. Called NANO2, the conference was a major debriefing on all changes and aspects of nanotechnology over the past decade.
Mark Tuominen, director of the National Nanomanufacturing Network and a professor at the Univ. of Massachusetts, Amherst, spoke at conference, discussing developments in the synthesis, processing, and manufacturing nanoscale components.
“As top-down techniques are really reaching fundamental limits we really have to look at bottom-up techniques such as self-assembly and chemical methods to manufacture the things we desire,” said Tuominen. “We really have to develop low-cost process to advance nanotechnology, and in some cases these low-processes have to be high-throughput.”
Tuominen cited major advances in nanotechnology, from colloidal nanoparticles to nano-patterning of surface, but stressed that before transitions to industry can be made we need to understand the first principles of materials synthesis and develop better metrology solutions.
The promise that is now being realized, says Liddle, is that the range and combination of materials properties available is no longer tied to the intrinsic properties of a material, but is instead expanded by our developing abilities to precisely control the size, shape, local composition, etc. of nanostructures and then to combine them in particular ways. Others see this transition as well.
“Overall, we are experiencing a transition from “bulk” nanotechnology, e.g., materials using bulk nanopowders, CNTs, etc. to “organized” nanotechnology, e.g., materials using organized nanostructures,” says Hart. Use of organized nanostructures, he says, requires a greater understanding of synthesis and assembly of nanostructures, and often the structures have to be directly organized by the manufacturing process that is integrated with the device. Basic techniques for synthesis and assembly of nanostructures are becoming sufficiently known to enable research on nanoscale systems, for example materials with responsive properties that change depending on their environment (such as load and temperature).
One of the priorities of the National Nanotechnology Initiative is helping businesses successfully identify products that could benefit from nanotechnology. The problem, unfortunately, is that many of these materials promise so much it’s difficult to determine what property should be leveraged first.
“One major challenge is to find an important application of new nanostructured materials that can give a significant performance advantage and remain cost-competitive with existing solutions as production volume scales up,” says Hart.
According to the U.S.-based Project on Emerging Nanotechnologies (PEN), the database of consumer nano-products has grown 300% in the last year. But the total is still miniscule compared to the investments being made, and only one industry, semiconductor, can be presented as an example where nanotechnology is in full production. Current generations of microprocessors and memory chips would not be possible without nanostructured substrates and dielectric materials along with new nanoscale transistor designs, which the industry integrated fully into their front-end manufacturing processes throughout the last decade.
The semiconductor industry is also a major supporter of continuing research into novel nano-enabled materials, devices, and system architectures, which are essential to maintaining Moore’s Law.
“Nanomaterials is another interesting area,” says Caprio. Some materials are used in high-volume applications, such as metal oxides with uses from sunscreens to industrial coatings, and ultra-thin polymer films protecting eyeglasses. Other materials, like carbon nanotubes, are being used in niche applications and are moving into the latter stage of development for much broader purposes, such as like transparent conductive coatings for electronic displays, or as performance-enhancers in concrete roadways.
According to Hart, there are many products that rely on “bulk” nanotechnology including composite tennis racquets and golf club shafts with nanoparticles and/or CNTs, cosmetics with nanopowders, window glass, and contact lenses with nanolayer coatings.
“These are being manufactured now, and in some cases the performance gain is small, but the product can sound much more attractive because it is ‘nano’,” he says. High performance batteries (for example, from A123 systems) have been made possible by nanoparticulate electrode materials. Hart also thinks major areas for new products are energy (batteries, solar cells) and medicine (anti-microbial coatings, and lab-on-a-chip sensors using nanomaterials).
“CNT manufacturers have been quite successful in scaling up production and in creating differentiated products,” says Caprio. “Then there are highly specialized materials that take longer to develop and may have to go through extensive regulatory reviews, like many of the materials in development for drug delivery and medical diagnostics.”
The notion that one breakout product will help transition nanotechnology to the mainstream, however, is increasingly appearing to be a fallacy. Nanotechnology appears to be far too broadly applicable to be dominate any one product category.
“Actually, a question here is whether there is a killer app for [nanotechnology] at all,” says Kalinin. “Electronics seem to be out until they can be grown of necessary chirality and necessary position. Structural materials require large scale production and low cost, and are unlikely to go forward until costs are significantly lowered and environmental and health safety are ascertained.”
