Safety and Health on Bridge Repair, Renovation and Demolition Projects

Table of Contents

R. Leroy Mickelsen

R. Leroy Mickelsen

ABSTRACT

A basic need for nearly all steel structures is protection from corrosion.  Historically, lead-containing coating systems were used because they were low cost, aesthetically appealing, and corrosion resistant.  Periodically, such coating systems must be maintained or completely replaced.  To adequately prepare the steel surface to receive a new coating system, the old coating is usually removed.⁽¹⁾  The cleaning process has traditionally been achieved by unconfined abrasive blasting.  Abrasive blast devices are designed to deliver a high-velocity stream of abrasive to remove the coating as well as impart an anchor pattern on the metal surface.  The workers direct the blasting nozzles at the surface to be cleaned.  As the paint is removed, small particles of lead paint, silica (silica from abrasive or from surface coatings), and other debris become airborne.  Workers exposed to lead, silica, arsenic, chromium, nickel, zinc, and other health hazards during the paint removal processes frequently exhibit health effects linked to their workplace exposure.⁽²⁾

Two environmental requirements have been driving forces for contractors to contain paint chips, dust, and used abrasive during paint removal processes.  The Resource Conservation and Recovery Act (RCRA) requires that waste material must be collected, tested, and classified as hazardous or not hazardous.⁽³⁾ Another requirement, the Clean Air Act, limits levels of particles with an aerodynamic diameter less than or equal to a nominal 10 micrometers (PM10) to a maximum of 150 mg/m³ average concentration over a 24-hour period.⁽⁴⁾  The Clean Air Act also limits the amount of airborne lead to 1.5 mg/m³, evaluated as a maximum arithmetic mean averaged over a calendar quarter.  Containing environmental emissions has concentrated the contaminants in and around the paint removal containment structures and increased workers' risk of occupational exposure to lead and other waste materials.  The primary goal of most ventilation systems at lead paint removal sites is to reduce environmental release, not reduce the occupational exposure.

Occupational exposure to airborne lead at construction sites is regulated by OSHA to no more than 50 mg/m³ as an 8-hour time weighted average.⁽⁵⁾  The regulation prescribes air monitoring for workers potentially exposed to lead, blood lead testing for workers exposed to 30 mg/m³ or greater, and medical removal conditions to protect workers who have elevated blood lead levels.  The OSHA lead in construction standard should be consulted for details of blood lead monitoring and other requirements.

Exhaust ventilation systems in industrial settings are of two types: local exhaust and general exhaust.  Local exhaust systems capture the contaminant near the source of generation; local exhaust systems are not evaluated in this study.  The local exhaust system is the preferred control method because a large portion of the contaminant can be captured before it reaches the worker.  Examples of local exhaust systems in the paint removal industry are vacuum blast systems and ventilated power tool systems.

General exhaust systems can be used to remove contaminants from the work space by flushing out large quantities of air and replacing it with uncontaminated air.  The clean air mixes with the contaminated air resulting in the average concentration of contaminant in the work space being reduced.  A ventilation system must supply clean makeup air, maintain air flow within the containment, and exhaust contaminated air.

General ventilation systems are not as effective in controlling contaminants as are local exhaust systems.⁽⁶⁾  General ventilation requires large volumes of air for dilution of contaminants.  For most abrasive blasting processes, it is impractical to supply the amount of air required to dilute airborne lead concentrations to permissible levels.  Good general ventilation design necessitates locating the supply inlets and exhaust outlets so that the worker stays upwind from the dust generation source.  In processes where the worker is stationary, this may be easily accomplished.  However, for lead paint removal where the workers are constantly on the move through a complex structure, it is practically impossible to stay upwind from the dust generation source at all times.

CASE STUDIES

Sampling Methods

The control techniques were evaluated by collecting and analyzing bulk samples of paint and abrasive along with personal air samples of workers performing the blasting.  The bulk samples were collected to determine the amount of lead in the paint and abrasive.  The air samples were collected to determine the level of exposure.

Old paint was collected from the steel structures by scraping the surface with a sharp chisel.  The bulk paint collection process removed the top and intermediate paint coatings, leaving a metal surface with only traces of the primer coating (less than 10 percent of the surface was covered by thin traces of paint).  Bulk samples of unused abrasive were obtained from freshly opened packages.  The used abrasive samples were collected by mixing abrasive from five locations within the storage bins or from the containment floor, and taking one sample from the mixture for analysis.

Air samples were collected and analyzed using NIOSH method 7300.⁽⁷⁾  Filter cassettes were placed outside the loose-fitting hood respirator of blast workers.  Earlier personal sampling at abrasive blast work sites resulted in a large number of lost samples.  In an attempt to reduce the likelihood of having the cassettes knocked off the workers' lapels and to reduce the likelihood of pinching the tubing between the worker and the scaffolding or steel members, cassettes were attached to the back of the blasters' outer clothing, rather than on the front lapels.  This is a deviation from NIOSH testing methods and should not be done if at all possible.

