effect of moisture on the ultrasonic welding of carbon ... · supplement to the welding journal,...

Introduction The request for affordable, light-weight, and high-performance engi-neering structures has promoted thedevelopment of fiber-reinforced poly-mer composites (Refs. 1, 2). In particu-lar, carbon-reinforced polyamide 66(Cf/PA66) is being increasingly used inautomotive and aerospace applicationsdue to its recyclability, thermal stabili-ty, and good mechanical properties(Refs. 3–5). To obtain complex Cf/PA66components, which often consist ofsmaller parts, joining techniques suchas mechanical fastening, adhesivebonding, and fusion bonding are usual-ly used (Refs. 6, 7). Ultrasonic welding(UW), which belongs to the fusion cate-

gory, has been used in industrial massproduction for many years because ofits efficiency, reliability, and good cos-metic quality (Refs. 8, 9). There are many influential factorsin the UW process, such as welding pa-rameters, material dimensions, andmechanical properties (Refs. 10–12).Mechanical properties of the materialbecomes the determining factor whenthe welding parameters and materialdimensions are optimal. However, themechanical properties of Cf/PA66composite can be easily affected by hy-grothermal exposure (Refs. 13–15). Ithas been reported in the literaturethat the imbibed moisture into materi-al may bring reversible and/or irre-versible changes in specific properties,

such as hydrolysis, plasticization, mi-crocracking, and even the glass transi-tion temperature (Refs. 16–18). Thesechanges would alter the moduli of thematerial, which is of great importanceto its weldability. Chaichanawong (Ref. 19) discov-ered that the microstructure of thefracture surface changed to ductilefracture from brittle fracture, and theelastic modulus decreased after im-mersion in water. Timmaraju (Ref. 20)revealed the significant plasticizationeffect of water on the moduli anddamping factor with an increase in im-bibed moisture. Sateesh (Ref. 21) dis-covered the flexural modulus of a wa-ter-bathed specimen was significantlydecreased due to the degradation ofthe glass-fiber-reinforced plastics.Typically, it has been established thatthe weld quality of the joint is relatedto the moduli (i.e., storage and lossmodulus) of the material (Refs. 10,22). The loss modulus of the materialis proportional to the viscoelasticheating (Ref. 23). As a result, the vari-ations in the material moduli couldproduce an influential effect on heatgeneration during the UW process,which would produce a profound effecton the weld area of the joint and jointstrength, accordingly. For normal storage of Cf/PA66 com-posite, the humidity and temperaturelevels undergo variations in differentseasons, weather conditions, and places.To store the material in a fixed humidityand temperature state the whole timewould definitely increase the financialburden. Based on this point of view, thispaper reports experimental findings ofan investigation regarding the effect of

WELDING RESEARCH

JUNE 2017 / WELDING JOURNAL 185-s

SUPPLEMENT TO THE WELDING JOURNAL, JUNE 2017Sponsored by the American Welding Society and the Welding Research Council

Effect of Moisture on the Ultrasonic Welding ofCarbonFiberReinforced Polyamide 66 Composite

Amplitude was proposed as an indication in monitoring weld quality

BY Q. ZHI, X. R. TAN, AND Z. X. LIU

ABSTRACT The influence of moisture absorption on the weldability of carbonfiberreinforcedPolyamide 66 (Cf/PA66) was investigated via ultrasonic welding in a lapjoint configuration. Results showed the joint strength decreased significantly when moisture absorptionwas up to 1 wt% and decreased slightly with low moisture. The water molecules decreased the hydrogen bridge bond in polymer chains, which not only plasticized the composite but also changed the loss modulus of the composite. The fracture surface observations showed the increase in moisture resulted in a decreased weld area, increasedvoids, and severe deformation around the voids, especially for high moisture content (> 1wt%). As a result, the joint with high moisture exhibited interfacial fracture. This deleterious effect of moisture on the weldability of Cf/PA66 can be reversed by completelyredrying before welding. The amplitude during welding decreased with increasing moisture and displayed a similar varying tendency as that of tensile strength and weld area ofjoint. The joint strength increased linearly with the amplitude and was influenced slightlyby moisture when the amplitude was larger than 100 m during welding. Therefore, theamplitude was proposed as an indication in monitoring weld quality.