This video capture of Au/Ni (20/80 at. % nominal composition of thin film) has been heated at 520°C and exposed to 0.187 Pa overpressure of flowing C2H2. The scale bar is actually 5 nm. Like macro materials processes, nanomanufacturing techniques will require widely varied thermal and pressure changes that will complicate QA efforts. Credit: Renu Sharma, National Institute of Standards and Technology
For James Hussey, chief executive officer at NanoInk Inc., Skokie, Ill., which makes dip-pen nanolithography (DPN) equipment that can write nanoscale features on a broad range of substrates and chemistries, the primary challenge for any company involved in nanotechnology is scale up.
In the case of DPN, he says, “we have almost no problems with the basic technology. You need time, money, and people to perfect the process. The key is scaling up, moving the technology to meet the requirements and demands of the market. You have to deliver the product to market, and do it at a cost and throughput to capture market share.”
At Xerox, Santokh Badesha has spent a career researching activities on a small scale. From inventing an early sol-gel process to helping transition his company to a world that is increasingly moving away from traditional paper printing, Badesha has seen the development of nanotechnology.
“In a typical Xerox marking technology,” such as what the company is currently marketing, says Badesha, “there is a powder, which is a complex set of materials. In addition to the powder, there is an additive fixer, colorants, and more.”
In the case of some of Xerox’s newest printing technologies, the powder is carefully structured on the nanoscale to interact electrostatically with various surfaces it comes in contact with during the print cycle. Nanoscale design parameters were required for the entire process, which begins when the familiar green button is pressed. Users see the flash, create a virtual image, and mechanical motions begin and end with fusing and fixing of the polymer nanoparticles to the paper surface.
“Nanotechnology plays very key role in the not only the design of the materials that are place on the paper, but also the surfaces,” says Badesha, where the particles make contact. There are four different types of surfaces, and each has multiple functionalities to help the nanoparticles assume the correct shape and color on the page.
But this type of technology took many years for Xerox to develop—and this is a large company. The ability for a nanotechnology company to scale up is highly dependent on the particular material that is being investigated, according to Liddle. For example, the feature sizes in integrated circuits are so small now, that they qualify as nanoscale. Yet there is clearly a large and efficient manufacturing infrastructure in existence to produce them. Similarly, he says, there are many material systems, such as block copolymers, that have characteristic length scales in the few-nanometer range that are synthesized in industrial quantities, and have been for many years. Both of these examples don’t use machine tools in the conventional sense. They instead rely on optical lithography in the case of integrated circuits and synthetic chemistry in the case of block copolymers.
Long lead times
“We all take the introduction of faster, cheaper, more functional devices for granted now, but the transistor was invented in 1947 and the integrated circuit in 1958, so it’s taken another 50 years to reach this point, even though, very crudely speaking, everyone has been working on one device (the transistor) made out of one material (silicon), all this time,” says Liddle.
When new materials are introduced to the IC manufacturing process, the lead time is 10 to 15 years. This is not a promising role model for nanotechnology, which sees new materials inventions regularly. But the comparison to the semiconductor industry can only be a partial one. Because nanotechnology is broad, employing a nearly limitless array of possible materials, the goals for synthesis, assembly and processing of future nanotechnology products must be different.
Tuominen, speaking for NSF, said the goals going forward would include both top-down and bottom-up solutions, including self-assembly, bioinspired synthesis, roll-to-roll production, and heterogeneous multicomponent 3-D nanosystems. Science-based process-structure-property relationships for nanoscale synthesis and processing will need to be adopted, and whatever processes are scalable will be pursued.
Through these strategies, agencies like NSF hope to give a nanotechnology the push that will spur growth elsewhere. Analytical instrumentation vendors will likely both pave their own direction and follow the lead of industry demand in the ceaseless effort to improve measurement techniques, which will be mandatory for nanoscale process structure.
“To borrow an adage from the optics industry, if it can be measured, it can be made,” says Liddle.
Meanwhile, the broad growth of the nanotechnology market, whether it’s in the billions or the trillions, should reflect how effectively these tools—instruments and techniques—are pulled together to create useful products.
“The market pull or investment will determine how quickly we scale up nanomaterials manufacturing, and how rapidly the related quality control tools/protocols are developed in concert,” says Hart.
Ultimately, manufacturers will require closed-loop control of process before the revolution on nanotechnology can truly take place. But to get they must first hurdle the barriers of knowledge and scalability. It will be a more an endurance race than a sprint.
Published in R & D magazine: Vol. 52, No. 6, October, 2010