Dear Sir,
Contribution of the Phase-Matrix Interface in the Behaviour of Aluminium Filled Epoxies 1
Pedro V. Vasconcelos1, F. Jorge Lino2, Rui J. L. Neto3, Paula Henrique3
ESTG/ IPVC - Instituto Politécnico de Viana do Castelo, Ap. 574, 4900-908 Viana do Castelo, Portugal, Tel: 258819700, Fax: 258.827636, [email protected] 2 FEUP – Faculdade de Engenharia da Universidade do Porto, DEMEGI – Departamento de Engenharia Mecânica e Gestão Industrial, Rua Dr. Roberto Frias, 4200-465 Porto, Portugal. Tel: 225081704 (42), Fax. 229537352, [email protected] 3 INEGI – Instituto de Engenharia Mecânica e Gestão Industrial, CETECOFF – Unidade de Fundição e Novas Tecnologias, Rua do Barroco, 174-214, 4465-591 Leça do Balio, Porto, Portugal. Tel: 229578714, Fax. 229537352, [email protected] Keywords: Filled epoxy resins, composites, interfaces, rapid tooling.
Abstract. Polymeric materials present mechanical and thermal limitations that disable their use in the mould manufacturing. Nevertheless, an adequate selection of the polymeric matrix and the dispersed materials allows the possibility to achieve performances closer to the metals and their alloys. These new materials are attractive solutions for applications that have less demanding mechanical properties, as is the case of the rapid tooling applications where a short number of parts are required. The composites were obtained from a mixture of an epoxy resin with fine and coarse aluminium particles and glass and carbon fibres. One can state that besides the dispersed phase resistance overcome the matrix one, its contribution to the global resistance of the composite is restricted, because the fracture surface lies basically in the matrix and interfaces. As the matrix section under load is reduced, with the increment of the dispersed phase, the composite properties turn out to be dependent on the interfaces quality and resistance. This is particularly true, when fine aluminium particles and milled fibres with lengths shorter than the critical length are used. The interface contribution to the global composite properties depends basically on two parameters: - The binding quality between the matrix and the dispersed phase (unitary adhesion resistance), and; - The interface extension per unit volume. It is well known that the adhesion mechanism between the aluminium particles and the epoxy matrix depends on a complex combination of physical, mechanical and chemical interactions at the interface. This communication studies the main contribution of the phase-matrix interface in the mechanical behaviour of aluminium filled epoxy. A similar approach applied to the fibres will be object of other future communication. Introduction Aluminium filled resins are frequently employed in the area of the Rapid Prototyping (RP) and Rapid Tooling (RT) to manufacture moulds for production of small series of plastic parts [1-3]. The strength of these composites is very sensitive to the different phase properties and to the respective concentrations. Furthermore, as shown in this study, the interface also seems to be an important parameter in the composite mechanical behaviour. The adhesive resistance through the interface depends essentially on the extension and quality of the adhesive bonding. The chemical complexity of the epoxy systems and the respective interfacial interactions with the aluminium surfaces represents an extra difficulty to the interpretation of the adhesive bonding mechanism. Different theories are under development to explain the adhesion mechanism, and how it affects the interface resistance [4, 5].
Experimental Two epoxy systems for high (A) and medium (B) temperature (see Table 1) were mixed with aluminium particles of two different classes, fine (F) and coarse (C), with a medium equivalent diameter of 45.5 µm and 1400 µm, respectively. Eight composites were manufactured, A1 to A4 and B1 to B4. The matrix epoxy system is represented by a letter and the aluminium class by a number (Table 2 and Fig. 1). Table 1. Characteristics of the two epoxy matrix systems. Epoxy system Hardener Mixture Main Curing Process Viscosity Cycloaliphatic 700-900 - Addition of amine to the epoxy group. A – based on polyamine [mPa.s] - Homopolymerization of epoxide and aromatic cyclization. glycidyl amine Cycloaliphatic 1200-1600 - Addition of amine to the epoxy group. B – based on polyamine [mPa.s] bisphenol A/F + epichlorhydrin
Tg [ºC] 200
120
Table 2. Aluminium particles dimensions and compositions employed in the polymeric matrix composites. Composite 1 Composite 2 Composite 3 Composite 4 Dispersed Equivalent phase medium diameter [µm] Mixture: F - Fine 45.5 100% fine Al Mixture: Aluminium Vp = 0.41 50% coarse Al 75% coarse Al + 50% fine Al + 25% fine Al 100% coarse Al C - Coarse 1400 Vp = 0.50 Vp = 0.50 Vp = 0.50 aluminium Vp = particle volume fraction
(a) (b) (c) Fig. 1. Optical micrographs of aluminium particles (a) fine, (b) coarse and (c) mixture The flexural strength and Charpy impact tests were performed according to ASTM D790-02 and D5942-98 standards, respectively. The ASTM D1002-94 and D2093-95 standards were used to determinate the shear strength of the aluminium-resin interface adhesive bonding of and the aluminium surface preparation for the test, respectively. Two different methods for the aluminium surfaces preparation were used:, namely a mechanical method which used an aluminium oxide abrasive, and a chemical one with an acid etching solution (sulfuric acid/sodium dichromate).