KEYWORDS • CarbonFiberReinforced Polyamide 66 • Ultrasonic Welding • Moisture • Amplitude

Zhi Supplement June 2017.qxp_Layout 1 5/10/17 3:51 PM 85

moisture uptake on the weldability ofultrasonic welded Cf/PA66 composite.The changes in the material propertiesand the microstructure of the weld jointare also examined. To determine thecritical water content, which has a slightinfluence on weld strength, an indica-tion of amplitude during welding is proposed.

Experimental Procedure

Materials

Commercial pure PA66 pellets and 2-mm-long carbon fiber (24K, T300 type,Toray Carbon Magic Co. Ltd.) were driedin an oven for 8 h at 80°C. The driedpellets and carbon fiber were injectionmolded into the 30 wt-% carbon-fiber-reinforced PA66 plaques with dimen-sions of 138 × 38 × 2.3 mm.

Moisture Absorption

The specimens were conditioned ina climatic chamber, where both tem-perature and relative humidity can bemonitored, to absorb different levelsof water in the specimens. These con-ditioned specimens were divided intoseven groups that were marked as W0,W1, W2, W3, W4, W5, and W6, respec-tively. The moisture absorption equi-librium in air is much less than in wa-ter (Ref. 20), thus this study revealedthe workpiece with moisture contentup to 2.6 wt-%. The specimens in the W0 group

were dried in a vacuum oven at 70°Cfor 48 h to remove surface-absorbedvapor, and W1 specimens were storedin the normal environment for 30days. Specimens for W2 to W5 wereconditioned in various humid environ-ments to obtain a certain humidity de-gree. W6 specimens were the redriedW4 specimens in the vacuum oven at70°C for 96 h to desorb water com-pletely. Since conditioning time mayaffect the crystallization of the car-bon-fiber-reinforced polyamide, allspecimens were conditioned for thesame duration of 48 h. The specimenswere weighed before and after themoisture absorption to ensure thatwater content was consistent with theconditioning. The percentage of themoisture uptake, M, was calculated foreach measurement as follows:

where mt is the mass of the sample af-ter being conditioned and m0 is the

mass of the sample in the dry state.The resulting level of humidity in de-pendency of the conditioning proce-dure is listed in Table 1.

Ultrasonic Welding

Ultrasonic welding was performedusing a KZH-2026 multifunction UWmachine with a nominal power of 2.6kW and a nominal frequency of 20 kHz.The welding setup used in this study isschematically shown in Fig. 1. Thepiezoelectric converter converts theelectrical signal into mechanical vibra-tions. To transfer the ultrasonic wavesto the workpiece, the transducer wasconnected to the horn that was placedat right angles in contact with the work-pieces to be welded. The support frameof the transducer-booster-horn systemwas attached to a pneumatic piston thatprovided vertical movement along withthe static force (i.e., weld pressure) ap-plied through the horn to the work-pieces. The machine was also equippedwith data acquisition systems that com-

M %( )= mt –m0m0

100 1( )

WELDING RESEARCH

WELDING JOURNAL / JUNE 2017, VOL. 96186-s

Fig. 1 — Schematic for ultrasonic welding of injectionmolded carbonfiber/polyamide 66 composites (dimensions in mm).

Fig. 2 — Schematics of the standard dogbone and single lapshear specimens (dimensions in mm).

Table 1 — Moisture Absorptions for Specimens under Different Conditioning Procedures

Conditioning Group Temperature Relative Humidity Moisture Content (°C) (%) (wt%)

W0 70 — 0 W1 30 50 0.4 W2 50 60 0.9 W3 50 80 1.7 W4 50 100 2.6 W5 70 60 2.4 W6 70 — 0

A

B

Zhi Supplement June 2017.qxp_Layout 1 5/10/17 3:51 PM Page 186

bined with pressure sensor, displace-ment sensor, and timer were integratedin the controller of the ultrasonic weld-ing machine. The output amplitude of the weldingmachine was set to 25 m. The ampli-tude of the horn was also 25 m whenvibrated in air. The amplitude curveduring welding was collected from thehorn displacement. When the ultrason-ic vibration was conducted to theCf/PA66 composite, the high-frequencyvibration resulted in a large amplitudein the Cf/PA66 composite due to thehysteresis effect of the composite andapplying it to the horn. Therefore, themeasured amplitude was the combina-tion of the horn and composite surfaceamplitudes. The horn amplitude wasrelatively small; thus, the measured am-plitude can be regarded as the ampli-tude of the composite surface duringwelding (hereinafter referred to as am-plitude). The welding horn used in thisstudy was made of enhanced 7075 alu-minum alloy with a size of 18 mm in di-ameter and a total length of 134 mm aswell as a goose length of 54 mm asshown in Fig. 1.

