Structure-property relationships of 3D-printed carbon nanotube/polymer composites

26 December 2017
Huseini S. Patanwala and Anson W. K. Ma
A dimensionless volumetric flow rate is found to be a critical parameter for understanding the microstructure and subsequent mechanical properties of 3D-printed composites.

Fused deposition modeling (FDM) is one of the most common methods of 3D printing. The technique is based on micro-extruding thermoplastic polymers in a raster pattern through a nozzle. To achieve better processibility, most FDM methods use thermoplastics, such as poly(lactic acid) (PLA) and acrylonitrile butadiene styrene.

However, although FDM parts are extremely useful for rapid-prototyping purposes, they lack the physical properties—e.g., mechanical strength and thermal stability (due to weak inter-road bonding and low heat-deflection temperature, respectively)—required for practical applications. There are two general approaches to enhancing the properties of FDM parts: by choosing polymers with better properties (e.g., polyaryletherketone1 or liquid crystalline polymers2); or by incorporating fillers in the neat polymers.3, 4 Indeed, identifying appropriate polymer grades and developing new formulations for 3D printing constitute an active area of research. Compared to conventional processing methods (e.g., injection molding), structure–processing–property relationships in 3D printing are not well established. Many studies have attempted to optimize mechanical properties by varying the process parameters used (e.g., deposition speed, nozzle temperature, the gap between the print nozzle and build plate, the infill pattern, infill density, and part-slicing layer thickness).5, 6 However, because there are many parameters to choose from, and due to the absence of a standardized methodology, optimizing the properties of FDM parts is challenging.

To improve the thermal and mechanical properties of PLA-based FDM parts, we have explored the use of carbon nanotubes (CNTs) as short-fiber fillers. Furthermore, we have investigated the effects of CNT concentration on the structure and subsequent properties of the 3D-printed parts. Finally, in order to lay down the foundation for property optimization, we have proposed an approach that uses a non-geometric process parameter. This parameter, the volumetric flow rate, is capable of influencing the microstructure of the printed composite at a given layer thickness and print speed.

We chose to use CNTs as reinforcement material because of their excellent intrinsic mechanical, thermal, and electrical properties.7–9 Furthermore, CNTs are available in powder form and can therefore be directly blended into PLA and extruded into feedstock filaments for the FDM printer. This process is different from printing continuous-fiber composites, for which a specialized printer and carefully chosen polymers (e.g., Markforged) are required.10,11

In 3D-printed parts, the width of a printed road (see Figure 1) depends on the interplay between the printing flow rate, the velocity of the print nozzle (relative to the build plate), and the gap between the nozzle and the build plate during printing. We therefore evaluated the use of a dimensionless volumetric flow rate (Qr) to predict the microstructure of the 3D-printed part. This parameter is defined as the ratio of the actual volumetric flow rate to the ideal volumetric flow rate (i.e., the rate required to completely fill a given gap between the nozzle and the build plate at a given relative velocity of the print head). Qr is particularly important because it helps to map out regions of ‘under-flow’ and ‘over-flow,’ indicated by values of Qr<1 and Qr>1, respectively. These values enable prediction of the microstructure, and therefore the mechanical properties, of the 3D-printed PLA-CNT composites. For a fixed road-to-road distance, Qr>1 leads to a wider neck, whereas Qr<1 results in a narrower neck, or a lack of bonding between the roads: see Figure 1(a) and (b).

Schematic diagrams of hypothesized factors leading to carbon nanotube (CNT) misalignment during 3D printing: (a) radial flow occurring as the material overfills the gap, and (b) geometry-based fusion between adjacent roads. Qr: Dimensionless volumetric flow rate.

In the absence of CNTs, we found that the Young's modulus and tensile strength of the PLA samples increased as a function of increasing Qr. We attribute this result to the reduction of the void fraction and better bonding between individual PLA roads. However, in the case of the CNT-PLA samples, we observed the highest Young's modulus and tensile strength at a Qr close to 1 (Qr≈1). The trends for the control (i.e., 0% CNT-PLA) and CNT-PLA samples are shown in Figure 2. As in the neat PLA case, we found that increasing the volumetric flow rate for underfilled samples (Qr<1) reduces the void fraction between roads and thus increases the stress transfer between individual roads under tension. Theoretically, as the volumetric flow rate is increased, the shear rate within the liquefier should also increase, thus leading to a higher degree of CNT alignment and consequently a higher modulus. However, we observed the opposite trend in terms of CNT alignment because overfilling (Qr>1) leads to randomization of CNT orientation.

Young's modulus (black), tensile strength (red), and scanning electron microscope (SEM) images of the fractured surface as a function of dimensionless volumetric flow rate (Qr) for CNT/poly(lactic acid) (CNT-PLA) composites with varying CNT content (0–5%). (a) 0% CNT-PLA, (b) 0.5% CNT-PLA, (c) 2.5% CNT-PLA, and (d) 5% CNT-PLA, respectively. The scale shown in (a) applies to all SEM images.

To quantify the degree of CNT alignment, we carried out x-ray diffraction analysis on the fabricated samples, both within printed roads and at the intersections between the roads. To this end, we used full width at half-maximum (FWHM) measurements of the azimuthal ring integral of the 2D diffraction pattern of single-layer FDM samples. As shown in Figure 3, the degree of CNT alignment decreased as a function of increasing Qr. This result is counterintuitive, since a higher Qr would lead to a higher wall shear rate and therefore a higher degree of alignment. These experimental observations can be explained, however, by a combination of flow and geometry-induced effects, as illustrated in Figure 1(a) and (b).

