Dual Piezoresistive Sensor System
Wearable Pulse Wave Monitoring

Synergistic Micropyramid & Porous Sensing with PPG Integration

📄 Co-Author Publication: ACS Applied Materials & Interfaces, 2024, 16, 65402-65413
DOI: 10.1021/acsami.4c17710
Journal: ACS Applied Materials & Interfaces (Impact Factor ~9.5)
Role: Co-author - Sensor fabrication, characterization, and data analysis
8.2 kPa⁻¹
Sensitivity (0-1 kPa)
0.79 kPa⁻¹
Sensitivity (1-20 kPa)
95/145 ms
Response/Recovery Time
5000 cycles
Durability Test
217 g
Wearable System Weight
Sensor Schematic
Figure 1: Schematic of the dual piezoresistive pressure sensing system
Wearable Prototype
Figure 2: Compact wearable signal collection prototype (217g)

Project Overview

This research presents a novel dual piezoresistive pressure sensor that synergistically employs two complementary resistance-changing mechanisms to balance the trade-off between high sensitivity and linear response. The sensor combines a micropyramid component (for high sensitivity in low-pressure regime) and a porous carbon black-coated filter paper component (for extended range and linearity). This work also introduces a fully wearable, compact signal collection device (217g) capable of simultaneously collecting signals from multiple sensors, including PPG integration for localized vascular health monitoring.

Motivation & Gap Addressed

Despite successful development of piezoresistive pressure sensors for pulse waveform acquisition, transitioning from laboratory to clinical trials remains challenging due to:

  • Testing schemes that don't mirror real-world wearable functionality
  • Complexity and size of signal acquisition devices (bulky sourcemeters)
  • Challenges related to scaling and manufacturing
  • Trade-off between high sensitivity and linear response

Sensor Design & Components

Dual Sensing Mechanisms:

Component Mechanism Pressure Range Sensitivity
Micropyramid Array Contact area change with IDE 0-1 kPa (Low) 13 kPa⁻¹ (standalone)
Porous Carbon Black Filter Paper Tunneling effect through porous structure 1-20 kPa (Mid) 0.7 kPa⁻¹ (standalone)
Dual Sensor (Synergistic) Both mechanisms combined 0-20 kPa (Full range) 8.2 kPa⁻¹ (low), 0.79 kPa⁻¹ (mid)

Materials Used:

  • Micropyramid Component: PDMS with 200nm gold sputter coating
  • Porous Component: Whatman filter paper (grade 4, 25μm pore size) + carbon black (20 wt%) + silicone grease
  • Electrode: Screen-printed interdigitated electrodes (IDEs)
  • Packaging: Kapton tape
  • Signal Acquisition: ESP32-S, Arduino Nano BLE 33, ADS1256 (24-bit ADC), 1000 mAh 7.2V LiPo battery
  • Additional Sensor: Photoplethysmography (PPG) sensor for cross-validation

Fabrication Process

Micropyramid Component:

  1. Silicon Mold Creation: Laser mask writer (μPG101) → photolithography → dry etching (RIE) → wet etching (KOH bath at 80°C for 4 hours)
  2. Surface Treatment: Silanization with perfluorooctyl trichlorosilane (1 hour in vacuum desiccator)
  3. PDMS Casting: 10:1 ratio (cross-linker to base), degassed 30 min, cured at 20°C for 72 hours
  4. Metal Deposition: 200nm gold sputter coating via PVD

Porous Component:

  1. Conductive Grease Preparation: Carbon black (20 wt%) mixed with silicone grease
  2. Coating: 0.1 mL conductive grease coated on filter paper (1×1 cm²)
  3. Drying: 100°C on hot plate for 24 hours
  4. Storage: Vacuum desiccator to maintain dryness

Sensor Assembly:

  1. Micropyramid and porous components positioned side-by-side facing IDE
  2. External wires soldered to IDE terminals
  3. Encapsulation with Kapton tape
Micropyramid SEM
Figure 3: SEM images of micropyramid array (100μm base, ~70μm height)
Porous Material SEM
Figure 4: Cross-sectional SEM of porous carbon black component

Finite Element Analysis (FEA)

FEA was conducted to understand the relationship between output current and pressure for materials of varying conductivities.

