Sage Journals: Discover world-class research

Abstract

Shape memory materials are an innovative type of materials that reversibly store a temporary shape and recover back to the original dimensions with the application of an external mechanism such as heat. Shape memory polymers (SMP), specifically thermoplastic SMP (e.g. shape memory polyurethane (SMPU)) have received much attention during the past decade because of the promising future applications and advantages such as ease of processability for thermoplastic SMP (e.g. by 3-D printing), cost, and biocompatibility. In the biomedical field, applications such as stents, surgical sutures, and orthodontic devices, amongst others have been proposed. The addition of fillers to the material can modify the material to improve their load bearing capabilities. Bio-based fillers such as cellulose nanocrystals (CNC) have been proposed in a variety of reinforcing applications. The present work focuses on the experimental description of the addition of nonmodified CNC to SMPU. The work studied the effect on melt-extruded ribbons, for 0, 0.5, 1, 2, and 4 wt%. An increase of yield point, toughness, flexural modulus, recovery rate, and decrease of total time showed that SMPU/CNC nanocomposites are a potential candidate to use in future biomedical applications.

Keywords

Nano crystalline cellulose (NCC)shape memory nanocomposite shape memory polyurethane (SMPU)shape memory effect melt-extrusion

Introduction

Shape memory materials (SMM) are materials with the ability to store a predefined permanent shape and restore to it upon the application of a triggering mechanism.^1,2 Some of the most common triggering mechanisms are temperature, moisture, and ultraviolet (UV) light.¹ The ability of SMM to recover their original/preconditioned shape is called a Shape memory effect (SME). The SME is based on three main steps: (i) setting of the desired/temporary shape; (ii) application of the triggering mechanism; and (iii) recovery of the original/permanent shape.^3,4

Shape memory polymers (SMP) are a subcategory of SMM. Commonly, their SME is activated by heat, that is, increase in temperature.⁵ Although both thermoplastic and thermoset SMP exist, thermoplastic SMP are studied more commonly, because they offer broader possibilities for real-life applications due to their ease of tailor-ability, recyclability, ease of free-form manufacturing and economical costs.¹ Among SMP, polyurethane-based SMP, also called shape memory polyurethanes (SMPU)⁶ are arguably the most commonly used/reported SMP. SMPU can be manufactured to a desired activation temperature by altering the soft to hard segment ratio within their block copolymer structure.⁷ Although many research groups synthesize their own SMP,⁸ there are also commercial suppliers such as the different types and varieties that were developed by Hayashi and are commercially available through SMP Technologies™ DiaPlex (Tokyo, Japan).^9

–12 However, there is still much to be developed in the SMP field, since this commercially available polymer has only a one-way SME. Newer novel laboratory developed materials have been reported such as two-way,¹³ triple,¹⁴ multi-way SME,¹⁵ among others.

SMP have shown favorable applications in the medical, bioengineering, and aerospace fields such as self-assembly mechanisms, mechanical sensors, biosensors, and self-deployable mechanisms.^16

–19 However, properties of thermoplastic SMPU must be tailored to fulfill demanding requirements of engineering industrial applications. Alteration of mechanical, thermal, and shape recovery properties has been reported to be successful through the addition of nanofiller particles to a polymeric matrix.^20

–23 Although successful in mechanical reinforcement, most commonly studied fillers have been carbon based such as graphene, carbon nanotubes, and carbon black particles.^24
–26

Cellulose is a vast resource in our natural environment that can be extracted from plants, cell walls, and living organisms, such as bacteria.²⁷ Cellulose nanocrystals (CNCs) are a relatively new type of cellulose-based nanoparticle group. This nanosized biomass is a more sustainable, biodegradable, and renewable alternative to carbon-based fillers. Due to their rod-like shape with high aspect ratios (i.e. diameters of approximately 8–10 nm and lengths of 100–200 nm), similar to carbon nanotubes, CNCs offer the ability to increase the mechanical properties of a polymeric matrix.^22,28,29 When designing a shape memory polymer composite (SMPC), a trade-off between mechanical strengthening and total shape recovery percentage must be accounted for different types of applications.^{1,22,24,30
–32}

