Current Strategies for Engineering Controls in Nanomaterial Production and Downstream Handling Processes/Nanotechnology Processes and Engineering Controls

by National Institute of Occupational Safety and Health

CHAPTER 3

Nanotechnology Processes and Engineering Controls

3.1 Primary Nanotechnology Production and Downstream Processes

Currently, nanomaterials are produced using a variety of methods that provide conditions for the formation of desired shapes, sizes, and chemical composition. These production processes can be separated into six categories [HSE 2004; NNI, no date]:

• Gas phase processes, including flame pyrolysis, high-temperature evaporation, and plasma synthesis. This process involves the growth of nanoparticles by homogenous nucleation of supersaturated vapor. Nanoparticles are formed in a reactor at high temperatures when source material in solid, liquid, or gaseous form is injected into the reactor. These precursors are supersaturated by expansion and cooled prior to the initiation of nucleated growth. The size and composition of the final materials depend on the materials used and process parameters.
• Chemical vapor deposition (CVD). This process has been used to deposit thin films of silicon on semiconductor wafers. The chemical vapor is formed in a reactor by pyrolysis, reduction, oxidation, and nitridation and deposited as a film with the nucleation of a few atoms that coalesce into a continuous film. This process has been used to produce many nanomaterials including TiO2, zinc oxide, silicon carbide, and, possibly most importantly, CNTs. The use of fluidized bed technology has been adopted as a way to prepare CNTs on a large scale at low cost [Wang et al. 2002]. This technology fluidizes CNT agglomerates and produces high yields necessary for larger-scale operations.
• Colloidal or liquid phase methods. Chemical reactions in solvents lead to the formation of colloids. Solutions of different ions are mixed to produce insoluble precipitates. This method is a fairly simple and inexpensive way to produce nanoparticles and is often used for the synthesis of metals (e.g., gold, silver). These nanomaterials may remain in liquid suspension or may be processed into dry powder materials often by spray drying and collection through filtration.
• Mechanical processes including grinding, milling, and alloying. These processes create nanomaterials by a “top-down” method that reduces the size of larger bulk materials through the application of energy to break materials into smaller and smaller particles. This technique has been referred to as nanosizing or ultrafine grinding.
• Atomic and molecular beam epitaxy. Atomic layer epitaxy is the process of depositing monolayers (i.e., layers one molecule thick) of alternating materials and is commonly used in semiconductor fabrication. Molecular beam epitaxy is another process for depositing highly controlled crystalline layers onto a substrate.
• Dip pen lithography. A “bottom-up” method is a production process that involves depositing a chemical on the surface of a substrate using the tip of an atomic force microscope (AFM). The AFM tips are coated with the chemical, which is directly deposited on a substrate in a specific pattern.

Downstream processes use engineered nanomaterials for product application and development. Examples of these tasks or operations include weighing, dispersion/sonication, mixing, compounding/extrusion, electro-spinning, packaging, and maintenance. These activities should be evaluated for potential sources of exposure.

3.2 Engineering Control Approaches to Reducing Exposures

Engineering controls are used to remove a hazard or place a barrier between the worker and the hazard, and though costs of engineering controls may be higher than that of administrative controls or PPE initially, over the long term, operating costs are often lower. A major advantage of engineering controls is that, when properly designed, they require little or no user effort or training to be effective. Many industries have implemented engineering controls to reduce exposure and risk of disease among their workers. The pharmaceutical industry uses hazardous (i.e., biologically active) liquids and powders that often do not have OELs. To address these hazards, the pharmaceutical industry has adopted a performance-based strategy using exposure control limits. This approach is based on establishing qualitative or semiquantitative criteria for assessing risk associated with the compounds and matching that information with known exposure-control approaches [Naumann et al. 1996].

Many of the processes used in pharmaceutical production are similar to those used in the nanoparticle industries discussed above and include blending, mixing, and handling of hazardous compounds in liquid and powder form. The general control concepts required for working with hazardous materials include specification of general ventilation, LEV, maintenance, cleaning and disposal, PPE, IH monitoring, and medical surveillance [Naumann et al. 1996]. Particular work practices, such as using HEPA-filtered vacuums instead of dry sweeping, are required. In addition, routine IH and medical monitoring ensure that work practices and engineering controls are effective.

Source containment is considered the highest level in the containment hierarchy and is used by the pharmaceutical industry [Brock 2009]. This category contains many options including elimination, substitution, product modifications, process modifications, and equipment modifications. These steps could include reworking the process to reduce the number of times material is transferred or keeping the product in solution to minimize aerosolization potential. The next level of control for capturing process emissions is the use of engineering controls such as glove boxes, downflow booths, and local exhaust ventilation.

Genaidy et al. [2009] conducted a detailed hazard analysis of a CNF manufacturing process and suggested the following potential sources of workplace exposure to nanomaterials:

• Leakage and spillage from reactors and powder processing equipment
• Manually harvesting product from reactors
• Discharging product into containers
• Transporting containers of intermediate products to the next process
• Charging the powders into processing equipment
• Weighing out powder for shipment
• Packaging material for shipment
• Storing material between operations
• Cleaning equipment to remove debris stuck to side walls
• Changing filters on dust collection systems and vacuum cleaners
• Further processing of products containing nanomaterials (e.g., cutting, grinding, drilling)

This detailed analysis, along with the review of exposure assessment studies in nanomaterial production and downstream user facilities described below, identify common processes that may lead to worker exposure to nanomaterials. This section provides some information on engineering control approaches that may be applicable for these common processes/tasks. Table 1 shows a generic process list along with applicable engineering controls and references. The engineering control column provides a framework for identifying exposure controls for particular processes. The third column shows the industry in which these control approaches have been tested. References are listed in the fourth column for studies that apply to each of these processes and controls.

