Hydrothermal stability of Cr/polyimide interfaces

The hydrothermal stability of both Cr/polyimide and C₇₅Cr₂/polyimide interfaces has been studied using Fourier transform infrared spectroscopy (FTIR), X-ray photoelectron spectroscopy (XPS), and peel testing. It was found that RF sputter treatment of the polyimide surface prior to metal deposition leads to an enhancement of adhesion through chemical bonding of the metal with polyimide. Owing to the enhanced chemical bonding at the interface, failure always occurs cohesively in the polyimide. The RF sputter treatment of the polyimide surface also simultaneously modifies the polyimide underneath the surface. The adhesion strength of the Cr/polyimide interface is degraded significantly upon exposure to high temperature and high humidity (85°C/81% relative humidity, T/H) environment. It is suggested that this degradation results from the hydrolysis of polyimide. The hydrolysis is facilitated by the presence of unstable modified polyimide near the interface. This degradation of adhesion strength can be minimized by converting the unstable modified polyimide to a stable state by reheating the Cr/polyimide interface at 400°C for 40 min before exposure to the T/H environment.     

Surface modification of polyimide films by argon plasma for copper metallization on microelectronic flex substrates

Surface modification of polyimide films such as Kapton E(N) and Upilex S by argon plasma was investigated because of the enhanced adhesive strength with sputtered copper. Peel tests demonstrated this improvement, with a peel strength of 0.7 and 1.2 g/mm for unmodified Kapton E(N) and Upilex S, respectively, and 110.3 and 98 g/mm for argon plasma–modified Kapton E(N) and Upilex S, respectively, in certain plasma conditions. This study showed that the enhanced adhesive strength of polyimide films with sputtered copper by argon plasma was strongly affected by the surface characteristics such as surface morphology and surface energy of polyimide films. Atomic force microscopy and the sessile drop method indicated that the surface roughness and surface energy of the polyimide films were greatly increased by argon plasma, resulting in highly increased peel strength of the polyimide films with sputtered copper. It was observed in electron spectroscopy for chemical analysis (ESCA) that the increased surface energy of the polyimide films from argon plasma was a result of more of the surface being composed of O and N and of the increased number of C? O, C?O, and C? N chemical bonds. © 2005 Wiley Periodicals, Inc. J Appl Polym Sci 99: 744–755, 2006     

Surface modification of polyimide film by coupling reaction for copper metallization

In this study, Upilex-S [poly(biphenyl dianhydride-p-phenylene diamine)], one of polyimide films, was modified by coupling reactions with N,N-carbonyldiimidazole (CDI) to increase adhesion to copper for flexible copper clad laminate (FCCL). Imidazole groups show strong interaction with copper metal to make charge transfer complexes. Because polyimide film did not have active site with coupling agent, the film surfaces were modified by aqueous KOH solutions and reacted with dilute HCl solutions.Surface modified Upilex-S was analyzed by X-ray photoelectron spectroscopy (XPS) to examine the surface chemical composition and film morphology and investigated by scanning electron microscopy (SEM) and atomic force microscopy (AFM). Changes in the wettability were evaluated by measuring contact angle with the sessile drop method. After deposition of copper on surface modified Upilx-S, the adhesion strength of the copper/polyimide system was measured by a 90° peel test using the Instron tensile strength tester. The peel strength of the copper/polyimide system increased from 0.25 to 0.86 kg_f/cm by surface modification. This result confirmed that the CDI coupling reaction is an effective treatment method for the improvement of the adhesion property between copper metal and polyimide film.     

