Evaluation of interfacial properties of the silane blend sized glass fiber

Abstract

The performance of a composite material system depends critically on the interfacial characteristics of the reinforcement and the matrix material. In this study, the interfacial adhesion was tailored by the creation of textures on the glass fiber surface using inorganic-organic silane blends. A single-fiber microdroplet test was conducted to assess the interfacial properties between the textured glass surface and an epoxy matrix. The load–displacement curves from microdroplet tests were analyzed. The stress-based and energy-based micromechanic models of interfacial debonding and corresponding adhesional parameters (apparent and ultimate interfacial shear strength, friction stress, critical energy release rate, work of adhesion, and adhesional pressure) were applied for theoretical calculations. The results showed a clear trend for the impact of different silane blends on the interfacial properties. The specimens containing 75:25 and 50:50 of inorganic–organic silane blends show the most effective improvement in the interfacial adhesion properties between glass fiber and epoxy resin. Scanning electron microscopy was used to visualize the failure surface of the specimen after the microdroplet test. The scanning electron microscopic images indicated that the failure in the blend sized treated glass fiber–epoxy matrix specimen runs predominantly along the interphase and combines both cohesive failure in resin (the presence of some resin fragments) and adhesive failure (some bare fiber surfaces can be seen).

Keywords

Glass fibers interphase interfacial adhesion micromechanic models microdroplet test

Introduction

Continuous glass fiber–epoxy composites have widely used for structural applications due to their excellent mechanical properties and relatively good price-to-performance ratio.¹ The performance of fiber-reinforced composites is strongly influenced by the functionality of composite interphase.^2–4 In case of glass fibers, applying appropriate sizing on fiber surface serves to increase the interaction between fiber and resin and protect the fibers from damage during weaving as well as environmental degradation.⁵ Therefore, sizing, i.e. functional coating is tailored to improve the transfer of stress from the matrix to the fiber reinforcement by enhancing fiber wettability, adhesion, compatibility, etc.⁶ The sizing applied to the glass fibers consists of a silane-based network that chemically binds to the fiber and other compounds that are adsorbed onto the glass surface.⁶

The role of silane coupling agents in increasing the fiber–matrix interfacial adhesion is well documented in various literatures^5,7–9 but research related to influence of mechanical interlocking in optimizing the fiber–matrix interfacial properties is limited. Also, as far as the author is aware, no work has been undertaken regarding the quantitative description of interfacial properties via micromechanical models, in order to evaluate the influence of silane blend textured glass fiber surface on interfacial properties of glass fiber–epoxy resin.

In the present work, glass fiber surface was textured using sol–gel synthesis of multiple alkoxy silanes. Using blends of tri-alkoxy silane coupling agents employed in traditional sizing systems (glycidoxypropyltrimethoxy silane, GPS) and tetraethoxysilane (TEOS), a variety of in situ textured surfaces were created on the glass fibers surface. GPS is very widely used since it has epoxide functional group on the silane structure that can react with the amine curing agent in the epoxy-amine resin and enhance the adhesion between glass fiber and epoxy resin in the composite.^10,11 The TEOS, as a precursor of silica nanoparticle, is used to create the mechanical interlocking between fiber and resin through forming in situ islands on the glass fiber surface.^12–14

The interfacial characterization of glass fiber–epoxy system was determined experimentally at the microscale using the single fiber microdroplet test. Then, the load–displacement curves from microdroplet test were analyzed using one-dimensional shear-lag analysis. Micromechanical models (stress- and energy-based models) have been used to determine the interfacial parameters such as apparent and ultimate interfacial shear strength, friction stress, critical energy release rate, work of adhesion, and adhesional pressure. One of the main results of this work was that the structure of the silane interphase on glass fibers formed from blends of silanes can be tailored by the choice of the ratio of silanes in such a blend.

