Effectiveness of dust control by atomisation of water sprays on handheld demolition and soil compacting equipment

ABSTRACT


A field study was conducted to investigate the effectiveness of atomisation of water as a method of dust control during the use of jackhammers and plate compactors. The study consisted of two substudies, both designed as intervention studies, where task-based respirable quartz concentrations were monitored. In addition, real-time dust concentrations were monitored.

In a first (sub)study, three workers located in a hall used a heavy jackhammer cutting concrete slabs. Each activity was performed four times, twice with and twice without atomisation. Two other workers were involved in scrapping tiles from bathroom walls using a light jackhammer. The activity was performed twice, with and with atomization in identical bathrooms. For both types of jackhammers, significant reductions of quartz exposures were shown. On average quartz exposure decreased 86% with use of atomisation during scrapping tiles, whereas the use of atomisation during demolition of concrete slabs reduced quartz exposure with 64%. However, average quartz exposure during the activity still exceeded Dutch OEL (MAC). 

In the second (sub) study soil was compacted in an indoor hall using two types of plate compactors: a heavy and a light weight type. For both types of plate compactors 8 data-points (4 with and 4 without atomisation) were generated. For the light plate compactor, atomisation reduced quartz exposure with approximately 88%, where no significant reduction could be observed for the heavy weight type.

During atomisation the average quartz exposure using the light type of plate compactor was below the OEL. During the use of the heavy plate compactor the average exposure level during the work exceeded the OEL.

For both jackhammers and plate compactors it was concluded that a substantial reduction of the quartz exposure could be achieved by atomization using low water volumes, however average levels of exposure were still high compared to the OEL. Optimization of the atomisation and or additional control measures, e.g. exhaust ventilation mounted on the equipment, might be necessary for a further reduction of exposure. 

MATERIAL AND METHODS

Experimental

The study consisted of two sub-studies, both designed as intervention studies. During both sub-studies, a series of measurements was conducted with water atomisation switched off and another with atomisation switched on. One sub-study (sub-study I) consisted of demolition activities, whereas the other (sub-study II) consisted of compacting soil.

Sub-study I

Two types of electrical powered jackhammers were involved in the demolition sub-study. A heavy (high capacity) jackhammer [Wacker EHB 7, beat frequency 1300 - 2000 b/min, weight approximately 9 kg] was used while cutting concrete slabs. The slabs were placed in an open container of approximately 15 m2, located in a hall of 30 x 25 m and a height of 5.5 m. Doors of the hall were kept close during the experiments. Two hollow cone spray nozzles (BEX Sproci Technick, Rotterdam, The Netherlands) operating at a 5 bar pressure and a flow rate of approximately 0.085 L/min for each nozzle, i.e. in total 0.17 L/min. were mounted to the hammer in the vicinity of the chisel. Water was supplied from a backpack 15 L water tank.

Three workers performed activities during replicates of approx. 34 min (range 32 to 88 min). For each worker two replicates with and without atomisation were taken. Water consumption during the replicates was determined by weighing the water tank prior and following the experiments. A light (low capacity) jackhammer (Hitachi 3/4" Hex. 21/32" Round Skank, beat frequency 3000 b/min, weight approximately 4.8 kg), provided with a single spray nozzle (operating at a 5 bar pressure and water flow rate approx. 0.19 L/min) was used to remove tiles from bathroom walls and floors. The water supply tank was located in the vicinity of the workplace. Water consumption during the replicates was also determined by weighing the water tank prior and following the experiments. The bathrooms (n=4) were located at the first floor of houses that were under reconstruction and had similar dimensions (ground surface area 1.95 m2 , height 2.4 m). Each bathroom bad a door (2.1 x 0.8 m) and a window (0.3 4 0.4 m) that was kept closed during the activities.

However, during the removal of wall tiles without atomisation the window had to be opened after approximately 30 min for better view, as this was obscured by high dust levels. Therefore, this procedure was repeated in each experiment. The total surface area of the wall tiles was about 10.5 m2. Two workers performed two types of activities. Two replicate samples were taken during removal of all wall tiles (+- 10.5 m2) from a single bathroom with and without atomisation and two samples during removal of a part of the floor tiles (+- 0.3 m2) from a single bathroom with and without atomisation. During this study, the surface area treated was kept constant for the different experiments instead of a fixed period of time.

Sub-study II

The experiments for compacting soil using plate compactors were conducted in an indoor examination hall of a training centre for pavement workers. The ground surface area of the hall was approx. 100 x 25 m and the height 5 m, whereas the test field for actual soil compaction was 270 m2. Doors were kept closed during the experiments. The quartz content of the soil was += 31 % (w/w). For two types of plate compactors in total 16 replicates of 30 min were collected: 8 for each type consisting of 4 replicates with and 4 replicates without atomisation. The lightweight type soil compactor (Errut PC 400), weighed 81 kg and had a flat plate and a tamping frequency of 60000 b/min. The heavy weight type was a Wacker DPU 6055 type with a tamping frequency of 4140 b/min. It had a V-type of plate and weighed 449 kg. At all four sides flat spray nozzles were mounted with a spraying angle of 90 — 95°, operating at a pressure of 1 bar. Water was supplied directly from the tap by a hose. The water supply flow rate was set at 1 L/min assuming to result in a flow rate of 0.23 L/min for each spray nozzle.

The study protocol was approved by the Medical Ethics Committee, and all test subjects volunteered to participate in the study. Prior to enter the study the test subjects were informed in writing about object and methods of the study and an informed consent form was completed. Prior to each replicate the test subjects were provided with clean work clothing to prevent contamination from previous activities.

