Top Navigation Bar

Article Request Page ASABE Journal Article

Evaluation of Low-Cost Hydrogen Sulfide Monitors for Use in Livestock Production

J. M. Beswick-Honn, T. M. Peters, T. R. Anthony


Published in Journal of Agricultural Safety and Health 23(4): 265-279 (doi: 10.13031/jash.12530). Copyright 2017 American Society of Agricultural and Biological Engineers.


Submitted for review in July 2017 as manuscript number JASH 12530; approved for publication by the Ergonomics, Safety, & Health Community of ASABE in October 2017.

The authors are Jessica M. Beswick-Honn, Research Assistant, Thomas M. Peters, Professor, and T. Renée Anthony, ASABE Member, Associate Professor, Department of Occupational and Environmental Health, University of Iowa, Iowa City, Iowa. Corresponding author: T. Renée Anthony, CPHB, 145 N. Riverside Drive, Iowa City, IA 52246; phone: 319-335-4429; e-mail: renee-anthony@uiowa.edu.

Abstract.  Direct-reading gas monitors warn workers of the risk of potentially fatal hydrogen sulfide (H2S) exposures that may arise during manure handling. Low-cost, low-maintenance H2S monitors are available from many manufacturers, but differences in their features and performance make selection challenging for farmers. Moreover, little information is available on the practical maintenance and performance of these devices in agricultural environments. The objective of this study was to provide information to agricultural workers to aid in the selection, maintenance, and use of low-cost H2S monitors. This laboratory study evaluated the performance of several low-cost monitors over a simulated period of use of one year in a swine barn. Four models were exposed to H2S concentrations of 1 to 10 ppm over 18 weeks to examine the drift in reported concentration and changes in the alarm reaction time. Over the simulated barn year, the performance of alarm-only monitors declined faster than that of monitors displaying the H2S concentration. Of concern was the high-level (20 ppm) alarm failures after an equivalent of 139 days (Altair) and 289 days (BW Clip) in a swine barn, well within the monitor’s reported shelf-life. Models displaying concentration exhibited fewer failures but were inaccurate in the displayed concentration when challenged with 20 ppm of H2S. The T40 Rattler provided consistently higher readings (+2.3 ppm), and the Pac 3500 showed consistently lower readings (-3.4 ppm) when challenged with 20 ppm. This study confirms the need for routine bump tests for these low-cost monitors to ensure that the monitor reacts to the presence of H2S, even if the manufacturer does not recommend this procedure. Most importantly, agricultural workers should inspect and bump test these monitors prior to any potentially high-risk activity, such as manure agitation, pumping, or pressure washing, to ensure that the monitor appropriately detects and warns users.

Keywords.Calibration, Direct-reading monitors, H2S, Hydrogen sulfide monitors, Manure gas, Safety, Sensors.

Agricultural workers continue to be exposed to fatal hydrogen sulfide (H2S) concentrations during manure handling activities (Adekoya and Myers, 1999; Hendrickson et al., 2004; Beaver and Field, 2007; Riedel and Field, 2013). Livestock operations generate and store large quantities of manure, and its anaerobic decomposition generates H2S. When manure is undisturbed, background H2S concentrations in surrounding areas remain low, typically at or below 1 part per million (ppm) (Swestka, 2010; Reeve et al., 2013; Anthony et al., 2015; Guarrasi et al., 2015). However, when manure is agitated during pressure washing or prior to manure pumping, H2S is released, presenting substantial risk of illness or death (Donham et al., 1982). When the manure is agitated to improve the pumping needed to empty storage pits, H2S concentrations can rise within seconds to higher than 500 ppm (Donham et al., 1982) and have been observed well over 1000 ppm (Popendorf, 1991; Fabian-Wheeler et al., 2017).

The smell of H2S is highly recognizable and easily detected at low concentrations (~1 ppm), where its “rotten egg” odor is identifiable. Acute exposures to low concentrations of H2S can lead to headaches, nausea, and dizziness. However, when concentrations reach 100 to 150 ppm and above, olfactory paralysis occurs, preventing detection of H2S by smell. At concentrations ranging from 100 to 500 ppm, exposure can result in neurological symptoms. When the H2S concentration reaches 700 to 1000 ppm, exposure can cause rapid unconsciousness, leading to death with just a few breaths, often called “knock-down” (NRC, 2010).

The Occupational Safety and Health Administration (OSHA) has established permissible exposure limits (PELs) for H2S applicable to workers in general industry, construction, and shipyards (OSHA, 2017). OSHA regulations do not specify H2S exposure limits for agricultural workers, but exposures to H2S in livestock operations would be covered by the OSHA general duty clause, which requires that all workers be protected from recognized hazards, and acutely fatal H2S exposures from livestock manure are well recognized in this industry. The general industry PELs for H2S were established in 1970 and specify that exposures be maintained below a 20 ppm ceiling across a work shift, although OSHA allows a single exposure to reach 50 ppm for up to 10 min when there is no other exposure (0 ppm) during a shift. For construction and shipyard workers, exposures to H2S must be maintained below 10 ppm as a full-shift (8 h) time-weighted average.

