Northwestern University's latest innovation in materials science and neuroengineering marks a pivotal moment in bridging synthetic electronics with biological neural networks. Engineers at the McCormick School of Engineering and Applied Science have developed flexible, printed artificial neurons capable of generating electrical signals that mimic those of living brain cells. These devices not only imitate neural activity but actively communicate with real brain tissue, eliciting responses from living neurons in a way that closely parallels natural synaptic interactions. This achievement, detailed in a recent study published in Nature Nanotechnology, opens new avenues for brain-machine interfaces, neuroprosthetics, and energy-efficient computing systems inspired by the human brain.

The research, led by Mark C. Hersam, Walter P. Murphy Professor of Materials Science and Engineering, showcases how additive manufacturing techniques can produce heterogeneous, dynamic structures reminiscent of the brain's complex architecture. Traditional silicon-based electronics rely on billions of identical transistors, resulting in rigid, power-hungry systems. In contrast, the brain's neurons form intricate, three-dimensional networks that process information with extraordinary efficiency—operating on just about 20 watts of power compared to the gigawatts consumed by modern AI data centers.

🧠 The Science Behind Printed Artificial Neurons

At the heart of this breakthrough are electronic inks composed of nanoscale flakes of molybdenum disulfide (MoS₂), serving as the semiconductor, combined with graphene as the conductor. These inks are deposited onto flexible polymer substrates using aerosol jet printing, a precise, low-cost method that allows for scalable production without the waste associated with subtractive manufacturing like etching silicon wafers.

A critical innovation lies in the partial decomposition of the polymer within the inks during printing. This creates an inhomogeneous structure where, upon applying current, further localized decomposition forms a conductive filament. This filament constricts the current flow into a narrow region, producing abrupt electrical responses that replicate neuron firing patterns: single action potentials, continuous repetitive firing, and even bursting activity. These multi-order complexities enable the artificial neurons to convey rich, information-dense signals, much like their biological counterparts.

Unlike previous artificial neurons made from organic materials (which spiked too slowly) or metal oxides (too quickly), these devices operate within the biologically relevant temporal range—typically milliseconds—allowing seamless synchronization with living neural circuits. The flexibility of the substrate ensures biocompatibility, a longstanding challenge in neural implants where rigid materials provoke inflammatory responses and degrade over time.

Experimental Validation: Interfacing with Living Tissue

To demonstrate functionality, the team collaborated with neurobiologist Indira M. Raman, Bill and Gayle Cook Professor of Neurobiology at Northwestern's Weinberg College of Arts and Sciences. Raman's group applied signals from the printed neurons to ex vivo slices of mouse cerebellum, a region rich in diverse neural activity patterns.

The results were striking: the artificial spikes matched the timing, duration, and shape of biological action potentials, reliably triggering calcium transients and downstream activity in living Purkinje cells and granule cells. This biohybrid interface represents the first demonstration of printed devices eliciting circuit-level responses in neural tissue, paving the way for closed-loop systems where synthetic neurons both read from and write to the brain.

Such precision addresses key limitations in current brain-computer interfaces (BCIs), like those from Neuralink or Blackrock Neurotech, which often suffer from signal degradation due to gliosis—the brain's scarring response to foreign implants. The soft, conformable nature of these printed neurons minimizes mechanical mismatch, potentially enabling long-term implantation.

Researchers Driving the Innovation

Mark C. Hersam, a prolific researcher with over 500 publications and leadership of Northwestern's Materials Research Science and Engineering Center (MRSEC), has long pioneered hybrid nanomaterials for electronics. His group previously developed memtransistors in 2018—devices combining memory and transistor functions to emulate synaptic behavior—laying groundwork for this neuron work. Co-lead Vinod K. Sangwan, research associate professor, specializes in nanoelectronics and neuromorphic systems, contributing expertise in 2D materials like MoS₂ and graphene.

Indira M. Raman brings deep knowledge of cerebellar neurophysiology, having pioneered studies on ion channel biophysics and synaptic integration in brainstem and cerebellum slices. Funded by the National Science Foundation, this interdisciplinary effort exemplifies Northwestern's strength in converging engineering, materials science, and neuroscience.

"The brain is five orders of magnitude more energy efficient than digital computers," Hersam noted, highlighting the motivation amid AI's escalating power demands—training models like GPT-4 consumes energy equivalent to thousands of households annually.