Experimental

Materials

Continuous E-glass fiber bundles containing approximately 2500 filaments with an average individual fiber diameter of 20 µm were supplied by Taishan, China. The silanes used were 3-glycidoxypropyltrimethoxysilane (GPS) and tetraethoxysilane (TEOS) from Silicone New Material Co., China. Ethanol and deionized water used for hydrolyzing the silane provided from Merck. Commercial hydrochloric acid (HCl, 37%) and acetic acid (CH₃COOH, 100%) were of analytical grade purchased from Merck and also used without further purification. The epoxy resin used in this study was Araldite LY-564, and the curing agent was Aradur HY 560 (Huntsman, Switzerland).

Methods

Surface treatment of glass fibers

As-received glass fibers were heat-cleaned at 450℃ for 1.5 h to remove sizing and organic impurities from the glass fiber surface. The silane blend solution was prepared by hydrolyzing the silanes for an hour at room temperature in 75/25 (v/v) solutions of ethanol and deionized water with a pH adjusted to 4.0 using acetic acid. The ratios of GPS and TEOS silanes were varied according to Table 1.

Table 1.

Blend ratios of silanes.

Sample code	GPS content (%)	GPS: TEOS ratio
A	1	100:0
B	1	75:25
C	1	50:50
D	1	25:75

All of the silane blends designed to treat the glass surface contains the same amount of compatible silane (GPS) to ensure proper chemical bonding between the fiber and resin. Once hydrolyzed, a tow of E-glass fibers immersed in the designed silane solutions for 15 min. The tow was removed and allowed to air-dry overnight at room temperature, followed by oven drying for 1 h at 100℃.

Single fiber microdroplet pull-out test

The single fiber microdroplet test was selected to quantify the interfacial properties of the different silane solution treated fibers using an epoxy matrix drop. Figure 1 shows a schematic presentation of a specimen making. First, E-glass fibers were stretched and fixed on a thin paper frame with a central longitudinal slot of fixed gage length. Then a small droplet of resin was formed on the E-glass fiber. Once deposited onto the glass fiber, the epoxy resin droplet was allowed to gel at room temperature for 24 h followed by a post cure at 80℃ for 2 h and 110℃ for another 2 h. Before the test, each specimen was studied under an optical microscope to make sure they were within the acceptable size range and relatively symmetric. Also, the fiber diameter measured.

After selecting the symmetrical droplets by optical microscope, the paper frame with the E-glass fiber and epoxy microdroplet were attached to a load cell by a hook (Zwick universal testing machine-1446 60, load cell 20 N). The paper frame was cut away before testing. A vernier caliper was used to grip the epoxy droplet. The gap between vernier caliper’s knives was adjusted so that the fiber, but not the droplet, moved between them. Tests performed at a speed of 0.5 mm/s. The microdroplet test setup and schematic of the test procedure is shown in Figure 2.

Micromechanical analysis

Stress-based and energy-based models are used to describe stress distribution and interfacial failure in the fiber–matrix system. Using a simple analysis, that assumes a uniformly distributed state of pure shear, the interfacial shear strength of the microcomposite can be determined as given by the following equation^15,16

τ_{app} = \frac{F_{max}}{π d_{f} l_{e}}

(1)

where

F_{max}

is the debond force, d_f is the fiber diameter, l_e is the embedded length, and

τ_{app}

is the apparent interfacial shear strength. However,

τ_{app}

is strongly affected by specimen geometry, and cannot characterize the interfacial bond strength adequately.¹⁷ In fact, the shear stress distributions actually vary along the interface (due to the end conditions and the loading). Therefore, a local (ultimate) interfacial shear strength,

τ_{ult}

, may be obtained using the following equation¹⁸

τ_{ult} = \frac{d F}{π d_{f} d l_{e}} = [\frac{τ_{app} l_{e} β}{\tanh (β l_{e})} + E_{f} \frac{d_{f}}{4} β (α_{m} - α_{f}) Δ T \tanh (\frac{β l_{e}}{2})]