Sampling and analysis

Respirable dust samples were collected on Millipore mixed cellulose ester filters (0.8 µm, 25mm) using cyclones (BCIRA-type, SKC, UK) as sampling heads in combination with Gillian Gilair constant flow pumps at a flow rate of 1.9 L/min. The pump flow was checked prior to and following the sampling using a calibrated rotameter. Blank field samples were collected during each sampling period in an uncontaminated room close to the test sites.

The filters were weighed prior and post sampling and sent to an external laboratory (Miljo Kemi, Galtren, Denmark) to determine the content of crystalline quartz in the respirable dust. The analysis was performed by IR-photo ionisation detection (wavelength 700-80 cm -1) according to NIOSH method 7602, adjusted by the laboratory. The filters were incinerated during 24 hours at 200 °C followed by a period of 48 hours at 370 °C. After incineration the ash was mixed potassium bromide in a mortar and pelletised. The limit of quantification was 2 µg (CV 10-20%), resulting in a LOQ for a 30 min sampling period of 0.018 mg/m3,

Material samples, collected at each test site, were also analysed for quartz. However, the incineration step was skipped during the sample preparation

In addition, the test subjects were provided with a DataRAM (Type 1000 MLE Personal Dataram, MIE Inc, USA) to assess real-time dust concentrations. This is a direct-reading aerosol monitor based on detection by (IR) light scattering that responds to particles in the range of 0.1-10 µm with a dynamic range of 0.001 to 400 mg/m3. Sampling intervals of 1 s were chosen. The DataRAM is calibrated with Arizona Road Dust and not with dust from construction sites, so the results cannot be considered representative for respirable construction dust. Therefore the results may be used for mutual comparison only.

After collection the logged data were read into a personal computer using an EXCEL (Microsoft) spread sheet program. Temperature and relative humidity were monitored over each min during the sampling periods, using a Vaisala HMP 31 UT probe in combination with a Grant Squirrel 1201 datalogger. During the experiments material samples were collected for the determination of the quartz content.  Additionally the moisture content of the soil was determined in Sub-study II by comparison of sample weights prior and after period of residence of minimum 5 hours in an exsiccactor. Samples were collected by scooping some soil after compaction (up to a depth of 2 cm).

All observations, raw data and the results of the chemical analysis were transferred to an EXCEL spreadsheet. Descriptive statistics were performed, analytical results reported as below LOQ were substituted by values 1/2 LOQ. For datasets consisting of results for the same persons with and without atomisation t-tests for paired samples were performed. For all other datasets t-tests for means with unequal variances were used. A significance level of p < 0.05 was used. Pearson correlation was calculated between quartz concentration and average dust concentrations as determined by the real-time dust monitoring. Reduction of exposure was calculated using the arithmetic means of either quartz concentrations or real-time dust concentrations.

Reduction (%) = [1 - 1/(AM atomiser off/ AM atomiser on)] * 100 (1)
RESULTS

Test conditions

The experiments during demolition using the heavy type of jackhammer were conducted on two consecutive days where temperature and RH were 12°C +-1 and 66% +- 2.5 respectively. The quartz content of the slabs was approximately 16% (w/w).

The experiments with the light type of hammer were performed within a three weeks period on 4 separate days. The time needed to complete the job differed significantly (p = 0.024) between the experiments with (average 60 min) and without atomisation (average 81 min).

The mean temperature ranged from 14 to 18 °C and the RH from 61 to 76%. During the experiments both wind speed (4 m/s) and direction (SSW) were almost similar for all days. The quartz content of the scrap was approx. 14% (w/w), whereas the quartz content of the floor below the tiles was approx. 15% (w/w).

The experiments during soil compacting were performed during a four weeks period on 6 separate days. Average temperatures during the experiments ranged from 20 to 31 °C and RH varied from  54 to 72%. Prior to the experiment two soil samples were taken. The moisture content of the soil was of 0.60 and 0.53%, respectively. During four days a soil sample was taken at the end of the experiments for the heavy type of compactor (atomisation on). The moisture percentage of the soil ranged from 1.5 to 1.9%. For two occasions this could be compared with the results of samples taken prior to the experiments, yielding an increase of 0.94 and 12%.

Exposure 

Gravimetrical analysis of the filters revealed many observations below the LOD, therefore no reliable respirable dust concentrations could be obtained. A summary of the results of both the respirable quartz exposure sampling and real-time dust monitoring are presented in Table I. During the activities the TWA exposures to quartz ranged from < LOQ to 1.747 mg/m3. The highest exposures were observed during the use of the light jackhammer without atomisation, where tiles were removed in (very) small bath rooms, whereas the lowest exposures were observed during the use of the light plate compactor with the atomiser on. Only 5 out of 36 samples showed an average concentration below the Dutch exposure limit (0.075 mg/m3 TWA 8 hr). Average dust concentrations as determined by real-time air monitoring ranged from 0.5 to 87.6 mg/m3. The highest concentrations were also observed during the use of the light jackhammer. For all 36 samples the correlation between quartz exposure and average 'dust' concentration as determined by real-time air monitoring was rP = 0.78, but ranged between rP = 0.53 (for the heavy type of soil compactor) and rP = 0.86 (for the light type of soil compactor).

Arithmetic mean quartz concentrations tor all types of equipment during atomisation switched on and off are illustrated in Fig. 1. Except for the heavy type of plate compactor (CH) the mean quartz concentration differed significantly between the series with atomisation on and off.

Table II summarises the effectiveness of atomisation for the different types of equipment. Based on personal exposure sampling for quartz, the average reduction of exposure ranged from 64 to 88%, whereas reduction based on ‘real-time’ dust concentrations ranged from 73 - 87%. Since no significant differences for concentrations could be observed between atomisation on/off a reduction factor for the heavy type of plate compactor (CH) could be derived only for 'real-time' dust concentrations (56%).
