UV stepwise cured fabrication of glass fiber/acrylate composites: Effects of exposure dose on curing uniformity and interlaminar shear strength

Abstract

To determine the quality of composites cured with UV light ply by ply, the effects of exposure dose on the cure uniformity and interlaminar shear strength of composite laminates were investigated. This was done by using UV power meter, differential scanning calorimetry, and the fiber optic strain sensors. According to differential photocalorimety results, the accumulative curing effect is remarkable in the stepwise UV-curing process, which results in gradual improvement of the degree of cure of the underlying ply. The cure uniformity increased with increasing exposure dose, but higher exposure dose led to a decrease of the interlaminar shear strength. This was due to increased shrinking strain and residual stress according to the results from the fiber optic strain sensors.

Keywords

UV curing composite cure uniformity shrinking strain interlaminar shear strength

Introduction

UV curing is a promising technology because it presents a number of unique advantages, like fast cure speed, solvent-free formulations, and room temperature operations.^1,2 In recent years, UV-curing technology was introduced in composites manufacturing and has attracted more and more attention.^3,4 Studies such as Yuan et al.⁵ fabricated glass fiber reinforced polyester by vacuum bag infusion and UV curing, the results showed that UV curing was a cost-effective process that did not compromise mechanical properties. Li et al.^6,7 developed a method to repair damaged composite laminates by using UV-curing technology. The results showed that the residual tensile strength of UV-curing resin repaired samples was 1.08 times that of unrepaired samples after low velocity impact test. The research from Compston et al.⁸ highlights the potential of UV-curing composites as providing excellent mechanical performance while reducing manufacturing time and improving the manufacturing environments. Nowadays the UV-curing process has been applied in fabricating and repairing wind turbine blades and manufacturing large watercrafts.^9–11

In the UV-curing process, the UV light intensity, the composite formulation (resin type, fiber volume fraction), and the laminate thickness all have important influence on the curing process.^12,13 Thicker composite laminates need higher UV exposure dose to obtain a full cure throughout the laminate, which normally results in yellowing and degradation of the laminates’ surface.¹⁴ However, the thicker laminates still cannot get fully cured because of the lower penetrated UV dose, when the time of UV exposure was 60 min. This had an obvious impact on the capacity for the adhesive to transfer load.¹⁵ Moreover, the traditional one-shot UV-curing process required longer curing period to fully penetrate and cure a limited thickness (up to 13 mm) composites structure,⁵ thereby reducing the production efficiency.¹⁶ In order to solve these problems, a stepwise fabrication process combined automated fiber placement technology with UV in situ curing process which was researched by Abulizi and Yugang¹⁷ and Di Pietro and Compston.^18,19 And a cost-effective in situ UV-curing process for polymer–matrix composites was developed and the mechanical properties of the composite were tested by Yugang et al.²⁰ All the above studies indicated that stepwise curing process is an efficient and low-cost way to fabricate composites. The exposure dose in the UV stepwise curing process is important to the cure uniformity and the performances of composites cured layer by layer, but the relation between the exposure dose and the curing uniformity is seldom reported. In this paper, the effects of exposure dose on the uniformity of cure and the interlaminar shear strength (ILSS) of UV stepwise curing composite laminates are investigated to get high-quality UV-cured composites on the basis of early research.²¹

Experimental

Materials

The UV prepreg was made by solution impregnation method. The resin content of the prepreg was about 36 ± 3% that measured according to ISO 11667:1997.²² The E-glass fiber of UV prepreg was produced by Nanjing Fiberglass R&D Institute of China. The density of fiber glass was 2.5–2.7 g/m³, the tensile strength was 3140 MPa, and the Young’s modulus was approx. 69–76 GPa. The UV-curable resin was prepared according to the following formula (wt%): epoxy acrylate CN104NS 70%, 1,6-hexanedioldiacrylate (HDDA) 22%, trimethylolpropane triacrylate (TMPTA) 5%, and photo-initiator 651 (Benzil Dimethyl Ketal) 3%. The CN104NS was supplied by SATOMAR, USA. The HDDA and TMPTA were purchased from Taiwan External Chemical Industry Co., Ltd. The photo-initiator was purchased from Jingjiang Hongtai Chemical Engineering Co., Ltd of China.

UV transmission measurement

The method used for UV transmission measurement is shown in Figure 1. The FieldMate power meter (Coherent Inc., USA) was covered with a quartz glass slice (thickness: 0.5 mm, 365 nm UV transmittance: 95 ± 1%), and the prepreg was placed on the slide and was irradiated layer by layer using the UV-LED light. The transmitted light power was measured by the power meter. And the transmission coefficient T_c was calculated by equation (1)

T_{c} = P_{t} / P_{i}

(1)

where P_t is the transmitted UV power and P_i is the incident UV power.