Table 1. Engineering controls and associated tasks for various industries

Process/task	Engineering control	Industry	Reference
Reactor fugitive emissions	Enclosure	Nanotechnology	Tsai et al. 2009b Lee et al. 2011
Product harvesting	Glovebox	Nanotechnology	Yeganeh et al. 2008
Reactor cleaning	Spot LEV system/fume extractor	Nanotechnology	Methner 2008
Small-scale weighing	Chemical fume hood	Nanotechnology	Tsai et al. 2009a Ahn et al. 2008 Tsai et al. 2010
	Biological safety cabinet	Nanotechnology and laboratory	Cena and Peters 2011 Macher and First 1984
	Glovebox isolator	Pharmaceutical	Walker 2002 Hirst et al. 2002
	Nano fume hood	Pharmaceutical
	Air curtain isolation hood	Nanotechnology/research	Tsai et al. 2010
Product discharge/bag filling	Discharge/collar hood	Silica and pharmaceutical	ACGIH 2013 HSE 2003e Hirst et al. 2002
	Continuous liner	Pharmaceutical	Hirst et al. 2002
	Inflatable seal	Pharmaceutical	Hirst et al. 2002
Bag/container emptying	Bag dump station	Silica	HSE 2003d Heitbrink and McKinnery 1986 Cecala et al. 1988
Large-scale weighing/handling	Ventilated booth	Pharmaceutical	Hirst et al. 2002 Floura and Kremer 2008 HSE 2003b
Nanocomposite machining	High velocity-low volume	Woodworking
	Wet suppression	Nanotechnology	Bello et al. 2009
Air filter change-out	Bag in-bag out	Pharmaceutical

3.3 Ventilation and General Considerations

It is important to confirm that the LEV system is operating as designed by regularly measuring exhaust airflows. A standard measurement - hood static pressure - provides important information on the hood performance, because any change in airflow results in a change in hood static pressure. For hoods designed to prevent exposures to hazardous airborne contaminants, the ACGIH Industrial Ventilation: A Manual of Recommended Practice for Operation and Maintenance recommends the installation of a fixed hood static pressure gauge [ACGIH 2010].

In addition to routinely monitoring the hood static pressure, additional system checks should be completed periodically to ensure adequate system performance, including smoke tube testing, hood slot/face velocity measurements, and duct velocity measurements using an anemometer. A dry ice test is another method of evaluation designed to qualitatively determine the containment performance of fume hoods. These system evaluation tasks should become part of a routine preventative maintenance schedule to check system performance. It is important to note that the collection and release of air contaminants may be regulated; companies should contact agencies responsible for local air pollution control to ensure compliance with emissions requirements when implementing new or revised engineering controls. To reduce the risk of exposure to nanomaterials, a few standard precautions should be followed in areas where exposures may occur:

• Isolate rooms where nanomaterials are handled from the rest of the plant with walls, doors, or other barriers.

• Maintain production areas where nanomaterials are being produced or handled under negative air pressure relative to the rest of the plant.

• Install hood static pressure gauges (manometers) near hoods to provide a way to verify proper hood performance.

• When possible, place hoods away from doors, windows, air supply registers, and aisles to reduce the impact of cross drafts.

• Provide supply air to production rooms to replace most of the exhausted air.

• Direct exhaust air discharge stacks away from air intakes, doors, and windows. Consider environmental conditions, especially prevailing winds.

3.4 Exposure Control Technologies for Common Processes

In a review of exposure assessments conducted at nanotechnology plants and laboratories, Brouwer [2010] determined that activities that resulted in exposures included harvesting (e.g., scraping materials out of reactors), bagging, packaging, and reactor cleaning. Downstream activities that may release nanomaterials include bag dumping, manual transfer between processes, mixing or compounding, powder sifting, and machining of parts that contain nanomaterials. Particle concentrations during production activities ranged from about 103–105 particles/cm3. Most studies showed bimodal particle distributions with modes of about 200–400 nm and 1,000–20,000 nm, indicating that the emissions are dominated by aggregates and agglomerates. With the exception of leakage from reactors when primary manufactured nanoparticles may be released, workers are believed to be primarily exposed to agglomerates and aggregates.

Methner et al. [2010] summarized the findings of exposure assessments conducted in 12 facilities with a variety of operations: seven were R&D labs, one produced CNTs, one produced nanoscale TiO2, one produced nanoscale metals and metal oxides, one produced silica-iron nanomaterials, and one manufactured nylon nanofibers. The most common processes observed at these facilities were weighing, mixing, collecting product, manual transfer of product, cleaning operations, drying, spraying, chopping, and sonicating.

Engineering controls used included portable vacuums with filters, laboratory fume hoods, portable LEV systems, ventilated walk-in enclosures, negative pressure rooms, and glove boxes. Tasks such as weighing, sonicating, and cleaning reactors showed evidence of nanomaterial emissions. The highest nanoparticle exposures measured occurred inside spray booth-type enclosures and during a spray dryer collection drum change-out. Other activities that resulted in higher exposures include reactor cleanout tasks (e.g., brushing and scraping slag material). Incidental (nonprocess) ultrafines were measured from a variety of sources, including electric arc welding, operating a propane-powered forklift, and the exhaust of a portable vacuum outfitted with filters.

From a review of published studies, some common sources of nanoparticles and fine particles can be identified. As expected, those processes that require material handling resulted in worker exposure to nanomaterials. Other activities that require operator interface with the reactor can result in nanoparticle exposure, and background concentrations may increase as a result of leakage from reactors under positive pressure. In addition, several studies found that evaluation of process emissions and exposure should take into account major sources of incidental nanoparticles that may be present in the workplace and also sources of natural nanoparticles, e.g., tree pollen brought into the work area through the facility HVAC system. Common incidental sources include diesel exhausts in outdoor air, welding fumes, forklifts, and gasfired heaters. Several studies showed that the use of engineering controls can reduce operator exposure, while one study showed that a poorly designed enclosure actually increased exposure [Cena and Peters 2011; Methner et al. 2007; Tsai et al. 2009a, 2010; Yeganeh et al. 2008].

The following sections describe applicable engineering controls for common processes used by nanotechnology companies described in the literature. For each control, a background is given along with a summary of relevant research conducted on their performance. Many of the control concepts discussed in this section come from the HSE Control Guidance Sheets in COSHH Essentials: Easy Steps to Control Chemicals [HSE 2003a,b,c,d] and the ACGIH Industrial Ventilation Manual [ACGIH 2013]. Table 2 lists common processes and tasks, along with potential emission points and the section or figure(s) that address those processes.

