Interaction of Surface Preparation and Cure Parameters on the Interface Reliability of Flexible Encapsulation in Flexible Hybrid Electronics Applications

Encapsulants are liquid resins made up of filler particles and epoxy that are coated and cured on flexible hybrid electronics to protect the substrate from moisture, dust, and other environmental variables, hence enhancing the durability of the flexible hybrid electronics. They are often soft materials, which allows us to have a large bending radius while also reducing the differential stresses caused by bending and twisting, which extends the fatigue life of flexible hybrid electronics (FHE) [1,2]. The current applications of these encapsulants are in light-emitting diode displays, which are fabricated on flexible thin-film device layers [3], and have mechanical weak points such as thin-film transistors [4,5]. These materials often crack at low strain, reducing their functionality and performance [6]. The encapsulants are used as a second layer on brittle thin film devices to reduce bending strain by shifting the device close to the bending neutral plane [7]. These encapsulated devices have shown to work on much smaller radii down to 1 mm [8]

Researchers have investigated various methods to achieve the robustness of flexible devices. Some of the approaches are spin coating the surface with encapsulation layers, low temperature curing or heat treating the encapsulant by Ultraviolet exposure [8–10]. The encapsulants are all viscoelastic materials with polymer properties, the Young's modulus and coefficient of thermal expansion are affected by temperature changes. The temperature near the glass transition (Tg) was discovered to significantly impact mechanical characteristics [11,12]. Most of the work on encapsulation is done on organic thin-film light-emitting diodes. There is a gap in the literature to observe the process parameters of binding these encapsulants with the substrates. The researchers have observed no prior work on quantifying interface adhesion between the encapsulation layer and substrates. This paper aims to fill that void using peel strength as a function of cure properties.

The influence of cure profile and substrate type on the peel strength of the material is the subject of this article. The investigation includes two substrates with varying stiffnesses and four encapsulants with varying viscosities that will be utilized to differentiate them. The peel strength of the material influences the dependability performance of the restraint mechanism. The encapsulant specimen was subjected to a 90 deg peel test, and the findings were quantified as a factor of peel strength, providing us with reliable performance parameters utilized in flexible hybrids.

The specimen used to test peel strength was rectangular in shape, and it was made up of two parts: a substrate and an encapsulant substance. The sample dimensions are 25 mm × 12 mm × 5 mm. At least three specimens were evaluated for each material configuration investigated in this work. A total of four encapsulants, comprising developing materials V, X, Y, and Z, were examined (Table 1). In the proposed mix-formulation, the ingredients have a wide range of initial viscosity values ranging from 28,000 cps to 67,000 cps. All of the formulations described in the study are two-part formulations containing a resin and a hardener. The approved formulation for the resin-hardener mix ratios was followed in all situations. The snap-time of the examined formulations is comparable for all formulations in the 8-9.5 min range. Furthermore, the tack-free period for all formulations evaluated is between 15 and 18 min. The specific gravity of the materials is in the restricted range of 1.12–1.26.

The peel-test methodology is configured to peel at a constant rate of 2 mm/min. The experimental settings employed in the test are shown in Table 2. The peel angle is maintained at 90 deg during the peel by attaching both surfaces in a way that permits them to glide to maintain a consistent peel angle when the interface surface is pressured. For the samples, the bond-line width was held constant at 12 mm. The bond-line thickness is in the 5 mm range. The bond-line length is in the vicinity of 25 mm. The published data in the study has been standardized in terms of width to enable for scalability of the reported results to other widths and more general comparison of the given results with other sample geometries.

The effect of substrate preparation prior to encapsulation on both Polyethylene terephthalate (PET) and Polymide substrates has been explored, including the effect of cleaning with isopropyl alcohol (IPA) and plasma cleaning. As demonstrated in Table 3, each of the flexible encapsulants examined had a unique cure profile. The usage of four cure profiles, designated as cure profiles A, B, C, and D, has been investigated. Cure profile-A calls for a cure temperature of 50 °C for 3 h. Cure profile-B calls for a cure temperature of 50 °C for 2 h.

Figure 1 depicts the peel test arrangement. The first jig was a 90 deg peel rail, and the second jig was a tensile jig that was attached to the Instron 3367 machine, which included a 1 kN load sensor. In this experiment, only one peel-arm was attached to the tensile jig, while the other side was bonded to the peel rail's steel plate with Loctite 414 (Fig. 2). Proper adherence of the bottom surface to the peel-rail is thus critical to ensuring that the reported peel strength values are for the interface of interest rather than the adhesion contact with the peel-rail. The experiment was carried out with the tougher peel arm attached to the tensile jig and the softer material using Loctite to the rail guide. Debonding of at least 12 mm length was employed in all tests to capture the adhesive peel strength of materials and the highest values were recorded as the peel strength of the materials. Figure 3 depicts the specimen peel test arrangement with the 90 deg peel in a cohesive failure mode.