Key Findings:

  • Conductivity below 1 S/m → signal output below 100 nA → poor SNR
  • High conductivity coatings (10⁵ S/m, e.g., 200nm gold) → milliamp range current outputs → enables wearable signal acquisition
  • Decreasing conductivity from 10³ to 1 S/m reduces sensitivity from 1 to 0.66 kPa⁻¹

Performance Characterization

Pressure Sensitivity Comparison:

Sensor Type Low Pressure (0-1 kPa) Mid Pressure (1-20 kPa) High Pressure (20-90 kPa)
Micropyramid Only 13.0 kPa⁻¹ 0.41 kPa⁻¹ 0.12 kPa⁻¹
Porous Filter Paper Only - 0.70 kPa⁻¹ 0.09 kPa⁻¹
Dual Piezoresistive Sensor 8.2 kPa⁻¹ 0.79 kPa⁻¹ 0.11 kPa⁻¹

Dynamic Response:

  • Response Time: 95 ms (loading at 10.1 kPa)
  • Recovery Time: 145 ms (unloading at 10.1 kPa)
  • Durability: 5000 two-stage compression cycles
  • Stepwise Response: Reliable from 9.25 to 86 kPa
  • Frequency Response: Tested at 0.5, 1.0, and 1.5 Hz (30-90 BPM)
Sensitivity Graph
Figure 5: Pressure response comparison of three sensor configurations
Cyclic Test
Figure 6: 5000-cycle durability test with two-stage waveforms

Wearable Signal Acquisition System

System Components:

  • Microcontrollers: ESP32-S + Arduino Nano BLE 33
  • ADC: ADS1256 (24-bit external analog-to-digital converter)
  • Power: 1000 mAh 7.2V LiPo battery + LM2596 DC-DC step-down (to 5.5V)
  • Weight: 217 grams
  • Battery Life: 6+ hours continuous operation
  • Data Resolution: 30 kHz maximum
  • Data Transmission: Bluetooth Low Energy (BLE)

Integrated PPG Sensor:

  • Purpose: Cross-validation of piezoresistive signals
  • Benefit 1: Rejects motion artifacts that may affect one sensor but not the other
  • Benefit 2: Enables simultaneous measurement of blood volume (PPG) and pressure waveforms
  • Benefit 3: Provides robust performance under varied environmental conditions
  • Benefit 4: Allows assessment of vascular stiffness and pulse wave velocity (PWV)

Pulse Wave Monitoring Results

Wrist Pulse Waveform:

The sensor successfully captured full pulse waveforms from the wrist artery with three characteristic peaks:

  • Percussion Peak (P-wave): Systolic peak from ventricular contraction
  • Tidal Peak (T-wave): Reflected wave from upper body
  • Diastolic Peak (D-wave): Reflection from lower body

Multi-Site Monitoring:

  • Wrist: Successful pulse waveform acquisition with all three peaks resolved
  • Ankle (Foot): Comparable signal quality to wrist measurement
  • Simultaneous PPG + Piezoresistive: Forward-going vs backward-going wave analysis

Comparison with Benchtop Sourcemeter:

The wearable system demonstrated strong correlation with Keithley 2460 sourcemeter readings, confirming the reliability and accuracy of the compact wearable system for practical cardiovascular monitoring.

