Occupational exposure and indoor environmental quality evaluation from operating multiple desktop 3D printers in an office environment within a library.

Jay L. Peterson¹, Kevin Dyrdahl², and Stacy Bruss³

¹Office of Safety, Health, and Environment (OSHE), Gaithersburg Safety, Health and Environment Division (GSHED), National Institute of Standards and Technology (NIST); email: [email protected]

²Office of Safety, Health, and Environment (OSHE), Gaithersburg Safety, Health and Environment Division (GSHED), National Institute of Standards and Technology (NIST); email: [email protected]

³Earth System Research Laboratory (ESRL), Boulder Labs Library, National Oceanic and Atmospheric Administration (NOAA); ORCID: 0000-0002-4652-8519

ABSTRACT

Based on previous air quality studies focused on environmental test chambers or the characterization of filament emissions, desktop 3D printers have been shown to generate particle emissions and volatile organic compounds during printing operations.  Due to their continued increase in popularity and use, the printers may be operated in varied settings, including libraries, classrooms, and homes.  Additionally, they may be placed around varied workplace personnel and general population groups, including office staff employees or school-age children.  In this study, air quality and noise levels were analyzed during the simultaneous operation of three desktop 3D printers in an office environment, within a library, in order to evaluate for occupational exposure and also provide for a more comprehensive indoor environmental quality assessment.

            Particle emissions and volatile organic compounds were detectable above background levels throughout the duration of the printing.  Particle emissions, characterized as the primary emission type, were generally stabilized at 30,000 to 60,000 particles per cubic centimeter above background levels during the majority of printing operations.  General room ventilation, including a 9.45 air changes per hour rate, reduced the particle concentration levels by one half after 12 min to 13 min; and, to near background levels within 18 min to 20 min.  Average sound pressure levels were 12 dB above background readings while all three printers were in operation.  While it is unlikely that an occupational exposure limit would be exceeded, particle emissions, volatile organic compounds, and sound pressure levels were detectable above background levels throughout the printing operation.  Additional considerations related to exposure criteria and ventilation may be needed based on the location and operation of a desktop 3D printer in proximity to other persons (non-employees) and personnel who are not accustomed to changes in indoor environmental quality parameters.

INTRODUCTION

Desktop three-dimensional printers (3D printers) are becoming commonplace in many industrial, commercial, and residential settings. It is estimated that around 278,000 desktop 3D printers (under $5,000) were sold in 2015, and around 528,000 were sold in 2017, close to doubling the number in 2 years.[1] 3D printers may be operated in many different settings, including laboratories, libraries, classrooms, offices, and homes. Some have estimated that there may be more than 800 3D printers in libraries across the world.[2] Additionally, the operations are likely to occur near a variety of workplace personnel types, including facility employees, researchers, and office staff as well as sectors of the general public, including school-age children. Because of the materials (and filament emissions) involved in the printing process, there is a potential for personnel in the proximity of a 3D printer to be affected during printing operations by conditions from several indoor environmental quality (IEQ) parameters, including indoor air quality (IAQ), ventilation, and noise.

In reviewing literature on this topic, similar studies and evaluations have focused on 3D printer emissions in environmental test chambers, single 3D printers, filament types, and theoretical calculations. Some studies have evaluated single 3D printer emissions in ventilated and unventilated rooms.[3] Additionally, other studies have evaluated 3D printing operations and emissions utilizing different filament types, including acrylonitrile butadiene styrene (ABS) and polylactic acid (PLA).[4] Others have evaluated volatile organic compound (VOC) emission components from ABS filament. [5][6]

Moreover, there have been some studies focusing on occupational (employee-related) exposures related to emissions from 3D printers in the workplace, with a focus on overall worker safety and health. One evaluation was performed at a facility that manufactures products using 3D printing.[7] Research is also continuing on emissions from 3D printers in various occupational settings and scenarios, and installing and evaluating potential safety engineering controls (e.g., filtration devices) to reduce any potential emissions from operating 3D printers.[8]

The purpose of this study was to evaluate for occupational exposure and also expand on and advance the current knowledge of IEQ parameters and emissions from operating multiple desktop 3D printers in an office environment (a library) as a real-world example. Specifically, IEQ parameters were evaluated from operating three 3D printers (with ABS filament), simultaneously, in an office setting with known ventilation parameters. Emissions, from the perspective of both occupational exposure (worker/employee) and general IAQ parameters (general population/non-employee), were evaluated by utilizing multiple monitoring methods and equipment. Additionally, noise measurements were obtained to provide for a more comprehensive IEQ evaluation for an office environment.

