See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/270049786 A technique for real-time detecting, locating, and quantifying damage in large polymer composite structures made of... Article in Structural Health Monitoring · January 2015 DOI: 10.1177/1475921714546063 CITATIONS READS 9 84 2 authors: Ali Naghashpour S. V. Hoa 18 PUBLICATIONS 57 CITATIONS 286 PUBLICATIONS 2,899 CITATIONS Concordia University Montreal SEE PROFILE All content following this page was uploaded by Ali Naghashpour on 21 June 2015. The user has requested enhancement of the downloaded file. Concordia University Montreal SEE PROFILE Original Article A technique for real-time detecting, locating, and quantifying damage in large polymer composite structures made of carbon fibers and carbon nanotube networks Structural Health Monitoring 2015, Vol. 14(1) 35–45 Ó The Author(s) 2014 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/1475921714546063 shm.sagepub.com Ali Naghashpour1,2 and Suong Van Hoa1,2 Abstract A significant safety concern preventing extensive use of composite materials for large polymer composite structures is the ability to detect, locate, and quantify damages that occur at one or several locations in large polymer composite structures. Real-time health monitoring of large polymer composite structures improves their performance, durability, and reliability while minimizing the life cycle cost. In this article, we present a new, practical, and real-time structural health monitoring technique for detecting, locating, and quantifying damages in large polymer composite structures made of carbon fibers and carbon nanotube networks. In this technique, electrically conductive epoxy resin was prepared by dispersing multiwalled carbon nanotubes into epoxy matrix. This modified epoxy matrix was then incorporated with long carbon fibers to make large composite plates. Two sets of grid points made from silver-epoxy paste were mounted on the surface of the large plates. The first set was used to apply the constant electric current, and the second set was utilized to measure the electric potential. The electric potentials across the second set of grid points on the undamaged plate were measured and used as a reference set. Two different damages were created by drilling holes and by applying impact loading on the large plates. It is found that the electric potential between the contact points surrounding the damage changes. The significant change in electric potential corresponds to the damage location in the plates. As such, drilled holes, impact damages, and barely visible impact damages are detected, located, and quantified. Keywords Damage location, damage detection, carbon nanotubes, large polymer matrix composite structures, electrical properties Introduction Carbon fiber–reinforced polymer composites (CFRPCs) have received attention in many industrial applications due to their high strength-to-weight and stiffness-toweight ratios. One of the critical challenges in the practical use of CFRPCs is to monitor the health of large polymer composite structures (LPCSs) in real time due to their susceptibility to different types of damages.1 The long-term use of the composites depends on the ability to detect and locate the damages in the structures. Various non-destructive evaluation (NDE) techniques such as X-ray tomography,2,3 ultrasonic C-scanning,2,4 liquid penetrant,2,5 acoustic emission,2,4,6 piezoelectric active sensors,2,7 fiber optics,2,4,8 and measuring electrical conductivity along the direction of carbon fibers2,9 have been used for detecting damage in CFRPC structures. Techniques such as liquid penetrant, ultrasound scanning, and X-ray can detect damages in composites, but these techniques can only work in a laboratory setting. They cannot be used on real composite structures during the operation of the engineering component. Techniques such as acoustic emission can 1 Concordia Centre for Composites (CONCOM), Department of Mechanical & Industrial Engineering, Concordia University, Montreal, QC, Canada 2 Center for Applied Research on Polymers and Composites (CREPEC), Montreal, QC, Canada Corresponding author: Ali Naghashpour, Concordia Centre for Composites (CONCOM), Department of Mechanical & Industrial Engineering, Concordia University, 1455 De Maisonneuve Blvd. W., Montreal, QC H3G 1M8, Canada. Email: al_nag@encs.concordia.ca Downloaded from shm.sagepub.com at CONCORDIA UNIV LIBRARY on June 15, 2015 36 Structural Health Monitoring 14(1) detect