Effect of bubble based degradation on the physical properties of

See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/306262439

Effect of bubble based degradation on the physical properties of Single Wall Carbon Nanotube/Epoxy Resin composite and new... Article in Composites Part A Applied Science and Manufacturing · August 2016 DOI: 10.1016/j.compositesa.2016.08.015

CITATIONS

READS

2

103

2 authors: Seyyed Alireza Hashemi

Seyyed Mojtaba Mousavi

Shiraz University of Medical Sciences

Shiraz University of Medical Sciences

13 PUBLICATIONS 8 CITATIONS

30 PUBLICATIONS 12 CITATIONS

SEE PROFILE

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

ELECTRON BEAM View project

EPOXY AND UNSATURATED POLYESTER NANOCOMPOSITE View project

All content following this page was uploaded by Seyyed Mojtaba Mousavi on 06 October 2017.

The user has requested enhancement of the downloaded file.

Composites: Part A 90 (2016) 457–469

Contents lists available at ScienceDirect

Composites: Part A journal homepage: www.elsevier.com/locate/compositesa

Effect of bubble based degradation on the physical properties of Single Wall Carbon Nanotube/Epoxy Resin composite and new approach in bubbles reduction Seyyed Alireza Hashemi a,⇑, Seyyed Mojtaba Mousavi b a b

Department of Mechanical Engineering, Najafabad Branch, Islamic Azad University, Najafabad, Iran School of Chemical and Petroleum Engineering, Shiraz University, Shiraz 71345-1789, Iran

a r t i c l e

i n f o

Article history: Received 10 June 2016 Received in revised form 27 July 2016 Accepted 16 August 2016 Available online 17 August 2016 Keywords: Degradation Void Bubble Composite Single Wall Carbon Nanotube

a b s t r a c t In this study the effect of bubble based degradation on the physical and structural properties of Single Wall Carbon Nanotube (SWCNT)/epoxy resin composite samples were investigated and new method based on vacuum shock was presented. For this purpose, with two different methods samples with and without bubble based degradation were fabricated and effect of degradation on the value of electrical conductivity and the amount of Electromagnetic (EM) waves absorption were investigated. Which vacuum shock technique can improve above mentioned properties about 58319.594 and 63.921 percentage for sample without degradation in comparison with destroyed sample due to the bubbles based voids effect. Moreover, the main factors in the bubbles formation and migration during the manufacturing process and their behavior in the matrix with the help of optical and SEM images were examined and their effect on structural properties of composite samples with Micro Raman Spectroscopy was evaluated. Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction Bubbles and their resulting voids are among most destructive factors during the fabrication of composite materials that can lead to significant decrease in overall properties and huge destruction in composite samples structure due to increase in gas concentration of bubbles. Creation of bubbles depend on the balance between their inner and outer pressure. Also bubbles growth and formation may occur due to diffusion of air or by agglomeration with other bubbles. Moreover, among the most effective factors that can affect bubbles mobility, we can refer to bubbles size, viscosity of polymeric matrix, fluid surface tension and fillers network configuration. Despite of that, simultaneous usage of vacuum and vibration with low frequency about 10–30 Hz can highly reduce the overall amount of bubbles and their size in the matrix [1]. In addition, many factors such as temperature, time, pressure, heating transfer and curing process conditions are involved in bubbles reduction [2,3]. As well as above mentioned conditions, molecularly disperse of bubbles into the polymeric matrix can lead to reduction in overall amount of bubbles in the whole structure [3]. In addition, some production situations such as operating time, ⇑ Corresponding author. E-mail addresses: [email protected] (S.A. Hashemi), kempo.smm@gmail. com (S.M. Mousavi). http://dx.doi.org/10.1016/j.compositesa.2016.08.015 1359-835X/Ó 2016 Elsevier Ltd. All rights reserved.

concentration and flow rate of the surfactant and superficial gas velocity can affect bubbles size and air hold up in the suspension. Despite of that, superficial velocities and surfactant concentration can lead to generation of medium-sized bubbles and increase in air hold up values [4]. Besides, surface reaction and gas transfer rate from liquid to bubbles interface are two main reasons for bubbles growth during photo electrochemical and electrochemical conversions respectively. In spite of that, bubbles growth rate for photo electrochemical conversion and electrochemical conversion are due to small effective solid surface engagement and big effective solid surface participating [5]. Moreover, bubbles have a strong desire to combine with each other and create larger bubbles that can cause huge destruction due to increase in their inner pressure after firs curing step of composite samples [6]. In addition, bubbles can grow due to the diffusion enhanced by the Thomson-Freundlich effect and pressure oscillation [7]. In aerospace and other industries, bubbles and their resulting voids content would be better not to be more than 1 and 5 wt% respectively [8,9]. Furthermore, the presence of bubbles and their resulting voids in a large numbers and with large diameters as well as increase in volume fraction, can act as a stress concentration areas and lead to reduce in the mechanical properties of composite samples include tensile, flexural and shear strength and also bending, tension and fracture toughness [3,10–13]. Despite of that,

