Abstract Crude oil may cause adverse dermal effects therefore dermal exposure is an exposure route of concern. Galea et al. (2014b) reported on a study comparing recovery (wipe) and interception (cotton glove) dermal sampling methods. The authors concluded that both methods were suitable for assessing dermal exposure to oil-based drilling fluids and crude oil but that glove samplers may overestimate the amount of fluid transferred to the skin. We describe a study which aimed to further evaluate the wipe sampling method to assess dermal exposure to crude oil, with this assessment including extended sample storage periods and sampling efficiency tests being undertaken at environmental conditions to mimic those typical of outdoor conditions in Saudi Arabia. The wipe sampling method was then used to assess the laboratory technicians’ actual exposure to crude oil during typical petroleum laboratory tasks. Overall, acceptable storage efficiencies up to 54 days were reported with results suggesting storage stability over time. Sampling efficiencies were also reported to be satisfactory at both ambient and elevated temperature and relative humidity environmental conditions for surrogate skin spiked with known masses of crude oil and left up to 4 h prior to wiping, though there was an indication of reduced sampling efficiency over time. Nineteen petroleum laboratory technicians provided a total of 35 pre- and 35 post-activity paired hand wipe samples. Ninety-three percent of the pre-exposure paired hand wipes were less than the analytical limit of detection (LOD), whereas 46% of the post-activity paired hand wipes were less than the LOD. The geometric mean paired post-activity wipe sample measurement was 3.09 µg cm−2 (range 1.76–35.4 µg cm−2). It was considered that dermal exposure most frequently occurred through direct contact with the crude oil (emission) or via deposition. The findings of this study suggest that the wipe sampling method is satisfactory in quantifying laboratory technicians’ dermal exposure to crude oil. It is therefore considered that this wipe sampling method may be suitable to quantify dermal exposure to crude oil for other petroleum workers. crude oil, dermal exposure, laboratory workers, recovery sampling method Introduction Crude oil is a mixture of a wide variety of constituents. It consists primarily of hydrocarbons and contains hundreds of substances including benzene, chromium, iron, mercury, nickel, nitrogen, oxygen, sulphur, toluene and xylene. The main dermal health effect associated with exposure to crude oil is irritation. Prolonged skin contact with crude oil may cause skin reddening, edema and burning of the skin (U.S. National Library of Medicine, 2017), as well as dryness, irritation and hyperkeratosis (ATSDR, 1999). Dermal exposure to crude oil is therefore an exposure route of concern to workers. To our knowledge, there have been no published studies reporting measurements for dermal exposure to crude oil, although there is some published literature reporting dermal exposure to heavy fuel oils (HFO), which are produced in oil refineries from crude oil (Christopher et al., 2011), as well as consumer exposure to diesel and lubricating oils (Galea et al., 2014a). The collection of samples to quantify dermal exposure in this sector would aid in the identification of exposure scenarios where dermal exposure to crude oil is of particular concern, as well as assist in the evaluation of any risk management measures introduced to reduce exposure. Previous laboratory validations identified that Hypaclean® Clinical wipes (13 × 13 cm), moist in 70% mass percentage of isopropyl alcohol, were a suitable material to assess drilling fluids and crude oil exposure from the hands of workers (Galea et al., 2014b) and also HFO (Christopher et al., 2011). One of the conclusions in Galea et al. (2014b) was that field evaluation was warranted to confirm their appropriateness and suitability in the working environment. This article describes a further study to evaluate the wipe sampling method, including an assessment of longer term storage stability and sampling efficiency testing under environmental conditions typical of the outdoor working environment in Saudi Arabia. A field measurement campaign involving petroleum laboratory workers who handle crude oil samples (amongst other products) was undertaken using the wipe sampling method to determine their dermal exposure and the suitability of the protective gloves worn during their exposure activity, this being laboratory analysis. It was originally planned that production workers who may work both in indoor and outdoor settings would be involved in the field measurement campaign, but this was not possible to undertake within project timescales. Methodology Laboratory validation study The International Organization for Standardization (ISO) describes the requirements against which sampling methods for determining dermal exposure need to be assessed, which include assessing the method’s sampling efficiency, recovery efficiency, and sample stability (ISO, 2011). The methodology described in Galea et al. (2014b) was broadly applied in this further validation, although for completeness the methods and new additions are summarised below. A representative sample of crude oil was supplied by the company for use in the laboratory validation tests. The vapour pressure of the crude oil samples ranged from 0.67 to 