METHODS

3D Printing Operations and Materials

During this evaluation, three commercially available desktop 3D printers were operated in an automated mode (open, no covers), both simultaneously and continuously, using a 3 mm black ABS filament. Based on the manufacturer’s printing profile guidance, the extruder temperature was set at 245°C and the print bed temperature was set at 110°C. The printing object, measuring 10 cm x 10 cm x 1 cm, was the National Institute of Standards and Technology (NIST) additive manufacturing test artifact.[9] The NIST artifact was used since it is designated as a standardized testing object in various manufacturing sectors.[5] The print time was 208 minutes. The IEQ parameter measurements, per datalogging, were taken during the duration of printing (208 minutes), and also included a time period both before printing (initial background levels and the printer warm-up stage) and after printing (return to background levels), for a total of 288 minutes of measurements. To further detail the measurements during the warm-upstage, as a relation to background levels and the actual start of the print, there was an intended manual pause on the printers (13 min) directly after the completion of the printer warm-up stage (12 min, no printing) before the initial printing started (not typical and is a distinctive element to this study). Because the room utilizes dilution ventilation, representative measurements were repeated on three days to evaluate for any significant variations in concentrations, with additional IEQ datalogging measurements obtained as background levels for 208 minutes (with no printers in operation) on another day. All instrumentation used in obtaining measurements were within the 1 year recommendation period for manufacturer calibration.

Location

Room

The room, measuring 5.85 m x 2.27 m x 3.18 m, for this evaluation is set-up exclusively for 3D printing operations, with no other printers or emission sources located in the room (Figure 1). The room consists of a carpeted floor, painted drywall walls extending from floor to ceiling, and a tiled drop ceiling. There is one door and no windows. There was no vacuuming or cleaning of the room between measurement days. During the evaluation, the room was unoccupied, and the door was closed—parameters intended to simulate the worst-case scenario of printing operations and concentrations that may occur.

Figure 1. Sketch of room used for the 3D printing operations. The microphone of the tripod-mounted sound level meter (SLM) was placed at the midpoint between printers #1 and #2, 60.96 cm (24 inches) from the front edge of printers, and 121.92 cm (48 inches) above the floor. The distance from the SLM to printer #3 was 121.92 cm (48 inches).

IAQ measurements were taken on a tripod at a midpoint between printers #1 and #2, 91.44 cm (36 inches) from the front edge of printers, and 121.92 cm (48 inches) above the floor. The distance from the tripod to printer #3 was 121.92 cm (48 inches).

Ventilation

In this room, general room ventilation (dilution) is the sole means used to minimize the buildup of emissions. There are two ducted, fresh-air supply vents located at the ceiling level. There are no ducted, return (exhaust) air vents in the room, and there is one non-ducted, passive return air grill/louver located along one wall. Additionally, there are no local exhaust ventilation (LEV) connections or specialized ventilation equipment (e.g., portable air filtration units) located in the room.

To determine an air changes per hour rate (AC/H), air flow measurements were obtained at the supply vents with a Balometer (Alnor Balometer ABT701, TSI Incorporated, Shoreview, MN). Additionally, a qualitative (visual) smoke tube test was used to evaluate differential air flow for the room.

Filament and Emissions (Air)

There are many filament types used in commercial desktop 3D printing operations, with ABS and PLA as two of the most commonly used (typically with filament diameters of 1.75 mm or 3 mm).[6] Emissions from ABS filament are expected to be higher than those from PLA based on chemical composition. As per the thermal processing of 3 mm ABS at the operational temperatures noted during this evaluation, the focus on the monitoring for this study was primarily related to the particulate (and particle) emissions, as expected based on the overall thermal processing of creating a new object from raw materials, and also the primary ABS chemical composition components: acrylonitrile, 1,3-butadiene, and styrene.[10] For the air emissions, all IAQ measurements were taken on a tripod (approximately the same height as the printers) at a midpoint between printers #1 and #2, 91.44 cm (36 inches) from the front edge of printers, and 121.92 cm (48 inches) above the floor. The distance from the tripod to printer #3 was 121.92 cm (48 inches).

