Effect of Thermal Cycling on Composite Honeycomb Sandwich

Structures for Space Applications

Sandesh Rathnavarma Hegde* and Mehdi Hojjati Concordia Centre for Composites, Department of Mechanical, Industrial and Aerospace

Engineering, Concordia University, Montreal, Quebec, Canada * Corresponding author (s_hegd@encs.concordia.ca)

ABSTRACT

The present work is focused on laboratory test, which involves subjecting the sandwich material

made of aramid honeycomb core and carbon fiber reinforced polymer (CFRP) facesheet to thermal

environment ranging from -185°C (cold case, cryogenic) to +150°C (hot case), and then study the

formation/growth of microcracks on the sample. Test plan was developed to subject the samples

to the above mentioned thermal conditions. The samples were then inspected for microcrack

formation at its cross-section The observation was done on two sides for each sample, ribbon

direction side (the side where the cut was made along the ribbon) and cross ribbon direction side

(cut perpendicular to the ribbon) of the core. Microscopic inspection was conducted every half a

cycle (after cold cycle and hot cycle) to observe the part of the cycle that contributes more towards

the formation and growth of microcracks. Microcracks were quantified by parameters such as

crack density and crack length, with the increase in cycles.

INTRODUCTION

Over the past few decades, the use of composites has increased tremendously for a wide range of

purposes. Within this past decade however, we have seen composites being used not only for non-

structural applications but have now become primary structures in many spacecraft and aircraft

designs that are currently in service today. The use of thermoset composites can be advantageous

compared to metallic structures due to their high strength to weight ratio, stiffness and their

resistance to harsh environments provided the correct epoxy and coatings are applied.

Figure 1: Composite honeycomb sandwich construction [1].

Honeycomb core

Adhesive

film

Facesheet

The Carbon Fiber Prepreg Honeycomb Sandwich Panel design are widely used in aerospace

applications considering its contribution towards weight reduction and functionality. It mainly

comprises of a core available in various forms, usually foam and honeycomb, sandwiched between

facesheet bonded with suitable structural adhesive. It is primarily used for space applications such

as spacecraft primary structural elements, in the form of panels and support structures for solar

arrays and antenna. A spacecraft during its operation in space is subjected to thermal fatigue

loading as high as ±185 °C. When the composite material is exposed to extreme temperature

environment microcracks are induced which effects the structural integrity. In the past, similar

study had been conducted to study the effects of microcracks [2-8] but primarily on solid laminates.

This paper is mainly focussed on the study of formation/growth of microcracks and quantification

for composite honeycomb sandwich structure.

MATERIAL AND TEST PLAN

Each facesheet comprises of two laminas, 45º ply on the outer surface and cross ply on the core

side. The sandwich panel was made of 6.4 mm (0.25inch) thick phenolic resin coated Kevlar

honeycomb core with 0.25 mm thick facesheets made of 5HS carbon fiber fabric with cyanate

ester resin. The facesheets were cured at the laminate level and are then bonded to the core using

a modified epoxy film adhesive cured at 120 ºC.

Figure 2: Temperature vs time plot for different rate of cooling a) Direct dipping in LN2 b) Non-contact

cooling.

To achieve cryogenic environment, liquid nitrogen (LN2) was used, to get to higher temperature,

the materials were placed in the convection oven. Two test setups were developed for the cold

case, the first setup involves direct dipping of samples in LN2 (higher rate of cooling), which is

similar to the condition experienced by sandwich structures used in the cryogenic fuel tank [8],

where the cryogenic fuel is in direct contact with the material. The second setup involves non-

contact cooling where samples are subjected to the cryogenic environment without direct contact

with LN2, this condition subjects the samples to the thermal environment (lower rate of cooling),

experienced by the materials used in spacecraft’s primary structures and support brackets in eclipse

region of the orbit. As shown in figure 2a it took close to 20 seconds to reach cryogenic temperature

-200

-150

-100

-50

0

50

100

150

0 500 1000 1500 2000

Tem

per

ature

(ºC

)

Time (Seconds)

-200

-150

-100

-50

0

50

100

150

0 1000 2000

Tem

per

ature

(º

C)

Time (Seconds)

for direct dipping in LN2, and close to 8 minutes to reach the cryogenic temperature in the non-

contact type of cooling.

A total of four test coupons were prepared each of size 25.4 mm by 25.4 mm. Two samples each

were subjected to thermal cycling at the fast and slow rate of cooling, followed by exposure to the

elevated temperature to complete one cycle. The microscopic inspection was done using an optical

microscope, at the cross section after every half cycle, to observe the growth and propagation of

microcracks.

MICROSCOPIC INSPECTION

(a)

(b)

(c)

(d)

Facesheet

Core wall

Adhesive

fillet

Microcrack

Microcrack

Microcrack

(e)

(f)

Figure 3: Microscopic images of cross-section at a) zero cycle b) 10th cycle c) 20th cycle d) 30th

cycle e) 40th cycle f) 60th cycle.

Microscopic images were taken on the ribbon direction and transverse ribbon direction side cross-

section of the sample, this was done in-order to observe the effect of material anisotropy of

sandwich structure on microcrack formation and growth. Figure 3a shows the various aspects of

composite honeycomb sandwich. Figure 3b to 3f presents the microscopic images comprising of

microcrack after subsequent thermal cycling. Most cracks appeared between facesheet and core in

the form of delamination cracks, some small transverse cracks appeared in the 90º tow after 30

thermal cycles.