458

S.A. Hashemi, S.M. Mousavi / Composites: Part A 90 (2016) 457–469

among this mechanical properties, voids can seriously affect shear and compression strength [14]. Moreover, cure cycle can highly affect the mechanical properties of composite samples. By means of ILSS tests, for void content about 1 wt% the ILSS decreased due to the voids volume fraction [15]. Also increase in the overall amount of temperatures and pressures during the cure cycle can affect the bubbles content and size distribution. In addition, increase in the magnitude of applied pressure to the composite samples and also removal of the gaps inside the mold can lead to significant decrease in the overall amount of bubbles content and their average diameter [16]. Besides, cure cycles can affect both bubbles content and mechanical properties of composite samples. Furthermore, increase in the pressure can lead to decrease in the amount of bubbles and their resulting voids content in the matrix [13]. In a work by Li et al. [3] they have investigated the formation mechanism of voids and also their effect on the mechanical properties of composite samples. According to their study, the amount of bubbles volume fraction in the composite samples decreased due to increase in the magnitude of cure cycle pressure. Furthermore, increase in the amount of cure cycle temperature had no effect on the amount of bubbles inside the matrix context, but it can affect the location of the bubbles. In addition, cycle temperature can affect the amount of bubble based voids, dimensional structure and spatial distribution of voids in the matrix. Besides, elongation factor of bubble based voids that is based on their length and average diameter will increase due to voids size and volume. Despite the fact that most of the voids were the result of air entrapment and wrinkles. Furthermore, higher pressure in regions can lead to the migration of resin as well as bubbles in the polymeric matrix. In a work that had been conducted by Nie et al. [2], they have investigated the bubbles behavior in the polymeric matrix and also bubbles formation reasons. Then they have compared experimental and theoretical results with each other. Some experimental and production methods indicates that bubbles can shrink and also disappear due to certain conditions. In this situation, bubbles will disperse into the polymeric matrix molecularly and the stress concentration due to these bubbles will not exist any longer. Moreover, the gas diffusion is very important in bubbles formation. Gas molecules are migrating from high concentration zones toward lower ones and also the bubbles volume fraction can change due to the migration of gas molecules. In addition, by increase in the temperature and pressure and decrease in the viscosity of polymeric matrix, the amount and average diameter of bubbles and their resulting voids will decrease [2]. Furthermore, bubbles and their resulting voids can be formed at the bond interface due to incomplete flowing of the matrix and gas remnant from the neighbor polymer may fill empty spaces to form gas bubbles [2,17,18]. In another works by DeValve et al. [19] and Frishfelds et al. [20], they have simulated the bubbles formation in the liquid composite and also their motion through non-crimp fabrics respectively. Based on their study, bubbles can highly affect and distribute the structural shape of composite samples. Moreover, while bubbles reaching the resin flow front, their mobility will increase significantly. The highest bubbles mobility occur while the bubbles are near to the resin flow fronts. When the flow cell has a moving flow front, the bubbles mobility was found to be greater than saturated flow cells. This is due to the dominance of the surface tension at flow front while bubbles approaching to the flow front. Also decrease in the thickness of the resin substrate can lead to increase in the migration speed of the resin flow and thus the migration speed of bubbles in the flow front [21]. In addition, natural fillers in the composite structure can increase the speed of degradation in the composite structure [22]. In this study for better investigation about the impact of bubbles and their resulting voids on the physical properties of composite samples such as electrical conductivity, Electromagnetic (EM)

waves absorption and composite structural shape, different production methods were used for fabrication of composite samples with and without bubbles and their resulting voids in the whole composite structure. Then composite samples containing Single Wall Carbon Nanotubes (SWCNTs) with and without voids were fabricated and compared with each other and effect of bubble based voids on the above mentioned properties of composite samples were examined. Then the behavior of bubbles before and after fully curing of composite samples were examined and samples structure were evaluated with SEM and optical images and Micro Raman Spectroscopy. Bubbles and their resulting voids are among most destructive factors in the fabrication of composite samples, this study was conducted for better investigation about bubbles behavior and their creation reasons in order to present a multistep method based on the vacuum shock technique for removal of bubbles and their resulting voids from the composite samples. 2. Experimental section 2.1. Materials and instruments In this study SWCNTs was Supplied by US Research Nanomaterial. Average external and inner diameter and length of these SWCNTs are about 1–2 nm, 0.8–1.6 nm and 5–30 lm respectively. The purity of these SWCNTs is more than >96%. Moreover, for better dispersion of these SWCNTs in the polymeric matrix, ACOOH (Carboxyl) functional group were applied on their surface by acidic method. Other specification of these SWCNTs can be seen in Table 1. In addition, NPEL-128 (Nan Ya-128) epoxy resin and Acetone were used as polymeric matrix and dispersant agent respectively. The epoxy resin and Acetone dispersant agent were supplied by Nan Ya Plastics Corporation and Merck respectively. Moreover, for measuring the amount of Electromagnetic waves absorption and electrical conductivity, Vector Network Analyzer (manufactured by ROHDE & SCHWARZ) by frequency range between 10 MHz and 20 GHz and IV-CV parameter analyzer (model K361, manufactured by Keithy) were used. Also for examination of composite samples Field Emission Scanning Electron Microscope (FESEM) (model HITACHI S-4160) analysis, 1000 Digital Microscope (manufactured by Rohs) and Micro Raman Spectroscopy (model SENTERRA (2009) manufactured by BRUKER CO) were used. 2.2. Composite samples production methods In this study, for the production of composite samples with and without degradation that was caused by bubbles and their resulted

Table 1 Specification of SWCNTs that had been used in this study. SWCNTs specification

Measurement instruments

OD

1–2 nm

ID

0.8–1.6 nm

Average diameter Length Tap density True density SSA Ash Electric conductivity Thermal conductivity Color Ig/Id

1.1 nm 5–30 lm 0.14 g/cm3 2.1 g/cm3 >380 m2/g 100 S/cm 50–200 W/m K Black >9

From HRTEM and Raman Spectroscopy From HRTEM and Raman Spectroscopy From Raman Spectroscopy TEM – – BET HRTEM and TGA – – – Raman Spectroscopy


voids, two different manufacturing methods were designed and used. By these two methods, effect of bubbles on the suspension before curing and their resulting degradation after completion of first curing step on the physical properties of composite samples such as Electromagnetic (EM) wave absorption, electrical resistivity, volume resistivity and electrical resistance as well as structural properties were investigated. The first step for the production of composite samples containing bubbles and defects is as follows: In the first production method for fabrication of composite samples containing large amount of voids and defects, first SWCNTs were dispersed in the Acetone solvent for 1 h by ultra-sonic mixer under 120 W in order to transform the agglomerated bundles of SWCNTs into individual ones. Then the SWCNTs/Acetone suspension and epoxy resin were poured into the Vacuum Erlenmeyer Flask. Then the suspension was stirred with magnet under vacuum (25 cmHg) for 4 h in order to decrease the level of Acetone in the SWCNTs/Acetone/epoxy resin suspension. Then the curing agent was added to the SWCNTs/Acetone/epoxy resin suspension and the suspension was poured into the mold until the curing process getting done. In this stage, due to removal of bubble reduction step and remain of Acetone solvent in the above mentioned suspension, a large number of bubbles will appear in the composite samples that can lead to huge destruction in the structural shape of composite samples due to increase in gas concentration and inner pressure of bubbles [2]. A view of this process can be seen in Fig. 1. In the second manufacturing process, for the fabrication of composite samples without or with very low amount of bubble based voids and defects, a multi stage manufacturing process was used. For this purpose, first the SWCNTs were put into the humidity absorbing chamber for further humidity reduction. Humidity is a destructive factor in the fabrication of composite samples that can lead to agglomeration of fillers and decrease in the overall