the occurrence of damage in composite structures. However, it is very difficult for this technique to locate the damage due to confusion created by extraneous sources. Techniques such as the use of fiber optical sensors can measure strains in the composite structures. The optical fibers may pick up the change in strains due to the occurrence of defects. However, the diameter of the optical fiber is large (52 mm, which is more than six times the diameter of a carbon fiber). When the optical fibers are incorporated within the composite structures, they create stress concentration that can degrade the mechanical properties of the composite structures. Besides, the optical fibers are fragile and tend to break fairly easily.9,10 Over the past few years, measuring the electrical conductivity of the carbon fibers was used as a technique to indicate the presence of damage by many researchers.11–24 Since carbon fibers are electrically conductive along the fiber direction, by applying an electric current over two probes at two points along the direction of the fibers, the change in electric potential can be taken as an indication of damage in CFRPC structures. Schulte and Baron11 first proposed electrical resistance change measurement (ERCM) for structural health monitoring (SHM), which can be used to monitor the internal damage of CFRPCs. Wang et al.20,21 showed that the electric potential change measurement (EPCM) is more sensitive than ERCM to detect internal damage in CFRPC plates where the electric potential probes are close to the electric current probes (current input locations). The problem with this technique is that since the resin is not conductive, one cannot use the technique to detect resin cracks. The majority of damage at the relatively low loads is due to matrix cracking and delamination, rather than fiber breakage. As such, the usefulness of this technique is limited. Recent advent of polymer nanocomposites in which carbon nanotubes (CNTs) are added in polymers has provided the impetus for scientists and researchers in producing functionally tailored matrix and fibers. This is because CNTs possess outstanding properties, including structural, mechanical, electrical, and thermal properties.25 The outstanding properties of CNTs combined with their small size offer the potential to not only modify polymers but also detect both strain and subsequently failure in polymer matrix composites (PMCs). Adding CNTs at small concentrations in a polymer matrix to form electrically conductive networks distributed around the structural fiber reinforcement displays piezoresistive behavior. The piezoresistive behavior of the CNT networks enables their use as highly responsive sensors to monitor initiation and detection of matrix cracks in the structures.26– 28 Zhang et al.29 embedded CNTs in graphite fiber/ epoxy laminates to improve their electrical conductivity in the thickness direction due to continuous electrical conduction pathways made by CNTs in between graphite fibers. This approach was used to detect delamination created by inserting a Teflon film in the laminates. They found that there is a good correspondence between the delamination length and changes in through-thickness electrical resistance. Kostopoulos et al.30 dispersed CNTs in carbon fiber/epoxy composite not only for improving the electrical conductivity of the composite in transverse direction but also for detecting matrix cracks in the composite. They found that the addition of CNTs in the composites acts as direct sensors with high damage sensitivity to detect matrix damage accumulation during monotonic and cyclic tensile loading. The above works illustrate very interesting and innovative attempts to monitor damage in CFRPC coupons with small size using CNT networks. No technique has existed to detect and locate damage in LPCSs made of carbon fibers and CNT networks. Here, we provide a technique to detect, locate, and quantify damage in large carbon fiber/epoxy/CNT composite plates. Experimental methods Materials Multiwalled carbon nanotubes (MWCNTs) with 95% purity, diameters of 2–20 nm, and lengths of 1 mm to more than 10 mm were purchased from Bayer MaterialScience. Plain weave woven carbon fabric (5.7 oz/yd2) purchased from US Composites, Inc. with thickness of 0.01 in, Epon 862 and EPIKURE W purchased from Miller-Stephenson Chemical Company were utilized as reinforcement, epoxy resin, and curing agent, respectively. Methods Fabrication of composite plates. To manufacture