PPESK-Modified Multi-Functional Epoxy Resin and Its Application To

polymers

Article PPESK-Modiﬁed Multi-Functional Epoxy Resin and Its Application to the Pultrusion of Carbon Fiber

Liwei Wang 1,2, Jinyan Wang 1, Fengfeng Zhang 1, Yu Qi 1, Zhihuan Weng 1,* and Xigao Jian 1,*

1 Liaoning High Performance Resin Engineering Research Center, Department of Polymer Science and Materials, Dalian University of Technology, Dalian 116024, China; [email protected] (L.W.); [email protected] (J.W.); [email protected] (F.Z.); [email protected] (Y.Q.) 2 PetroChina Co., Ltd., Jilin Petrochemical Branch, Jilin 132022, China * Correspondence: [email protected] (Z.W.); [email protected] (X.J.); Tel.: +86-411-8498-6107 (Z.W.)

Received: 24 July 2018; Accepted: 23 September 2018; Published: 26 September 2018

Abstract: Multi-functional epoxy resins are generally brittle due to their high crosslinking densities, which can limit their use for applications that require impact resistance. Pultruded poly(phthalazinone ether sulfone ketone) (PPESK)-modified epoxy resins were prepared and their curing behaviors, heat resistance properties, and viscosity changes investigated. The glass transition temperature of these resins was found to increase with increasing PPESK content; however, these values were still compatible with the pultrusion process. Little change in the tensile strength and elongation lengths at breaking point were observed for blended PPESK/multi-functional epoxy resin containing 4–6% PPESK, and its viscosity levels were still within the requirements of the pultrusion process. Carbon fiber/multi-functional epoxy resin/PPESK (CF/E/PPESK) composites were also prepared and their performance investigated. The bending radius of these PPSEK-modified composites could reach up to 55 D with no cracking or peeling observed in their surface layers. The fatigue frequency of the sinusoidal waveforms for the composite did not change after one million fatigue test cycles, meaning that a strength retention rate of >90% was achieved. Therefore, this study describes a powerful approach for preparing toughened multi-functional epoxy resins that are well suited to pultrusion processes.

Keywords: carbon fiber; multi-functional epoxy resin; polyethersulfone resin; pultrusion process; curing

1. Introduction The advantages of using multi-functional epoxy resins include high temperature resistance, low shrinkage, good stability, and high mechanical strength. However, the brittle nature of these resins after curing, their poor resistance to impact, and the appearance of stress cracks over time limits their application under harsh conditions [1–3]. Modiﬁed multi-functional epoxy resins have been widely studied in recent years, resulting in tough high-temperature epoxy resins now being available [4–6]. Many methods have also been developed to successfully enhance the fracture toughness of epoxy resin [7–9], with rubbers often used as effective toughening additives for epoxy resin. However, the presence of nanorubber particles dispersed throughout the resin matrix can result in a large increase in its viscosity, which can cause difﬁculties when it is used for pultrusion. Another powerful alternative for toughening the epoxy resin is to use thermoplastics, which can maintain both the mechanical and thermal properties of the resin. These toughened epoxy resins are produced through blending epoxy resin with high-performance engineering thermoplastics, such as polysulfone [10], poly(ether sulfone) [8,11], poly(ether imide) [12,13], poly(amide-amidic acid) [14] and poly(ether ketone) [15,16]. Reaction-induced phase separation is the major toughening mechanism currently used to engineer thermoplastic/epoxy blend composites [12,17]. We have previously

Polymers 2018, 10, 1067; doi:10.3390/polym10101067 www.mdpi.com/journal/polymers Polymers 2018, 10, 1067 2 of 10 reported a series of engineering phthalazinone-derived thermoplastics, whose twisted aromatic structures confer high mechanical strengths and good thermal properties on the toughened epoxy resins [18–21]. For example, a novel thermoplastic poly(phthalazinone ether nitrile ketone) (PPENK) ◦ 0 with a Tg of 280 C[22] was blended with an epoxy-based tetraglycidyl 4,4 -diaminodiphenyl-methane (TGDDM)/bisphenol A-type novolac (F-51) resin that was cured using 4,40-diaminidiphenysulfone (DDS) [20]. This resulted in the formation of a composite with numerous intermolecular and intramolecular hydrogen cross-links throughout the TGDDM/F-15/PPENK blend. These hydrogen bonding interactions resulted in good compatibility between the PPENK and the epoxy resin, with the composite’s critical stress intensity factor and impact strength values reaching a maximum when 10 phr PPENK was used. Poly(phthalazinone ether sulfone ketone) (PPESK) is a new type of high-performance, temperature-resistant resin characterized by its resistance to high temperatures and a good solubility profile, with a glass transition temperature of between 250–310 ◦C. The good toughness, high modulus, heat resistance, and excellent overall performance of PPESK resins makes them suitable as modifiers for strengthening multi-functional epoxy resins [2]. PPESK-modified multi-functional epoxy resins could be potentially used for pultrusion of high-temperature composites that should exhibit good temperature resistance, improved resistance to impact, and reduced stress cracking. In this study, the curing behavior, compatibility, viscosity change, and related mechanical properties of CF/PPESK-modified multi-functional epoxy resins were investigated. These resins were found to exhibit good mechanical and torsional strength properties, excellent resistance to fatigue, and good aging properties.

2. Materials and Methods

2.1. Materials Polyphenylene ether sulfone resins (PPESK) with an average molecular weight of 4.52 × 104 g/mol were purchased from Dalian Polymer New Materials Co. Ltd. (Dalian, China). Carbon ﬁbers (CF, JHTD-45, 12 K) with a tensile strength of 4.9 GPa were obtained from the Carbon Fiber Factory of China Petroleum Jilin Petrochemical Company (Jilin, China). Multi-functional epoxy resins (CP02A) and an acid anhydride curing agent (CP02B) were purchased from Changshu Jiafa Resin Co. Ltd. (Changshu, China). Reagent-grade chloroform (99%) was obtained from Shanghai Youshi Chemical Co. Ltd. (Shanghai, China). A release agent (1890 M) was purchased from Beijing Kesila Technology Co. Ltd. (Beijing, China).