459

properties of composite samples. Also humidity can increase the amount adhesive force between the Carbon Nano Tubes (CNTs) due to increase in the Van der Waals forces that can lead to agglomeration of CNTs in the suspension [23]. Then in order to transform the agglomerated bundles of SWCNTs into the individual ones, SWCNTs were mixed in the Acetone solvent by ultra-sonic mixer for 1 h under 120 W power and with 45 °C temperature limit. Ultra-sonic mixer can provide cavitation that can lead to homogenous dispersion of the fillers in the matrix. Then the SWCNTs/Acetone suspension and epoxy resin were poured into the Vacuum Erlenmeyer Flask and mixed by magnet under vacuum shock between magnitude 25–60 cmHg for 3 h. Then for fully reduction of the Acetone from SWCNTs/Acetone/epoxy resin suspension, 70 °C heat was added to the previous stage conditions and then the process was continued until 1 h. Then the resulting suspension was poured into the beaker for accurately calculation of the curing agent weight percentage with respect to the epoxy resin weight. Besides, due to the high viscosity of the resulting suspension, some of it will attach to the container and if the curing agent added to the above mentioned suspension more than certain amount of 100:58, this can lead to increase in the amount of bubbles and make the composite samples brittle. Afterward, for further degassing and bubbles reduction, the curing agent and SWCNTs/epoxy resin suspension were placed into the vacuum oven separately for 2 h under vacuum shock between magnitude of 10–60 cmHg. In the next stage, the curing agent was added to the SWCNTs/epoxy resin suspension and then the resulting suspension was poured into the mold. In this stage the vacuum shock technique was used for further bubbles reduction. As mentioned in Section 1, highest bubbles mobility occur while they are near to the resin flow front [21,24]. In the vacuum shock technique, negative pressure of vacuum was added and then removed constantly, a view of this process can be seen in Fig. 2. This tech-

Fig. 1. First manufacturing method for fabrication of composite samples full of defects and bubble based voids. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

460


Fig. 2. A view of vacuum shock technique. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

nique forced the bubbles inside the suspension to move toward the surface and while the bubbles reaching the resin flow front, their mobility will increase and thus all the trapped airs and bubbles can be removed from the bottom of the cast and suspension respectively. The aim of this technique is to increase in the bubbles mobility for their faster reduction. While the viscosity of the resin increase, bubbles could not move toward the surface easily, so by adding and removing the negative pressure on the samples, it can lead to the movement of bubbles toward the surface and by doing this constantly, bubbles movement from the bottom and all other locations of the samples toward the surface will occur. By removing the vacuum negative pressure and due to the high viscosity of resin, bubbles will not turn back to their primary location and thus with this technique, step by step movement of the bubbles toward the surface can be achieved. Moreover after these steps, for the first and second curing steps, composite samples were put at room temperature for 4 h and then

in heat oven under 80 °C for 1 h respectively. A view of second production method can be seen in Fig. 3. Moreover there is a good method for ensuring about fully reduction of Acetone from the SWCNTs/Acetone/epoxy resin suspension. As can be seen in Fig. 3 step 3–4, if the magnitude of negative pressure goes more than 60 cmHg, it can lead to increase in the Acetone evaporation speed and also bubbles reduction. After step 4 and in order to be ensure about fully reduction of Acetone from the suspension, if the vacuum magnitude goes more than 60 cmHg and there were no bubbles on the surface of the suspension, it means that all of the Acetone solvent was removed from the suspension and the resulting suspension is ready for the next step. A view of this step can be seen in Fig. 4. In addition, for measuring the effect of bubbles and their resulting voids on the amount of Electromagnetic waves absorption by SWCNTs/epoxy resin Composite in the X-band (8.2–12.4 GHz), Vector Network Analyzer by frequency range between 10 MHz

Fig. 3. A view of production method. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

461

S.A. Hashemi, S.M. Mousavi / Composites: Part A 90 (2016) 457–469 Table 2 Specification of composite samples.

Fig. 4. A view of step 3 and 4 bubble reduction mechanism. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

and 20 GHz was used. For this purpose, first the instrument was calibrated with three usual calibration method includes math load, short load and open load. In the math load calibration step, with some calibration instrument, Vector Network Analyzer was calibrated through the coaxial cables for precise calculation of absorption rate. In the short load and open load calibration processes, the waveguide was calibrated. Then once the amount of Electromagnetic waves absorption was measured without putting any sample in the waveguide and then the achieved numbers was saved. Then the samples were put into the waveguide one by one and then the value of Electromagnetic waves was measured for each one of them. Afterward, the amount of Electromagnetic waves absorption for each one of the samples were subtracted from the value of Electromagnetic waves absorption that was achieved without putting any sample in the waveguide. Resulting data shows the amount of Electromagnetic waves absorption by each one of samples in the X-band frequency range. Despite of that, for measuring the amount of electrical conductivity of composite samples, the IVCV parameter analyzer (model K361 that had been manufactured by Keithy) was used. In the first step, various amount of voltage from range of 100/+100 V was applied to the samples from longitudinal direction and then the amount of electrical resistance, volume resistivity and electrical conductivity were measured. 2.3. Physical properties 2.3.1. Electromagnetic waves absorption In this section effect of bubble based voids on the amount Electromagnetic (EM) wave absorption were investigated. In Table 2 specification of composite samples can be seen. In addition, dimension of samples is based on WR-90 Standard (0:9 in: 0:4 in: ¼ 2:286 cm 1:016 cm) and their thickness is about 0.5 cm (0.196 in.). In Fig. 5 the amount of Electromagnetic waves absorption of samples A-D can be seen. In addition, manufactured composite samples that had not been disposal under vacuum shock after molding step, had some voids at their bottom that is due to air entrapment and removal of bubble reduction steps based on vacuum shock technique. The average amount of Electromagnetic waves absorption for all samples in the whole X-band can be seen in Table 3. As can be seen in Table 3, the average amount of Electromagnetic waves absorption due to the destruction of sample with bubble based voids (for sample A in comparison with sample B), decrease about

