ARTÍCULO 27

MONITORING OF THE CURING PROCESS OF ANINDUSTRIAL COMPOSITE STRUCTURE BY TJS AND FUT

M. Torres , F. Collombet , B. Douchin , L. Crouzeix , R. Bazer-Bachi ,Y-H. Grunevald , J. Lubin , T. Camps , X. Jacob , S. Rodriguez , K-T. Wu .

1 2 2 2 2

3 4 4 5 5 6

1

2

3

4

5

6

INSTITUTO POLITÉCNICO NACIONAL, ESIMEUNIVERSITÉ DE TOULOUSE, INSA, UPS, ICA (INSTITUT CLÉMENT ADER),

COMPOSITES, EXPERTISE & SOLUTIONSUNIVERSITÉ DE TOULOUSE, LABORATORY FOR ANALYSIS AND ARCHITECTURE OF SYSTEMS

UNIVERSITÉ DE TOULOUSE, LABORATORY OF HUMAN BEING PHYSICS APPLIED TO THE ENVIRONMENTINDUSTRIAL MATERIALS INSTITUTE

INSTITUTO TECNOLÓGICOde saltillo

SALTILLO, COAHUILA 23, 24 Y 25 DE OCTUBRE DEL 2013.

MONITORING OF THE CURING PROCESS OF AN INDUSTRIAL COMPOSITE STRUCTURE BY TJS AND FUT

M. Torres*1, F. Collombet2, B. Douchin2, L. Crouzeix2, R. Bazer-Bachi2, Y-H. Grunevald3,

J. Lubin4, T. Camps4, X. Jacob5, S. Rodriguez5, K-T. Wu6.1 Instituto Politécnico Nacional, ESIME U.P. Ticomán, Av. Ticomán 600, 07340, México D.F., México

2 Université de Toulouse, INSA, UPS, ICA (Institut Clément Ader),

133 C Av. de Rangueil, F-31077 Toulouse, France.

3 Composites, Expertise & Solutions, 4 Rue Georges Vallerey, F-31320 Castanet Tolosan, France.

4 Université de Toulouse, Laboratory for Analysis and Architecture of Systems, Toulouse, France.

5 Université de Toulouse, Laboratory of Human Being Physics Applied to the Environment,

118 Route de Narbonne, 31062, Toulouse, France.

6 Industrial Materials Institute, 75 Bvld de Mortagne, Boucherville, Quebec J4B6Y4, Canada.

ABSTRACT

In this work, a multi-instrumentation setup is conceived in order to have a better

understanding of composites curing by using different devices: tunnelling junction sensor

(TJS), flexible ultrasonic transducers (FUT) and thermocouples. The sensors are

simultaneously embedded or placed on the surface of a thick carbon-epoxy plate for

monitoring its autoclave cycle. By means of their signals, the evolution of temperature related

with the polymerization state is described. A qualitative statement is provided to have a first

approach of the initial state of the composite plate in order to improve its mechanical

properties.

RESUMEN

En este trabajo, un experimento de multi-instrumentación es concebido para tener un mejor

entendimiento de la polimerización de estructuras compósitas. En él se utilizan diferentes

dispositivos: sensores a efecto túnel (TJS), transductores ultrasónicos flexibles (FUT) y

termopares (TC). Los sensores son embebidos o puestos en la superficie de una placa

carbono-epoxi para monitorear su manufactura dentro del autoclave. Mediante las señales

recuperadas, se describe la evolución de la temperatura y su relación con el estado de

polimerización del compósito. Una aproximación cualitativa es obtenida a fin de tener el

estado inicial de la placa compósita con el fin de mejorar sus propiedades mecánicas.

Keywords: curing process, carbon-epoxy plate, tunnelling junction sensor, flexible ultrasonic

transducers.

*Corresponding Author, e-mail: mtorresar@ipn.mx, Tel: (55)5729 6000 ext 56104.

1. Introduction

The nature of long-fibre composite structures allows the possibility to implement in-core

instrumentation for monitoring since the manufacturing process until their final application.