Particulates (and Particles)

Particulates (and particles) were evaluated utilizing both a direct reading instrument and sampling cassettes (area samples). The direct reading instrument, a condensation particle counter (CPC) measuring particle concentrations (size range of 0.01 μm to > 1.0 μm), has a concentration range of 0 to 100,000 particles per cubic centimeter (cm³) (CPC 3007, TSI Incorporated, Shoreview, MN). The CPC, nonspecific to material type, was programmed to datalog for 1 minute, with measurements taken each second, and record the average of those 60 data points every 1 minute, for a total of 288 minutes, including a print time of 208 minutes. For the area samples, low-volume air pumps (SKC Aircheck TOUCH, SKC, Eighty Four, PA) were used in accordance with sampling method 0500, from the National Institute of Occupational Safety and Health (NIOSH), to collect two samples for particles not otherwise regulated (PNOR) analysis, as referenced in the NIOSH Manual of Analytical Methods (NMAM).[11]

Based on the filament quantities and active room ventilation, particulates, as a general mass-based material categorization (collected on a filter), including nuisance dusts and particles, were not expected to exceed an employee-based occupational exposure limit (OEL), as per OSHA or the American Conference of Governmental Industrial Hygienists (ACGIH). OSHA generally defines PNOR as total dust, including inert or nuisance dusts not specifically listed elsewhere.[12] Additionally, ACGIH has a category related to particles (insoluble or poorly soluble) not otherwise specified (PNOS), including inhalable, deposited anywhere in the respiratory track, and respirable, deposited in the gas-exchange region of the lungs.[13]

Particles, including ultrafine particles (UFP), which are generally considered to be less than 0.1 μm (100 nm), are an additional category that may be referenced during dust/particulate evaluations.[14] Particles, including the UFP category, provide for additional input beyond evaluating for an OEL, and UFPs would be detected by the CPC during this evaluation. Because of the active room ventilation and use of a programmed print time, datalogging parameters included overall particle concentrations, peaks in concentration, and time required to return to background levels.

Volatile Organic Compounds (VOCs) & Chemical-Specific Analytes

VOCs were evaluated utilizing both a direct-reading instrument and sorbent tubes (area samples). The direct reading instrument, a photoionization detector (PID) measuring VOC concentrations with a 10.6 eV lamp, has a concentration range of 0.00229 mg/m³ to 22,947 mg/m³ (1 part per billion (ppb) to 10,000 ppm) (ppbRAE 3000, RAE Systems, San Jose, CA). The 10.6 eV lamp will not detect acrylonitrile, but will detect 1,3-butadiene and styrene. The PID was programmed to datalog for 1 min, taking a reading every 1 second, and record the average of 60 measurements every 1 minute, for a total of 288 minutes, including a print time of 208 minutes.

For the VOC area samples, low-volume air pumps (SKC Aircheck TOUCH, SKC, Eighty Four, PA) were used in accordance with NIOSH method 1500 to collect two samples for hydrocarbons.[15] For the chemical-specific analytes, two samples were collected for the three separate analytes (acrylonitrile, 1,3-butadiene, and styrene) using the low-volume air pumps, as per sampling method 37 from OSHA for acrylonitrile; and, OSHA sampling methods 56 and 86 for 1,3-butadiene and styrene, respectively.[12]

Noise

As an additional IEQ parameter, decibel (dBA) levels, as per sound pressure level (SPL) measurements, were evaluated using a sound level meter (SLM) (SoundTrack LxT1, Larson Davis, Depew, NY). The SLM, programmed on the A-weighted scale, was used to measure: average levels or sound level equivalent (LAeq), minimum sound level (LASmin), and maximum sound level (LAmax).[16] To capture variations in sound levels, the SLM was programmed to record every 1 second for the total time of each sampling run.

To evaluate for occupational exposure and to approximate the potential sound level exposure of an operator, the SLM was placed in what would be the hearing zone (HZ) of an operator. Specifically, the microphone of the tripod-mounted SLM was placed at the midpoint between printers #1 and #2, 60.96 cm (24 inches) from the front edge of printers, and 121.92 cm (48 inches) above the floor. The distance from the SLM to printer #3 was 121.92 cm (48 inches).