Although some studies have focused on the reinforcement of SMPU with CNC,^22,33
–35 the present study focuses on completing and verifying some of the experimental trends observed on the effect of reinforcing a SMPU matrix with nonmodified CNC, among those characteristics the flexural, toughness (from stress–strain curves), and shape recovery speed. The study will investigate the behavior of melt-extruded CNC/SMPU ribbon composites with 0, 0.5, 1, 2, and 4 wt% (weight percent content) of CNC for future application on the biomedical industry.

Experiment

Materials

A thermoplastic polyurethane block copolymer SMP material, specifically MM4520, from SMP Technologies Inc. (www.smptechno.com), a subsidiary of Mitsubishi, a common trade name is DiAPLEX (Tokyo, Japan). The mentioned material was used as the polymeric matrix. As provided by the manufacturer’s MSDS, MM4520 is compounded of diphenylmethane-4,4′-diisocyanate, adipic acid, ethylene glycol, ethylene oxide, polypropylene oxide, 1,4-butanediol and bisphenol A. MM4520 came in pellet form and has a prescribed glass transition temperature (T _g) of 45°C, temperature which can be used in numerous applications inside the human body.

The CNC particles were supplied by InnoTech Alberta (formerly known as Alberta Innovates Technology Futures (AITF)) located in Edmonton, Alberta, Canada). AITF reported that their process to obtain CNCs consists of shredding Whatman No.1 filter paper into small pieces prior to producing a solution of 10 wt% solid in deionized water. The shredded filter paper was added into 65 wt% of sulfuric acid and later diluted to 10 times the acid solution. The solution was then neutralized with 2 wt% of sodium carbonate and continuously purified via dialysis with deionized water. A dry powder of CNC is later obtained by freeze drying. The reported aspect ratios of these particles are 11.9 in length to diameter with ±3.1 in standard deviation.³⁶

Sample preparation

A Lindberg/Blue M™ vacuum was used to vacuum dry SMPU prior to extrusion. Pellets were dried for 12 h at a constant temperature of 80°C as recommended by the manufacturer.³⁷ Similarly, CNCs were placed in the oven to vacuum dry for the same amount of time at 80°C. Drying times were determined as per manufacturer’s recommendation. Physical mixing of CNC and SMPU was carried out immediately prior to the melt-extrusion with CNC loadings of 0, 0.5, 1, 2, and 4 wt%, to produce SMPU/CNC nanocomposite ribbons.

Melt-extrusion was performed using a HAAKE™ MiniLab Rheomex CTW5 twin-screw extruder. The equipment’s screws are conically shaped, with a length of 109.5 mm and an initial conical diameter of 5 and a final diameter of 14 mm. An extrusion die of 4 mm by 0.5 mm was used to prepare ribbons for testing. A constant shear rate of 70 Ncm was maintained throughout the extrusion process. Extrusion temperature and speed were kept at 195°C and 50 r min⁻¹, respectively. Using a Dynisco™ LME take-up system, the pulling speed of the process was measured using a US Digital encoder (4096 counts per revolution quadrature encoder) and was set at 9.74 r min⁻¹. To avoid plasticization by moisture on SMPU,^38,39 samples were oven dried at $T g + 5 ° C$ for a period of 12 h prior to testing.

Characterization

Differential scanning calorimetry analysis

Differential scanning calorimetry (DSC) analyses were carried out using DSC Q1000 (TA Instruments). Testing was done on samples weighting 2–5 mg. Testing procedure was performed at a heating rate of 20°C fmin⁻¹ for a temperature range of −40°C to 260°C.