Sample code

SWCNTs weight percentage

Sample condition

A

0.01 wt%

B

0.01 wt%

C

0.01 wt%

D

0.0 wt% – (pure resin)

E

0.0 wt% – (pure resin)

With very low amount of defect or voids/manufacturing method 2 Sample was destroyed by bubble based voids with large diameter/manufacturing method 1 Some voids with low diameter exist at sample bottom and cross section/manufacturing method 2 without putting in the vacuum chamber after pouring in the mold and usage of vacuum shock technique With very low amount of defect or voids/manufacturing method 2 Affected by Acetone/with large amount of bubble based voids in the whole structure/manufacturing method 1

63.921%. In addition, appearance of void at the bottom of sample can reduce the average amount of absorption about 3.102% (for sample A in comparison with sample C). Besides, the amount of increase in the average absorption for samples A, B and C in comparison with sample D are about 33.818%, 20.780%, 29.791% respectively. This results shows that bubbles based degradation can highly affect the overall amount of Electromagnetic waves absorption, thus we would be better avoid this bubbles in the whole structure in order to gain a structure without voids. Besides, by comparison between samples A, B and C it can be concluded that increase in the average diameter of bubbles can decrease the overall amount Electromagnetic waves absorption. Furthermore, bubble based degradation can lead to significant decrease in overall amount of Electromagnetic waves absorption. These overall amount of Electromagnetic waves absorption for sample B is even lower than sample D (sample containing pure epoxy resin). These results shows that bubbles and their resulting voids can highly affect the overall amount of absorption rate. Despite of that, absorption loss is depend on the physical characteristics of the composite layer and is independence from the type of the electromagnetic source. In addition, while an electromagnetic wave passes through a composite layer, its amplitude decrease exponentially. This absorption loss is due to the induced current in the composite layer that can generate ohmic losses and heating of the material, which can be expressed as E1 ¼ E0 et=d and H1 ¼ H0 et=d respectively [25]. Besides, the required distance for the waves to be attenuate is defined as 1=e or 37% of skin depth. Furthermore, the absorption term of SEA can be expressed as follow [25]:

SEA ¼ 20ðt=dÞ log e ¼ 8:69ðt=dÞ ¼ 131 t

qffiffiffiffiffiffiffiffiffi f lr

ð1Þ

which t, f, l, r and d are the thickness of the shield in mm, frequency in MHz, relative permeability, electrical conductivity relative to copper and skin depth, respectively. In spite of that, skin depth can be expressed as follow [25]:

1 d ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffi pf lr

ð2Þ

Furthermore, the skin depth effect is very important for lower frequencies where the field are more likely to be predominantly magnetic with lower wave impedance than 377 X. For better

462

S.A. Hashemi, S.M. Mousavi / Composites: Part A 90 (2016) 457–469 8

A C Log. (A) Log. (C)

7

B D Log. (B) Log. (D)

Absorption (dB)

6 5 4 3 2 1

8.00E+20 8.10E+20 8.20E+20 8.30E+20 8.40E+20 8.50E+20 8.60E+20 8.70E+20 8.80E+20 8.90E+20 9.00E+20 9.10E+20 9.20E+20 9.30E+20 9.40E+20 9.50E+20 9.60E+20 9.70E+20 9.80E+20 9.90E+20 1.00E+21 1.01E+21 1.02E+21 1.03E+21 1.04E+21 1.05E+21 1.06E+21 1.07E+21 1.08E+21 1.09E+21 1.10E+21 1.11E+21 1.12E+21 1.13E+21 1.14E+21 1.15E+21 1.16E+21 1.17E+21 1.18E+21 1.19E+21 1.20E+21

0

Frequency (Hz) Fig. 5. Comparison between samples A–D. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Table 3 The average amount of Electromagnetic waves absorption for samples A–D. Samples code

Average amount of Electromagnetic waves absorption (dB)

A B C D

3.659 2.166 3.549 2.734

absorption rate, composite sample should containing high electrical conductivity and permeability with sufficient thickness to achieve required number of skin depth [25]. Despite of that, creation of bubble based voids can decrease the thickness of the material as well as the effect of skin depth. Decrease in these two factors can highly affect the overall amount of Electromagnetic waves absorption rate, a view of this phenomenon can be seen in Fig. 6. Furthermore, as can be seen in Eq. (1), electromagnetic absorption rate is highly depend on the skin depth and decrease in the skin depth due to the creation of bubble based voids and decrease in available thickness of the absorber layer can highly decrease the SEA .