(2)

Variables α_m and α_f are the coefficients of thermal expansion of the matrix and fiber ( $α_{m} = 72.5 \times 10 - 6 / \circ K$ , $α_{f} = 4.7 \times 10 - 6 / \circ K$ ), respectively. ΔT is the variation of temperature ( $Δ T = T_{free} - T_{test}$ ), d_f is the fiber diameter, and E_f is the fiber Young’s modulus. β is the shear-lag parameter that can be calculated using Cox’s shear-lag analysis by the following equation

β 2 = \frac{8 G_{m}}{E_{f} d_{f}^{2} \ln (2 \frac{r_{m}}{d_{f}})}

(3)

where r_m is the droplet radius, and G_m is the matrix shear modulus.

After debonding, the friction is governed by the friction coefficient μ and the interfacial pressure P. For glass fiber and a thermoset matrix, Piggott and Xiong¹⁹ estimated the value of the μ to be between 0.35 and 1.8. This coefficient represents the physical and mechanical interactions between fiber and matrix. Moreover, the interfacial pressure P originates from the matrix shrinkage during crosslinking, and the residual stresses developed during cooling. This parameter is given by the following equation²⁰

P = \frac{(1 + ν_{f LT}) (α_{m} - α_{f}) Δ T}{(\frac{1 - ν_{fT} - 2 ν_{fL}^{2}}{E_{fT}} + \frac{1 + ν_{m}}{E_{m}})}

(4)

where ν, α, and E are Poisson’s coefficient, the thermal expansion coefficient, and Young’s modulus, respectively (the subscripts f, m, L, and T denote the fiber, matrix, longitudinal, and transverse, respectively). ΔT is the difference between test temperature and stress-free temperature. The simple linear expression linking friction stress, friction coefficient μ, and the interfacial pressure P often used is

τ_{fric} = μ . P

(5)

The true value of $τ_{fric}$ is related to the value of $τ_{app}$ by equation (6) ²¹

τ_{fric} = \frac{τ_{app} E_{f} d_{f}}{E_{f} d_{f} + 4 τ_{app} (1 + 2 l_{e})}

(6)

In the energy-based models were assumed that the debonding zone extends when the energy release rate, G, reaches its critical value, G_ic. Thus, G_ic is the specific interfacial parameter in this approach, and can be estimated by equations (7) to (9)^22,23

G_{ic} = \frac{d_{f}}{4} [C_{33 S} τ_{app}^{2} + 2 τ_{app} D_{3 S} Δ T + \frac{(D_{3 S} Δ T) 2}{C_{33 S}}]

(7)

with C_{33 S} = \frac{1}{2} (\frac{1}{E_{f}} + \frac{ν_{f LT}}{ν_{m} E_{m}})

(8)

and D_{3 S} = \frac{1}{2} (α_{f} - α_{m})

(9)

where ν is the Poisson’s coefficient (glass fiber ν_f = 0.22; epoxy ν_m = 0.35), E is the Young’s modulus, and α is the coefficient of thermal expansion with f for fiber, m for matrix, L for longitudinal, and T for transverse. ΔT is the temperature range as defined earlier. The interfacial shear strength

τ_{app}

was determined by equation (1).

In the failure of a fiber–matrix adhesional contact, external forces normal to the interface do a work equal to the work of adhesion, W_A, to produce debonding. A simple connection between W_A and $τ_{ult}$ has been postulated as follows²⁴

τ_{ult} = (2 \times 109) \sqrt{\frac{E_{m}}{E_{f}}} . W_{A}

(10)

This work is done against the forces of adhesive pressure, which are normally directed with respect to the interface and equal $σ_{ult} = τ_{ult} (E_{f} / E_{m}) 1 / 2$ . Thus, $λ = W_{A} / σ_{ult}$ is the distance at which the work against adhesion forces is done.^24,25 However, to pull a fiber out of a matrix, considerably larger work than W_A should be done, namely G_ic. Using G_ic instead of W_A in the above equation, the interphase thickness (d_i) can be determined as follows²⁶

G_{ic} = d_{i} . σ_{ult}

(11)

Contact angle measurement

Fiber–matrix contact angles were also evaluated on the micro debonding specimens following an approach proposed by Carroll.²⁷ As schematically represented in Figure 3, it is supposed that the drop assumes a symmetric elliptical shape around the fiber.

