Effect of Intermediate Heat Treatment on the Mechanical Propertiesof 3003/4343 Aluminum Clad Sheet Manufactured by Strip Casting/Clad Rolling

Shin-Cheon Yun1, Kyu-Sik Kim1, Kwang Jun Euh2, Hyoung Wook Kim2 and Kee-Ahn Lee1,+

1School of Advanced Materials Engineering, Andong National University, Andong-Si, Gyeongbuk 760-749, Korea2Korea Institute of Materials Science, Changwon-Si, Gyeongnam 642-831, Korea

Effect of intermediate heat treatment on the microstructure evolution and mechanical properties of A3003/A4343 clad sheet manufacturedthrough new processes were investigated. Raw materials® both A3003 and A4343 alloys®were fabricated by twin roll strip casting process,and then cold rolling and roll bonding were performed with final thickness of 215µm. In the manufactured clad sheets, the thickness of 3003alloy was found to be 191µm, and that of 4343 alloy, 24 µm. Rolled and recovered microstructures appeared at low heat treatment temperatures(260°C, 280°C), and recrystallization/grain growth were noted from the A3003/A4343 interfaces at 573K or higher temperatures, with the area(recrystallization/grain growth) gradually increasing as heat treatment temperatures increased further. According to the results of tensile tests,yield strength was 223.7MPa, and tensile strength was 270MPa in as-rolled state. As the intermediate heat treatment temperature increased,strengths continuously decreased, whereas elongation exhibited an increasing tendency. Until the final tensile fracture, 3003 alloy and 4343 alloywere not separated but maintained well in bonded state. Based on the results of microstructure evolution by which microstructures were dividedinto three areas and changed following intermediate heat treatment, the tensile properties of A3003/A4343 clad sheets were analyzed anddiscussed. [doi:10.2320/matertrans.M2014298]

(Received August 18, 2014; Accepted November 7, 2014; Published December 19, 2014)

Keywords: twin-roll strip casting, clad material, intermediate heat treatment, microstructure, mechanical properties

1. Introduction

Clad metals are materials manufactured by bonding two ormore different metals using a special bonding process (i.e,cladding) and have the advantage of enabling the compositeuse of the properties of two or more kinds of metals.Therefore, clad metals are applied in diverse areas such as theautomobile industry, aviation industry, offshore plant in-dustry, and electronics industry.1­4) Among them, aluminumclad metals are lightweight materials with advantages such ashigh strength to weight ratio, high corrosion resistance, highthermal conductivity and high formability.1) They are appliedas materials of air conditioner tubes, evaporators, and car heatexchanger components such as condensers nowadays.

The A3003/A4343 clad metal used in heat exchangershas been conventionally manufactured through ingot casting,rolling and roll clad bonding processes. Specifically, A3003alloy and A4343 alloy manufactured through the ingotcasting process were made into respective sheets through hotrolling and cold rolling, and the two different kinds of sheetswent through stacking and rolling to be manufactured into thefinal clad material. Thereafter, to control the microstructuresand properties of the material, post-heat treatment processesare implemented. The abovementioned processes are rela-tively complex, leading to low energy efficiency and highprocess costs. In addition, due to the application of diverseprocesses, disadvantages such as non-uniform microstruc-tures of the clad material and difficulties in property controlmay also arise.5)

The strip casting process enables manufacturing thin sheetsdirectly from molten metal, thereby enabling the simplifica-tion of complicated subsequent rolling processes that wereindispensable in existing processes. In addition, it enablesobtaining fine structures due to the high cooling rate whenstrips are manufactured.6) Therefore, efforts to apply the strip

casting process as a clad metal manufacturing processhave been made. Recently, the possibility of manufacturingclad metals that incur low manufacturing costs (comparedto manufacturing raw materials through the existing ingotcasting processes) but show excellent properties® bymanufacturing the raw materials of clad metals through thestrip casting process® has been reported in relation toA3527/A4343 and A1050/Al-12%Si (mass%) clad materi-als.6,7) Nonetheless, no study has been conducted so far inrelation to the manufacture of A3003/A4343 clad metalusing raw materials manufactured through the strip castingprocess.