Table 2. Process/tasks and emission

Process/task	Potential emission/exposure points	See section	See figures
Production of bulk nanomaterials	Reactor fugitive emissions Product harvesting Reactor cleaning	3.4.1 3.4.1 3.4.1	7, 8 12
Downstream processing	Product discharge/bag filling Bag/container emptying Small-scale weighing Machining of products	3.4.3.1 3.4.3.2 3.4.2 3.4.3.4	14, 15, 16 17 10, 11, 12, 13
Product packaging	Small-scale weighing/handling Large-scale weighing/handling Product packaging	3.4.2 3.4.3.3 3.4.3	10, 11, 12, 13 18 14, 15, 16, 18
Maintenance	Facility equipment cleaning Air filter change-out Spill clean-up	3.4.4 3.4.4.1 3.4.4.2	19

3.4.1 Reactor Operation and Cleanout Processes

Harvesting material from reactors has been identified as a potentially high exposure activity in several manufacturing plants [Demou et al. 2008; Lee et al. 2010, 2011; Methner 2008; Yeganeh et al. 2008]. In addition, cleanout of reactors has contributed to increasing facility concentrations and exposures to operation and maintenance workers. Leakage from pressurized reactors can also contribute to background concentrations and result in exposure to employees throughout the facility. When the reactors are small, some facilities have placed them inside fume hoods to help control fugitive emissions. Two studies have shown that when the reactor is housed in a well-designed and operated fume hood, particle loss to the work environment is low [Tsai et al. 2009b; Yeganeh et al. 2008]. When the reactors are larger, enclosures can be built that isolate the reactor from the environment and seek to reduce fugitive emissions (Figure 7).

Methner et al. [2010] summarized airborne measurements in 12 facilities that processed nanomaterials, including manufacturers and research and development labs. The authors found that some of the highest measured exposures occurred during reactor cleanout tasks, which included brushing and scraping slag material from the reactor walls and during torch cleaning. Demou et al. [2008] evaluated exposure to nanoparticles at a pilot-scale nanomaterial production facility. The major emission source was determined to be the production unit as the airborne particle concentrations rose when the unit was started and fell when production rate was decreased. The other task that resulted in substantial particle release was cleaning of the reactor using a vacuum cleaner not fitted with a HEPA filter. Evans et al. [2010] studied nanoparticle concentrations in a facility that manufactured and processed carbon nanofibers (CNFs). During the thermal treatment of the CNFs in a reactor under positive pressure, elevated concentrations of non-CNF ultrafines were released.

Figure 7. A large-scale ventilated reactor enclosure used to contain production furnaces to mitigate particle emissions in the workplace (Used with permission from Flow Sciences, Inc.)

Lee et al. [2010] conducted personal, area, and real-time sampling in seven CNT plants. Results showed that nanoparticles and fine particles were most frequently released upon opening the chemical vapor deposition (CVD) furnace. Catalyst preparation and the opening of the CVD furnace resulted in the release of nanoparticles in the range of 20–50 nm. Lee et al. [2011] also evaluated workplace exposures to nanoscale TiO₂ at manufacturing plants. In one TiO₂ plant, the reactor was small and was placed in a fume hood; the entire process was conducted in that hood. Even though the reactor was located in the hood, high concentrations of nanoparticles were measured outside the hood. Worker exposure increased during product harvesting because the worker put his head into the hood to brush out the product powder. A second TiO₂ plant isolated the large-scale reactor with a vinyl curtain and used a glove box for the harvesting of product from the reactor. Overall, airborne particle concentrations were fairly stable during production although increases occurred during both the operation of a process vacuum pump and welding conducted in the facility.

Yeganeh et al. [2008] evaluated a small facility producing carbonaceous nanomaterials including fullerenes. The process involved the production of materials in an arc furnace that was enclosed in a ventilated fume hood. This hood had a plastic front face shield and ports that allowed worker access during the process. The process involved placing graphite rods into the furnace, volatilizing the graphite in the furnace, producing raw soot, and using a scoop and brush to remove raw soot into a jar. At the beginning or end of each day, the reactors were completely cleaned by manual sweeping and vacuuming to remove residual soot. Real-time particle analyses showed that physical handling of material (sweeping of the reactor) resulted in the aerosolization of ultrafine particles. Measurements inside and outside the reactor enclosure (i.e., fume hood), however, showed that the hood was effective at containing particulates.

Methner [2008] evaluated the use of a portable LEV unit for controlling exposure during cleanout of a vapor deposition reactor used for producing nanoscale metal catalytic materials comprised of manganese, cobalt, or nickel. Following the automated collection of product materials, an operator cleaned out slag and waste product from the reactor using brushes and scrapers. Initial measurements had shown this task to be a high-exposure task for the operator. A follow-up survey was conducted at the facility using a commercially available fume extraction unit with HEPA filtration to pull airborne dusts away from the operator during cleanout. Analysis of real-time instrumentation and filter samples analyzed for metals showed an average reduction in airborne concentrations of 88%–96% during three cleanout procedures.

Emission sources related to reactor operations, harvesting, and maintenance can be categorized as fugitive or task–based. The approaches that have been used for controlling fugitive emissions from the reactor have primarily been ventilated enclosures. Laboratory fume hoods and glove boxes can be used when the reactor is small, typical of R&D or pilot operations. Where the production reactors are larger, custom-fabricated enclosures often constructed from a polycarbonate, transparent thermoplastic material, or vinyl curtains have been used to reduce emissions (Figure 7). When designing these types of enclosures, it is necessary to consider reactor access needs, determination of exhaust airflows capable of maintaining a negative pressure (even during the opening of access doors), and accommodation of heat loads generated by the process. Failure of containment can result from not carefully addressing these key design needs. When looking at pressure differentials, it is important to study the airflow to minimize turbulent situations that can actually increase particle release rather than containing the particles.

Figure 8. A canopy hood used to control emissions from hot processes When a process is heated, the use of canopy hoods (Figure 8) may be another reasonable alternative as long as the design meets the operational and facility exposure control requirements [ACGIH 2013; McKernan and Ellenbecker 2007]. Even if the process does not involve heat, contaminant capture velocities suitable for gas/vapor contaminants (rather than coarse particulates) may be sufficient, as ultrafine and nanoparticles possess negligible inertia and follow the flow stream well.

