ChineseJournal of Aeronautics

Chinese Journal of Aeronautics 22(2009) 449-452 www.elsevier.com/locate/cja

TTT Diagram Used to Control Phase Structure of 2/4 Functional Epoxy Blends for Advanced Composites

Zhang Ming, An Xuefeng, Tang Bangming, Yi Xiaosu*

National Key Laboratory of Advanced Composites Materials, Beijing Institute of Aeronautical Materials, Beijing, 100095, China

Received 1 August 2008; accepted 21 October 2008

Abstract

This article uses a 2/4 functional epoxy blend system (E-54/AG-80) cured with diaminodiphenyl sulphone(DDS) as a raw ma-terial and develops a methodological procedure to establish a cure kinetic model with isothermal and dynamic differential scan-ning calorimeter(DSC) method and a gelation model with round-disk compression mode dynamic mechanical analyzer(DMA), thus acquiring a series of experimental data. Characteristic temperatures such as initial glass transition temperature Tg0, gelation glass transition temperature gelTg, and infinite glass transition temperature Tg are determined. The cure degree at gelation is turned out to be 0.45, while gelTg is found to be 70.2 C. The data are then used to form time-temperature-transition(TTT) dia-gram of the system, which serves to be a tool for process optimization of epoxy-matrix composites. A new cure processing is therefore derived from the TTT diagram. The final phase structure obtained from a controllable method is identified through scanning electron microscope(SEM) photographs to be of “ex-situ” phase morphology.

Keywords: composites; epoxy resins; TTT diagram; phase structure

1. Introduction1

Time-Temperature-Transition (TTT) diagram was originally developed to describe the phase transition behavior of steel. Afterwards, the TTT concept was extended to the depiction of the complete curing be-havior of epoxies as a function of temperature and time[1-2]. A typical TTT diagram for reactive thermo-setting systems is characterized by three transition temperatures: infinite glass transition temperature Tg , initial glass transition temperature Tg0, and gelation glass transition temperature gelTg, as well as three tran-sition lines: vitrification line, gelation line, and iso-conversion line. This article develops a methodological procedure to establish a cure kinetic model and a gela-tion model with a series of experimental data. The models and the data are then used to form a TTT dia-gram of the bi-component epoxy system, which is ex-pected to be a tool for process optimization of ep-oxy-matrix composites. A controllable method of achieving “ex-situ” phase structure is derived from the TTT diagram, and the results of the new technique *Corresponding author. Tel.: +86-10-62496740. E-mail address: xiaosu.yi@gmail.com Foundation item: National Basic Research Program of China

(2003CB615604) 1000-9361/$ - see front matter © 2009 Elsevier Ltd. All rights reserved. doi:10.1016/S1000-9361(08)60124-7

have been validated by means of scanning electron microscope (SEM) photographs.

2. Experimental

2.1. Materials

The basic resin system used in this study comprised 2/4 functional epoxy blends of dialycidal ether of bisphenol-A(DGEBA), E-54, from Wuxi Resin Fac-tory, Jiangsu province, China and tetraglycidyl me-thylene dianiline(TGMDA), AG-80, from Shanghai Institute of Synthetic Resins, China, with diaminodi-phenyl sulphone (DDS) as a curing agent. They were mixed in the ratio of E-54 AG-80 DDS = 2 3 2 to form an in-situ epoxy system. The toughening agent was poly aromatic ether ketone(PAEK) film produced by authors’ laboratory.

2.2. Blending procedure and sample preparation

Both components of the epoxy system were first mixed and dissolved in tetrahydrofuran (THF) under agitation at room temperature to form a homogenous solution. A vacuum pump was employed to extract the solvent out of the resin system for 72 hours. The pre-pared resin was preserved at 20 C. For thermal me-chanical analysis (dynamic mechanical analyzer (DMA) in compression mode), round-disk samples, 15 mm dia-

· 450 · Zhang Ming et al. / Chinese Journal of Aeronautics 22(2009) 449-452 No.4

meter, were fabricated out of glass fabric as a carrier. The resin was also poured into a rectangle mold of 20 mm 8 mm 5 mm. A solid PAEK film was then laid on the resin surface to co-cure in the procedure: 130 C/1 h + 180 C/2 h + 200 C/2 h based on the results from the TTT study. The cured pieces for SEM study were fractured in liquid-nitrogen and chemically etched by immersing them in methylene chloride (preferential solvent for PAEK) at room temperature for 48 hours.

2.3. Techniques

A differential scanning calorimeter (DSC), Q10, from Thermal Analysis & Rheology Instruments Inc. and a DMA, Q800, from the same manufacturer were used to carry out the experiments.