Mechanical characterization

Tensile tests

Tensile testing was performed using an Electroforce® 3200 Series by TA Instruments™. Ten samples were tested from CNC 0, 0.5, 1, 2, and 4 wt% content. Prior to testing, specimens were cut to 25 mm in length. Testing jigs required 10 mm for sample gripping, leaving a gage length of 5 mm. Testing was performed using a stretching rate of 0.02 mm s⁻¹ from initial length to 100% strain. Testing jigs were tightened to 17.6 lb-in at each side using a torque wrench. Elastic moduli of all samples were determined using ISO 527-1 “Determination of tensile properties.”⁴⁰ Displacements were verified with optical strain measurement (OSM).⁴¹

Flexural tests

Flexural testing procedure was determined with respect to “Plastics determination of flexural properties” code (ISO178:2010).⁴⁰ A three-point bending test was performed again using the Electroforce 3200 Series by TA Instruments with a three-point bend test jig. Ten specimens were tested for each CNC wt% content. The span length to thickness ratio of the test was maintained at 40.⁴² Specimens’ dimensions were constrained to space limitations of the machine and to prevent effects by shear stress distributions at the midspan.⁴² To comply with the mentioned ratio, specified testing dimensions were set to 20 mm for span length for extruded ribbons of 4 × 0.5 mm. The supports used for the tests are cylindrical with 2 mm in radius. Flexural speed of the test was set to 0.01 mm s⁻¹. Figure 1 shows a general representation of the flexural tests conducted. The recorded data were obtained at the midspan of the sample.

Figure 1.

Picture of performed flexural test on a representative SMPU/CNC.

Shape recovery characterization

The shape recovery characterization was performed in two stages: (i) programming of the temporary shape by applying a thermomechanical load and (ii) activation of the SME by applying a temperature gradient. Prior to testing, samples were cut to 45 mm in length to account for 10 mm of gripping length at each side, therefore having a remaining gage length of 25 mm similar to the tensile tests.

The programming phase was performed using the tensile testing machine (Electroforce 3200 Series by TA Instruments) equipped with a forced air convection–heating chamber. Figure 2(a) and (b) shows the programming phase of a sample. During the test, the sample was attached to the machine’s grips at room temperature (RT). Then, the ambient temperature was raised to 50°C (T _g + 5°C) (Figure 2(a)) until the equilibrium is reached. The sample was then stretched to 100% strain with a rate of 0.2 mm s⁻¹. While maintaining the displacement, the samples were then cooled using forced air to RT and were kept at that temperature to prevent recovery of the sample prior to activation (Figure 2(b)).

Figure 2.

Shape recovery procedure showing: (a) and (b) shape programming and (c, d, e) shape recovery activation.

The activation phase was performed in a radiation-based oven (Lindberg/Blue M). The oven was set to 70°C (T _g + 25°C). The activation phase can be observed in Figure 2(c) to (e), where Figure 2(c) depicts the programmed sample placed inside the oven after it has reached (T _g + 25°C). In the figure, the sample has a minor load of approximately 0.2 N placed at the free end simply to avoid sample curling. Figure 2(d) shows the recovered sample after 2 min of heat exposure. Figure 2(e) shows the final recovered shape at RT. Strain recovery was measured by image processing for images collected at 4 Hz.

Results and discussion

Activation temperature

DSC analyses were conducted to determine the effect of CNC on the T _g of SMPU. DSC was performed on 0, 1, 2, and 4% SMPU/CNC composites. Figure 3(a) shows sample DSC curves. Figure 3(b) shows the T _g versus CNC contents of different samples. As can be seen from the figure, there is a slight decrease and later increase with the addition of CNC to the polymeric matrix. Although slight changes exist, the average effect of addition of CNC (seen by the dotted blue line) to the matrix does not alter the T _g for general application purposes (the activation temperature). Studies on T _g for different thermoplastic polyurethanes reinforced with CNC have encountered similar results.^22,33,35

Figure 3.

(a) DSC curves obtained for different nanocomposites zoomed in at T _g; (b) obtained T _g for CNC wt% of 0, 1, 2, and 4. Data were obtained from DSC curves.