1.3 kPa (5–10 mmHg) at room temperature. Suitable markers of the crude oil were identified using gas chromatography-mass spectrometry (GC/MS) to allow for calibration and quantification of exposure. The recovery efficiency is a measure of how well the analytical method can recover the crude oil marker from the sampling medium (ISO, 2011). Spiked samples were prepared by accurately weighing aliquots of the crude oil onto the wipes. Wipes were spiked (pipetted) at three different levels (three repeats for each) to represent ‘low (~10 mg), medium (~40 mg) and high’ (74–84 mg) levels of contamination. Spiked wipes were placed inside cone capped 30 ml glass jars. The spikes were left for ~1 h then desorbed. To assess storage stability, spiked wipes were prepared by pipetting known amounts of crude oil (range 2–88 mg) onto the wipes, which were subsequently stored in 30 ml glass jars with solvent suitable cone caps and the caps sealed with Parafilm®. Three different spiking levels were used (three repeats for each). The jars were stored at room temperature for 0, 7, 14, and 28 days prior to analysis. Due to delays in receiving the field survey samples, additional spiked wipes were prepared to reflect the longest period between sample collection and analysis (i.e. 54 days). Pig trotters were used as a surrogate for human skin to determine the sampling efficiency. The pig trotters were spiked at two levels considered to represent likely dermal contamination based on expert judgement (range 16–92 mg; low/medium: ≤30 mg and high: ≥50 mg). Three repeats were undertaken for each loading (and time period). Crude oil was placed in a septum capped glass vial. A capillary tube, open at both ends, was inserted through the septa in the cap and the initial weight recorded. The capillary tube was used to dispense spots of the crude oil onto the trotter. After dispensing, the capillary tube was returned to the same vial and the final weight recorded (final weight). The amount spiked onto the trotter was calculated as the difference between the initial and the final weights of the vial with tube. The pig trotters were then left for set periods of time (10–15 min, 2, and 4 h) before wiping to reflect likely time periods between hand washing. A total of six samples were therefore collected per time period. A standardized wiping pattern was employed. This consisted of five horizontal and five vertical wipes across the spiked surface of the upper surface of the trotter (estimated to be 36 cm2), followed by a wipe in the clockwise direction. This procedure was repeated with two different wipes, with the wipes being stored together for analysis. Two researchers carried out all the sampling efficiency trials. The researchers were experienced in such trials and fully trained on the wipe sampling protocol. It is considered that the researchers applied no significant differences in pressure during the wiping trials. At the time of undertaking the laboratory evaluation, it was intended that a field measurement campaign would be undertaken in Saudi Arabia involving both petroleum laboratory workers and plant operators working outdoors. Sampling efficiency tests were therefore undertaken at ambient laboratory conditions (temperature ranging from 21 to 24°C, relative humidity (RH) ranging from 28 to 42%) and also elevated temperature (45°C) and RH (60–80%) to reflect typical outdoor environmental conditions typically encountered in Damman, Saudi Arabia (World Weather and Climate Information, 2017). The elevated temperature and RH testing was performed in a climatic chamber in accordance with BS EN-IEC 60068-2-78: 2002. Six samples were again collected per time period. Field sampling campaign The occupational hygienists responsible for the collection of the wipe sample measurements and contextual information successfully completed a dermal exposure assessment workshop prior to the start of the field measurement campaign. This workshop included ‘Hands-on’ training sessions on the collection of measurements using the wipe sampling technique to be employed. Exposure measurements were collected from petroleum laboratory technicians who worked in the Crude Unit of the Laboratory. Their laboratory analysis activity involved the handling and analysis of crude oil samples and included a range of tasks. Dermal samples were collected from the palm and dorsal regions of the technicians’ hands. Prior to commencing work, technicians were asked to thoroughly wash and dry their hands following a standardised hand washing protocol (NHS, 2007). Pre-exposure hand wipes were collected to capture any residual background levels of exposure. The technician then undertook the task, working as they would normally. Upon completion of the activity, post-exposure wipe samples were collected prior to the participants washing their hands. A standardized wiping technique was used for both the pre- and post- activity samples. For each hand, five horizontal and five vertical wipes across the surface of the palm of the hand (including the fingers) were made, followed by a wipe in the clockwise direction. This procedure was then repeated for the dorsal region of the hand, with each finger then being wiped, taking care to wipe in between the fingers. This process was repeated with two different wipes, with the wipes being stored together for analysis in the same labelled 30 ml glass jar with solvent suitable cone caps before