Photo by Peter Burdon on Unsplash

Implications for Brain-Machine Interfaces

Brain-computer interfaces represent a burgeoning field, with the global market projected to exceed $3 billion in 2026 and grow at a 16% compound annual growth rate (CAGR) through 2034, driven by applications in paralysis restoration, sensory prosthetics, and cognitive enhancement. Current BCIs, often rigid electrode arrays, face biocompatibility issues, with 50-70% signal loss within months due to encapsulation by astrocytes.

Northwestern's printed neurons offer a paradigm shift toward biohybrid systems, where synthetic components integrate seamlessly with host tissue. For patients with spinal cord injuries or neurodegenerative diseases like Parkinson's, these could enable precise neural stimulation for movement restoration or tremor suppression. In auditory prosthetics, they might interface directly with cochlear neurons, improving sound fidelity over traditional cochlear implants.

A detailed overview of BCI advancements can be found in recent industry reports, underscoring the need for scalable, flexible solutions like those from Northwestern. Precedence Research BCI Market Forecast.

Revolutionizing Neuromorphic Computing

Beyond medicine, these neurons advance neuromorphic computing—hardware mimicking neural architectures for AI tasks. Traditional von Neumann processors separate memory and computation, creating an "energy wall" where data movement consumes 90% of power. Neuromorphic chips, like Intel's Loihi or IBM's TrueNorth, reduce this by colocating them, achieving up to 1,000 times better energy efficiency for pattern recognition.

Northwestern's devices push further with printable, spiking networks exhibiting multi-order dynamics, potentially enabling edge AI in wearables or IoT without cloud dependency. As AI data centers strain grids—projected to consume 8% of US electricity by 2030—these brain-inspired systems promise sustainability.

Overcoming Key Challenges in Neural Engineering

• Biocompatibility: Flexible polymers and nanoscale inks reduce foreign body response compared to silicon probes.
• Signal Fidelity: Tunable spiking matches biological timescales, unlike prior memristors limited to binary states.
• Scalability: Aerosol printing enables mass production at low cost, democratizing access for research and clinical use.
• Energy Efficiency: Filament-based dynamics minimize components needed for complex behaviors.

Previous Hersam lab work on memtransistors addressed synaptic plasticity, but this iteration achieves neuron-level autonomy. Challenges remain in in vivo longevity and human trials, yet the ex vivo success bodes well.

Northwestern's Role in US Higher Education Research

Northwestern exemplifies US leadership in neuroengineering, with MRSEC funding fostering cross-disciplinary hubs. Similar efforts at MIT (neurogrid), Stanford (neurogrid 2), and UC Berkeley (Tianqiong) complement this, but Northwestern's printable approach stands out for manufacturability. The university's Feinberg School of Medicine integration accelerates translation from bench to bedside.

For students, programs like Northwestern's Neuroengineering PhD track prepare the next generation, with alumni securing roles at Neuralink, Kernel, and Synchron.

Photo by National Cancer Institute on Unsplash

Future Outlook and Actionable Insights

Looking ahead, Hersam's team aims to scale networks for full cortical interfaces and integrate with organoids—mini-brains grown from stem cells—for drug testing. Clinical trials could emerge within 5-10 years, targeting epilepsy or depression via closed-loop modulation.

For researchers and students: Explore MoS₂-based memristors for theses; faculty openings in materials neuroengineering abound. Institutions like Northwestern prioritize interdisciplinary hires, blending EE, materials, and neuroscience.

This work not only advances science but positions US universities at the forefront of a transformative field. For deeper reading, the original paper provides technical details: Nature Nanotechnology DOI: 10.1038/s41565-026-02149-6.

Stakeholders in higher education should invest in nanofab facilities and neuro labs to capitalize on BCI's projected $13B market by 2035.

Career Pathways in Neuroengineering

The rise of biohybrid tech fuels demand for experts. Postdocs in Hersam-like labs gain skills in 2D materials printing, leading to tenure-track positions. Industry roles at Medtronic or Boston Scientific offer $150K+ starting salaries for BCI specialists.

• Pursue MS/PhD in materials science or neuroengineering.
• Gain hands-on printing via MRSEC programs.
• Collaborate across disciplines for holistic training.

Northwestern's ecosystem, with proximity to Chicago's biotech hub, exemplifies opportunity.