Three-dimensional printing with nano-enabled filaments releases polymer particles containing carbon nanotubes into air

Abstract

Fused deposition modeling (FDM™) 3-dimensional printing uses polymer filament to build objects. Some polymer filaments are formulated with additives, though it is unknown if they are released during printing. Three commercially available filaments that contained carbon nanotubes (CNTs) were printed with a desktop FDM™ 3-D printer in a chamber while monitoring total particle number concentration and size distribution. Airborne particles were collected on filters and analyzed using electron microscopy. Carbonyl compounds were identified by mass spectrometry. The elemental carbon content of the bulk CNT-containing filaments was 1.5 to 5.2 wt%. CNT-containing filaments released up to 10¹⁰ ultrafine (d < 100 nm) particles/g printed and 10⁶ to 10⁸ respirable (d ~0.5 to 2 μm) particles/g printed. From microscopy, 1% of the emitted respirable polymer particles contained visible CNTs. Carbonyl emissions were observed above the limit of detection (LOD) but were below the limit of quantitation (LOQ). Modeling indicated that, for all filaments, the average proportional lung deposition of CNT-containing polymer particles was 6.5%, 5.7%, and 7.2% for the head airways, tracheobronchiolar, and pulmonary regions, respectively. If CNT-containing polymer particles are hazardous, it would be prudent to control emissions during use of these filaments.

Keywords: 3-D printing; carbon nanotubes; elemental carbon; emission rate; lung deposition modeling; polymer.

FIGURE 1

Scanning electron micrographs of surfaces of commercially available unused filaments with and without CNTs: (A) ABS_CNT, (B) ABS, (C) PLA_CNT, (D) PLA, (E) PC_CNT, and (F) PC. Note that scale bars differ among images

FIGURE 2

Transmission electron micrographs of cross-sections of commercially available filaments labeled for sale with or without CNTs: (A) ABS_CNT, (B) ABS, (C) PLA_CNT, (D) PLA, (E) PC_CNT, and (F) PC. Note that scale bar differs among images

FIGURE 3

Particle emission yields by filament type: (A) number from condensation nuclei counter data (20 nm to 1 μm), (B) number from all FMPS size channels (5.6 to 560 nm), and (C) number from all APS size channels (0.5 to 20 μm). The lower boundary of a box is the 25^th percentile, the line within a box is the median, and the upper boundary of a box is the 75^th percentile. Whiskers (error bars) below and above a box indicate the 10^th and 90^th percentiles. Horizontal square bracket = statistical difference (P < 0.05). Note the break in the y-axis scale in each panel

FIGURE 4

Particle emission rates by filament type: (A) number from condensation nuclei counter data (20 nm to 1 μm), (B) number from all FMPS size channels (5.6 to 560 nm), and (C) number from all APS size channels (0.5 to 20 μm). The lower boundary of a box is the 25^th percentile, the line within a box is the median, and the upper boundary of a box is the 75^th percentile. Whiskers (error bars) below and above a box indicate the 10^th and 90^th percentiles. Horizontal bracket = statistical difference (P < 0.05). Note the break in the y-axis scale in each panel

FIGURE 5

Peak particle number concentration (N_peak) by filament type: (A) number from condensation nuclei counter data (20 nm to 1 μm), (B) number from all FMPS size channels (5.6 to 560 nm), and (C) number from all APS size channels (0.5 to 20 μm). The lower boundary of a box is the 25^th percentile, the line within a box is the median, and the upper boundary of a box is the 75^th percentile. Whiskers (error bars) below and above a box indicate the 10^th and 90^th percentiles. Horizontal bracket = statistical difference (P < 0.05). Note the break in the y-axis scale in each panel

FIGURE 6

Scanning electron micrographs of aerosol particles released during FDM 3-D printing using commercially available filaments with and without CNTs. Printing with nano-enabled filaments released particles that contained CNTs (indicated by arrows), but printing with base polymer filaments did not: (A) ABS_CNT, (B) ABS, (C) PLA_CNT, (D) PLA, (E) PC_CNT, and (F) PC. Note that scale bars differ among images

FIGURE 7

Scanning electron micrographs of surfaces of printed objects. Objects printed with nano-enabled filaments had CNTs visible on surfaces, but objects printed with base polymer filaments did not: (A) ABS_CNT, (B) ABS, (C) PLA_CNT, (D) PLA, (E) PC_CNT, and (F) PC. Note that scale bars differ among images