Adhesion behavior of PDMS-containing polyimide to glass

The effect of the incorporation of poly(dimethyl siloxane) (PDMS) segments into a poly[N,N'-(p,p'-oxydiphenylene) pyromellitimide] (PMDA-ODA) polyimide backbone on the adhesion between PMDA-ODA polyimide and glass was investigated using X-ray photoelectron spectroscopy (XPS), infrared (IR) spectroscopy, contact angle measurements, and the peel test. The peel energy of PMDA-ODA polyimide to glass was significantly improved when low molecular weight PDMS (248.5 g/mol) was incorporated, while little improvement was observed for the incorporation of high molecular weight PDMS (900 or 1680 g/mol). Exposure to air resulted in a considerable deterioration in the peel energy for the pure PMDA-ODA polyimide, while no deterioration was observed for the PDMS-containing polyimides. The improvement in peel strength was successfully achieved by the incorporation of very small quantities of PDMS such as 2 wt%. Based on XPS, IR spectroscopy, and contact angle measurements, it is suggested that the incorporated PDMS segments migrated from the bulk polyimide to the polyimide/glass interface and chemically bonded to the glass surface, which resulted in enhancement of the peel energy. However, a weak boundary layer was formed between the bulk polyimide and glass when a high molecular weight PDMS (900 or 1680 g/mol) was incorporated and thus the peel energy deteriorated.     

Adhesion Properties of Acrylic Copolymers Modified with Si-Containing Copolymer

We made clear the cause of the increase in peel strength of pressure sensitive (PS) adhesives as a function of contact time, and investigated how to modify PS adhesives to maintain a low and constant peel strength for a long time. It was found that polar groups in the adhesive orient to the interface between the adhesive and the (stainless steel) metal substrate (SUS 304) so as to minimize interfacial free energy during adhesion, and the orientation increased affinity between the adhesive and the metal material and increased the peel strength as a result. The use of modifier which contained both P(MMA-co-SiMA) and PDMS showed an excellent modification effect, although modification with only PDMS or P(MMA-co-SiMA) was not sufficient. It was suggested that PDMS which migrated to the surface was extended uniformly over the surface by PDMS segments of P(MMA-co-SiMA) and that the enriched layer of PDMS on the adhesive surface worked as a barrier to prevent the orientation of polar groups in bulk. Therefore, low and constant peel strength could be achieved.     

Adhesion Properties of Acrylic Copolymers Modified with Si-Containing Copolymer

We made clear the cause of the increase in peel strength of pressure sensitive (PS) adhesives as a function of contact time, and investigated how to modify PS adhesives to maintain a low and constant peel strength for a long time. It was found that polar groups in the adhesive orient to the interface between the adhesive and the (stainless steel) metal substrate (SUS 304) so as to minimize interfacial free energy during adhesion, and the orientation increased affinity between the adhesive and the metal material and increased the peel strength as a result. The use of modifier which contained both P(MMA-co-SiMA) and PDMS showed an excellent modification effect, although modification with only PDMS or P(MMA-co-SiMA) was not sufficient. It was suggested that PDMS which migrated to the surface was extended uniformly over the surface by PDMS segments of P(MMA-co-SiMA) and that the enriched layer of PDMS on the adhesive surface worked as a barrier to prevent the orientation of polar groups in bulk. Therefore, low and constant peel strength could be achieved.     

Reactive bonding of natural rubber to metal by a nitrile–phenolic adhesive

The mechanism of adhesive bonding of rubber to metal using an interlayer of bonding agent (adhesive) is discussed with respect to various physical and chemical events such as adsorption at the metal surface, chemical crosslinking within the adhesive, interdiffusion, and formation of interpenetrating networks at the rubber–adhesive interface. An investigation on the peel strength of a natural rubber (NR)–adhesive–metal joint, made by vulcanization bonding using nitrile–phenolic adhesive containing various concentrations of toluene diisocyanate–nitrosophenol (TDI–NOP) adduct, is presented. A single‐coat adhesive, consisting of a p‐cresol phenol formaldehyde resin, nitrile rubber (NBR), and vulcanizing agents in methyl ethyl ketone solvent, was selected for the study. Considerable improvement in the peel strength was obtained by the incorporation of TDI–NOP adduct into the nitrile–phenolic adhesive. The peel strength increases as the concentration of TDI–NOP adduct in the adhesive composition increases, then levels off with a transition from interfacial failure to cohesive tearing of rubber. The peel strength improvement is believed to be attributed to the interfacial reactions between the bonding agent and natural rubber, when TDI–NOP adduct is incorporated. © 2001 John Wiley & Sons, Inc. J Appl Polym Sci 81: 2597–2608, 2001     