2.2. Preparation of Samples Preparation of blended resins. Blended solutions were prepared using 0, 2, 4, 6, and 8 phr of PPESK in CP02A chloroform, where phr represented the weight of PPESK per hundred parts of CP02A. The solvent was then removed using vacuum evaporation at 70 ◦C. The resultant PPESK was then mixed with CP02B and 1890 M and magnetically stirred to produce a homogeneous solution. The blended resin was poured into a mold, vacuum defoamed, and cured at 90 ◦C/2 h, 160 ◦C/4 h and 200 ◦C/2 h sequentially. Samples were then cooled to room temperature and demolded. This general procedure was used to prepare blended resins containing different doses of PPESK of 0, 2, 4, 6, and 8 phr that were classified as E/PPESK-0, E/PPESK-2, E/PPESK-4, E/PPESK-6 and E/PPESK-8, respectively. Preparation of pultruded carbon fiber bars. An overview of the method used to produce the carbon fiber pultruded bars is shown in Figure1 and Table1. Different mass ratios of PPESK/CP02A, CP02B and 1890 M (internal release agent containing 1–2% CP02A) were stirred for 10–20 min. Each of the blended resin mixtures was then transferred to a dipping tank, and a pultrusion machine used to perform dipping and pultrusion. For example, a ϕ25 carbon fiber bar was prepared using 700 shafts of JHTD45-12K carbon fiber that was then passed through the dipping tank to create a preformed mold Polymers 2018, 10, 1067 3 of 10

Polymers 2018, 10, x FOR PEER REVIEW 3 of 11 that was heateddipping and tank then to create processed a preformed through mold that a tractor was heated and and cutting then processed machine. through The resulting a tractor and composites were classiﬁedcutting as machine. CF/E/PPESK. The resulting composites were classified as CF/E/PPESK.

Creel Collecting Cut-off stand Guide Soaking Heating yam Preformed unit roller tank mold mold

Tractor

FigureFigure 1. Carbon 1. Carbon ﬁber fiber barbar pultrusion process. process.

Table 1. PultrusionTable 1. Pultrusion conditions conditions for carbon for carbon fiber/poly(phthalazinone fiber/poly(phthalazinone ether ether sulfone sulfone ketoneketone (CF/PPESK) (CF/PPESK) resin/multi-functionalresin/multi-functional epoxy epoxy resin resin composites. composites. Phase 1 (°C) Phase 2 (°C) Phase 3 (°C) Mold Length (m) Pultrusion Rate (m/min) Resin Content (%) ◦ ◦ ◦ Phase 1 ( C)170 Phase 2 ( C)180 Phase 3 (195C) Mold 0.9–1.2 Length (m) Pultrusion 0.4–0. 6 Rate (m/min) 20–24 Resin Content (%) 170 180 195 0.9–1.2 0.4–0.6 20–24 2.3. Characterization 2.3. CharacterizationThe thermal stabilities of the epoxy resins were tested using differential scanning calorimetry (DSC, Mettler Toledo, Zurich, Switzerland) and thermogravimetric analysis (TGA, Mettler Toledo, The thermalZurich, Switzerland). stabilities For of theDSC epoxyanalysis, resins the glass were transition tested temperature using differential of cured 10 mg scanning samples was calorimetry (DSC, Mettlermeasured Toledo, on a Zurich,Mettler DSC1 Switzerland) differential andscanning thermogravimetric calorimeter, using a analysisheating rate (TGA, of 10 Mettler°C/min Toledo, over a temperature range of 50–350 °C. Thermogravimetric analysis (TGA) was carried out using a Zurich, Switzerland).Mettler TGA1 Forthermal DSC gravimetric analysis, analysis the glass instrume transitionnt under temperature a flow of nitrogen of cured (50 mL/min) 10 mg samplesat a was ◦ measured onheating a Mettler rate of DSC1 10 °C/min differential over a temperature scanning calorimeter,range of 30–700 using °C. aThe heating tensile ratestrength of 10 of theC/min over a temperaturecomposites range was of 50–350 determined◦C. Thermogravimetricusing a WSM-50KN elec analysistric universal (TGA) testing was carriedmachine out(Intelligent using a Mettler TGA1 thermalInstrument gravimetric Equipment, analysis Changchun, instrument China) according under to a the flow GB/T1447-2005 of nitrogen fiber-reinforced (50 mL/min) plastic at a heating ◦experimental protocol, using a displacement rate of 2◦ mm/min for five repeat samples. Torsional rate of 10 strengthsC/min overwere determined a temperature according range to the of 30–700standard C.GB/T10128-2007 The tensile method strength using of thea NWS1000 composites was determinedmicrocomputer-controlled using a WSM-50KN electrictorsion test universal machine testing (Changchun machine Research (Intelligent Institute Instrumentfor Mechanical Equipment, Changchun,Science China) Co., accordingLtd., Changchun, to the China) GB/T1447-2005 operating in chuck fiber-reinforced angle control mode plastic at a experimental rate of 6°/min. protocol, using a displacementTensile fatigue rate tests of were 2 mm/min conducted foron a five SDS-20 repeat0 electro-hydraulic-servo samples. Torsional fatigue strengths testing weremachine determined (Changchun Research Institute for Mechanical Science Co., Ltd., Changchun, China) at 25 °C using a according10 to Hz the frequency standard for one GB/T10128-2007 million cycles. This method testing cycle using involved a NWS1000 a pull, release, microcomputer-controlled press, release and torsion testpull machine cycle, using (Changchun a pull force of Research70 kN and a Institutedownward for force Mechanical of 10 kN. This Sciencemechanical Co., process Ltd., could Changchun, China) operatingbe represented in chuck as a angle sine-wave, control with mode its maximum at a rate wave of 6◦ peak/min. corresponding Tensile fatigue to the tests maximum were conducted on a SDS-200tensile electro-hydraulic-servo force, and its lowest wave fatiguepeak corresp testingonding machine to the minimum (Changchun pressure. Research Repetition Institute of for several cycles resulted in a decrease in the mechanical properties of the materials, with the Mechanical Science Co., Ltd., Changchun, China) at 25 ◦C using a 10 Hz frequency for one million resultant decrease in tensile force and pressure measured by software analysis of changes to the cycles. Thissine-wave testing curve. cycle involvedThe morphology a pull, of the release, blended press, resin releasewas determined and pull by spraying cycle, using the fracture a pull force of 70 kN andsurface a downward of the resin force with of 10gold, kN. followed This mechanical by analysis process using a couldKYKY2800B be represented scanning electron as a sine-wave, with its maximummicroscope wave(SEM, KYKY peak Technology corresponding Co., Ltd., to theBeijing, maximum China). Toughness tensile force,tests were and conducted its lowest wave peak correspondingaccording to tothe theGB29324 minimum standard pressure. by winding Repetition the composite of resin several onto a cycles 55 D cylinder resulted using in aa test decrease in rotation speed of 3 r/min. the mechanical properties of the materials, with the resultant decrease in tensile force and pressure measured3. by Results software analysis of changes to the sine-wave curve. The morphology of the blended resin was determined by spraying the fracture surface of the resin with gold, followed by analysis 3.1. Curing Behavior and Viscosity of E/PPESK Resin Blends using a KYKY2800B scanning electron microscope (SEM, KYKY Technology Co., Ltd., Beijing, China). Toughness testsProduction were conducted of composite according materials using to the pultrusi GB29324on confers standard significant by winding efficiencies the associated composite resin with fast molding and high production rates, with the ideal process requiring low mixing viscosities, onto a 55 Da cylinderslow increase using in aviscosity test rotation over time speed at room of 3 temperature, r/min. and a rapid heat-curing reaction. Consequently, the curing behavior of E/PPESK epoxy resin blends was examined to determine the 3. Results