analysis. The procedure was then repeated for the second hand. The wipes collected from the right hand were classed as one sample, with the wipes collected from the left hand being classed as a separate sample. Field blanks were also collected and comprised of two wipes that were handled in the same way as the exposure samples but were not used to wipe a hand. Two occupational hygienists and one occupational hygiene technician observed the laboratory technician’s activities. Contextual information concerning the technicians work undertaken during the measurement period was recorded using a standardized proforma. The occupational hygienists and technician participated in the aforementioned workshop. The protocol was devised with due consideration of the input parameters required to use DREAM (van Wendel de Joode et al., 2003) and RISKOFDERM (Warren et al., 2006) dermal exposure models. This included recording information on the use of protective gloves, products handled, task information, as well as the occupational hygienist’s assessment of the frequency and intensity of dermal exposure observed for the various pathways of exposure including surface transfer [contact with contaminated surfaces, deposition (contact to airborne material), and emission (direct contact with the substance)]. The field sample collection was carried out at a room temperature of 21 to 22°C. The samples were stored in a refrigerator at 5°C for 4 days before being couriered in a cool box to the laboratory for analysis. Chromatographic conditions, sample preparation, and analysis Each wipe was placed into an individual 30 ml glass jar and 12 ml of dichloromethane pipetted into the wipe vial, ensuring the wipe was completely submersed. The solution was ultrasonicated for 10 min and allowed to stand for 1 h, before being agitated further with a Pasteur pipette. The solution was then filtered into a 2-ml glass vial, glass inserts were used in the vial when filtering was difficult. A Shimadzu GC/MS QP2010S using electron impact (EI) ionization in full scan mode was used for the analysis, fitted with a Phenomonex ZB-Semivolatiles 30 m, 0.25 mm id, 0.25 µm film thickness column. The run time was 35 min. The column temperature program for the samples was as follows: 40°C for 2 min to 320°C at 10°C/min then hold for 5 min. Standard solutions of crude oil were run, together with spiked samples and the instrument conditions adjusted to optimise results. Quality control included (i) the running of a standard repeat after every 10 samples to monitor any calibration drift during the analysis; (ii) preparation of a control sample by a separate analyst to check the calibration standards had been prepared correctly, and (iii) 1 in 10 samples being reanalysed. All calibration standards for this study were prepared from stock solutions of crude oil in dichloromethane. Dilutions of these stocks were used as calibration standards and the weights calculated from the weight of the crude oil. The n-Nonadecane peak from the crude oil (Rt ca 20.5 min) was used as the marker peak in both standards and samples for quantification of the wipes due to its low volatility. In each case, the area under the n-nonandecane peak was integrated to provide the calibration and the results for the samples. The limit of detection (LOD) was 0.52 mg per sample. Data analysis The recovery, sampling, and storage efficiencies were determined by calculating the recovered weights of crude oil spiked on the wipes/surrogate skin compared to the nominal weight of crude oil spiked, expressed as a percentage. Recovery, storage and sampling efficiencies of the wipe sampling methods for the different time periods, spike levels, and environmental conditions for the sampling efficiency tests were compared using analysis of variance (ANOVA), Kruskal–Wallis, or non-parametric Wilcoxon rank-sum tests to assess trends over time. For the field measurement results, these were corrected for the field blank, as well as with the sampling efficiency and storage stability results, as necessary. For samples with surface loadings less than the analytical LOD suitable methods for dealing with these were considered (Croghan and Egeghy, 2003). The decision was subsequently taken to substitute those results less than the analytical LOD with a value of the LOD/√2. This approach is usual practice when there are a high number of values below the LOD (Curwin et al., 2007). The results are expressed as estimates of the exposure mass present in the skin area wiped at the time of sampling (µg cm−2) using the average hand surface for an adult (70 kg is 420 cm2 [US EPA, 2011]). Summary statistics for the dermal exposure results are presented for the right and left hand wipe sample results individually and combined (paired wipe sample results). Using the corresponding contextual information recorded by the occupational hygienists at the time of collection of the wipe sample measurements, regression analysis was undertaken including the categorical variables of interest, namely: filling mixing and loading duration, wiping duration, general ventilation, adequate local exhaust ventilation (LEV), % of both hands contaminated, and emission, deposition, and surface transfer frequency and intensity of exposures, respectively. All statistical analyses were performed using STATA 13 (StataCorp, College Station, TX). Results Laboratory validation study Chromatograms of the supplied crude