Area air samples for total lead were collected using the same equipment as the personal samples, but were located at fixed points within or adjacent to the containment structure.  In addition to the integrated area samples, a Real-time Aerosol Monitor (RAM) was used to measure the respirable dust concentrations inside the containments immediately following abrasive blasting.  Output from this instrument is qualitative in nature, providing only relative measures of respirable dust concentrations.

Ventilated Containment for an Oil Refinery Process Tank

Lead-based paint was removed from a process tank at an oil refinery.  The tank, about 4.5 feet (1.4 m) in diameter and 12 feet (3.6 m) tall, was mounted on a concrete, 2-foot high (0.6 m) octagonal base.  Piping extended above the tank, reaching about 20 feet (6.2 m) above grade (Figure 1).  Abrasive blasting with Starblast7 XL (DuPont Company), a staurolite sand typically containing less than 1 percent quartz, was used inside a ventilated containment.  The blaster used a type CE, continuous-flow, air-supplied blasting respirator.  An adjacent tank similar in size had been blasted previously; however, a quick blast was needed to remove the rust prior to painting.  An aluminum scaffold approximately 17 feet (5.2 m) long, 9 feet (2.7 m) wide, and 21 feet (6.4 m) high was erected around the two tanks.  A scaffold extension approximately 6 feet (1.8 m) long and 3 feet (0.9 m) wide was added to one side providing room for a ladder to access two plywood platforms which were constructed approximately 8 and 15 feet (2.4 and 4.6 m) from the ground.  Personnel from the painting contractor constructed the containment by enclosing the scaffold with 6-mil (0.15 mm) nylon-reinforced polyethylene; a double flap entry was provided in the ladder extension area.

The enclosure was exhausted through three high efficiency particulate air (HEPA) filters by means of pneumatic driven blowers attached to 12-inch (0.3 m) wire-reinforced, polyethylene ducts installed in the east, north, and west sides of the containment.  Two of the ducts were preceded by a 2-foot (0.6 m) square prefilter inside the enclosure.  Each blower was rated at 2000 cubic feet per minute (cfm) (57 m³/min) when driven by 90 psi (6.2*10⁵ N/m²) air.  Fresh air entered through two slits cut into the top of the containment.  The slits formed an X-shape opening approximately 4 square feet (0.4 m²).

The enclosure and exhaust ventilation system contained the particulate and prevent visible environmental release except on a few occasions when the blasting nozzle was directed at the entry flaps or at a weak seal in the containment; at these times visible emissions were observed.  Four of five area air samples, located outside and adjacent to the containment, measured less than 1 mg/m³ of lead.  The fifth sample measured 4 mg/m³ of lead.  Assuming an air flow of 6000 cfm (170 m³/min) evenly distributed over the cross-sectional area of the containment (130 ft² [12 m²]), the average velocity would be about 46 feet (14 m) per minute (fpm).  Measured air velocities using a hot-wire anemometer ranged from 20 to 120 (6 to 37 m per minute) fpm (average = 80 fpm) at the opening between the platforms and the structure.

One area sample, located inside the containment on the first plywood platform, collected a lead concentration of 10,000 mg/m³during the blasting process.  The personal exposure sample outside the helmet of the blaster was 22,000 mg/m³; higher than the area result since the personal sampler was closer to the lead generation source throughout the blasting process.  The bulk samples of old paint from the refinery tank averaged 25 (range 23-26) percent lead by weight.  Dust levels, as measured by a respirable particulate counter, inside containment decayed rapidly when blasting ceased, dropping 90 percent in the first minute.

Large Ventilated Containment at a Bridge Site

Two separate engineering controls were evaluated at one bridge site consisting of two separate containment, ventilation, and abrasive blasting systems.  One control system consisted of abrasive blasting with low silica (<1% by weight) sand, Starblast7 XL (DuPont Company), in a large enclosure made of interconnected canvas tarps suspended from the top of the roadway down to the ground, creating an enclosure with a volume of approximately 200,000 cubic feet (5700 m³) (Figure 2).  The enclosure was ventilated with two 20-inch (0.5 m) diameter exhaust ducts which were suspended from a street light pole; this placed the duct openings approximately 12 feet (3.7 m) above the ground while the abrasive blasting proceeded approximately 40 feet (12.2 m) above the ground.  There were no provisions for supply inlets; the seams of the side tarps and other unplanned openings (most openings were at ground level where the side tarps were not adequately fastened to the ground tarps) acted as supply air inlets.  Dust was observed regularly escaping this large-bridge containment.  The blast equipment area, 15 feet (4.6m) from the containment, had an airborne lead concentration of 210 mg/m³.  The abrasive was used one time and allowed to settle to the floor tarps along with other wastes created by the paint removal process, then it was collected for disposal.  The workers donned a continuous-flow, loose-fitting hood respirator prior to entering the containment.