Numerous studies have proven the value added of composites in-core instrumentation to

follow through-the-thickness properties [1-4]. However, composites in-core instrumentation

still reveals worries which retard its plenty industrial application. Sensor’s size, autonomy and

wire connexions are the major issues to deal with in order to fulfil this approach in the

industry. Several researchers have shown their interest on the thermal phenomena which

occur during composites curing [3-5]. A deeper knowledge of the curing reactions will aid to

reply on many snags that appear in the fabrication stage and, in consequence, to improve

the mechanical properties of the composite structure.

This works belongs to the project called “Multi-sensor Instrumentation for the Composite

Materials and Structures (I2MC)”. It puts together the expertise of seven research teams with

the goal to study the in-core instrumentation of composites structures [7,8]. In this

collaboration, a multi-instrumentation set up is done for monitoring a composite structure

during its curing process. For instance, one of the goals is to compare the signals from

different embedded devices such as tunnelling junction sensors (TJS) [9-11], and flexible

ultrasonic transducers (FUT) [12, 13] in order to infer the curing stages of the composite

structure.

2. Experimental set up

The composite structure studied here is a M10 HS300 carbon-epoxy plate of 230 x 640 mm

with two drop-offs. The composite plate has three zones with different thickness. In first

place, the “current zone” has 20 plies, in second place the “thick zones” have 36 plies and

finally the “over-thick zone” has 50 plies. The lay-up sequences for the three different zones

are mentioned in table 1. The drop-off zones are built with a 2.5 mm single latter step from

the 11th ply to the 26th ply. This type of structure is representative of the design singularities

currently found in aeronautics.

Table 1. Stacking sequences for the three zones of the carbon-epoxy plate.

Current zone [0/45/0/-45/0/45/0/-45/0/90]sThick zone [0/45/0/-45/0/45/0/-45/0/90/0/45/0/-45/0/45/0/-45]s.

Over-thick zone [90/02/90/0/45/0/-45/0/45/0/-45/0/90/0/45/0/-45/0/45/0/-45/0/902/0]s

The multi-instrumentation of the composite structure can be divided in two sets: the in-core

instrumentation and the surface instrumentation. The in-core instrumentation involves two

monitoring patches [14] with a tunnelling junction sensor (TJS) inside each one. The surface

instrumentation involves five pairs of flexible ultrasonic transducers (FUT) and nine

thermocouples (TC) at different locations of the composite plate. A general view of the

sensors arrangement on the composite plate is illustrated in figure 1. Figure 2 show some

illustrations of the experimental set-up and the placement of the plate inside the autoclave.

Picture 2-a shows the TJS being encapsulated in the monitoring patch. Picture 2-b exhibits

on of the FUT with a silicone seal. Finally, picture 2-c shows the final disposition of the plate

before its placement inside the autoclave for curing.

Figure 1. Instrumentation scheme for monitoring the curing cycle of the composite plate.

Figure 2. Illustrations of the instrumentation procedure for the composite plate.

Patch 2

TJ sensor

FUT

a)

b)

c)

TC2 TC3

TC5

TC8 TC9

TC4 TC7

Embedded thermocouple

Surface thermocouple

TC1

TC6

FUT at mould

FUT atcomposite

Large thick zoneCurrent

zoneOver-thick zone Thick zone Drop-off

mm27035 170 40 80 40

FUT 1

FUT 2

FUT 4

FUT 5

FUT 3

Monitoring patch (TJS)

Patch 2Patch 1

The polymerization cycle begins at room temperature (25ºC). Then a 2ºC/min heating slope

is provided to reach a 90ºC plateau during 45 min. After that, the temperature rises with the

same 2ºC/min slope, to have a 120ºC plateau for 2 hrs. When the second plateau ends, the

cooling phase is fixed with a -1ºC/min slope to get back at room temperature. For the

pressure, we apply 2 bar from the beginning of the cycle until the end of the first temperature

plateau. Then, the pressure is increased to 5 bar until the end of the polymerization cycle.

During all the autoclave cycle, void pressure is fixed at 0.9 bar.

3. Signal of embedded sensors: Tunnelling junction sensors (TJS)

Tunnelling junction sensors (TJS) are electronic devices which take advantage of inverse

polarization of P – N junctions, known as Symmetric Thresholds Element (STE). To stimulate

the P – N junctions, the TJS is polarized continuously. Between 100 nA and 10 mA, the

increment of energy produces the shrinkage of the potential barrier, allowing the electrons to

pass through it and the voltage increases gradually. This fact is a pure quantum physics

phenomenon, known as “tunnelling effect conduction”.

