Predicting production outcomes in manufacturing relies on accurate data that calculates the variables that actually matter to things like reliability, performance and longevity of the product. There are countless tests to examine every conceivably measurable aspect of production.
Metrology is inextricable from quality assessment. But are manufacturers measuring everything they need to guarantee a zero defect production process?
If all the aspects that determine the quality of a process like adhesion aren’t measured and controlled, then how can certainty that every product is flawless be attained?
It’s no secret that adhesion, in all its many forms (e.g. bonding, coating, printing, painting, gluing, laminating, overmolding, sealing, etc.) is a concern for nearly all manufacturers in every major industry. And adhesion, which is reliant upon the adhesive or coating, the curing process and the quality of the surface being adhered to, is difficult to understand and even harder to control. That is, unless you have the proper way to manage the three elements adhesion is reliant upon.
Manufacturers have a firm grasp on the proper composition of the adhesive or coating they need for a given application. Usually there are few to no appropriate substitutes. They also understand how to cure their given adhesive or coating, and they have many accurate ways of measuring how successful the curing process was.
What’s typically left to chance is the quality of the bond surface. The chemical characteristics of the top 1-5 molecular layers of a metal, polymer, composite or other substrate can be all that stands between perfect, strong adhesion that meets all performance requirements and weak, unpredictable bonds that fail intermittently.
Chemical cleanliness is a specific state of cleanliness where the molecules present on a surface are adhesion-promoting, or at least not detrimental to adhesion. This is a difficult characteristic to measure without the right method and there are very few techniques that are sensitive to the incredibly small contaminants that can result in weak bonds. There are even fewer devices that make these measurements possible in production or “near to action” on real parts where it is needed the most.
Using Contact Angle Measurements to Predict Adhesion
Contact angle measurements are sensitive to extremely minute surface energy changes and therefore can detect, with ruthless accuracy, how prepared a surface is for adhesion. Surface energy fluctuates as the molecular makeup of the surface changes due to treatment, cleaning, abrasion, contamination, degradation of a clean surface or dozens of other scenarios that can alter the reactivity of the surface. Every place in a production process where the surface has an opportunity to change in a way that affects adhesion outcomes is called a Critical Control Point (CCP).
When contact angle measurements are taken at each CCP, then manufacturers can be certain they know whether the surfaces of their materials are within the tolerances for excellent adhesion.
To learn more about building a medical device production process that verifies the surface quality of each and every product, download our eBook: How to Streamline Process Design to Production for Medical Devices
The sensitivity of these measurements makes them perfect for root cause analysis as they can easily highlight where in the process something is going awry. One of the world’s leading manufacturers of high-performance wire and cable was experiencing inexplicable and intermittent failures of adhesion that needed to be dealt with. Once they employed contact angle measurements to measure surface energy and get an accurate picture of what their surface quality looked like, the solution became clear and decisive action was able to be taken.
The Cost of Adhesion Failure
A couple years ago, a manufacturer of medical cable and connectors made a critical change to their process.
The cables they produce have silicone overmolded connectors on each end. The connectors are typically chrome-plated. To increase manufacturing efficiency, they decided to remove a chrome-plating step and instead started molding silicone over the chrome plating, but it wasn’t adhering consistently
For about a year they experienced intermittent bonding failures, some days resulting in as much as 60% of the day’s production being scrapped and then the next day only 3% would fail -- an untenable gap in reliability.
Reworking hundreds of parts was taking resources away from producing good quality parts the first time around. Every cable that is reworked equals one cable that is not being produced. The cycle that is run for the reworked cable is the same cycle run for a new cable. So, each reworked part represents a double loss.
The manufacturer conducted rigorous root cause analysis by examining their cleaning and handling process. They enlisted the help of a laboratory to see if they could understand what was making the bond fail. Unfortunately, the data they received was vague and consisted mostly of a verbal report that nothing unusual was found. This dead end left the manufacturer on their own again to try to figure out the origins of the problem.
Subsequent Failure Modes and Effects Analysis (FMEA) efforts led them to expend effort reducing the amount of touches each part would experience in order to reduce the risk of contamination. Further, they went through an exercise monitoring temperature and humidity in their facility to see if there was a meaningful impact. Unfortunately, none of these efforts seemed to have an effect on the overmold’s ability to adhere to the chrome plating.
Last year, they reached out to Brighton Science, kicking off a partnership between Brighton Science Scientists and the engineers at the cable manufacturer to help investigate the issue, diagnose the root cause, and implement a permanent solution.
Through trial and error, they hadn’t seen success in understanding where the problem was originating and it led to the over molded connector becoming the number one reject in the plant. Once they read about Brighton Science they concluded they needed to start monitoring surface quality.
Through examining exactly how the parts were failing - at the interface between the silicone and metal - and how extra cleaning and priming steps, used during the reworking of failed parts, greatly increased the success of the bond, it became clear that unidentified contaminants present on the connector were the likely root cause of the over molding failure.