To further look at possible effects of CNC, the melting peaks observed in Figure 3 were also analyzed. A numerical integration was done to obtain the total released heat in the crystallization process. From Figure 3, there is a slight change in the crystallization process due to the addition of CNC particles. The enthalpy of crystallization of a 100% crystalline polyurethane material was considered as $Δ H_{m} = 196.8 J \cdot g^{- 1}$ ,⁴³ and the method of calculation was done according to $% crystallinity = [Δ H_{m} - Δ H_{c}] / Δ H_{m}$ .⁴⁴ Figure 4 shows the multiplication factors by which crystallinity increases with respect to 0 wt% CNC; data points were obtained from the DSC curves; a general increase in crystallinity is observed because CNC allows for nucleation, which affects the mechanical properties of the material.⁴⁵

Figure 4.

Graph for crystallinity obtained from the melting peak of the DSC curves.

Mechanical properties

Tensile properties

Figure 5(a) shows the representative experimental stress–strain curves for different CNC wt% contents. From the experiments, the elastic moduli were calculated for the SMPU matrix and SMPU/CNC nanocomposites. An average of approximately 717.1 MPa was found for the elastic modulus of the non-reinforced SMPU polymer matrix. An analysis of variance (ANOVA) with 95% confidence interval (CI) was conducted to determine the significance of the addition of CNC as reinforcing fillers. Figure 5(b) shows the statistical analysis of the elastic modulus for CNC wt% of 0, 0.5, 1, 2, and 4 by means of box plots, where the red data within the box plot are the mean values 717.1, 647.6, 760.7, 797.4, 651.2 MPa, respectively. On the graph, red crosses are depicted as outliers (a similar analogy will be carried out for all box plots). A resulting p value of 0.4627 > 0.05 determined that the addition of CNC is not statistically significant on the elastic modulus of the polymer. Similar results when adding CNC to thermoplastic polyurethane matrices have shown little to no effect on the Young’s modulus. However, the studies did show a significant effect with respect to the elongation of the material before yield failure.³⁵

Figure 5.

(a) Typical tensile tests’ curves obtained and (b) tensile modulus with respect to CNC wt%.

A similar statistical analysis was conducted for the yield stress of the materials. The yield stress point was considered for the analysis as the maximum stress reached below 10% in strain. The obtained mean yield stresses for 0, 0.5, 1, 2, and 4 wt% are 35.7, 37.6, 31.5, 35.2, and 26.21, respectively. These results are portrayed in Figure 6. A Welch ANOVA returned a p value of approximately 1.25e−05 < 0.05, this shows that the addition of CNC has a significant effect on the yield stress.⁴⁶ Figure 6 shows the obtained results, showing that CNC 0.5 wt% had the most effect (approximately an increase of 5% before material failure). In the graph, a change in variance of the nanocomposites with respect to the raw material is also perceivable; this can be due to a poor dispersion of fillers within the matrix. As suggested by Auad et al.,³³ an increase in maximum stress could possibly mean stress dissipation from the filler particles, which by assuming large strains, we can infer that internal crack growth appears; however, fillers can dissipate crack growth direction and stresses at the crack tip, this phenomenon increases the required force to reach yield of the material. Although an increase in force to yield means an increase in the maximum reached stress, it does not translate into an increase in elastic modulus. Analogically, a decrease in maximum reached stress stands closely related to the existence of agglomerations, as they can be related to large imperfections within the polymer, this can be seen at CNC wt% of 4.

Figure 6.

Resultant box plot for the obtained yield stress from the ANOVA.

Toughness

Since we have only considered the engineering strain in this study, to support the argument that CNC does significantly affect the yield point and to verify the dispersion observed from the statistical deviation, a calculation on toughness was acquired from the stress–strain uniaxial tensile curves. For means of comparison, the point up to where toughness is calculated was considered at 90% of the yield stress (portrayed in Figure 7(a)). Obtained mean values for toughness are portrayed in Figure 7(b). A p value of 6.9947e−06 < 0.05 showed that the addition of CNC does have a significant effect on the material’s toughness. Figure 7(b) shows the box plot results from the obtained ANOVA. A smaller variance in comparison to maximum reached stress can be observed in the graph, additionally 0.5 wt% in content of CNC showed to increase approximately 60% of the original value, whereas 1, 2, and 4 wt% of CNC show a poor dispersion within the samples. An increase in toughness can be related to an increase in strain energy by strain hardening of the material. Possibly strain hardening of the material could be related to strain-induced crystallization.³³

Figure 7.