3. Results and Discussion

3.1. Gelation study

Gelation is a transition phenomenon characterized by beginning of cross-linking reaction of reactive polymers. The gelation time tgel is meant the time when gelation occurs, which can be experimentally deter-mined with DMA in compression mode. Table 1 shows tgel measured on the bi-component epoxy system. Fig.1 illustrates tgel in the form of ln tgel plotted against 1/T , which is fit by using Eq.(1)[2] into a linear relationship out of otherwise scattered points. ln tgel is calculated from Eq.(1), which is the mathematical description of the gelation model with given parameters A and Ea.

agelln Et A

RT (1)

Table 1 Gelation time at different isothermal tempera-tures

Isothermal temperature/ C Gelation time/s 130 8 189.4 140 5 706.6 150 3 745.2 160 1 177.8 170 1 047.6 180 750.0

Fig.1 Linear fitting of gelation time against temperature.

3.2. Cure kinetic and vitrification line

Table 2 lists all the information of the resin system collected from the first and second scans of the fresh resin by DSC, including the results calculated with the software developed by Thermal Analysis & Rheology Instruments Inc. It should be made clear that the values of Ea and A0 were computed with Borchardt-Daniels method[3]. Let the nearby peak reaction temperature Tpeak of 180 C be the isothermal temperature, cure the samples in the DSC furnace at this temperature for different lengths of time, and detect the corresponding Tg(T, t) and the residual heat Hres. Table 3 lists the results inclusive of cure degrees of conversion calcu-lated with Eq.(2).

res

total( , ) 1 100%HT t

H (2)

where is the cure degree of conversion.

Table 2 Information from DSC analysis

Subject Value Subject Value Tg0/ C 19.41 Htotal/(J·g 1) 563.35 Tg / C 220.97 Ea/(kJ·g 1) 87.54

Tinitial/ C 141.40 A0/s 1 1.376×107 Tpeak/ C 173.55 n 1.602 Tfinal/ C 241.48

Table 3 Analysis of DSC curves for different lengths of

isothermal time at 180 C

Isothermal time/min Tg/ C Hres/(J·g 1) 5 / 392.1 0.304 0

10 56.60 169.6 0.394 9 20 114.33 178.5 0.651 8 30 126.50 167.2 0.683 1 40 158.53 95.12 0.831 2 60 179.23 83.40 0.852 0 80 197.93 55.0 0.885 4

Because the relationship between Tg and is inde-

pendent of the isothermal cure temperature[2], the em-pirical DiBenedetto equation (Eq.(3)) can be used to correlate the scattered points of Tg and , where is an adjustable structure-dependent parameter between 0 and 1. Fig.2 shows the relationship between Tg and through the DiBenedetto equation, where = 0.715.

Fig.2 Relationship between cure degree and temperature

using DiBenedetto equation.

No.4 Zhang Ming et al. / Chinese Journal of Aeronautics 29(2009) 449-452 · 451 ·

Eq.(3) defines the relationship between Tg and . The relationship between and (T, t) can be expressed by Eq.(4). Then the relationship between Tg and (T, t) can be obtained by combining Eq.(4) and Eq.(3). Ac-cording to the definition of glass transition temperature, let Tg = T in the new equation (Eq.(5)), a correspond-ing time tvitrification can be solved. Plot a curve of T ver-sus tvitrification, a sigmoid vitrification line can be ulti-mately diagramed as shown in Fig.3.

g g0g g0

( )1 (1 )T T

T T (3)

11a

01 1 (1 ) expnEn A t

RT (4)

g g0

11a

g g0 0

11a

0

( ) 1 1 (1 ) exp

1 (1 ) 1 1 (1 ) exp

n

n

T T

ET T n A tRT

En A tRT

(5)

Fig.3 TTT diagram of the epoxy resin.

3.3. TTT diagram and phase structure control

On the basis of gelation model (Eq.(1)) and vitrifi-cation model (Eq.(4)), the TTT diagram is derived from the contours of the logarithmic time for gelation and to vitrification as a function of the reaction tem-perature (Fig.3). The intersection of gelation line and vitrification line, gelTg, is found to be 70.2 C, meaning the gelation time of 134.7 h, relatively long compared to the manufacture process circle. Isoconversion curves have been determined through numerical inte-gration of the kinetic model accounting differential coefficient. Isoconversion curves for = 0.30, 0.45 and 0.70 are plotted. The isoconversion curve for

= 0.45 results from fitting the scattered gelation points.

This diagram shows the different stages of the cure process. Cross-linking starts from gelation line to the vitrification, during which a transition line to a glassy solid occurs. In the vicinity of vitrification, the seg-mental mobility of the macromolecule chains de-creases, and the overall rate of reaction might be con-trolled by the limited diffusion of reaction species[4-7]. Thereafter, the reaction goes on quite slowly. Cross- linking continues at a tardy rate until vitrification. The analysis suggests that the nearby area of gelation line can provide possibilities to control the viscosity and diffusion of the resin system so as to form the specific phase structure. The traditional cure processing— 130 C/0.5 h+coating PAEK atop the resin+ 180 C/2 h +200 C/2 h—is modified into 130 C/ 1 h+ coating PAEK atop the resin+ 180 C/2 h+200 C/2 h (see Fig.4). The pre-gelation time is prolonged in the pre-venient cure procedure to further restrict the diffusion and flow of the solution. The morphology of the resin casts cured with different processings is investigated by means of SEM.