Quasistatic Test

Quasi-static tests were performedby loading each specimen to failure inan MTS 810 tensile tester according toASTM D1002-2001. For the quasi-static tensile specimens of the com-posite treated with various moisture(nonwelded specimen), standard dog-

bone specimens with a gauge of 25mm in length, 6 mm in width, and 2.3mm in thickness based on ASTM D638were used, as shown in Fig. 2A. For thewelded joints, single-lap specimenswith the dimensions of 138 × 38 × 2.3mm with an overlap of 25 mm wereused, as shown in Fig. 2B. To minimize the bending stressesinherent in the testing of single-lapwelded specimens, filler plates were at-tached onto both ends of the specimenusing masking tape to accommodatethe sample offset. Load vs. displace-ment results were obtained, as thespecimens were loaded at a stroke rateof 2.0 mm/min. Joint strengths of thewelded workpieces were evaluated bypeak load. Three replicates were per-formed, and the average jointstrengths were reported.

Characterization of the Material

The morphologies of the weld be-fore/after tensile testing were charac-terized by scanning electron mi-croscopy (SEM, JSM 6700F). All thesamples were sputter-coated with plat-inum for 50 s before SEM analysis toinduce conductivity. The crystal struc-tures of the dry and wet specimenswere identified by x-ray diffraction(XRD, X’Pert PROX) with a copper tar-get. The scanning speed was 10deg/min and scanning angle (2)was from 15 to 35 deg for all thespecimens. The dynamic mechanical analysis

was carried out to investigate the ef-fects of temperature and frequency onviscoelastic properties of the carbon-fiber-reinforced polyamide 66 compos-ite. Specimens with dimensions of 38× 8.5 × 2.3 mm were subjected tothree-point bending with a spanlength of 20 mm. An oscillating forcewas applied (maximum 4 N) to giveconstant amplitude of deflection of 30m. Measurements were conductedover the temperature range of 23° to200°C with a heating rate of 2°C /minand under fixed frequencies (1, 2, 5,10 Hz). The moduli at 20 kHz were ex-trapolated by the time temperaturesuperposition.

ResultsEffect of Moisture Absorptionon Weld Quality

To investigate the influence of hu-midity levels with various tempera-tures on the weldability of carbon-fiber-reinforced polyamide 66, all theconditioned specimens were ultrasonicwelded with the same welding parame-ters of 3000 J weld energy under aweld pressure of 0.17 MPa. The resultsare listed in Table 2. As shown, thepeak load of the welded joints wereslightly affected by the humidity whenthe humidity level was low (i.e., 1 wt-%), whereas the peak load started todecrease significantly when the mois-ture content was above 1.7 wt-%. Thepeak load of the joint made with the

WELDING RESEARCH


Fig. 3 — Static failure modes of the ultrasonic welded carbonfiber/polyamide composite with 30 wt% fiber (dry and wet): A —Workpiece breakage; B — interfacial failure.

Fig. 4 — Effect of water absorption on the tensile stressstrain ofcarbonfiberreinforced polyamide 66.

A

B


redried sample (i.e., W6) was similar tothat of the dried sample (i.e., W0).Combined with the results listed inTable 1, it can be found that the ten-sile property of the joint varied withthe moisture content of the specimen,while the temperature had a slight in-fluence on the peak load. Comparingthe tensile properties of W0 with W4,the peak load decreased by approxi-mately 40% when the moisture con-tent was up to 2.6 wt-%. These charac-teristics confirmed that the polyamidecomposite was sensitive to moisture(Refs. 15, 17). It is noted that the weld area of thejoint is closely related to the jointstrength (Refs. 24, 25). In this case,the weld area of the joint decreasedwith the increasing humidity, but toan obvious lesser degree when com-pared to that of the peak load. Thisodd phenomenon may be related tothe mechanical property of the materi-al and the microstructure of the fayinginterface, which will be discussed later.