Meanwhile, in the case of aluminum-based clad sheetsused as heat exchanger parts, appropriate control of me-chanical properties is important due to the necessaryconditions for forming into parts and durability of final parts.Therefore, the manufactured clad sheets are subjected tointermediate heat treatment before being formed into parts intheir final shapes. However, the effects of intermediate heattreatment on the mechanical properties of clad metals appliedwith strip casting have not been reported so far.

In this study, A3003 alloy and A4343 alloy weremanufactured into raw materials using the strip castingprocess and then into A3003/A4343 clad sheets using theroll bonding process and the effect of intermediate heattreatment on the manufactured clad sheets was thenexamined.

2. Experimental Procedure

Al-Mn-based A3003 alloy was used as core material, andAl-Si-based A4343 alloy was applied as filler material. Thecompositions of the individual alloys are shown in Table 1.The raw materials were manufactured by Choil AluminumCo. (Korea) using the twin roll strip casting process, and themanufactured materials were 8mm thick sheets. 3003 alloyunderwent a 723 k, 1 hour annealing on a strip casted 8mm+Corresponding author, E-mail: keeahn@andong.ac.kr

Materials Transactions, Vol. 56, No. 2 (2015) pp. 242 to 248©2014 The Japan Institute of Metals and Materials EXPRESS REGULAR ARTICLE

thick sheet. Meanwhile, 4343 alloy underwent a 723 k, 1 hannealing and further went through additional cold rollinguntil 1mm thick (rolling ratio: 87.5%). To bond the twomaterials (8mm thick 3003 sheet + 1mm thick 4343 sheet)together, the facing surfaces were brushed, and they werebonded through rolling until 0.215mm thickness (rollingratio 97.6%) at room temperature (298K). The width of3003/4343 clad sheet was 120mm. A schematic diagramof the A3003/A4343 clad sheet manufacturing processis presented in Fig. 1. Thereafter, the clad sheets wereintermediate heat-treated at different temperatures of 533,553, 573, 593 and 613K for 14 h.

To observe the microstructures of the clad sheets, thesamples were ground using SiC paper and 1 µm-level Al2O3

slurry, and etched using a 35% Barker solution thereafter.OM (Optical Microscope) and SEM (Scanning ElectronMicroscope) were used for observation. In addition, microX-ray diffractometer (D/MAX RAPID-S), EDS (EnergyDispersion Spectroscopy) and EPMA (Electron Probe MicroAnalyzer) analyses were conducted to analyze the micro-structures before and after heat treatments.

To obtain the mechanical properties of the clad sheets,hardness and tensile tests were conducted. The hardnesswas measured 12 times repeatedly using load of 10 g, andthe average hardness value of 10 times of measurementsexcluding the highest and lowest values was used. Thespecimens used in the tensile tests were processed accordingto the ASTM E8M standard. The shapes and sizes are shownin Fig. 2. The specimens were prepared from the clad sheetcenter area placed in parallel to the rolling direction. Tensiletests were conducted using Instron 8801, and the tests wereperformed twice for each condition under an initial strain ratecondition of 4 © 10¹4 s¹1 at room temperature. The averageof the two tests was used as the tensile result. Tensilefractured surfaces were also observed using SEM.

3. Results and Discussions

3.1 Microstructure evolution with heat treatmentFigure 3(a) presents the initial microstructures of the

A3003/A4343 clad sheets (³0.215mm) made through thestrip casting/clad rolling processes. The area observed in thefigure is the center area of width direction. In the corematerial (A3003), approximately 191 µm thicknesses werenoted, and typical rolled microstructures (deformation lines)were observed. The filler material (A4343, upside) was foundto have been attached uniformly to 3003 alloy approximately24 µm thick. Figure 3(b) is the XRD analysis resultsperformed on 3003 side and 4343 side of the clad sheetthat are on the opposing surfaces. As a result, peaks of ¡-Al,

Fig. 1 Schematic diagram of manufacturing processes for A3003/A4343clad material.

(a)

(b)

Fig. 2 (a) Tensile specimen prepared from the clad sheet center area, and(b) morphology and size of specimen for the tensile test.

Fig. 3 (a) Cross-sectional microstructure, (b) XRD analysis results, and (c) SEM observation results of newly manufactured A3003/A4343 clad material.