3.1. Curing Behavior and Viscosity of E/PPESK Resin Blends Production of composite materials using pultrusion confers signiﬁcant efﬁciencies associated with fast molding and high production rates, with the ideal process requiring low mixing viscosities, a slow increase in viscosity over time at room temperature, and a rapid heat-curing reaction. Consequently, the curing behavior of E/PPESK epoxy resin blends was examined to determine the impact of the PPESK content on the curing process. A step-up temperature program was used to investigate the Polymers 2018, 10, 1067 4 of 10

Polymers 2018, 10, x FOR PEER REVIEW 4 of 11 curing behavior of composite epoxy resins, with the curing behavior of different PPESK mass ratios impact of the PPESK content on the curing process. A step-up temperature program was used to (0, 2, 4, 6, and 8 phr) determined using DSC at a heating rate of 10 ◦C/min (see Figure2). The heat investigate the curing behavior of composite epoxy resins, with the curing behavior of different releasedPPESK from mass the curing ratios (0, reaction 2, 4, 6, and and 8 thephr) curingdetermined temperature using DSC wereat a heating found rate to increaseof 10 °C/min slightly (see as the PPESKFigure content 2). increased.The heat released This is from because the curing PPESK reaction has aand relatively the curing high temperature glass transition were found temperature to (280 ◦C),increase withsmall slightly amounts as the PPESK of PPESK content serving increased. to plasticizeThis is because the networkPPESK has structure a relatively of high the epoxyglass resin. This plasticizationtransition temperature is beneficial (280 to°C), the with overall small amounts flow of theof PPESK epoxy serving resin, to which plasticize results the network in an increase in the cross-linkingstructure of the density epoxy resin. [23] andThis heatplasticization resistance is beneficial of the polymer to the overall blend. flow The of the high epoxy glass resin, transition which results in an increase in the cross-linking density [23] and heat resistance of the polymer temperatureblend. of The PPESK high meansglass transition that distortion temperature of the of network PPESK structuremeans that of distortion the epoxy of resin the chainsnetwork becomes increasinglystructure difficult of the epoxy as the resin PPESK chains content becomes increases. increasingly difficult as the PPESK content increases.

1.5

E/PPESK-0 E/PPESK-2 1.1 E/PPESK-4 E/PPESK-6 E/PPESK-8

0.7

0.3 Heat Flow (mW/mg) Flow Heat

-0.1 50 100 150 200 250 o Temperature ( C) Figure 2.FigureDynamic 2. Dynamic differential differential scanning scanning calorimetry calorimetry(DSC) (DSC) curves of of resin resin systems systems at scanning at scanning rate rate of of 10 °C/min. 10 ◦C/min. Epoxy resins that exhibit high temperature resistance normally have a relatively high Epoxy resins that exhibit high temperature resistance normally have a relatively high Polymersroom-temperature 2018, 10, x FOR PEERviscosity. REVIEW The viscosity of epoxy resins can be controlled by heating during5 of 11 room-temperatureprocessing, which viscosity. enables good The fluidity viscosity to be of maintained epoxy resins and thus can improves be controlled infiltration by of heating the resin during processing,into the which carbon enables fibers. good An effective fluidity pultrusion to be maintained process requires and thus the blended improves resin infiltration to be stable of at the resin into thetemperatures carbon fibers. belowTable An 40 °C.2. effective Effect The bestof PPESK pultrusionviscosity content stabi on processlity ther levelsmal properties requires we observed of the epoxy were blended resin. for blended resin E/PPESK to be stable at resins containing 6◦ phr PPESK over a 6 h period (see Figure 3a). It was found that addition of PPESK temperatures below 40GlassC. TheTransition best viscosity Initial stability Decomposition levels we observed Maximum were for Weight blended Loss E/PPESK hadSamples little effect on the viscosity of the resin, which was also relatively stable at higher temperatures; resins containing 6 phrTemperature PPESK over (°C) a 6 h periodTemperature (see Figure (°C)3a). It was foundTemperature that addition (°C) of PPESK this is likely to be due to no steric hindrance being present. However, the gelation time of the system E/PPESK-0 225 385 420 had littlewas effect found on to theincrease viscosity significantly of the when resin, the which content was of PPESK also relatively exceeded 6% stable (Figure at higher3b), which temperatures; was E/PPESK-2 226 386 421 this is likelydetrimental to be dueto the to pultrusion no steric process hindrance and also being affected present. the curing However, behavior the of gelationthe resin. time of the system E/PPESK-4 230 387 425 was found toThe increase heat resistance significantly of pultruded when epoxy the contentresins is ofusually PPESK investigated exceeded by 6%measuring (Figure their3b), glass which was E/PPESK-6 232 385 417 transition temperatures, with higher glass transition temperatures normally affording pultruded detrimentalE/PPESK-8 to the pultrusion235 process and also affected 386 the curing behavior of the 421 resin. resins with greater thermal stabilities at higher temperatures. The glass transition temperatures of the blended resins were found to increase with increasing PPESK content (see Table 2), with TGA measurements showing that their initial thermal decomposition and maximum weight loss temperatures were essentially unchanged. The blended resins remained intact below temperatures of 385 °C, indicating that addition of PPESK did not adversely affect the overall thermal stability of the epoxy resins.

