The effects of selected printing parameters on the fire properties of additively produced composites from neat polylactic acid (PLA) and wood/PLA filaments were investigated. The reaction to fire of the 3D-printed specimens was tested according to the ISO 5660-1 cone calorimeter test method. The results showed that the properties of the specimens when exposed to fire were significantly affected by the incorporation of wood flour into the PLA filament. It was also interesting that PLA specimens had much better reactions to fire than the wood/PLA specimens. Time to ignition was found to be much longer in the 3D-printed PLA specimens. Although the maximal heat release rate was a little higher in the PLA than the wood/PLA specimens, the duration of HRR was longer for the wood/PLA specimens. The initial mass of the specimens was smaller in the wood/PLA composites, but during the radiant heat exposure the mass typically decreased slower than in the PLA specimens.
3D printing, also known as additive manufacturing (AM), has recently gained significant attention in many industries for enabling rapid prototyping. This technology enables users to produce complex shaped products based on a virtual computer model. Furthermore, there is no material waste in this technology, which ensures efficient use of expensive materials [
The filaments used in the 3D printers can be produced from different materials, such as polymers, metals, and polymer composites. The use of natural fibres in thermoplastic filaments as reinforcing filler, and particularly wood, has increased significantly over the last decade due to wood’s significant advantages, such as being a nontoxic, environmentally friendly and sustainable material with a low-cost, easy supply, and good mechanical properties, causing less abrasive damage to processing equipment [
Fossil- and bio-based thermoplastics have a relatively high flammability and tendency to produce toxic gases that can harm human health [
The fire properties of PLA and PLA-based composites have been widely investigated in previous studies [
The effects of printing parameters on the physical and mechanical properties of 3D-printed wood/PLA composites have been extensively investigated in previous studies [
3D-printed wood/PLA composites can be produced using different printing parameters such as the printing layer thickness and infill rate. Undoubtedly, the reaction to fire of the resulting composites are affected by the changes in the layer thickness and infill rate. An extensive search of the literature did not reveal any study related to the effects of the printing parameters on the reaction to fire of 3D-printed wood/PLA composites. The effects of the layer thickness and infill rate on the reaction to fire of wood/PLA composites were thus investigated according to the ISO 5660-1 cone calorimeter test method. The results were compared with the properties of neat PLA specimens.
Commercially fabricated wood/PLA filaments (30 wt% wood and 70 wt% PLA) having a diameter of 1.75 mm were used in the manufacture of 3D-printed specimens. The wood/PLA filaments were purchased from a commercial 3D filament seller in Slovenia. The 3D-printed wood/PLA composite specimens with dimensions of 100 mm × 100 mm × 10 mm were produced at different processing parameters, varying the printing layer thickness and porosity (infill rate). A Zortrax M200 3D printer (Zortrax, Poland) was used to print the specimens for the cone calorimetry tests (
Printing parameters | Unit | Specimen code | |||||||||
---|---|---|---|---|---|---|---|---|---|---|---|
A | B | C | D | E | F | G | H | I | |||
Printing layer thickness | 0.29 | 0.19 | 0.09 | 0.29 | 0.19 | 0.09 | 0.29 | 0.19 | 0.09 | ||
Infill ratio | 40 | 40 | 40 | 60 | 60 | 60 | 80 | 80 | 80 | ||
Number of printing layers | mm | 2 (0.58)* | 3 (0.57) | 6 (0.54) | 2 (0.58) | 3 (0.57) | 6 (0.54) | 2 (0.58) | 3 (0.57) | 6 (0.54) | |
Printing time | hour:min | 7:59 | 4:52 | 3:35 | 8:32 | 5:09 | 3:41 | 9:55 | 5:38 | 4:07 | |
Used material | g | 75 | 64 | 56 | 92 | 79 | 68 | 111 | 98 | 80 | |
Set thickness of specimens (mm) | mm | 10 | 10.20 | 10.60 | 10 | 10.20 | 10.60 | 10 | 10.20 | 10.60 |
Note: *Total thickness of solid material at top and bottom is a multiply of printing layer thickness and number of solid layers (0.29 mm ×2 = 0.58 mm).
3D-printed PLA specimens of the same dimensions were prepared in the same way as the wood/PLA 3D-printed specimens. The neat PLA filaments were purchased from a commercial 3D filament seller in Slovenia.
