FIRE AND MATERIALS Fire Mater. 2013; 37:91–99 Published online 31 January 2012 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/fam.2115
A highly efficient fire-retardant nanomaterial based on carbon nanotubes and magnesium hydroxide Chase C. Knight1,2, Filbert Ip1,3, Changchun Zeng1,2,*,†, Chuck Zhang1,2 and Ben Wang1,2 1
Florida State University, High-Performance Materials Institute, Tallahassee, FL 32310, USA FAMU–FSU College of Engineering, Department of Industrial & Manufacturing Engineering, Tallahassee, FL 32310, USA 3 FAMU–FSU College of Engineering, Department of Chemical and Biomedical Engineering, Tallahassee, FL 32310, USA 2
SUMMARY Hybrid buckypapers (HBP) were developed and showed potential as efficient fire-retardant materials by implementing multiple fire retardance mechanisms. The fabrication of HBP was performed using multiwalled carbon nanotubes (MWCNTs) and magnesium hydroxide (Mg(OH)2) nanoparticles. The Mg(OH)2 nanoparticles were well dispersed throughout the CNTs network, as revealed by scanning electron microscopy and Energy Dispersive X-ray spectroscopy. Thermogravimetric analysis and differential scanning calorimetry both confirmed the decomposition of magnesium hydroxide in the HBPs and heat absorption under elevated temperatures. Our initial results indicated that when used as a skin layer, the HBP has the potential to significantly improve the fire-retardant properties of epoxy carbon fiber composites. Copyright © 2012 John Wiley & Sons, Ltd. Received 18 December 2010; Revised 2 August 2011; Accepted 9 December 2011 KEY WORDS:
polymer composites; carbon nanotubes; nanoparticles; nanocomposites
1. INTRODUCTION Polymeric materials are highly combustible because of their chemical structures, which consist mainly of hydrogen and carbon [1]. Therefore, the fire performance of these materials needs to be improved for various applications where fire safety is an issue. Potential improvements in fire performance can be achieved by reducing the availability of the essential elements that sustain the combustion cycle (the fire triangle): combustibles (reducing agent, typically organic volatiles from decomposition of the polymer), combustives (oxidizing agent, typically oxygenation in air), and heat. Previous research has demonstrated that when applied to the surface of polymer matrix composites, carbon nanotube (CNT) buckypaper serves as an effective fire-retardant shield and reduces fire hazard [2,3]. Buckypaper is a thin membrane consisting of a dense network of entangled CNT ropes [4]. The main working mechanism for the fire retardancy is via the reduction of transport of both the combustibles and combustives. The dense network of nonflammable CNTs acts as a physical barrier to the diffusion of oxygen and also slows the escape of combustion products from the decomposition of the polymer matrices. When heated above 340 C, magnesium hydroxide (Mg(OH)2) decomposes to form magnesium oxide and water vapor with a heat of absorption of 1300 J/g (the reaction is endothermic). Considering that the buckypaper is applied as surface skin, this will lead to the lowering of the temperature of the polymers underneath and reduce the decomposition rate and release of the combustibles. Furthermore, lowering the temperature of the buckypaper will improve the survivability of the buckypaper by delaying the *Correspondence to: C. Zeng, High Performance Materials Institute, Florida State University, Tallahassee, FL 32310, USA. † E-mail:
[email protected] Copyright © 2012 John Wiley & Sons, Ltd.