Failure Mode Analysis

The effect of humidity level on fail-ure mode of the joint was also exam-ined and initial observations showedthere were two main failure modes forthe welded joints. Figure 3 shows thetypical fracture modes of workpiecebreakage (WB) and interfacial failure(IF) of the joint. From the experimen-tal examinations, the workpiece break-age was likely for the joints made withthe workpieces with moisture absorp-tion below 1 wt-% (i.e., W0, W1, W2)whereas the interfacial fracture was

dominant for the workpieces withhigher moisture content. This differ-ence in failure mode was likely relatedto the mechanical properties of thepolyamide 66 composite and the weldjoint microstructure. Figure 4 presents the tensile curvesof the composite with various watercontents. Referring to Fig. 4, tensilestrength, elastic modulus, and strain atfailure of the composites were influ-enced by the amount of absorbed water.The tensile strength and the elasticmodulus decreased while the elongationat break of composite increased withfurther increase in water absorption.This change in the mechanical propertyof carbon-fiber-reinforced polyamide 66was mainly because of its hydrophilicnature. When the polyamide compositeencounters water, the polar nature ofthe water molecules cause them to beattracted to the amide groups and de-crease the hydrogen bridge bond inpolymer chains as shown in Fig. 5 (Refs.19, 26). To analyze the effect of waterabsorption on the microstructure ofpolyamide 66 composite, the dry and

wet Cf/PA66 composites were analyzedusing XRD technology. The XRD pat-terns of the dry and wet specimens areshown in Fig. 6. There are two peaks (2 = 20.4,23.7), 1 and 2, corresponding to(110) and (010)/(110) crystal faces, re-spectively, in polyamide 66. The inten-sities of the peaks decreased with theincreasing water absorption, indicat-ing the crystalline regions reduced andthen confirmed that the polar attrac-tion between the amide groups inpolymer chains became weaker. There-fore, the stiffness of the polyamidecomposite diminished and the ductili-ty increased, which was also observedin other research (Refs. 13, 20). As canalso be seen in Fig. 4, the stress-straincurve of the redried sample (W6) wassimilar to that of the as dry as molded(W0), which indicated the water des-orption somewhat recovered the stiff-ness, elastic modulus, and strength ofthe Cf/PA66 composite. Thus, the peakload of the joint made with the redriedsample was close to that of the dryjoint as shown in Table 2.

WELDING RESEARCH


Fig. 5 — The interaction between water and polyamide. Fig. 6 — The XRD patterns of the dry and wet Cf/PA66 composite.

Table 2 — Peak Load, Weld Area, and Failure Mode for Joint Made with Specimens underDifferent Conditioning Procedures

Conditioning Group Peak Load Weld Area Failure Mode (kN) (mm2)

W0 6.3 424 WB W1 6.2 416 WB W2 5.9 412 WB W3 5.2 377 IF W4 3.8 340 IF W5 4.1 354 IF W6 6.2 421 WB


Because the weld joint is the mostvaluable part to investigate, the mi-crostructures of the sectional surfaceand the workpiece-to-workpiece inter-face of the joint made with dry andwet specimens were examined and theresults presented in Fig. 7. The mi-crostructure of the weld joint was dif-ferent and not as compact as the ma-trix shown in Fig. 7A. There existedsome random voids at the fracturedfaying surface of the joint made withdry workpieces (i.e., W0), which waslikely formed during manufacture ordue to ultrasonic cavitation (Ref. 27). With the increase in moisture con-tent, the voids increased in quantityand the polyamide 66 resin deformedaround them as presented in speci-mens W2, W3, and W4. For W4, wherethe specimens were conditioned under100% relative humidity (RH) at 50°C,the plastic deformation of thepolyamide 66 resin around the voids