oil analysed show a typical hydrocarbon series pattern, with components exhibiting a wide range of volatilities. The n-Nonadecane peak was selected for quantification as it was observed that there were losses of the more volatile components quantified. The Hypaclean® Wipes exhibited a range of VOCs components at low concentrations, with only one significant component, di-n-undecylamine (similarity index 87%), being observed in the chromatograms of oil recovered from wipe samples. However, sufficient chromatographic separation was observed between the n-Nonadecane and the residual di-n-undecylamine from the wipe. The calibration curves were considered acceptable [correlation coefficient (R2) >0.99]. The recovery efficiencies (Day 0) and storage stabilities for the crude oil spiked directly onto the wipes for Days 7, 14, 28, and 54 are reported in Table 1. The average recovery efficiency (Day 0) from the wipes using crude oil as a calibration standard was 97.3%, which is considered acceptable and typical for this type of sampling. Table 1. Percentage (%) recovery efficiency (Day 0) and storage stabilities (Days 7–54). Day Mean SD Min Max 0 97.3 5.4 91.1 105.9 7 93.4 17.3 59.1a 106.1 14 103.3 5.4 96.0 111.1 28 94.6 2.9 91.1 99.3 54b 98.3 7.6 91.1 106.2 Day Mean SD Min Max 0 97.3 5.4 91.1 105.9 7 93.4 17.3 59.1a 106.1 14 103.3 5.4 96.0 111.1 28 94.6 2.9 91.1 99.3 54b 98.3 7.6 91.1 106.2 Mean: arithmetic mean; Min: minimum; Max: maximum; SD: standard deviation. aConsidered to be spiking error. bBased on three spikes, all other storage stability periods are based on nine spiked samples. View Large Table 1. Percentage (%) recovery efficiency (Day 0) and storage stabilities (Days 7–54). Day Mean SD Min Max 0 97.3 5.4 91.1 105.9 7 93.4 17.3 59.1a 106.1 14 103.3 5.4 96.0 111.1 28 94.6 2.9 91.1 99.3 54b 98.3 7.6 91.1 106.2 Day Mean SD Min Max 0 97.3 5.4 91.1 105.9 7 93.4 17.3 59.1a 106.1 14 103.3 5.4 96.0 111.1 28 94.6 2.9 91.1 99.3 54b 98.3 7.6 91.1 106.2 Mean: arithmetic mean; Min: minimum; Max: maximum; SD: standard deviation. aConsidered to be spiking error. bBased on three spikes, all other storage stability periods are based on nine spiked samples. View Large The spiked samples show good storage stability for 7, 14, 28 and 54 days (these being 93.4, 103.3, 94.6, and 98.3%, respectively). Two samples from Day 7 showed poor recovery, which explains the high standard deviation. As Days 14, 28 and 54 samples all showed good recoveries, the Day 7 data anomaly is thought to be due to a spiking error rather than losses during storage. A comparison of mean recovery efficiency percentages on each day did not find any statistically significant differences (P = 0.232). The data were normally distributed, except for Day 7, which had a high SD, due to the spiking error. When removing the Day 7 data, there was a significant difference (P = 0.009) between the mean levels; though there was no evidence of a trend over time (P = 0.523). The mean recovery efficiencies for low, medium, and high spiking levels were 100.9, 100.5, and 90.2%, respectively, and were significantly different when testing with ANOVA (P = 0.003). Table 2 presents the sampling efficiency tests for three time periods carried out at both ambient and elevated environmental conditions. Mean sampling efficiencies at both ambient and elevated environmental conditions were 102, 96, and 91% for 10–15 min, 2 h, and 4 h, respectively. No statistical differences were found at mean ambient conditions (P = 0.669), and a statistically significant difference was identified at the elevated (P = 0.001) environmental conditions, with evidence of a declining trend over time (P = 0.001). Sampling efficiency was unchanged between spiking levels at both ambient (P = 0.895) and elevated (P = 0.627) environmental conditions. Table 2. Sampling efficiency (%) of the wipes removing crude oil from pig trotters at ambient and elevated environmental conditions (six repeats per test). Environmental conditions 10–15 min 2 h 4 h Mean SD Min Max Mean SD Min Max Mean SD Min Max Ambienta 98.2 11.3 84.0 113.1 93.5 7.6 84.7 105 93.8 10.7 80.3 109.3 Elevateda 104.8 2.5 101.2 108.6 97.6 8.8 88.2 110.2 87.9 5.9 81.6 96.6 Overall 101.5 8.5 84.0 113.1 95.6 8.1 84.7 110.2 90.9 8.8 80.3 109.3 Environmental conditions 10–15 min 2 h 4 h Mean SD Min Max Mean SD Min Max Mean SD Min Max Ambienta 98.2 11.3 84.0 113.1 93.5 7.6 84.7 105 93.8 10.7 80.3 109.3 Elevateda 104.8 2.5 101.2 108.6 97.6 8.8 88.2 110.2 87.9 5.9 81.6 96.6 Overall 101.5 8.5 84.0 113.1 95.6 8.1 84.7 110.2 90.9 8.8 80.3 109.3 Mean: arithmetic mean; Min: minimum; Max: maximum; SD: standard deviation. aSix repeats per test. View Large Table 2. Sampling efficiency (%) of the wipes removing crude oil from pig trotters at ambient and elevated environmental conditions (six repeats per test). Environmental conditions 10–15 min 2 h 4 h Mean SD Min Max Mean SD Min Max Mean SD Min Max Ambienta 98.2 11.3 84.0 113.1 93.5 7.6 84.7 105 93.8 10.7 80.3 109.3 Elevateda 104.8 2.5 101.2 108.6 97.6 8.8 88.2 110.2 87.9 5.9 81.6 96.6 Overall 101.5 8.5 84.0 113.1 95.6 8.1 84.7 110.2 90.9 8.8 80.3 109.3 Environmental conditions 10–15 min 2 h 4 h Mean SD Min Max Mean SD Min Max Mean SD Min Max Ambienta 98.2 11.3 84.0 113.1 93.5 7.6 84.7 105 93.8 10.7 80.3 109.3 Elevateda 104.8 2.5 101.2 108.6 97.6 8.8 88.2 110.2 87.9 5.9 81.6 96.6 Overall 101.5 8.5 84.0 113.1 95.6 8.1 84.7 110.2 90.9 8.8 80.3 109.3 Mean: arithmetic