the carbon fiber/epoxy/MWCNT composite plates, 0.3 wt% MWCNT (as the optimal quantity of MWCNT)27 was dispersed into epoxy resin mixed with curing agent (26.4 wt%) using three-roll milling (EXAKT 80E; EXAKT Technologies, Inc.). The mixture was heated up to 60°C for 20 min in a vacuum oven to remove air bubbles. The modified epoxy matrix was dispersed in six layers of plain weave carbon fabric by hand layup. The composite plates were cured using an autoclave. Electrical measurement strategy. Electrical measurement can be performed by two methods for detecting damage in composite structures. One is a two-probe method and the other is a four-probe method. In the two-probe method, two electrical contacts are used to apply constant electric voltage and to determine electrical resistance based on the measured electric current. In the Downloaded from shm.sagepub.com at CONCORDIA UNIV LIBRARY on June 15, 2015 37 Naghashpour and Hoa Figure 1. Schematic illustrating the strategy for EPM using the four-probe method. EPM: electric potential measurement. four-probe method, four electrical contacts are used. One pair of electrical contacts is utilized to apply constant electric current between the two probes, while the other pair is used to measure electric potential over two points in the vicinity of the first two points. Contact resistance exists between the probes and the plate. In situations where the intrinsic resistance of the material is much larger than the contact resistance at the probes (such as glass fiber/epoxy composites containing CNT), the effect of the contact resistance is small and the twoprobe method can be used effectively. However, in situations where the intrinsic resistance of the material is small compared to the contact resistance at the probes (such as the case of carbon fiber/epoxy composites containing CNT), contact resistance is more dominant and the two-probe method is not suitable. In this case, four-probe method is more suitable since electric potential does not depend on the contact resistance at the probes. Wang et al.22 demonstrated that four-probe method is very effective and accurate compared to twoprobe method for detecting damage in CFRPC structures. As such, in this work, four-probe method is used to minimize contact resistance. Another issue in using electrical method for damage detection in twodimensional plate is current spreading that reduces damage sensitivity. High sensitivity for detection of minor damage in the structures can be achieved where current input location is close to the electric potential line.20,21 This close distance between electric current line and electric potential line provides high current density while minimizing current spreading. A new strategy for electric potential measurement (EPM) is proposed to overcome the aforementioned issues for in situ damage monitoring in LPCSs made of carbon fibers and CNT networks. The EPM strategy is described in Figure 1. In this figure, two sets of grid points are applied to avoid the contact resistance issue. The first set is used to inject constant electric current by pairs of I1I,2I, I1I,3I, I2I,4I, I3I,4I, I2I,3I, and I1I,4I (not shown for clarity). The second set is utilized to measure electric potentials between adjacent pairs of V1V,2V, V1V,3V, V2V,4V, V3V,4V,V2V,3V, and V1V,4V (not shown for clarity), where I and V stand for electric current and potential. The subscript numbers, separated by a space, represent the associated first and second sets of grid points, respectively. Composite plate specification and electrical measurement strategy. The specification of composite plate and a new strategy for EPM for in-situ damage monitoring are schematically illustrated in Figure 2. In this figure, two Downloaded from shm.sagepub.com at CONCORDIA UNIV LIBRARY on June 15, 2015 38 Structural Health Monitoring 14(1) Figure 2. Schematic illustration of composite plate specification and strategy for EPM. EPM: electric potential measurement. sets of grid points were mounted on the surface of CFRPC structures. Each set consists of 40 electrical contact points. Each contact point was made from electrically conductive silver-epoxy paste. The first set of grid points labeled from 1I to 40I spaced 3 in apart was used to apply a constant electric current to CFRPC structures. The second set of grid points labeled from 1V to 40V spaced 3 in apart was diagonally shifted by 0.2 in with respect to the first set to measure electric potentials. Electrical wires