FigureFigure 3. (a) viscosity3. (a) viscosity against against time time curves curves at ambient at ambient temperature temperature and and 40 40◦ C;°C; ( b(b)) gelation gelation timetime against PPESKagainst content PPESK curve. content curve.

3.2. Mechanical Properties of E/PPESK Resin Blends Little change in the tensile strength and elongation length at breaking point was observed for the blended resins that contained 4–6% PPESK content, which reflects the good physical compatibility between these two components (see Figure 4). In contrast, the mechanical properties of the system changed significantly when the content of PPESK exceeded 6%. SEM images of E/PPESK epoxy resins containing varying PPESK content revealed obvious separation between the two phases of the resin when the PPESK content was 8% (see Figure 5). It is known that poor compatibility between the two phases of a resin used for pultrusion results in the binding forces of their composite materials being decreased dramatically. Consequently, the physical properties of pultruded CF/E/PPESK epoxy resins containing ≥8% PPESK dosage were not investigated in this study.

Polymers 2018, 10, 1067 5 of 10

The heat resistance of pultruded epoxy resins is usually investigated by measuring their glass transition temperatures, with higher glass transition temperatures normally affording pultruded resins with greater thermal stabilities at higher temperatures. The glass transition temperatures of the blended resins were found to increase with increasing PPESK content (see Table2), with TGA measurements showing that their initial thermal decomposition and maximum weight loss temperatures were essentially unchanged. The blended resins remained intact below temperatures of 385 ◦C, indicating that addition of PPESK did not adversely affect the overall thermal stability of the epoxy resins.

Table 2. Effect of PPESK content on thermal properties of epoxy resin.

Glass Transition Initial Decomposition Maximum Weight Loss Samples Temperature (◦C) Temperature (◦C) Temperature (◦C) E/PPESK-0 225 385 420 E/PPESK-2 226 386 421 E/PPESK-4 230 387 425 E/PPESK-6 232 385 417 E/PPESK-8 235 386 421

3.2. Mechanical Properties of E/PPESK Resin Blends Little change in the tensile strength and elongation length at breaking point was observed for the blended resins that contained 4–6% PPESK content, which reﬂects the good physical compatibility between these two components (see Figure4). In contrast, the mechanical properties of the system changed signiﬁcantly when the content of PPESK exceeded 6%. SEM images of E/PPESK epoxy resins containing varying PPESK content revealed obvious separation between the two phases of the resin when the PPESK content was 8% (see Figure5). It is known that poor compatibility between the two phases of a resin used for pultrusion results in the binding forces of their composite materials being decreased dramatically. Consequently, the physical properties of pultruded CF/E/PPESK epoxy resins containingPolymers≥8% 2018 PPESK, 10, x FOR dosagePEER REVIEW were not investigated in this study. 6 of 11

FigureFigure 4. The 4. effectThe effect of PPESKof PPESK content content onon tensile strength strength and and elongation elongation at break. atbreak.

(a) (b)

Figure 5. Scanning electron microscope (SEM) micrographs of the fractured surface of failed specimens for epoxies modified with (a) 6% and (b) 8% PPESK.

3.3. Physical Properties of CF/E/PPESK Epoxy Resin Composites The tensile strength and interlaminar shear strength of CF/E/PPESK-6 epoxy resin composites were found to decrease slightly (see Figure 6 and Table 3), with only a small decrease in the binding of the resin to the carbon fibers being observed after addition of PPESK. The surface morphologies of the carbon fibers and the composite are shown in Figure 7, while images of fractured composites that were produced in stress tests are shown in Figure 8. The composite fractures are filamentous, with their clusters exhibiting a divergent-like appearance. However, fractures became flaky and loosely distributed when the PPESK content was too high or too low, resulting in a non-uniform distribution throughout the resin. Modification of the resin with 6% PPESK resulted in the bending performance of the composite being significantly improved, enabling a value of 55 D to be achieved without any cracking or peeling of its surface layer occurring. This level of bending performance is highly desirable, because continuous pultruded composite materials are often incorporated into coiled materials.

Polymers 2018, 10, x FOR PEER REVIEW 6 of 11

Polymers 2018, 10, 1067 6 of 10 Figure 4. The effect of PPESK content on tensile strength and elongation at break.

(a) (b)

Figure 5.FigureScanning 5. Scanning electron electron microscope microscope (SEM) (SEM) micrographs micrographs of the of fracturedthe fractured surface surface of failedof failed specimens specimens for epoxies modified with (a) 6% and (b) 8% PPESK. for epoxies modiﬁed with (a) 6% and (b) 8% PPESK.