Adhesion and interface studies between copper and polyimide

The adhesion and interface structure between copper and polyimide have been studied. Polyimide films were prepared by spinning a polyamic acid solution (Du Pont PMDA-ODA) in an NMP solvent onto a Cu foil, followed by thermal curing up to 400°C. The adhesion strength was measured by a 90° peel test. The peel strength of 25 μm thick Cu foil to 25 μm thick polyimide substrate was about 73 g/mm with the peel strength decreasing with increasing polyimide thickness. Cross-sectional TEM observation revealed very fine Cu-rich particles distributed in the polyimide. Particles were not found closer than 80-200 nm from Cu boundary. These Cu-rich particles were formed as a result of reaction of polyamic acid with Cu during thermal curing. We attribute the high peel strength to interfacial chemical bonding between Cu and polyimide. This behavior is in contrast to vacuum-deposited Cu onto fully cured polyimide.     

Temperature effect on PI/Cu interface

The interfacial reaction and peel strength of polyimide with copper foil at various cure schedules have been investigated by scanning electron microscopy (SEM), transmission electron microscopy (TEM), X-ray photoelectron spectroscopy (XPS), and peel test to determine the temperature effect on polyimide/copper interface diffusion and adhesion. SEM studies indicate that the polyimide/copper interfaces are fairly smooth for all samples studied in this experiment. The TEM microstructure reveals the existence of a copper-polyimide interaction zone at the interface when it is cured at a temperature higher than 250°C, which also results in a high peel strength. XPS spectra revealed higher copper and carbonyl carbon contents at the polyimide interface when it is cured at a high-temperature schedule (350°C). From the results of these interface studies, it is concluded that chemical bonding resulting from the interaction of copper oxide and polyimide carbonyl group provides the binding force for polyimide and copper foil. © 1993 John Wiley & Sons, Inc.     

Temperature and humidity effects on adhesion of polyimide film to a silicon substrate

A 90° peel tester with substrate heating capability was built to evaluate the adhesion strength of polyimide films to a silicon substrate. The effects of polyimide film thickness and peel rate on polyimide adhesion to a silicon substrate under high or low humidity, and at elevated temperatures, have been evaluated. In a high humidity environment, a low peel strength was measured. The influence of moisture on the peel strength increases with decreasing peel rate. Peeling at elevated temperature reduces the moisture effect even under high humidity conditions. Using a low peel rate in a high humidity environment, the measured peel strength showed a maximum as the polyimide film thickness increased. No striations in peeled polyimide films were observed for peeling in a high humidity environment.     

Adhesion of polyimide to annealed 301 stainless-steel films

—Several chemical treatments for the surface preparation of 301 austenitic stainless steel for bonding to polyimide films of pyromellitic dianhydride-oxydianiline (PMDA-ODA) are compared and characterized by means of surface analysis. The role of surface roughness, chromium surface enrichment by selective etching of the alloy, and the use of amine-terminated or styryl-terminated silanes in adhesion promotion are also described. Finally, 90° peel strengths for the polymer/metal interfaces developed by the various surface preparation methods are compared and discussed.     

Surface analysis of press dried-CTMP paper samples by electron spectroscopy for chemical analysis

The effect of press-drying temperature on the surface chemistry of chimicothernomechanical pulp fibers has been studied using electron spectroscopy for chemical analysis (ESCA). The chemical composition showed no significant variation for press-dried samples at temperatures between 25 and 140°C. On the other hand, ESCA showed that lignin content increased whereas hemicelluloses content decreased on the surface of press-dried samples at 175°C. By its hydrophobic nature, lignin gives to paper and paperboards better dimensional stability and resistance to moisture and water. However, lignin does not intervene in fiber bonding because the specific bond strength does not vary with press-drying temperature. © 1996 John Wiley & Sons, Inc.     