The interest on the tunnelling effect conduction lays onto its dependence to temperature

changes [9, 10]. The sensibility to temperature changes becomes evident when the

polarization current gets a value within the tunnel effect range. For a fixed polarization

current, the voltage will fluctuate as regards on the increment or decrease of temperature.

This condition makes the TJS affordable for monitoring temperature changes.

The TJS of patch 1 is connected to a SMU 2612 Keithley® power source controlled by

Labtracer software. A total of 45 TJS polarizations, from 1 nA to 10 mA, are induced during

the curing cycle in order to follow the temperature changes inside the composite plate. The

space of measurement of the TJS is detected between 2 mA and 5 mA, where the sensitivity

to temperature gets quantifiable. The voltage dependence to the temperature variation is

illustrated in figure 3.

In order to establish the link between the voltage variations of the TJS and the temperature

changes of the composite plate, a temperature – voltage – time diagram is built. Figure 4

shows the TJS voltage variations for two current values, 2.75 mA and 3.8 mA, and the

temperature measurements of the thermocouple TC5 during the curing cycle.

Figure 3. Measurement range for the tunnelling junction sensor (TJS).

On the diagram of figure 4, voltage decreases when the temperature rises and voltage

increases when temperature diminishes. The TJS voltage has an inverse relation to the

composite plate temperature. On one side, the TJS sensitivity at 2.75 mA is -8 mV/ºC; on the

other side, the TJS sensitivity at 3.8 mA is -7 mV/ºC. With these results, the trustworthiness

of using tunnelling junction sensors for composites curing monitoring is pointed.

Figure 4. Temperature – voltage – time diagram for the tunnelling junction sensor (TJS).

4

4.5

5

5.5

6

6.5

7

0

20

40

60

80

100

120

140

0 50 100 150 200 250 300 350

Volta

ge (V

)

Tem

pera

ture

(°C)

Curing cycle (min)

TC 52.75 mA

3.8 mA

2.75 mA 3.8 mA

0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

8.0

2.0E-03 3.0E-03 4.0E-03 5.0E-03

Volta

ge (V

)

Current (A)

T 20,6°C T 24,2°C T 26,8°CT 30°C T 33,4°C T 38,5°CT 42,2°C T 43,7°C T 54,2°CT 63°C T 71,4°C T 80,1°CT 87,4°C T 91,8°C T 92,9°CT 93,5°C T 99,6°C T 115,1°CT 120,6°C T 126,6°C T 127,6°CT 106,7°C T 133,7,6°C T 126,7°CT 123,5°C T 122,6°C T 122°CT 121,7°C T 121,6°C T 121,6°CT 121,6°C T 121,6°C T 121,6°CT 120,2°C T 108,8°C T 94,9°CT 88,9°C T 84,6°C T 79,4°CT 72°C T 68,9°C T 63,5°CT 59,8°C

4. Signal of embedded sensors: Flexible Ultrasonic Transducers (FUT)

Flexible ultrasonic transducers (FUT) are one of the non-destructive techniques (NDT) which

can be used to reveal the presence of internal defects in a structure. FUT works thanks to

the piezoelectric effect of their constitutive micro-layers. By receiving an electrical

stimulation, the FUT produce a high-frequency wave that can travel through any material

which the FUT is in contact to. The parameters of the ultrasonic wave after travelling are

unique and depend only on the atomic composition and the physical state of each material.

The FUT employed on this work are composed by bismuth titanate (BIT) powders dispersed

by sol-gel spray onto a lead-zirconate-titanate (PZT) substrate of 4 mm diameter. Then, this

compound is placed onto a stainless steel foil of 12.7 mm length. This configuration, as

shown in figure 5, gives the FUT their main characteristics: flexibility, good ultrasonic

coupling and functioning at relatively high-temperature (< 500ºC). This last characteristic

makes the FUT promising for monitoring composites during curing processes.

Figure 5. Schemes of the FUT assembly and the different echoes produced by the FUT.

