- Email: [email protected]
Fire resistance and mechanical properties of enamelled aluminium foam
Accepted Manuscript Fire resistance and mechanical properties of enamelled aluminium foam
Stefano Rossi, Lorenzo Bergamo, Vigilio Fontanari PII: DOI: Reference:
S0264-1275(17)30654-8 doi: 10.1016/j.matdes.2017.06.064 JMADE 3182
To appear in:
Materials & Design
Received date: Revised date: Accepted date:
22 March 2017 27 June 2017 28 June 2017
Please cite this article as: Stefano Rossi, Lorenzo Bergamo, Vigilio Fontanari , Fire resistance and mechanical properties of enamelled aluminium foam, Materials & Design (2017), doi: 10.1016/j.matdes.2017.06.064
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Fire resistance and mechanical properties of enamelled aluminium foam Stefano Rossi*, Lorenzo Bergamo, Vigilio Fontanari Department of Industrial Engineering, University of Trento, Via Sommarive 9, 38123 Trento Italy.
SC RI PT
*Corresponding author: [email protected]; Tel.: +39-0461-924069
Abstract
PT
ED
MA
NU
Aluminium foams show very interesting properties for a wide range of applications, in particular good mechanical properties and fire resistance. The particular cellular structure reduces the corrosion resistance of these materials in several environments. The application of a vitreous enamel layer is a possible solution to obtain a good protective barrier that can also improve the resistance at high temperatures. In this work these properties are evaluated for different porcelain enamelled aluminium foam samples. Starting from plate samples of aluminium foam produced by Alporas method, different enamel systems are studied. A frit without vanadium suitable for aluminium substrate was used. Microstructure analysis was made to highlight the adhesion between enamel and metallic substrate, as well as the filling of the open cells and the presence of defects in the glassy layers. The response to flame exposure of enamelled samples was tested in comparison to the uncoated foam. Finally, mechanical properties were studied by 4-points bending tests. This study has shown that the presence of the vitreous enamel coating increases the fire resistance of the aluminium foam. In addition, it was found that the flexural rigidity increases strongly and it is connected with the particle dimension in the frit.
AC
CE
Keywords Aluminium foam; vitreous enamel coating; fire resistance; mechanical properties
ACCEPTED MANUSCRIPT 1
Introduction
AC
CE
PT
ED
MA
NU
SC RI PT
The unique properties of metallic foams are useful in a number of potential application fields including damping, heat exchange, sound insulation, sound absorption and energy absorption [1-6]. Aluminium foam is one of the main cellular structures that has been developed in recent years thanks to the evolution of a number of cost-effective production processes [7,8]. These structures are increasingly used in marine, transportation, building construction and high-performance engineering applications [9-15] because of their high specific stiffness, specific strength, environmental resistance and thermal insulation features [16,17]. Other application is the use of this material as internal walls of tunnel and subway where it is important to have a highly glossy and cleanable surface, as well as corrosion resistance and fire resistance. Compared to dense materials the foam panels present a very interesting reduction of specific weight, which permits a lighter and more manageable structures. However the presence of opened cells are a dangerous critical point as concern the corrosion behaviour. In addition, the limited corrosion resistance of aluminium foams in several environments is an important problem that limits their wide application, as for example in presence of aggressive ions or in environments with a low pH value [4,5,18,19]. An approach widely used to improve the surface properties employs coating technologies to obtain protective surface layers [20,21]. Unfortunately, due to the complex surface morphology of aluminium foams most coating techniques that are well suited for bulk materials cannot be applied on metal foams. A coating deposited only on internal surfaces of the opened cells results insufficient to guarantee an optimal corrosion protection, because of the unavoidable presence of defects [22,23]. Figure 1 shows a 3D image of the surface morphology obtained with the software NIS Elements 4.20 using a an optical stereomicroscope Nikon SMZ25.
Fig. 1. 3-D Foam surface morphology.
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
MA
NU
SC RI PT
In this study, a protective vitreous enamel coating deposited to improve corrosion resistance is evaluated with the aim of preserving both the incombustibility and the high temperature resistance of the material. Enamel coatings have not the target to be a heat shield for components. Porcelain enamel is a vitreous inorganic coating that can be applied to metallic structures, such as steel and aluminium alloys [24]. Enamel has been used as a decorative material for over 4500 years. In the 19 th century enamelling became industrialized. Today it has been used both for decorative purpose and for protection against corrosion as well as for its engineering properties [25]. Porcelain enamel is characterized by several interesting properties, such as good durability due to high corrosion resistance and excellent resistance to UV radiation, chemical and high temperature resistance [26,27], that allow many applications in different fields. The corrosion resistance of this type of coating is due to its glassy nature (intrinsic chemical inertia), absence of defects, high thickness and excellent adhesion obtained during the firing process [24]. Aluminium components used in the field of road infrastructures, interior of tunnels, subway walls, façades, building constructions and ships are extensively protected using this type of coatings, in particular where high durability and fire resistance are required. Usually an aluminium alloy can be defined enamelable when, after the firing treatment, a good adhesion between metal and enamel with low defectiveness has been obtained. [2836] However, for aluminium alloy substrates some critical aspects should be taken into consideration. Considering the low meting point of aluminium, low firing temperature is requested. In this situation the adhesion between substrate and coating could be affected. For this reason, enamels developed for aluminium and its alloys should have a low melting temperature (since the substrate cannot exceed 600°C in the firing step). Since 30 years vanadium-based enamels are used on aluminium alloy to guarantee a good adhesion with substrate. Various vanadium compounds, as the divanadium pentoxide (V2O5), are the mostly utilized. However, this compound is harmful to human health and toxic to the environment [37]. Therefore vanadium-free enamels have been developed and used in this research. There are several deposition methods to obtain an enamel glassy layer on metallic substrate characterized by smooth surfaces [27]. Due to the complex geometry of the foam, the surface coverage is a critical point for enamel application and a suitable deposition process has to be developed and optimized. Normally to have a good deposition of glassy enamel it is necessary to prepare the surface eliminating the contaminants and producing an oxide layer, which favours the adhesion between substrate and coating. In addition the presence of contaminants on the substrate could produce bubbles during the firing process, detrimental for the quality of the layer. Surface preparation is not easy in the presence of opened cells; in addition clearly it is difficult to obtain a uniform layer without bubbles. In this work, several different types of vitreous enamel coating systems have been developed. The most important properties required for traditional aluminium foam in infrastructure and transportation applications, such as flexural resistance and fire
ACCEPTED MANUSCRIPT
2.1
Materials processing and samples preparation Materials processing
MA
2
NU
SC RI PT
behaviour, are evaluated to observe how the presence of different solutions can affect these properties. Fire resistance and thermal insulation properties are the main issues for the fireretardant structure that are requested in road infrastructure or in interior of tunnels applications [21]. The fire resistance means the ability to prevent the spread of the fire. Therefore, the fire-retardant structure must insulate the thermal radiation of the fire to have a safe time during fire events. In order to reduce the thermal transmittance through the foam, an enamel coating was deposited on only one foam surface and the heat propagation was investigated. Since several coated panels may undergo to flexural loads in service, the knowledge in this context need to be improved. While mechanical properties of aluminium foams such as compressive behaviour has been extensively studied in particular where the energy absorption results important [38-43], other studies are focused on the flexural properties of foams when this material is used as a core of sandwich panels [44-47]. Considering the brittle nature of deposits and the final application of the investigated enamelled foams, the mechanical characterization has been focused on flexural tests.
AC
CE
PT
ED
The foams object of this research is produced following Alporas industrial process and supplied by Foamtech Co. Ltd (Seoul, Korea). They are closed-cell aluminium foams with a density about 0.2 g/cm3. The Alporas process is a direct foaming technology of metals where a blowing agent (usually 1.6 wt% TiH2) is added to the melt. The blowing agent, under the influence of heat, decomposes and releases gas acting as propellant of the foaming process. The exact procedure is described in literature [7]. Briefly, the manufacturing process to obtain the panels consists of five stages: thickening, foaming, cooling, foamed block and slicing. In the melted aluminium at 680°C calcium compounds are added to increase the viscosity of the melt. Titanium hydride is then added and hydrogen is thus released. Consequently the melted material fills the mould, due to expansion process. In this step the cell structure is obtained. After aluminium solidification during the following cooling step the foam block is obtained. Sheets of different thickness can be finally produced from this block by mechanical slicing This operation necessary to obtained panels from the foamed block leaves surface opened cells. The enamel were deposited starting from a low temperature frit without vanadium oxide, supplied by Emaylum Italia (Chignolo d’Isola, BG, Italy), whose composition is reported in Table 1. The frit used for the different layers deposition is characterised by the same flake shapes and composition, in order to minimize the number of parameters adopted during the deposition process.
ACCEPTED MANUSCRIPT Table 1: Frit composition expressed as wt% of oxides Glass Formers Melting agent Opacifiers Modifiers
55 wt% 35 wt% 5 wt% bal.
Samples production
SC RI PT
2.2
SiO2 + B2O3 Na2O + K2O + Li2O TiO2 Al2O3 + ZnO + P2O5 + SrO
AC
CE
PT
ED
MA
NU
All samples used in this research were produced in laboratory. Each aluminium foam panel (100 x 50 mm2) was pretreated by immersion in water solution with 20 wt% NaOH for 50 seconds, followed by distilled water rinse and drying by air flow. This is the common process to prepare an aluminium surface eliminating the contaminates and activating the surface to increase adhesion. As commonly requested by the different typical applications, only one face of the foams was coated in order to limit the weight increase of the sample. Table 2 resumes the investigated porcelain enamelled systems. The choice of the vitreous enamel systems is dictated to seal the opened porosity, to guarantee a good adhesion between the deposit and the aluminium substrate and finally to obtain a defectless layer of sufficient thickness to perform as a protective barrier. Typical enamel systems are made of a ground layer, to improve the adhesion with the substrate and to enable a levelling effect of the topcoat, that must show good resistance in acid and basic environments, colour stability, corrosion protection as well as other specific properties dictated by the application. Following this standard the studied systems were initially composed by two layers, a ground one, that could favour the adhesion of the second layer named topcoat, that could guarantee the protection properties covering the whole surface. In the studied samples the ground layer, in addition to guarantee a good adhesion of the enamel, has also the function to seal the opened cell. Considering the difficulty of this operation two different frit grain dimensions are used. In order to improve the not optimal adhesion observed in the first studied samples, in a second time a first layer, named primer, was applied before ground deposition with the aim to enhance the adhesion between ground and substrate. Table 2: Studied samples Sample name
Primer
A1
Ground (type)
Topcoat
Number of layers
✓ (A)
✓
2
Number of firing treatment 2
A2
✓
✓ (A)
✓
3
2
A3
✓
✓ (A)
✓
3
3
✓ (B)
✓
2
2
✓ (B)
✓
3
2
B1 B2
✓
ACCEPTED MANUSCRIPT
MA
NU
SC RI PT
For the ground layers a frit with the chemical composition shown in Table 1 was used. Vitreous particles with two different size, 500 µm for ground A and 350 µm for ground B, were used (Fig. 2). These particles are obtained by milling and sieving the frit flakes. 10 wt% clay, 0.5 wt% sodium aluminate and 20 wt% water were added to the frit to produce a slurry used for wet spraying deposition of layer. Drying (200°C for 15 minutes) and firing treatments at a temperature of 580°C for 20 minutes were carried out in a heating chamber.