Table 1 Chemical compositions of A3003 and A4343 alloys used in thisstudy (in mass%).

Si Fe Cu Mn Mg Ti Zn Al

A3003 0.53 0.419 0.192 1.40 0.005 0.0015 Bal.

A4343 6.91 0.318 0.014 0.022 0.003 Bal.

Effect of Intermediate Heat Treatment on the Mechanical Properties of 3003/4343 Aluminum Clad Sheet 243

Si and AlFeSi phases were identified on the 4343 alloy side,and peaks of ¡-Al, Al6Mn and AlFeMnSi phases wereidentified on 3003 alloy side. SEM and EDS analyses wereconducted to observe distribution and shape of constituentphases, with the results shown in Fig. 3(c). In the case of theA4343 filler material (upside), Si phase (black arrow) wasobserved in ¡-Al matrix. The Si phase was thought to beeutectic Si appearing in general Al-Si-based alloys. TheAlFeSi phase (dotted circle), known to be one of theconstituent phases appearing in A4343 alloy,8­10) wasadditionally observed. In the case of the A3003 core material,¡-Al matrix and intermetallic phases (white arrows) werenoted. Figure 4 presents the results of analysis of distributionof alloy elements using EPMA. The existence and distribu-tion of the same phases as those in the EDS analysis results inFig. 3 could be identified. Within A3003 alloy area, a phase(Al6Mn phase suspected) distributed on areas where Al andMn are identified, and a phase (AlFeMnSi phase suspected)with Al, Mn as well as Fe were identified. Based on the XRDand SEM/EDS analysis results of Fig. 3(b), (c) and EPMAanalysis results of Fig. 4, it is inferred that ¡-Al, Si andAlFeSi phases exist in 4343 alloy, and ¡-Al, Al6Mn andAlFeMnSi phases exist in 3003 alloy.

The A3003/A4343 clad material showing the above-mentioned microstructures was heat-treated for 14 h at 533,553, 573, 593 and 613K, and the observation results of themicrostructures are shown in Fig. 5. The thickness of heattreated clad materials was measured in the range from209µm to 217 µm. In the Figure, under the 533³553K heattreatment temperature condition, microstructures deformed/recovered from A3003 alloy were observed. Under the573³593K temperature condition, recrystallized/graingrowth microstructures were observed in the interfaces ofA4343/A3003 alloys; the area of the recrystallized/graingrowth microstructure was expanded as the temperatureincreased. In areas where recrystallized/grain growth micro-structures did not appear (at high temperature), deformed/recovered microstructures similar to those observed at lowheat treatment temperatures were observed. In the case ofhigh heat treatment temperature condition of 613K, coarsegrain structures wherein recrystallization/grain growth oc-

curred in general were observed in the A3003 alloy areas.The thicknesses of A3003 and A4343 alloys were shown tobe similar to the initial thicknesses regardless of intermediateheat treatment.

Figure 6 illustrates the results of phase analyses (XRD,SEM) conducted for individual intermediate heat treatmenttemperatures. According to the results of XRD phaseanalyses (Fig. 6(a)), the alloys were composed of the samephases as those in as-rolled state without the formation of anynew phase even after the heat treatment (compared to theXRD results in Fig. 3(b)). As can be seen from the SEMobservation results in Fig. 6(b)­(d), the light gray Fe-intermetallic phase and the dark grey eutectic Si phaseappeared similar to the as-rolled state, and the sizes anddistributions of the phases did not show big differences.

3.2 Tensile properties of A3003/A4343 clad sheets afterheat treatment

The tensile properties of the initial clad material and thoseof the material after intermediate heat treatments are shown inFig. 7(a). The yield strength and tensile strength of the initialclad material were found to be 223.7MPa and 270MPa,respectively, and the elongation was revealed to be 1.79%.On the other hand, the yield strength and tensile strength ofthe clad material after intermediate heat treatment at 613Kwere 106.8MPa and 160MPa, respectively, and the elonga-tion was 9.91%. In the A3003/A4343 clad material, as heattreatment temperatures increased, yield strength and tensilestrength continuously decreased, and elongation increased,which is a general tendency. When heat treatment wasconducted at 573³593K, however, whereas strength fol-lowed the general tendency of decreasing continuously,elongation showed a distinctive tendency of decreasingslightly (Fig. 7(b)).