(a) Toughness calculation and (b) box plot from the obtained ANOVA.

Flexural properties

Figure 8 shows the obtained flexural properties from the conducted tests. Figure 8(a) shows the representative curves of the flexural strengthening of SMP/CNC nanocomposites. The figure shows representative curves of the flexural strengthening with respect to the deflection of the nanocomposite samples at the midspan.

Figure 8.

Flexural properties of SMP/CNC nanocomposites, showing (a) representative curves for flexural strengthening and (b) flexural modulus with respect to the added CNC wt%.

From the curves, the flexural modulus was calculated at the midspan, Figure 8(b) shows box plots of CNC wt% content with respect to the obtained flexural modulus. Mean values for SMPU/CNC nanocomposites resulted in 1136.0, 2178.2, 2531.3, 2115.0, 1968.0 MPa for CNC wt% 0, 0.5, 1, 2, and 4 wt%, respectively. From the results, an ANOVA was performed and a p value of 9.357e−06 < 0.05 returned from the analysis indicated that the addition of CNC is significant on the flexural modulus of the material. The graph shows a maximum increase of approximately 60% in flexural modulus at CNC wt% of 1. Additionally, Figure 8(a) shows a smaller experimental variance in comparison to those observed in tensile tests, since nanocomposite variance is close to that of the polymeric neat material.

An increase in flexural properties can be explained by assuming that the composite has small particle agglomerations but has good dispersion and (ii) that most of the CNC particles are heavily agglomerated in such a way that can be accounted for as small voids within the polymeric matrix. For the first case, a possible explanation is that pure bending occurs and therefore beam theory holds. Figure 9 shows a loaded beam subjected to a force F at the midspan. As such, everything above the neutral axis can be considered in compression while everything below can be considered in tension. It is therefore seen that the compression modulus is different from the tensile modulus. It was experimentally proven from the tensile tests that CNC particles do not contribute to an increase in the modulus. However, above the neutral axis, particles are compressed and therefore clumped closer together, which increases the required force to reach the same strain, increasing the flexural modulus of the SMPC.

Figure 9.

SMPC beam in flexural mode.

Although this explanation could be relevant to the observed phenomena, the addition of too many particles and poor dispersion will create an overcrowding effect, which will cause voids to be created and therefore will decrease the material’s properties.

Shape recovery properties

Strain recovery was characterized by the final shape recovery percentage, R _r obtained from the processed data. The recovery percentage was calculated from equation (1) and the measured recovered strain ε_r was obtained from Figure 2(c) to (e). The assigned strain ε_a for the present study was assigned to 100% (equation (1); Figure 2(b) and (c)).

R_{r} = \frac{(ε_{a} - ε_{r})}{ε_{a}}

Shape recovery properties were determined only for the extreme cases of CNC wt% 0 and 4 content. The main parameters considered for the study were maximum amount of recovery, rate of recovery, and total recovery time. Figure 10(a) shows two strain recovery representative curves for SMP and SMP/CNC at 4 wt% loading. The figure shows a quicker slope for the composite material in comparison to the neat SMPU material. Additionally, the nanocomposite material quickly recovers but later expands, decreasing in total strain recovery. An analogy for this phenomenon is that CNC rods could possibly act as springs within the structure, which during the programming phase of the composite may have suffered plastic deformation. However, such springs can still allow the polymeric matrix to recuperate and they can also contribute to the material’s recovery with their remaining elastic stored strain. Nevertheless, after the maximum stored strain has been recuperated, the plastic deformation of the fillers will counteract the material’s stored strain energy, decreasing in total strain recovery.

Figure 10.

Effect of CNC on the shape recovery properties of SMP MM4520 (a) demonstrates two representative curves for strain recovery and (b) shows box plot from ANOVA for 0, 4 wt% CNC.