When controlling exposures during operations such as product harvesting and reactor cleanout, solutions such as spot LEV systems (e.g., a fume extractor) or containment may be acceptable alternatives. Manual harvesting of product materials may be better suited for higher-level enclosure controls such as a glove box or a specially designed enclosure to provide good capture while minimizing loss of product materials. The use of a commercially available fume extractor has been shown to be effective in reactor cleanout and provides a flexible solution that may meet facility needs across a range of operations [Methner 2008]. Selection of any control should be evaluated to ensure worker acceptance and use as well as verifying that it meets the exposure control objectives.

3.4.2 Small-scale Weighing and Handling of Nanopowders

Small-scale weighing and handling of nanopowders are common tasks; examples include working with a quality assurance/control sample and processing small quantities in downstream industries. During these operations, workers may weigh out a specific amount of nanomaterials to be added to a process such as mixing or compounding. The tasks of weighing out nanomaterials can lead to worker exposure primarily through the scooping, pouring, and dumping of these materials. Many different types of commercially available laboratory fume hoods can be employed to reduce exposure during the handling of nanopowders. Other controls have also been used in the pharmaceutical and nanotechnology industries for containment of powders during small-quantity handling and manipulation. They include glove boxes, glove bags, biological safety cabinets or cytotoxic safety cabinets, and homemade ventilated enclosures.

Methner et al. [2007] evaluated a university-based research lab that used CNFs to produce high-performance polymer materials. Several processes were evaluated during the survey: chopping extruded materials containing CNFs, transferring and mixing CNFs with acetone, cutting composite materials, and manually sifting oven-dried CNFs on an open benchtop. Real-time monitoring did not identify any process as a substantial source of airborne CNF emissions; however, weighing/mixing of CNFs in an unventilated area resulted in elevated particle concentrations compared to background. Other studies have shown that benchtop activities such as probe sonication of nanomaterials in solution can also result in emission of airborne particles [Johnson et al. 2010; Lee et al. 2010]. Producing dispersions by sonication is a primary operational step, and the industrial hygiene assessment should address the sound level exposure as well as the potential exposure to aerosols of nanomaterials from the sonication. Maintaining the sonicator/dispersion process within an enclosure such as a hood can be an effective means for mitigating the noise and aerosol exposure.

3.4.2.1 Fume Hood Enclosures

In 2006, a survey was conducted of international nanotechnology firms and research laboratories that reported manufacturing, handling, researching, or using nanomaterials [Conti et al. 2008]. All organizations participating in the survey reported using some type of engineering control. The most common exposure control used was the traditional laboratory fume hood with two-thirds of firms reporting the use of a fume hood to reduce exposure to workers. These devices have been used for many years in research laboratories to protect workers from chemical and biological hazards. The design and operation of the fume hood is an important factor when considering good exposure control. Traditional designs for laboratory fume hoods create airflow patterns that form recirculation regions inside the hood. In addition, airflow around the worker, as shown in Figure 9, creates a negative pressure region downstream of the worker, which may provide a mechanism for the transport of materials out of the hood as well as into the breathing zone of the worker.

Recent research has shown that the laboratory fume hood may allow the release of nanomaterials during their handling and manipulation [Tsai et al. 2009a]. This research evaluated exposures related to the handling (i.e., scooping and pouring) of powder nanoalumina and nanosilver in a constant air volume (CAV) hood, a bypass hood, and a variable air volume (VAV) hood. This study showed that the CAV fume hood, in which face velocity varies inversely with sash height, allowed the release of significant amounts of nanoparticles during pouring and transferring activities involving nanoalumina. The particles that escaped the fume hood were circulated to the general room air and were not cleared by the general ventilation system for 1/2–2 hours. Sash heights both above and below the recommended height (corresponding to a face velocity of 80–120 ft/min) may lead to increased potential exposure for the user. In contrast, more modern hoods such as the VAV hood, which is designed to maintain the hood face velocity in a desired range regardless of sash height, yielded better containment of nanoparticles than the other hoods tested.

Fume hood airflow with body

Figure 9. Schematic illustration of how wakes caused by the human body can cause transport of air contaminants into the worker's breathing zone A meta-analysis of fume hood containment studies was conducted to identify the important factors that affect the performance of a laboratory fume hood [Ahn et al. 2008]. An analysis of factors affecting the containment performance of the hoods showed that worker exposures to air contaminants can be greatly impacted by a variety of operational issues. Increasing the distance between the contaminant source and the breathing zone leads to reduced exposure. Exposures can also be reduced by limiting the height/area of the sash opening; increasing the height of the sash opening increased the risk of hood containment failure. The presence of a manikin/human subject in front of the hood caused the greatest risk of hood failure among factors studied. This indicates that containment testing should include an operator or manikin to adequately assess hood performance. Face velocity did not make a significant difference in hood performance unless it was extremely high or low (> 150 ft/min or < 60 ft/min). Several hood operating factors showed an effect but were not statistically significant, including sash movement, hand and arm movement, pouring/weighing, and thermal load.

New fume hoods specifically designed for nanotechnology are being developed primarily based on low-turbulence balance enclosures, which were initially developed for the weighing of pharmaceutical powders. The use of bench-mounted weighing enclosures, as seen in Figure 10, is common for the manipulation of small amounts of material. These fume hood-like LEV devices typically operate at airflow rates lower than those in traditional fume hoods and use airfoils at enclosure sills to reduce turbulence and potential for leakage. They also have face velocity alarms to alert the user to potentially unsafe operating conditions. Based on the hazards assessment, these fume hood-like LEV devices can be outfitted with HEPA filtration or connected to the ventilation exhaust system.