Correlation of residual stress and adhesion on copper by the effect of chemical structure of polyimides for copper‐clad laminates

BACKROUND: Polyimide films coated on copper are a potential new substrate for fabricating printed circuit boards; however, adhesion between the copper and polyimide films is often poor. The relations between residual stress and adhesion strength according to the development of molecular orientation of polyimide films with different chemical backbone structure coated on copper were studied. RESULTS: The effect of chemical structures on properties including the residual stress and the adhesion strength were widely investigated for four different polyimides. Diamine 4,4′‐oxydianiline (ODA) and dianhydrides 1,2,4,5‐benzenetetracarboxylic dianhydride (PMDA), 4′‐(hexafluoroisopropylidene)diphthalic anhydride (6FDA), 4,4′‐oxydiphthalic anhydride (ODPA) and 3,3′,4,4′‐benzophenone tetracarboxylic dianhydride (BTDA) were used to synthesize polyimide. In an attempt to quantify the interaction of thermal mismatch with the polyimide films depending on various structures, residual stress experiments between polyimide film and Cu? Si wafer were carried out over a range of 25–400 °C using in situ thin film stress analysis. A universal test machine was used to conduct 180° peel test (ASTM D903‐98) of polyimide film from cooper foil. The residual stress on Cu? Si (100) wafer decreased in the order 6FDA‐ODA > BTDA‐ODA > ODPA‐ODA > PMDA‐ODA, and the interfacial adhesion strength decreased in the order BTDA‐ODA (5 N mm^?2) > ODPA‐ODA > PMDA‐ODA > 6FDA‐ODA. The results may suggest that the morphological structure, degree of crystallinity of chain orientation and packing significantly relate to the residual stress and adhesion strength in polyimide films. Wide‐angle X‐ray diffraction was used for characterizing the molecular order and orientation and X‐ray photoelectron spectroscopy was used for the analysis of components on copper after polyimide films were detached to confirm the existence of copper oxide chemical bonding and to measure the binding energy of elements on the copper surface. CONCLUSION: In this research, it is demonstrated that BTDA‐ODA polyimide has a low residual stress to copper, good adhesion property, good thermal property and low dielectric constant. Therefore, BTDA‐ODA would be expected to be a promising candidate for a two‐layer copper‐clad laminate. Copyright © 2007 Society of Chemical Industry     

Changes in the adhesion strength between copper thin films and polyimide substrates after heat treatment

The adhesion strength between a copper (Cu) thin film and a polyimide [pyromellitic dianhydride-oxydianiline (PMDA-ODA)] substrate is reduced by heat treatment at 150°C in air. In this work, we determined the changes in adhesion strength between Cu films and polyimide substrates using Auger electron spectroscopy (AES), attenuated total reflection Fourier-transform infrared spectroscopy (FTIR), and X-ray photoelectron spectroscopy (XPS). The analysis showed that the weak boundary layer (WBL) shifted towards the Cu interface as the heat treatment time was increased. To confirm this shift, we looked at two other polyimide substrates: biphenyl dianhydride-p-phenylene diamine (BPDA-PDA) and biphenyl dianhydride-oxydianiline (BPDA-ODA). Comparing the adhesion strength for the Cu thin film, the adhesion strength was high for the Cu/PMDA-ODA and Cu/BPDA-ODA laminates, but very low for the Cu/BPDA-PDA laminate. One of the possible reasons for this behavior could be that the ether moiety between the two benzene rings in ODA is related to the adhesion between a Cu film and an 0₂-plasma-treated polyimide (PI) substrate. The relationship between the adhesion strength and chemical bonding states is also discussed. We conclude that a Cu thin film sputtered onto a PI substrate is apt to peel at the oxidized interface, due to the heat treatment.     