Fig.4 Comparison of the modified cure process and tradi-

tional cure process.

From the SEM micrographs (see Fig.5), it can be observed two distinct phase structures stemmed from traditional and proposed cure processings, respectively. The former is of a typical toughened thermoset phase state with the fracture surface having many small en-ergy-absorbing parts akin to the “in-situ” toughened phase structure. This is because the PAEK film wholly melted into the resin system is equivalent to addition of a toughening agent into it. When the resin is kept in

· 452 · Zhang Ming et al. / Chinese Journal of Aeronautics 22(2009) 449-452 No.4

Fig.5 Phase morphology comparison of laminated resin structure after different cure processings.

isothermal state for as long as 60 min to approach the gelation line, not only the liquid fails to solve the en-tire thermoplastic because of the higher viscosity caused by further cross-linking; but also the phase reversion induced by chemical reaction forms a co-continuous phase structure[8-10]. After the system has cured, a transitional surface is also left behind be-tween the toughened epoxy and pure epoxy. This is identified to be the “ex-situ” phase structure by a lot of previous works [11-14].

4. Conclusions

A TTT diagram has been derived for a 2/4 func-tional epoxy blend system (E-54/AG-80) cured with DDS. Characteristic temperatures such as Tg0, gelTg, and Tg are determined. Mathematical gelation and vitrifying models are established with experimental data from DMA and DSC respectively. A new cure processing is derived from the TTT diagram by pro-longing the pre-gelation time to restrict the diffusion and flow of the solution. The final phase structure is identified through SEM micrographs to be the “ex-situ” phase morphology as distinct from the tradi-tional “in-situ” phase structure.

References

[1] Gao J W, Shen K, Gao Z M. The cure behavior of tetra- glycidyl diaminodiphenyl methane with diaminodi-phenyl sulfone. Thermochimica Acta 2000; 352-353: 153-158.

[2] N ñez L, Fraga F, Castro A, et al. TTT cure diagram for an epoxy system diglycidyl ether of bisphenol A/1,2 diamine cyclphexane/calcium carbonate filler. Polymer 2001; 42(8): 3581-3587.

[3] Rosu D, Must t F, Cascaval C N. Investigation of the curing reactions of some multifunctional epoxy resins

using differential scanning calorimetry. Thermo-chimica Acta 2001; 370(1-2): 105-110.

[4] Hseih H K, Su C C, Woo E M. Cure kinetics and in-ter-domain etherification in an amine cured phenoxy/ epoxy system. Polymer 1998; 39(11): 2175-2183.

[5] López J, López-Bueno I, Nogueira P, et al. Effect of poly (styrene-co-acrylonitrile) on the curing of an epo- xy/amine resin. Polymer 2001; 42(4): 1669-1677.

[6] van Assche G, van Hemelrijck A, Rahier H, et al. Modulated temperature differential scanning calo-rimetry: cure, vitrification, and devitrication of ther-mosetting systems. Thermochimica Acta 1997; 304- 305: 317-334.

[7] van Assche G, Swier S, van Mele B. Modeling and ex-perimental verification of the kinetics of reacting poly-mer systems. Thermochimica Acta 2002; 388: 327-341.

[8] Inoue T. Reaction-induced phase decomposition in polymer blends. Progress in Polymer Science 1995; 20: 119-153.

[9] Girard-Reydet E, Sautereau H, Pascault J P, et al. Re-action-induced phase separation mechanisms in modi-fied thermosets. Polymer 1998; 39(11): 2269-2280.

[10] Oyangurn P A, Galante M J, Andromaque K, et al. Development of bicontinuous morphologies in poly-sulfone-epoxy blends. Polymer 1999; 40(19): 5249- 5255.

[11] Yi X S, An X F. Effect of interleaf sequence on impact damage and residual strength in a graphite/epoxy laminate. Journal of Material Science 2004; 39(9): 3253-3255.

[12] Yi X S, An X F, Tang B M, et al. Ex-situ formation of periodic interlayer structure to improve significantly the impact damage resistance of carbon laminates. Ad-vanced Engineering Materials 2003; 5(10): 729-732.

[13] An X F, Ji S Y, Tang B M, et al. Toughness improve-ment of carbon laminates by periodic interleaving thin thermoplastic films. Journal of Material Science Let-ters 2002; 21(22): 1763-1765.

[14] Long W, Xu Y H, Yi X S, et al. Preliminary study on resin transfer molding of high-toughnessed graphite laminates by ex-situ method. Journal of Material Sci-ence 2004; 39(6): 2263-2266.

Biography:

Zhang Ming As a Ph.D. candidate, he is an assistant engi-neer at the National Key Laboratory of Advanced Composites Materials, Beijing Institute of Aeronautical Materials. He received his B.S. degree from Beijing University of Aero-nautics and Astronautics in 2003. His research interests lie in thermosetting resin matrix composites, thermoplastic resin composites, RTM, RFI and other liquid composite molding process. As the first author, he has published eight academic papers. E-mail: zhangming621@gmail.com