Biodegradable electronics: Help for the waste problem

The fast pace of technology results in tremendous amounts of waste. New research into biodegradable electronics could reduce waste, both at manufacturing processes and in end-of-life trash.

We have a waste problem. Not only do consumers throw away electronics every time a new model comes along, but the processes used to manufacture electronics produce vast amounts of hazardous waste. We need to do something about that. Enter Ravinder Dahiya, professor or Electronics and Nanoengineering at the University of Glasgow James Watt School of Engineering.

Dahiya and his team are investigating ways to make biodegradable electronics that can naturally return to the Earth, this reducing waste, saving money, and making electronics more environmentally friendly. EE World spoke with Prof. Dahiya about his research and how to reduce electronic waste.

The researchers have been developing flexible electronics that go beyond making flexible phone screens. Some of the processes used to manufacture flexible electronics can be extended to make flexible transistors. The materials used to make flexible transistors could also result in electronics that biodegrade. While such products are years away, it’s a start.

EE World: Is there a relationship between flexible electronics and flex PCBs?
Dahiya: Flexible, bendable electronics are different from flex PCBs where the substrate is flexible but not the components on it. With flexible electronics, the transistors themselves are flexible. We try to use the conventional microfabrication tools. That is, we use metal deposition, photolithography, and etching etc. in a conventional way.

The problem with these processes is that they lead to a huge amount of chemical and material waste, both in the fabrication process and at the end of a product’s lifetime. We have a problem with growing electronics waste and its impact on health and environment. It’s attracting more attention because of the growth of personal electronics in our daily life. We don’t have a policy other than managing the electronics after they’re fabrication. By then, damage has already been done. Chemicals, electricity, and water have already been used.

We should be proactive and look at the problem both from a manufacturing and disposal perspective. We should not respond reactively only. Besides looking at how to extract materials from waste electronics, we should also be looking at material selection and manufacturing processes from the start. Can we use sustainable materials? If we put them all together, can we retain the performance of today’s electronics? If so, then we can bring the much-needed change in electronics fabrication and make it sustainable. With these questions in mind, we wrote this project and secured funding for research.

EE World: How do you control the degradation process so that devices don’t degrade too quickly? This sounds like a reliability engineer’s nightmare (and perhaps a marketer’s dream of controlled failure so consumers will have to replace products)?
Dahiya: The answer lies in proper encapsulation. We could use materials that degrade more slowly than the active devices themselves. We looked into materials for substrates and semiconductors. Some materials can degrade under specific temperature and humidity conditions. It may depend on the application. It’s important to realize that you don’t want to make electronics that degrade after one use, or a few uses.

EE World: What about a method for initiating a degradation, say some condition that wouldn’t likely occur in normal use that would case premature degradation?
Dahiya: All materials degrade over time. We looked into materials such as Poly Lactic-co-Glycolic Acid (PLGA), Poly (l-lactic acid) (PLLA), degradable polyethylene terephthalate (PET), and so on. We are looking at conditions such as temperature, pH, and light to initiate a controlled degradation. Take PLGA, for example. It degrades about 1% to 2% per day and could degrade in 50 to 100 days, depending on material thickness. The degradation also affects a material’s flexibility.

Metal foils could be used instead of the polymers when the processing temperature is high and slow degradation is needed. In these cases, a thin dielectric or insulation layer is deposited on the foils before fabrication of electronics. Some metals are also degradable, such as Magnesium. We want to avoid using gold or silver because they don’t help with sustainability. We want to look at metals such as magnesium, which degrade and are readily available.

EE World: What factors affect degradability?
Dahiya: Degradability factors include into what you immerse the electronics. For example, an aqueous material such as water. The degradable materials could also be placed into soil. For aqueous materials, temperature, pH, and the properties of aqueous medium could affect degradation. For other materials such as soil, humidity, temperature, pressure, insects, and microbes can also become degradation factors.

EE World: What about the up-front processes? What can we do to reduce waste? What do we need to study?
Dahiya: We need to develop manufacturing processes that don’t depend or have least dependence on the traditional micro-nano fabrication. In the Bendable Electronics and Sensing Technologies (BEST) group at University of Glasgow, we have developed our own printing machine to print silicon-based nanostructures. By “nanostructures,” refer to nanowires and nanoribbon. We print on the flexible substrates, such as paper or degradable plastic at room temperature. That’s important because many of these substrate materials can’t withstand high temperatures, which we need in conventional semiconductor fabrication processes.

We use a printing process that lets us perform “directional printing.” That is, longer sides of nanostructures align along the desired direction while they are printed. Once we have the electronic layer, we print the dielectric and metal lines with our high-resolution printers. That’s how we develop the transistors.

Key elements for the transistor include:

• Substrate
• Semiconductor material
• Dielectric material
• Metal

All of these materials are printed. We’re using the materials as needed instead of using a removal process.

EE World: When you flex materials such as metals, they can break. It sounds like a reliability nightmare. How do we circumvent that problem even now?
Dahiya: That’s more of a problem for flexible PCBs. You are trying to integrate materials of dissimilar properties. Think of it like trying to glue a piece of wood to a brick. Sometimes, you see a crack. With flexible electronics, the lines are so thin that they can withstand stretching at the micro level. When we print the metal line, they can conform to the micro-level radius that can occur.