Photo by NIOSH

Figure 10. Nano containment hood adapted from a pharmaceutical balance enclosure

3.4.2.2 Biological Safety Cabinets

The Centers for Disease Control and Prevention (CDC) divides biological safety cabinets (BSCs) into three classes: Class I, Class II, and Class III. The Class II BSCs are further divided into four subcategories (A1, A2, B1, B2) [DHHS 2009]. These hoods are used for processes that require operator and product protection. The BSC pulls air into the hood to protect the operator while providing a downward flow of HEPA-filtered air inside the cabinet to minimize cross-contamination along the work surface (see Figure 11). The most common BSC (Type II/A2) uses a fan to provide a curtain of HEPA-filtered air over the work surface. The downward moving air curtain splits as it approaches the work surface; some of the air is drawn to the front exhaust grille and the remainder to the rear grille. The air is then drawn back up to the top of the cabinet where it is recirculated or exhausted from the cabinet. In general, 70% of the air is HEPA-filtered and recirculated while 30% is filtered and then exhausted from the cabinet. The make-up air is drawn through the front of the cabinet. The air being drawn in acts as a barrier to protect the workers from contaminated air leaking out of the hood.

Figure 11. A tabletop model of a Class II, Type A2 biological safety cabinet (BSC) (Used with permission from ASHRAE [2011].) Cena and Peters [2011] evaluated the effectiveness of ventilated enclosures including a Class II, Type A2 BSC and a custom fume hood during the manual sanding of epoxy test samples reinforced with CNTs. Sanding of CNT-epoxy materials released respirable-sized (micronsized) particles but generally no nano-sized particles. The respirable mass concentration in the operator’s breathing zone while using the BSC was approximately two orders of magnitude lower than the concentration when using the custom fume hood. The use of the custom fume hood resulted in an increase of breathing zone concentrations of about one order of magnitude compared to the use of no controls. The custom fume hood had a low average face velocity of about 45 ft/min with high variability across the hood face. The authors suggested that the poor performance of the custom fume hood may have been due to its rudimentary design, which did not include a front sash or rear baffles. The lack of these common fume hood features along with the low average face velocity may have resulted in poor airflow distribution across the face and increased leakage.

Macher and First [1984] evaluated the effect of airflow rates and operator activity on containment effectiveness for a Class II, Type B1 biological safety cabinet using bacterial spores released by two 6-jet collison nebulizers. The hood sash height correlated negatively with the containment effectiveness; that is, the higher of two sash heights provided better containment of the aerosol. In addition, working in the front half of the cabinet provided better protection than working in the rear half of the cabinet. The authors postulated that working in the rear of the hood caused the operator to move closer to the hood opening, blocking the opening and causing more turbulence and leakage around the sides of the hood. The operator withdrawing his arms from the hood caused significantly more leakage than moving arms side to side within the hood. The authors concluded that testing BSCs with persons working at them provides more information than static testing alone and that even well-designed cabinets lose a small fraction of aerosols.

3.4.2.3 Glove Box Isolators

A glove box isolator fully isolates (contains) a small-scale process and is sometimes referred to as a primary protection device (Figure 12) [HSE 2003a]. The design can be either the more typical hard unit or a soft, flexible containment unit (often referred to as a glove bag). Glove boxes provide a high degree of operator protection but at a cost of limited mobility and size of operation. In addition, cleaning the glove box may be difficult, and, to prevent exposures, operators should use caution when transferring materials and equipment into and out of the glove box. In general, glove boxes include a pass-through port, which allows the user to move equipment or supplies into and out of the enclosure.

The performance of a glove box containment system was evaluated during weighing activities of fine lactose powder (a common pharmaceutical surrogate test material). Air samples were collected at four locations: inside the glove box, in the pass-through, in front of the glove box, and at the exit of the recirculating HEPA filter [Walker 2002]. The results of sampling a 10-minute task showed the average concentration measured inside the glove box was 298 µg/ m³, the average concentration in the integral pass-through was 35 µg/m³, and concentrations measured in the room, including downstream of the glove box exhaust, were below the analytical limit of detection of 1 µg/m³. Sample swabs of interior surfaces showed dust contamination within both the main glove box and pass-through. These results indicated that, although internal surfaces were contaminated with the materials, no leakage from the glove box was detected.

Figure 12. A glove box isolator for handling substances that require a high level of containment (Contains public sector information published by the Health and Safety Executive and licensed under the Open Government License v1.0.)

3.4.2.4 Air Curtain Fume Hood

A recent fume hood design addresses the known issues surrounding the recirculating flow patterns both inside the fume hood and around the operator (Figure 9). The air curtain-isolated fume hood, as shown in Figure 13, uses a push-pull ventilation configuration created by a narrow planar jet from the sash to an exhaust slot along the base of the hood opening. Tsai et al. [2010] evaluated the performance of this hood during handling and manipulation of nanoparticles. In this test, measurements in the worker's breathing zone were taken while nanoalumina powders were manually transferred or poured between several 400-ml beakers. The air-curtain hood had very low particle release during all tested conditions (i.e., varying sash heights) with low but measurable release occurring at the lowest sash position. This same study showed that the particle leakage from two traditional fume hoods (both a CAV and VAV hood) exhibited substantial particle release during similar nanomaterial handling operations. This study suggested that the air curtain isolated hood may provide better containment performance during typical handling procedures.

Figure 13. Air curtain safety cabinet hood that uses push-pull ventilation (Used with permission from Huang et al. [2007a].)

3.4.2.5 Summary

Overall, the published studies suggest that the selection of a fume hood with improved operating characteristics such as a VAV hood provides better operator protection than conventional fume hoods when handling dry nanomaterials. When using any hood, the worker should strive to maintain the face velocity in the recommended range of 80–120 ft/ min [ACGIH 2013]. Additionally, proper use of the engineering control by the operator and validation of the performance of the control equipment are essential for risk mitigation. Newer nano hoods based on pharmaceutical weigh-out enclosures may be a reasonable alternative to larger fume hoods when only small-scale, benchtop manipulation of powders is needed. These hoods have proven effective in the pharmaceutical industry but need more thorough evaluation to assess the impact of lower airflow rates on containment performance, especially in the industrial environment. BSC-type hoods are commonly used for containing hazardous powders in hospitals for hazardous drug formulation and may have features designed to improve containment performance over traditional fume hoods. However, there are few published studies on their effectiveness for containing nanomaterials. The selection of a BSC appropriate for use with nanomaterials is essential. Considerations including how to clean the cabinet after use, how to maintain the BSC during required maintenance such as filter change-outs, and proper exhaust configuration (sending exhaust out of the production area versus recirculating exhaust air) should be considered prior to use. Glove box isolators typically provide a greater level of worker protection but at a cost of reduced access and limited operational scale. Newer hood designs, such as the air curtain fume hood, have shown excellent containment performance in initial studies and may be potential control options in the future.