ED
Fig. 2. Micrographs of vitreous frit particles used for ground layer deposition: a) max size 500 µm for Ground A; b) max size 350 µm for Ground B.
AC
CE
PT
The topcoat was subsequently deposited. The frit (the composition is shown in Table 1) was milled together with 3.5 wt% inorganic dyes (CoO) and 0.5 wt% sodium aluminate and then the enamel powder was sieved down to 120 µm. After the addition of 38 wt% of water, the obtained slurry was deposited by wet spraying. The topcoat enamel was dried at 200°C for 15 minutes and fired at 560°C for 15 minutes in a heating chamber. To improve the adhesion between the ground and the substrate surface in some cases a primer layer before the ground deposition was applied. The primer was obtained by milling the frit together with 0.5 wt% sodium aluminate, sieving down the enamel powders to 120 µm and adding 38 wt% of water. The obtained slurry was then deposited by wet spraying application. The enamel was dried at 200°C for 20 minutes in a heating chamber. Only for the sample A3 a firing treatment at a temperature of 560°C for 15 minutes was carried out in a heating chamber for the primer layer to evaluate the effect on the mechanical and thermal properties of the enamelled aluminium foam. For the other samples the first firing step was made after ground application. Considering the adhesion between aluminium substrate and glassy layers, the firing treatment of the primer layer on sample A3 did not produce observable improvement in comparison of sample A2. Its main drawback is the negative increase of production complexity. Indeed the extra fire treatment increases the process time and cost. This research is carried out in university and industrial laboratories and then the samples are not of industrial production. However, considering the industrial use of the enamelled coatings it is possible to have an estimate
ACCEPTED MANUSCRIPT of the presumable cost. An indication cost for a deposit of 100 µm is about 0.02 Euro/dm2. The thermal treatment cost is comparable to the cost of a curing treatment of a paint. For each protection system four different samples were prepared. The density of uncoated foam is 0.2 gr/cm3. Clearly with the presence of the glassy deposits an increase of density was expected. The final density was equal to 0.43 gr/cm3 for sample A1, 0.44 gr/cm3 for samples A2, B1 and B2 and finally 0.49 gr/cm3 for samples A3. In the last case the three firing treatments produce a denser deposit system.
3.1
Experimental methods
SC RI PT
3
Microstructural analysis
The microstructural characterization was carried out on the cross-section using a JEOL IT300 scanning electron microscope. Fire resistance test
NU
3.2
AC
CE
PT
ED
MA
The flame test was performed considering the process described by A. Lakshmanam et al. [48]. Enamelled foam was subjected to a fully engulfing fire with an average flame temperature of 600°C for 10 minutes. The setup of the test is shown in Fig. 3a. The burner was operating with propane. The size of the samples was 50 x 50 x 13 mm. Samples were kept vertically, with the coated surface exposed to fire. The system represents a fully affected fire with a homogeneous temperature distribution on the fireside of the samples. Two appropriately machined refractory bricks (Fig. 3b) were used to prevent heat and mass loss though the borders of the investigated assembly. In this way a better comparison between the results of the different samples was reached. Fig. 3c shows the sample assembled in the refractory insulation materials. The distance between the coated foam surface and the burner was adjusted to 100 mm, so that an average flame temperature of 600°C reaches the sample surface. Temperature measurement was performed by Type-K (chromel alumel) probe-style thermocouple (model Type K Thermocouple producer Thermometrics - Norhridge US), which was connected to a computer and LabVIEW interface via a data acquisition control unit. This thermocouple (Chromel - 90% nickel and 10% chromium - Alumel - 95% nickel, 2% manganese, 2% aluminium and 1% silicon) shows a measure range between -200°C to +1260°C with tolerance in the measurement experimental range of 0,07%. A thermocouple was located to the rear face of the sample, as shown in fig. 3c. The junction was fixed in intimate contact with the specimen by means of aluminium foil adhesive. An infrared thermo-camera (FLIR T-Series T460) was implemented to capture the sample temperature. The camera was positioned on a tripod with the imaging field capturing one of the side cross section of the sample.
CE
PT
ED
MA
NU
SC RI PT
ACCEPTED MANUSCRIPT
3.3
AC
Fig. 3. a) Fire test setup; b) machined refractory bricks; c) sample assembly in refractory insulation materials; d) view of the enamelled surface of the sample, exposed to the direct flame.
Static bending tests
Four-point bending tests were performed to measure the flexural stiffness, defined as bending stiffness, , following the ASTM D6272-10 Standard. This test was conducted on a Tinius Olsen H10KT electromechanical testing machine at the loading rate of 2 mm/min. The size of the samples was 100 mm in length, 20 in width and 13 mm in thickness; the outer span during the test was 80 mm and the inner span was 20 mm. The specimen was positioned to have the coated surface under tensile stress. The test was replicated using 4 samples for each type.
ACCEPTED MANUSCRIPT The load-displacement curves were acquired; the samples were considered failed when the curves reached a steadily declining slope. The bending stiffness was calculated using Eq. (1) [49]. (1)
ED
MA
NU
SC RI PT
Where L is the size of the outer span, P is the load, is the displacement of the beam in correspondence of the loading points A is the geometric parameter represented in Fig. 4. The four-point bending strength values which were reported in this work, were the averages of four tests.
CE
4.1
Result and discussion
Characterization of enamel layers
AC
4
PT
Fig. 4. 4-Point bending test configuration (the black line represents the enamel layer).
Fig. 5 shows the cross section of the different samples. Considering the microstructural analysis of the samples, it is possible to clearly observe for the sample A1 (Fig. 5a, dark area between Al substrate and light area indicated as Ground A) a debonding between the coating and the substrate. During the thermal treatment, the sintering of the vitreous particles involves a volumetric shrinkage of the ground layer producing a lack in adhesion between enamel coating and substrate. In addition, the ground is characterized by a huge amount of internal porosity, due to the lower degree of sintering, probably connected with the presence of big frit particles. The microstructural observation of the samples with ground A highlights the presence of high volume vitreous particles that probably were not melted during the firing process. For sample A2 (Fig. 5b) the presence of a primer layer increases the coating-substrate adhesion. Nevertheless, it is not enough to completely counteract the effect of ground
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
MA
NU
SC RI PT
shrinkage during the heat treatment. Observing the Fig. 5b (dark area between white primer and Al substrate) it is possible to see a debonding within the primer, where one part of primer remains adherent to the cell surface of the substrate and the other part is connected to the ground. This lack of adhesion is due to the shrinkage force of vitreous particles obtained during the sintering process. For sample A3 (Fig. 5c) the application of three firing treatments involves an improvement of the adhesion between substrate and coating, in comparison with that observed in sample A2. This improvement probably is due to the presence of a consolidate primer, that prevents the shrinkage of the ground during the sintering of the particles. For sample B1 it is possible to observe a debonding between coating and substrate (Fig. 5d, the grey area between Al substrate and ground layer) occurred during the thermal cycle of the sintering process. Therefore, this debonding is characterized by a lower volumetric shrinkage of the ground connected with the smaller size of the vitreous particles (compared with ground A), that allows to achieve a higher density before firing. This increase of density is related to better particle packing because the smaller particles fill the voids between the bigger ones. The density is influenced by the ratio between big and small particles size [50]. The better sintering process produces higher compactness and smaller amount of pores for the ground B compared to the ground A. With smaller particles size lower temperature and time are requested to obtain a certain sintering degree because of the enhancement of the driving force for the process due to higher surface area [50]. For this reason the opened cells of the samples B are filled to an higher extent than those of samples A. Furthermore, the ground B shows lower difference of morphology with the topcoat, probably due to a better interdiffusion between the vitreous particles of the two different layers during the firing treatment. Sample B2 (Fig. 5e) presents the highest quality characterized by the best adhesion and open cells filling. This coating system shows a good homogeneity between the three different layers thanks to the better sintering of the vitreous particles. The internal porosity is characterized by voids with a small average diameters. Differently from sample A2, for this sample the presence of a dry primer layer is sufficient to counteract the ground shrinkage during the vitreous particles sintering as a consequence of the lower ground shrinkage occurring during the process. Finally, using the ground B with the number of firing treatments it is possible to obtain a good filling of the open cells, during enamel deposition. The function of the ground layer is to seal the opened porosity. The difference of frit particles dimension does not influence in the thickness of the deposits. Considering this aspect it is not easy to define and measure exactly the thickness due to the opened cell. However, if the surface obtained by completely filling the opened cells is considered as reference level, the thickness of the deposits ranges between 370 and 500 μm for all samples.
AC
CE
PT
ED
MA
NU
SC RI PT
ACCEPTED MANUSCRIPT
Fig. 5. Microstructural cross-sections of vitreous enamel coatings deposited on aluminium foam substrate: a) sample A1; b) sample A2; c) sample A3; d) sample B1; e) sample B2; 4.2
Flame test experimental results and discussions
Sequential infrared (IR) images of enamelled samples and of bare foam during flame test are displayed in Fig. 6. As can be seen, bare aluminium foam reached thermal
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
MA
NU
SC RI PT
saturation in about 60 seconds, while more than 90 seconds are necessary for the enamelled aluminium foams to reach the stationary condition. The high flame retardant behaviour of aluminium foam with closed cells is connected to the high quantity of porosity and to low heat permeability of air, which restricts fire and prevents the heat transfer [50]. In case of enamelled aluminium foam the increased flame retardant properties of the foam are associated to thermal barrier behaviour of vitreous coating. Considering the random distribution of the opened cell and of the aluminum walls under the enamel layer together with the resolution of infrared camera it is not possible to exactly correlate the local temperature with the pore structures. However the temperature evolution during flame exposure permits to differentiate the samples behavior.