Figure 4 depicts an average-load/width-displacement plot based on data acquired from the Instron during the 90 deg peel test. Figures 5–8 quantify the peel strength and sample variance at various temperatures for distinct material types. In the experiment, we observed that raising the cure temperature weakens adhesive bonding while the other components remained constant. The substance Y has the greatest peel strength value of all the encapsulants, and it should also be mentioned that it has the highest viscosity, which equates to the highest quantity of filler content, hence peel strength directly varies with the amount of filler content in our experiment.

The data provided below show that the materials were cured for the same amount of time, the substrates were manufactured of the same material, and the substrates were cleaned under the same conditions.

Figures 9–12 illustrate the cure time on the adhesive bond between the encapsulant material and the substrate. Our findings reveal that increasing the cure time while maintaining the constant temperature results in a greater value of binding strength between the encapsulant substrate. The variation between samples is also displayed, with substance Y having the highest variance of any encapsulant material. The overall cure time of 3 h at 50 °C provides the best case scenario of peel strength across every encapsulant material. Our idea is that cure time allows the polymer to realign to long chains, improving strength, but cure temperature induces fast polymerization, lowering strength.

Figures 13–16 demonstrate the cure profile's influence on the encapsulant's steady-state strength owing to cure profiles.

Consider Fig. 13 and material Y; one can see that the peak strength is 72% greater than the steady-state strength at 60 °C, which is undesirable for real-time applications; therefore, the best case scenarios should be those with a low percent relative change between the peak strength and the steady-state strength. Higher temperatures cause a large difference in strength between the two; however, longer cure times diminish the difference between peak strength and steady-state strength.

Figure 17 depicts the effect of surface cleaning on the adhesive binding of the encapsulant on the surface of the PET substrate. Our observations have shown that plasma cleaning of the substrate results in a better adhesive connection between the encapsulant and the substrate than IPA cleaning. The cure profile applied in this case is cure profile A, which is the typical cure profile standard. This feature could be attributed to the high surface energies produced due to plasma cleaning leading to better bond strength between the surfaces (Fig. 18).

Figure 19 depicts the influence of substrate type on the encapsulant's adhesive bond. According to the results, the PET substrate provides a superior bonding on the encapsulant (Fig. 20). This experiment uses a conventional cure profile A to quantify data.

Figure 21 depicts the effect of silver ink printing on the substrate's adhesive bond. The printing was completed on an Optomec AJP 300 machine. The printed substrate was then cured using cure profile A, and a peel test was performed to understand and quantify the material's peel strength. Our observations indicate that there has been an influence on adhesive bonding, but it may not be as substantial because it is so near to the initial bare substrate peel value.

Data indicates that lower temperature 50 °C and higher cure time of 3 h enables the encapsulants to better bond with the substrate. The plasma cleaning provides superior adhesion than IPA cleaning because it increases the surface energy of the underlying substrate. The material Y has best performance regarding interface peel strength which correlates to having higher viscosity among all the material.

In this article, we measured the influence of substrate alteration, cure profile, and substrate cleaning on several encapsulant materials, and we discovered that the material Y has the best peel strength values, while the PET substrate has a superior adhesion property than the PI substrate. The material Y has the highest viscosity of all the materials. All of the encapsulant materials have cohesive failure, which is inherently better than adhesive failure because the wearable circuits will not be exposed to harmful environment, so one must consider the type of application these encapsulants will be used for; if there are no small crevices and gaps in the design of the hybrid substrate, it is better to use the encapsulant with the highest viscosity, and If there are small gaps where the encapsulant must flow, viscosity becomes a significant element. Strength must be sacrificed for improved flow-ability of material, and our advice would be to utilize material V in such types of applications.

The project was sponsored by the NextFlex Manufacturing Institute under PC4.1 Project titled – Encapsulation & Overmolding of FHE Devices. This material is based, in part, on research sponsored by Air Force Research Laboratory under Agreement No. FA8650-15-2-5401, as conducted through the flexible hybrid electronics manufacturing innovation institute, NextFlex. The views and conclusions contained herein are those of the authors and should not be interpreted as necessarily representing the official policies or endorsements, either expressed or implied, of Air Force Research Laboratory or the U.S. Government.

The authors attest that all data for this study are included in the paper.