Results
20
120
B
100
Impact str. (kJ.m-2)
Flexural strength (MPa)
Flexural strength and Charpy impact tests have demonstrated that the mechanical behaviour of the two resin systems, are strongly dependent on the aluminium particles addition (Fig. 2). The unfilled A system that is characterized by high structural heterogeneity and crosslinking density, shows a lower strength and toughness. In this system, fine aluminium addition not only improves the impact strength but also the flexural strength. This composite surface fracture exhibits an intergranular nature, due to the particle/epoxy bond. With aluminium addition, both systems tend to exhibit similar flexural and impact strength due to the fact that these characteristics are interface dependent. Based on figure 1 results, one can conclude that the aluminium/epoxy interface quality of the in A system is better than in the B system.
80 60
A
40 20 0 0
0.2
0.4
Aluminium particle fraction
15
B
10
A
5 0 0
0.2
0.4
Aluminium particle fraction
(a)
(b) Fig. 2. The effect of the aluminium concentration (fine class) in the mechanical behaviour of A1 and B1 composites: (a) flexural strength, (b) Charpy impact strength. Interface Quality. The shear strength of the aluminium-resin interface, determined by the adhesive bond test, prove that the A system shows better interfacial resistance with the aluminium particles (Fig. 3). Recent studies, based in XPS analysis, revealed that interaction of the epoxy resins nitrogen molecules with the oxidised and hydroxylated aluminium surfaces are the main contribution for the adhesion in these epoxy systems [6, 7]. Considering that A system has an amine group concentration of about 6% and that B system only has 1,5%, it seems that this system tend to develop greater chemical interaction with the aluminium particles than the B system, which means higher interfacial strength revealed in the experimental tests. The lower viscosity of A system also contributes for a better wettability and bonding. Interface Extension. Image processing and quantitative analysis allow the determination of the total particles perimeter divided by the total test area, LA, that can be related with the interface area per unit composite volume, Sv (composite interfacial specific area) [8]. Sv = 4 LA / π
(1)
The Sv parameter can be used as a measure of the aluminium-resin interface area. This value depends on the aluminium concentration, particle shape and particle size distribution. The two particle classes have very different size distributions, demanding different amplification factors in the image acquisition processing. For this reason, the composite Sv parameter prepared with the mixture of the two classes, was estimated based on the results obtained from the ones prepared with only fine and only coarse aluminium particles (Table 3).
Adhesive shear strength (MPa)
4 3 2 1 0
A
B Epoxy system
mechanical finishing
chemical finishing
Fig. 3. Adhesive shear strength of single-lap-joint of A and B epoxy systems with aluminium and different surface preparation. The particle specific area (per unit particle volume), sv, is obtained dividing the Sv value by the particle volume fraction, Vp. When the Sv results from A1 to A4 and B1 to B4 composites are compared, one can figure out that there is a relation of the flexural strength and the impact strength with the interface specific surface (Fig. 4). The mechanical behaviour indicated by the impact and flexural curves is very similar, in both systems. This result shows a potential correlation between the Sv parameter (that measure the interface extension) and the composite strength. Table 3. Medium size and sv medium parameter of the two aluminium particle classes and the Sv parameter for the different composites, determined by image processing. Composites
Aluminium particles Particle
Medium
sv (mm-1)
Type
sv (mm-1)
Vp
Sv = Vp x sv (mm-1)
A1/B1
171.2
0.41
70.2
A2/B2
90.0
0.5
45.0
class
Size (µm)
Fine
45.5
171.2
A3/B3
51.6
0.5
25.8
Coarse
1400
13
A4/B4
13.0
0.5
6.5
Conclusions The composite interfacial specific area, Sv, assess the degree of interaction in the interface and the respective contribution to the composite mechanical behaviour. Considering that the interface extension is proportional to the respective interactions to transfer the stresses between the matrix and the dispersed phases, the specific area improvement contributes to a better mechanical performance of the particle filled composite, as it was intention to prove in quantitative terms. The adhesion degree at the interface, experimentally determined by the shear test proves to be in accordance with the flexural and impact strength results of the particle filled epoxies. In general terms one can conclude that epoxy composites with high aluminium particle concentrations display a mechanical behaviour dependent on the interface properties (adhesive strength and interface extension per unit volume).