NIOSH researchers, used a fog generator to visually evaluate the air-flow patterns within each containment.  Fog within several feet of the exhaust duct vents flowed into the vents.  But throughout the rest of the large containment, there was no obvious airflow pattern (mostly stagnant air was observed).  The air velocity at the face of each exhaust vent was 5,500 feet (1700 m) per minute (fpm), and the volumetric flow rate was 24,000 cubic feet (680 m³) per minute (cfm) through the two 20-inch (0.5 m) diameter exhaust ducts.  Within the large containment, the average lead exposure of the four blasters' personal samples was 6200 mg/m³.  The dust levels took 17 minutes to decrease 90 percent after dry abrasive blasting ceased.  Two concentration decay curves are shown in Figure 3.  Quick reductions in dust levels after blasting ceases can result in lower exposures to personnel who enter the containment after blasting.  A quick reduction in dust level was obtained at the oil refinery case study site where supply air and air-flow patterns through the containment were considered in the design of the containment and ventilation system.

Small Containment at a Bridge Site

Another system at the bridge site consisted of abrasive blasting with recyclable steel grit in a 3,000 ft³ (85 m³) enclosure (8' X 8' X 48') made of an alloy piping for the frame, rigid corrugated polycarbonate panels for the sides, and an aluminum grating floor.  The enclosure was suspended from two adjacent I-beams under the bridge (Figure 4).  This small enclosure had a supply air fan at one end and an exhaust duct and fan at the other.  The sloping walls under the grate flooring directed the used abrasive and wastes to the bottom center of the containment where the materials were removed from the containment.  The steel grit was cleaned in an air wash and stored for reuse.

The flow of air within the containment ran perpendicular to structural cross members that connected the two I-beans.  Air flow near these cross members was stagnant in some areas and turbulent in others.  The average air velocities across the working cross-section were 110 fpm (34 m/minute).  The quantity of exhaust air was 6000 cfm (170 m³/minute) and was moved through a 20-inch (0.5 m) diameter duct.

Abrasive and dust escaped from the small-bridge containment only rarely.  The workers donned a  continuous-flow, loose-fitting hood respirator prior to entering the containment. Within the small containment, personal lead concentration for three blaster workers ranged from 6,300 to 58,000 mg/m³ with a geometric mean of 20,000 mg/m³.  The paint samples from the bridge contained approximately 58 (range 57-60) percent lead.

Finally, higher airborne lead concentrations may result from using steel grit that was not adequately cleaned prior to reblasting.  Concentrations of lead in the abrasive (based on one sample each) were:  3700 ppm in the dirty grit, 1700 ppm in the cleaned (recycled) grit, and 100 ppm in the new grit.  An improved abrasive cleaning system may help reduce airborne lead and dust exposures during abrasive blasting.  The amount of exposure attributed to the contaminated abrasive was not determined.

Conclusions and Recommendations

Results from all three ventilation evaluations: the oil refinery containment, the large-bridge containment, and small-bridge containment, demonstrate that general ventilation alone is not an effective means for controlling the high concentrations of airborne lead generated during abrasive blasting of lead-coated steel structures.  Abrasive blasting systems currently being used require the operator to work close to the lead and dust generation source.  For significant reduction of worker exposure, lead particulate must be controlled before it reaches the blast operator.

Properly designed and functioning general ventilation systems provided for rapid decay of respirable dust levels when blasting ceases.  Rapid dust decay may help reduce exposure during re-entry activities such as cleanup; however, it is insufficient to protect the worker from lead exposures during abrasive blasting.

General ventilation may not reduce concentrations of airborne lead and particulate below OSHA standards during abrasive blasting inside containment structures.  Therefore, respiratory protection is needed to reduce worker exposure to hazardous airborne substances such as lead.  The selection of respiratory protection should be based on workplace exposure data.  Respirators should provide enough protection to reduce airborne exposures to acceptable levels.  The use of respirators should supplement the continued use of engineering controls and good work practices, and they should not be used as the only means of exposure reduction.

Alternative methods for steel structures maintenance include overcoating, vacuum blasting, vacuum power tools, wet blasting, automated blasting, and chemical stripping.  These methods may help to reduce worker lead exposures.  Each removal method has particular strengths and weakness.  When implementing new workplace procedures or controls, worker exposure monitoring should be conducted to determine the resulting change in exposure.