Cone calorimeter testing allows characterization of many parameters, including ignition time, heat release rate, smoke release rate, and mass loss of specimens exposed to a heat flux. In this a cone heating element provided a radiative flux to the specimen (
The cone calorimeter tests were conducted in the horizontal orientation with the conical radiant electric heater set at a heat flux level of 25 kW/m2. The specimens were tested in the optional retainer frame but without the wire grid over the specimens. One specimen was tested for each type of 3D printing.
The ignitability of the test specimens was determined according to the EN ISO 11925-2 [
A comparison of the heat release rate (HRR) of the 3D-printed wood/PLA specimens produced at different printing parameters is given in
Specimen code | Infill ratio (wt%) (%) | Printing layer thickness (mm) | Heat release rate (kW/m2) | TSI (s) | THR600 (MJ/m2) | AEHOC* (MJ/kg) | ASEA** (m2/kg) | |||
---|---|---|---|---|---|---|---|---|---|---|
Max value | Average value over | |||||||||
60 s | 180 s | 300 s | ||||||||
A | 40 | 0.29 | 289 | 198 | 228 | 205 | 76 | 68.8 | 18.4 | 29.8 |
B | 0.19 | 281 | 198 | 216 | 222 | 81 | 79.1 | 18.3 | 25.1 | |
C | 0.09 | 293 | 194 | 212 | 226 | 77 | 95.9 | 18.5 | 20.3 | |
D | 60 | 0.29 | 298 | 206 | 212 | 229 | 85 | 87.2 | 18.5 | 26.2 |
E | 0.19 | 276 | 197 | 205 | 216 | 92 | 98.2 | 17.8 | 19.9 | |
F | 0.09 | 259 | 193 | 210 | 210 | 90 | 112.2 | 18.1 | 19.4 | |
G | 80 | 0.29 | 281 | 198 | 206 | 217 | 90 | 101.5 | 18.2 | 22.6 |
H | 0.19 | 273 | 192 | 207 | 210 | 117 | 110.7 | 18.3 | 22.4 | |
I | 0.09 | 272 | 194 | 208 | 210 | 106 | 110.6 | 18.4 | 21.1 |
Note: *Time between 150 in 350 s. **Time between 150 and 250 s. TSI: time to sustained ignition. THR: total heat released. AEHOC: average effective heat of combustion. ASEA: average specific extinction area.
Based on the curves of the heat release rate, the recorded observations were the maximum value of HRR (PHRR, kW/m2) and the heat release rates averaged over 60, 180 and 300 s (AHRR-60, AHRR-180 and AHRR-300, kW/m2) after sustained ignition. The HRR of the PLA specimens was found to be higher than that of the wood/PLA specimens (
Specimen code | Infill ratio (wt%) (%) | Printing layer thickness (mm) | Heat release rate (kW/m2) | TSI (s) | THR600 (MJ/m2) | AEHOC* (MJ/kg) | ASEA** (m2/kg) | |||
---|---|---|---|---|---|---|---|---|---|---|
Max value | Average value over | |||||||||
60 s | 180 s | 300 s | ||||||||
A | 40 | 0.29 | 285 | 208 | 249 | 228 | 177 | 73.1 | 15.5 | 92.1 |
B | 0.19 | 327 | 203 | 258 | 258 | 183 | 84.2 | 16.0 | 163.9 | |
C | 0.09 | 334 | 191 | 226 | 262 | 181 | 100.0 | 16.2 | 119.4 | |
D | 60 | 0.29 | 313 | 191 | 240 | 257 | 178 | 91.1 | 15.3 | 188.7 |
E | 0.19 | 342 | 190 | 222 | 259 | 194 | 93.0 | 15.4 | 269.5 | |
F | 0.09 | 337 | 193 | 225 | 242 | 209 | 103.9 | 14.7 | 143.7 | |
G | 80 | 0.29 | 327 | 203 | 230 | 261 | 228 | 100.0 | 10.6 | 248.2 |
H | 0.19 | 346 | 204 | 231 | 246 | 240 | 95.4 | 10.1 | 263.9 | |
I | 0.09 | 350 | 167 | 218 | 223 | 220 | 91.0 | 11.9 | 129.9 |
Note: *Time between 150 in 350 s. **Time between 150 and 250 s. TSI: time to sustained ignition. THR: total heat released. AEHOC: average effective heat of combustion. ASEA: average specific extinction area.