Tensile fractured surfaces were observed after the tensiletests, with the results shown in Fig. 8. Regardless of whetherheat treatment has been carried out or not as well as theheat treatment temperature conditions, ductile fracture modecomposed of well-developed dimples generally observed inaluminum alloys were noted. As can be seen in Fig. 8(a)­(d),micro-cracks could not be found in the interfaces between

Fig. 4 Distribution of alloying elements (EPMA analysis) in A3003/A4343 clad material.

S.-C. Yun, K.-S. Kim, K. J. Euh, H. W. Kim and K.-A. Lee244

A3003 alloy and A4343 alloy. Therefore, the two alloys weredetermined to have bonded well with each other throughthe rolling bonding process. In general, unstable bondinginterfaces are reported to be a potential cause of deteriorationof mechanical characteristics.11,12) In the case of the cladmaterial manufactured in this study, the deterioration oftensile properties due to interface weakness can be judged tobe minimal. Meanwhile, a tendency of decreases in the areaof tensile rupture surfaces could be seen as the intermediateheat treatment temperatures increased; this is consistent withthe results that showed increases in elongation. As mentionedearlier in relation to Fig. 7, however, the results that showeddecreases in elongation under the 573³593K conditions aredifferent from those showing continuous increases in thereduction ratio of rupture area. Upon reviewing Fig. 3carefully in relation to this, the formation and expansion ofnew microstructures (that still have partial areas compared tothe entire deformed structures) where recrystallization/graingrowth occurred can be identified centering on the interfacesbetween A3003 alloy and A4343 alloy. The small decrease

in elongation in the 573³593K heat treatment temper-ature conditions shown in Fig. 7 is considered attributableto the plastic deformation instability related to the transi-tional development of the non-uniform (structures of 3003divided into two areas in general) microstructures shown inFig. 3.

3.3 Correlations between microstructures and tensileproperties of A3003/A4343 clad sheet

As factors that greatly affect clad materials’ tensileproperties, grain sizes, dislocation density, interface charac-teristics, and phase changes (kinds, sizes, and distribution)can be considered. As mentioned in the microstructures(Fig. 6) and the fracture surfaces observation (Fig. 8) results,the kinds, sizes, and distributions of second phases showedfew differences in relation to intermediate heat treatment. Inaddition, because the interface between the two alloys wasstable, there was no harmful effect of interface on tensileproperties (no crack occurred in the interface). Therefore,grain sizes and dislocation density can be judged to have the

Fig. 5 Optical micrographs showing changes of microstructure in A3003/A4343 clad materials after intermediate heat-treatment at(a) 533K, (b) 553K, (c) 573K, (d) 593K and (e) 613K for 14 h.

Fig. 6 Changes of microstructure in A3003/A4343 clad materials after intermediate heat-treatments; (a) XRD analysis results, and SEMobservation results after (b) 553K, (c) 573K, (d) 613K heat-treatments.

Effect of Intermediate Heat Treatment on the Mechanical Properties of 3003/4343 Aluminum Clad Sheet 245

major effects on the tensile properties of the A3003/A4343clad metal. In relation to this, X. P. Xing, et al.13) explainedthat dislocation density and grain sizes would change greatlyaccording to the heat treatment conditions of the deformedA3003 alloy (not clad material); thus having considerableeffect on the mechanical characteristics of the material.

Figure 9 presents the measured values of the microhardness of the core material (A3003) and filler material(A4343) constituting the clad sheets at various heat treatmenttemperatures in comparison with each other. The hardnessvalues of A3003 alloy and A4343 alloy were found to be68.6Hv and 78.0Hv, respectively, in the case of initialspecimens, and the general tendency of decreases in hardnessalong with increases in heat treatment temperatures wasnoted. In the case of 613K, the highest heat treatmenttemperature condition, the hardness values were 42.4Hv(A3003) and 45.6Hv (A4343), showing great decreasescompared to the initial hardness values.