were attached to the two grid points to make electrodes. Then, the electrodes were connected to the data acquisition system (Vishay Micro-Measurements System 7000) and a computer with program. The EPM strategy corresponding to nearest points 13I and 13V is described in Figure 3. In this figure, the pairs of first set of grid points corresponding to point 13I identified by solid red lines to inject constant electric current are I13I,7I, I13I,8I, I13I,9I, I13I,12I, I13I,14I, I13I,17I, I13I,18I, and I13I,19I, while the pairs of second set of grid points corresponding to point 13V identified by dashed blue lines to measure electric potentials are V13V,7V, V13V,8V, V13V,9V, V13V,12V, V13V,14V, V13V,17V, V13V,18V, and V13V,19V. The current input location to measure electric potential between one point and its nearest neighbors (e.g. for points 13I and 13V) is presented in Table 1. In this table, for example, constant electric current is injected between points 13 and 7 (I13I,7I), while electric potential is measured between points 13 and 7 (V13V,7V). Another example is to inject constant electric current between points 13 and 8 (I13I,8I) while measuring electric potential between points 13 and 8 (V13V,8V). Electric potential measurement. The EPM is used to detect damage in carbon fiber/epoxy/MWCNT composite plates. The EPM adopts the electrically conductive carbon fibers and CNTs themselves as self-sensing materials. The EPM was performed by four-probe method using Keithley 6220 DC, Keithley 218A, and Downloaded from shm.sagepub.com at CONCORDIA UNIV LIBRARY on June 15, 2015 39 Naghashpour and Hoa Drilled holes and impact tests Two damage types were introduced in carbon fiber/ epoxy/MWCNT composite plates. One was drilled holes of different sizes at different locations in the plates. The other was impact damages caused by collision with high-velocity projectiles and drop weights on the plates. The plates were subjected to high-velocity impacts with energy of 78 J produced by a 318 mg aluminum particle traveling at 700 m/s using a gas gun. The details of high-velocity impact test set-up to detect and locate damage are shown in Figure 4. The drop weights were applied on the clamped plates placed on electrically non-conductive rigid supports to create the low-velocity impacts. Result and discussion Figure 3. Schematic illustration for describing pairs of electrical contacts corresponding to points 13I and 13V to inject constant electric current (solid red lines) and to measure electric potentials (dashed blue lines). Table 1. Pairs of electrical contacts nearest to points 13I and 13V. Pairs of electrical contacts Constant current input location Measured electric potential 13–7 13–8 13–9 13–12 13–14 13–17 13–18 13–19 I13I,7I I13I,8I I13I,9I I13I,12I I13I,14I I13I,17I I13I,18I I13I,19I V13V,7V V13V,8V V13V,9V V13V,12V V13V,14V V13V,17V V13V,18V V13V,19V Vishay Micro-Measurements System 7000. A constant current (100 mA) was directly applied by the mounted first grid points through the plate using Keithley 6220 DC. Then, the electric potential across the second grid points was measured using Keithley 218A. Electric potential change (EPC) is expressed as DV (%) = Vf, iv, jv  VI, iv, jv 3100 VI, iv, jv ð1Þ where VI, iv, jv and Vf, iv, jv represent the initial and final electric potential values between grid points iv and jv, respectively. The electric potential values between pairs of second set of grid points spaced 3 in apart before damage for 22 3 13 in2 carbon fiber/epoxy/0.3 wt% MWCNT composite plate 1 were measured and used as reference values. The electric potential distribution of plate 1 is shown in Figure 5(a). Coefficient of variation in percent (100 3 standard deviation divided by average electric potential) as a measure of uniformity of electric potential distribution for carbon fibers/epoxy/0.3 wt% MWCNT composite plate 1 before damage was determined to be 14%. Drilled holes and impacted areas A hole of size 1/16 in was drilled in plate 1. The electric potential values were measured after the hole was drilled. These values were compared against the reference values. The difference between the electric potential values before and after drilled hole was calculated based on equation (1). The hole is detected and located based on the significant local variation in the distribution of the EPC, as shown in Figure 6(a). Subsequently, another hole of size 2/16 in was drilled in plate 1. It is observed from Figure 6(b) that the significant local