Fig. 6. Sequential IR images of bare foam (sample 0) and coated foams (samples A1, A2, A3, B1, B2) during flame test showing temperature profile in side wall. Table 3 shows the data fitting of temperature-time curves, detected by the thermocouple (Fig. 7). Using the Eq. 2 (that describes the temperature vs time curve considering the intrinsic properties of the material) it is possible to obtain the parameter bi (Eq. 3). Then, using the Eq. 4, it is also possible to calculate the apparent thermal diffusivity coefficient [51] that allows a comparison between the thermal behaviour of the different samples.
ACCEPTED MANUSCRIPT (2)
(3)
SC RI PT
(4)
AC
CE
PT
ED
MA
NU
Where T is the temperature detected by the thermocouple, is the temperature at saturation time, Ti is the initial temperature of the foam face opposite to that exposed to fire (that for a better fitting, in this work, is considered as 100 °C, avoiding then the initial transient part of the temperature-time curve), is the apparent thermal conductivity, is the foam density, Cp specific heat of aluminium and d is the foam thickness (13 mm). A lower value of the heat diffusivity inside of the foam structure increases the time employed by the foam to reach the saturation temperature.
Fig. 7. Temperature-time curves of fire test.
Table 3: Values of parameter -bi
ACCEPTED MANUSCRIPT -bi
Sample name 0 A1 A2 A3 B1 B2
0.0338 0.0123 0.0154 0.0162 0.0192 0.0231
SC RI PT
It is possible to observe (Table 4) a remarkable increase in thermal insulating for enamelled samples compared to bare samples. The highest thermal insulation is shown by samples A, probably connected with the higher internal porosity. Due to the low adhesion between enamel layer and aluminium walls and a lower sintering level of ground A more voids are present at interface and inside the glassy mater. The presence of primer layer lowers the thermal insulating property as a result of reduction of internal porosity. Table 4: Apparent thermal diffusivity
Static bending test
MA
ED
4.3
0.00044 0.00016 0.00020 0.00021 0.00025 0.00030
PT
0 A1 A2 A3 B1 B2
NU
Sample name
AC
CE
A preliminary test using a bare aluminium foam beam specimen was carried out for comparing the flexural strength with the enamelled ones. The typical load-displacement curves measured by four-point bending tests for all the samples are shown in Fig. 8. The reported curves represent the typical behaviour, considering a replication on 4 specimens, of the samples. The stiffness values were determined using eqn. 1. To this purpose a linear fitting was performed to determine the slope of the curve in the range between 5 and 40 N, where the curves can be considered as approximately linear. The calculated values are reported in Table 5 as relative values, by comparing the coated samples results with those of the bare material (Sample 0). It is possible to observe an impressive bending stiffness increase for enamelled foams compared to the bare foam. This effect is given by hard vitreous enamel coating, characterized by glassy nature. The application of a primer layer allows to further increase the stiffness. In addition, samples B are characterized by a higher stiffness due to the more homogeneous and compact ground layer that delays the early crack formation at the porosity tips. This behaviour is also influenced by the higher adhesion level between glassy mater and aluminium walls.
ACCEPTED MANUSCRIPT
90 80 70
50 40
Sample 0 Sample A1 Sample A2 Sample A3 Sample B1 Sample B2
30 20 10 0 0.0
0.2
0.4
0.6
0.8
1.0
SC RI PT
Load (N)
60
1.2
1.4
1.6
1.8
2.0
NU
Displacement (mm)
MA
Fig. 8. Load-displacement curves from 4-point bending test.
ED
Table 5: Relative slope load-displacement curves
4.4
1.52 3.49 2.62 2.47 4.51
CE
AC
A1/0 A2/0 A3/0 B1/0 B2/0
PT
Sample name/Sample 0
Enamel coating brittleness
The brittleness of porcelains coating could be investigated by evaluating the point of “First Failure Crack” for each different deposited enamel coating obtained during the fourpoint bending test. To individuate the first crack failure it is considered the instant where a sudden decrease of load displacement curve was observed. The crack behaviour of samples was studied using the load-displacement curves shown in Fig. 8, extrapolating the values of displacement in which the first coating crack occurs. From the literature [52] is known that FCF (First Crack Failure) is the parameter used for describing the crack behaviour of the coating. It is calculated using the following equation:
ACCEPTED MANUSCRIPT (5) Where is the coating (coatings plus glass filled open porosity) thickness and is the displacement at the first crack onset. Then FCF is an adimensional value that describes the enamel coating brittleness. The higher the FCF the lower is the enamel coating brittleness. The average values for each sample are reported in Table 6. Table 6: FCF, ηfcf and δs values FCF
ηfcf (mm)
δs (mm)
A1 A2 A3 B1 B2
0.25 0.095 0.11 0.603 0.437
0.472 0.252 0.487 1.17 0.810
1.86 2.66 4.30 1.94 1.85
SC RI PT
Sample name
Conclusion
ED
5
MA
NU
It is possible to observe that using the ground B the coating is characterized by a lower brittleness probably as a result of the higher homogeneity and compactness of the coating itself. On the other hand, the deposition of a primer layer produces the increase of the enamel brittleness, probably as a result of the raise of sample rigidity.
AC
CE
PT
The results of the characterization of different enamel coating systems applied to aluminium foams, can be summarized as follows: The different vitreous enamel systems studied in this work allow a good degree of coverage of the foam surface; The smaller size of the frit vitreous particles, improves the sintering, during the heat treatment, thus producing a more homogeneous and compact layer; The deposition of a primer layer improves the adhesion between the substrate and the coating. On the other hand, it increases the cost and time of the production process as well as the weight of the coated foam; Each sample retains the incombustibility during fire exposure. The enamelled aluminium foams show a lower heat propagation due to the barrier effect guaranteed by the presence of vitreous coating. Samples characterized by ground A and by the lack of a primer layer shows better heat insulation performances; The application of the enamel coating highly increases the bending stiffness of the foam. In particular the use of ground B and the application of a primer layer produce a higher increase in the bending stiffness of the foam; For the adopted testing configuration (coating on the tensile side during test), the samples with ground B and without primer layer show less brittle behaviours;
ACCEPTED MANUSCRIPT
The deposition process adopted for the sample A3 highly increases time and cost of the process; in addition, this process doesn’t improve the mechanical and thermal properties as requested by the possible applications. For this reason this deposition process should be not taken in account for further studies. The evaluated protection systems are a representative view of the porcelains enamel layers normally applied and then they could show the properties, which could be obtained with a glassy enamel deposition.
SC RI PT
Acknowledgements The authors are grateful to Attilio M. Compagnoni, (Emaylum Italia, Italy) for supplying the enamel samples and useful discussion. Thanks also to Alberto Bettini (Vaber Industriale, Italy) and to Sylove Won (Foam Tech Global, Korea), for aluminium foam materials supply.
NU
References
MA
[1] T.J. Lu, J.M. Ong, Characterization of closed-celled aluminium alloys, Journal of Materials Science, 2001, Vol. 36, pp. 2773-2786.
ED
[2] M.F. Ashby, A.G. Evans, N.A. Fleck, L.J. Gibson, J.W. Hutchinson, Metal Foams: A Design Guide, Butterworth-Heinemann, Woborn, USA, 2000.
PT
[3] T.J. Lu, C. Chen, Thermal transport and fire retardance properties of cellular aluminium alloys, Acta mater.1999, Vol. 47, pp. 1469-1485.
CE
[4] F. Stergioudi, C.A. Vogiatzis, K. Gkrekos, N. Michailidis, S.M. Skolianos, Electrochemical corrosion evaluation of pure, carbon-coated and anodized Al foams, Corrosion Science, 2015, Vol. 91, pp 151–159.
AC
[5] N. Movahedi, A. Habibolahzadeh, Effect of plasma electrolytic oxidation treatment on corrosion behavior of closed-cell Al-A356 alloy foam, Materials Letters, 2016, Vol. 164, pp. 558–561. [6] Y. Zhou, Y. X. Li, Oxidation kinetics of oxide film on bubble surface of aluminum foams produced by gas injection foaming process, Trans. Nonferrous Met. Soc. China, 2015, vol. 25, pp. 2781−2788. [7]J. Banhart, Manufacture, characterisation and application of cellular metals and metal foams, Progress in Materials Science, Vol. 46, 2001, pp. 559-632.
ACCEPTED MANUSCRIPT [8] K. Heim, G. S. Vinod-Kumar, F. Garcia-Moreno, J. Banhart, Stability of various particlestabilised aluminium alloys foams made by gas injection, J Mater Sci, 2017, Vol. 52, pp. 6401–6414. [9] A.E. Simone, L.J. Gibson, Effects of solid distribution on the stiffness and strength of metallic foams, Acta Materials, Vol. 4, 1998, pp. 2139-2150.
SC RI PT
[10] H.N.G. Wadley, Cellular metals manufacturing, Adv. Eng. Mater., Vol. 4, 2002, pp. 726-733. [11] H.P. Degisher, B. Kriszt, Handbook of Cellular Metals: Production, Processing, Applications, Wiley-VCH, Weinheim, Germany, 2002.
NU
[12] Y. Wang, J.Y. R. Liew, S. Chin Lee, X. Zhai, W. Wang, Crushing of a novel energy absorption connector with curved plate and aluminum foam as energy absorber, Thin Walled Structures 2017, Vol. 111, pp 145–154.
MA
[13] J. Baumeister, J. Banhart, M. Weber, Aluminium foams for transport industry, Materials Design, 1997, Vol.18, pp 217- 220. [14] F. García-Moreno, Commercial Applications of Metal Foams: Their Properties and Production, A review, Materials, 2016, Vol. 85, n.9, pp. 1-27.
PT
ED
[15] E. Solórzano, J.A. Reglero, M.A. Rodríguez-Pérez, D. Lehmhus, M. Wichmann, J.A. de Saja, An experimental study on the thermal conductivity of aluminium foams by using the transient plane source method, International Journal of Heat and Mass Transfer, 2008, Vol. 51, pag. 6259–6267.