The shape of the fiber is governed by the Laplace equations (12a) and (12b), neglecting the effect of gravity force²⁷

- \frac{d r}{d z} = \frac{[(r_{m}^{2} - r 2) (r 2 - a 2 r_{f}^{2})] \frac{1}{2}}{r 2 + {ar}_{f} r_{m}}

(12a)

a = \frac{r_{m} \cos θ - r_{f}}{r_{m} - r_{f} \cos θ}

(12b)

where r_f is the fiber radius, r_m is the maximum radius of the drop, θ is the contact angle, and r and z represent the radial and the longitudinal directions, respectively. By measuring r_f and r_m with an optical microscope, and integrating along the length L with an iterative procedure, the fiber–matrix contact angle θ can be evaluated on the basis of equations (12a) and (12b). A minimum of five microdroplets analyzed for each sample.

SEM examination

Surface properties of treated glass fibers and fracture surfaces were characterized by scanning electron microscope (SEM). The samples were sputter-coated with a thin layer of gold in an automatic sputter coater (KYKY-SBC 12, China) and observed with a KYKY-EM3200 scanning electron microscope.

Results and discussion

Study of contact angle and aspect ratio of microdroplets

Wettability can affect the interfacial adhesion between fiber and matrix; better wetting can enhance the interfacial shear strength by improving the work of adhesion. The silane blend of TEOS and GPS is used in this study to treat the glass fiber. So, it is nice to gain some insights into how the addition of TEOS in the coating affects the wettability of the epoxy resin to the coated glass. For this purpose, fiber-epoxy drop contact angles were determined by optical measurements of the parameters involved in the Laplace equation (equations (12a) and (12b)). In Table 2, the contact angles between glass fibers and epoxy matrix, as well as aspect ratios (embedded length/diameter) are summarized. It found that the coatings on the glass fiber does not change the contact angle between the different coating coated glass and cured epoxy, significantly. Moreover, the addition of TEOS in the silane blend does not change the wetting behavior of epoxy resin. The aspect ratio shows good reproducibility between samples and correlates with the trend followed by contact angles.

Table 2.

Contact angles between glass fiber and resin and aspect ratio of microdroplets.

Sample code	GPS:TEOS ratio	Contact angles (°)	Aspect ratio (L_e/D_g₎
A	100:0	50.67 ± 4.1	1.26 ± 0.10
B	75:25	52.16 ± 3.5	1.21 ± 0.12
C	50:50	50.87 ± 2.2	1.24 ± 0.09
D	25:75	52.38 ± 2.6	1.20 ± 0.10

Microdroplet test results

During the microdroplet pull-out test, the force, and the crosshead displacement were measured, and the load–displacement plot was recorded. A typical load versus displacement curve obtained from the microdroplet pull-out test is shown in Figure 4. It can be seen that the curve shows four segments. Linear behavior corresponding to elastic energy storage is seen until the force reaches the crack initiation force (F_max) (segment A). At the second stage (segment B), the force continues to increase with the fiber-end dislocation (or with crack length). In fact, the interfacial frictional force in the debonded region is added to the total bonding force from the intact part of the interface. After an ultimate debonding force F_max is reached, the whole embedded interface fully debonds and the energy previously stored is released through fast interfacial cracking with a nearly constant friction force (segment C). From this point to complete pull out, the recorded force is from the frictional resistance between the glass fiber and the surrounding matrix (segment D).