92
C. C. KNIGHT ET AL.
nanotube network breakdown from thermal oxidation [2]. Additionally, the evolved water vapor from the decomposition acts as a fire inhibitor and diluent for the combustion in the gas phase. The spread of fire is further inhibited by the formation of an inert magnesium oxide (MgO) layer, which may work coherently with the CNT network to enhance impediment of transport of decomposed volatiles and air. Mg(OH)2 particles have previously been utilized as a fire-retardant additive by being incorporated in the polymer matrix as a filler [5–8]. Nano-sized Mg(OH)2 particles have been shown to be more effective as compared with larger particle sizes [5]. However, high-weight loading (more than 60%) is required to achieve satisfactory fire retardancy, which is undesirable because the high concentration of particles may give rise to processing issues and deteriorate the composite mechanical properties [9]. In this study, Mg(OH)2 nanoparticles were combined with MWCNT buckypaper to form a hybrid fire-retardant membrane. This new hybrid material was expected to allow for a reduction in the amount of Mg(OH)2 required to achieve satisfactory fire performance of composites without substantially altering the buckypaper manufacturing process and jeopardizing other properties of buckypaper. The hybrid buckypaper was incorporated as the skin material in epoxy carbon-fiber reinforced composites, and the excellent fire-retardant properties were demonstrated.
2. EXPERIMENTAL Mg(OH)2 nanoparticles (particle size 5.8–20 nm by transmission electron microscopy) were purchased from Sigma Aldrich (St. Louis, MO, USA), and the multi-walled carbon nanotubes (MWCNTs) were purchased from CNano Inc (San Francisco, CA, USA). HBPs with different concentrations of Mg(OH)2 were prepared. Thus, an appropriate amount of MWCNTs and Mg(OH)2 were dispersed in isopropyl alcohol using a high shear, high impact microfluidic processer (Microfluidics Corp, M110-P). The suspension was then vacuum-filtered through a membrane. The HBP was peeled from the membrane after completely drying. Table I shows the respective recipes used for fabrication. Buckypaper without Mg(OH)2 was also fabricated as the control material. Despite the difference in total mass, there were no appreciable differences in buckypaper thickness. This suggests that most of the Mg(OH)2 was embedded within the buckypaper. Figure 1 shows the photos of the buckypaper and hybrid buckypaper (30 wt% Mg(OH)2), and no appreciable difference was observed. For the fire-retardant properties of these materials to be tested, they were included as the skin layer of an epoxy carbon fiber composite. The epoxy used was Epon 862 (diglycidyl ether of bisphenol F) with curing agent EPICURE W (diethylene toluene diamine) (both from Miller-Stephenson Chemical Company, Inc., Danbury, CT, USA). Twelve layers of IM-7 carbon fiber fabric (5HS weave, Hexcel, Stamford, CT, USA) were used as the reinforcement. The composites were fabricated by vacuum-assisted resin transfer molding. Three types of composites were fabricated: composite with no skin layer (C-NS), composite with buckypaper skin layer (C-BPS), and composite with hybrid buckypaper skin layer (C-HBPS). For C-BPS and C-HBPS, one layer of BP or HBP was placed at the bottom of the carbon fiber layers on a mold. After resin infusion, the composites were cured at 121 C for 2 h and at 177 C for an additional 2 h and then cooled to ambient temperature. Table I. Hybrid buckypaper (HBP) fabrication recipes. ID# (mg) (mg) (mg) 1 2 3 4
Sample
Weight% Mg(OH)2
Mass of MWCNTs
0 10 20 30
90 90 90 90
Mass of Mg(OH)2 Total Mass HBP HBP HBP HBP
MWCNT MWCNT—10% Mg(OH)2 MWCNT—20% Mg(OH)2 MWCNT—30% Mg(OH)2
0 10 22.5 38.6
90 100 112.5 128.6
MWCNT, multi-walled carbon nanotube. Copyright © 2012 John Wiley & Sons, Ltd.
Fire Mater. 2013; 37:91–99 DOI: 10.1002/fam
A HIGHLY EFFICIENT FIRE-RETARDANT NANOMATERIAL
93
Figure 1. Photos of pure buckypaper (left) and hybrid buckypaper with 30% magnesium hydroxide (right).