Adhesion behavior of CuCr alloy/polyimide films under 350°C and 85° C /85 % RH environments*

CuCr alloys with varying Cr content were sputter-deposited onto polyimide films and the metal/polyimide films were exposed to a 350°C/N₂ environment for up to 10 h or to 85°C/85% relative humidity (RH) (T/H) conditions for up to 840 h for reliability measurements. The Cr contents of the alloy layers (x) prepared were 0, 2.5, 8.5, 17, 25, 34, and 100 atomic %. Before exposure to hostile environments, the peel strength increased proportionally with the Cr content in the alloy layer up to x = 17 and saturated around 550 J/m², and failure occurred within the polyimide near the metal/polyimide interface, except for the specimen with no Cr (x = 0). After exposure to 350°C, the peel strength dropped for all specimens, but most drastically for the specimens with lower Cr contents (x = 8.5) which failed along the Cr-oxide/polyimide interface. The general trend was the same in the case of the T/H treatment, where interfacial failure along the CuCr-oxide/polyimide interface was found for the specimens with lower Cr content (x 17). The extent of interfacial failure over the peeled metal surfaces was found to increase with the T/H treatment time and was inversely related to the peel strength.     

Tantalum and chromium adhesion to polyimide. Part 2. Peel and locus of failure analyses

Ta and Cr adhesion to 3,3'-4,4'-biphenyl tetracarboxylic acid dianhydride-p-phenylenediamine derived (BPDA-PDA) polyimide (PI) surfaces has been studied before treatment, and after CF₄ reactive ion etching (RIE), and Ar sputtering. The initial peel adhesion results for both metals on the BPDA-PDA surfaces are comparable and show increased peel adhesion as a function of the surface treatment in the following order: virgin (no treatment) < Ar sputter < CF₄ RIE ~ CF₄ RIE followed by Ar sputter. The surface roughness effect on metal/PI adhesion has also been investigated. The data suggest that the surface roughness does not primarily affect peel adhesion. In this case, it is the removal of the weak boundary layer and the cracking of the residual PI on the metal peel interface surface during the peeling process which cause the increase in the peel strength. It is also proposed that the changes observed in the peel strength as a function of the surface treatment are due to differences in the fracture toughness of the modified PI layers rather than differences in the surface roughness.     

Changes in the Peeled Surfaces of Copper Thin Films Deposited on Polyimide Substrates After Heat Treatment

Peel strength between a copper (Cu) thin film and a polyimide (pyromellitic dianhydride-oxydianiline, or PMDA-ODA) substrate is reduced by heat treatment at 150°C in air. In this work, we investigated the peel strength, the morphology of the interface between Cu films and polyimide substrates using optical microscopy and electron microscopy, and chemical change of the interface using Auger electron spectroscopy (AES) and micro X-ray photoelectron spectroscopy (XPS). The analysis showed that CuO “lumps” were present on the peeled surface of PMDA-ODA after heat treatment at 150°C in air. The peeled surfaces of other polyimide substrates were also analyzed: biphenyl dianhydride-para phenylene diamine (BPDA-PDA) and biphenyl dianhydride-oxydianiline (BPDA-ODA). CuO lumps were present on the peeled surface of BPDA-ODA after the heat treatment, but not that of BPDA-PDA. Compared with the adhesion strength for the Cu thin film, the adhesion strength was high for the Cu/PMDA-ODA and Cu/BPDA-ODA laminates, but the adhesion strength was very low for the Cu/BPDA-PDA laminate. This low strength is the reason that CuO lumps were not detected on the peeled surface of the BPDA-PDA substrate. These CuO lumps were related to the adhesion degradation of the Cu/polyimide laminates after the heat treatment.     

