Particle and vapor emissions from vat polymerization desktop-scale 3-dimensional printers

Abstract

Little is known about emissions and exposure potential from vat polymerization additive manufacturing, a process that uses light-activated polymerization of a resin to build an object. Five vat polymerization printers (three stereolithography (SLA) and two digital light processing (DLP) were evaluated individually in a 12.85 m³ chamber. Aerosols (number, size) and total volatile organic compounds (TVOC) were measured using real-time monitors. Carbonyl vapors and particulate matter were collected for offline analysis using impingers and filters, respectively. During printing, particle emission yields (#/g printed) ranged from 1.3 ± 0.3 to 2.8 ± 2.6 x 10⁸ (SLA printers) and from 3.3 ± 1.5 to 9.2 ± 3.0 x 10⁸ (DLP printers). Yields for number of particles with sizes 5.6 to 560 nm (#/g printed) were 0.8 ± 0.1 to 2.1 ± 0.9 x 10¹⁰ and from 1.1 ± 0.3 to 4.0 ± 1.2 x 10¹⁰ for SLA and DLP printers, respectively. TVOC yield values (µg/g printed) ranged from 161 ± 47 to 322 ± 229 (SLA printers) and from 1281 ± 313 to 1931 ± 234 (DLP printers). Geometric mean mobility particle sizes were 41.1-45.1 nm for SLA printers and 15.3-28.8 nm for DLP printers. Mean particle and TVOC yields were statistically significantly higher and mean particle sizes were significantly smaller for DLP printers compared with SLA printers (p < 0.05). Energy dispersive X-ray analysis of individual particles qualitatively identified potential occupational carcinogens (chromium, nickel) as well as reactive metals implicated in generation of reactive oxygen species (iron, zinc). Lung deposition modeling indicates that about 15-37% of emitted particles would deposit in the pulmonary region (alveoli). Benzaldehyde (1.0-2.3 ppb) and acetone (0.7-18.0 ppb) were quantified in emissions from four of the printers and 4-oxopentanal (0.07 ppb) was detectable in the emissions from one printer. Vat polymerization printers emitted nanoscale particles that contained potential carcinogens, sensitizers, and reactive metals as well as carbonyl compound vapors. Differences in emissions between SLA and DLP printers indicate that the underlying technology is an important factor when considering exposure reduction strategies such as engineering controls.

Keywords: 3-dimensional printing; digital light processing; stereolithography; ultrafine particles; vat polymerization; volatile organic compounds.

Figure 1.

Time series plots (in minutes) of: panels (a) and (c) the mean (μ_t) and panels (b) and (d) standard deviation (σ_t) of two 3-D printer emission tests. Panels (a) and (b) are results for the Pegasus SLA printer, and panels (c) and (d) are results for the M-One DLP printer. The gray area indicates the printing period. Black dots are mean values with Bayesian 95% credible intervals. All y-axes are log-scale.

Figure 2.

Morphology of exemplar particles emitted during operation of vat polymerization printers: (a) nanoscale-cluster particle observed in emissions from all printers; (b) aggregate particle emitted by the Form 1+ printer; (c) compact irregular particle emitted by the Pegasus Touch printer; (d) discrete spherical particle emitted by the Nobel 1.0A printer; (e) discrete spherical particle emitted by the Titan 1 printer; and (f) flake-like particle emitted by the M-One printer.

Figure 3.

Elemental spectra illustrating the composition of exemplar particles emitted during operation of vat polymerization printers (spectra correspond to the particles shown in Figure 2): (a) representative carbon/oxygen particle observed in emissions from all printers; (b) iron-containing particle emitted by the Form 1+ printer; (c) calcium- and sulfur-containing particle emitted by the Pegasus Touch printer; (d) iron- and nickel-containing particle emitted by the Nobel 1.0A printer; (e) iron-, phosphorous- and zinc-containing particle emitted by the Titan 1 printer; and (f) aluminum- and silicon-containing particle emitted by the M-One printer.