2.3.2. Electrical conductivity In this section the amount of electrical resistance, volume resistivity and electrical conductivity were measured and presented. For this purpose, then the average amount of electrical resistance between range of 36/+36 V was calculated, because if the amount of voltage goes more than 36/+36 V, SWCNTs will move or vibrate on their position and this phenomenon will shows itself by vibration in the electrical resistance graph, a view of this phenomenon can be seen in Fig. 7. After measuring the amount of electrical resistance, the volume resistivity and then electrical conductivity of composite samples were calculated based on the formula (3) and (4).

q¼

r¼

RA L

ð3Þ

1

ð4Þ

q

which q, A, L and r are volume resistivity, surface area, distance between to crocodile clips and electrical conductivity, respectively. In addition, dimensions of samples is according to the dimensions related to the previous section (0:9 0:4 0196 in:3 ¼ 2:286 1:016 0:5 cm3 ). In this case, A and L related to Eq. (3) are 2:286 1:016 cm2 and 2:286 cm respectively. Based on the above mentioned method, the amount of electrical resistance, volume resistivity and electrical conductivity of composite samples were measured and can be seen in Figs. 8–10 respectively.

6.00E+09

Electrical Resistance (Ω)

5.00E+09 4.00E+09 3.00E+09 2.00E+09 1.00E+09 0.00E+00 -1.00E+09 -2.00E+09 -3.00E+09 -4.00E+09 -100 -88 -76 -64 -52 -40 -28 -16 -4 8 Voltage

Fig. 6. Part 1 shows the absorption phenomenon in an absorber layer and part 2 shows the effect of bubble based voids on thickness of the absorber layer. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

20 32 44 56 68 80 92

Fig. 7. Vibration in the electrical resistance graph for sample containing 0.01 wt% SWCNT, as can be seen in this figure, by increase in the value of applied voltage more than range of 36/36 V, the amount of Vibration will increase significantly. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Electrical Resistance (Ω)

S.A. Hashemi, S.M. Mousavi / Composites: Part A 90 (2016) 457–469 2.00E+08 1.80E+08 1.60E+08 1.40E+08 1.20E+08 1.00E+08 8.00E+07 6.00E+07 4.00E+07 2.00E+07 0.00E+00

1.85E+08

7.58E+07

3.03E+07 9.91E+06

3.17E+06

A

B

C

D

E

Sample Code

Volume Resistivity (Ω.Cm)

Fig. 8. The average amount of electrical resistance for composite samples between range of 36/+36 V. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

2.00E+08 1.80E+08 1.60E+08 1.40E+08 1.20E+08 1.00E+08 8.00E+07 6.00E+07 4.00E+07 2.00E+07 0.00E+00

1.88E+08

7.70E+07

3.08E+07 1.01E+07

3.22E+06

A

B

C

D

E

Sample Code

Electrical Conductivity (S/Cm)

Fig. 9. The amount of volume resistivity of composite samples. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

3.50E-07

3.11E-07

3.00E-07 2.50E-07 2.00E-07 1.50E-07 9.93E-08

1.00E-07 3.24E-08

5.00E-08

1.30E-08

5.32E-10

0.00E+00 A

B

C

D

E

Sample Code Fig. 10. The amount of electrical conductivity of composite samples. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

As can be seen in Figs. 8–10, bubble based voids degradation in composite samples can highly affect the electrical properties of composite samples. By comparison between samples A and B, it can be seen that bubble based voids have a great role in the physical properties of composite samples. The amount of electrical conductivity for sample A in comparison with samples B and C increased about 58319.594% and 858.094% respectively. Moreover, the electrical conductivity of sample containing pure epoxy resin that had been affected by Acetone (sample E) increased about 664.826% in comparison with sample containing pure epoxy resin (sample D). Also addition of Acetone to the epoxy resin and curing agent (without removing it) can lead to change in the color and thus change in the chemical properties of the final sample that can lead to increase in the amount on electrical conductivity and also bubble based voids in comparison with the sample containing pure epoxy resin without any Acetone. It is interesting to note that the amount of electrical conductivity of pure epoxy resin that had been affected by Acetone and with presence of large amount of bubble based voids in the whole structure is much higher than

463

the sample containing pure epoxy resin without any Acetone. But for the production of composite samples containing CNTs, we would better to remove all of the Acetone solvent from the suspension by the manufacturing method-2, because presence of the Acetone at the cold curing step can affect the curing agent and lead to change in the color and amount of bubbles in the whole structure. In addition, the amount of electrical conductivity of samples A, B and C were increased about 2293.920%, 99.590% and 149.862% in comparison with pure epoxy resin (sample D) respectively. These results shows that the value of electrical conductivity for sample B is even lower than the sample containing pure epoxy resin (sample D). So based on the above mentioned results, significant effect of bubble based voids on the electrical conductivity of samples can be seen. Effect of these bubble based voids on the electrical properties of composite samples is much higher than their effect on the composite samples absorption rate. 2.3.3. Structural examination In this section, behavior of bubbles in the composite structure were examined. There are several factors that can cause and affect the bubbles formation and mobility such as matrix viscosity, curing step temperature and pressure, thickness of composite samples, mold surface shape, air diffusion, matrix surface tension, filler configuration and inner & outer pressure of bubbles. As mentioned in introduction, bubbles tend to migrate from high concentration zones toward lower ones. Also due to their surface tension, bubbles tend to attach and combine with each other. This phenomenon can lead to creation of bubbles with high diameter and thus severe destruction in composite structure. This destruction is due to increase in the inner gas concentration of bubble based voids that can lead to crack growth and brittle fracture. In Fig. 11, this phenomenon can be seen. As can be seen in this figure, number 1 shows bubbles combination and also a bridge between two bubbles border shows that the bubbles tend to attract each other in order to create bigger bubble. Fig. 11 numbers 2, 3 and 4 shows the bubbles desire in migration toward less concentration zones and also their combination with each other. Fig. 11 numbers 5 and 6 shows the combination of bubbles with each other and creation of bubbles with higher diameter. Fig. 11 number 7 shows the migration of bubbles from higher concentrated area toward lower ones (huge bubbles are the result of small size bubbles combination and also they show the high concentrate areas). Fig. 12 shows the border between bubbles that can lead to bubbles combination and thus creation of huge size bubbles. Huge size bubbles can act as a stress concentration areas that can lead to severe degradation and crack growth along their radial direction.