The interfacial shear strength is expected to be dependent on the strengths of the interactions (H-bonding, covalent) between the silane blends sized glass fiber and epoxy-amine resin in the interphase region. Determination of the interphase strength allows for better understanding of the sizing effects on the fiber–matrix adhesion and the mechanical response of the interphase. In the present study, the apparent shear stress ( $τ_{app}$ ) and the ultimate shear stress ( $τ_{ult}$ ) have been determined. The results presented in Figure 5 for different silane sized glass fiber–epoxy model composites.

These results show a clear trend for the impact of different silane blends on the interfacial properties. For the unsized glass fiber–epoxy system, only a small amount of increased force is needed to overcome the physical interactions between fiber and resin for further crack propagation. As the glass fiber is sized by compatible silane such as GPS (sample A), chemical bonds are formed between glass fiber and epoxy resin. In fact, GPS silane has an epoxide functional group at one end that can react with the resin, and a methoxy group on the other end which can be hydrolyzed and react with the fiber surface. GPS will act as a bridge to connect fiber and matrix together. After the interphase partially debonded, more force is needed to break the chemical bond in the bonded region which results in an increase in the interfacial shear strength. When the glass fibers are sizing with 75:25 and 50:50 ratios of GPS: TEOS (samples B and C), the in situ texture tends to grow on the surface which can significantly change the fiber surface roughness since TEOS uses as a precursor of silica nanoparticle.²⁸ After the debonding occurs, more loads are needed not only to break the chemical bonding in the bonded region, but also to overcome the frictional effect from the debonded region. It can be concluded that the interfacial shear strength is sensitive to both chemical bonding and mechanical interlocking between glass fiber and epoxy resin.¹⁴ Figure 5 also shows that the fiber treated with the 25:75 ratios of GPS: TEOS silane blend (sample D) undergoes a drop in the interfacial shear strength when compared to the other silane blends. The large molar fraction of TEOS in the silane blend may possibly result in the formation of weak and brittle aggregates that can be sheared off relatively easily, which does not contribute to an improvement of the interfacial shear strength.

The results reveal that the ultimate strength values are significantly higher than the apparent shear strengths. The $τ_{ult}$ is a local stress corresponding to the edge of the knife loading zone and the point at which the fiber enters the droplet. It includes the development of residual thermal stresses (α and ΔT).^23,29 Figure 6 shows the typical force versus displacement for different silane treated glass fiber–epoxy system obtained from the microdroplet test.

Following the debonding process, the frictional force can be related to P (interfacial pressure) by

τ_{fric} = μ P

as before. The calculated coefficients of friction (μ) and the friction stress (

τ_{fric}

) for different samples are presented in Table 3. One-way analysis of variance (ANOVA) test, followed by a Duncan’s test, for multiple comparisons between means to determinate significant differences was used at a significance level set at 0.05, for analysis of the experimental results.

Table 3.

Study of friction after debonding.

Sample code	GPS: TEOS ratio	Friction stress, τ_fric (MPa)^a	Interfacial pressure, P (MPa)	Friction coefficient (μ)^a
Unsized fibers	0	6.14 ± 0.38^b	16.29	0.377 ± 0.03
A	100:0	9.84 ± 0.40^b	16.29	0.604 ± 0.06
B	75:25	10.68 ± 0.34^b	16.29	0.655 ± 0.04
C	50:50	12.42 ± 0.66^b	16.29	0.762 ± 0.07
D	25:75	8.78 ± 0.33^b	16.29	0.540 ± 0.03

Mean ± Std. Deviation.

The mean difference is significant at the 0.05 level.