was more severe, which could have re-sulted from the extensive mobility ofthe polymer chains caused by moisture(Ref. 17). The large scale of plastic de-formations that occurred at the frac-ture surface of W4 revealed the frac-ture mode shifted from brittle fractureto ductile fracture, implying the plasti-cization effect of imbibed moisture onthe composite (Refs. 20, 26). For the joint made with redriedspecimens (i.e., W6), the appearanceof the fracture surface was similar tothat of the dried specimen (i.e., W0).This characteristic showed the recov-erability of the material microstruc-ture by desorption of water with com-plete drying, which was consistentwith the tensile results. Based on the aforementioned re-sults, the specimens with low humiditycontent (e.g., W1, W2) exhibited work-piece breakage, suggesting that the weldjoint could still bear load. Thus, the

slight decrease in peak load of the jointswas mainly because of the decrease inthe ultimate tensile strength of thecomposite. When the water absorptionincreased to a certain level (i.e., above1.7 wt-%), the loading capacity of theweld joint decreased due to the in-creased voids. Moisture absorption not only de-creased the mechanical properties ofthe composite but also deteriorated theweld microstructure. As a result, thejoints made with the specimens thatimbibed a relatively small amount ofmoisture displayed workpiece breakagewhile the joints with large moisturecontent fractured from the deteriorat-ed weld joint during the tensile test.On the other hand, though the weldarea of the wet joint had a smaller de-crease in comparison to peak load withthe increased water absorption, the mi-crostructure at the faying interface be-came loose and the mechanical proper-ties decreased. Therefore, the changein the weld area was not consistentwith that of peak load.

Relationship between WaterContent and Peak Load Since water absorption of Cf/PA66greatly influences the weld quality, itis essential to establish the relation-ship between water absorption andweld quality. In this section, a Voigt-Kevin model is proposed to thoroughlyunderstand the ultrasonic weldingprocess. The relationship between theamplitude during welding and watercontent is found and a threshold levelfor water absorption is established.

Modeling the Ultrasonic System

Although a certain amount of mois-ture absorption has a significant influ-ence on the weldability of carbon-fiber-reinforced polyamide 66, it is dif-ficult and complex to measure the wa-ter content in Cf/PA66 composite dur-ing normal storage. In this study, a re-lationship between the workpiecestrain during welding and water con-tent was found. The welding system ofCf/PA66 composite in this investiga-tion can be simulated by a Voigt-Kevinmodel (Ref. 22) as shown in Fig. 8. Thewelding machine and workpieces arerepresented in terms of springs anddampers. The equations of motion forthe welding system were

WELDING RESEARCH


Fig. 7 — SEM images of the sectional fracture of the joint with 2.6 wt% moisture content (W4): A — Fracture surfaces for the welded joints after tensile tests; B — W0; C —W2; D — W3; E — W4; F — W6.


where m0 is the mass of the horn; k1, k2

are the spring constants; and c1, c2 arethe dampings of the materials. x1, x2

are the displacements of the weldingmachine/workpiece interface duringwelding. x1, x2, and x1 are the first andsecond derivative of the displacement.F is the force that was applied to thehorn, consisting of static normal andoscillating holding forces. The upper

and lower workpieces are the samematerial, the value of k1, c1 equals thatof k2, c2; therefore, x1 = 2x2. ThenEquation 2 can be rewritten as

are the stresses of thespring and damper. The input dis-placement for the welding machine tothe upper workpiece is sinusoidal,

For ultrasonic welding, = 2f, thevibration frequency was large, which in-dicated that x1 was much larger than x1.Moreover, the difference between stor-age and loss moduli (i.e., k1, c1) was lessthan two orders of magnitude (Ref. 2).Thus, the damping property of the ma-terial was the key factor in ultrasonicwelding of thermoplastics. The damp-ing part determined the measured am-plitude and the larger amplitude indi-cated greater heating in the faying inter-face, thus the weld area was larger. Inconclusion, the heat generation at thefaying interface mainly depended on theloss modulus of the material.

m0x1 + x1k1 + x1c1

x2k1 x2c1 = F 2( )

k1 + k2( )x2 + c1 + c2( )x2

=x1k1 = x1c1 = 0 3( ) x1k1 + x1c1 = 2( F=m0 x1 ) 4( )

x1k1 , x1c1

x1 = Asin t 5( )

x1= A cos t 6( )

WELDING RESEARCH


Fig. 8 — VoigtKelvin model for the welding system (Ref.22).