Fig. 11. Bubbles behavior in the polymeric matrix. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

464


Fig. 12. Border between the bubbles. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Moreover, pure epoxy resin has yellow color and by adding Acetone solvent to the epoxy resin by presence of curing agent, it can change the color of resin to orange and also lead to creation of large amount of bubbles in the whole structure. Therefore, before adding the curing agent to the epoxy resin, it would be better to remove the Acetone solvent from the suspension completely. But it is worth mentioning that by presence of too many voids inside the pure epoxy resin that were created due to presence of the Acetone, the electrical conductivity of affected sample is higher than the pure epoxy resin that was affected by Acetone. This can be due to the change in the chemical properties of epoxy resin. In Fig. 13, structure of pure epoxy resin after fully curing with and without Acetone can be seen. Besides, fillers configuration has a great role in bubbles and void formation. As mentioned in the production section, in order to transform agglomerated bundles of SWCNT into individual ones, SWCNTs were mixed in the Acetone solvent with ultra-sonic mixer for 1 h. Individual SWCNTs have higher interaction with matrix in comparison with agglomerated bundles that can lead to increase in the matrix viscosity. By increasing in the suspension viscosity, bubbles and trapped airs could not leave the suspension and thus the air diffusion will increase. Increase in the air diffusion will increase the bubbles formation rate that can lead to severe destruction in the composite structure due to the increase in the bubbles diameter and their internal gas concentration. A view of this phenomenon can be seen in Fig. 14. In addition, bubble based voids and fillers configuration with each other can lead to severe destruction in the fully cured composite samples. A view of samples A and B can be seen in Fig. 15.

Moreover, mold with non-smooth surface and scrape can lead to air entrapment and thus bubbles formation at the bottom of the composite samples. So regarding to that and due to the destructive factors of bubble based voids on the composite structure, it would be better to be sure about the mold quality and also reduction of bubbles from bottom of the suspension before pouring the suspension into the cast. This goal can achieved due to the vacuum shock technique as mentioned in production section. A view of this problems can be seen in Fig. 16. Moreover, the overall amount of bubble based voids in the surface and cross section of the samples A-E can be seen in Fig. 17. Besides, in Fig. 17 volume of all samples are 0:9 0:4 0196 in:3 ¼ 2:286 1:016 0:5 cm3 (Lengthwidththickness). Lowest amount of voids, related to the samples A and D, in these two samples voids number and also their diameter is very small and have no considerable effect on the composite samples properties. Highest amount of bubble based voids exist in the sample C but these samples has higher electrical properties and absorption rate than sample B and E that is due to small diameter of voids. It is true that sample C has much higher voids, but these voids have very small diameter. But in sample B and E, diameter of bubble based voids is very higher than other samples. This results indicates that increase in the diameter of bubble based voids due to their combination can lead to significant decrease in physical properties of sample and severe destruction in composite sample structure. Besides, effect of increase in diameter of voids on the overall properties of composite samples is much higher than effect of voids number and distribution along the composite structure. Furthermore, order of voids diameter values for composite samples are as follow:

B>E>C>A¼D

Fig. 13. A view of pure epoxy resin with and without bubble based voids in the whole structure that is due to remain of Acetone in the suspension in the presence of curing agent. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Based on the gas diffusion theory, the main driving force of the gas diffusion in polymeric matrix is the concentration gradient. Bubbles are tend to migrate from high concentrated areas toward lower ones and combining with each other in order to create bigger bubbles. C 0 , C s and C 1 are the gas concentration inside of the bubble, the concentration on the interface between the bubble and matrix and concentration of the bulk polymer respectively. In addition, C 0 is much higher than gas solubility and is constant under invariable temperature and pressure. Besides, C s is the saturated concentration and is much higher than C 1 . According to the gas diffusion theory, dissolved gas in the interface between the bubble and matrix (C s ) will flow into the matrix. Moreover, at the same time, existed gas in the bubble will diffuse into the interface in order to complete the gas flowing. Furthermore, by continuing this process, the amount of dissolved gas in the matrix will increase and it will continue until C 1 be equals to C s that can lead to the bubble collapses [2]. This phenomenon can be reversed and lead to increase in the diameter and gas concentration inside the bubble. By bubbles com-


465

Fig. 14. Transformation of the SWCNTs agglomerated bundle into the individual ones can lead to increase in interaction between the resin and CNTs and thus increase in the viscosity of final suspension. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 15. Samples with and without bubble based voids, right side shows the surface area of samples and left side shows the cross section of samples. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

bination, bigger size bubble will create and this can lead to increase in the gas concentration (C 0 ) as well as increase in the diameter of the bubble. Diffusion driving force due to the concentration gradient can be predicted by Fick’s law. In a work that had been conducted by Nie et al. [2], they have predicted bubbles behavior in the radial direction according to the Fick’s second law. Based on their study, relation between the bubble diameter and gas concentration is as follow:

R¼

sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi D R20 ðC s C 1 Þt

q

ð5Þ

In Eq. (5), R is the ultimate diameter of the bubble, R0 is the initial diameter of the bubble, D is the diffusion coefficient for Fick’s second law, q is the gas density within the bubble in invariable temperature and pressure and t is the passed time for gas diffusion. Based on Eq. (1), increase in the diameter of bubbles due to their combination can lead to increase in the gas concentration inside the bubble (C 0 ). In addition, increase in the gas concentration inside the bubble can lead to increase in the inner pressure of the bubble and this phenomenon will provide the overcoming possibility for C 0 against C s . Besides, increase in the inner pressure of

the bubble will increase the pressure on the interface areas around the bubble that can lead to crack growth and severe destruction in the composite structure. A view of this phenomenon can be seen in Fig. 18. Moreover, severe destruction due to this phenomenon around the bubble based voids can be seen in Fig. 19. In addition, bubbles effect and their resulting degradation occurring in two different stages. Bubbles will from due to the air diffusion in the suspension (in liquid state). Furthermore, formation of bubbles and their behavior have different effect on the suspension (in liquid state). While bubbles creating, they tend not only to migrate from high concentrated areas to lower ones but also combining with each other and create larger size bubbles. Increase in bubbles diameter and size due to their combination can increase their inner pressure, this inner pressure will push back and affect its neighboring areas (in liquid state). After completion of curing step, this bubbles will transform to void, pressure inside this voids are higher than their neighboring areas which can lead to crack growth and brittle fracture of their surrounding area, a view of this phenomenon can be seen in Fig. 19. In summary, the first phenomenon (in liquid state) is the effect of bubbles on its surrounding area which can lead to the second phenomenon after completion of curing step and creation of voids (in solid state).