In general, the relationship between yield strength andhardness can be expressed through the following Taborexpression:14,15)

Hv ¼ ~M·y ð1ÞWhere Hv is the Vickers hardness value, ~M is the Taborcoefficient (this may vary with the materials and theirconditions but generally shows values in a range of 2.5³3.8)and ·y is the yield strength. Applying the written relational

expression above, an attempt to examine the effects of themicrostructures of the A3003/A4343 clad material®whichchange according to intermediate heat treatment® on the

(a)

(b)

Fig. 7 Tensile properties of A3003/A4343 clad materials; (a) stress-straincurves and (b) variation of strengths and elongation with heat treatmenttemperature.

Fig. 8 Tensile fractographies of A3003/A4343 clad materials; (a) as-rolled, heat treated at (b) 553K, (c) 573K, (d) 613K, and (e) changes ofreduction of area after intermediate heat-treatment.

Fig. 9 Hardness of A3003/A4343 clad materials before and afterintermediate heat-treatment.

S.-C. Yun, K.-S. Kim, K. J. Euh, H. W. Kim and K.-A. Lee246

strength of the material was made, and the simple rule ofmixture relations was basically applied. Generally applied toclad materials, the rule of mixture relations is expressedconsidering only those alloys constituting the clad materials.However, the A3003/A4343 clad sheets subjected tointermediate heat treatment in this study (section 3.1) wereobserved to have been composed of three large parts (i.e., 1.A4343 area, 2. A3003 alloy deformation/recovery area, 3.A3003 alloy recrystallization/grain growth area). Thedistinguished areas were calculated using Image-Pro PLUS.This process was repeated ten times on various points of theclad material and their average was obtained. Then theaverage value was divided by the overall area of 3003/4343clad to obtain the area fraction. The area fractions of themicrostructure that changed after heat treatment were shownin Fig. 10.

Using the hardness values and area fractions of individualareas, the relations of yield strength were expressed asfollows:

·YSðcladÞ ¼ f3003,deformed �Hv3003,deformed=M1

þ f3003,rex �Hv3003,rex=M2

þ f4343 �Hv4343=M3 ð2ÞWhere f refers to the ratios by area, Hv is the Vickershardness values by area, and M pertains to Tabor coefficientsby area. Yield strength values according to changes intemperature conditions were predicted using expressionno. (2), with the results compared with the yield strengthvalues in actual tensile test results (Fig. 11). The coefficientsfitted here were shown to be M1 = 2.98, M2 = 3.66, andM3 = 3.69. As can be seen from the figure, the predictedvalues according to heat treatment changed similarly to actualyield strength. These results can be considered to be evidenceshowing that changes in the mechanical properties of the3003/4343 clad sheets according to heat treatment aredetermined by the degree of softening and fraction changes ofthe three areas (in the two alloys). In addition, throughcomparisons of the Tabor coefficients of the two alloys andthree areas, the fact that the 3003 deformation/recoveryarea®wherein relatively less recrystallization/grain growthoccurred® contributed to the strength of the clad sheets themost (smallest M value) can be inferred.

4. Conclusion

In this study, the microstructures and mechanical proper-ties of the 3003/4343 clad material manufactured using thenew processes® strip casting/clad rolling® according tointermediate heat treatment temperatures were examined; thefollowing conclusions were drawn:

By applying the new processes, i.e., strip casting/cladrolling, 215 µm thick A3003/A4343 clad sheets could bemanufactured. As the intermediate heat treatment temper-atures increased, recrystallized/grain grown microstructuresbegan to be formed first in areas close to the 3003/4343interface of 3003 alloy, and recrystallization/grain growthareas were expanded to the entire 3003 alloy at 613K. As forthe precipitates, ¡-Al and Fe-intermetallic phases appearedcommonly in both alloys regardless of intermediate heattreatment temperatures; in addition, the Al6Mn phase wasidentified in Al 3003 alloy, and eutectic Si phase, in Al 4343alloy. However, there was almost no difference in phasedistribution or shapes according to the intermediate heattreatment temperatures.