Sound level measurements were obtained while all three printers were in operation in order to capture a worst-case exposure scenario, along with variations in sound levels associated with the operational phases of the printers. Initial, short-term background measurements were obtained immediately prior to the warm-up stage. Background sound pressure levels were also measured for 205 min and 235 min when no printers were operating.

IEQ Datalogging and Area Samples

Overall, three separate sets of IEQ datalogging measurements, with all three printers in simultaneous operation, were collected on three separate days, excluding VOC measurements, of which two separate sets were collected on two separate days because of limitations with the PID (will not detect acrylonitrile). A set of background measurements (no printers in operation) was collected on a separate day.

The analysis for all the area samples was performed by an American Industrial Hygiene Association (AIHA) accredited laboratory (Analytics Corporation, Ashland, VA). Because the area sample results were expected to be below the laboratory reporting limits, only one set of area samples were collected.

RESULTS

The generalizability of results from this study are limited by the overall study design and parameters specific to this room, including the use of dilution ventilation to minimize the buildup of air emissions. Figures noted for the particle concentration, VOC concentration, and sound pressure levels are from a representative day of datalogging.

Ventilation

The AC/H for the room was determined to be 9.45, with a supply air flow of 7.14 m³/min (252 cubic feet per minute (CFM)) and a room volume of 45 m³ (1,600 cubic feet). The differential air flow for the room was positive (smoke moving out towards the adjacent corridor).

Particulates (and Particles)

The particle concentrations from the CPC measurements, displaying 1-min averaging measurements (Figure 2), indicate the particle levels were above background levels throughout the entire printing operation. The highest levels occurred during both the initial 12 min warm-up stage (no printing) and then the initial start of the printing (which were separated by the 13 min intended manual pause). The levels observed during the warm-up stage are likely due to residual filament on the print bed and/or extruder. The printing process lasted 208 minutes, and the datalogging was conducted for 288 minutes, which included 10 minutes before the start of the printing and continued for 44 minutes once the printing stopped. Once printing started the levels then peaked at 154,737 particles per cm³ and then steadily decreased until stabilizing around 30,000 to 60,000 particles per cm³ for the majority of the printing time. Once the printing stopped, the particle levels were reduced by approximately one half after 12 to 13 min, and to near background levels within 18 to 20 min.

The initial particle concentration levels (over 100,000 particles per cm³) were comparable for the three separate particle datalogging sets. In referencing a consistent data point for error analysis, the particle concentration at the start of the printing was compared to the concentration after 60 min of printing (steadily decreasing and stabilizing). The concentration at the start of the printing (and concentration after 60 minutes of printing) was 154,000 particles per cm³ (53,000 particles per cm³) for the run shown in Figure 2; and 138,000 particles per cm³ (22,100 particles per cm³) and 119,000 particles per cm³ (21,300 particles per cm³), respectively, for the other two data sets (Table 1). Additionally, the mean particle level concentration from background level monitoring (208 minutes), conducted during a separate day with no printing activity, was 1,380 particles per cm³ with a standard deviation of 300 particles per cm³.

For the area samples (sampling cassettes), results for PNOR (Table 2) were below the detection limit, as reported by the analytical laboratory. Results for the area samples were not evaluated further.

Figure 2. Particle concentrations, based on 1-minute averaging, for 288 minutes (print time 208 minutes, as indicated by the shaded area between minutes 36 and 244). Datalogging included initial background level sampling from minutes 1-10, followed by a printer warm-up stage (no printing) from minutes 10-22, followed by a manual pause after the warm up stage from minutes 22-36 before print start.

Table 1. CPC measurements on three separate days, including the initial particle concentration at the start of the printing and then after 60 minutes of printing.

Table 2. Results for the sampling cassettes (area samples for PNOR) were below the reporting limit (<) for analysis.