mean; Min: minimum; Max: maximum; SD: standard deviation. aSix repeats per test. View Large However, when combining the two environmental conditions, three time periods and spiking levels in a linear regression model (as categorical variables), there was no significant overall difference in sampling efficiency between the ambient and elevated samples (P = 0.584). Nevertheless, the 4-h period still had significantly lower efficiencies (β = −10.6; P = 0.005) than the 10–15 min period, with a non-significant decrease in the 2-h period (β = −5.9; P = 0.104) compared to the 10–15-min period. There was no apparent difference in sampling efficiency based on low/medium or high spike levels (P = 0.656). Field dermal exposure measurement survey The laboratories provide analytical services and technical support including laboratory analyses, field testing, sampling and testing related to Southern/Central Arabia oilfield and support facilities, gas plants and pipelines and oil operations. The laboratories receive and analyse various samples from the petroleum facilities, processing on average 5000 samples monthly and 156 000 tests annually. The laboratory technicians work 8-h shifts (day or afternoon shift). During their shifts, they undertake various laboratory analysis tasks on crude oil samples including analysis of: octane number, flash point, cloud point; chloride, sodium and metals, hydrogen sulfide, sulphur, nitrogen, mercaptan and lead, penetration and viscosity tests etc. In addition to crude oil, laboratory analysis of compounds including diesel, asphalt, jet fuel, kerosene, naphtha, gasoline and kerosene takes place. Laboratory technicians perform physical tests, electrochemical, chemical, chromatographic, spectroscopic, and titrimetric analyses. This involves the preparation of samples, standards and of reagents which include pouring and transferring of samples and reagents using pipet and or manual pump; mixing and manual shaking of liquids; dispensing of solvent reagents. The conduct of the various analyses entail the use of a chemical fume hood, common laboratory apparatus such as water bath, centrifuge, oven, titration device, mixer, gas chromatograph, IR spectrometer and X-ray fluoroscope. Rags are used for cleaning and housekeeping of the work bench. Washing and rinsing of glassware involves the use of toluene and acetone for drying. All the laboratory technicians wore Ansell TouchNTuff 92–600 single use nitrile chemical protection gloves. The nitrile gloves are replaced immediately if they become contaminated with crude oil, otherwise they are removed and disposed of once the task is completed. In addition, the laboratory technicians wore safety glasses, fire-resistant clothing/apron and safety shoes. The field wipe sample measurement campaign took place during 19–23 February 2017, with the samples being collected from the laboratory technicians during the course of their normal work laboratory analysis activity. Nineteen laboratory technicians participated, providing in total 35 pre- and 35 post-activity paired samples. Several technicians provided multiple activity related samples, with two technicians providing three-activity related sample sets, nine technicians providing two activity-related sample sets and one technician providing five sets of activity-based measurements. For the wipe samples which were collected prior to the laboratory work being carried out, 33 (93%) were less than the LOD for n-Nonadecane. The two wipe samples with detectable concentrations reported concentrations of 7.3 and 8.06 µg cm−2. It was assumed that the pre-wipes collected before work took place resulted in the hand being clean (due to the high wipe sampling efficiencies reported in the validation study). Therefore, the data analysis was based on the post-wipe results (i.e. no adjustments were made for the pre-wipe value). Sixteen post-activity paired wipe samples were <LOD (46%), with detectable masses being detected on the remaining paired wipes. There was no difference in the concentrations on the left or right hand (P = 0.783). The results expressed as estimates of the exposure mass present in the skin area wiped at the time of sampling (µg cm−2) are presented in Table 3. Table 3. Post-laboratory analysis activity wipe sample results (µg cm−2) quantified using crude oil (n-nonadecane marker peak) (n = 35). % µg cm−2 <LOD Min Max 95 %ile GM GSDa AM SD Left hand wipe 51.4 0.88 23.1 17.5 1.54 2.3 2.63 4.6 Right hand wipe 54.3 0.88 17.3 12.3 1.49 2.2 2.39 3.6 Paired wipes 45.7 1.76 35.4 34.8 3.09 2.2 5.02 8.0 % µg cm−2 <LOD Min Max 95 %ile GM GSDa AM SD Left hand wipe 51.4 0.88 23.1 17.5 1.54 2.3 2.63 4.6 Right hand wipe 54.3 0.88 17.3 12.3 1.49 2.2 2.39 3.6 Paired wipes 45.7 1.76 35.4 34.8 3.09 2.2 5.02 8.0 AM, arithmetic mean; GM, geometric mean; GSD, geometric standard deviation; Min, minimum; Max, maximum; 95% ile, 95th percentile; SD, standard deviation. aGSD is dimensionless. View Large Table 3. Post-laboratory analysis activity wipe sample results (µg cm−2) quantified using crude oil (n-nonadecane marker peak) (n = 35). % µg cm−2 <LOD Min Max 95 %ile GM GSDa AM SD Left hand wipe 51.4 0.88 23.1 17.5 1.54 2.3 2.63 4.6 Right hand wipe 54.3 0.88 17.3 12.3 1.49 2.2 2.39 3.6 Paired wipes 45.7 1.76 35.4 34.8 3.09 2.2 5.02 8.0 % µg cm−2 <LOD Min Max 95 %ile GM GSDa AM SD Left hand wipe 51.4 0.88 23.1 17.5 1.54 2.3 2.63 4.6 Right hand wipe 54.3 0.88 17.3 12.3 1.49 2.2 2.39 3.6 