In the case of initial clad specimens, 3003 alloy and 4343alloy showed hardness values of 68.6Hv and 78.0Hv,respectively; the hardness of the alloys gradually decreasedas intermediate heat treatment temperatures increased. In thecase of 613K heat treatment, the hardness values of thealloys were identified to be 42.4Hv (A3003) and 45.6Hv(A4343). Through tensile tests, yield strength of 223.7MPa,tensile strength of 270.07MPa, and elongation of 1.79%were obtained from the A3003/A4343 clad sheets; the yieldstrength and tensile strength continuously decreased, whereasthe elongation gradually increased as intermediate heattreatment temperatures increased. Finally, values® tensilestrength of 160MPa, yield strength of 106.8MPa andelongation of 9.91%®were recorded under the intermediateheat treatment 340°C condition. The tensile fractography ofthe clad sheets showed the typical ductile fracture mode. Thebonding interfaces of the two alloys were not separated, withthe bonding maintained well until the clad sheets were totallyfractured.

Based on the measured hardness values of individualalloys according to intermediate heat treatment temperatures

Fig. 10 Area fractions of three different microstructure areas withintermediate heat-treatment temperature.

Fig. 11 Comparison between estimated strength and experimental strengthof A3003/A4343 clad materials.

Effect of Intermediate Heat Treatment on the Mechanical Properties of 3003/4343 Aluminum Clad Sheet 247

and the measured fraction values of the three observed areas(3003 alloy deformation/recovery area, 3003 alloy recrystal-lization/grain growth area, 4343 alloy area), Tabor expres-sion and simple rule of mixture formula were applied topredict changes in the strength of the clad sheets; based onthe results, the changes were analyzed well. Accordingly, itcould be seen that the mechanical properties of A3003/A4343 alloy are mainly determined by the microstructuresoftening behavior of each alloy appearing after intermediateheat treatment.

Acknowledgments

We would like to acknowledge the financial support fromthe R&D Convergence Program of MSIP (Ministry ofScience, ICT and Future Planning) and ISTK (KoreaResearch Council for Industrial Science and Technology) ofRepublic of Korea (Grant B551179-11-02-00).

REFERENCES

1) W. S. Miller, L. Zhuang, J. Bottema, A. J. Wittebrood, P. De Smet, A.Haszler and A. Vieregge: Mater. Sci. Eng. A 280 (2000) 37­49.

2) J. Lacaze, S. Tierce, M.-C. Lafont, Y. Thebault, N. Pebere, G.Mankowski, C. Blanc, H. Robidou, D. Vaumousse and D. Daloz:Mater. Sci. Eng. A 413­414 (2005) 317­321.

3) E. A. Starke, Jr. and J. T. Staley: Prog. Aerospace Sci. 32 (1996) 131­172.

4) S. M. Lee, S. M. Kim, S. N. We, D. H. Bae, G. A. Lee, J. S. Lee, Y. B.Kim and D. S. Bae: Korean J. Met. Mater. 50 (2012) 316­323.

5) H. X. Jiang, H. T. Zhang, K. Qin and J. Z. Cui: Trans. Nonferrous Met.Soc. China 21 (2011) 1692­1697.

6) T. Haga and S. Suzuki: J. Mater. Process. Technol. 138 (2003) 366­371.

7) K. Euh, S. H. Kim, H. W. Kim, D. B. Kim and Y. M. Oh: Trans. Mater.Process. 22 (2013) 264­269.

8) J. P. Martins, A. L. M. Carvalho and A. F. Padilha: J. Mater. Sci. 44(2009) 2966­2976.

9) J. M. Howe: Metall. Trans. A 17 (1986) 593­605.10) A. M. Samuel, J. Gauthier and F. H. Samuel: Metall. Mater. Trans. A 27

(1996) 1785­1798.11) K. S. Lee, D. H. Yoon, H. K. Kim, Y. N. Kwon and Y. S. Lee: Mater.

Sci. Eng. A 556 (2012) 319­330.12) S. Roy, B. R. Nataraj, S. Suvas, S. Kumar and K. Chattopadhyay:

Mater. Des. 36 (2012) 529­539.13) Z. P. Xing, S. B. Kang and H. W. Kim: J. Mater. Sci. 39 (2004) 1259­

1265.14) J. D. Currey and K. Brear: J. Mater. Sci. Mater. Med. 1 (1990) 14­20.15) J. R. Cahoon, W. H. Broughton and A. R. Kutzak: Met. Trans. 2 (1971)

1979­1983.

S.-C. Yun, K.-S. Kim, K. J. Euh, H. W. Kim and K.-A. Lee248