VOCs & Chemical-Specific Analytes

The VOC levels, measured with the PID (datalogging) indicated an overall increase above background levels throughout the printing operations but generally stabilized and consistent throughout the entire monitored period of 288 minutes, including the print time of 208 minutes (Figure 3). Once printing began, the VOC levels increased approximately 0.46 to 1.15 mg/m³ (200 ppb to 500 ppb) in periodic time intervals throughout, compared to the background level monitoring, which was conducted on a separate day for 208 minutes, during which no printers were operating. Summarized results from the monitoring included minimum, maximum, and average mg/m³ levels as 0.25, 2.5, and 1.53 (ppb levels as 109, 1089, and 667), respectively, for Figure 3; and, 1.03, 1.58, and 1.44 (ppb levels as 447, 690, and 626), respectively, during a second set of monitoring data, with a standard deviation of 0.047 (20.5 ppb) on the averages of the two data sets (Table 3). For the background level monitoring conducted on a separate day, the minimum, maximum, and average mg/m³ levels were 0.1, 0.69, and 0.49 (ppb levels as 44, 302, and 212) respectively, with a standard deviation of 0.08 (35.06 ppb).

Limitations with the PID measurements and overall error analysis should be noted. Namely, the PID will not detect acrylonitrile, one of the ABS components. Additionally, there is an initial increase in VOC levels prior to the printing start time (printer warm-up/preparation), which was noted during both data sets, and also during the background level monitoring with no printers in operation or identified VOC sources. Periodic, brief fluctuations were noted, including during the background level monitoring. As such, extrapolating VOC data with representative results within the ppb range (for comparison, occupational exposure limits are generally listed on a higher scale in the ppm range, including 85.19 mg/m³ (20 ppm) as an 8-hour limit for styrene, as an example) and correlating scale, as comparatively to the overall quantity of ABS filament used during printing operations, should be evaluated further. From this evaluation, the results, at a minimum, indicate VOC levels were above background levels during printing operations; however, additional analysis is warranted to (1) further quantify correlations, within the ppb range, to specific time periods and printing activities, and (2) provide for more input in error analysis.

For the area samples (sorbent tubes), results were below the detection limit, as noted by the reporting limit (<) for analysis noted by the analytical laboratory, for both the VOCs (hydrocarbons) and individual chemicals (Table 4). Results for the area samples were not evaluated further.

Figure 3. VOC concentrations, based on 1-minute averaging, for 288 minutes (print time 208 minutes, as indicated by the shaded area between minute 36 and 244). Datalogging included initial background level sampling from minutes 1-10, followed by a printer warm-up stage (no printing) from minutes 10-22, followed by a manual pause from minutes 22-36.

Table 3. PID measurements on two separate days.

Table 4. Results for the sorbent tubes (area samples for VOCs and chemical specific analytes) were below the reporting limit (<) for analysis.

Noise

Sound pressure levels ranged from 53.8 dBA (LASmin) to 63.2 dBA (LASmax), respectively, with an overall average of 57.4 dBA (LASeq) when all 3 printers were operating. Datalogging collected 17,300 individual measurements over the entire sampling period of 288 minutes (Figure 4) that included: 6 minutes with printers off, 10 minutes during the printer warm-up stage, a pause in standby mode for 13 minutes, the entire printing operation of 208 minutes, and 44 minutes after the printing had stopped.

Background sound pressure levels in the room ranged from 44.3 dBA (LASmin) to 53.5 dBA (LASmax), respectively, with an overall average of 45.7 dBA (LASeq) with a standard deviation of 0.42 dBA, while no printers were in operation for 209 minutes (12,400 datalogging measurements, per every 1 second). Both operational and background sound pressure levels are summarized in Table 5 and Figure 4, with instrumentation error listed as +/- 0.75 dBA for all values.

Figure 4. Sound pressure level equivalent, based on 1-second averaging, for 288 minutes (print time 208 minutes, as indicated by the shaded area between minute 36 and 244). Datalogging included initial background level sampling from minutes 1-10, followed by a printer warm-up stage (no printing) from minutes 10-22, followed by a manual pause from minutes 22-36 minutes.

Table 5. SLM measurements with all 3 printers in operation, averaged over 2 seconds, compared to background sound levels.

DISCUSSION

Ventilation and Emissions

The general ventilation in a room, when a LEV or other specialized ventilation equipment is not used, will have an impact on reducing emissions related to particles and/or chemicals. During this evaluation, a 9.45 AC/H rate correlated to the particle levels stabilizing during the printing operations and then lowering to near background levels with 18 to 20 minutes once printing stopped. The 9.45 AC/H rate is generally considered higher than is typically found in an office type setting (non-laboratory), where some guidance recommends a minimum of 6 AC/H for a laboratory.[17] With any reduction in AC/H, there is likely to be (1) a higher concentration of emissions in the room during the printing, and (2) an increase in the time for the concentration to return to background levels. While this room was supplied with fresh air (at the supply vents), a room with only recirculating air (common in offices and many homes) would likely have exponentially increased any potential emissions (and remain in the room for extended periods).