The interfacial friction coefficient between the matrix and the glass fibers was the key parameter to determine the load needed to pull the glass fibers out after their debonding from the matrix. The calculated coefficient of friction for the microdroplet slipping along the unsized glass fiber is 0.37 while μ_A = 0.604. Moreover, the friction stress and friction coefficient increase with the addition of TEOS into the silane blend. In fact, the in situ texture was induced to grow on the fiber surface through the phase separation of silane blends with the addition of different TEOS contents into the fiber coating. The values of $τ_{fric}$ and frictional components indicate that the interface between fiber and matrix in the sample C is mechanically stronger than that of the other systems during and following pull-out experiments. The increase in frictional forces for this sample can be related to the morphology of the fiber that was roughened by the addition of TEOS into the silane blend. The rough fiber surface below the bead (leading edge) also impedes frictional sliding of the debonded microdroplet along the fiber, leading to higher values of $τ_{fric}$ . Higher degree of effective surface texturing increases the average shear stress, during the stage where relative sliding occurs between the debonded fiber and resin droplet.

Table 4 presents the adhesional parameters and interphase characteristics for epoxy resin/glass fibers treated with different ratio of silane blends. There is some discussion of whether energy-based models are better suited to analyze this type of test.¹³ It suggests that stress based models are better for initiation, and short debond lengths while energy models are preferable to describe the propagation.²³ The critical strain energy release rate, G_i_c, the work of adhesion, W_A, and the adhesional pressure,

σ_{ult}

were calculated for different silane blends coated glass fiber–epoxy samples as shown in Table 4. As seen, experimental results can conclude that the strongest adhesion bond between matrix and glass fiber exists in the samples of B and C compared with the others. Statistical analysis has been carried out by one-way analysis of variance (ANOVA) and a Duncan’s test. The results of Duncan’s test analysis are shown in Table 4.

Table 4.

Adhesional parameters and interphase characteristics for glass fiber–epoxy systems.

Sample code	GPS:TEOS ratio	G_ic (mJ/m²)	W_A (mJ/m²)	$σ_{ult} (MPa)$	λ (Å)	d (nm)
Unsized fibers	0	599.28 ± 21.10^a	168.26 ± 7.30^a	336.5 ± 13.25^a	5.0004	1.78 ± 0.07
A	100:0	748.54 ± 44.91^a	195.46 ± 15.64^a	390.9 ± 27.36^a	5.0003	1.92 ± 0.09
B	75:25	869.87 ± 49.60^a	236.99 ± 17.03^a	474.1 ± 34.06^a	4.9998	1.82 ± 0.10
C	50:50	788.51 ± 25.23^a	226.08 ± 9.95^a	452.2 ± 19.94^a	4.9995	1.74 ± 0.06
D	25:75	704.64 ± 35.79^a	199.91 ± 8.59^a	399.81 ± 19.89^a	5.0003	1.76 ± 0.05

The mean difference is significant at the 0.05 level.

G_ic: critical energy release rate for crack initiation; W_A: work of adhesion; $σ_{ult}$ : adhesional pressure; λ: range of action of intermolecular forces at the interface; d: effective thickness of the interphase.

Surface morphology of sized glass fibers

The SEM and the optical microscopy were used to observe the failure surface of the specimen after the microdroplet test. Since there is a capillary force acting on the liquid droplet ends when the resin drop is deposited on the fiber, there is always a meniscus formed at the end of the droplet (Figure 7(a)). It was reported that meniscus failure, usually, occurred before the ultimate debonding failure resulting in a small portion of the polymer bead remaining adhered to the fiber. It was observed routinely in our tests as shown in Figure 7(b). The failure modes occur in a sequence of the meniscus, then progressive debonding with local sliding followed by sliding of the entire drop over long distances.

When the shearing force exceeds the interfacial bond strength, pull out occurs, and the droplet is displaced downward along the axis of the fiber. The lack of resin on the fiber following debonding indicates adhesive failure of the interface. The only cohesive failure of the resin is shown by the small cone remaining at the top of the fiber following the pull-out experiment. Figure 8(a) shows the fracture surface of 1% GPS treated E-glass ﬁber embedded in an epoxy sample after debonding. The presence of large resin fragments on the surface of the fiber after pull-out demonstrates that a cohesive failure of the matrix occurs.

