A key part of my experimental workflow involves micro/nanofabrication using the Microwriter DMO, a high-precision maskless lithography system. This system allows direct writing of custom patterns onto substrates using a focused UV laser, eliminating the need for physical photomasks — which makes it ideal for prototyping and fine-tuning device structures.

At the core of lithography lies the principle of selective light exposure on a photoresist-coated substrate, followed by development to create patterns for subsequent processing like etching or metal deposition. With the DMO, I typically use a 0.6 µm lens, which gives me the resolution needed for extremely small features.

My Lithography Work:

• Patterning sub-micron to 2 µm wide current channels for Hall bar devices.

• Performing lithography directly on exfoliated 2D materials (like flakes of graphene or TMDs), aligned carefully under a microscope.

• Implementing grayscale lithography to fabricate structures with variable depth and profile — an advanced technique especially useful on uneven flake surfaces.

• Working on ultra-small substrates, as tiny as 1 mm × 1 mm and 2 mm × 2 mm, where handling and alignment become extremely challenging.

Despite the difficulty of working with such small-scale substrates and fragile 2D materials, I successfully developed a robust and repeatable recipe — a result of countless iterations, failed patterns, and deep troubleshooting.

Photoresist & Process Details:

• I use AZ5214E, a versatile image reversal photoresist, compatible with the fine-resolution patterns I require.

• The development process uses a compatible developer (like AZ 726 MIF) carefully timed to avoid over- or under-development.

Challenges & Breakthroughs:

Working on such a micro-scale often led to:

• Incomplete development,

• Pattern distortion due to substrate curvature,

• Lift-off failure due to adhesion issues.

But over time, through extensive process optimization — including adjusting baking times, exposure doses, and development parameters — I overcame these hurdles. I now have a reliable process for high-resolution lithography on both conventional and exotic substrates.

I'd love to show a few images of failed vs. successful patterns — they tell a story of trial, error, and eventual success.

When Lithography Fails — Root Causes and Critical Factors

Lithography is a cornerstone process in micro/nanofabrication, and when it fails, it often does so in subtle but devastating ways. Working extensively with the Microwriter DMO, I've encountered and resolved many challenges while patterning sub-micron features, Hall bar structures, and 2D material-based devices.

Here are the key factors that can cause patterning failures or defects, based on hands-on troubleshooting:

1. Substrate Issues

• Substrate Size & Handling
  When working with very small substrates (e.g., 1 mm × 1 mm, 2 mm × 2 mm), even minor misalignments or motion during spin coating or exposure can lead to total pattern failure.

• Surface Contamination
  Dust, moisture, or even leftover exfoliation tape residue can prevent photoresist adhesion or cause uneven development.

2. Photoresist Coating Problems

• Non-uniform Spin Coating
  On small or non-standard substrates, achieving a uniform photoresist layer is difficult. This can lead to incomplete exposure or irregular pattern formation.

• Incorrect Bake Parameters
  Overbaking or underbaking (softbake/hardbake) can alter resist sensitivity or cause cracks, especially when using AZ5214E for image reversal processes.

3. Exposure Parameters

• Wrong Focus Height or Dose
  In the Microwriter DMO, incorrect z-height calibration, lens choice, or exposure dose can result in:

  • Overexposed (merged) features,

  • Underexposed (faint or incomplete) lines.

• Feature Size vs Lens Resolution Mismatch
  Using a larger lens for fine patterns can blur or distort small features (I typically use the 0.6 µm lens for sub-micron Hall bars and fine current channels).

4. Development Failures

• Incorrect Developer Concentration or Time
  Using AZ 726 MIF with AZ5214E, if development time is off by even 10–20 seconds, it can lead to:

  • Overdevelopment: lifting fine features or damaging edges.

  • Underdevelopment: pattern not fully cleared.

• Aggressive Agitation or Rinsing
  Especially on exfoliated flakes or grayscale patterns, excessive agitation during development can cause pattern delamination or loss.

5. Grayscale & Flake Patterning Specific Challenges

• Non-Uniform Substrate Topography
  Patterning on 2D flakes involves aligning on non-flat, sometimes transparent regions. This makes both focus and exposure intensity tricky to control.

• Resist Thickness vs Depth Profile
  For grayscale lithography, if the resist thickness or exposure gradient isn't tuned properly, the resulting 3D structures are unpredictable or fail to form.