Paired wipes 45.7 1.76 35.4 34.8 3.09 2.2 5.02 8.0 AM, arithmetic mean; GM, geometric mean; GSD, geometric standard deviation; Min, minimum; Max, maximum; 95% ile, 95th percentile; SD, standard deviation. aGSD is dimensionless. View Large The maximum paired post activity measurement result were obtained from a technician who processed many hydrocarbon samples during that day. This technician only replaced their disposable gloves once they had completed all the sample analyses assigned to them. It was considered that there was prolonged contact with contaminated gloves. The laboratory workers undertook various tasks within their laboratory analysis activity which included filling/mixing and loading; wiping; or mechanical treatment (no dispersion, spraying or immersion activities were reported). The majority of these tasks reportedly lasted 5 min in duration (n = 24; 68.5%), although in some instances the tasks reportedly lasted 1 min (n = 1; 2.9%), 1–2 min (n = 3; 8.6%), 5–10 min (n = 4; 11.4%), with three tasks reportedly lasting 5 h in duration (8.6%). Wiping tasks were reported in all instances and typically involved wiping apparatus and the work bench used after the analysis and general housekeeping, with three instances reportedly lasting 1 min in duration (8.6%) and 1–2 min in duration for the remaining 32 instances (91.4%). Less information on mechanical treatment tasks was gathered; however, it was reported that these tasks included use of the titration system and flash point apparatus, with mechanical tasks lasting 2 min in duration, when duration was reported (n = 2). It was reported in all instances that both the right and left hands (i.e. gloves) were observed to be contaminated through direct contact with crude oil. It was also reported that 10% of the hand surface area was contaminated in 27 of the measured activities (77%), whereas 40% of the hand surface area was contaminated for the other 8 activities measured (23%). Table 4 summarises the reported frequency of exposure and intensity of exposure for emission (direct contact with the substance), deposition (contact to airborne material) and surface transfer (contact with contaminated surfaces) dermal exposure pathways. Table 4. Frequency and intensity of exposure for emission, deposition, and surface transfer pathways for the 35 tasks observed. Frequency of exposure N (%) Intensity of exposure N (%) Unlikely Occasionally Repeatedly Almost constantly Small amount Medium amount Large amount Emission 23 (66) 9 (26) 2 (6) 1 (3) 32 (91) 3 (9) 0 Deposition 26 (74) 7 (20) 2 (6) 0 (0) 33 (94) 2 (6) 0 Surface transfer 29 (83) 3 (9) 2 (6) 1 (3) 32 (91) 3 (9) 0 Frequency of exposure N (%) Intensity of exposure N (%) Unlikely Occasionally Repeatedly Almost constantly Small amount Medium amount Large amount Emission 23 (66) 9 (26) 2 (6) 1 (3) 32 (91) 3 (9) 0 Deposition 26 (74) 7 (20) 2 (6) 0 (0) 33 (94) 2 (6) 0 Surface transfer 29 (83) 3 (9) 2 (6) 1 (3) 32 (91) 3 (9) 0 Unlikely, <1% of task duration; occasionally, 1–<10% of task duration; repeatedly, 10–<50% of task duration; almost constantly, >50% of task duration; small amount, <10% of hands; medium amount, 10–50% of hands; large amount, >50% of hands. View Large Table 4. Frequency and intensity of exposure for emission, deposition, and surface transfer pathways for the 35 tasks observed. Frequency of exposure N (%) Intensity of exposure N (%) Unlikely Occasionally Repeatedly Almost constantly Small amount Medium amount Large amount Emission 23 (66) 9 (26) 2 (6) 1 (3) 32 (91) 3 (9) 0 Deposition 26 (74) 7 (20) 2 (6) 0 (0) 33 (94) 2 (6) 0 Surface transfer 29 (83) 3 (9) 2 (6) 1 (3) 32 (91) 3 (9) 0 Frequency of exposure N (%) Intensity of exposure N (%) Unlikely Occasionally Repeatedly Almost constantly Small amount Medium amount Large amount Emission 23 (66) 9 (26) 2 (6) 1 (3) 32 (91) 3 (9) 0 Deposition 26 (74) 7 (20) 2 (6) 0 (0) 33 (94) 2 (6) 0 Surface transfer 29 (83) 3 (9) 2 (6) 1 (3) 32 (91) 3 (9) 0 Unlikely, <1% of task duration; occasionally, 1–<10% of task duration; repeatedly, 10–<50% of task duration; almost constantly, >50% of task duration; small amount, <10% of hands; medium amount, 10–50% of hands; large amount, >50% of hands. View Large Emission and deposition frequencies were borderline significant when the regression model was run with all variables included to assess effects on the concentration of n-Nonadecane. A log-linear model was also run. This identified that ‘occasional’ emission and deposition frequencies lead to an average of 2.89 (P = 0.018) and 2.58 (P = 0.033) greater n-Nonadecane exposures, respectively, compared to the ‘unlikely’ frequency. Although the ‘none’ category was significantly associated with reduced concentrations when compared with ‘Normal or Good’ (P = 0.040) ventilation, the ‘none’ category was only recorded once. To confirm, there were no statistically significant associations of n-Nonadecane concentrations with filling, mixing and loading duration, wiping duration, adequate LEV, % of both hands contaminated, the surface transfer frequency and intensity of exposure, or either of the deposition or emission frequency or intensity of exposures. The log-linear model was deemed a better fit using the Bayesian Information Criterion. Discussion In this article, we describe a study that aimed to evaluate further a wipe sampling method to assess crude oil dermal exposure, as described by Galea et al. (2014b), as well as