From the explained analogy, the maximum strain recovery was considered at 90% below the maximum strain recovery. Figure 10(b) shows the results of a statistical analysis of 90% of maximum strain recovery. From the graph, it is clear that the addition of CNC significantly decreases the maximum recovery. However, from Figure 11, CNC particle addition alters the rate of recovery and recovery time (decreasing the average recovery time to approximately 14.79 s) in a favorable and desirable manner for possible applications.

Figure 11.

Effect of CNC on the shape recovery rate % s⁻¹ with addition of CNC wt%.

Figure 11 shows the statistical analysis (of the recovery rate and recovery time) after performing a one-way ANOVA on the obtained data. For Figure 11(a), the recovery rate increases significantly with a p value of 0.02 < 0.05. Figure 11(b) shows the statistical analysis of the time required for the material to reach 90% of its maximum recovery. A p value of 0.00435 < 0.5 showed that the addition of CNC influences the recovery time of the sample.

Conclusions

Nanocomposites made of unmodified CNC and SMPU matrices were melt-extruded into ribbons to mechanically characterize their properties as well as their shape recovery capabilities. SMPU/CNC loading percentages of 0, 0.5, 1, 2, 4 wt% were manufactured for the tests. Although no statistically significant effect was observed on the elastic modulus of the composite, the addition of CNC did show to have a statistical significance on the yield point of the material, increasing by approximately 5%. Toughness of the material showed to increase with the addition of CNC and the material also achieved a best dispersion with the addition of 0.5 wt% CNC. Flexural properties of the material were also significantly improved at 1 wt%. Moreover, CNC fillers showed to possibly act as embedded springs within the material, allowing for a faster recovery rate and therefore a decrease in the recovery time. However, total recovery percentage was affected when adding 4 wt% CNC.

A clear trade-off between the percent addition of the filler (wt% CNC) and the total recovery percentage must be considered when designing applications with this nanocomposite. Further studies must be done to determine an exact trade-off between the total shape recovery and recovery rate. Since a clear improvement on the recovery speed was observed, possible future work will cover a more detailed description of the recovery force. This work is part of a broader project focusing on biomedical application design such as braided SMP/CNC catheters and sutures.

Footnotes

Acknowledgements

The authors thank and acknowledge the support from Alberta Innovates (BFC16006)—CNC Challenge 2.0 in this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors would like to acknowledge the funds provided by Natural Sciences and Engineering Research Council of Canada (NSERC)—Discovery Grants (418533) and the University of Alberta Mechanical Engineering Department that allowed this research to be conducted. The authors also would like to acknowledge the funds provided by Canadian Foundation for Innovation (CFI—Project #31500) to purchase the testing equipment used for mechanical tests. The authors thank and acknowledge the support from Alberta Innovates (BFC16006)—CNC Challenge 2.0 in this article.

ORCID iD

Irina T Garces

References

Leng

. Multifunctional Shape-Memory Polymers and Actuation Methods In: Shape-memory polymers and multifunctional composites. Boca Raton: CRC Press, 2010; 4(74003): 133–201.

Mather

Luo

Rousseau

. Shape memory polymer research. Annu Rev Mater Res 2009; 39: 445–471.

Huang

Zhao

, et al. Mechanisms of the shape memory effect in polymeric materials. Polymers 2013; 5: 1169–1202.

Leng

Liu

, et al. Shape-memory polymers—A class of novel smart materials. MRS Bull 2009; 34: 848–855.

Moore

Bruck

. A fundamental investigation into large strain recovery of one-way shape memory alloy wires embedded in flexible polyurethanes. Smart Mater Struct 2002; 11: 130–139.

Huang

Yang

Zhao

, et al. Thermo-moisture responsive polyurethane shape-memory polymer and composites: a review. J Mater Chem 2010; 20(17): 3367–3381.

Behl

Lendlein

. Shape-memory polymers. Mater Today 2007; 10(4): 20–28.

