A Novel Approach to Monitor the Curing of Composite Materials in Closed Tools by the Use of Ultrasonic Spectroscopy

With an ever broadening use of composite materials manufacturers are in high demand of 1 efficient curing cycles to reduce costs and speed up production cycles. One method to to archive this 2 goal is active cure monitoring to determine the exact time of curing needed. This article provides a 3 novel method to measure the cure inside of closed tools by using ultrasonic spectroscopy. For this a 4 simple experiment is used to show the change of the ultrasonic spectrum during the cure of an epoxy. 5 The results clearly show a direct correlation of amplitude and state of cure where the amplitude 6 reaches a global minimum at the glass point. 7


Introduction
Composite materials are the basis of many constructions of our times.They offer high stiffness at low weight.That makes them the ideal material for aerodynamic and space applications where weight is a serious concern.As a result of the growing ecological awareness of the population the demand of composite materials has grown significantly not only in the aerospace industry.Especially the automotive industry sees them as a way to reduce fuel consumption by lowering the general weight.The broader application has resulted in a high demand for large quantities of lower priced composite materials with a consistent quality.The result of this development has set new challenges on manufacturers.Especially the inconstancy of curing times are a big challenge for faster production cycles.
Currently most manufacturers use the standard curing times with high security factors to determine the cure duration.By doing so they accept unnecessary high curing times and thus a lower productivity or face the danger of not fully cured products.A possible solution is the use of cure monitoring.This would allow manufacturers to optimize curing cycles, resulting in higher production values.
There are many different cure monitoring techniques on the market like dielectric cure monitoring ), which is already well established, and fibre optic cure monitoring ( [3] & [4]).A limited number of cure monitoring techniques is able to detect cure changes without direct contact between sensor and composite material.Direct contact may result in unnecessary wear of the sensor and the tool, reducing the overall lifetime and the creation of unwanted visible production marks.Ultrasonic is one of the few techniques that can detect cure changes without direct contact.On of those techniques is ultrasonic cure monitoring.[5] Ultrasonic waves detect the change of acoustic impedance and damping during the curing process.This paper is focused on a new approach on ultrasonic cure by the active use of acoustic tool resonances similar to [6] but with a low energy and low price approach on the subject.

Materials and Methods
To test the viability of ultrasonic spectroscopy to monitor the curing of epoxy materials a simple experiment is prepared.In this experiment a general epoxy (RTM 6 of Hexcel) is applied to a aluminium plate.This epoxy stands as an example for many matrix systems with similar properties.
The plate has the dimensions of 250x250x20 [mm].
On the backside of this plate two piezos made of PIC255 are applied close to the centre with 40 [mm] distance to each other.One functions as a transmitter, the other one as a receiver.The epoxy used to fix the sensor and actuator is RTM 6 as well.It was cured before at 180 [C] under vacuum.
On the front side a pool is created using a general vacuum sealant.In the pool centre a standard thermal sensor is applied.The plate is than heated to 80 During the whole time the ultrasonic spectrum is measured using a 65 [kHz] and 90 [kHz] discrete 300 point swept sine actuation with 20 [V] peak to peak amplitude.This frequency range is in direct proximity to the first resonance frequency in the direction of the plate thickness direction.The resonance frequency range was calculated using the following formula Where f is the resonance Frequency, c is the speed of sound and d is the thickness of the plate and n a numeric to determine the resonance order.By using the known sound parameter for aluminium and the thickness of 20 [mm] the first resonance frequency is calculated at 77 [kHz] for shear waves.

Results
Figure 2 shows the results of a single cure measurement using ultrasonic spectroscopy in a range The fifth change in the impedance spectrum occurs when the oven is opened at 6500[s] and the system is rapidly cooled.Because of the difference in the specific thermal expansion parameters, a high shearing force between the metal and the epoxy occurs, leading to complete detachment.This is clearly indicated by the jump in amplitude and was visibly confirmed during the experiment.The epoxy was kept at room temperature.The results clearly show, that the minimum of the mean spectrum is very close to the calculated point of cure.Differences can be a result of calculation, and epoxy variations.

Discussion
The results presented in chapter 3 clearly show a very clear correlation between amplitude and point of cure visible in figures 2 and 3.There is a small deviation between calculation and measurement.
A possible reason for this deviation is a wrong temperature measurement.The temperature sensor was very close to the aluminium plate.This might have created a faster temperature rise.Another possibility is a fitting error for the curing lines.The deviation is however quite small and can be ignored.
The results of the experiment clearly show the effectiveness of the presented method.Measuring the resonance spectrum allows a good detection of the glass transition point.This allows the tracking of epoxy cure even in closed tools without a direct contact between the epoxy and sensor.By removing a need for a direct sensor epoxy contact, high abrasion before, after or during the cure are possible It also allows sensing without leaving any extra production marks.

Figure 1 .
Figure 1.Test setup for ultrasonic spectroscopy cure monitoring

Figure 2 .
Figure 2. Change in the response spectrum during cure over time

Figure 3 .
Figure 3. Measured temperature (black-dotted) and calculated degree of cure (blue) of the epoxy during the experiment with added support lines (red) to determine glass transition point (intersection), Mean spectrum amplitude of figure 2 green-dashed