Modification of a nitrile-phenolic bonding agent by addition of an interfacial agent: effect on the practical adhesion between nitrile rubber and metal

—The mechanism of vulcanization bonding of a nitrile rubber (NBR) elastomer to metal with a single-coat nitrile-phenolic bonding agent is discussed. A nitrile-phenolic bonding agent consisting of NBR, phenol formaldehyde (PF) resin, and vulcanizing agents was modified with an interfacial agent (p-cresol formaldehyde resin) and the effect of interfacial agent addition on the practical adhesion between metal and the NBR elastomer after vulcanization was investigated. The adhesion strength was measured in terms of the metal-to-NBR elastomer peel strength using the bonding agent. The addition of p-cresol formaldehyde (PCF) resin to the bonding agent with a proportionate reduction of PF resin initially improved the peel strength; a maximum was reached at about 20% PCF content and then decreased with a further increase in the PCF content. The improvement in peel strength produced by the addition of PCF resin is attributed to the increased chemical bonding between NBR and the phenolic resin. The drop in peel strength above 20% PCF content is explained by the increased diffusion of the bonding agent into the NBR elastomer, away from the bond line, leading to a starved glue line. The mechanism for the optimum performance at about 20% PCF resin content is believed to be due to the balance of diffusion and chemical crosslinking.     

Effects of moisture and temperature ageing on reliability of interfacial adhesion with black copper oxide substrate

The reliability of adhesion performance of bare Cu, as-deposited and surface-hardened black oxide coatings on Cu substrates was studied. The interfacial adhesion with a polyimide adhesive tape and an epoxy moulding compound was measured using the button shear and tape peel tests after hygrothermal ageing in an autoclave, high temperature ageing and thermal cycles. Moisture adsorption and desorption studies at different aging times suggested that the black oxide coating was effective in reducing the moisture adsorption. The bond strengths for all substrates remained almost unchanged after thermal ageing at 150°C for 8 h. Thermal cycling between ?50°C and 150°C for 500 cycles reduced by about 20% the button shear strength of the as-deposited black oxide substrate, but it did change much the bonding performance of the bare Cu substrate. Hygrothermal ageing at 121°C/100% RH in an autoclave was most detrimental to adhesion performance because of the combined effect of elevated temperature and high humidity. The reduction in button shear strength after the initial ageing for 48 h was 50–67%, depending on the type of coating. In all accelerated ageing tests, the residual interfacial bond strengths were consistently much higher for the black-oxide-coated substrates than the bare Cu surface, confirming a higher reliability of black oxide coating. Fracture surfaces analysis of tape-peeled bare copper substrates after 500 cycles of thermal loading revealed a transition in failure mechanism from interfacial to cohesive failure. In contrast, the failure mechanism remained unchanged for black-oxide-coated substrates. The observations made from the button shear and tape peel tests were generally different because of the different fracture modes involved.     

A New Concept for Adhesion Promotion in Metal–Polymer Systems by Introduction of Covalently Bonded Spacers at the Interface

A new concept for molecular interface design in metal–polymer systems is presented. The main features of this concept are the replacement of weak physical interactions by strong covalent bonds, the flexibilization of the interface for compensating different thermal expansions of materials by using long-chain flexible and covalently bonded spacers between the metal and the polymer as well as its design as a moisture-repellent structure for hindering diffusion of water molecules into the interface and hydrolysis of chemical bonds. For this purpose, the main task was to develop plasmachemical and chemical techniques for equipping polymer surfaces with monotype functional groups of adjustable concentration. The establishing of monotype functional groups allows grafting the functional groups by spacer molecules by applying usual wet-chemical reactions. Four processes were favoured for production of monotype functional groups by highly selective reactions: the plasma bromination, the plasma deposition of plasma polymers, the post-plasma chemical reduction of O-functionalities to OH-groups, and the chemical replacement of bromine groups by NH₂-groups. The grafting of flexible organic molecules as spacers between the metal layer and polymer improved the peel strength of the metal. To obtain maximal peel strength of aluminium coatings to polypropylene films and occurrence of cohesive failure in the polypropylene substrate, about 27 OH groups per 100 C-atoms or 6 COOH groups per 100 C-atoms were needed. Introducing C_6–11-aliphatic spacers 1 OH or COOH group per 100 C-atoms contributed about 60% of the maximal peel strength of the Al–PP system, i.e. 2 or 3 spacer molecules per 100 C-atoms were sufficient for maximal peel strength.