The peak value of HRR correlates with the initial mass of the specimen. Increasing the initial mass of specimens increased the maximum HRR value in PLA specimens and decreased the maximum HRR value for the wood/PLA specimens.
The results indicated that the infill rate and printing layer thickness significantly affected the results of the cone calorimeter tests. Increasing the infill rate extended the HRR curve of the specimens, as reflected in the results for the HRR time duration. As for the wood/PLA specimens, the maximum HRR decreased with a higher infill rate, while it increased for the PLA specimens (
The TSI data were also recorded. The TSI from the start of the testing time of 3D-printed neat PLA and wood/PLA composite specimens is presented in
In general, the ASEA of the specimens decreased with increasing infill rate in the wood/PLA specimens and increased in the PLA specimens. The smoke is the major fire hazard, which occurs due to the incomplete combustion. The specific extinction area is based on the smoke obscuration where the decrease in light transmission is measured by a laser beam through the exhaust duct [
The highest total HRR was found in the specimens produced with a 0.09 mm layer thickness and 80% infill rate, while the lowest total HRR in the first 600 s of testing time (THR600s) was found in the specimens produced with a 0.29 mm layer thickness and 40% infill rate (
Two simulated parameters were observed for comparison with the intermediate reaction to fire test (EN 13823). The simulation was carried out with the Development tool for Euroclasses according to EN 13501-1-ConeTools. Simulated classification parameter FIGRA values of the neat PLA and wood/PLA specimens are presented in
The edge and surface small flame exposure results (ISO 11925-2) for the vertically oriented test specimens of 90 mm by 200 mm, 10 mm thick are given in
Filament | Specimen code | Infill rate (wt%) (%) | Printing layer thickness (mm) | TSI (s) | Flame | Droplets | ||||
---|---|---|---|---|---|---|---|---|---|---|
Height (cm) | End of flaming (s) | Flaming | Not flaming | None | ||||||
PLA | A | 40 | 0.29 | 2 | 8 | >60 | x | |||
PLA | I | 80 | 0.09 | 4 | 2 | >60 | x | |||
Wood/PLA | A | 40 | 0.29 | 1 | 20 | >60 | x | |||
Wood/PLA | I | 80 | 0.09 | 2 | 12 | >60 | x |
Filament | Specimen code | Infill rate (wt%) (%) | Printing layer thickness (mm) | TSI/s | Flame | Droplets | ||||
---|---|---|---|---|---|---|---|---|---|---|
Height/cm | End of flaming (s) | Flaming | Not flaming | None | ||||||
PLA | A | 40 | 0.29 | 13 | 4 | 30 | x | |||
PLA | I | 80 | 0.09 | 14 | 3 | 31 | x | |||
Wood/PLA | A | 40 | 0.29 | 10 | 13 | >60 | x | |||
Wood/PLA | I | 80 | 0.09 | 13 | 3 | 30 | x |
The cone calorimeter results gave the detailed information about the fire behaviour of 3D-printed wood/PLA specimens produced with different printing setting parameters such as infill rate and printing layer thickness. According to the test results, printing layer thickness and infill rate significantly affected the TSI and HRR of the specimens through the quantity of material. In both wood/PLA and PLA specimens the time during which the HRR was above 50% of its peak value increased with the higher initial mass of the specimens. This time was significantly longer in wood/PLA specimens compared to PLA specimens. However, the peak value of HRR was notably higher in PLA specimens compared to wood/PLA specimens. According to the results, it can be said that the mass of the specimens is the most influential factor with regard to their HRR. Time to ignition was found to be much longer in the PLA specimens than the wood/PLA specimens. With the small flame test it was observed that flaming droplets occurred when burning the vertically oriented PLA specimen. On the other hand, the same experiment with wood/PLA did not show any dripping. Contact with the pilot flame during the small flame test caused a high flame in the wood/PLA material, which for the lightest wood/PLA specimen exceeded the allowed maximum value for the E classification according to EN 13501-1. In conclusion, the density of printing (i.e., the amount of material used per unit volume) affects the burning properties of both PLA and wood/PLA specimens in the same way—at small flame exposure (test ISO 11925-2) specimens with higher density ignite later and burn with a lower flame during the testing time. On the other hand, due to the larger amount of material the heat release rate of specimens with a higher density is greater in the cone calorimeter test (ISO 5660-1).