Figure 1.

Schematic for measurement of the transmitted UV light power.

Fabrication of laminate

Figure 2 shows the principles of the UV stepwise curing fabrication process. According to ASTM 2344, the thickness of laminate is 6 ± 0.3 mm, which required 30 plies prepreg. The prepregs were pressed on the mold by the compacting roller and were irradiated by the UV at the same time. The exposure dose was calculated corresponding to the speed of the tape placement head. The relationship between the speed and the exposure dose is determined by equation (2)

E = Hb / V_{s}

(2)

where H is the irradiance of UV-LED light, its value is 2.5 ± 0.1 mW/mm²; b is the width (mm) of UV irradiation area, which is parallel to the moving direction of tape placement head; and V_s is the speed of the tape placement head. The compacting force was 200 ± 10 N (the compaction cylinder diameter is 40 mm, and the air pressure was about 0.16 MPa). The UV-LED light source was provided by UVATA China: the wavelength is 365 nm, the maximum output power is approx. 10.5 mW/mm², and the UV emitting area is 8 mm(width) × 50 mm(length). The UV-LED was placed as close as possible to the curing area, so the emitting area is almost as same as the irradiation area.

Figure 2.

Principle of UV stepwise curing process.

Tests of degree of cure (DoC)

The DoC of composite samples was tested by differential scanning calorimetry (DSC). The temperature range of DSC test was from 25 to 400℃, the heating rate was 10℃/min. The curing behavior of the prepreg irradiated by small UV exposure doses was studied by the differential photocalorimetry (DPC). The DSC with DPC function (model: DSC1) was made by METTLER TOLEDO. The DoC was calculated by equation (3)

C_{r} = 100 (h_{max} - h_{i}) / H_{max} %

(3)

where C_r is the DoC, h_max is the full exothermic energy of an uncured prepreg sample, and h_i is the residual exothermic energy of a part cured laminate sample at sampling point i. The h_max and h_i are all tested by DSC.

Tests of the cure uniformity

The DoC of samples along the depth in the laminate was tested to evaluate the curing uniformity. Figure 3 shows the location of tested points in the laminate. The composite laminate was machined by a milling machine to get samples at different depths. Three samples from each depth were taken and the DoC of the samples was tested via DSC. And the curing uniformity of a laminate is calculated according to equation (4)

η = 1 - (C_{max} - C_{min})

(4)

where η denotes the curing uniformity of a laminate, and C_max and C_min are the maximum DoC and the minimum DoC among the four tested points in the laminate.

Figure 3.

The schematic of sampling point for DSC.

Test of the ILSS

The ILSS was tested by a short-beam three-point bend fixture at 1.0 mm/min on the INSTRON 1195 according to ASTM D2344. The sample dimension was 40(length) × 12(width) × 6(thickness) mm. The ILSS (τ) was calculated using equation (5)

τ = 3 P / 4 wt

(5)

where P is the maximum load, w is the width of specimen, and t is the thickness of the specimen. Five samples were taken in the same composite laminates and the average results were reported for every test point.

Test of shrinking strain

The micro strain of stepwise curing was tested by the fiber optic strain sensors. The sensors were embedded between cured and uncured prepreg with the angle perpendicular and parallel to the fibers (as shown in Figure 4). The test method was as follows: first, the prepreg was laid up on the mold and was cured with UV. Then the fiber optic strain sensors were fixed on it along two directions. Next, the second layer prepreg was placed and compacted by compaction roller without curing. Finally, when the second layer prepreg was cured without compaction, the shrinking strains were detected by the sensors. Fiber optic strain sensor was made by FISO Technologies Inc. (Canada) with range from −5000 to 5000 μɛ. The individual samples were made in order to test the shrink strain of the laminates irradiated by different UV exposure dose.

Figure 4.

The schematic of embedded fiber optic strain sensor in the laminate.

Results and discussion

The accumulative curing effect

Figure 5 shows the DoC variation of the bottom layer (the firstly placed layer) increasing with each subsequent ply at the 0.83 mJ/mm² level of exposure dose. The reason could be explained by Figure 6, by which the change of transmission coefficient versus ply number by incident exposure dose of 0.83 mJ/mm² is shown. It can be noted in Figure 6 that the UV light could penetrate five layers maximum, and the average transmission coefficient decreases exponentially with increasing of penetrated layers, the average transmission coefficient of the first placed layer is about 0.445, and the average transmission coefficient of total five layers under the top layer is about 0.047. This means the first placed layer still receive approx. 0.039 mJ/mm² exposure dose (0.83 × 0.047) when the exposure dose on the fifth placed layer is 0.83 mJ/mm², which could improve the DoC of the first (bottom) layer, and it is the same for every five layers, this conclusion could be verified from Figure 7.

Figure 5.

The accumulative DoC curves of the bottom layer.

Figure 6.

Relationship of transmission coefficient and layers.

Figure 7.

The DPC curves of a multiple step irradiation.

Figure 7 shows the stepwise UV-curing reaction curves from DPC. It can be seen from Figure 7 that when the prepreg sample is first irradiated by 0.83 mJ/mm² of UV light and then by 0.031 mJ/mm² (the minimum UV dose of the DPC) for four times, the curing reaction heat of the sample is 191.96, 22.66, 16.29, 10.22, and 3.32 J/g, respectively. This is because the prepreg was not fully cured under the first irradiation of UV, so when the next step UV irradiated on the prepreg sample, the reaction heat was detected by DPC once again. These results indicated that even tiny UV exposure dose could still induce the curing reaction of the prepreg. Moreover, the h_max (the full exothermic energy of an uncured prepreg sample) was about 264.52 J/g, and the total exothermic energy of the five irradiations was 244.45 J/g. These results meant if the sample got more UV irradiation it still would produce more reaction heat. But the UV just penetrated five plies in this study, so a higher UV dose was needed to trigger more reaction heat, which resulted in a higher DoC. It also can be noted that with the increase of the irradiation times, the heat from curing reaction decreased gradually, that is to say, the DoC of the UV prepreg improved stepwise. From the discussion above, during the UV stepwise curing process, the UV light can penetrate the layers until the fifth layer below, which could improve the DoC of the UV prepreg layers step by step gradually.

Effect of the exposure dose on curing uniformity

Figure 8 shows the DoC difference of the four test points for different exposures. It can be seen from Figure 8 that the DoC difference between the maximum value and minimum value of the four tested points decreases with the increasing exposure dose. This indicates that the curing uniformity is improved gradually with the increasing of exposure dose. That means the more exposure dose, the more uniform of the DoC. At 4.00 mJ/mm² level of exposure doses, the average DoC was 92.33%, and the curing uniformity was 96.46%.

Figure 8.

The DoC plot of the four tested points of different exposure doses.

It can be seen from Figure 8 that the maximum DoC appears in the test point 4, the DoC in the point 3, point 2, and point 1 gradually decreases. The reason was that the point 4 (bottom surface) was irradiated more times than the other points because of the accumulative curing effect.

Effect of the exposure dose on ILSS

Figure 9 shows the variation of the ILSS with different UV exposure doses. It can be seen that the ILSS of UV stepwise curing laminate increases first, then deceases with the increasing of the exposure dose. When the exposure dose is 0.83 mJ/mm², the maximum ILSS of the laminate is 30.26 MPa.

Figure 9.

Correlation between the ILSS and exposure dose.

Figure 10(a) and (b) shows the strain variation of perpendicular and parallel to the fiber direction, respectively. It can be seen from Figure 10 that the curves have similar variation in both directions. There are three stages in the curing process. In the initial stage, the UV irradiates on the prepreg layers and activates the polymerization, and the reaction heat results in thermal expansion of the prepreg. Positive strains were detected during the stage, which results in the convex portion of the curves in Figure 10. After that, the matrix resin started to shrink because of the polymerization, which is relatively high resulting in negative strain. Finally, the resin achieves the gel state and the residual strain stabilized gradually. From Figure 10, the maximum strain is approx. 63, 98, and 552 μɛ when exposure dose is 0.42, 0.83, and 4.00 mJ/mm², respectively. These results indicate that the more exposure dose was irradiated on the prepreg, the more shrinking strain was introduced. It is proposed that the difference of DoC between the upper layer and lower layer became larger with the exposure dose increasing, when the upper layer began to cure, the lower layer with higher curing degree would restrict the cure shrinkage of curing layer, which induced residual strain. The more the exposure dose, the “restriction effect” is more obvious, accordingly the more the residual strain. This would lead to decrease of the ILSS. However, according to the discussion in “Effect of the exposure dose on curing uniformity” section, the high exposure dose is beneficial to the curing uniformity, so a suitable exposure dose should be considered for a balance between ILSS and curing uniformity.

Figure 10.

The curves of strain in the directions of perpendicular (a) and parallel (b) to the fibers with different exposure doses.