AC
CE
[16] Lee DG, Suh NP, Axiomatic design and fabrication of composite structure applications in robots, machine tools and automobiles, New York, Oxford University Press, 2006. [17] Yu HN, Kim SS, Do Suh J, Lee DG, Axiomatic design of the sandwich composite endplat for PEMFC in fuel cell vehicles, Compos Struct, 2010:92:1504-11. [18] J.R. Davis, Corrosion of Aluminium and Aluminium Alloys, ASM International, Materials Park, Ohio, USA, 1999. [19] X. Xia, Z. Zhang, J. Wang, Compressive characteristics of closed-cell aluminium foams after immersion in simulated seawater, Materials and Design, 2015, Vol. 67, pp 330-336. [20] C.M. Cotell, J.A. Sprague, F.A. Smidt, Surface Engineering, Vol. 5; ASM International, Novelty, OH, 1994.
ACCEPTED MANUSCRIPT [21] T. Abdulla, A. Yerokhim, R. Goodall, Effect of Plasma Electrolytic Oxidation coating on the specific strength of open-cell aluminium foams, Materials and Design, 2011, Vol. 32, pp 3742-3749. [22] S. Rossi, M. Fedel, L. Da Col, F. Deflorian, S. Petrolli, Coatings to increase the corrosion behaviour of aluminium foam, Surface Engineering, 2017, Vol. 33, pp 405-409.
SC RI PT
[23] S. Rossi, M. Calovi, M. Fedel, Corrosion protection of aluminium foams by cataphoretic deposition of orgnaic coatings, Progress in Organic Coatings, 2017, Vol 109, pp 144-151. [24] S. Rossi, A. Quaranta, L. Tavella, F. Deflorian, A.M. Compagnoni, Innovative Luminescent Vitreous Enameled Coatings, Intelligent Coatings for Corrosion Control, Chapter 7, Elsevier, Oxford, UK, 2015, pp. 251-282.
NU
[25] A. Umbertazzi, N. Wojciechowski, Vitreous Enamel, Ulrico Hoepli Editore, Milano, 2002.
MA
[26] S. Pagliuca, W.D. Faust, Porcelain (vitreous) Enamels and Industrial Enamelling Processes, The International Enamellers Institute, Mantova, Italy, 2011.
ED
[27] A.I. Andrews, Porcelain Enamels: The Preparation, Application and Properties of Enamels, 3rd ed., Tipografia Commerciale, Mantova, Italy, 2010. recent
research
and
PT
[28] S.L. Lee, N.E. Shanmugam, Aluminium structures: developments, Elsevier applied science, New York, USA, 1991.
CE
[29] C. Vargel, Corrosion of aluminium, Elsevier, Amsterdam, 2004.
AC
[30] M. Stacey, C. Bayliss, Aluminium and Durability: reviewed by inspection and testing, Materials Today: Proceedings, 2015, Vol 2, pp 5088 – 5095. [31] M. Pathirana, N. Lam, S. Perera, L. Zhang, D. Ruan, E. Gad, Damage modelling of aluminium panels impacted by windborne debris, Journal of Wind Engineering & Industrial Aerodynamics, 2017, Vol 165, pp 1–12. [32] T.Theodosiou, K.Tsikaloudaki, D.Bikas, Analysis of the Thermal Bridging Effect on Ventilated Facades, Procedia Environmental Sciences, 2017 Vol 38, pp 397 – 404. [33] D. Izydorczyk, B. Sędłak, B. Papis, P. Turkowski, Doors with specific fire resistance class, Procedia Engineering, 2017, Vol 172, pp 417 – 425.
ACCEPTED MANUSCRIPT [34] F.L. Chen, X. He, H.M. Yin, Manufacture and multi-physical characterization of aluminum/high-density polyethylene functionally graded materials for green energy building envelope applications, Energy and Buildings, 2016, Vol 116, pp 307–317. [35] B. Sędłak, P. Sulik, D. Izydorczyk, M. Łukomski, Fire-stop wraps and collars with intumescent materials –performance comparison, Procedia Engineering 2017, Vol 172, pp 961 – 968.
SC RI PT
[36] T. Ihara, B. P. Jelle, T. Gao, A. Gustavsen, Accelerated aging of treated aluminum for use as a cool colored material for facades, Energy and Buildings, 2016 Vol 112, pp 184– 197. [37] A.M. Compagnoni, Reduction or elimination of vanadium in enamel for aluminium, in: Proceedings of the XXI International Enamellers Congress, IEC, Shanghai, 18-22 May, 2008.
MA
NU
[38] X. Xia, X. Chen, Z. Zhang, X. Chen, W. Zhao, B. Liao, B. Hur, Compressive properties of closed-cell aluminium foams with different contents of ceramic microspheres, Materials and Design, 2014, Vol. 56, pp 353-358.
ED
[39] J. Lázaro, E.Solórzano, M.A.Rodríguez-Pérez, A.R. Kennedy, Effect of solidification rate on pore connectivity of aluminium foams and its consequences on mechanical properties, Materials Science & Engineering A 6, 2016, Vol. 72, pp. 236–246.
PT
[40] J. Kováčik, Ľ. Orovčík, J. Jerz, High-temperature compression of closed cell aluminium foams, Kovove Mater., 2016, Vol. 54, pp. 429–440.
CE
[41] J. Lazaro, E. Solórzano, M. A. Rodríguez-Pérez, F. Garcıa-Moreno, Pore connectivity of aluminium foams: effect of production parameters, J Mater Sci, 2015, Vol. 50, pp. 3149– 3163.
AC
[42] I. Duarte, J.M.F. Ferreira. Composite and nanocomposite metal foams. Review. Materials 2016, Vol. 9(2), n. 79, pp. 1-34; doi:10.3390/ma9020079. [43] X. Xia, H. Feng, X. Zhang, W. Zhan, The compressive properties of closed-cell aluminium foams with different Mn additions, Materials and Design, 2013, Vol. 51, pp 797802. [44] A.M. Harte, N.A. Fleck, M.F. Ashby, Sandwich panel design using aluminum alloy foam, Advanced Engineering Materials, 2000, Vol. 2, pp 219-222. [45] G. Zu, B. Song, Z. Zhong, X. Li, Y. Mu, G. Yao. Static three-point bending behavior of aluminum foam sandwich, Journal of Alloys and Compounds, 2012, Vol. 540, pp 275-278.
ACCEPTED MANUSCRIPT [46] L.L. Yan, B. Han, B. Yu, C.Q. Chen, Q.C. Zhang, T.J. Lu. Three-point bending of sandwich beams with aluminum foam-filled corrugated cores, Materials and Design, 2014, Vol. 60, pp 510-519. [47] I. Duarte, M. Vesenjak, L. Krstulović-Opara. Dynamic and quasi-static bending behaviour of thin-walled aluminium tubes filled with aluminium foam. Composite Structures. 2014, Vol. 109, pp 48-56.
SC RI PT
[48] A. Lakshmanan, Sintering of Ceramics: New Emerging Techniques, InTech, Rijeka, Croatia, 2012. [49] E. Carrera, G. Giunta, M. Petrolo, Beam Structures: Classical and Advanced Theories, Wiley, Chichester, UK, 2011.
NU
[50] E. Placido, M.C. Arduini-Schuster, J. Kuhn, Thermal properties predictive model for insulating foams, Infrared Physics and Technology, 2005, Vol. 46, pp. 219-231.
MA
[51] J.H. Lienhard IV, J.H. Lienhard V, A Heat Transfer Textbook, Cambridge, 3 rd edition, Massachusetts, USA, 2001.
AC
CE
PT
ED
[52] A. Chelli, R. Poletti, L. Pignatti, F. Bruscoli, G. Pasqualetti, F. Bruni, A. Zucchelli, L. Rossetti, F. Lotti, G. Minak, V. Dal Re, S. Curioni, Composite Enameled Steel Elements for Air Preheaters and Gas-gas Heaters: an Integrated Approach from Sheet Forming and Enamelling to Basket Assembly, XXI International Enamellers Congress, 18-22 May 2008, Shanghai, China.
ACCEPTED MANUSCRIPT Figures captions list Fig. 1. 3-D Foam surface morphology. Fig. 2. Micrographs of vitreous frit particles used for ground layer deposition: a) max size 500 µm for Ground A; b) max size 350 µm for Ground B. Fig. 3. a) Fire test setup; b) machined refractory bricks; c) sample assembly in refractory insulation materials; d) view of the enamelled surface of the sample, exposed to the direct
SC RI PT
flame. Fig. 4. 4-Point bending test configuration (the black line represents the enamel layers). Fig. 5. Microstructural cross-sections of vitreous enamel coating deposited on aluminium foam substrate: a) sample A1; b) sample A2; c) sample A3; d) sample B1; e) sample B2; Fig. 6. Sequential IR images of bare foam (sample 0) and coated foams (samples A1, A2,
Fig. 7. Temperature-time curves of fire test.
NU
A3, B1, B2) during flame test showing temperature profile in side wall.
MA
Fig. 8. Load-displacement curves from 4-point bending test.
ED
Table list Table 1: Oxide composition frits
CE
PT
SiO2 + B2O3 Na2O + K2O + Li2O TiO2 Al2O3 + ZnO + P2O5 + SrO
55 wt% 35 wt% 5 wt% bal.
AC
Glass Formers Melting agent Opacifiers Modifiers
Table 2: Studied samples Sample name
Primer
A1
Ground (type)
Topcoat
Number of layers
✓ (A)
✓
2
Number of firing treatment 2
A2
✓
✓ (A)
✓
3
2
A3
✓
✓ (A)
✓
3
3
✓ (B)
✓
2
2
✓ (B)
✓
3
2
B1 B2
✓
ACCEPTED MANUSCRIPT
Table 3: Values of parameter -bi
0 A1 A2 A3 B1 B2
0.0338 0.0123 0.0154 0.0162 0.0192 0.0231
Table 4: Apparent thermal diffusivity coefficient
MA
0.00044 0.00016 0.00020 0.00021 0.00025 0.00030
ED
0 A1 A2 A3 B1 B2
NU
Sample name
SC RI PT
-bi
Sample name
AC
A1/0 A2/0 A3/0 B1/0 B2/0
CE
Sample name/Sample 0
PT
Table 5: Relative slope load-displacement curves
1.52 3.49 2.62 2.47 4.51
Table 6: FCF, ηfcf and δs values Sample name
FCF
ηfcf (mm)
δs (mm)
A1 A2 A3 B1 B2
0.25 0.095 0.11 0.603 0.437
0.472 0.252 0.487 1.17 0.810
1.86 2.66 4.30 1.94 1.85
NU
SC RI PT
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
MA
Graphical abstract
ACCEPTED MANUSCRIPT Highlights
CE
PT
ED
MA
NU
SC RI PT
Enamel coatings improve the corrosion resistance of aluminium foam; Enamels increase the fire resistance of the foam maintaining the incombustibility; The bending stiffness results higher for enamel coated foam than for bare foam.