The failure surface in the GPS: TEOS (75:25) treated glass fiber–epoxy specimen after debonding is observed in Figure 8(b). There is some remainder of the resin droplet left on the fiber surface, but the shear traces in the micrograph indicate that the locus of failure has moved to the fiber–matrix interface. In the other word, the failure in the blend sized treated glass fiber–epoxy matrix specimen is a combination of both cohesive failure in resin and adhesive failure.

When the crack went through the interphase, its propagation would be hindered by the fiber surface texture that has been created through the in situ island growth from the phase separation between GPS and TEOS silane, which will force the crack propagate along a tortuous path. This tortuous failure path will help to slow down the crack propagation and absorb more energy during relative sliding of the fiber to the resin. The failure mode of these two silanes treated glass fiber system explains why GPS: TEOS ratio of 75:25 shows higher interfacial shear strength compared to 1% GPS sized one.

Conclusion

In this work, the glass fiber surface was treated with different silane blends including different ratios of GPS and TEOS. To gain some insights into how the addition of TEOS in the coating affects the wettability of the epoxy resin to the coated glass, fiber–epoxy drop contact angles were determined by Carroll’s equations. The results revealed that the contact angle between the coated surface and resin does not change significantly. A small contact angle between coated fiber surface and resin means that there is no big problem for the resin to wet any of these four different coated fiber surfaces. The adhesion properties of the glass fibers and the epoxy matrix were studied through a single fiber microdroplet pull-out test, and the load–displacement curves from pull-out tests were analyzed. The results from stress- and energy-based models of interfacial debonding showed that the specimens containing 75:25 and 50:50 of GPS:TEOS show the most effective improvement in interfacial adhesion properties between glass fiber and matrix. The failure surface of glass fiber–epoxy specimens after debonding was observed using SEM. The SEM images showed that failure mode propagates through the textured interphase in a more tortuous path, which results in a greater degree of energy absorption during fiber–matrix pull out. This investigation demonstrates that the creation of mechanical interlocking between the resin and fiber is a promising route to tailor properties of the composite.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Tanwer

. Mechanical properties testing of uni-directional and bi-directional glass fibre reinforced epoxy based composites. Int J Res Advent Technol 2014; 2: 34–39.

Kim

Mai

. Engineered interfaces in fibre reinforced composites, London: Elsevier, 2006.

Jancar

. Review of the role of the interphase in the control of composite performance on micro- and nano-length scales. J Mater Sci 2008; 43: 6747–6757.

Pukanszky

. Interfaces and interphases in multicomponent materials: Past, present, future. Eur Polym J 2005; 41: 645–662.

Dey

Deitzel

Gillespie

Jr . Inﬂuence of sizing formulations on glass/epoxy interphase properties. Compos Part A: Appl Sci Manuf 2014; 63: 59–67.

Mallick

. Fiber reinforced composites: Materials, manufacturing, and design, New York: CRC Press, Taylor & Francis Group, LLC, 2008.

Yang

Thomason

. Effect of silane coupling agent on mechanical performance of glass ﬁbre. J Mater Sci 2013; 48: 1947–1954.

Sockalingam

Nilakantan

. Fiber-matrix interface characterization through the microbond test. Int J Aeronaut Space Sci 2012; 13: 282–295.

Reilly SP and Thomason JL. Effects of silane coating on the properties of glass fibre and glass fibre reinforced epoxy resin. In: 14th European conference on composite materials, ECCM14, 7–10 June 2010, Budapest, Hungary.

10.

Safi

Zadhoush

Masoomi

. Effects of chemical surface pretreatment on tensile properties of a single glass fiber and the glass fiber reinforced epoxy composite. Polym Compos 2014; 37: 1–10.

11.

Sever

Sarikanat

Seki

. Concentration effect of glycidoxypropyltri- methoxysilane on the mechanical properties of glass fiber-epoxy composites. Polym Compos 2009; 30: 1251–1257.