variations in the distribution of the EPC reveal the locations of holes of sizes 1/16 and 2/16 in drilled in plate 1. Other holes with different sizes were then made in plate 1. Observing Figures 6(c) and 6(d), good correspondence is observed between the significant local variations in the distribution of the EPC and holes of sizes 1/16, 2/16, 3/16, 4/16, 5/16, and 6/16 in, respectively, drilled in plate 1. Figure 6(e) shows the effect of hole volume on the change in electric potential. In this figure, the numbers below the curve represent the pairs of electric potential probes. This pair of probes is closest to the hole. A clear relationship between hole volume and change in electric potential is observed in Downloaded from shm.sagepub.com at CONCORDIA UNIV LIBRARY on June 15, 2015 40 Structural Health Monitoring 14(1) Figure 4. Experimental set-up to detect and locate damage due to high-velocity impact test. Figure 5. (a) Distribution of electric potential of plate 1. The black dots, labeled 1–40, are representations of electric potential contacts. (b) Close-up view of electrical contact points. Figure 6(e). This reveals the capability of the technique to determine severity of the damages. High-velocity projectiles were impacted using a gas gun in plate 2. Figures 7(a) and 7(b) show the locations and values of the change in electric potential due to the collision with high-velocity projectiles (78 J each). By comparing Figures 7(b) and 7(c), it is clear that impact damages 1 and 2 are detected and located according to the sharp local variations in the distribution of the EPC. This indicates that the technique is capable of detecting and locating impact damages. To explain the obtained experimental results, from Ohm’s law, the electrical resistance R is defined by R= V I ð2Þ where R, V , and I are electric resistance, electric potential, and electric current, respectively. The relationship between electrical resistivity as a material property and electrical resistance is expressed as Downloaded from shm.sagepub.com at CONCORDIA UNIV LIBRARY on June 15, 2015 R= rl wt ð3Þ 41 Naghashpour and Hoa Figure 6. Electric potential change distribution of plate 1 after drilling: (a) hole 1 (1/16 in), (b) holes 1 and 2 (1/16 and 2/16 in); (c) holes 1–6 (1/16, 2/16, 3/16, 4/16, 5/16, and 6/16 in); (d) plate 1 (22 3 13 in2) after six drilled holes; and (e) the effect of hole volume on the change in electric potential (data are presented as mean 6 standard deviation from three experiments). where r, w, t, l, and R represent electrical resistivity, width of sample, thickness of sample, distance over which the electrical resistance is determined, and electrical resistance, respectively. Substituting equation (2) into equation (3) I= V r wtl ð4Þ In the case of electric potential in plate measured using the four-probe method, constant electric Downloaded from shm.sagepub.com at CONCORDIA UNIV LIBRARY on June 15, 2015 42 Structural Health Monitoring 14(1) Figure 7. Electric potential change distribution of plate 2 after (a) impact damage 1 (78 J) and (b) impact damages 1 and 2 (78 J each). (c) Plate 2 (22 3 13 in2) after two impact damages. current is injected. The electric current flowing through the plate before damage (I1) and after damage (I2) was kept constant. Equation (4) can be written as l V2 r2 w22t2 = V1 r1 wl1t 1 1 ð5Þ where subscript numbers 1 and 2 represent before and after damage, respectively. If the resistivity, the distance over which the electric potential is measured, and the thickness of plate do not change, equation (5) can be given as V2 w 1 = V1 w 2 ð6Þ If a hole (broken fibers and matrix) is made in plate, this hole would result in a reduction in width (w2 \w1). This reduction would cause an increase in electric potential (V2 . V1) according to equation (6). The fraction of change in electric potential is given as DV V2  V1 w1  w2 = = V1 V1 w2 ð7Þ where w2 is smaller than w1 due to hole. From a physical point of view, as the amount of electric charge flowing at a given time (electric current) is constant, the energy to push the amount of electric charge flowing through the plate (electric potential) is increased due to hole. As such, any discontinuity such as hole in plate would cause an increase in electric potential. Downloaded from shm.sagepub.com at CONCORDIA UNIV LIBRARY on June 15, 2015 43 Naghashpour and Hoa Figure 8. (a) Schematic illustration of 22 3 13 in2 plate 3 impacted at three regions using drop weights with impact