AC
Stefano Rossi, Lorenzo Bergamo, Vigilio Fontanari PII: DOI: Reference:
S0264-1275(17)30654-8 doi: 10.1016/j.matdes.2017.06.064 JMADE 3182
To appear in:
Materials & Design
Received date: Revised date: Accepted date:
22 March 2017 27 June 2017 28 June 2017
Please cite this article as: Stefano Rossi, Lorenzo Bergamo, Vigilio Fontanari , Fire resistance and mechanical properties of enamelled aluminium foam, Materials & Design (2017), doi: 10.1016/j.matdes.2017.06.064
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Fire resistance and mechanical properties of enamelled aluminium foam Stefano Rossi*, Lorenzo Bergamo, Vigilio Fontanari Department of Industrial Engineering, University of Trento, Via Sommarive 9, 38123 Trento Italy.
SC RI PT
*Corresponding author: [email protected]; Tel.: +39-0461-924069
Abstract
PT
ED
MA
NU
Aluminium foams show very interesting properties for a wide range of applications, in particular good mechanical properties and fire resistance. The particular cellular structure reduces the corrosion resistance of these materials in several environments. The application of a vitreous enamel layer is a possible solution to obtain a good protective barrier that can also improve the resistance at high temperatures. In this work these properties are evaluated for different porcelain enamelled aluminium foam samples. Starting from plate samples of aluminium foam produced by Alporas method, different enamel systems are studied. A frit without vanadium suitable for aluminium substrate was used. Microstructure analysis was made to highlight the adhesion between enamel and metallic substrate, as well as the filling of the open cells and the presence of defects in the glassy layers. The response to flame exposure of enamelled samples was tested in comparison to the uncoated foam. Finally, mechanical properties were studied by 4-points bending tests. This study has shown that the presence of the vitreous enamel coating increases the fire resistance of the aluminium foam. In addition, it was found that the flexural rigidity increases strongly and it is connected with the particle dimension in the frit.
AC
CE
Keywords Aluminium foam; vitreous enamel coating; fire resistance; mechanical properties
ACCEPTED MANUSCRIPT 1
Introduction
AC
CE
PT
ED
MA
NU
SC RI PT
The unique properties of metallic foams are useful in a number of potential application fields including damping, heat exchange, sound insulation, sound absorption and energy absorption [1-6]. Aluminium foam is one of the main cellular structures that has been developed in recent years thanks to the evolution of a number of cost-effective production processes [7,8]. These structures are increasingly used in marine, transportation, building construction and high-performance engineering applications [9-15] because of their high specific stiffness, specific strength, environmental resistance and thermal insulation features [16,17]. Other application is the use of this material as internal walls of tunnel and subway where it is important to have a highly glossy and cleanable surface, as well as corrosion resistance and fire resistance. Compared to dense materials the foam panels present a very interesting reduction of specific weight, which permits a lighter and more manageable structures. However the presence of opened cells are a dangerous critical point as concern the corrosion behaviour. In addition, the limited corrosion resistance of aluminium foams in several environments is an important problem that limits their wide application, as for example in presence of aggressive ions or in environments with a low pH value [4,5,18,19]. An approach widely used to improve the surface properties employs coating technologies to obtain protective surface layers [20,21]. Unfortunately, due to the complex surface morphology of aluminium foams most coating techniques that are well suited for bulk materials cannot be applied on metal foams. A coating deposited only on internal surfaces of the opened cells results insufficient to guarantee an optimal corrosion protection, because of the unavoidable presence of defects [22,23]. Figure 1 shows a 3D image of the surface morphology obtained with the software NIS Elements 4.20 using a an optical stereomicroscope Nikon SMZ25.
Fig. 1. 3-D Foam surface morphology.
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
MA
NU
SC RI PT
In this study, a protective vitreous enamel coating deposited to improve corrosion resistance is evaluated with the aim of preserving both the incombustibility and the high temperature resistance of the material. Enamel coatings have not the target to be a heat shield for components. Porcelain enamel is a vitreous inorganic coating that can be applied to metallic structures, such as steel and aluminium alloys [24]. Enamel has been used as a decorative material for over 4500 years. In the 19 th century enamelling became industrialized. Today it has been used both for decorative purpose and for protection against corrosion as well as for its engineering properties [25]. Porcelain enamel is characterized by several interesting properties, such as good durability due to high corrosion resistance and excellent resistance to UV radiation, chemical and high temperature resistance [26,27], that allow many applications in different fields. The corrosion resistance of this type of coating is due to its glassy nature (intrinsic chemical inertia), absence of defects, high thickness and excellent adhesion obtained during the firing process [24]. Aluminium components used in the field of road infrastructures, interior of tunnels, subway walls, façades, building constructions and ships are extensively protected using this type of coatings, in particular where high durability and fire resistance are required. Usually an aluminium alloy can be defined enamelable when, after the firing treatment, a good adhesion between metal and enamel with low defectiveness has been obtained. [2836] However, for aluminium alloy substrates some critical aspects should be taken into consideration. Considering the low meting point of aluminium, low firing temperature is requested. In this situation the adhesion between substrate and coating could be affected. For this reason, enamels developed for aluminium and its alloys should have a low melting temperature (since the substrate cannot exceed 600°C in the firing step). Since 30 years vanadium-based enamels are used on aluminium alloy to guarantee a good adhesion with substrate. Various vanadium compounds, as the divanadium pentoxide (V2O5), are the mostly utilized. However, this compound is harmful to human health and toxic to the environment [37]. Therefore vanadium-free enamels have been developed and used in this research. There are several deposition methods to obtain an enamel glassy layer on metallic substrate characterized by smooth surfaces [27]. Due to the complex geometry of the foam, the surface coverage is a critical point for enamel application and a suitable deposition process has to be developed and optimized. Normally to have a good deposition of glassy enamel it is necessary to prepare the surface eliminating the contaminants and producing an oxide layer, which favours the adhesion between substrate and coating. In addition the presence of contaminants on the substrate could produce bubbles during the firing process, detrimental for the quality of the layer. Surface preparation is not easy in the presence of opened cells; in addition clearly it is difficult to obtain a uniform layer without bubbles. In this work, several different types of vitreous enamel coating systems have been developed. The most important properties required for traditional aluminium foam in infrastructure and transportation applications, such as flexural resistance and fire
ACCEPTED MANUSCRIPT
2.1
Materials processing and samples preparation Materials processing
MA
2
NU
SC RI PT
behaviour, are evaluated to observe how the presence of different solutions can affect these properties. Fire resistance and thermal insulation properties are the main issues for the fireretardant structure that are requested in road infrastructure or in interior of tunnels applications [21]. The fire resistance means the ability to prevent the spread of the fire. Therefore, the fire-retardant structure must insulate the thermal radiation of the fire to have a safe time during fire events. In order to reduce the thermal transmittance through the foam, an enamel coating was deposited on only one foam surface and the heat propagation was investigated. Since several coated panels may undergo to flexural loads in service, the knowledge in this context need to be improved. While mechanical properties of aluminium foams such as compressive behaviour has been extensively studied in particular where the energy absorption results important [38-43], other studies are focused on the flexural properties of foams when this material is used as a core of sandwich panels [44-47]. Considering the brittle nature of deposits and the final application of the investigated enamelled foams, the mechanical characterization has been focused on flexural tests.
AC
CE
PT
ED
The foams object of this research is produced following Alporas industrial process and supplied by Foamtech Co. Ltd (Seoul, Korea). They are closed-cell aluminium foams with a density about 0.2 g/cm3. The Alporas process is a direct foaming technology of metals where a blowing agent (usually 1.6 wt% TiH2) is added to the melt. The blowing agent, under the influence of heat, decomposes and releases gas acting as propellant of the foaming process. The exact procedure is described in literature [7]. Briefly, the manufacturing process to obtain the panels consists of five stages: thickening, foaming, cooling, foamed block and slicing. In the melted aluminium at 680°C calcium compounds are added to increase the viscosity of the melt. Titanium hydride is then added and hydrogen is thus released. Consequently the melted material fills the mould, due to expansion process. In this step the cell structure is obtained. After aluminium solidification during the following cooling step the foam block is obtained. Sheets of different thickness can be finally produced from this block by mechanical slicing This operation necessary to obtained panels from the foamed block leaves surface opened cells. The enamel were deposited starting from a low temperature frit without vanadium oxide, supplied by Emaylum Italia (Chignolo d’Isola, BG, Italy), whose composition is reported in Table 1. The frit used for the different layers deposition is characterised by the same flake shapes and composition, in order to minimize the number of parameters adopted during the deposition process.
ACCEPTED MANUSCRIPT Table 1: Frit composition expressed as wt% of oxides Glass Formers Melting agent Opacifiers Modifiers
55 wt% 35 wt% 5 wt% bal.
Samples production
SC RI PT
2.2
SiO2 + B2O3 Na2O + K2O + Li2O TiO2 Al2O3 + ZnO + P2O5 + SrO
AC
CE
PT
ED
MA
NU
All samples used in this research were produced in laboratory. Each aluminium foam panel (100 x 50 mm2) was pretreated by immersion in water solution with 20 wt% NaOH for 50 seconds, followed by distilled water rinse and drying by air flow. This is the common process to prepare an aluminium surface eliminating the contaminates and activating the surface to increase adhesion. As commonly requested by the different typical applications, only one face of the foams was coated in order to limit the weight increase of the sample. Table 2 resumes the investigated porcelain enamelled systems. The choice of the vitreous enamel systems is dictated to seal the opened porosity, to guarantee a good adhesion between the deposit and the aluminium substrate and finally to obtain a defectless layer of sufficient thickness to perform as a protective barrier. Typical enamel systems are made of a ground layer, to improve the adhesion with the substrate and to enable a levelling effect of the topcoat, that must show good resistance in acid and basic environments, colour stability, corrosion protection as well as other specific properties dictated by the application. Following this standard the studied systems were initially composed by two layers, a ground one, that could favour the adhesion of the second layer named topcoat, that could guarantee the protection properties covering the whole surface. In the studied samples the ground layer, in addition to guarantee a good adhesion of the enamel, has also the function to seal the opened cell. Considering the difficulty of this operation two different frit grain dimensions are used. In order to improve the not optimal adhesion observed in the first studied samples, in a second time a first layer, named primer, was applied before ground deposition with the aim to enhance the adhesion between ground and substrate. Table 2: Studied samples Sample name
Primer
A1
Ground (type)
Topcoat
Number of layers
✓ (A)
✓
2
Number of firing treatment 2
A2
✓
✓ (A)
✓
3
2
A3
✓
✓ (A)
✓
3
3
✓ (B)
✓
2
2
✓ (B)
✓
3
2
B1 B2
✓
ACCEPTED MANUSCRIPT
MA
NU
SC RI PT
For the ground layers a frit with the chemical composition shown in Table 1 was used. Vitreous particles with two different size, 500 µm for ground A and 350 µm for ground B, were used (Fig. 2). These particles are obtained by milling and sieving the frit flakes. 10 wt% clay, 0.5 wt% sodium aluminate and 20 wt% water were added to the frit to produce a slurry used for wet spraying deposition of layer. Drying (200°C for 15 minutes) and firing treatments at a temperature of 580°C for 20 minutes were carried out in a heating chamber.