Pulse Waveform
Figure 7: Simultaneous pulse waveform from dual piezoresistive and PPG sensors

Material Characterization

Raman Spectroscopy (Conductive Grease):

  • G-band (1600 cm⁻¹): Graphitic carbon (sp² hybridized)
  • D-band (~1340 cm⁻¹): Defects/disorder (sp³ hybridized carbon)

FTIR Spectroscopy:

  • 3400 cm⁻¹: O-H stretching (water absorption in porous carbon black)
  • 2800-3000 cm⁻¹: C-H stretching (aliphatic hydrocarbons from grease)
  • 1750 cm⁻¹: C=O stretching (esters from synthetic/natural oils)
  • 1630 cm⁻¹: C=C stretching (unsaturated hydrocarbons)

Equipment Used

  • μPG101 Laser Mask Writer
  • Reactive Ion Etching (RIE)
  • Physical Vapor Deposition (PVD) Sputtering
  • JSM-F100 Scanning Electron Microscope (SEM)
  • FT/IR-4100 (FTIR Spectroscopy)
  • MultiView 2000TS Raman Spectroscopy
  • Mark-10 Force Gauge
  • Keithley 2460 Sourcemeter
  • ESP32-S / Arduino Nano BLE 33
  • ADS1256 24-bit ADC

Downloads

Key Innovations

  • First dual piezoresistive sensor combining micropyramid and porous mechanisms for balanced sensitivity and linearity
  • Synergistic design prevents signal saturation while maintaining high sensitivity across full pressure range
  • Compact wearable signal acquisition system (217g, 6+ hour battery) replaces bulky benchtop sourcemeters
  • Integrated PPG + piezoresistive sensors for cross-validation and localized vascular health monitoring
  • Multi-site monitoring capability (wrist and ankle) for comprehensive vascular assessment
  • Scalable manufacturing via reusable silicon mold for micropyramids

Technology Readiness Level (TRL) Advancement

This study specifically aimed to enhance the technology readiness level of piezoresistive pressure sensors by addressing:

  • Testing Protocols: Real-world wearable functionality testing instead of benchtop-only
  • Signal Acquisition: Compact, battery-powered system vs. bulky sourcemeters
  • Scaling: Reusable molds and simple fabrication processes
  • Linearity vs Sensitivity: Novel dual-mechanism approach to balance trade-offs

Comparison with State-of-the-Art

The dual piezoresistive sensor outperforms many recent piezoresistive sensors in both sensitivity and linearity across the critical blood pressure range (0-20 kPa), as detailed in Supporting Information Table S1.

Funding

  • National Science Foundation (NSF) Awards #2423124 and #2412500
  • NSF PATHS-UP ERC (Award #1648451)
  • Dissertation Year Fellowship (DYF) - Florida International University (B.J. and A.H.C.)
  • Doctoral Evidence Acquisition (DEA) Fellowship - Florida International University (B.J.)

Related Publications

  • Jafarizadeh, B., Chowdhury, A.H., Sozal, M.S.I., Cheng, Z., Pala, N., Wang, C. "Wearable System Integrating Dual Piezoresistive and Photoplethysmography Sensors for Simultaneous Pulse Wave Monitoring." ACS Applied Materials & Interfaces, 2024, 16, 65402-65413.
  • Jafarizadeh, B., Chowdhury, A.H., Khakpour, I., Pala, N., Wang, C. "Design Rules for a Wearable Micro-Fabricated Piezo-Resistive Pressure Sensor." Micromachines, 2022, 13, 838.
  • Chowdhury, A.H., Jafarizadeh, B., Pala, N., Wang, C. "Paper-Based Supercapacitive Pressure Sensor for Wrist Arterial Pulse Waveform Monitoring." ACS Applied Materials & Interfaces, 2023, 15, 53043-53052.

Conclusion

This study successfully developed a dual piezoresistive pressure sensor that synergistically combines micropyramid and porous carbon black-coated filter paper components to achieve high sensitivity (8.2 kPa⁻¹ in low range, 0.79 kPa⁻¹ in mid range) and improved linearity across the blood pressure range (0-20 kPa). The sensor demonstrated excellent durability (5000 cycles) and fast response time (95/145 ms). A compact wearable signal acquisition system (217g) was developed, replacing bulky benchtop sourcemeters, and successfully captured full pulse waveforms from both wrist and ankle locations. Integration with PPG sensors enables cross-validation and localized vascular health monitoring. This work significantly advances the technology readiness level of piezoresistive pressure sensors for practical, continuous cardiovascular monitoring applications.

← Back to All Projects