3.3. Physical3.3. Physical Properties Properties of CF/E/PPESK of CF/E/PPESK Epoxy Epoxy Resin Resin CompositesComposites The tensile strength and interlaminar shear strength of CF/E/PPESK-6 epoxy resin composites Thewere tensile found strength to decrease and slightly interlaminar (see Figure shear 6 and strength Table 3), ofwith CF/E/PPESK-6 only a small decrease epoxy in the resin binding composites were foundof the to resin decrease to the carbon slightly fibers (see being Figure observed6 and after Table addition3), with of PPESK. only a The small surface decrease morphologies in the of binding of the resinthe carbon to the fibers carbon and the fibers composite being are observed shown in after Figure addition 7, while images of PPESK. of fractured The surface composites morphologies that of the carbonwere produced fibers and in stress the compositetests are shown are in shown Figure in 8. FigureThe composite7, while fracture imagess are of filamentous, fractured compositeswith that weretheir produced clusters exhibiting in stress a testsdivergent-like are shown appearance. in Figure However,8. The compositefractures became fractures flaky and are loosely filamentous, distributed when the PPESK content was too high or too low, resulting in a non-uniform distribution with their clusters exhibiting a divergent-like appearance. However, fractures became flaky and loosely throughout the resin. Modification of the resin with 6% PPESK resulted in the bending performance distributedof the when composite the PPESK being significantly content was improved, too high enab orling too alow, value resulting of 55 D to in be a achieved non-uniform without distribution any throughoutcracking the or resin. peeling Modification of its surface of thelayer resin occurring. with 6%This PPESK level of resulted bending inperformance the bending is highly performance of the compositedesirable, because being significantlycontinuous pultruded improved, composite enabling materials a value are of often 55 Dincorporated to be achieved into coiled without any crackingmaterials. or peeling of its surface layer occurring. This level of bending performance is highly desirable, because continuous pultruded composite materials are often incorporated into coiled materials. Polymers 2018, 10, x FOR PEER REVIEW 7 of 11

(a) (b)

Figure 6.FigureTensile 6. Tensile strength strength (a) and(a) and interlaminar interlaminar shearshear strength strength (b ()b of) ofcarbon carbon fiber/multi-functional ﬁber/multi-functional epoxy resin (CF/E) and CF/E/PPESK-6. epoxy resin (CF/E) and CF/E/PPESK-6.

(a)Table 3. Physical properties (b) of CF/E/PPESK before and (c) after toughening.

Properties CF/E CF/E/PPESK-6 Tensile strength (MPa) 2300 ± 142 2200 ± 153 Linear expansion coefﬁcient ≤2.0 × 10−6 ≤2.0 × 10−6 Interlaminar shear strength (MPa) 82.7 ± 4.8 81.6 ± 4.2 30 KN radial pressure resistance No cracking or peeling No cracking or peeling

Bending performance, 55D (D: bar No cracking or peeling on the No cracking or peeling on the Figure 7. (a) SEM micrograph of the surface of carbon fiber (CF-12K); (b) CF/E epoxy resin diameter) surface below 55D surface at 55 D composite; and (c) CF/E/PPESK-6 epoxy resin composite. Tensile strength at 190 ◦C ≥ the Tensile strength at 190 ◦C ≥ the High temperature tensile strength (a)value of 90% at room (b) temperature value of 90% at room temperature

Figure 8. Pictures of fractured composites: (a) CF/E epoxy resin composite; and (b) CF/E/PPESK-6 epoxy resin composite.

Table 3. Physical properties of CF/E/PPESK before and after toughening.

Properties CF/E CF/E/PPESK-6 Tensile strength (MPa) 2300 ± 142 2200 ± 153 Linear expansion coefficient ≤2.0 × 10−6 ≤2.0 × 10−6 Interlaminar shear strength 82.7 ± 4.8 81.6 ± 4.2 (MPa) 30 KN radial pressure No cracking or peeling No cracking or peeling resistance Bending performance, 55D No cracking or peeling on the surface No cracking or peeling on the surface (D: bar diameter) below 55D at 55 D High temperature tensile Tensile strength at 190 °C ≥ the value Tensile strength at 190 °C ≥ the value strength of 90% at room temperature of 90% at room temperature

3.4. Mechanical Properties of CF/E/PPESK Epoxy Resin Composites Torsion strength tests revealed no obvious change in torque for the PPESK-modified resin CF/E/PPESK-6 when compared to unmodified multi-functional epoxy resin CF/E within a 10° twist range (see Figure 9). Unmodified epoxy resin was found to lengthen and fracture as the torque angle

Polymers 2018, 10, x FOR PEER REVIEW 7 of 11 Polymers 2018, 10, x FOR PEER REVIEW 7 of 11

(a) (b) (a) (b)

Polymers 2018,Figure10, 1067 6. Tensile strength (a) and interlaminar shear strength (b) of carbon fiber/multi-functional 7 of 10 Figure 6. Tensile strength (a) and interlaminar shear strength (b) of carbon fiber/multi-functional epoxy resin (CF/E) and CF/E/PPESK-6. epoxy resin (CF/E) and CF/E/PPESK-6.

(a) (b) (c) (a) (b) (c)

Figure 7. (a) SEM micrograph of the surface of carbon fiber (CF-12K); (b) CF/E epoxy resin Figure 7.Figure(a) SEM 7. (a micrograph) SEM micrograph of the surfaceof the surface of carbon of carbon ﬁber (CF-12K);fiber (CF-12K); (b) CF/E (b) CF/E epoxy epoxy resin resin composite; composite; and (c) CF/E/PPESK-6 epoxy resin composite. and (c) CF/E/PPESK-6composite; and (c) CF/E/PPESK-6 epoxy resin composite.epoxy resin composite.

(a) (b)