In accordance with the ventilation parameters, there are other variables that would impact emissions, which may include: physical location of the 3D printers, occupants in the room, supply and exhaust vent locations, filament type, residual filament, printer enclosures, proximity to the printer, printing times, complexity of the printed object(s), and other emission sources in the room. The use of other chemicals (in proximity to the printers) should also be reviewed, as a synergistic effect may need to be evaluated, including any adjustments to the OEL, as applicable. There are also other operational considerations on potential emission variables. Though the operating temperatures were not evaluated, comparatively, during this evaluation, as 3D printing technologies continue to evolve, altered and/or increased processing temperatures may generate, at a minimum, a review of additional emission types. In a prior study related to ABS and combustion, twenty-seven chemical compounds were identified as combustion products, including hydrogen cyanide (HCN) and carbon monoxide (CO).[18] Additionally, emissions related to UFPs, and comparatively to nanoparticles, have shown to increase with bed temperatures.[5]

Exposure Criteria - Limitations of Applying an OEL

For this evaluation, the personnel typically operating the 3D printers are primarily office employees, which would entail applying the applicable workplace OEL, including OSHA permissible exposure limits (PELs), for occupational exposure, at a minimum. The exposure criteria would also include referencing OELs for the typical workday (8 hours). However, because of the increasing prevalence of 3D printers in various settings (libraries, classrooms, etc.) and various persons (school-age children, guests, etc.) and workplace personnel (office, laboratory, etc.) involved, an additional review and interpretation of applicable OELs, and other potential health hazards may be needed depending on the intended use and location of each 3D printer. Also, additional considerations may be needed if sensitive populations (e.g., asthmatic) are present.

Applicable chemical based OELs commonly considered during workplace evaluations, including this evaluation, are the PELs from OSHA and the Threshold Limit Values (TLVs) from ACGIH. NIOSH has also developed recommended exposure limits, or RELs, which should be reviewed and considered. For chemicals, OELs are typically set as an 8-hour time-weighted average (TWA), Ceiling (C) limit, or Short Term Exposure Limit (STEL), in parts per million (ppm) or milligrams per meter cubed (mg/m³). As a comparison, for styrene, the 8-hour TWA from OSHA is 425.97 mg/m³ (100 ppm); and, from ACGIH is 85.19 mg/m³ (20 ppm).[19][13] The styrene results (area samples) from this evaluation were less than the detection limit (< 0.81 mg/m³ (< 0.189 ppm)); however, the PID VOC levels (which would include styrene as one of the combined VOC analytes), measured in ppb, were detected above room background levels during the printing operation.

As ACGIH explains in abbreviated terms, TLVs-TWAs refer to airborne concentrations under which it is believed that nearly all workers may be repeatedly exposed, day after day, over a working lifetime, without adverse health effects’, and these limits are not expected to adequately protect all workers.[13] For example, workers who have chemical allergies, sensitivities, or unique health conditions, etc. may not be protected, and other factors must be considered. Because of the varied 3D printer locations and difference in personnel involved, OELs may be one of many considerations when determining the exposure criteria; other non-occupational based health hazard information may also need to be considered. For example, as noted by the International Agency for Research on Cancer (IARC), acrylonitrile is classified as Group 2B (possibly carcinogenic to humans); and, 1, 3-butadiene is classified as Group 1 (confirmed human carcinogen).[20]

Further, as more 3D printers are placed in various settings (classrooms, offices), there is a potential for the printers to be located around sensitive populations (e.g., asthmatic), which may include school age children or office employees, or others not typically acclimated to emissions or odors. Chemical odors and irritation may also be reported, even if OELs are not expected to be approached or exceeded. The reported odor threshold range for acrylonitrile is 3.47 to 47.74 mg/m³ (1.6 to 22 ppm), with an odor characteristic of onion, garlic; 1,3-butadiene is 0.22 to 168.13 mg/m³ (0.099 to 76 ppm) with aromatic, rubber; and, for styrene is 0.01193 to 259.84 mg/m³ (0.0028 to 61 ppm) with sharp, sweet.[21] Additionally, styrene can be irritating to the eyes and skin.[22]