10
0 0 10 20 30 40 50 60 70
10
50 30
5
10
0
-2
B
0 10 20 30 40 50 60 70 -1
Particle specific area (mm ) Flexion
15 (KJ.m )
5
30
70
Impact strength
50
Flexural Strength (MPa)
10
-2
A
(KJ.m )
15 Impact strength
Flexural strength (MPa)
70
Impact
-1
Particle Specific Area (mm ) Flexion
(a)
Impact
(b)
Fig. 4. Flexural strength and impact strength vs Sv in composites based in: (a) A epoxy system and, (b) B epoxy system. Acknowledgements The authors would like to acknowledge the financial support from FEDER through the project POCTI/EME/41199/2001, “Development of an Indirect Rapid Tooling Process Based in Polymeric Matrix Composites”, approved by the Fundação para a Ciência e Tecnologia (FCT) and POCTI. References [1] F. Lino, F. Braga, M. Simão, R. Neto and T. Duarte; Protoclick – Prototipagem Rápida, (Protoclick, Porto, Portugal, 2001). [2] P. Hilton and P. Jacobs; Rapid Tooling, Technologies and Industrial Applications, (Marcel Dekker, Inc., USA, 2000). [3] P.Vasconcelos, F. Lino and R. Neto; Materials Science Forum (2003), p.169. [4] A. Pocius; Adhesion and Adhesives Technology, (Hanser Pub., NY, 1997). [5] E. Pisanova and E. Mader; J. Adhes. Sci. Technol. Vol. 14 (2000), p. 415. [6] M. Barthes-Labrousse; Journal of. Adhesion Vol. 57 (1996), p. 65. [7] C. Fauquet, P. Dubot, L. Minel, M. Barthes-Labrousse, et al; Appl. Surf. Sci. Vol. 81 (1994), p. 435. [8] E. Underwood; in Metallography and Microstructures, Vol.9, (ASM Handbook, 1985).
Pedro V. Vasconcelos1, F. Jorge Lino2, Rui J. L. Neto3, Paula Henrique3
ESTG/ IPVC - Instituto Politécnico de Viana do Castelo, Ap. 574, 4900-908 Viana do Castelo, Portugal, Tel: 258819700, Fax: 258.827636, [email protected] 2 FEUP – Faculdade de Engenharia da Universidade do Porto, DEMEGI – Departamento de Engenharia Mecânica e Gestão Industrial, Rua Dr. Roberto Frias, 4200-465 Porto, Portugal. Tel: 225081704 (42), Fax. 229537352, [email protected] 3 INEGI – Instituto de Engenharia Mecânica e Gestão Industrial, CETECOFF – Unidade de Fundição e Novas Tecnologias, Rua do Barroco, 174-214, 4465-591 Leça do Balio, Porto, Portugal. Tel: 229578714, Fax. 229537352, [email protected] Keywords: Filled epoxy resins, composites, interfaces, rapid tooling.
Abstract. Polymeric materials present mechanical and thermal limitations that disable their use in the mould manufacturing. Nevertheless, an adequate selection of the polymeric matrix and the dispersed materials allows the possibility to achieve performances closer to the metals and their alloys. These new materials are attractive solutions for applications that have less demanding mechanical properties, as is the case of the rapid tooling applications where a short number of parts are required. The composites were obtained from a mixture of an epoxy resin with fine and coarse aluminium particles and glass and carbon fibres. One can state that besides the dispersed phase resistance overcome the matrix one, its contribution to the global resistance of the composite is restricted, because the fracture surface lies basically in the matrix and interfaces. As the matrix section under load is reduced, with the increment of the dispersed phase, the composite properties turn out to be dependent on the interfaces quality and resistance. This is particularly true, when fine aluminium particles and milled fibres with lengths shorter than the critical length are used. The interface contribution to the global composite properties depends basically on two parameters: - The binding quality between the matrix and the dispersed phase (unitary adhesion resistance), and; - The interface extension per unit volume. It is well known that the adhesion mechanism between the aluminium particles and the epoxy matrix depends on a complex combination of physical, mechanical and chemical interactions at the interface. This communication studies the main contribution of the phase-matrix interface in the mechanical behaviour of aluminium filled epoxy. A similar approach applied to the fibres will be object of other future communication. Introduction Aluminium filled resins are frequently employed in the area of the Rapid Prototyping (RP) and Rapid Tooling (RT) to manufacture moulds for production of small series of plastic parts [1-3]. The strength of these composites is very sensitive to the different phase properties and to the respective concentrations. Furthermore, as shown in this study, the interface also seems to be an important parameter in the composite mechanical behaviour. The adhesive resistance through the interface depends essentially on the extension and quality of the adhesive bonding. The chemical complexity of the epoxy systems and the respective interfacial interactions with the aluminium surfaces represents an extra difficulty to the interpretation of the adhesive bonding mechanism. Different theories are under development to explain the adhesion mechanism, and how it affects the interface resistance [4, 5].