Zhou

Chen

, et al. Development of shape memory polyurethane based on polyethylene glycol and liquefied 4,4′-diphenylmethane diisocyanate using a bulk method for biomedical applications. Polym Int 2015; 64(4): 477–485.

Takahashi

Hayashi

. Structure and properties of shape-memory polyurethane block copolymers. J Appl Polym Sci 1996; 60(7): 1061–1069.

10.

Tobushi

Hayashi

Ikai

, et al. Thermomechanical properties of shape memory polymer of polyurethane series and their applications. J Phys IV 1996; 6(1): C1-377–C1-384.

11.

Tobushi

Hashimoto

Hayashi

, et al. Thermomechanical constitutive modeling in shape memory polymer of polyurethane series. J Intell Mater Syst Struct 1997; 8(8): 711–718.

12.

Huang

Yang

Bin

. Polyurethane shape memory polymers. Boca Raton: CRC Press, 2011.

13.

Chen

Yang

, et al. Electroactive two-way shape memory polymer laminates. Polym Compos 2015; 36(3): 439–444.

14.

Ban

Zhu

Chen

, et al. The effect of 4-octyldecyloxybenzoic acid on liquid-crystalline polyurethane composites with triple-shape memory and self-healing properties. Materials (Basel) 2013; 9: 792.

15.

Chen

Yuan

, et al. Insights into liquid-crystalline shape-memory polyurethane composites based on an amorphous reversible phase and hexadecyloxybenzoic acid. J Mater Chem C 2014; 2(6): 1041–1049.

16.

Leng

Sun

, et al. Deployable morphing structure based on shape memory polymer. Aircraft Eng Aerospace Technol 2015; 87(3): 218–223.

17.

Barbarino

Bilgen

Ajaj

, et al. A review of morphing aircraft. J Intell Mater Syst Struct 2011; 22(9): 823–877.

18.

Zhu

Huang

, et al. Recent advances in shape–memory polymers: structure, mechanism, functionality, modeling and applications. Prog Polym Sci 2012; 37(12): 1720–1763.

19.

Kimionis

Isakov

Koh

, et al. 3D-printed origami packaging with inkjet-printed antennas for RF harvesting sensors. IEEE Trans Microw Theory Tech 2015; 63(12): 4521–4532.

20.

Bhattacharya

Kamal

Gupta

. Fundamental issues in nanocomposite synthesis. München: Carl Hanser Verlag GmbH & Co. KG, 2007.

21.

Liang

Gou

Shen, Y

, et al. Carbon fiber reinforced shape memory nanocomposites incorporated with highly conductive carbon nanopaper for electro actuation. In: ASME 2013 Conference on Smart Materials, Adaptive Structures and Intelligent Systems, SMASIS 2013; 2.

22.

Auad

Mosiewicki

Richardson

, et al. Nanocomposites made from cellulose nanocrystals and tailored segmented polyurethanes. J Appl Polym Sci 2010; 115(2): 1215–1225.

23.

Phillips

Baur

, et al. Analysis of shape-memory polymer composites with embedded microvascular system for fast thermal response. J Compos Mater 2015; 49(15): 1881–1893.

24.

Yao

Lin

. Carbon-based reinforcement in shape-memory polymer composite for electrical actuation. Pigment Resin Technol 2013; 43(1): 26–34.

25.

Gunes

Jimenez

Jana

. Carbonaceous fillers for shape memory actuation of polyurethane composites by resistive heating. Carbon 2009; 47(4): 981–997.

26.

Casalini

Bogoslovov

Qadri

, et al. Nanofiller reinforcement of elastomeric polyurea. ADDMA 2012; 53(6): 1282–1287.

27.

Klemm

Heublein

Fink

, et al. Cellulose: fascinating biopolymer and sustainable raw material. Angew Chem Int Ed Engl 2005; 44(22): 3358–3393.

28.

Miao

Hamad

. Cellulose reinforced polymer composites and nanocomposites: a critical review. Cellulose 2013; 20(5): 2221–2262.

29.

Postek

Vladar

Dagata

, et al. Cellulose nanocrystals the next big nano-thing? In: Postek

Allgair

(eds). SPIE, 2008, pp. 70420D–11.

30.

Lan

Liu

, et al. Fiber reinforced shape-memory polymer composite and its application in a deployable hinge. Smart Mater Struct 2009; 18(2): 024002.

31.

Sonseca

Camarero-Espinosa

Peponi

, et al. Mechanical and shape-memory properties of poly(mannitol sebacate)/cellulose nanocrystal nanocomposites. J Polym Sci A Polym Chem 2014; 52(21): 3123–3133.

32.

Ohki

Ohsako

, et al. Mechanical and shape memory behavior of composites with shape memory polymer. Compos A Appl Sci Manuf 2004; 35(9): 1065–1073.

33.

Auad

Contos

Nutt

, et al. Characterization of nanocellulose-reinforced shape memory polyurethanes. Polym Int 2008; 57(4): 651–659.

34.

Saralegi

Gonzalez

Valea

, et al. The role of cellulose nanocrystals in the improvement of the shape-memory properties of castor oil-based segmented thermoplastic polyurethanes. Compos Sci Technol 2014; 92: 27–33.

35.

Navarro-Baena

Kenny

Peponi

. Thermally-activated shape memory behaviour of bionanocomposites reinforced with cellulose nanocrystals. Cellulose 2014; 21(6): 4231–4246.

36.

Going

Sameoto

Ayranci

. Cellulose nanocrystals: dispersion in co-solvent systems and effects on electrospun polyvinylpyrrolidone fiber mats. J Eng Fibers Fabrics 2015; 10(3): 155–163.

37.

Villacres

Ayranci

Nobes

. Additive manufacturing of shape memory polymers: effects of print orientation and infill percentage on mechanical properties. Rapid Prototyp J 2018; 24(4): 1–34.

38.

Yang

Huang

, et al. Effects of moisture on the thermomechanical properties of a polyurethane shape memory polymer. Polymer 2006; 47(4): 1348–1356.

39.

Yang

Bin

. Effects of moisture on electrically conductive polyurethane shape memory polymers. In: Polyurethane Shape Memory Polymers. Boca Raton: CRC Press, 2011.

40.

Materials

ASFTA

. ISO and IEC selected standards for the plastics industry. ASTM Intl, 1996.

41.

Aldrich

Ayranci

Nobes

. OSM-classic: an optical imaging technique for accurately determining strain. SoftwareX 2017; 6: 225–230.

42.

Schneeweiß

Felber

. Review on the bending strength of wood and influencing factors. Am J Mater Sci 2013; 3(3): 41–54.

43.

Pistor

de Conto

Ornaghi

, et al. Microstructure and crystallization kinetics of polyurethane thermoplastics containing trisilanol isobutyl POSS. J Nanomater 2012; 2012(2): 1–8.

44.

Sichina

. DSC as problem solving tool: measurement of percent crystallinity of thermoplastics. Norwalk: PerkinElmer instruments, 2000.

45.

Mariano

Chirat

Kissi El

, et al. Impact of cellulose nanocrystal aspect ratio on crystallization and reinforcement of poly(butylene adipate-co-terephthalate). J Polym Sci B Polym Phys 2016; 54(22): 2284–2297.

46.

Trujillo-Ortiz

Hernandez-Walls

. Welch’s ANOVA: Welch ANOVA test for unequal variances, http://www.mathworks.com/matlabcentral/fileexchange/37121-welchanova (2012, accessed 17 October 2017).

Cellulose nanocrystals (CNC) reinforced shape memory polyurethane ribbons for future biomedical applications and design

Abstract

Keywords

Introduction

Experiment

Materials

Sample preparation

Characterization

Differential scanning calorimetry analysis

Mechanical characterization

Tensile tests

Flexural tests

Shape recovery characterization

Results and discussion

Activation temperature

Mechanical properties

Tensile properties

Toughness

Flexural properties

Shape recovery properties

Conclusions

Footnotes

Acknowledgements

Funding

ORCID iD

References