466


Fig. 16. Bubble based voids at the bottom of the composite samples that were caused due to the suspension flowing into the cast and mold scrapes. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Number of bubble based voids

800

Surface

730

700

Cross Section

600

Total

500 391

400

362

339

300

253 164

200 70

100 3

1

109

94 7

4

8 15

0 A

B

C

D

E

Sample Code Fig. 17. Number of bubble based voids at the surface and cross section of composite samples. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Besides, the first phenomenon could not cause any degradation and the degradation will occur in the solid state and after completion of first curing step. 2.3.3.1. SEM examination. SEM images of composite samples with and without defect and voids shows the smooth and non-smooth surfaces respectively. Fig. 20 part-A shows the surface of sample B and upper part of bubble based voids. As can be seen in this part, bubbles based voids can seriously spoil the composite surface and thus decrease the overall properties of composite samples. Fig. 20 part-B shows the surface of composite sample without any defect (sample A), as can be seen in this part, this sample has smooth surface and without any defect and also this image shows good dispersion of fillers near the surface of sample. In addition, voids can act as a stress concentration areas and cause crack growth along the voids direction that can lead to brittle fracture and decrease in the overall properties of composite samples. A view of crack growth along the composite surface can be

Fig. 18. As can be seen, number 1 and 2 shows the migration of bubbles and their combination respectively, number 3 shows the creation of bigger size bubble due to combination by other bubbles, number 4 shows the effect of bigger bubble with higher amount of inner gas concentration on the composite structure that can lead to increase in the crack growth and severe destruction within the composite structure. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)


467

Fig. 19. destruction in composite sample structure due to increase in the gas concentration inside the bubbles. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 20. SEM images from surface of composite samples, (A) sample-B, (B) sample-A. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

seen in Fig. 21. Moreover in Figs. 21 and 22, wrapped SWCNTs along their direction can be seen. These images shows that CNTs are flexible materials with high length to diameter ratio. So regarding to these SEM images it can be concluded that bubble based degradation can cause severe destruction along the composite structure that can seriously decrease the overall properties of composites, specially mechanical and electrical properties. In addition, in Figs. 20–22, CNTs are not individually dispersed.

2.3.3.2. Micro Raman Spectroscopy analysis. Micro Raman Spectroscopy peaks shows that degradation has caused change in the Raman peaks in the whole band. As can be seen in Fig. 23, sample A has lower peaks intensity in comparison with sample B that is due to degradation of sample B. As can be seen in this figure, degradation has caused increase in the Raman Intensity. For CNT based composites, less Raman Intensity shows better dispersion and higher weight percentage of carbon based fillers, because by

468


Fig. 21. Crack growth along the voids direction on the composite sample surface. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 22. Wrapped SWCNTs along their direction. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

1600

A

B

Raman Intensity

1400 1200 1000 800 600 400 200 0

Wavenumber Cm-1 Fig. 23. Micro Raman Spectroscopy analysis from samples A and B. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

increase in the weight percentages of CNTs in the composite structure the amount of Raman Intensity will decrease due to the beams absorption by CNTs. Therefore, we can conclude that degradation in the composite structure can decrease the amount of beams absorption as well as mechanical, electrical and magnetic properties. 3. Conclusion In this study with two different methods, composite samples with and without bubble based voids and defects were prepared

in order to examine the effect of bubbles based degradation on the physical and structural properties of composite samples. Also a new approach in the bubbles reduction based on the vacuum shock was presented that is very useful for bubbles reduction during the manufacturing process of composite samples. Also the effect of bubble based voids on the electrical properties and the amount of Electromagnetic waves absorption in the composite structure were measured. Results shows that bubble based voids can lead to severe decrease in the overall amount of electrical conductivity and Electromagnetic waves absorption. Furthermore, by optical examination, it can clearly be seen that bubbles have a great desire to combine with each other that can lead to formation of huge size bubbles. Huge size bubbles can cause severe destruction in the composite structure and decrease the overall properties of composite samples due to increase in their internal gas concentration and diameter. In addition, bubbles are tend to migrate from high concentration zones toward lower ones that can cause destruction in edge of composite samples. Moreover, SEM images shows that bubble based voids can lead to crack growth along their direction that can cause decrease in the mechanical properties of composite samples. Despite of that, some factor such as matrix viscosity, fillers concentration and cast scrapes can cause air entrapment at the bottom of the cast and thus increase in the amount of bubbles and degradation at the bottom of the composite samples. In addition, increase in the voids diameter due to their combination can lead to severe destruction in the composite samples. Besides, voids diameter has much higher


effect on the physical properties of composites than overall amount of voids in the whole structure. Moreover, Micro Raman Spectroscopy analysis shows that bubble based degradation can increase the Raman Intensity. This increase in the peaks is due to the less absorption of beams by CNTs that was caused by composite degradation. As a result, by increase in the temperature and pressure of the curing cycle and by decreasing the matrix viscosity and with help of vacuum shock technique, composite samples with very low amount of voids and defects can achieved. Results of this paper is very helpful for further investigation about the bubbles behavior and also developing composite fabrication methods for decreasing the overall amount of bubble based voids in the whole structure.