6. Equipment or Environmental Issues

• Stage Drift or Vibration
  During long exposures or multi-layer alignment, any mechanical instability causes pattern shift or blur.

• Optical Path Dust or Misalignment
  A dirty lens or misaligned beam in the Microwriter DMO can cause uneven exposure across the pattern.

Personal Insight

After working with dozens of 2D flakes and nanoscale devices, I've developed a reliable process recipe — but it came after many rounds of optimization. I've also documented and analyzed failed patterns to identify causes, including:

• Lift-off artifacts,

• Residual resist in narrow gaps,

• Feature collapse during drying.

Sharing those failure cases and learning from them has been essential to building robust, reproducible fabrication processes.

Wire bonding is a critical technique in micro/nanoelectronics — and especially so in spintronic devices, where device dimensions are small and signal fidelity is highly sensitive to contact quality. In my research, wire bonding plays an essential role in establishing reliable connections between fabricated devices and external measurement systems, such as probe stations and cryogenic setups.

What is Wire Bonding?

Wire bonding is a method of creating electrical interconnects between a microchip or substrate and a larger circuit, using thin metal wires. This is typically achieved using a combination of pressure, heat, and ultrasonic energy, depending on the bonding method.

The choice of bonding technique and wire type depends on device geometry, material stack, operating conditions (such as cryogenic temperatures), and signal requirements.

Bonding Techniques & Tools I’ve Worked With:

I have hands-on experience working with:

• West-Bond (manual)

• TPT Wire Bonder

• F&S Bondtec (manual and semi-automated systems)

And I've used the following wire bonding methods:

• Wedge Bonding: Common for fine-pitch layouts and aluminum wires.

• Ball Bonding: Ideal for gold wires; quicker and more automated.

• Ribbon Bonding: Used in high-frequency applications due to its broader contact area.

Wire Types I’ve Used:

• Aluminum wire – typically for wedge bonding, suitable for low-temperature and spintronic setups.

• Gold wire – used in ball bonding for high-quality, low-resistance connections.

• Gold ribbon – occasionally used for RF/microwave applications.

Ensuring Bond Quality in Spintronic Devices

Since I work with micro and nanoscale spintronic devices, bond quality is not just important — it's critical. A weak or unstable bond can drastically affect:

• Contact resistance, leading to errors in resistance and Hall measurements.

• Signal-to-noise ratio, especially in cryogenic setups.

• Repeatability and reliability of low-temperature measurements.

To ensure bond integrity, I routinely perform:

• Visual inspection under a microscope – to check bond placement, loop height, and adhesion.

• Resistance measurement across bonded leads – to verify low-resistance connections.

• Occasional pull testing – to assess mechanical stability when needed.

Repair & Troubleshooting Experience

I also have practical experience in maintaining and troubleshooting bonding equipment. Notably, I successfully repaired a TPT wire bonder, diagnosing and fixing issues in the force cylinder and motherboard. This helped restore full functionality and minimized experimental downtime. 

When Wire Bonding Fails — Critical Factors

Even with a good tool and process, achieving a perfect bond can be challenging. Based on my hands-on experience, here are the key factors that can cause bonding failure:

1. Surface Contamination

   • Oxidized or dirty bond pads (from dust, resist residue, or environmental exposure) prevent proper wire adhesion.

   • Even minor contamination can drastically increase contact resistance.

2. Improper Tool Settings

   • Incorrect force, ultrasonic power, or bonding time can lead to weak or lifted bonds.

   • Different substrates and wire types require specific tuning.

3. Poor Alignment or Pad Geometry

   • Small or uneven pads (common in nano-devices or exfoliated flakes) make accurate bonding more difficult.

   • Misalignment can result in the wire not bonding at all or shorting adjacent features.

4. Inadequate Wire Looping or Tail Length

   • Loops that are too short or too long can break during cooling (especially in cryogenic setups).

   • Poor looping can also introduce mechanical strain on small substrates.

5. Tool Condition and Wear

   • A worn or misaligned capillary can result in consistent bonding issues.

   • I have diagnosed such issues in our lab and even repaired a TPT bonder, fixing the force cylinder and motherboard to restore optimal bonding performance.

Conclusion

Wire bonding may seem like a standard tool, but in cutting-edge spintronics — where device features are tiny, substrates are fragile, and measurements are highly sensitive — mastering it is critical. My experience spans not just routine bonding, but also process optimization, troubleshooting, and equipment repair, ensuring that my devices are measurement-ready with reliable electrical contacts.