This led to a closer examination of the cleaning process that the metal components were subjected to. During one of the first visits to the manufacturer’s facility, Brighton Science was able to show their technicians that if you took a long enough time to scrub the side of the metal housing, you could have a positive impact on surface energy. This simple demonstration made it evident that whatever contaminant was on the surface, it was removable. However, sitting there and scrubbing down every component in production was obviously not a production viable situation.
After Brighton Science was able to identify the contaminants on the metal housing as a hydrocarbon and a silicone, we took a look at strategies for optimizing the cleaning processes the manufacturer was currently using.
The Challenges of Process Changes in Medical Device Manufacturing
In medical device manufacturing, OEMs are limited in the changes they can implement. Once a certain cleaning chemical or piece of equipment is approved, any changes to that chemistry or equipment needs to go through additional vetting and approval processes. These product verification procedures can require significant effort and can be time intensive, so any modifications to the process need to be fully verified as correct ones to make before going down this road.
This cable manufacturer had been using a vapor degreaser for the primary cleaning step before conducting the over molding process. Using the Surface Analyst (a handheld contact angle measurement gauge) to take molecularly sensitive contact angle measurements, the investigative team was able to see major inconsistencies in surface energy levels on parts that came out of the vapor degreaser.
Root Cause Analysis using FTIR and Contact Angle Measurements
Samples of the manufacturer’s vapor degreaser cleaning fluids were sent to the Brighton Science Surface Laboratory to see if the contaminants found on the metal surfaces were being transferred during the cleaning process.
Through FTIR spectroscopy analysis of the degreaser bath solution sent to Brighton Science, we were able to determine that there was silicone in the cleaning fluids and thus we were able to identify the probable source of contamination.
The chart below shows the results of our analysis of used degreaser chemistry versus the same brand of pristine, unused cleaner. The peaks highlighted in orange and green indicate the presence of silicones which are notorious for creating weak bonds by making a less reactive (low surface energy) surface than required for strong adhesion.
The chart below displays analysis of the connector housings that failed to bond and were sent to our lab (top half of the graph). The bottom half of the graph shows analysis of the housings after going through the typical cleaning process. The silicones we found in the used wash solution (as seen in the graph above) was transferring to the housings. The silicones show up on these spectra in the green and orange peaks.
Using this knowledge to make informed manufacturing process decisions, the engineers at the cable manufacturer began overhauling their preventative maintenance (PM) program. More frequent changes of the cleaning chemistry, as well as more thorough cleaning of the interior of the tank once the fluid is removed, helped reduce the risk of cross contamination among parts from washer fluid drag out.
The Surface Analyst was used in two ways to assist in this process. First, as a routine quality control metric to ensure parts leaving the degreaser were sufficiently clean and second as a predictive trending metric to let the quality engineering team anticipate when cleaning tank changeovers were necessary.
Over time, using this new data-driven approach, it became clear that the vapor degreaser just wasn’t able to produce the cleanliness levels needed. So, instead of spending the money to change out their equipment, and the pain of going through the expensive requalification process, the engineering team opted to convert the vapor degreaser into an ultrasonic washer.
In the chart below you can see the data we gathered from the Surface Analyst to evaluate the effectiveness of the cleaning process. The vapor degreaser was showing too much variability to be confident in its cleaning ability. Contact angles the manufacturer measured using the Surface Analyst on new parts were showing the highest numbers (i.e. lowest surface quality) yet.
In the chart below we measured the surface quality of the new samples after cleaning them with an ultrasonic washer and a handheld plasma pen treatment. The results showed that, with proper monitoring, the ultrasonic cleaning would result in surfaces ready for bonding. But they also showed that 30 seconds of plasma treatment would bring contamination to unquestionably low levels.
Measuring Surface Cleanliness Leads to Measurably Better Adhesion Outcomes
Using the Surface Analyst to test 100% of the metal housings before they go into and after they come out of the washer, and then correlating the surface quality data the Surface Analyst provides, has equipped this renowned cable manufacturer with a cleaning process that is finally producing the level of successfully bonded connectors they need to satisfy their customers’ demands.
They began seeing scrap rates drop as much as 95% for these connectors, something that seemed impossible just months earlier. They were able to accomplish their production and quality goals while also avoiding a significant capital expenditure acquiring and commissioning new cleaning equipment.
Unlike the first laboratory they looked to for help, the comprehensive data supplied by Brighton Science allowed them to make knowledgeable decisions and see measurable gains from the new approach.
There's no programmable setting on a molding machine that's going to guarantee it will bond. That would make manufacturers’ lives much much easier. Instead, if you can guarantee the housing is clean 100% of the time, then bonding will be successful 100% of the time. Once the cable manufacturer was able to measure the cleanliness of their metal surfaces and verify that each one was always as clean as necessary, then their bonding procedure was naturally far more successful.
To learn more about building a medical device production process that verifies the surface quality of each and every product, download our eBook about streamlining process development. In this free guide, you’ll learn how to meet FDA reliability testing requirements faster and easier than ever before. Download your copy today: How to Streamline Process Design to Production for Medical Devices.