Many options are available to facilities that require worker protection during small-scale material handling operations. The best option for a given process depends on several factors including scale of handling operations, physical properties of materials being handled (size, density, wet or dry formulation), work environment (lab versus plant, cross drafts, nearby activity), equipment requirements (size of equipment/operation being enclosed), and level of protection required. Independent of the control selected, users should also adopt good work practices, such as using the smallest possible quantities of materials. Other procedures, such as wiping down and sealing containers before they are removed from the enclosure, are recommended. In addition, using care when working with powders, such as refraining from dropping dust from height, helps to prevent dust generation and to reduce operator exposure. The proper positioning of these workstations away from doors, windows, air supply registers, and aisleways will help to reduce the impact of cross drafts.

3.4.3 Intermediate and Finishing Processes

Exposures resulting from the manual handling of powdered materials are common in industry. Reduction in worker exposure through implementation of careful work practices and appropriate engineering controls would benefit these operations. Dumping bags of powdered materials has been commonly reported in the literature for production and processing. Typically, a worker dumps the ingredients for one process into a hopper and then compacts or disposes of the empty bags. Ventilated bag-dumping stations have been used successfully in a variety of industries and applications. The transfer of large quantities of nanomaterials requires different solutions adapted to the particular process. However, a few controls that are applicable to these common processes are available and have been evaluated for similar industrial operations.

After the completion of production, many nanomaterials are sent for further processing. The powder product may be refined through a common process such as spray drying [Lindeløv and Wahlberg 2009]. Other studies have documented collection of fugitive emissions of nanomaterials from the process reactors using devices such as baghouse air filters [Evans et al. 2010]. In both of these operations, the nanomaterials are collected in a barrel or drum following the completion of these production steps. Several examples of engineered drum or bag filling solutions have been described elsewhere and could be implemented to reduce such releases [ACGIH 2013; Hirst et al. 2002]. These engineering controls consist of enclosing the product off-loading process by temporarily sealing the drum/bag to the filling vessel above and/or overbagging through a continuous liner type bagging system. The addition of a local exhaust ventilation hood near the drum/bag opening could also be used to capture airborne nanomaterials.

Evans et al. [2010] studied nanoparticle concentrations in a facility that manufactured and processed carbon nanofibers (CNFs). The authors discussed four discrete events that resulted in elevations in airborne particle concentrations. The largest increases in particle concentration measured within the plant were related to manual handling processes, such as dumping product into lined drums and manual change-out and closing bags of final treated CNF product. Increases in particle concentrations were the result of the change-out and closing of the collection bag containing approximately 15 lbs of CNFs. Emissions from this event were almost entirely due to aerosolized CNFs. Tamping of the bag to settle contents (so that it could be adequately closed) and subsequent closing appeared to efficiently aerosolize CNF material through the bag opening into the workplace environment. This resulted in an increase in respirable mass concentration and a dark visible airborne CNF plume.

A few studies have been conducted to look at the emission of nanoparticles from downstream products during machining of nanocomposites. Methner et al. [2007] reported increases in total carbon (a marker for nanoparticles), particle number, and mass concentration during the wet sawing of a CNF-impregnated composite. However, the increase in particle concentration was primarily of particles greater than 400 nm in diameter. Vorbau et. al. [2009] evaluated nanoparticle release from oak and steel panels coated with polyurethane mixed with zinc oxide nanoparticles. A standard abrasive test rig was used to provide uniform conditions for testing the release of particles from the surface of the panels. During the abrasion tests, no significant release of particles below 100 nm was observed. However, the nanoscale zinc oxide particles were embedded in the aerosols with larger surface area. Bello et al. [2009] evaluated the release of nanoscale particles during dry and wet cutting of nanocomposite materials. Two composites were used for evaluation: a CNT-enhanced graphite prepreg laminate sheet and a woven alumina fiber cloth with CNTs grown on the surface of the fibers. Significant exposures to nanoscale particles were generated during dry cutting of all composites with emission levels being related to composite material and thickness; wet cutting reduced exposures to background levels.

For all processes/tasks discussed, engineering controls should be adapted for the specific process. Acceptable exhaust volumes and capture velocities may differ from currently available guidance due to differences in materials being handled. Pilot testing of any controls should be conducted to evaluate proper control operation and verify that exposures are controlled to desired levels.

3.4.3.1 Product Discharge/Bag Filling

The process of filling bags with nanomaterials is commonly done following large-scale production or refining processes. The off-loading of product after spray drying, for example, may be a significant source of exposure when post-processing nanomaterials. In the spray-drying process, a mixture of liquid and powder ingredients (slurry) is sprayed within a large sealed tank. Heat within the tank dries the slurry droplets, leaving a powder as the finished product. When the process is completed, the powder product is commonly discharged into a bulk tote or drum before packaging. Methner et al. [2010] reported exposure measurements at 12 facilities and noted that the highest background-adjusted concentration was observed during spray dryer drum changeout. Evans et al. [2010] reported exposures related to changing out a drum that collected fugitive CNF materials from a process reactor using a baghouse filtration system. Even though the processes differ, the tasks for each of these steps are similar and include the removal of the drum from the process outlet. These drums are often sealed to the process outlet minimizing exposure during production but potentially expose workers when removing the drums or liners.

A ventilated, collar-type hood around the discharge point can help minimize worker exposure to dust. Figure 14 presents a control approach for filling bags with solid powder materials [HSE 2003c]. The control includes the specification of a ventilated enclosure around the powder discharge outlet and applies to filling smaller product bags as well as intermediate bulk containers. This design guidance recommends an inward air velocity of 200 ft/min (1.0 m/s) into the enclosure. The ACGIH Industrial Ventilation Manual [ACGIH 2013], Design plate VS-15-02, Bag Filling, is similar in design to the HSE exhaust hood (Figure 14) but specifies an overall hood flow rate of 400–500 ft³/min for nontoxic dust or 1,000–1,500 ft3/ min for toxic dust with a maximum inward air velocity of 500 ft/min. These flow rates have been specified for common industrial powders and may need to be adjusted based on the process and properties of the nanoscale materials being addressed to prevent excessive loss of product.