ED
Fig. 2. Micrographs of vitreous frit particles used for ground layer deposition: a) max size 500 µm for Ground A; b) max size 350 µm for Ground B.
AC
CE
PT
The topcoat was subsequently deposited. The frit (the composition is shown in Table 1) was milled together with 3.5 wt% inorganic dyes (CoO) and 0.5 wt% sodium aluminate and then the enamel powder was sieved down to 120 µm. After the addition of 38 wt% of water, the obtained slurry was deposited by wet spraying. The topcoat enamel was dried at 200°C for 15 minutes and fired at 560°C for 15 minutes in a heating chamber. To improve the adhesion between the ground and the substrate surface in some cases a primer layer before the ground deposition was applied. The primer was obtained by milling the frit together with 0.5 wt% sodium aluminate, sieving down the enamel powders to 120 µm and adding 38 wt% of water. The obtained slurry was then deposited by wet spraying application. The enamel was dried at 200°C for 20 minutes in a heating chamber. Only for the sample A3 a firing treatment at a temperature of 560°C for 15 minutes was carried out in a heating chamber for the primer layer to evaluate the effect on the mechanical and thermal properties of the enamelled aluminium foam. For the other samples the first firing step was made after ground application. Considering the adhesion between aluminium substrate and glassy layers, the firing treatment of the primer layer on sample A3 did not produce observable improvement in comparison of sample A2. Its main drawback is the negative increase of production complexity. Indeed the extra fire treatment increases the process time and cost. This research is carried out in university and industrial laboratories and then the samples are not of industrial production. However, considering the industrial use of the enamelled coatings it is possible to have an estimate
ACCEPTED MANUSCRIPT of the presumable cost. An indication cost for a deposit of 100 µm is about 0.02 Euro/dm2. The thermal treatment cost is comparable to the cost of a curing treatment of a paint. For each protection system four different samples were prepared. The density of uncoated foam is 0.2 gr/cm3. Clearly with the presence of the glassy deposits an increase of density was expected. The final density was equal to 0.43 gr/cm3 for sample A1, 0.44 gr/cm3 for samples A2, B1 and B2 and finally 0.49 gr/cm3 for samples A3. In the last case the three firing treatments produce a denser deposit system.
3.1
Experimental methods
SC RI PT
3
Microstructural analysis
The microstructural characterization was carried out on the cross-section using a JEOL IT300 scanning electron microscope. Fire resistance test
NU
3.2
AC
CE
PT
ED
MA
The flame test was performed considering the process described by A. Lakshmanam et al. [48]. Enamelled foam was subjected to a fully engulfing fire with an average flame temperature of 600°C for 10 minutes. The setup of the test is shown in Fig. 3a. The burner was operating with propane. The size of the samples was 50 x 50 x 13 mm. Samples were kept vertically, with the coated surface exposed to fire. The system represents a fully affected fire with a homogeneous temperature distribution on the fireside of the samples. Two appropriately machined refractory bricks (Fig. 3b) were used to prevent heat and mass loss though the borders of the investigated assembly. In this way a better comparison between the results of the different samples was reached. Fig. 3c shows the sample assembled in the refractory insulation materials. The distance between the coated foam surface and the burner was adjusted to 100 mm, so that an average flame temperature of 600°C reaches the sample surface. Temperature measurement was performed by Type-K (chromel alumel) probe-style thermocouple (model Type K Thermocouple producer Thermometrics - Norhridge US), which was connected to a computer and LabVIEW interface via a data acquisition control unit. This thermocouple (Chromel - 90% nickel and 10% chromium - Alumel - 95% nickel, 2% manganese, 2% aluminium and 1% silicon) shows a measure range between -200°C to +1260°C with tolerance in the measurement experimental range of 0,07%. A thermocouple was located to the rear face of the sample, as shown in fig. 3c. The junction was fixed in intimate contact with the specimen by means of aluminium foil adhesive. An infrared thermo-camera (FLIR T-Series T460) was implemented to capture the sample temperature. The camera was positioned on a tripod with the imaging field capturing one of the side cross section of the sample.
CE
PT
ED
MA
NU
SC RI PT
ACCEPTED MANUSCRIPT
3.3
AC
Fig. 3. a) Fire test setup; b) machined refractory bricks; c) sample assembly in refractory insulation materials; d) view of the enamelled surface of the sample, exposed to the direct flame.
Static bending tests
Four-point bending tests were performed to measure the flexural stiffness, defined as bending stiffness, , following the ASTM D6272-10 Standard. This test was conducted on a Tinius Olsen H10KT electromechanical testing machine at the loading rate of 2 mm/min. The size of the samples was 100 mm in length, 20 in width and 13 mm in thickness; the outer span during the test was 80 mm and the inner span was 20 mm. The specimen was positioned to have the coated surface under tensile stress. The test was replicated using 4 samples for each type.
ACCEPTED MANUSCRIPT The load-displacement curves were acquired; the samples were considered failed when the curves reached a steadily declining slope. The bending stiffness was calculated using Eq. (1) [49]. (1)
ED
MA
NU
SC RI PT
Where L is the size of the outer span, P is the load, is the displacement of the beam in correspondence of the loading points A is the geometric parameter represented in Fig. 4. The four-point bending strength values which were reported in this work, were the averages of four tests.
CE
4.1
Result and discussion
Characterization of enamel layers
AC
4
PT
Fig. 4. 4-Point bending test configuration (the black line represents the enamel layer).
Fig. 5 shows the cross section of the different samples. Considering the microstructural analysis of the samples, it is possible to clearly observe for the sample A1 (Fig. 5a, dark area between Al substrate and light area indicated as Ground A) a debonding between the coating and the substrate. During the thermal treatment, the sintering of the vitreous particles involves a volumetric shrinkage of the ground layer producing a lack in adhesion between enamel coating and substrate. In addition, the ground is characterized by a huge amount of internal porosity, due to the lower degree of sintering, probably connected with the presence of big frit particles. The microstructural observation of the samples with ground A highlights the presence of high volume vitreous particles that probably were not melted during the firing process. For sample A2 (Fig. 5b) the presence of a primer layer increases the coating-substrate adhesion. Nevertheless, it is not enough to completely counteract the effect of ground
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
MA
NU
SC RI PT
shrinkage during the heat treatment. Observing the Fig. 5b (dark area between white primer and Al substrate) it is possible to see a debonding within the primer, where one part of primer remains adherent to the cell surface of the substrate and the other part is connected to the ground. This lack of adhesion is due to the shrinkage force of vitreous particles obtained during the sintering process. For sample A3 (Fig. 5c) the application of three firing treatments involves an improvement of the adhesion between substrate and coating, in comparison with that observed in sample A2. This improvement probably is due to the presence of a consolidate primer, that prevents the shrinkage of the ground during the sintering of the particles. For sample B1 it is possible to observe a debonding between coating and substrate (Fig. 5d, the grey area between Al substrate and ground layer) occurred during the thermal cycle of the sintering process. Therefore, this debonding is characterized by a lower volumetric shrinkage of the ground connected with the smaller size of the vitreous particles (compared with ground A), that allows to achieve a higher density before firing. This increase of density is related to better particle packing because the smaller particles fill the voids between the bigger ones. The density is influenced by the ratio between big and small particles size [50]. The better sintering process produces higher compactness and smaller amount of pores for the ground B compared to the ground A. With smaller particles size lower temperature and time are requested to obtain a certain sintering degree because of the enhancement of the driving force for the process due to higher surface area [50]. For this reason the opened cells of the samples B are filled to an higher extent than those of samples A. Furthermore, the ground B shows lower difference of morphology with the topcoat, probably due to a better interdiffusion between the vitreous particles of the two different layers during the firing treatment. Sample B2 (Fig. 5e) presents the highest quality characterized by the best adhesion and open cells filling. This coating system shows a good homogeneity between the three different layers thanks to the better sintering of the vitreous particles. The internal porosity is characterized by voids with a small average diameters. Differently from sample A2, for this sample the presence of a dry primer layer is sufficient to counteract the ground shrinkage during the vitreous particles sintering as a consequence of the lower ground shrinkage occurring during the process. Finally, using the ground B with the number of firing treatments it is possible to obtain a good filling of the open cells, during enamel deposition. The function of the ground layer is to seal the opened porosity. The difference of frit particles dimension does not influence in the thickness of the deposits. Considering this aspect it is not easy to define and measure exactly the thickness due to the opened cell. However, if the surface obtained by completely filling the opened cells is considered as reference level, the thickness of the deposits ranges between 370 and 500 μm for all samples.
AC
CE
PT
ED
MA
NU
SC RI PT
ACCEPTED MANUSCRIPT
Fig. 5. Microstructural cross-sections of vitreous enamel coatings deposited on aluminium foam substrate: a) sample A1; b) sample A2; c) sample A3; d) sample B1; e) sample B2; 4.2
Flame test experimental results and discussions
Sequential infrared (IR) images of enamelled samples and of bare foam during flame test are displayed in Fig. 6. As can be seen, bare aluminium foam reached thermal
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
MA
NU
SC RI PT
saturation in about 60 seconds, while more than 90 seconds are necessary for the enamelled aluminium foams to reach the stationary condition. The high flame retardant behaviour of aluminium foam with closed cells is connected to the high quantity of porosity and to low heat permeability of air, which restricts fire and prevents the heat transfer [50]. In case of enamelled aluminium foam the increased flame retardant properties of the foam are associated to thermal barrier behaviour of vitreous coating. Considering the random distribution of the opened cell and of the aluminum walls under the enamel layer together with the resolution of infrared camera it is not possible to exactly correlate the local temperature with the pore structures. However the temperature evolution during flame exposure permits to differentiate the samples behavior.