Locally, you don’t have same level of bendability as at the macro level. We also test the printed devices by twisting them thousands of times to evaluate their reliability. That’s macro-level twisting. The individual components see only micro-level twists, and that’s not much. We’ve published papers on this topic. [Ref. 1, 2, 3]

EE World: What about device performance?
Dahiya: We’ve seen little variation in device performance, but there’s always variation, even with devices manufactured with traditional processes. The piezoresistive effect comes into play when a material experiences bendability. During fabrication, some of the materials experience stress due to temperature and pressure conditions.

EE World: How could this research result in a reduction or elimination of waste?
Dahiya: Micro/nano fabrication processes are subtractive in nature. Any material removed is waste. A semiconductor will go through these steps several times. You deposit material and remove it from the unwanted areas. A wafer needs to be cleaned several times. Cleaning means washing with chemicals or with water. You must do that for each of the hundreds of steps in the manufacturing process. That leads to much waste. We don’t realize it.

Printing is an additive process. You only add as much material as you need. Some of our additive processes are dry. We don’t use any water. For printing our nanoribbons, all we use is the force between the substrate and the material to be transferred. You press the materials at a certain force and then a roller moves. The material bonds to the flexible substrate. Metal lines also act as a further bond (Figure 1).

EE World: Do you have to apply the roller at a controlled pressure and angle of attack? How do you do that?
Dahiya: We have two versions of printing machines. One we call contact printing and other we call the direct-roll transfer printing. The mechanisms are slightly different. In the case of contact printing, we grow nanowires atom by atom using a vapor-liquid-solid (VLS) mechanism on a donor substrate. The receiver substrate below sits on a load cell to measure the pressure. A self-aligning mechanism aligns the donor and receiver substrates. The receiver substrate then moves, and the nanowires are printed horizontally along the substrates. That makes the electronic layer. When you use this additive process, you add a few nanometers of material at a time.

With direct-roll transfer printing we realize nanoribbons on the wafer. It can be a silicon-on-insulator (SOI) wafer, or it can be a bulk wafer. Bulk wafers use a longer chemical etching process so there is more waste there. In this case, we try to use some of the nanofabrication processes so common today. That makes the process more attractive to industry. Gradually, you can switch to the SOI process.

Our direct-roll transfer process also overcomes the issues with conventional transfer processes such as low yield and poor registration. Conventional transfer processes use soft elastomer such as polydimethylsiloxane (PDMS) as transfer stamp, which can lead to registration issues. Even a 1 µm registration error can cause a transistor not to function as designed. Our direct-roll transfer process does not require PDMS stamps and hence we are able to achieve excellent nanoscale registration.

We’re using high-mobility materials, which is important for high-performance, fast switching transistor devices.

EE World: You mentioned switching transistors. Could this transfer process also work for making analog circuits?
Dahiya: It works for both digital and analog. It just depends on the design. Digital electronics are much easier to manufacture and operate, but the process is the same for digital or analog transistors.

EE World: What kind of geometries can you achieve?
Dahiya: Nanoribbons can be less than 100 nm thick, 50 µm to 100 µm long. and 4 µm to 10 µm wide. We can print them on to a large area, currently restricted by the size of the roll only. We’ve developed devices on an area that’s 9 cm@sup2;. Registration accuracy is 99.9%. That’s a mismatch of just a few nanometers. We can take that to a larger area because we didn’t use PDMS. We can make both N-type and P-type transistors and thus we can make CMOS circuits. With P-type transistors, we have an electron mobility of about 225. Conventional silicon will give you about 250. With N-type, we can get about 700 as where conventional processes can produce about 800 to 850 cm@sup2;/Vs. The performance of our devices is slightly lower the conventional silicon-based devices. We are also printing metal lines, not depositing them.

Most applications today are for communications that require switching at roughly 500 MHz to 1 GHz, say for IoT. These applications also require flexibility and hence our approach is ideal for this range. Currently this process might not be suitable for, say 6G because you need switching speeds in terahertz range. Gigahertz frequencies are easily possible. However, with advances in printing technology there is potential for further improvements in performance.

What materials have you tried, both for substrates and semiconductors?
Dahiya: For substrates, we’ve tried textiles, paper, PET, polymide, and metal foils. For the semiconductor material, we’ve tried metal oxides such as zinc oxide for nanowires, N-type and P-type silicon, and gallium arsenide (GaAs) because it’s attractive for power electronics. We have also explored metallic nanowires such as copper and silver.

References
1. Ayoub Zumeit, Abhishek Singh Dahiya, Adamos Christou, Dhayalan Shakthivel & Ravinder Dahiya ,”Direct roll transfer printed silicon nanoribbon arrays based high-performance flexible electronics,” https://www.nature.com/articles/s41528-021-00116-w
2. Ayoub Zumeit, Abhishek Singh Dahiya, Adamos Christou and Ravinder Dahiya, “High-performance p-channel transistors on flexible substrate using direct roll transfer stamping,” https://iopscience.iop.org/article/10.35848/1347-4065/ac40ab
3. Anastasios Vilouras, Adamos Christou, Libu Manjakkal, and Ravinder Dahiya, “Ultrathin Ion-Sensitive Field-Effect Transistor Chips with Bending-Induced Performance Enhancement,” https://pubs.acs.org/doi/abs/10.1021/acsaelm.0c00489.