Fig. 9 — The temperature dependence of loss modulus for Cf/PA66 composite at 20 kHz.

Fig. 10 — Effect of moisture absorption on the amplitude during welding.

Fig. 11 — Effect of moisture content on thew amplitude and weld area,peak load of the joint.

˙

˙ ˙ ˙̇


Effect of Moisture on the Damping Property

The damping property ofpolyamide 66 composite can be easilyinfluenced with the presence of water,and the loss modulus was the mostcommon index for damping. Figure 9presents the loss modulus for dry andwet carbon-fiber-reinforced specimensat 20 kHz. The loss modulus of theCf/PA66 composite at different tem-peratures at 20 kHz was obtained byusing the time-temperature superposi-tion (Ref. 2). The shapes of the lossmodulus curves with various watercontents were similar but shifted to alower temperature, which resulted in alarge difference of loss modulus atroom temperature (25°C). The shift inmaximum loss modulus with moistureabsorption was attributed to the hy-drophilic nature of polyamide 66. Thechains were frozen in fixed positionsin dry polyamide 66. The water ab-sorption from a humid environmentintroduced water molecules into theamide groups. Hence, the mobility ofthe polyamide chains increased, whichlowered the glass transition tempera-ture (Tg was the corresponding tem-perature at the maximum loss modu-lus (Ref. 28)) and the peak loss modu-lus toward the left. Because the welding time of the ul-trasonic process was short (1300 ms inthis study), and the temperature of thematerial rose quickly, which was tooshort for the state transformation dur-ing welding, a reference temperature of

25°C was chosen for the damping prop-erty of the material (Refs. 2, 16) be-cause of the hysteresis phenomenon.Referring to Fig. 9, the loss modulus at20 kHz in room temperature increasedwith water absorption, which may affectthe amplitude during welding.

Effect of Moisture on the Amplitudeduring Welding

To analyze the effect of water ab-sorption on the amplitude duringwelding, the amplitude with variousmoisture contents during ultrasonicwelding was measured and the resultsare shown in Fig. 10. The amplitudedecreased slowly and gradually to astable state. The decrease in the ampli-tude for approximately the first 500ms was probably because it was an un-steady phase, in which the small asper-ities at the surface were planished toobtain intimate contact between parts(Ref. 29). Afterward, the amplitudewas in a stable state. In addition, theamplitude curves of the wet work-pieces (the stable state) were lowerthan those of the dried workpieces(i.e., W0, W6). With the increasingmoisture content, the amplitude de-creased, which displayed the oppositetrend to that of the loss modulus.

Relation between the Peak Load andthe Amplitude

It is known that the amplitude de-termines the heat generation at thefaying interface of the welded work-

pieces (Refs. 25, 30). With water ab-sorption, the water molecules inCf/polyamide 66 composite influencethe loss modulus of the material whilethe loss modulus affects the ampli-tude. Hence, there may exist a correla-tion among the water content, ampli-tude, and the peak load weld area ofthe joint. These correlations are plot-ted as a function of moisture contentand are shown in Fig. 11. As shown,the amplitude exhibited the similarvarying tendency as that of the peakload and weld area of joints with in-creasing moisture content. Therefore,there would exist a relation betweenthe amplitude and peak load. Becausethe water content was difficult tomeasure, amplitude during weldingmay be an indication in controlling theweld quality. Figure 12 presents the connectionbetween the amplitude and peak loadof the joint. The red solid line in Fig.12 is the curve fitted to the data and ismeant to demonstrate the relation be-tween them. The correlation coeffi-cient (R2) is 0.98, which indicates thefitting is reliable. The relation betweenthe amplitude and peak load can be ex-pressed as

P = 0.07A – 0.94 (7)

where P is the peak load of the jointand A is the measured amplitude dur-ing ultrasonic welding. As seen, thepeak load increased linearly with theamplitude. Examining the results in Fig. 11,there existed a threshold for the ab-sorbed water content, 1 wt-% (based onpeak load), which corresponded to theamplitude of 100 m in Fig. 12. As longas the amplitude was above 100 m, theweld quality of the joint was desirable.However, when the amplitude wassmaller than the critical level (i.e., work-piece with high moisture content), theworkpiece needed to be desorbed withcomplete drying before welding.