12.

Kiyoharu

Koji

Keisuke

. Synthesis of monodispersed silica nanoparticles with high concentration by the Stober process. J Sol-Gel Sci Technol 2013; 68: 341–345.

13.

Zheng

. Inorganic-organic sol gel hybrid coatings for corrosion protection of metals. J Sol-Gel Sci Technol 2010; 54: 174–187.

14.

Jensen

McKnight

. Inorganic–organic fiber sizings for enhanced energy absorption in glass fiber-reinforced composites intended for structural applications. Compos Sci Technol 2006; 66: 509–521.

15.

Zinck

Wagner

Salmon

. Are microcomposites realistic models of the fibre/matrix interface? I. Micromechanical modelling. Polymer 2001; 42: 5401–5413.

16.

Kang

S-K

Lee

D-B

Choi

N-S

. Fiber/epoxy interfacial shear strength measured by the microdroplet test. Compos Sci Technol 2009; 69: 245–251.

17.

Zhandarov

Mader

Yurkevich

. Indirect estimation of fiber/polymer bond strength and interfacial friction from maximum load values recorded in the microbond and pull-out tests, Part 1: Local bond strength. J Adhes Sci Technol 2002; 16: 1171–1200.

18.

Zhandarov

Pisanova

. The local bond strength and its determination by fragmentation and pull-out tests. Compos Sci Technol 1997; 57: 957–964.

19.

Piggott

Xiong

. Visualization of debonding of fully and partially embedded glass fibres in epoxy resins. Compos Sci Technol 1994; 52: 535–540.

20.

Piggott

. Debonding and friction at fibre polymer interfaces. I. Criteria for failure and sliding. Compos Sci Technol 1987; 30: 295–306.

21.

Chua

Piggott

. The glass fibre-polymer interface: I-Theoretical consideration for single fibre pull-out tests. Compos Sci Technol 1985; 22: 33–42.

22.

Nairn

. On the use of shear-lag methods for analysis of stress transfer in unidirectional composites. Mech Mater 1997; 26: 63–80.

23.

Duigou

Davies

Baley

. Interfacial bonding of flax fibre/poly(l-lactide) bio-composites. Compos Sci Technol 2010; 70: 231–239.

24.

Nardin

Schultz

. Relationship between fiber-matrix adhesion and the interfacial shear strength in polymer-based composites. Compos Interf 1993; 1: 177–192.

25.

Zhandarov

Pisanova

Lauke

. Is there any contradiction between the stress and energy failure criteria in micromechanical tests? Part I.. Crack initiation: Stress controlled or energy controlled? Compos Interf 1998; 5: 387–404.

26.

Zhandarov

Mader

. Characterization of fiber/matrix interface strength:applicability of different tests, approaches and parameters. Compos Sci Technol 2005; 65: 149–160.

27.

Carroll

. The accurate measurement of contact angle, phase contact areas, drop volume, and Laplace excess pressure in drop-on-fiber systems. J Colloid Interf Sci 1976; 57: 483–495.

28.

Karmakar

Ganguli

. Dense silica microspheres from organic and inorganic acid hydrolysis of TEOS. J Non-Cryst Solids 2000; 272: 119–126.

29.

Duigou

Davies

Baley

. Exploring durability of interfaces in flax fibre/epoxy micro-composites. Compos Part A: Appl Sci Manuf 2013; 48: 121–128.

Evaluation of interfacial properties of the silane blend sized glass fiber–epoxy composite by the microdroplet test

Abstract

Keywords

Introduction

Experimental

Materials

Methods

Surface treatment of glass fibers

Single fiber microdroplet pull-out test

Micromechanical analysis

Contact angle measurement

SEM examination

Results and discussion

Study of contact angle and aspect ratio of microdroplets

Microdroplet test results

Surface morphology of sized glass fibers

Conclusion

Footnotes

Declaration of Conflicting Interests

Funding

References