Full width at half-maximum (FWHM) of: (a) 0.5% CNT-PLA, (b) 2.5% CNT-PLA, and (c) 5% CNT-PLA as a function of Qr. The error bar represents the standard deviation of at least three samples.

Further, Figure 3 shows that the CNTs were less aligned at the intersection of adjacent roads compared to those at the center of the road. A higher degree of CNT alignment is expected at the intersections due to the higher shear rates close to the wall within the liquefier. However, we did not observe this in our experimental results. We attribute the reduced alignment of CNTs at the intersection to the convergence zone in the nozzle. For a given volumetric flow rate, the flow velocity increases as the cross-sectional area decreases. This increase in the flow velocity leads to extensional deformations, which further align CNTs at the core,12 thus helping to explain the observed trend.

In summary, we have explored the 3D printing of CNT-PLA composites using an extrusion-based FDM method. By investigating the effects of volumetric flow rate and CNT concentration, we found that the dimensionless volumetric flow rate (Qr) was a useful parameter for understanding the microstructure and subsequent mechanical properties of FDM parts.13 In our future work, we intend to use a similar approach for other functional fillers (e.g., boron nitride nanotubes and cellulose nanofibers) to understand the structure–processing–property relationships for such systems.


Huseini S. Patanwala
Department of Chemical and Biological Engineering, University of Maine

Huseini Patanwala is a research associate whose research focuses on the additive manufacturing of polymer nanocomposites and biomaterials. He is a polymer scientist, and received multiple graduate awards at the University of Connecticut, CT, including a General Electric Graduate Fellowship for Innovation.

Anson W. K. Ma
Department of Chemical & Biomolecular Engineering, University of Connecticut

Anson W. K. Ma is an associate professor whose research focuses on improving the resolution and reliability of additive manufacturing. He is a rheologist, and has received prestigious awards from TA Instruments, the National Science Foundation, the Society of Rheology, 3M, and the American Association of University Professors.


  1. J. Lisagor, A. R. Miller, M. Johnson, N. Podgursky and L. Herran, System and method for 3d printing on permeable materials., US Patent 2016018504, 2016.

  2. R. W. Gray IV, D. G. Baird and J. H. Bøhn, Effects of processing conditions on short TLCP fiber reinforced FDM parts, Rapid Prototyping J. 4, pp. 14-25, 1998.

  3. J. M. Gardner, G. Sauti, J.-W. Kim, R. J. Cano, R. A. Wincheski, C. J. Stelter, B. W. Grimsley, D. C. Working and E. J. Siochi, 3-D printing of multifunctional carbon nanotube yarn reinforced components, Addit. Manuf. 12, pp. 38-44, 2016.

  4. S. Dul, L. Fambri and A. Pegoretti, Fused deposition modelling with ABS–graphene nanocomposites, Compos. Part A: Appl. Sci. Manuf. 85, pp. 181-191, 2016.

  5. O. Rishi, Feed rate effects in freeform filament extrusion,, 2013. Rochester Institute of Technology, New York

  6. O. S. Carneiro, A. F. Silva and R. Gomes, Fused deposition modeling with polypropylene, Mater. Des. 83, pp. 768-776, 2015.

  7. H. J. Qi, K. B. K. Teo, K. K. S. Lau, M. C. Boyce, W. I. Milne, J. Robertson and K. K. Gleason, Determination of mechanical properties of carbon nanotubes and vertically aligned carbon nanotube forests using nanoindentation, J. Mech. Phys. Solids 51 (11–12), pp. 2213-2237, 2003.

  8. P. Kim, L. Shi, A. Majumdar and P. McEuen, Thermal transport measurements of individual multiwalled nanotubes, Phys. Rev. Lett. 87, pp. 215502, 2001.

  9. E. Bekyarova, M. E. Itkis, N. Cabrera, B. Zhao, A. Yu, J. Gao and R. C. Haddon, Electronic properties of single-walled carbon nanotube networks, J. Am. Chem. Soc. 127, pp. 5990-5995, 2005.

  10. R. Matsuzaki, M. Ueda, M. Namiki, T.-K. Jeong, H. Asahara, K. Horiguchi, T. Nakamura, A. Todoroki and Y. Hirano, Three-dimensional printing of continuous-fiber composites by in-nozzle impregnation, Sci. Rep. 6, pp. 23058, 2016.

  11. N. Li, Y. Li and S. Liu, Rapid prototyping of continuous carbon fiber reinforced polylactic acid composites by 3D printing, J. Mater. Process. Technol. 238, pp. 218-225, 2016.

  12. B. P. Heller, D. E. Smith and D. A. Jack, Effects of extrudate swell and nozzle geometry on fiber orientation in Fused Filament Fabrication nozzle flow, Addit. Manuf. 12, pp. 252-264, 2016.

  13. H. S. Patanwala, D. Hong, S. R. Vora, B. Bognet and A. W. K. Ma, The microstructure and mechanical properties of 3D printed carbon nanotube-polylactic acid composites, Polym. Compos., 2017.

DOI:  10.2417/spepro.006981