energies of 1–3 J. Absolute electric potential change distribution of plate 3 after (b) BVI damage 1 (1 J) and (c) BVI damages 1–3 (1–3 J). BVI: barely visible impact. Barely visible impact damages Plate 3 was tested under low-velocity impact performed using drop weights with different energy levels ranging from 1 to 3 J. These energy levels were applied to create barely visible impact damages (BVIDs) that cannot be detected by visual observations. Plate 3 impacted at three different locations with different energy levels is schematically illustrated in Figure 8(a). Figures 8(b) and 8(c) show the locations and values of the changes in absolute electric potential for plate 3 due to lowvelocity impact tests. Comparing Figures 8(a) and 8(c), good correspondence is found between the significant local variations in absolute EPC distribution and barely visible damage zones. This reveals that BVIDs 1–3 produced by different energy levels at different locations in plate 3 are detected and located distinctly. From impact damage detection and location points of view, significant change in electric potential can be taken as an indication of impact damage detection and location in the large structures. However, complex electrical responses due to low-velocity impact damage were observed by many researchers for CFRPC structures.17–24,31,32 It was shown that an impact causes electric current path distortion, which results in a reduction in electric potential in the plane of CFRPC plate.20–22 Another explanation is that impact test may cause partial reduction in Downloaded from shm.sagepub.com at CONCORDIA UNIV LIBRARY on June 15, 2015 44 Structural Health Monitoring 14(1) the thickness direction of carbon fiber/epoxy composite laminates. This may result in fiber–fiber contacts at the interlaminar interface and minimizing contact resistance, which may cause a reduction in electric potential, and increasing electric current flow in interlaminar carbon fiber laminates. As such, electrical conductivity of the laminates increases with permanent local deformation of resin caused by impact.23,24,31,32 Conclusion A new, practical, and real-time SHM technique has been developed to detect, locate, and quantify damages in the large polymer composite plates made of carbon fibers and CNT networks. In this technique, electric current was applied through the large plate, and electric potentials were measured. This electric potential distribution in the undamaged plate was used as a reference map. Real-time monitoring of electric potential distribution over the surface of the plate was performed to provide an actual map of electric potential distribution. This map was compared against a reference map to identify significant changes in electric potential. These significant changes provide the ability for detection, location, and quantification of damages and BVIDs in the large plates. Acknowledgements The authors acknowledge the support from Natural Sciences and Engineering Research Council of Canada (NSERC) and Concordia Center for Composites (CONCOM). Funding This work was funded by the Natural Sciences and Engineering Research Council of Canada (NSERC) (grant number: RGPIN/413-2007) and Concordia Center for Composites (CONCOM). References 1. Chou TW. Microstructural design of fiber composites. Cambridge: Cambridge University Press, 1992. 2. Balageas D, Fritzen CP and Guemes A. Structural health monitoring. London: ISTE Ltd, 2006. 3. McCombe GP, Rouse J, Trask RS, et al. X-ray damage characterization in self-healing fibre reinforced polymers. Compos Appl Sci Manuf 2012; 43: 613–620. 4. Takeda S, Aoki Y, Ishikawa T, et al. Structural health monitoring of composite wing structure during durability test. Compos Struct 2007; 79: 133–139. 5. Masters JE. Damage detection in composite materials. Philadelphia, PA: American Society for Testing and Materials, 1992, p. 35. 6. Rizzo P and Di Scalea FL. Acoustic emission monitoring of carbon-fiber-reinforced polymer bridge stay cables in large-scale testing. Exp Mech 2001; 41(3): 282–290. 7. Park G, Faffar CF, di Scalea FL, et al. Performance assessment and validation of piezoelectric active-sensors in structural health monitoring. Smart Mater Struct 2006; 15: 1673–1683. 8. Kirkby E, de Oliveira R and Mänson JA. Impact localization with FBG for a self-healing carbon fibre composite structure. Compos Struct 2011; 94: 8–14. 9. Lee DC, Lee JJ and Yun SJ. The mechanical characteristics of smart composite structures with embedded optical fibre sensors. Compos Struct 1995; 32: 39–50. 