Conclusions 1. The joint strength of the UWjoint decreased significantly whenmoisture absorption was up to 1wt-% and decreased slightly with lowmoisture. 2. The water molecules decreasedthe hydrogen bridge bond in polymer

WELDING RESEARCH


Fig. 12 — Relation between peak load and amplitude during welding.


chains, which not only plasticized thecomposite but also changed the lossmodulus of the composite. 3. Fracture surface observationsshowed the increase in moisture re-sulted in a decreased weld area, in-creased voids, and more severe defor-mation around the voids, especially forhigh moisture content (>1 wt-%).Thus, joint with high moisture contentexhibited interfacial fracture. 4. The deleterious effect of waterabsorption on the weldability of Cf/PA 66 can be reversed by complete-ly redrying the workpiece before ultra-sonic welding. 5. The joint strength increased lin-early with the amplitude and was in-fluenced slightly by moisture when theamplitude was larger than 100 mduring welding. Therefore, the ampli-tude was proposed as an indication inmonitoring weld quality.

1. Powell, C. E., and Beall, G. W. 2006.Physical properties of polymer/claynanocomposites. Current Opinion in SolidState and Materials Science 10: 73–80. 2. Senthilvelan, S., and Gnanamoorthy,R. 2006. Damping characteristics of unre-inforced, glass and carbon fiber reinforcednylon 6/6 spur gears. Polymer Testing 25:6–62. 3. Karsli, N. G., and Aytac, A. 2013.Tensile and thermomechanical propertiesof short carbon fiber reinforced polyamide6 composites. Composites: Part B 51: 270–275. 4. Botelho, E. C., Figiel, L., Rezende, M.C., and Lauke, B. 2003. Mechanical behav-ior of carbon fiber reinforced polyamidecomposites. Composites Science and Technol-ogy 63: 1843–1855. 5. Diao, H., Robinson, P., Wisnom, M.R., and Bismarck, A. 2016. Unidirectionalcarbon fibre reinforced polyamide-12 com-posites with enhanced strain to tensile fail-ure by introducing fibre waviness. Compos-ites: Part A 87: 186–193. 6. Ageorges, C., Ye, L., and Hou, M.2001. Advances in fusion bonding tech-niques for joining thermoplastic matrixcomposites: A review. Composites: Part A32: 839–857. 7. Amanat, N., James, N. L., andMcKenzie, D. R. 2010. Welding methodsfor joining thermoplastic polymers for the

hermetic enclosure of medical devices.Medical Engineering & Physics 32: 690–699. 8. Chan, W. X., Ng, S. H., Li, K. H. H.,Park, W.-T., and Yoon, Y.-J. 2016. Micro-ul-trasonic welding using thermoplastic-elas-tomeric composite film. Journal of Materi-als Processing Technology 236: 183–188. 9. Qiu, J., Zhang, G., Asao, M., Zhang,M., Feng, H., and Wu, Y. 2010. Study onthe novel ultrasonic weld properties of het-erogeneous polymers between PC andPMMA. International Journal of Adhesion &Adhesives 30: 729–734. 10. Benatar, A., Eswaran, R. V., and Na-yar, S. K. 1989. Ultrasonic welding of ther-moplastics in the near-field. Polymer Engi-neering and Science 29: 1689–1698. 11. Rani, R. M., Suresh, K. S., Prakasan,K., and Rudramoorthy, R. 2007. A statisticalstudy of parameters in ultrasonic welding ofplastics. Experimental Techniques 31: 53–58. 12. Parmar, U., and Pandya, D. H. 2016.Experimental investigation of ultrasonicwelding on non-metallic material. ProcediaTechnology 23: 551–557. 13. Tanaka, K., Honda, S. M. H.,Katayama, T., and Enoki, S. 2013. Effect ofwater absorption on the mechanical prop-erties of carbon fiber/polyamide compos-ites. Journal of Solid Mechanics and Materi-als Engineering 7: 520–529. 14. Launay, A., Marco, Y., Maitournam,M. H., and Raoult, I. 2013. Modelling theinfluence of temperature and relative hu-midity on the time-dependent mechanicalbehaviour of a short glass fibre reinforcedpolyamide. Mechanics of Materials 56: 1–10. 15. Jia, N. Y., and Kagan, V. A. 2001.Mechanical performance of polyamideswith influence of moisture and tempera-ture — Accurate evaluation and better un-derstanding. Plastics Failure Analysis andPrevention, pp. 95–104. 16. Grammatikos, S. A., Zafari, B.,Evernden, M. C., Mottram, J. T., andMitchels, J. M. 2015. Moisture uptakecharacteristics of a pultruded fibre rein-forced polymer flat sheet subjected tohot/wet aging. Polymer Degradation andStability 121: 407–419. 17. Christian, S. J., and Billington, S. L.2012. Moisture diffusion and its impact onuniaxial tensile response of biobased com-posites. Composites: Part B 43: 2303–2312. 18. Sturm, R., Grimberg, R., Savin, A.,and Grum, J. 2015. Destructive and non-destructive evaluations of the effect ofmoisture absorption on the mechanicalproperties of polyester-based composites.Composites: Part B 71: 10–16. 19. Chaichanawong, J., Thongchuea, C.,and Areerat, S. 2016. Effect of moisture on