to investigate petroleum laboratory technicians’ dermal exposure during laboratory activities which involved a range of tasks involving the handling of crude oil and other related products. The collected wipe samples were quantified using calibration standards derived from samples of the extracted crude oil. The vapour pressure of the crude oil samples ranged from 0.67 to 1.3 kPa (5–10 mmHg) at room temperature. Overall, acceptable storage efficiencies up to 54 days were reported with the results suggesting that storage stability is stable over time. In the event of samples being stored for periods extending beyond 54 days, further work would be necessary to confirm the stability of these. Sampling efficiencies at both ambient and elevated temperature and RH environmental were also reported to be satisfactory (within the IOM in-house criterion of 85–115%), though with some indication of declines with time. Clearly this does not cover all environmental conditions under which samples may be collected, but does demonstrate that the methodology is satisfactory under more extreme conditions. Galea et al. (2014b) used a different marker to quantify for crude oil and the crude oil used in the current validation was obtained from different oil fields. The 1 day and 1 week storage stabilities reported in the current study compared favourably with Galea et al. (2014b); however, better storage stabilities were reported at 2 weeks for the current study (103 versus 83%). Overall the sample efficiency in the current study was higher than Galea et al. (2014b) at both 15 min (98 versus 92%) and 2-h time periods (94 versus 83%). Crude oil is a complex mixture of hydrocarbons and other chemicals. The composition varies widely depending on where and how the petroleum was formed. Hydrocarbons are the most abundant component of crude oil, accounting for 50–98% of the total composition (although the majority contain higher relative amounts of hydrocarbons). Hydrocarbons are typically present in petroleum at the following percentages: paraffins (15–60%), naphthenes (30–60%), aromatics (3–30%), with asphaltics making up the remainder (e.g. Petroleum.co.uk, 2015; Helmenstine, 2018). (N-Nonadecane falls within the paraffin category, as paraffin composition can represent up to 45%). Selection of suitable markers to quantify exposure was based on review of the chromatograms of the bulk crude oil samples (see the Online Supplementary Material). It is evident from the chromatograms used in the current study and Galea et al. (2014b) that the composition of crude oil samples differ significantly. This emphasizes the importance of highlighting that others, when undertaking similar future work, cannot automatically apply the markers previously used. It is important for researchers to ensure that they take due care and attention to ensure that they identify the most appropriate marker for use as the surrogate of the complex crude oil for their given situation. In addition, potential differences in permeation of the crude oil marker used in this study through deceased porcine skin to that of living human skin is unknown. However, in Galea et al. (2014b), it was deemed unethical to carry out sampling efficiency tests using human volunteers and therefore surrogate skin options were used in both studies. As to the author’s knowledge, this is the first study to quantify workers’ dermal exposure to crude oil rand so it is not possible to compare the results with previous studies. Christopher et al. (2007) reported laboratory technicians’ hand exposure to HFO as being < LOD, although it is important to highlight that this was based on two samples. It should also be noted that although these laboratory technicians undertook similar activities to those reported in the current study, they wore two pairs of gloves (cotton and nitrile), rather than the one pair (nitrile) worn in this study. Disposable nitrile gloves were worn by the petroleum laboratory technicians and were disposed of after use. The protective gloves worn are reportedly suitable for laboratory applications and industries including the chemical and petrochemical industryhowever the results from this study would suggest otherwise given that over half of the post-activity paired hand wipe measurements were above the LOD. It was observed in all the activities measured that both hands were contaminated through direct contact with crude oil, although no significant differences in exposure were reported, irrespective of the percentage of hand contamination reported. The exposure reported on the wipes was due to hands becoming contaminated with crude oil during sample preparation, such as filling or pouring, transferring, and measuring of crude oil. It was observed that the samples tend to spill during these tasks and the technicians did not always replace their gloves until the analysis was completed, thus permeation occurred over time with prolonged contact. Another instance occurred during manual pipetting of crude sample using a rubber bulb: crude oil contaminated the surface of the gloves and this took a few minutes of contact causing permeation. These cases are exacerbated by technicians wearing gloves nor providing the level of protection as reported. It was considered that dermal exposure most frequently occurred through direct contact with the substance (emission) or contact to airborne material (deposition). Therefore, changing the nature