Figure 8. Pictures of fractured composites: (a) CF/E epoxy resin composite; and (b) CF/E/PPESK-6 Figure 8. Pictures of fractured composites: (a) CF/E epoxy resin composite; and (b) CF/E/PPESK-6 Figure 8.epoxyPictures resin composite. of fractured composites: (a) CF/E epoxy resin composite; and (b) CF/E/PPESK-6 epoxy resinepoxy composite. resin composite. Table 3. Physical properties of CF/E/PPESK before and after toughening. 3.4. Mechanical PropertiesTable of CF/E/PPESK3. Physical properties Epoxy of CF/E/PPESK Resin Composites before and after toughening. Properties CF/E CF/E/PPESK-6 Properties CF/E CF/E/PPESK-6 Tensile strength (MPa) 2300 ± 142 2200 ± 153 TorsionTensile strength strength tests(MPa) revealed no obvious 2300 ± 142 change in torque for the 2200 PPESK-modified± 153 resin Linear expansion coefficient ≤2.0 × 10−6 ≤2.0 × 10−6 ◦ CF/E/PPESK-6Linear expansion when coefficient compared to unmodified≤2.0 × 10−6 multi-functional epoxy≤2.0 resin × 10−6 CF/E within a 10 Interlaminar shear strength Interlaminar shear strength 82.7 ± 4.8 81.6 ± 4.2 twist rangePolymers (see 2018 Figure(MPa), 10, x FOR 9). PEER Unmodified REVIEW epoxy82.7 resin ± 4.8 was found to lengthen 81.6and ± fracture4.2 as8 of the 11 torque (MPa)◦ angle exceeded30 KN 10radial; however,pressure the torque applied to the PPESK-modified composite could be increased 30 KN radial pressure No cracking or peeling No cracking or peeling ◦ exceededresistance 10°; however, the torque Noapplied cracking to thore peeling PPESK-modified composite No cracking could or be peeling increased to to >80 before itresistance finally fractured. Therefore, the tensile strength of the PPESK composites was much >80°Bending before performance, it finally fractured.55D No crackingTherefore, or peeling the tensile on the strength surface ofNo the cracking PPESK or composites peeling on thewas surface much greater thanBending that performance, of multi-functional 55D No cracking epoxy or resin,peeling withon the both surface materials No cracking fracturing or peeling suddenly on the surface when their greater(D: than bar diameter) that of multi-functional epoxybelow resi 55Dn, with both materials fracturingat 55 Dsuddenly when (D: bar diameter) below 55D at 55 D torque limitstheirHigh torque weretemperature limits exceeded. weretensile exceeded. Tensile strength at 190 °C ≥ the value Tensile strength at 190 °C ≥ the value High temperature tensile Tensile strength at 190 °C ≥ the value Tensile strength at 190 °C ≥ the value strength of 90% at room temperature of 90% at room temperature strength of 90% at room temperature of 90% at room temperature

3.4. Mechanical Properties of CF/E/PPESK Epoxy Resin Composites 3.4. Mechanical Properties of CF/E/PPESK Epoxy Resin Composites Torsion strength tests revealed no obvious change in torque for the PPESK-modified resin Torsion strength tests revealed no obvious change in torque for the PPESK-modified resin CF/E/PPESK-6 when compared to unmodified multi-functional epoxy resin CF/E within a 10° twist CF/E/PPESK-6 when compared to unmodified multi-functional epoxy resin CF/E within a 10° twist range (see Figure 9). Unmodified epoxy resin was found to lengthen and fracture as the torque angle range (see Figure 9). Unmodified epoxy resin was foundPristine to lengthen and fracture as the torque angle resin

Figure 9.FigureTorsion 9. strengthTorsion strength test on pristinetest on resinpristine (CF/E) resi andn (CF/E) resin blendedand resin with blended PPESK with (CF/E/PPESK-6). PPESK (CF/E/PPESK-6). 3.5. Fatigue and Aging Properties of CF/E/PPESK Composites 3.5. Fatigue and Aging Properties of CF/E/PPESK Composites A millionA million fatigue fatigue test test cycles cycles were were conducted conducted on the CF/E/PPESK-6 CF/E/PPESK-6 composite composite using a using testing a testing frequencyfrequency of 10 Hz,of 10 aHz, pull a pull force force of of 70 70 kN, kN, and a a downward downward force force of 10 ofkN. 10 Structural kN. Structural displacement displacement of of the compositethe composite was was found found to tobe be lessless than 1% 1% after after a amillion million cycles cycles (see (seeTable Table 4), with4), no with obvious no obvious changeschanges in its waveformin its waveform frequency frequency being being detecteddetected (s (seeee Figure Figure 10). 10The). strength The strength retention retention rate of a rate of a CF/E/PPESK-6CF/E/PPESK-6 composite composite afterafter the fatigue tests tests ha hadd been been completed completed was found was found to be >90% to be of >90% its of its original value. original value. Table 4. Parameters during the fatigue test for the composite CF/E/PPESK-6.

Pull Force Downward Force Upward Displacement Downward Displacement Cycles (kN) (kN) (mm) (mm) 3183 70.034 −9.961 0.942 0.134 6410 70.025 −10.044 0.96 0.131 142,036 70.025 −10.044 0.96 0.131 513,257 69.425 −10.044 0.98 0.132 784,601 69.925 −9.844 0.94 0.139 1,025,732 70.0855 −10.144 0.97 0.142

70KN (a)70KN (b)

-10KN Sine wave curve fatigue hundred -10KN Sine wave curve fatigue 1 million

Figure 10. Fatigue sinusoids (a) before and (b) after one million cycles.

The thermal, humidity (temperature 40 °C, humidity 60%, time 1200 h) and ultraviolet (wavelength 340 nm, temperature 50 °C, time 2160 h) aging properties of the CF/E/PPESK composites were subsequently determined. These studies revealed that the microstructure of the composite was affected by the tests, with composites experiencing some aging when treated with alternating high/low temperatures, high levels of moisture, and ultraviolet light over extended

Polymers 2018, 10, x FOR PEER REVIEW 8 of 11

exceeded 10°; however, the torque applied to the PPESK-modified composite could be increased to >80° before it finally fractured. Therefore, the tensile strength of the PPESK composites was much greater than that of multi-functional epoxy resin, with both materials fracturing suddenly when their torque limits were exceeded.

Pristine resin

Figure 9. Torsion strength test on pristine resin (CF/E) and resin blended with PPESK (CF/E/PPESK-6).

3.5. Fatigue and Aging Properties of CF/E/PPESK Composites A million fatigue test cycles were conducted on the CF/E/PPESK-6 composite using a testing frequency of 10 Hz, a pull force of 70 kN, and a downward force of 10 kN. Structural displacement of Polymers 2018the ,composite10, 1067 was found to be less than 1% after a million cycles (see Table 4), with no obvious 8 of 10 changes in its waveform frequency being detected (see Figure 10). The strength retention rate of a CF/E/PPESK-6 composite after the fatigue tests had been completed was found to be >90% of its original value.Table 4. Parameters during the fatigue test for the composite CF/E/PPESK-6.