Figure 14. Ventilated collar-type exhaust hoods for containing dust during product discharge or manual bag filling (Contains public sector information published by the Health and Safety Executive and licensed under the Open Government License v1.0.) In addition to ventilation solutions, other dust control approaches have been used in a variety of industries and should be applicable for nanomaterial production. For example, an inflatable seal can be used to create a dust-tight seal on the discharge outlet of a spray dryer during the product discharge/bag filling process (Figure 15). The seal inflates during the product transfer from the process to the packaging bag (providing the seal) and deflates once the transfer is completed to allow removal of the bags. These systems are available on many commercially available bulk bag filling systems [Hirst et al. 2002].

Another system that can be used to contain powders during process off-loading/emptying is the continuous liner system (Figure 16). Polypropylene liners are often used when products are discharged from the industrial processes into the intermediate or final product containers. In this operation, a sleeve of polypropylene liners is stowed around the circumference of the discharge outlet. The first liner, the bottom having been sealed, is pulled down into the overpack (usually a drum or a cardboard box). Product is discharged into the liner through a butterfly valve on the process outlet. Once full, the top of the first liner sleeve is closed using tape or a fastener, or it is heat sealed and cut. The product is sealed within the poly-lined container, and a new sealed poly liner is pulled down to start discharge into the next container. This continuous process seals off the primary leak paths for dust during unloading of an industrial blender or other equipment. These systems are commonly used in the pharmaceutical industry and may provide cost-effective alternatives to traditional local exhaust ventilation control systems for nanotechnology facilities.

Figure 15. An inflatable seal used to contain nanopowders/dusts as they are discharged from a process such as spray drying

Figure 16. A continuous liner product off-loading system that uses a continuous feed of bag liners fitted to the process outlet to isolate and contain process emissions and product (Used with permission from ILC Dover.)

3.4.3.2 Bag Dumping/Emptying

Technology used to control dusts during bag dumping has been in place for many years. The standard control—a ventilated bag dump station—consists of a hopper outfitted with an exhaust ventilation system to pull dusts away from workers as they open and dump bags of powdered materials.

This equipment eradicates the dust problems caused by manually emptying bags and the need to dispose of empty bags. This ensures a healthy environment is maintained in the process area as well as reduces maintenance and repair problems caused by powder contamination to surrounding areas. The basic equipment consists of a bag dump cabinet with a dust extraction outlet for connection to a separate dust collector or existing plant exhaust. A bag is placed on the mesh support shelf and manually slit, with the contents falling directly into the inlet of a flexible conveyor or mixing tank. A side-mounted empty bag compactor may also be included. Design examples for these devices are available from several manufacturers of industrial materials. The British HSE has developed a control approach for a ventilated station for emptying bags of solid materials [HSE 2003d]. The control includes the specification of a face velocity of 200 fpm (1.0 m/s) and includes a waste bag collection chute (Figure 17).

Figure 17. A ventilated bag-dumping station that reduces dust emissions when emptying product from bags into a process hopper (Contains public sector information published by the Health and Safety Executive and licensed under the Open Government License v1.0.)

Research into the effectiveness of these types of devices has shown that worker exposure to dust and vapors can be reduced. A review of commercially available units showed that their use with a variety of materials—including limestone, carbon black, and asbestos—controlled particle concentrations to acceptable levels [Heitbrink and McKinnery 1986]. However, particle contamination on the surface of the bag and handling/disposal of bags caused increased worker exposure. An integral pass through to a bag disposal chute/compactor can help reduce dust exposure resulting from bag handling. Further studies in mineral processing plants showed that the use of an overhead air supply also significantly decreased worker exposure [Cecala et al. 1988]. The ACGIH Industrial Ventilation Manual also has two designs that are applicable to the control of powder materials during bag dumping [ACGIH 2013]. Design plate VS-15-20, Toxic Material Bag Opening, is similar in design to the HSE station described above but recommends a slightly higher control velocity of 250 fpm at the face of the station opening. In addition, Design plate VS-50-10, Bin and Hopper Ventilation, requires a hood face velocity of 150 fpm. In general, higher velocities are specified to adequately capture dusts in a plant environment. While the materials used in the studies discussed above were not nanoscale, the application of the dust control concept is still relevant. However, the capture velocities specified in the Industrial Ventilation Manual may be excessive when attempting to contain nanomaterials; lower velocities may be warranted.

Figure 18. A unidirectional downflow booth for handling larger quantities of powders (Used with permission from Esco Technologies, Inc. [2012].)

3.4.3.3 Large-scale Material Handling/Packaging

Unidirectional flow booths, or downflow booths, as seen in Figure 18, are used in pharmaceutical applications for large-scale powder packing, process loading, and tray dryer loading [Hirst et al. 2002]. Similar applications have been proposed for handling hazardous dye powders. In general, these booths supply air from overhead (commonly at 100 fpm) over the full depth of the booth. Particles generated by processes carried out in the booths are captured and carried to the exhaust registers, which are located along the back wall of the booth. For the nanotechnology industry, these booths may provide a flexible solution for several common processes, including packaging of materials, transferring materials between process containers, or loading materials into containers for post processing.

Floura and Kremer [2008] evaluated a downflow booth used for transferring 25 kg of lactose (a surrogate pharmaceutical material) from drum to drum inside a downdraft booth. Air samples were collected in the operator’s breathing zone and around the perimeter of the process during the transfer operation. The operator scooped the lactose powder from the initial drum into the final product drum until it was nearly empty and then carefully inverted the bag to pour the remainder of the contents into the final container. With no active ventilation controls, the concentration within the operator breathing zone averaged 2,250 µg/m³. When the ventilation inside the booth was turned on, the breathing zone concentration was substantially reduced to an average of 1.01 µg/m³. Finally, the authors evaluated the downflow booth with a ventilated collar added. The ventilated collar surrounded the interface between the drums and exhausted air at a rate of 425 ft3/min. During this test, the initial drum was inverted and the powder materials were emptied by gravity with the operator massaging the materials into the final product drum. The operator’s breathing zone concentration averaged 0.03 µg/m³ during this process. This study showed that the use of a downflow booth significantly reduced operator exposure during powder transfer processes and that adding a second level of LEV, the ventilated collar, further reduced the exposure by two orders of magnitude.