MECHANICS OF MICROCRACK FORMATION

Microcracks form due to the difference in the CTE between the various elements of the composite

sandwich structure. The difference in CTE gives rise to thermal strain that is maximum during

exposure to cryogenic temperature, compared to elevated temperature, as implied by the equation

below.

αi�( Ajj Ni - Aij Nj ) / ( Aii Ajj - Aij2 )� �i/�� (1)

Equation 1 is the linear in-plane hygrothermal expansion co-efficient for layered structures [9],

where i,j=x,y, the principal directions of laminate, Aij are the in-plane stiffness, Ni are the

hygrothermally induced loads, αi is the CTE and εi are the in-plane strains. Matrix and adhesive

have positive and higher CTE compared to the carbon fiber tows, which has negative CTE, due to

this difference, stress is induced, resulting in micrcoracks. Initially microcracks appeared more on

the ribbon direction side compared to the transverse ribbon direction side, this is mainly due to

more positive CTE in the ribbon direction of the core, compared to transverse ribbon direction.

Microcrack

QUANTIFICATION OF CRACKS

Figure 3: Change in crack density with number of cycles

Figure 4: Change in crack length with number of cycles.

The microcracks were quantified by parameters such as crack density and crack length. Crack

density is the measure of the number of cracks per unit length, and crack length is the summation

of all cracks on the side under observation (ribbon side or transverse) [10]. Figure 3 and figure 4

presents the change in crack density and crack length respectively with the increase in number of

cycles. Sample 1 and 2 were subjected to a faster rate of cooling, and sample 3 and 4 were subjected

to a slower rate of cooling. It can be noticed that there is a marginal difference between, different

rates of cooling, on the crack density and crack length variation with increase in cycles.

It is interesting to notice from the plots, that the cracks grow faster in ribbon direction side

compared to transverse ribbon direction side. The crack density values saturate for all samples

after 30 thermal cycles, however, the crack length values saturate after 30 thermal cycles only for

sample 1 and 2, which were subjected to a faster rate of cooling compared to sample 3 and 4 which

were subjected to slower rate of cooling.

CONCLUSION

Most microcracks appeared between facesheet and core in the form of delamination crack. A

methodology of quantifying microcracks on sandwich composite material subjected to thermal

cycling is developed. Samples were subjected to different rates of cooling to observe the effect of

the same on the formation/growth of microcracks, it was found that the difference is marginal.

Microcracks formation and growth saturates after 30 thermal cycles indicating no further need to

perform more thermal cycles.

ACKNOWLEDGEMENT

The authors would like to acknowledge the financial support from the Natural Sciences and

Engineering Research Council of Canada (NSERC), Consortium de Recherche et D’innovation en

Aérospatiale au Québec (CRIAQ), MDA Corporation and Stelia North America.

REFERENCES

[1] Online: http://www.fibre-reinforced-plastic.com/2010/12/sandwich-composite-and-core-

material.html

[2] Q. Yu et al, "Effects of vacuum thermal cycling on mechanical and physical properties of high

performance carbon/bismaleimide composite," Mater. Chem. Phys., vol. 130, pp. 1046-1053,

2011.

[3] I. Tsukrov, B. Drach and T. Gross, "Effective stiffness and thermal expansion coefficients of

unidirectional composites with fibers surrounded by cylindrically orthotropic matrix layers," Int.

J. Eng. Sci., vol. 58, pp. 129-143, 2012.

[4] J. Sanborn and D. Morel, "Effects of thermal cycling on the mechanical and physical properties

of a space qualified epoxy adhesive," J Reinf Plast Compos, vol. 7, pp. 155-164, 1988.

[5] C. Henaff-Gardin and M. Lafarie-Frenot, "Specificity of matrix cracking development in CFRP

laminates under mechanical or thermal loadings," Int. J. Fatigue, vol. 24, pp. 171-177, 2002.

[6] T. Shimokawa et al, "Effect of thermal cycling on microcracking and strength degradation of

high-temperature polymer composite materials for use in next-generation SST structures," J.

Composite Mater., vol. 36, pp. 885-895, 2002.

[7] Gates, T. S., Su, X., Abdi, F., Odegard, G.M., Herring, H. M. “Facesheet delamination of

composite sandwich materials at cryogenic temperature”, Elsevier Ltd, Composites Science and

Technology, 2006; 66: 2423-2435.

[8] Brian W. Grimsley et al., “Hybrid Composites for LH2 Fluid Tank Structure”, NASA

Technical Report Server, https://ntrs.nasa.gov/search.jsp?R=20040086019.

[9] H. Chen et al., “Thermal Expansion of Honeycomb Sandwich Panels”, Reprinted from Thermal

Conductivity 25- Thermal Expansion 13-June 13-16,1999.

[10] Hegde. S. R, Hojjati. M, “Effect of Microcracks on Mechanical Property of Honeycomb

Sandwich Composite Subjected to Thermal Cycling”, SAMPE Conference, Long Beach,

California, USA, May 2018.