The amplitude and the delay of time for all echoes of each FUT are acquired. After the

polymerization process, only the echoes Lm+c and L2m can be employed for the curing

monitoring because they are strong enough. If we consider the difference between the depart

time and the return time of the echo, this “time delay” of the ultrasonic wave can be analysed

as a function of the physical state of the composite material. Time delay variations can be

attached to the chemical and physical phenomena happening inside the composite plies.

Figure 6 shows the evolution of the time delay for the duration of the curing process. There

are five recognizable phases which are described as follows. S0: phase where mould

temperature is a little above the room temperature. In this phase the resin has a lower

viscosity which provides a good coupling between the mould and the plies. The time delay

Al plate (6 mm)Teflon foil (1 mm)

Plate FUTComposite

plate

Al mould (15 mm)

Lm+cLm+3c L2m

L2m+2c

Steel rodMould FUT

Lm+cLm+3c

L2mL2m+2c

Echoes received by plate FUT

Echoes received by mould FUT

decreases and more ultrasonic waves go through the plies. S1: phase where temperature

continues rising till 90ºC. The viscosity of matrix decreases, the speed of sound diminishes

and the time delay increases until the point D1. This time delay increment is caused by a

partially melting of the resin causing an irregularly impregnation among fibres. S2: phase

corresponding to the 90ºC plateau. Theoretically, the time delay should not present any

change. However, the time delay diminishes gradually until the point D2. In this phase, it is

considered that the polymerization reaction starts. The composite morphology defined by the

mix of the solid state of the fibers and the liquid - rubbery state of the resin could be the

cause of ultrasonic fuzziness. Nonetheless, more experiments should been carried out in

order to confirm this hypothesis. S3: phase located during the second temperature slope,

from 90ºC to 120ºC. In this phase, the polymerization reaction is almost fulfilled. The

evolution of the time delay presents a point D3, probably related to the matrix gel point. After

this point, time delay decreases drastically, because of the matrix reticulation until the signal

stabilizes during the composite consolidation. S4: phase linked to the composite

consolidation. The time delay decreases steadily because of the autoclave cooling. This final

reduction is certainly related to the final composite solidification into a glassy state.

Figure 9. Evolution of the time delay in the composite plate during the curing cycle.

2

2.5

3

3.5

4

4.5

19.15

19.2

19.25

19.3

19.35

19.4

19.45

0 100 200 300 400

Tim

e del

ay in

com

posi

te (µ

s)

Tim

e del

ay o

f L2m

(µs)

Curing cycle (min)

Composite time delay

L2m

S1 S2 S3 S4

S0

D3 (Gel point)

D2

D1

Rubbery state Glassy stateLiquid state

6. Conclusions

A carbon-epoxy plate with two drop-offs is instrumented, in-core and at its surface, with

tunnelling junction sensors (TJS), flexible ultrasonic transducers (FUT), and thermocouples

(TC). This multi-instrumentation is designed to obtain information related to the

polymerization phenomena during the curing cycle.

For the tunnelling junction sensors (TJS), detection of temperature changes is confirmed and

their application in composites monitoring shows promising features. Temperature – voltage

– time diagram are built to link the temperature changes in the composite plate and the TJS

voltage changes. These sensors will get more attentions not only for their temperature

sensitivity, but for their capability for detecting strains too. The reliability of TJS for detecting

residual strains inside composites after curing cycles will be one of the key-points in the

incoming research.

An assembly of flexible ultrasonic transducers (FUT) are placed to follow the thickness

evolution of the composite plate. The measurements through-the-thickness of the ultrasonic

waves can give signs of the physical state of the composite plies. The evolution of the time

delay during the curing cycle is analysed and presents different phases. The curing phases

are identified with a unique signature beginning with the mould-composite coupling, passing

through resin reticulation and finishing with composite consolidation. With these recognizable

signatures, the viability of FUT to estimate the composite polymerization evolution is proved.

7. Acknowledgments

The present work is part of the research project “Multi-sensor Instrumentation for Composite

Materials and Structures (I2MC)” financially supported by the Thematic Advanced Research

Network for Aeronautic and Space Sciences & Technologies of Toulouse (RTRA STAE).