of tasks to reduce direct contact with the substance or airborne material would theoretically lead to a reduction in exposure. The findings of this study suggest that the wipe sampling method is satisfactory in quantifying laboratory technicians’ dermal exposure to crude oil. It was not possible to apply the wipe sampling method to plant operators and it would be beneficial to trial this approach for such workers. Additional field surveys would help establish the method’s suitability for these workers and also identify if any sampling efficiency trials are necessary at higher loading levels than reported in the current study. The collection of such data would also allow for the identification of exposure scenarios where dermal exposure to crude oil is of particular concern, as well as assist in the evaluation of any risk management measures introduced to reduce exposure. Supplementary Data Supplementary data are available at Annals of Work Exposures and Health online. Funding This project was funded by Saudi Aramco (contract number 6600035583). Two of the authors (A.M.A. and J.L.L.) are employed by Saudi Aramco. Acknowledgements We would like to thank Hilary Cowie (IOM) for her helpful comments and suggestions on earlier version of this manuscript. Thanks also to Prof. John Cherrie (IOM and Heriot Watt University) for delivering the dermal exposure assessment workshop prior to the field measurement campaign and Dr Anne Sleeuwenhoek (IOM) for her assistance with preparing the field sampling protocol and some of the workshop training materials. The environmental chamber used for the laboratory validation trials conducted at elevated temperature and relative humidity was located at Selex ES Ltd, Crew Toll, Edinburgh, UK. All authors have made substantial intellectual contributions to the design and execution of the research and in the writing, and subsequent revisions, of this manuscript. Its contents, including any opinions and/or conclusions expressed, are solely those of the authors. References ATSDR . ( 1999 ) Toxicological profile for total petroleum hydrocarbons (TPH) . US Department of Health and Human Service. Agency for Toxic Substances and Disease Registry . Available at http://www.atsdr.cdc.gov/ToxProfiles/tp123.pdf. Accessed 13 November 2013 . BS EN-IEC 60068-2-78 . ( 2002 ) Environmental testing. Test methods. Test Cab. Damp heat, steady state . London : British Standards Institution . Available at http://www.iom-world.org/pubs/IOM_TM0705.pdf. Accessed 11 May 2018. Christopher Y , van Tongeren M , Cowie H , et al. ( 2007 ) Occupational dermal exposure to heavy fuel oil . IOM Technical Memorandum TM/07/05 . Croghan CW , Egeghy PP. (2003) Methods of dealing with values below the limit of detection using SAS . Available at http://analytics.ncsu.edu/sesug/2003/SD08-Croghan.pdf. Accessed 11 May 2018. Curwin BD , Hein MJ , Sanderson WT , et al. ( 2007 ) Urinary pesticide concentrations among children, mothers and fathers living in farm and non-farm households in iowa . Ann Occup Hyg ; 51 : 53 – 65 . Google Scholar CrossRef Search ADS PubMed Galea KS , Davis A , Todd D , et al. ( 2014a ) Dermal exposure from transfer of lubricants and fuels by consumers . J Expo Sci Environ Epidemiol ; 24 : 665 – 72 . Google Scholar CrossRef Search ADS Galea KS , McGonagle C , Sleeuwenhoek A , et al. ( 2014b ) Validation and comparison of two sampling methods to assess dermal exposure to drilling fluids and crude oil . Ann Occup Hyg ; 58 : 591 – 600 . Helmenstine AM . ( 2018 ) Chemical composition of petroleum . Available at https://www.thoughtco.com/chemical-composition-of-petroleum-607575. Accessed 26 March 2018 . ISO . ( 2011 ) ISO/TR 14294:2011 Workplace atmospheres - measurement of dermal exposure - principles and methods . Geneva, Switzerland: International Organisation for Standardisation . NHS . ( 2007 ) Hand washing technique with soap and water . Adapted from World Health Organisation Guidelines on hand hygiene in health care . Available at: https://www.hey.nhs.uk/patient-leaflet/hand-hygiene-information/. Accessed 11 May 2018. Petroleum.co.uk . ( 2015 ) Petroleum composition . Available at http://www.petroleum.co.uk/composition. Accessed 26 March 2018 . U.S. EPA . ( 2011 ) Exposure factors handbook 2011 edition (final report) . Washington, DC : U.S. Environmental Protection Agency . U.S. National Library of Medicine . ( 2017 ) Crude oil . Available at http://toxtown.nlm.nih.gov/text_version/chemicals.php?id=73. Accessed 19 June 2017 . Warren ND , Marquart H , Christopher Y , et al. ( 2006 ) Task-based dermal exposure models for regulatory risk assessment . Ann Occup Hyg ; 50 : 491 – 503 . Google Scholar PubMed van Wendel de Joode B , Brouwer DH , Vermeulen R , et al. ( 2003 ) DREAM: a method for semi-quantitative dermal exposure assessment . Ann Occup Hyg ; 47 : 71 – 87 . Google Scholar PubMed World Weather and Climate Information . ( 2017 ) Average minimum and maximum temperature in Dammam (celsius) . Available at https://weather-and-climate.com/average-monthly-min-max-Temperature,dammam,Saudi-Arabi. Accessed 19 June 2017 . © The Author(s) 2018. Published by Oxford University Press on behalf of the British Occupational Hygiene Society. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
Annals of Work Exposures and Health (formerly Annals Of Occupational Hygiene) – Oxford University Press
Published: May 21, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera