Elsevier

Carbon

Volume 76, September 2014, Pages 321-329
Carbon

SWCNT-thin-film-enabled fiber sensors for lifelong structural health monitoring of polymeric composites - From manufacturing to utilization to failure

https://doi.org/10.1016/j.carbon.2014.04.083Get rights and content

Abstract

With glass, polyaramid, nylon and polyethylene terephthalate as the fiber substrate, a continuous spray coating and winding process was developed for fabricating SWCNT-FibSen sensor – single-walled carbon nanotube (SWCNT) thin film enabled 1D fiber sensor. The microstructures of the SWCNT-FibSen and its piezoresistive behavior were respectively characterized by scanning electron microscopy and Raman spectroscopy and evaluated by the coupled electromechanical tensile test. With embedding the SWCNT-FibSen sensor in the fiberglass prepreg laminates at different orientations and locations, its use for multipurpose sensing in life-long (from manufacturing to failure) structural health monitoring of high-performance polymeric composites was demonstrated. During the manufacturing process, the sensor was able to provide valuable local resin curing information. After the manufacturing process, the sensor was useful in mapping the stress/strain states of the host composite under different deformation modes – tension, bending, compression and failure. The integrity of the sensing performance of the embedded SWCNT-FibSen sensor was demonstrated upon a 10,000-cyclic coupled electromechanical test.

Introduction

Structural health monitoring (SHM) integrates sensors and sensor networks with structures, e.g., composite aviation vehicles and civil infrastructures, to provide a system-level technology for detection, identification, quantification and decision about their health states [1], [2]. The sensors and sensing system that have been traditionally adopted for SHM applications include metallic strain gage, semiconductor or metal oxide thin or thick films [3], [4], piezoelectric sensors [5], [6], optical fiber sensors [7], [8], [9], [10], eddy-current sensors [11], [12], and magnetostrictive sensors [13]. The recent discovery of the highly sensitive piezoresistivity of carbon nanotubes (CNTs) stimulates great interests in exploring this novel material for developing smart and multifunctional sensors for SHM applications [14], [15], [16]. With CNT thin film or buckypaper as the 2-dimensional (2D) sensing elements, the utility of CNTs for strain and corrosion sensing has been demonstrated [17], [18], [19]. In addition, by dispersing CNTs as 3-dimensional (3D) distributed sensing elements, Chou et al. systematically studied their usefulness for damage and failure sensing of glass fiber composites [20], [21], [22]. Using a 1D ensemble structure of CNTs referred to as CNT yarn, Zhao et al. [23] and Abot et al. [24] both demonstrated the value of CNT based 1D sensing elements as embedded sensors for monitoring the crack initiation and propagation in composite structures. With glass fiber as substrate, Sebastian et al. [25] and Zhang et al. [26] respectively applied chemical vapor deposition (CVD) and electrophoretic deposition method to fabricate CNT enabled 1D fiber sensors and explored their application as embedded sensors for damage detection of epoxy matrix or related composites.

The embeddable and non-invasive characteristics of CNT enabled fiber sensors make these novel 1D sensing elements highly valuable in SHM for mapping strain, detecting damage, and monitoring crack initiation and propagation. To further advance this emerging field, in this paper, we report single-walled carbon nanotube (SWCNT) thin film enabled fiber sensors SWCNT-FibSen that were fabricated through a continuous spray coating and winding process. In comparison to the previously developed CNT enabled 1D sensors [23], [24], [25], [26], the fabrication of SWCNT-FibSen is simple, cost-effective, and environmental benign. The sensing characteristics of SWCNT-FibSen can be tailored through manipulating the SWCNT structures in the dispersion [27], forming hybrids with other types of nanomaterials [28], and selecting different types of fiber substrates, such as glass fiber, polyaramid fiber, nylon fiber, and PET fiber. Moreover, when embedded into polymeric composites, the SWCNT-FibSen can be used as a multipurpose sensor for life-long (from manufacturing to failure) SHM of the host structure. During the composite manufacturing process, the resin curing can be in situ monitored by the resistance change of the SWCNT-FibSen. Given its embedded nature, the SWCNT-FibSen provides the resin curing information in the interior of the host composite structure, which cannot be readily accessed by other techniques, e.g., differential scanning calorimetry (DSC) [29]. This is considered a great advantage of SWCNT-FibSen for its use in improving the quality assurance of composite manufacturing process. After the composite manufacturing process, the same embedded SWCNT-FibSen is capable of detecting the different mechanical deformation modes – tension, bending, and compression as well as the failure of the host structure. The strain/stress mapping has been demonstrated with multiple SWCNT-FibSen sensors deployed at the prescribed orientations and/or locations of the host composite. Upon a 10,000-cyclic coupled electromechanical test, the superior robustness of the embedded SWCNT-FibSen sensor was demonstrated. After test, the sensor resistance and piezoresistive sensitivity only showed a minor change of 1.2% and 3.7%, respectively. The large-scale manufacturing and facile deployment capabilities of SWCNT-FibSen sensors as well as their non-invasiveness, robustness, and the ability for in situ manufacturing process monitoring make SWCNT-FibSen a vantage technique for accurate local damage detection and strain mapping in many complex engineered systems made with lightweight composites, such as a manned or unmanned vehicle airframe or a space satellite optical mirror support.

Section snippets

Preparation and characterization of aqueous dispersion of SWCNTs

The SWCNT dispersion was prepared by sonicating a mixture of 0.12 g SWCNT raw materials (Lot # PO258, Unidym™) and 2.4 g sodium dodecylbenzensulfonate (SDBS, CAS # 25155-30-10, Sigma–Aldrich) in 100 g deionized water in an ice bath using a Misonix 3000 probe sonicator (20 kHz). The sonicator was operated in a pulse mode (on 10 s, off 30 s) with the power set at 45 W for 6 h. The as-sonicated dispersion was characterized with the recently developed preparative ultracentrifuge method (Optima™ MAX-XP

Structural characterization of SWCNT-FibSen sensors

With the in-house developed continuous spraying coating process, we have successfully fabricated SWCNT-FibSen sensors on different types of fiber substrates, which included glass fiber, polyaramid fiber, nylon fiber, and polyethylene terephthalate (PET) fiber. In this paper, we focus on SWCNT-FibSen sensors formed on a single filament of glass fiber. Fig. 2a shows the optical photographs of the bare glass fiber bundles and the corresponding SWCNT coated SWCNT-FibSen sensor. The dark visual

Conclusions

A continuous spray coating process was developed to coat SWCNT thin films on different types of single fiber filament substrates, which included glass fiber, polyaramid fiber, nylon and PET fiber, for fabricating SWCNT-FibSen piezoresistive fiber sensors. With epoxy/glass fiber composites processed by vacuum bagging and prepreg laminates as examples, the SWCNT-FibSen sensors have been demonstrated to be readily embedded into the polymeric composite structures with conventional composite

Acknowledgement

The authors acknowledge the funding support provided by the Air Force Office of Scientific Research (AFOSR) and supervised by Dr. David S. Stargel.

References (41)

  • N. Hu et al.

    Investigation on sensitivity of polymer/carbon nanotube composite strain sensor

    Carbon

    (2010)
  • T.C. Theodosiou et al.

    Numerical investigation of mechanisms affecting the piezoresistive properties of CNT-doped polymers using multi-scale models

    Compos Sci Technol

    (2010)
  • C. Boller et al.

    Encyclopedia of Structural Health Monitoring

    (2009)
  • F. Chang

    Structural Health Monitoring 2013 – a Roadmap to Intelligent Structures

    (2013)
  • A.L. Window et al.

    Strain Gauge Technology

    (1992)
  • L. Moletn et al.

    Review of fatigue monitoring of agile military aircraft

    Fatigue Fract Eng Mater Struct

    (2000)
  • S.S. Kessler et al.

    Damage detection in composite materials using lamb wave methods

    Smart Mater Struct

    (2002)
  • X. Qing et al.

    Built-in sensor network for structural health monitoring of composite structure

    J Int Mat Sys Struct

    (2007)
  • N. Takeda

    Summary report of the structural health-monitoring project for smart composite structure systems

    Adv Comp Mat

    (2001)
  • B. Culshaw

    Optical fiber sensor technologies: opportunities and – perhaps – pitfalls

    J Lightwave Technol

    (2004)
  • Cited by (103)

    View all citing articles on Scopus
    View full text