Pull ForceTable 4. ParametersDownward during Force the fatigueUpward test for Displacementthe composite CF/E/PPESK-6.Downward Displacement Cycles (kN) (kN) (mm) (mm) Pull Force Downward Force Upward Displacement Downward Displacement Cycles 3183 70.034(kN) −9.961(kN) (mm) 0.942 (mm) 0.134 64103183 70.02570.034 −10.044−9.961 0.942 0.96 0.134 0.131 142,0366410 70.02570.025 −10.044−10.044 0.96 0.96 0.131 0.131 513,257142,036 69.42570.025 −10.044−10.044 0.96 0.98 0.131 0.132 784,601513,257 69.92569.425 −9.844−10.044 0.98 0.94 0.132 0.139 1,025,732784,601 70.085569.925 −10.144−9.844 0.94 0.97 0.139 0.142 1,025,732 70.0855 −10.144 0.97 0.142

70KN (a)70KN (b)

-10KN Sine wave curve fatigue hundred -10KN Sine wave curve fatigue 1 million

FigureFigure 10. Fatigue 10. Fatigue sinusoids sinusoids ( a(a)) beforebefore and and (b ()b after) after one one million million cycles. cycles.

The thermal,The thermal, humidity humidity (temperature (temperature 4040 ◦°C,C, hu humiditymidity 60%, 60%, time time 1200 1200h) and h) ultraviolet and ultraviolet (wavelength 340 nm, temperature 50 °C, time 2160 h) aging properties of the CF/E/PPESK (wavelength 340 nm, temperature 50 ◦C, time 2160 h) aging properties of the CF/E/PPESK composites composites were subsequently determined. These studies revealed that the microstructure of the were subsequentlycomposite was determined. affected by the Thesetests, with studies composites revealed experiencing that the some microstructure aging when treated of the with composite was affectedalternating by the high/low tests, withtemperatures, composites high experiencinglevels of moisture, some and aging ultraviolet when light treated overwith extended alternating high/low temperatures, high levels of moisture, and ultraviolet light over extended periods of time Polymers 2018, 10, x FOR PEER REVIEW 9 of 11 (see Figure 11). However, no damage leading to excessive cracking or oxidative corrosion (or other chemicalperiods reactions) of time was (see observed, Figure 11). with However, static tensionno damage tests leading revealing to excessive that the cracking composites or oxidative still retained 90% of theircorrosion mechanical (or other strength. chemical Thisreactions) indicates was observ that theseed, with composite static tension materials tests revealing exhibit good that the anti-aging and highcomposites corrosion still resistance retained 90% properties of their thatmechanical should strength. make themThis indicates suitable that for these challenging composite polymer materials exhibit good anti-aging and high corrosion resistance properties that should make them applicationssuitable in for relatively challenging harsh polymer environments. applications in relatively harsh environments.

(a) (b)

Figure 11.FigureMicrostructure 11. Microstructure of the sampleof the sample CF/E/PPESK CF/E/PPESK (a) before (a) before and (band) after (b) 50after days 50 of days hydrothermal of aging, ashydrothermal well as (c) beforeaging, as and well ( das) ( afterc) before 2160 and h ( ofd) ultravioletafter 2160 h of aging. ultraviolet aging.

4. Conclusions4. Conclusions This studyThis study has shownhas shown that that poly(phthalazinone poly(phthalazinone ether ether sulfone sulfone ketone) ketone) (PPESK) (PPESK) can be used can to be used improve the heat resistance of multi-functional epoxy resins, increasing their glass transition to improve the heat resistance of multi-functional epoxy resins, increasing their glass transition temperature by 5 °C, which enables them to be used as resins for pultrusion. Good bonding was observed between the two components in E/PPESK epoxy resins containing 4–6% PPESK, which meant the resins’ tensile strengths and elongation lengths at breaking point were essentially unchanged. The mixed viscosity stabilities of these modified resins proved well suited to both the pultrusion and molding processes. The bending performance of composite bars prepared from the modified epoxy resins was improved significantly, with no cracking or peeling being observed in their surface layer for a bending radius of 55 D. Fatigue tests (1 million cycles) revealed that the CF/E/PPESK epoxy resin could maintain its shape effectively through multiple expansions and contractions, while aging tests (heat, temperature, humidity, and ultraviolet) resulted in an aged sample that retained 90% of its original strength.

Funding: This research received no external funding.

Acknowledgments: We acknowledge financial support from the National Nature Science Foundation of China (No. 51673033, U1663226 and 51473025), and from the Fundamental Research Funds for the Central Universities (DUT17LK39).

Conflicts of Interest: The authors declare no conflicts of interest.

References

Polymers 2018, 10, 1067 9 of 10 temperature by 5 ◦C, which enables them to be used as resins for pultrusion. Good bonding was observed between the two components in E/PPESK epoxy resins containing 4–6% PPESK, which meant the resins’ tensile strengths and elongation lengths at breaking point were essentially unchanged. The mixed viscosity stabilities of these modified resins proved well suited to both the pultrusion and molding processes. The bending performance of composite bars prepared from the modified epoxy resins was improved significantly, with no cracking or peeling being observed in their surface layer for a bending radius of 55 D. Fatigue tests (1 million cycles) revealed that the CF/E/PPESK epoxy resin could maintain its shape effectively through multiple expansions and contractions, while aging tests (heat, temperature, humidity, and ultraviolet) resulted in an aged sample that retained 90% of its original strength.

Author Contributions: L.W., J.W., Z.W., and X.J. conceived and designed the experiments; L.W., F.Z., and J.W. performed the experiments and analyzed the data; L.W., Y.Q., and Z.W. contributed reagents, materials, and analysis tools; L.W. and Z.W. wrote the paper. Funding: This research received no external funding. Acknowledgments: We acknowledge financial support from the National Nature Science Foundation of China (No. 51673033, U1663226 and 51473025), and from the Fundamental Research Funds for the Central Universities (DUT17LK39). Conflicts of Interest: The authors declare no conflicts of interest.