HSE Control Guidance Sheet 202, Laminar Flow Booth, presents a design for powder-handling processes called a horizontal- or cross-flow design [HSE 2003b]. The concept behind the design is similar to the downflow booth except that air enters the booth from the booth face. Air moves across the back of the worker toward the back of the booth. An issue with the cross-flow design is the secondary airflow patterns caused by the presence of the operator in the booth. Additionally, if purity or cleanliness of the product is important, sweeping of the air across the operator could be problematic. These patterns may cause turbulent dispersion of dust in the booth and result in higher operator exposure or potential leakage, compared to the downflow booth, but may provide a reasonable control for some processes.

3.4.3.4 Nanocomposite Machining

Initial studies have shown that machining some nanocomposite materials can result in the release of nanoscale particles to the work environment. Engineering controls when machining materials are available for most common processes. They range from ventilation of handheld tools using a high velocity-low volume (HVLV) system to the use of wet cutting techniques commonly adopted for silica control during construction activities. The use of standard dust controls such as those described by the HSE for woodworking as well as those identified in the ACGIH Industrial Ventilation Manual for machining processes provide a source of guidance that can be used to identify controls for machining processes. Bello et al. [2009] showed that the use of wet suppression techniques during sawing of nanocomposites reduced exposures down to background levels.

3.4.3.5 Summary

Processing nanomaterials involves a variety of steps. Following the production process, bulk unrefined materials may be packaged and shipped for use or may be subject to further processing. These processes require handling and manipulation of nanomaterials and have been shown to be a point of exposure for workers. These processes typically are composed of a limited number of tasks that may result in exposure of workers to nanoparticles or their agglomerates.

Product discharge. When processes empty into a large container, there is a potential for exposure especially when removing the full drum. Several engineering controls are available for this process/task. Nonventilation controls, such as inflatable seals and continuous liner systems, reduce the possibility of exposure. Ventilation-based options include the ventilated collar or enclosure around the discharge point. These solutions have been used and evaluated in a variety of industrial settings and have been shown to effectively capture dusts when properly designed and implemented in the process.

Bag dumping/emptying. When raw, bulk nanomaterials receive further processing/refining, those materials are often dumped from containers such as drums or bags into hoppers that feed the downstream processing equipment. Ventilated bag dump stations have been in use in industry for many years and have been proven to be effective in controlling dusts. Several commercial vendors and sources of design guidance exist for these devices.

Large-scale handling/packaging. When nanomaterials are handled in quantities larger than those that can easily fit in a fume-type hood, a unidirectional flow booth can provide a suitable control to reduce worker exposure and mitigate a potential emission source. These booths are commonly used in the pharmaceutical industry and have also been employed for handling hazardous dye powders in industrial settings. They provide the flexibility for a variety of operations that require handling of nanomaterials from larger containers, such as drums. They can also be designed to provide local exhaust for specific operations that may occur inside the booths. These booths are available for a variety of commercial vendors or can be designed from sources of readily available guidance.

Machining of nanocomposites. When machining composite materials coated or impregnated with nanomaterials, good dust suppression techniques should be used. Guidance on dust suppression techniques from ventilation-based (woodworking-type) or mist/ water-based (silica/construction-type) controls may be adopted to reduce worker exposures. Exposures during machining should be carefully monitored and controlled. Standard engineering controls may need modification to properly control emissions. In addition to engineering controls, workers may need to wear appropriate respiratory protection.

3.4.4 Maintenance Tasks

Maintenance of the production facility and equipment can lead to exposures that are often overlooked. Demou et al. [2008] noted that maintenance procedures were a source of considerable particle emissions, specifically during the vacuuming of a reactor using a vacuum cleaner with a high-efficiency filter. However, other researchers have observed that cleaning the process area after CNT preparation reduced airborne particle concentrations [Lee et al. 2010]. Another typical activity not reported in the literature is the changeout of facility air filters. When local exhaust ventilation is used to contain nanomaterials and dusts, facilities will typically use air filtration prior to exhausting air from the building or recirculating into the work zone. When filters require change-out, the use of integral containment equipment and procedures can reduce maintenance worker exposure. Other general maintenance procedures, such as modifying ductwork or performing fan maintenance, will also require appropriate precautions to avoid exposing workers to nanomaterials settled in the equipment. In addition, general good housekeeping processes and written spill response procedures can help reduce the potential for worker exposure.

3.4.4.1 Filter Change–out–Bag In/Bag Out Systems

Bag in/bag out procedures are typically designed to protect workers performing maintenance on air filter change out. Bag in/bag out housings are specifically designed to allow for removal of a dirty air filter while minimizing worker exposure [Filtration Group Inc. 2012]. In these systems, a plastic liner is attached to a service port on the filter unit, as shown on the following page in Figure 19. When the filter is ready for replacement, the facility maintenance worker, wearing appropriate PPE, removes the filter into a liner. This process contains the filter with its contaminants so the worker is not exposed and the particulates are not resuspended in the workplace environment.

3.4.4.2 Spill Cleanup Procedures

An organized, clean workplace enables faster and easier production, improves quality control, and reduces the potential for exposure. It is important to maintain good general housekeeping practices so that leaks, spills, and other process integrity problems are readily detected and corrected. Proper practices regarding spills include:

• Allowing only individuals wearing appropriate protective clothing and equipment and who are properly trained, equipped, and authorized for response to enter the affected area until the cleanup has been completed and the area properly ventilated.

• Using HEPA-filtered vacuums, wet sweeping, or a properly enclosed wet vacuum system for cleaning up dust that contains nanomaterials.

• Cleaning work areas regularly with HEPA-filtered vacuums or with wet sweeping methods to minimize the accumulation of dust.

• Cleaning up spills promptly.

• Limiting accumulations of liquid or solid materials on work surfaces, walls, and floors, to reduce contamination of products and the work environment.

Figure 19. Removal of a dirty air filter from a ventilation unit into a plastic bag to minimize worker exposure to particles captured by the filter unit (Used with permission from Filtration Group Inc. [2012])