For noise OELs, ACGIH explains that TLVs refer to sound pressure levels and durations of exposure that represent conditions under which it is believed that nearly all workers may be repeatedly exposed without adverse effect on their ability to hear and understand normal speech; and, it should be recognized that the TLVs for noise will not protect all workers from the adverse effects of noise exposure.[13] For 8 hours, the ACGIH TLV is 85 decibels (dBA), as measured by a sound level meter set to the A-weighted scale with slow meter response; and, the OSHA PEL is 90 dBA.[13][23] For this evaluation, all SPL measurements were below the OSHA PEL and ACGIH TLV; however, similar to the chemical based OELs, other personnel and persons (e.g., sensitive populations) may be present. For example, placing multiple 3D printers in a typically quiet office, or educational environment, can raise the ambient noise levels by 12 dBA, which may impact personnel in performing of certain types of tasks (e.g., concentrating on detailed tasks and/or basic communication).

Particulates (and Particles)

As a general guideline related to IAQ and occupant comfort, particle concentration levels in an indoor room (commercial), with no occupants or active sources (equipment) of particle generation, would be expected to be consistently comparable, and maintainable, to background particle levels of the room. For 3D printing processes, elevated particle concentration levels, above the background levels of a room, would be indicative of active printing. For this particular room and set-up, the AC/H rate of 9.45 (two air supply vents located at the ceiling level; and, one passive, non-ducted return air louver located along a wall) reduced the particle concentration levels to near background within 18 to 20 minutes once printing had stopped.

As a basic parameter of standard air handling unit (AHU) operation and filtration of outdoor air at the air intake locations in a commercial building: indoor particle levels should be lower than outdoors. An increase in indoor particle concentration (compared to outdoors) is likely related to an indoor activity, which would include 3D printing operations. However, as compared to outdoor particles, the health effects of indoor exposure to particles are less well-understood as compared to outdoors.[24] During this evaluation, once stabilized, particles were generally 30,000 to 60,000 particles per cm³ above background levels during the printing operations. Indoor particles and UFP related health concerns continue to be identified and researched, both in an occupational setting and also the outdoor environment. In general, research on UFPs is showing an association with cardiovascular and respiratory health in humans.[24][25] From a perspective of an OEL, UFP related limits are still under review and discussion. Whereas weighing is used to assess the mass concentrations of respirable and inhalable particles, particle counts may be used when evaluating UFP exposure; whether a conversion between these different metrics is feasible for setting OELs has to be explored.[26]

CONCLUSIONS

Simultaneously operating three desktop 3D printers (with ABS filament) in an indoor office environment will generate particle-based and VOC emissions above background levels. A ventilation rate of 9.45 AC/H stabilized the particle concentration to generally 30,000 to 60,000 particles per cm³ above background levels during the majority of printing operations and returned the levels to near background within 18 to 20 minutes from when printing stopped. The noise levels increased by approximately 12 dBA, compared to background levels, when three printers were operating. Overall, occupational exposure, from an OEL and workplace/employee perspective, are not expected to be exceeded; however, emissions and sound pressure levels were detectable above background levels throughout the printing operations.

RECOMMENDATIONS

Ventilation and noise should be considered when placing a desktop 3D printer in an indoor environment. To minimize the potential for the build-up of particle and VOC emissions from operating a desktop 3D printer, the ventilation parameters around the printer location should be evaluated. Additionally, because of the variations in locations (offices, classrooms, libraries, etc.) and persons/personnel involved (employees, school age children, office staff) with desktop 3D printers, incorporating exposure criteria beyond the workplace/employee related OELs may be needed, especially as the overall use of 3D printers continues to increase.

ACKNOWLEDGEMENTS

Thank you to the NIST Research Library staff, including Mr. Keith Martin and Ms. Sarah Reeves, for their assistance and support. Additional thank you to the NIST Office of Safety, Health, and Environment (OSHE) staff for their support throughout.

Disclaimer: Certain commercial equipment, instruments, or materials are identified in this paper to foster understanding. Such identification does not imply recommendation or endorsement by the National Institute of Standards and Technology (NIST), nor does it imply that the materials or equipment identified are necessarily the best available for the purpose.

The opinions, recommendations, findings, and conclusions in this publication do not necessarily reflect the views or policies of NIST or the United States Government.