References

1. Park, S.Y.; Choi, W.J.; Choi, C.H.; Choi, H.S. The effect of curing temperature on thermal, physical and mechanical characteristics of two types of adhesives for aerospace structures. J. Adhes. Sci. Technol. 2018, 32, 1200–1223. [CrossRef] 2. Xu, Y.J.; Liao, G.X.; Gu, T.S.; Zheng, L.; Jian, X.G. Mechanical and morphological properties of epoxy resins modified by poly(phthalazinone ether sulfone ketone). J. Appl. Polym. Sci. 2008, 110, 2253–2260. [CrossRef] 3. Auvergne, R.; Caillol, S.; David, G.; Boutevin, B.; Pascault, J.P. Biobased thermosetting epoxy: Present and future. Chem. Rev. 2014, 114, 1082–1115. [CrossRef][PubMed] 4. Kumar, S.; Krishnan, S.; Samal, S.K.; Mohanty, S.; Nayak, S.K. Toughening of petroleum based (DGEBA) epoxy resins with various renewable resources based flexible chains for high performance application: A review. Ind. Eng. Chem. Res. 2018, 57, 2711–2726. [CrossRef] 5. Thunga, M.; Akinc, M.; Kessler, M.R. Tailoring the toughness and CTE of high temperature bisphenol E cyanate ester (BECy) resin. Express Polym. Lett. 2014, 8, 336–344. [CrossRef] 6. Marouf, B.T.; Mai, Y.W.; Bagheri, R.; Pearson, R.A. Toughening of epoxy nanocomposites: Nano and hybrid effects. Polym. Rev. 2016, 56, 70–112. [CrossRef] 7. Wu, S.Y.; Guo, Q.P.; Kraska, M.; Stuhn, B.; Mai, Y.W. Toughening epoxy thermosets with block ionomers: The role of phase domain size. Macromolecules. 2013, 46, 8190–8202. [CrossRef] 8. Fernandez, B.; Arbelaiz, A.; Diaz, E.; Mondragon, I. Influence of polyethersulfone modification of a tetrafunctional epoxy matrix on the fracture behavior of composite laminates based on woven carbon fibers. Polym. Compos. 2004, 25, 480–488. [CrossRef] 9. Chen, C.H.; Chen, P.H. Hybrid fibre reinforced epoxy composites for pultrusion: Mechanical and thermal properties. Polym. Polym. Compos. 2011, 19, 459–468. [CrossRef] 10. Li, G.; Huang, Z.B.; Li, P.; Xin, C.L.; Jia, X.L.; Wang, B.H.; He, Y.D.; Ryu, S.; Yang, X.P. Curing kinetics and mechanism of polysulfone nanofibrous membranes toughened epoxy/amine systems using isothermal DSC and NIR. Thermochim. Acta 2010, 497, 27–34. [CrossRef] 11. Zhang, J.; Guo, Q.P.; Fox, B.L. Study on thermoplastic-modified multifunctional epoxies: Influence of heating rate on cure behavior and phase separation. Compos. Sci. Technol. 2009, 68, 1172–1179. [CrossRef] 12. Giannotti, M.I.; Bernal, C.R.; Oyanguren, P.A.; Galante, M.J. Morphology and fracture properties relationship of epoxy-diamine systems simultaneously modified with polysulfone and poly(ether imide). Polym. Eng. Sci. 2005, 45, 1312–1318. [CrossRef] Polymers 2018, 10, 1067 10 of 10

13. Hourston, D.J.; Lane, J.M.; Zhang, H.X. Toughening of epoxy resins with thermoplastics: 3. An investigation into the effects of composition on the properties of epoxy resin blends. Polym. Int. 1997, 42, 349–355. [CrossRef] 14. Chen, H.M.; Lv,R.G.; Liu, P.; Wang, H.Y.; Huang, Z.Y.; Huang, T.; Li, T.S. An investigation of cure and thermal stability of poly(amide-amidic acid) modified tetraglycidyl 4,40-diaminodiphenylmethane/4,40-diaminodiphenylsulfone. J. Appl. Polym. Sci. 2013, 128, 1592–1600. [CrossRef] 15. Zhong, Z.K.; Zheng, S.X.; Huang, J.Y.; Cheng, X.G.; Guo, Q.P.; Wei, J. Phase behavior and mechanical properties of epoxy resin containing phenolphthalein poly(ether ether ketone). Polymer 1998, 39, 1075–1080. [CrossRef] 16. Luo, Y.; Zhang, M.; Dang, G.D.; Li, Y.; An, X.F.; Chen, C.H.; Yi, X.S. Toughening of epoxy resin by poly(ether ether ketone) with pendant fluorocarbon groups. J. Appl. Polym. Sci. 2011, 122, 1758–1765. [CrossRef] 17. Banea, M.D.; da Silva, L.F.M.; Campilho, R.D.S.G. Effect of temperature on the shear strength of aluminium single lap bonded joints for high temperature applications. J. Adhes. Sci. Technol. 2012, 28, 1367–1381. [CrossRef] 18. Liu, R.; Wang, J.Y.; Li, J.L.; Jian, X.G. An investigation of epoxy/thermoplastic blends based on addition of a novel coply(aryl ether nitrile) containing phthalazinone and biphenyl moieties. Polym. Int. 2015, 64, 1786–1793. [CrossRef] 19. Xu, Y.J.; Fu, X.J.; Liao, G.X.; Zhou, H.X.; Jian, X.G. Preparation, morphology and thermo-mechanical properties of epoxy resins modified by co-poly(phthalazinone ether sulfone). High Perform. Polym. 2011, 23, 248–254. [CrossRef] 20. Liu, R.; Wang, J.Y.; He, Q.Z.; Zong, L.S.; Jian, X.G. Interaction and properties of epoxy-amine system modified with poly(phthalazinone ether nitrile ketone). J. Appl. Polym. Sci. 2016, 133, 42938. [CrossRef] 21. Wang, L.W.; Wang, J.W.; Qi, Y.; Zhang, F.F.; Weng, Z.H.; Jian, X.G. Preparation of novel epoxy resin bearing phthalazinone moiety and their application as high-temperature adhesives. Polymers 2018, 10, 708. [CrossRef] 22. Wang, J.Y.; Wang, M.J.; Liu, C.; Zhou, H.X.; Jian, X.G. Synthesis of poly(arylene ether nitrile ketone)s bearing phthalazinone moiety and their properties. Polym. Bull. 2013, 70, 1467–1481. [CrossRef] 23. Song, Y.; Wang, J.Y.; Li, G.H.; Sun, Q.M.; Jian, X.G.; Teng, J.; Zhang, H.B. Synthesis, characterization and optical properties of cross-linkable poly(phthalazinone ether ketone sulfone). Polymer 2008, 49, 724–731. [CrossRef]

© 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).