References [1] Muric-Nesic J, Compston P, Noble N, Stachurski ZH. Effect of low frequency vibrations on void content in composite materials. Composites: Part A 2009;40 (1):548–51. http://dx.doi.org/10.1016/j.compositesa.2008.11.010. [2] Nie L, Lee K, Lee S, Shi T, Liao G. Void control in adhesive bonding using thermosetting polymer. Sens Actuat A: Phys 2011;167(2):398–405. http://dx. doi.org/10.1016/j.sna.2011.03.028. [3] Li Y, Li Q, Ma H. The voids formation mechanisms and their effects on the mechanical properties of flax fiber reinforced epoxy composites. Composites: Part A 2015;72(1):40–8. http://dx.doi.org/10.1016/j.compositesa.2015.01.029. [4] Reis AS, Barrozo MAS. A study on bubble formation and its relation with the performance of apatite flotation. Sep Purif Technol 2016. http://dx.doi.org/ 10.1016/j.seppur.2016.01.038. [5] Wang Y, Hu X, Cao Z, Guo L. Investigations on bubble growth mechanism during photoelectrochemical and electrochemical conversions. Colloids Surf A: Physicochem Eng Aspects 2016. http://dx.doi.org/10.1016/ j.colsurfa.2016.01.004. [6] Stabler WR, Tatterson GB, Sadler RL, El-Shiekh AH. Void minimization in the manufacturing of carbon fiber composites by RTM. SAMPE Q 1992;23 (2):38–42. [7] Muric-Nesic J, Compston P, Stachurski ZH. On the void reduction mechanisms in vibration assisted consolidation of fibre reinforced polymer composites. Composites: Part A 2011;42(1):320–7. http://dx.doi.org/10.1016/ j.compositesa.2010.12.011. [8] Liu L, Zhang BM, Wang DF, Wu ZJ. Effects of cure cycles on void content and mechanical properties of composite laminates. Compos Struct 2006;73 (3):303–9. http://dx.doi.org/10.1016/j.compstruct.2005.02.001. [9] Kocatepe K. Effect of low frequency vibration on porosity of LM25 and LM6 alloys. Mater Des 2007;28(6):1767–75. http://dx.doi.org/10.1016/ j.matdes.2006.05.004.

View publication stats

469

[10] Judd NCW, Wright WW. Voids and their effects on the mechanical properties of composites – an appraisal. SAMPE J 1978;14(1):10–4. [11] Cantwell WJ, Morton J. Significance of damage and defects and their detection in composite materials: a review. J Strain Anal 1992;27(1):29–42. [12] Varna J, Joffe R, Berglund LA, Lundstrom TS. Effect of voids on failure mechanism in RTM laminates. Compos Sci Technol 1995;53(2):241–9. http://dx.doi.org/10.1016/0266-3538(95)00024-0. [13] Ling Liu L, Zhang BM, Wang DF, Wu ZJ. Effects of cure cycles on void content and mechanical properties of composite laminates. Compos Struct 2006;73 (3):303–9. http://dx.doi.org/10.1016/j.compstruct.2005.02.00. [14] Jongwoo P, Taweeplengsangsuke J, Theis C, Osenbach J. Epoxy adhesive used in optical fiber/passive component: kinetics voids and reliability. In: Proceedings 51st electronic components and technology conference. p. 637–44. [15] Hernandez S, Sket F, Molina-Aldareguıá JM, Gonzalez C, LLorca J. Effect of curing cycle on void distribution and interlaminar shear strength in polymermatrix composites. Compos Sci Technol 2011;71(10):1331–41. http://dx.doi. org/10.1016/j.compscitech.2011.05.00. [16] Bodaghi M, Cristóvão C, Gomes R, Correia NC. Experimental characterization of voids in high fibre volume fraction composites processed by high injection pressure RTM. Composites: Part A 2016;82(1):88–99. http://dx.doi.org/ 10.1016/j.compositesa.2015.11.042. [17] Hirai Y, Konishi T, Yoshikawa T, Yoshida S. Simulation and experimental study of polymer deformation in nanoimprint lithography. J Vac Sci Technol B 2004;22(6):3288–93. http://dx.doi.org/10.1116/1.1826058. [18] Niklaus F, Decharat A, Forsberg F, Roxhed N, Lapisa M, Populin M, et al. Wafer bonding with nano-imprint resists as sacrificial adhesive for fabrication of silicon-on-integrated-circuit (SOIC) wafers in 3D integration of MEMS and ICs. Sens Actuat A: Phys 2009;154(1):180–6. http://dx.doi.org/10.1016/j. sna.2009.07.009. [19] DeValve C, Pitchumani R. Simulation of void formation in liquid composite molding processes. Composites: Part A 2013;51(1):22–32. http://dx.doi.org/ 10.1016/j.compositesa.2013.03.016. [20] Frishfelds V, Staffan Lundstrom TS, Jakovics A. Bubble motion through noncrimp fabrics during composites manufacturing. Composites: Part A 2008;39 (2):243–51. http://dx.doi.org/10.1016/j.compositesa.2007.10.020. [21] Gangloff Jr JJ, Hwang WR, Advani SG. Characterization of bubble mobility in channel flow with fibrous porous media walls. Int J Multiph Flow 2014;60 (1):76–86. http://dx.doi.org/10.1016/j.ijmultiphaseflow.2013.12.003. [22] Mousavi SM, Esmaeili H, Arjmand O, Karimi SH, Hashemi SA. Biodegradation study of nanocomposites of phenol novolac epoxy/unsaturated polyester resin/egg shell nanoparticles using natural polymers. J Mater 2015. http://dx. doi.org/10.1155/2015/131957. [23] Khan S, Pothnis JR, Kim JK. Effects of carbon nanotube alignment on electrical and mechanical properties of epoxy nanocomposites. Composites: Part A 2013;49:26–34. http://dx.doi.org/10.1016/j.compositesa.2013.01.015. [24] Gangloff Jr JJ, Daniel C, Advani SG. A model of two-phase resin and void flow during composites processing. Int J Multiph Flow 2014;65(1):51–60. http://dx. doi.org/10.1016/j.ijmultiphaseflow.2014.05.015. [25] Reddy BSR. Advances in nanocomposites – synthesis, characterization and industrial applications. Croatia: Intech; 2011. p. 443–96.