Fig. 6. Sequential IR images of bare foam (sample 0) and coated foams (samples A1, A2, A3, B1, B2) during flame test showing temperature profile in side wall. Table 3 shows the data fitting of temperature-time curves, detected by the thermocouple (Fig. 7). Using the Eq. 2 (that describes the temperature vs time curve considering the intrinsic properties of the material) it is possible to obtain the parameter bi (Eq. 3). Then, using the Eq. 4, it is also possible to calculate the apparent thermal diffusivity coefficient [51] that allows a comparison between the thermal behaviour of the different samples.
ACCEPTED MANUSCRIPT (2)
(3)
SC RI PT
(4)
AC
CE
PT
ED
MA
NU
Where T is the temperature detected by the thermocouple, is the temperature at saturation time, Ti is the initial temperature of the foam face opposite to that exposed to fire (that for a better fitting, in this work, is considered as 100 °C, avoiding then the initial transient part of the temperature-time curve), is the apparent thermal conductivity, is the foam density, Cp specific heat of aluminium and d is the foam thickness (13 mm). A lower value of the heat diffusivity inside of the foam structure increases the time employed by the foam to reach the saturation temperature.
Fig. 7. Temperature-time curves of fire test.
Table 3: Values of parameter -bi
ACCEPTED MANUSCRIPT -bi
Sample name 0 A1 A2 A3 B1 B2
0.0338 0.0123 0.0154 0.0162 0.0192 0.0231
SC RI PT
It is possible to observe (Table 4) a remarkable increase in thermal insulating for enamelled samples compared to bare samples. The highest thermal insulation is shown by samples A, probably connected with the higher internal porosity. Due to the low adhesion between enamel layer and aluminium walls and a lower sintering level of ground A more voids are present at interface and inside the glassy mater. The presence of primer layer lowers the thermal insulating property as a result of reduction of internal porosity. Table 4: Apparent thermal diffusivity
Static bending test
MA
ED
4.3
0.00044 0.00016 0.00020 0.00021 0.00025 0.00030
PT
0 A1 A2 A3 B1 B2
NU
Sample name
AC
CE
A preliminary test using a bare aluminium foam beam specimen was carried out for comparing the flexural strength with the enamelled ones. The typical load-displacement curves measured by four-point bending tests for all the samples are shown in Fig. 8. The reported curves represent the typical behaviour, considering a replication on 4 specimens, of the samples. The stiffness values were determined using eqn. 1. To this purpose a linear fitting was performed to determine the slope of the curve in the range between 5 and 40 N, where the curves can be considered as approximately linear. The calculated values are reported in Table 5 as relative values, by comparing the coated samples results with those of the bare material (Sample 0). It is possible to observe an impressive bending stiffness increase for enamelled foams compared to the bare foam. This effect is given by hard vitreous enamel coating, characterized by glassy nature. The application of a primer layer allows to further increase the stiffness. In addition, samples B are characterized by a higher stiffness due to the more homogeneous and compact ground layer that delays the early crack formation at the porosity tips. This behaviour is also influenced by the higher adhesion level between glassy mater and aluminium walls.
ACCEPTED MANUSCRIPT
90 80 70
50 40
Sample 0 Sample A1 Sample A2 Sample A3 Sample B1 Sample B2
30 20 10 0 0.0
0.2
0.4
0.6
0.8
1.0
SC RI PT
Load (N)
60
1.2
1.4
1.6
1.8
2.0
NU
Displacement (mm)
MA
Fig. 8. Load-displacement curves from 4-point bending test.
ED
Table 5: Relative slope load-displacement curves
4.4
1.52 3.49 2.62 2.47 4.51
CE
AC
A1/0 A2/0 A3/0 B1/0 B2/0
PT
Sample name/Sample 0
Enamel coating brittleness
The brittleness of porcelains coating could be investigated by evaluating the point of “First Failure Crack” for each different deposited enamel coating obtained during the fourpoint bending test. To individuate the first crack failure it is considered the instant where a sudden decrease of load displacement curve was observed. The crack behaviour of samples was studied using the load-displacement curves shown in Fig. 8, extrapolating the values of displacement in which the first coating crack occurs. From the literature [52] is known that FCF (First Crack Failure) is the parameter used for describing the crack behaviour of the coating. It is calculated using the following equation:
ACCEPTED MANUSCRIPT (5) Where is the coating (coatings plus glass filled open porosity) thickness and is the displacement at the first crack onset. Then FCF is an adimensional value that describes the enamel coating brittleness. The higher the FCF the lower is the enamel coating brittleness. The average values for each sample are reported in Table 6. Table 6: FCF, ηfcf and δs values FCF
ηfcf (mm)
δs (mm)
A1 A2 A3 B1 B2
0.25 0.095 0.11 0.603 0.437
0.472 0.252 0.487 1.17 0.810
1.86 2.66 4.30 1.94 1.85
SC RI PT
Sample name
Conclusion
ED
5
MA
NU
It is possible to observe that using the ground B the coating is characterized by a lower brittleness probably as a result of the higher homogeneity and compactness of the coating itself. On the other hand, the deposition of a primer layer produces the increase of the enamel brittleness, probably as a result of the raise of sample rigidity.
AC
CE
PT
The results of the characterization of different enamel coating systems applied to aluminium foams, can be summarized as follows: The different vitreous enamel systems studied in this work allow a good degree of coverage of the foam surface; The smaller size of the frit vitreous particles, improves the sintering, during the heat treatment, thus producing a more homogeneous and compact layer; The deposition of a primer layer improves the adhesion between the substrate and the coating. On the other hand, it increases the cost and time of the production process as well as the weight of the coated foam; Each sample retains the incombustibility during fire exposure. The enamelled aluminium foams show a lower heat propagation due to the barrier effect guaranteed by the presence of vitreous coating. Samples characterized by ground A and by the lack of a primer layer shows better heat insulation performances; The application of the enamel coating highly increases the bending stiffness of the foam. In particular the use of ground B and the application of a primer layer produce a higher increase in the bending stiffness of the foam; For the adopted testing configuration (coating on the tensile side during test), the samples with ground B and without primer layer show less brittle behaviours;
ACCEPTED MANUSCRIPT
The deposition process adopted for the sample A3 highly increases time and cost of the process; in addition, this process doesn’t improve the mechanical and thermal properties as requested by the possible applications. For this reason this deposition process should be not taken in account for further studies. The evaluated protection systems are a representative view of the porcelains enamel layers normally applied and then they could show the properties, which could be obtained with a glassy enamel deposition.
SC RI PT
Acknowledgements The authors are grateful to Attilio M. Compagnoni, (Emaylum Italia, Italy) for supplying the enamel samples and useful discussion. Thanks also to Alberto Bettini (Vaber Industriale, Italy) and to Sylove Won (Foam Tech Global, Korea), for aluminium foam materials supply.
NU
References
MA
[1] T.J. Lu, J.M. Ong, Characterization of closed-celled aluminium alloys, Journal of Materials Science, 2001, Vol. 36, pp. 2773-2786.
ED
[2] M.F. Ashby, A.G. Evans, N.A. Fleck, L.J. Gibson, J.W. Hutchinson, Metal Foams: A Design Guide, Butterworth-Heinemann, Woborn, USA, 2000.
PT
[3] T.J. Lu, C. Chen, Thermal transport and fire retardance properties of cellular aluminium alloys, Acta mater.1999, Vol. 47, pp. 1469-1485.
CE
[4] F. Stergioudi, C.A. Vogiatzis, K. Gkrekos, N. Michailidis, S.M. Skolianos, Electrochemical corrosion evaluation of pure, carbon-coated and anodized Al foams, Corrosion Science, 2015, Vol. 91, pp 151–159.
AC
[5] N. Movahedi, A. Habibolahzadeh, Effect of plasma electrolytic oxidation treatment on corrosion behavior of closed-cell Al-A356 alloy foam, Materials Letters, 2016, Vol. 164, pp. 558–561. [6] Y. Zhou, Y. X. Li, Oxidation kinetics of oxide film on bubble surface of aluminum foams produced by gas injection foaming process, Trans. Nonferrous Met. Soc. China, 2015, vol. 25, pp. 2781−2788. [7]J. Banhart, Manufacture, characterisation and application of cellular metals and metal foams, Progress in Materials Science, Vol. 46, 2001, pp. 559-632.
ACCEPTED MANUSCRIPT [8] K. Heim, G. S. Vinod-Kumar, F. Garcia-Moreno, J. Banhart, Stability of various particlestabilised aluminium alloys foams made by gas injection, J Mater Sci, 2017, Vol. 52, pp. 6401–6414. [9] A.E. Simone, L.J. Gibson, Effects of solid distribution on the stiffness and strength of metallic foams, Acta Materials, Vol. 4, 1998, pp. 2139-2150.
SC RI PT
[10] H.N.G. Wadley, Cellular metals manufacturing, Adv. Eng. Mater., Vol. 4, 2002, pp. 726-733. [11] H.P. Degisher, B. Kriszt, Handbook of Cellular Metals: Production, Processing, Applications, Wiley-VCH, Weinheim, Germany, 2002.
NU
[12] Y. Wang, J.Y. R. Liew, S. Chin Lee, X. Zhai, W. Wang, Crushing of a novel energy absorption connector with curved plate and aluminum foam as energy absorber, Thin Walled Structures 2017, Vol. 111, pp 145–154.
MA
[13] J. Baumeister, J. Banhart, M. Weber, Aluminium foams for transport industry, Materials Design, 1997, Vol.18, pp 217- 220. [14] F. García-Moreno, Commercial Applications of Metal Foams: Their Properties and Production, A review, Materials, 2016, Vol. 85, n.9, pp. 1-27.
PT
ED
[15] E. Solórzano, J.A. Reglero, M.A. Rodríguez-Pérez, D. Lehmhus, M. Wichmann, J.A. de Saja, An experimental study on the thermal conductivity of aluminium foams by using the transient plane source method, International Journal of Heat and Mass Transfer, 2008, Vol. 51, pag. 6259–6267.
AC
CE
[16] Lee DG, Suh NP, Axiomatic design and fabrication of composite structure applications in robots, machine tools and automobiles, New York, Oxford University Press, 2006. [17] Yu HN, Kim SS, Do Suh J, Lee DG, Axiomatic design of the sandwich composite endplat for PEMFC in fuel cell vehicles, Compos Struct, 2010:92:1504-11. [18] J.R. Davis, Corrosion of Aluminium and Aluminium Alloys, ASM International, Materials Park, Ohio, USA, 1999. [19] X. Xia, Z. Zhang, J. Wang, Compressive characteristics of closed-cell aluminium foams after immersion in simulated seawater, Materials and Design, 2015, Vol. 67, pp 330-336. [20] C.M. Cotell, J.A. Sprague, F.A. Smidt, Surface Engineering, Vol. 5; ASM International, Novelty, OH, 1994.