the mechanical properties of glass fiber re-inforced polyamide composites. AdvancedPowder Technology 27: 898–902. 20. Timmaraju, M. V., Gnanamoorthy,R., and Kannan, K. 2011. Influence of im-bibed moisture and organoclay on tensileand indentation behavior of polyamide66/hectorite nanocomposites. Composites:Part B 42: 466–472. 21. Sateesh, N., Rao, P. S., Ravishanker,D. V., and Satyanarayana, K. 2015. Effectof moisture on GFRP composite materials.Materials Today: Proceedings 2: 2902–2908. 22. Benatar, A. 1989. Ultrasonic weld-ing of PEEK graphite APC-2 composites.Polymer Engineering and Science 29:1705–1721. 23. Rani, M. R., and Rudramoorthy, R.2013. Computational modeling and experi-mental studies of the dynamic perform-ance of ultrasonic horn profiles used inplastic welding. Ultrasonics 53: 763–772. 24. Balkan, O., Demirer, H., andYildirim, H. 2008. Morphological and me-chanical properties of hot gas welded PE,PP and PVC sheets. Journal of Achievementsin Materials and Manufacturing Engineering31: 60–70. 25. Glowska, A. W., and Pietras, A.2012. Influence of the welding parameterson the structure and mechanical propertiesof vibration welded joints of dissimilargrades of nylons. Archives of Civil and Me-chanical Engineering 12: 198–204. 26. Zhu, L., Wang, C., and Qiu, Y. 2007.Influence of the amount of absorbed mois-ture in nylon fibers on atmospheric pres-sure plasma processing. Surface & CoatingsTechnology 201: 7453–7461. 27. Ye, L., Lu, M., and Mai, Y.-W. 2002.Thermal de-consolidation of thermoplasticmatrix composites — I. Growth of voids.Composites Science and Technology 62:2121–2130. 28. Vlasveld, D. P. N., Groenewold, J.,Bersee, H. E. N., Mendes, E., and Picken, S.J. 2005. Analysis of the modulus ofpolyamide-6 silicate nanocomposites usingmoisture controlled variation of the matrixproperties. Polymer 46: 6102–6113. 29. Fernandez Villegas, I. 2014.Strength development versus process datain ultrasonic welding of thermoplasticcomposites with flat energy directors andits application to the definition of opti-mum processing parameters. Composites:Part A 65: 27–37. 30. Levy, A., Le Corre, S., and Fernan-dez Villegas, I. 2014. Modeling of the heat-ing phenomena in ultrasonic welding ofthermoplastic composites with flat energydirectors. Journal of Materials ProcessingTechnology 214: 1361–1371.

WELDING RESEARCH


QIAN ZHI, XINRONG TAN, and ZHONGXIA LIU ([email protected]) are with the School of Physics and Engineering, Zhengzhou University, Zhengzhou, China.

References


effect of moisture on the ultrasonic welding of carbon ... · supplement to the welding journal,...

Documents