Experimental Two epoxy systems for high (A) and medium (B) temperature (see Table 1) were mixed with aluminium particles of two different classes, fine (F) and coarse (C), with a medium equivalent diameter of 45.5 µm and 1400 µm, respectively. Eight composites were manufactured, A1 to A4 and B1 to B4. The matrix epoxy system is represented by a letter and the aluminium class by a number (Table 2 and Fig. 1). Table 1. Characteristics of the two epoxy matrix systems. Epoxy system Hardener Mixture Main Curing Process Viscosity Cycloaliphatic 700-900 - Addition of amine to the epoxy group. A – based on polyamine [mPa.s] - Homopolymerization of epoxide and aromatic cyclization. glycidyl amine Cycloaliphatic 1200-1600 - Addition of amine to the epoxy group. B – based on polyamine [mPa.s] bisphenol A/F + epichlorhydrin
Tg [ºC] 200
120
Table 2. Aluminium particles dimensions and compositions employed in the polymeric matrix composites. Composite 1 Composite 2 Composite 3 Composite 4 Dispersed Equivalent phase medium diameter [µm] Mixture: F - Fine 45.5 100% fine Al Mixture: Aluminium Vp = 0.41 50% coarse Al 75% coarse Al + 50% fine Al + 25% fine Al 100% coarse Al C - Coarse 1400 Vp = 0.50 Vp = 0.50 Vp = 0.50 aluminium Vp = particle volume fraction
(a) (b) (c) Fig. 1. Optical micrographs of aluminium particles (a) fine, (b) coarse and (c) mixture The flexural strength and Charpy impact tests were performed according to ASTM D790-02 and D5942-98 standards, respectively. The ASTM D1002-94 and D2093-95 standards were used to determinate the shear strength of the aluminium-resin interface adhesive bonding of and the aluminium surface preparation for the test, respectively. Two different methods for the aluminium surfaces preparation were used:, namely a mechanical method which used an aluminium oxide abrasive, and a chemical one with an acid etching solution (sulfuric acid/sodium dichromate).
Results
20
120
B
100
Impact str. (kJ.m-2)
Flexural strength (MPa)
Flexural strength and Charpy impact tests have demonstrated that the mechanical behaviour of the two resin systems, are strongly dependent on the aluminium particles addition (Fig. 2). The unfilled A system that is characterized by high structural heterogeneity and crosslinking density, shows a lower strength and toughness. In this system, fine aluminium addition not only improves the impact strength but also the flexural strength. This composite surface fracture exhibits an intergranular nature, due to the particle/epoxy bond. With aluminium addition, both systems tend to exhibit similar flexural and impact strength due to the fact that these characteristics are interface dependent. Based on figure 1 results, one can conclude that the aluminium/epoxy interface quality of the in A system is better than in the B system.
80 60
A
40 20 0 0
0.2
0.4
Aluminium particle fraction
15
B
10
A
5 0 0
0.2
0.4
Aluminium particle fraction
(a)
(b) Fig. 2. The effect of the aluminium concentration (fine class) in the mechanical behaviour of A1 and B1 composites: (a) flexural strength, (b) Charpy impact strength. Interface Quality. The shear strength of the aluminium-resin interface, determined by the adhesive bond test, prove that the A system shows better interfacial resistance with the aluminium particles (Fig. 3). Recent studies, based in XPS analysis, revealed that interaction of the epoxy resins nitrogen molecules with the oxidised and hydroxylated aluminium surfaces are the main contribution for the adhesion in these epoxy systems [6, 7]. Considering that A system has an amine group concentration of about 6% and that B system only has 1,5%, it seems that this system tend to develop greater chemical interaction with the aluminium particles than the B system, which means higher interfacial strength revealed in the experimental tests. The lower viscosity of A system also contributes for a better wettability and bonding. Interface Extension. Image processing and quantitative analysis allow the determination of the total particles perimeter divided by the total test area, LA, that can be related with the interface area per unit composite volume, Sv (composite interfacial specific area) [8]. Sv = 4 LA / π
(1)
The Sv parameter can be used as a measure of the aluminium-resin interface area. This value depends on the aluminium concentration, particle shape and particle size distribution. The two particle classes have very different size distributions, demanding different amplification factors in the image acquisition processing. For this reason, the composite Sv parameter prepared with the mixture of the two classes, was estimated based on the results obtained from the ones prepared with only fine and only coarse aluminium particles (Table 3).
Adhesive shear strength (MPa)
4 3 2 1 0
A
B Epoxy system
mechanical finishing
chemical finishing
Fig. 3. Adhesive shear strength of single-lap-joint of A and B epoxy systems with aluminium and different surface preparation. The particle specific area (per unit particle volume), sv, is obtained dividing the Sv value by the particle volume fraction, Vp. When the Sv results from A1 to A4 and B1 to B4 composites are compared, one can figure out that there is a relation of the flexural strength and the impact strength with the interface specific surface (Fig. 4). The mechanical behaviour indicated by the impact and flexural curves is very similar, in both systems. This result shows a potential correlation between the Sv parameter (that measure the interface extension) and the composite strength. Table 3. Medium size and sv medium parameter of the two aluminium particle classes and the Sv parameter for the different composites, determined by image processing. Composites
Aluminium particles Particle
Medium
sv (mm-1)
Type
sv (mm-1)
Vp
Sv = Vp x sv (mm-1)
A1/B1
171.2
0.41
70.2
A2/B2
90.0
0.5
45.0
class
Size (µm)
Fine
45.5
171.2
A3/B3
51.6
0.5
25.8
Coarse
1400
13
A4/B4
13.0
0.5
6.5
Conclusions The composite interfacial specific area, Sv, assess the degree of interaction in the interface and the respective contribution to the composite mechanical behaviour. Considering that the interface extension is proportional to the respective interactions to transfer the stresses between the matrix and the dispersed phases, the specific area improvement contributes to a better mechanical performance of the particle filled composite, as it was intention to prove in quantitative terms. The adhesion degree at the interface, experimentally determined by the shear test proves to be in accordance with the flexural and impact strength results of the particle filled epoxies. In general terms one can conclude that epoxy composites with high aluminium particle concentrations display a mechanical behaviour dependent on the interface properties (adhesive strength and interface extension per unit volume).
10
0 0 10 20 30 40 50 60 70
10
50 30
5
10
0
-2
B
0 10 20 30 40 50 60 70 -1
Particle specific area (mm ) Flexion
15 (KJ.m )
5
30
70
Impact strength
50
Flexural Strength (MPa)
10
-2
A
(KJ.m )
15 Impact strength
Flexural strength (MPa)
70
Impact
-1
Particle Specific Area (mm ) Flexion
(a)
Impact
(b)
Fig. 4. Flexural strength and impact strength vs Sv in composites based in: (a) A epoxy system and, (b) B epoxy system. Acknowledgements The authors would like to acknowledge the financial support from FEDER through the project POCTI/EME/41199/2001, “Development of an Indirect Rapid Tooling Process Based in Polymeric Matrix Composites”, approved by the Fundação para a Ciência e Tecnologia (FCT) and POCTI. References [1] F. Lino, F. Braga, M. Simão, R. Neto and T. Duarte; Protoclick – Prototipagem Rápida, (Protoclick, Porto, Portugal, 2001). [2] P. Hilton and P. Jacobs; Rapid Tooling, Technologies and Industrial Applications, (Marcel Dekker, Inc., USA, 2000). [3] P.Vasconcelos, F. Lino and R. Neto; Materials Science Forum (2003), p.169. [4] A. Pocius; Adhesion and Adhesives Technology, (Hanser Pub., NY, 1997). [5] E. Pisanova and E. Mader; J. Adhes. Sci. Technol. Vol. 14 (2000), p. 415. [6] M. Barthes-Labrousse; Journal of. Adhesion Vol. 57 (1996), p. 65. [7] C. Fauquet, P. Dubot, L. Minel, M. Barthes-Labrousse, et al; Appl. Surf. Sci. Vol. 81 (1994), p. 435. [8] E. Underwood; in Metallography and Microstructures, Vol.9, (ASM Handbook, 1985).