Conclusions

In the UV stepwise curing process, the accumulative curing effect is remarkable due to the transmission behavior of the UV in the UV curable prepreg. This makes stepwise curing process an energy saving method to fabricate thicker composite laminate. The UV exposure dose has an important effect on the properties of the laminate. With the increasing of the UV dose, the curing uniformity of laminate is improved, but the ILSS decreased because of large shrinking strain. So a suitable exposure dose should be considered for a balance between ILSS and curing uniformity, which would improve the quality of UV-curing composite laminates with lower fabrication stress.

Footnotes

Funding

This research was supported by NSFC of China (51275393), Program for New Century Excellent Talents in University (NCET-11-0419), and Fundamental Research Funds for the Central Universities and Specialized Research Fund for the Doctoral Program of Higher Education (20120201110031).

Conflict of interest

None declared.

References

Decker

. The use of UV irradiation in polymerization. Polym Int 1998; 51: 133–141.

Decker

Moussa

. UV-curable acrylic resins for production of glass laminates. J Appl Polym Sci 1995; 55: 359–369.

Nwabuzor

Okol

. Preliminary assessment of the ultra violet curing of composites manufactured by the resin infusion between double flexible tooling process. Polym Compos 2006; 27: 417–424.

Xian

. Water absorption and hydrothermal ageing of ultraviolet cured glass-giber reinforced acrylate composites. Polym Compos 2012; 33: 1120–1128.

Yuan

Yang

Mai

. Ultraviolet curing of glass fibre reinforced polyester composites. Adv Compos Lett 2000; 9: 341–346.

Maricherla

Singh

. Effect of fiber orientation on the structural behavior of FRP wrapped concrete cylinders. Compos Struct 2006; 74: 475–483.

Pourmohamadian

Cygan

. Fast repair of laminated beams using UV curing composites. Compos Struct 2003; 60: 73–81.

Compston

Schiemer

Cvetanovska

. Mechanical properties and styrene emission levels of a UV-cured glass-fibre/vinylester composite. Compos Struct 2008; 86: 22–26.

Auscomposites, Auspreg UV for wind energy products. http://www.auscomposites.com.au/products_wind_energy.htm (2010, accessed 5 July 2014).

10.

Livesay MA. UV-VARTM fabrication of low cost composite structures. In: The proceedings of 44th international SAMPE symposium and exhibition, Long Beach, CA, 23--27 May 1999, pp.1338–1344. Covina, CA: SAMPE.

11.

Marsh

. Meeting the challenge of wind turbine blade repair. Reinf Plast 2011; 55: 32–36.

12.

Fan

Schumacher

Kristy

. Curing-light intensity and depth of cure of resin-based composites tested according to international standards. JADA 2002; 133: 429–434.

13.

Endruweit

Ruijter

Johnson

. Transmission of ultraviolet light through reinforcement fabrics and its effect on ultraviolet curing of composite laminates. Polym Compos 2008; 29: 818–829.

14.

Glauser

Johansson

Hult

. A comparison of radiation and thermal curing of thick composites. Macromol Mater Eng 2000; 274: 25–30.

15.

Peck

Jones

Pang

. UV-cured FRP joint thickness effect on coupled composite pipes. Compos Struct 2007; 80: 290–297.

16.

Endruweit

Johnson

Long

. Curing of composite components by ultraviolet radiation: a review. Polym Compos 2006; 27: 119–128.

17.

Abulizi D and Yugang D. A new method for glass-fiber reinforced composites manufacturing automated fiber placement with in-situ UV curing. In: IEEE international symposium on assembly and manufacturing, Tampere, Finland, 2011, pp.1–13. New York: IEEE.

18.

Di Pietro

Compston

. Automated fibre placement with in situ ultra-violet light curing concept testing. In: Wellnitz J, Subic A and Leary M (eds) Sustainable automotive technologies, Berlin Heidelberg: Springer, 2010, pp. 79–85.

19.

Di Pietro

Compston

. Shear properties of a glass-fibre vinylester composite cured with high intensity ultraviolet light for use in automated manufacturing processes. Adv Compos Lett 2011; 20: 45–48.

20.

Yugang

Wang

Tang

. Fabrication and mechanical properties of UV-curable glass fiber-reinforced polymer-matrix composite. J Compos Mater 2010; 45: 565–572.

21.

Yugang

Jia

Weihong

. Effect of compaction and UV exposure on performance of acrylate-glass-fiber composites cured layer by layer. J Appl Polym Sci 2012; 123: 3799–3805.

22.

ISO 11667:1997. Fibre-reinforced plastics – moulding compounds and prepregs – determination of resin, reinforced-fibre and mineral-filler content – dissolution methods. Switzerland, PA: ISO.