3. RESULTS AND DISCUSSION 3.1. Dispersion of Mg(OH)2 For the inclusion of Mg(OH)2 throughout the CNTs network to be confirmed, the cross-sections of the HBPs were observed using a JEOL JSM-7401 F scanning electron microscope (SEM). Figure 2 (top row) shows the micrographs. The distribution of Mg(OH)2 appears to be uniform over the cross-section, as only a small
Figure 2. Scanning electron microscope (SEM) and Energy Dispersive X-ray spectroscopy (EDS) analysis of hybrid buckypapers. Each column contains a set of image for a hybrid buckypaper with a particular Mg(OH)2 concentration (from left to right—10%, 20%, 30% Mg(OH)2). Each row is a comparison of a particular analysis for the three samples. First row—SEM images of the entire cross-section for EDS analysis (1800x); scale bar 5 mm. Second row—distribution of magnesium across the hybrid buckypaper. Signals were collected along the straight lines indicated in the SEM micrograph on the first row. The x-axis indicates the position along the cross-section (from top to bottom of the hybrid buckypaper). Third row—SEM images of higher magnification (20000x) showing the uniform dispersion of Mg(OH)2 particles within the multi-walled carbon nanotube network; scale bar 1 mm. ROI, region of interest. Copyright © 2012 John Wiley & Sons, Ltd.
Fire Mater. 2013; 37:91–99 DOI: 10.1002/fam
94
C. C. KNIGHT ET AL.
amount of aggregates of Mg(OH)2 particles (bright spots) are observed. Majority of the particles are well dispersed. Because of the small size (5–20 nm), they are not discernible under the magnification used. However, the presence of the Mg(OH)2 throughout the HBPs was unequivocally confirmed by Energy Dispersive X-ray spectroscopy (EDS), which is widely used for element analysis. EDS is based on the principle that the X-rays emitted by an atom upon being hit by charged particles (such as electrons beams in SEM) are characteristic of an element’s atomic structure and can be used to uniquely identify one element from another. Thus, the entire cross-section of the HBPs was scanned for elemental composition along a straight line, as shown in the first row of Figure 2. The region of interest counts for magnesium along each line was recorded and shown in the second row of Figure 2. These spectra represent the variation of the intensity of magnesium at each point of measurement along the straight line, and substantial presence of magnesium was observed throughout the buckypaper. Because of the rough and porous nature of the surface, the intensity profile is not suitable for quantitative analysis. Nevertheless, both EDS and SEM observations suggest that with the dispersion and fabrication technology employed, HBPs were successfully fabricated with well-dispersed Mg(OH)2 nanoparticles and distributed throughout the network of CNTs. 3.2. Mg(OH)2 decomposition and heat absorption of hybrid buckypaper Thermogravimetric analysis (TA Instruments Q50, New Castle, DE, USA) was performed to investigate the decomposition of Mg(OH)2. Figure 3 shows the results obtained for the buckypaper samples and the pure Mg(OH)2. The pure MWCNT buckypaper shows little weight loss over the temperature range. The weight loss of the hybrid buckypaper is more substantial and increases as Mg(OH)2 content increases. In all cases, the majority of the weight loss occurs between 300 and 400 C, with maximum weight loss ~340–350 C, similar to those of pure Mg(OH)2. These observations confirm that the weight losses in the hybrid buckypaper samples were due to the decomposition of Mg(OH)2. Differential scanning calorimetry (DSC) (TA Instruments Q100, New Castle, DE, USA) was performed on the HBPs, and the results are shown in Figure 4(a). In each case, a strong endothermic peak was observed, and the peak temperature was very similar to that of pure Mg(OH)2. This strongly suggests that the absorbed heat is associated with the decomposition of Mg(OH)2 in the HBPs. The total heat was obtained by integration using TA Universal Analysis software. The values were 91, 232, and 304 J g 1 for 10%, 20%, and 30% Mg(OH)2 weight loadings, respectively. The heat absorption when exposing to elevated temperature is the primary fire-retardant mechanism by MDH. The absorbed heat did not scale quantitatively with the amount of incorporated MDH. The difference may be due to the following. When performing peak integration, identification of baseline may have contributed to the discrepancy. Additionally, because DSC only uses minute amount of sample (