10. Seo DC and Lee JJ. Effect of embedded optical fibre sensors on transverse crack spacing of smart composite structures. Compos Struct 1995; 32: 51–58. 11. Schulte K and Baron CH. Load and failure analyses of CFRP laminates by means of electrical resistivity measurements. Compos Sci Tech 1989; 36: 349–356. 12. Chung DL. Structural health monitoring by electrical resistance measurement. Smart Mater Struct 2001; 10: 624–636. 13. Schueler R, Joshi SV and Schulte K. Damage detection in CFRP by electrical conductivity mapping. Compos Sci Tech 2001; 61(6): 921–930. 14. Todoroki A and Tanaka Y. Delamination identification of cross-ply graphite/epoxy composite beams using electrical resistance change method. Compos Sci Tech 2002; 62: 629–639. 15. Hou L and Hayes SA. A resistance-based damage location sensor for carbon-fibre composites. Smart Mater Struct 2002; 11: 966–969. 16. Wang S, Chung DDL and Chung JH. Impact damage of carbon fiber polymer–matrix composites, studied by electrical resistance measurement. Compos Appl Sci Manuf 2005; 36: 1707–1715. 17. Todoroki A, Tanaka Y and Shimamura Y. Multi-probe electric potential change method for delamination monitoring of graphite/epoxy composite plates using normalized response surfaces. Compos Sci Tech 2004; 64(5): 749–758. 18. Iwasaki A and Todoroki A. Statistical evaluation of modified electrical resistance change method for delamination monitoring of CFRP plate. Struct Health Monit 2005; 4(2): 119–136. 19. Wang D and Chung DDL. Comparative evaluation of the electrical configurations for the two-dimensional electric potential method of damage monitoring in carbon fiber polymer–matrix composite. Smart Mater Struct 2006; 15(5): 1332–1344. 20. Wang D, Wang S, Chung DDL, et al. Comparison of the electrical resistance and potential techniques for the selfsensing of damage in carbon fiber polymer–matrix composites. J Intell Mater Syst Struct 2006; 17: 853–861. 21. Wang D, Wang S, Chung DDL, et al. Sensitivity of the two-dimensional electric potential/resistance method for damage monitoring in carbon fiber polymer–matrix composite. J Mater Sci 2006; 41: 4839–4846. Downloaded from shm.sagepub.com at CONCORDIA UNIV LIBRARY on June 15, 2015 45 Naghashpour and Hoa 22. Wang S, Wang D and Chung DDL. Method of sensing impact damage in carbon fiber polymer–matrix composite by electrical resistance measurement. J Mater Sci 2006; 41: 2281–2289. 23. Todoroki A, Shimazu Y and Misutani Y. Electrical resistance reduction of laminated carbon fiber reinforced polymer by dent made by indentation without cracking. J Solid Mech Mater Eng 2012; 6(12): 1042–1052. 24. Suzuki Y, Todoroki A, Matsuzaki R, et al. Impact-damage visualization in CFRP by resistive heating: development of a new detection method for indentations caused by impact loads. Compos Appl Sci Manuf 2012; 43(1): 53–64. 25. Li C, Thostenson ET and Chou TW. Sensors and actuators based on carbon nanotubes and their composites: a review. Compos Sci Tech 2008; 68: 1227–1249. 26. Thostenson ET and Chou TW. Carbon nanotube networks: sensing of distributed strain and damage for life prediction and self-healing. Adv Mater 2006; 18: 2837–2841. 27. Naghashpour A and Hoa SV. A technique for real-time detection, location and quantification of damage in large 28. 29. 30. 31. 32. polymer composite structures made of electrically nonconductive fibers and carbon nanotube networks. Nanotechnology 2013; 24(455502): 1–9. Naghashpour A and Hoa SV. In-situ monitoring of through-thickness strain in composite laminates using carbon nanotube sensors. Compos Sci Tech 2013; 78: 41–47. Zhang W, Sakalkar V and Koratkar N. In situ health monitoring and repair in composites using carbon nanotubes additives. Appl Phys Lett 2007; 91(31): 1-3. Kostopoulos V, Vavouliotis A, Karapappas P, et al. Damage monitoring of carbon fiber reinforced laminates using resistance measurements, improving sensitivity using carbon nanotube doped epoxy matrix system. J Intell Mater Syst Struct 2009; 20: 1025–1034. Todoroki A, Mizutani Y and Suzuki Y. Fatigue damage detection of CFRP using the electrical resistance change method. Int J Aeronaut Space Sci 2013; 14(4): 350–355. Todoroki A, Haruyama D, Mizutani Y, et al. Electrical resistance change of carbon/epoxy composite laminates under cyclic loading under damage initiation limit. Open J Compos Mater 2014; 4: 22–31. Downloaded from shm.sagepub.com at CONCORDIA UNIV LIBRARY on June 15, 2015 View publication stats