M. Torres conveys his special appreciation to the National Council of Science and

Technology of Mexico (CONACYT) for the financial support.

8. References

1. F. Collombet, G. Luyckx, C. Sonnenfeld, Y-H. Grunevald, Y. Davila, M. Torres, X.

Jacob, K-T. Wu, S. Rodriguez, B. Douchin, L. Crouzeix, R. Bazer-Bachi, T.

Geernaert, J. Degrieck, F. Berghmans. “Cure monitoring of an autoclave

manufactured industrial part: Added value of complementary instrumentation”.

Proceedings of the 19th International Conference on Composite Materials. 2013,

Quebec, Canada, 1178-1187.

2. C. Sonnenfeld, G. Luyckx, F. Collombet, Y-H. Grunevald, B. Douchin, L. Crouzeix, M.

Torres, T. Geernaert, S. Sulejmani, K. Chah, P. Mergo, H. Thienpont, F. Berghmans.

“Cure cycle monitoring of laminated carbon fiber-reinforced plastic with fiber Bragg

gratings in microstructured optical fiber”. Proceedings of the 19th International

Conference on Composite Materials. 2013, Quebec, Canada, 3327-3335.

3. Hernández H, Collombet F, Douchin B, Choqueuse D, Davies P, González JL. “Entire

life time monitoring of filament wound composite cylinders using Bragg grating

sensors: II. Process Monitoring”. Applied Composites Materials, 2009; 16/4: 197-209.

4. Mulle M, Zitoune R, Collombet F. “Through-the-thickness material properties

identification in a technological specimen using 3D-DIC and embedded FBG

measurements”. Experimental Mechanics, 2008; 1340-1343.

5. Mulle M, Zitoune R, Collombet F, Olivier P, Grunevald YH. “Thermal expansion of

carbon–epoxy laminates measured with embedded FBGS – Comparison with other

experimental techniques and numerical simulation”. Composites Part A, 2007; 38:

1414-1424.

6. Kim S, Murayama H, Kageyama K, Uzawa K, Kanai M. “Study on the curing process

for carbon/epoxy composites to reduce thermal residual stress”. Composites Part A:

Applied Science and Manufacturing, 2012; 43/8: 1197-1202.

7. Collombet F, Grunevald YH, Zitoune R, Mulle M. “Economical value added of Multi

Instrumented Technological Evaluators for the development of composite civil

aircraft”. Proceedings of the 16th National Journeys on Composites (JNC16),

Toulouse, France, 2009.

8. Grunevald YH. Value Added of MITE toolbox on development of a full composite

railway bogie, comparison with a classical approach. Proceedings of the 7th

International Conference on Rapid Product Development, Saint Dié des Vosges,

France, 2009.

9. Dumonteuil M. “Solution générique pour l'adressage matriciel de micro-actionneurs

thermiques et optimisation de micro-sources thermiques”. PhD Thesis (in French).

Laboratory for Analysis and Architecture of Systems (LAAS). University of Toulouse,

2007.

10. Marty B. “Conception, réalisation et mise en œuvre d’une plateforme

d’instrumentation thermique pour des applications microfluidiques”. PhD Thesis (in

French). Laboratory for Analysis and Architecture of Systems (LAAS). University of

Toulouse, 2009.

11. Lubin J. “Déploiement d'un réseau de capteurs enfouis dans des multi-plis carbone-

époxy pour l'instrumentation in-situ de structures composites pour l'avionique”. PhD

Thesis (in French). Laboratory for Analysis and Architecture of Systems (LAAS).

University of Toulouse, 2012.

12. Kobayashi M, Jen CK. “Piezoelectric thick bismuth titanate/PZT composite film

transducers for smart NDE of metals”. Smart Materials Structures 2004; 13: 951-956.

13. Kobayashi M, Jen CK, Bussiere JF, Wu KT. "High-temperature integrated and flexible

ultrasonic transducers for non-destructive testing”. NTD & E International, 2009; 42:

157-161.

14. Torres M, Crouzeix L, Collombet F, Douchin B, Grunevald Y-H. Mechanical

characterization of an alternative technique to embed sensors in composite

structures: the monitoring patch. Applied Composites Materials, 2012; 19: 379 – 391.