ACCEPTED MANUSCRIPT [21] T. Abdulla, A. Yerokhim, R. Goodall, Effect of Plasma Electrolytic Oxidation coating on the specific strength of open-cell aluminium foams, Materials and Design, 2011, Vol. 32, pp 3742-3749. [22] S. Rossi, M. Fedel, L. Da Col, F. Deflorian, S. Petrolli, Coatings to increase the corrosion behaviour of aluminium foam, Surface Engineering, 2017, Vol. 33, pp 405-409.
SC RI PT
[23] S. Rossi, M. Calovi, M. Fedel, Corrosion protection of aluminium foams by cataphoretic deposition of orgnaic coatings, Progress in Organic Coatings, 2017, Vol 109, pp 144-151. [24] S. Rossi, A. Quaranta, L. Tavella, F. Deflorian, A.M. Compagnoni, Innovative Luminescent Vitreous Enameled Coatings, Intelligent Coatings for Corrosion Control, Chapter 7, Elsevier, Oxford, UK, 2015, pp. 251-282.
NU
[25] A. Umbertazzi, N. Wojciechowski, Vitreous Enamel, Ulrico Hoepli Editore, Milano, 2002.
MA
[26] S. Pagliuca, W.D. Faust, Porcelain (vitreous) Enamels and Industrial Enamelling Processes, The International Enamellers Institute, Mantova, Italy, 2011.
ED
[27] A.I. Andrews, Porcelain Enamels: The Preparation, Application and Properties of Enamels, 3rd ed., Tipografia Commerciale, Mantova, Italy, 2010. recent
research
and
PT
[28] S.L. Lee, N.E. Shanmugam, Aluminium structures: developments, Elsevier applied science, New York, USA, 1991.
CE
[29] C. Vargel, Corrosion of aluminium, Elsevier, Amsterdam, 2004.
AC
[30] M. Stacey, C. Bayliss, Aluminium and Durability: reviewed by inspection and testing, Materials Today: Proceedings, 2015, Vol 2, pp 5088 – 5095. [31] M. Pathirana, N. Lam, S. Perera, L. Zhang, D. Ruan, E. Gad, Damage modelling of aluminium panels impacted by windborne debris, Journal of Wind Engineering & Industrial Aerodynamics, 2017, Vol 165, pp 1–12. [32] T.Theodosiou, K.Tsikaloudaki, D.Bikas, Analysis of the Thermal Bridging Effect on Ventilated Facades, Procedia Environmental Sciences, 2017 Vol 38, pp 397 – 404. [33] D. Izydorczyk, B. Sędłak, B. Papis, P. Turkowski, Doors with specific fire resistance class, Procedia Engineering, 2017, Vol 172, pp 417 – 425.
ACCEPTED MANUSCRIPT [34] F.L. Chen, X. He, H.M. Yin, Manufacture and multi-physical characterization of aluminum/high-density polyethylene functionally graded materials for green energy building envelope applications, Energy and Buildings, 2016, Vol 116, pp 307–317. [35] B. Sędłak, P. Sulik, D. Izydorczyk, M. Łukomski, Fire-stop wraps and collars with intumescent materials –performance comparison, Procedia Engineering 2017, Vol 172, pp 961 – 968.
SC RI PT
[36] T. Ihara, B. P. Jelle, T. Gao, A. Gustavsen, Accelerated aging of treated aluminum for use as a cool colored material for facades, Energy and Buildings, 2016 Vol 112, pp 184– 197. [37] A.M. Compagnoni, Reduction or elimination of vanadium in enamel for aluminium, in: Proceedings of the XXI International Enamellers Congress, IEC, Shanghai, 18-22 May, 2008.
MA
NU
[38] X. Xia, X. Chen, Z. Zhang, X. Chen, W. Zhao, B. Liao, B. Hur, Compressive properties of closed-cell aluminium foams with different contents of ceramic microspheres, Materials and Design, 2014, Vol. 56, pp 353-358.
ED
[39] J. Lázaro, E.Solórzano, M.A.Rodríguez-Pérez, A.R. Kennedy, Effect of solidification rate on pore connectivity of aluminium foams and its consequences on mechanical properties, Materials Science & Engineering A 6, 2016, Vol. 72, pp. 236–246.
PT
[40] J. Kováčik, Ľ. Orovčík, J. Jerz, High-temperature compression of closed cell aluminium foams, Kovove Mater., 2016, Vol. 54, pp. 429–440.
CE
[41] J. Lazaro, E. Solórzano, M. A. Rodríguez-Pérez, F. Garcıa-Moreno, Pore connectivity of aluminium foams: effect of production parameters, J Mater Sci, 2015, Vol. 50, pp. 3149– 3163.
AC
[42] I. Duarte, J.M.F. Ferreira. Composite and nanocomposite metal foams. Review. Materials 2016, Vol. 9(2), n. 79, pp. 1-34; doi:10.3390/ma9020079. [43] X. Xia, H. Feng, X. Zhang, W. Zhan, The compressive properties of closed-cell aluminium foams with different Mn additions, Materials and Design, 2013, Vol. 51, pp 797802. [44] A.M. Harte, N.A. Fleck, M.F. Ashby, Sandwich panel design using aluminum alloy foam, Advanced Engineering Materials, 2000, Vol. 2, pp 219-222. [45] G. Zu, B. Song, Z. Zhong, X. Li, Y. Mu, G. Yao. Static three-point bending behavior of aluminum foam sandwich, Journal of Alloys and Compounds, 2012, Vol. 540, pp 275-278.
ACCEPTED MANUSCRIPT [46] L.L. Yan, B. Han, B. Yu, C.Q. Chen, Q.C. Zhang, T.J. Lu. Three-point bending of sandwich beams with aluminum foam-filled corrugated cores, Materials and Design, 2014, Vol. 60, pp 510-519. [47] I. Duarte, M. Vesenjak, L. Krstulović-Opara. Dynamic and quasi-static bending behaviour of thin-walled aluminium tubes filled with aluminium foam. Composite Structures. 2014, Vol. 109, pp 48-56.
SC RI PT
[48] A. Lakshmanan, Sintering of Ceramics: New Emerging Techniques, InTech, Rijeka, Croatia, 2012. [49] E. Carrera, G. Giunta, M. Petrolo, Beam Structures: Classical and Advanced Theories, Wiley, Chichester, UK, 2011.
NU
[50] E. Placido, M.C. Arduini-Schuster, J. Kuhn, Thermal properties predictive model for insulating foams, Infrared Physics and Technology, 2005, Vol. 46, pp. 219-231.
MA
[51] J.H. Lienhard IV, J.H. Lienhard V, A Heat Transfer Textbook, Cambridge, 3 rd edition, Massachusetts, USA, 2001.
AC
CE
PT
ED
[52] A. Chelli, R. Poletti, L. Pignatti, F. Bruscoli, G. Pasqualetti, F. Bruni, A. Zucchelli, L. Rossetti, F. Lotti, G. Minak, V. Dal Re, S. Curioni, Composite Enameled Steel Elements for Air Preheaters and Gas-gas Heaters: an Integrated Approach from Sheet Forming and Enamelling to Basket Assembly, XXI International Enamellers Congress, 18-22 May 2008, Shanghai, China.
ACCEPTED MANUSCRIPT Figures captions list Fig. 1. 3-D Foam surface morphology. Fig. 2. Micrographs of vitreous frit particles used for ground layer deposition: a) max size 500 µm for Ground A; b) max size 350 µm for Ground B. Fig. 3. a) Fire test setup; b) machined refractory bricks; c) sample assembly in refractory insulation materials; d) view of the enamelled surface of the sample, exposed to the direct
SC RI PT
flame. Fig. 4. 4-Point bending test configuration (the black line represents the enamel layers). Fig. 5. Microstructural cross-sections of vitreous enamel coating deposited on aluminium foam substrate: a) sample A1; b) sample A2; c) sample A3; d) sample B1; e) sample B2; Fig. 6. Sequential IR images of bare foam (sample 0) and coated foams (samples A1, A2,
Fig. 7. Temperature-time curves of fire test.
NU
A3, B1, B2) during flame test showing temperature profile in side wall.
MA
Fig. 8. Load-displacement curves from 4-point bending test.
ED
Table list Table 1: Oxide composition frits
CE
PT
SiO2 + B2O3 Na2O + K2O + Li2O TiO2 Al2O3 + ZnO + P2O5 + SrO
55 wt% 35 wt% 5 wt% bal.
AC
Glass Formers Melting agent Opacifiers Modifiers
Table 2: Studied samples Sample name
Primer
A1
Ground (type)
Topcoat
Number of layers
✓ (A)
✓
2
Number of firing treatment 2
A2
✓
✓ (A)
✓
3
2
A3
✓
✓ (A)
✓
3
3
✓ (B)
✓
2
2
✓ (B)
✓
3
2
B1 B2
✓
ACCEPTED MANUSCRIPT
Table 3: Values of parameter -bi
0 A1 A2 A3 B1 B2
0.0338 0.0123 0.0154 0.0162 0.0192 0.0231
Table 4: Apparent thermal diffusivity coefficient
MA
0.00044 0.00016 0.00020 0.00021 0.00025 0.00030
ED
0 A1 A2 A3 B1 B2
NU
Sample name
SC RI PT
-bi
Sample name
AC
A1/0 A2/0 A3/0 B1/0 B2/0
CE
Sample name/Sample 0
PT
Table 5: Relative slope load-displacement curves
1.52 3.49 2.62 2.47 4.51
Table 6: FCF, ηfcf and δs values Sample name
FCF
ηfcf (mm)
δs (mm)
A1 A2 A3 B1 B2
0.25 0.095 0.11 0.603 0.437
0.472 0.252 0.487 1.17 0.810
1.86 2.66 4.30 1.94 1.85
NU
SC RI PT
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
MA
Graphical abstract
ACCEPTED MANUSCRIPT Highlights
CE
PT
ED
MA
NU
SC RI PT
Enamel coatings improve the corrosion resistance of aluminium foam; Enamels increase the fire resistance of the foam maintaining the incombustibility; The bending stiffness results higher for enamel coated foam than for bare foam.
AC