More recent consensus exposure limits have been established based on current health hazard evidence, irrespective of occupational sector. The American Conference of Governmental Industrial Hygienists (ACGIH) annually publishes threshold limit values (TLVs) (ACGIH, 2017), and the National Institute of Occupational Safety and Health (NIOSH) periodically revises recommended exposure limits (RELs) to protect workers (NIOSH, 2007). These organizations recommend that short-term exposures not exceed a 10 min ceiling of 10 ppm H2S (NIOSH) and a 15 min average of 5 ppm (ACGIH). In addition, ACGIH recommends that 8 h average H2S exposures be maintained below 1 ppm (ACGIH, 2017) to protect against chronic diseases. NIOSH also establishes exposure criteria for chemicals that are immediately dangerous to life and health (IDLH), defined as a concentration above which exposed workers must rely on supplied-air respirators for protection to escape a hazardous environment (NIOSH, 2007). For H2S, the IDLH is currently 100 ppm, a reduction from the pre-1994 recommendation of 300 ppm.

Hydrogen sulfide can be a major component of manure pit gases, which are liberated from manure during agitation prior to pumping and which can occur in the air above poorly ventilated storage spaces. Indoor manure storage spaces typically meet the criteria for confined spaces because of the high probability of lethal concentrations of H2S combined with limited entry and egress. While U.S. health and safety regulations do not require agricultural workers to conform to either H2S exposure limits or confined space practices, preventing access to areas with high H2S concentrations and monitoring the air quality to ensure safe breathing can protect health and save lives.

Direct-reading gas monitors are useful tools for providing real-time warnings to livestock producers. These monitors can rapidly assess the environment and warn workers in real time to leave areas where the H2S concentration is unsafe. Most commonly, direct-reading monitors for H2S incorporate electrochemical sensors that detect H2S through a reaction between the airborne gas and an electrolyte, generating an electrical signal that corresponds to the gas concentration in ppm. The monitor detects the signal voltage, translates it into a concentration value, and triggers an audible alarm if the concentration exceeds a preset threshold. In the past ten years, manufacturers of gas detection equipment have developed and marketed “low-maintenance” or “maintenance-free” monitors that can provide early hazard warnings to agricultural workers at a fraction of the cost of traditional gas monitors.

ASABE and NIOSH have published guidance on the use of gas monitoring equipment in agriculture, including investigating the safety of manure pits prior to entry (ASABE, 2011; NIOSH 1990). However, the use of gas monitors remains rare, with surveys indicating that only 1.3% to 6% of farm operations test the air quality prior to pit entry (Murphy and Manbeck, 2014). Adoption of gas monitors is also low in allied agricultural services, e.g., only 5% of Iowa manure applicators indicated that they have used H2S monitors, although 31% indicated they might purchase them in the future (ISU, 2016).

Increased use of gas monitors by agricultural workers could provide important protection against toxic exposure to H2S. Traditional multi-gas monitors ($600 to $5000), often used for confined space entry, may be too costly and difficult for livestock producers to maintain. Low-cost, low-maintenance monitors for H2S are available from many manufacturers, but differences in their features and performance make selection challenging. Moreover, little information is available on the practical maintenance and performance of these devices in agricultural environments.

This purpose of this study was to identify and evaluate, in a controlled environment, several models of low-maintenance or maintenance-free monitors that can be used as personal alarms to warn of H2S exposure risks. This study presents an evaluation of the long-term reliability of these monitors in an anticipated worst-case maintenance scenario, i.e., leaving the monitor inside a livestock production building and exposed to low concentrations of H2S without performing maintenance throughout its useful life.

Methods

In summer 2015, low-cost (<$250) personal H2S monitors marketed as “low-maintenance” or “maintenance-free” were identified. Farm supply catalogs had no gas monitors for sale, so a broader internet search identified single-gas H2S monitors available through online U.S. retailers. A panel of industrial hygienists reviewed the resulting web searches for monitors described as needing no initial calibration or maintenance. To prioritize the selection, the panel identified models from recognizable manufacturers of robust high-cost equipment. Next, the top-ranked monitors that were available from at least three online retailers were selected for inclusion. Four monitors were selected for evaluation, representing four manufacturers and a range of cost and features (table 1). Two monitors provided “qualitative” information, with displays indicating the remaining monitor life, in months, but not the H2S concentration. The other two monitors were “quantitative,” displaying the H2S concentration detected in ppm. All monitors had audible and visual alarm warnings at two levels that were preset by the manufacturer.

Table 1. Test monitors selected for the study, with comparison of key features.
SpecificationQualitative MonitorsQuantitative Monitors
Altair[a]BW Clip[b]Pac 3500[c]T40 Rattler[d]
Cost$109$110$209$220
Battery
TypeLithiumInternal (2 years)LithiumAA (1500 h)
ReplaceableNoNoYesYes
Display
ConcentrationNoNoYesYes
Default low alarm10 ppm10 ppm10 ppm10 ppm
Default high alarm15 ppm15 ppm20 ppm20 ppm
Monitor response
Reported accuracy[e]NSNS=1% drift
per month
NS
Maximum detection100 ppm100 ppm100 ppm500 ppm
Warranty periodTwo years or
18 h alarm
Two years with
2 min alarm
per day
Two yearsTwo years from
shipping date
Shelf-lifeOne year prior
to activation
One year prior
to activation
Two yearsOne year prior
to activation
Manufacturer recommendations[e]
Bump testRecommendedPossible[f]RecommendedPossible
CalibrationPossible but
not required
Recommended
if alarmed[f]
Possible[f]Possible

    Concentration

NS40 ppmNS25 ppm
Flow rate0.25 L min-1NS0.5 L min-1NS

    [a] MSA Safety, Cranberry Township, Pa.

    [b] Honeywell International, Morris Plains, N.J.

    [c] Dräger Safety, Inc., Pittsburgh, Pa.

    [d] Industrial Scientific Corp., Pittsburgh, Pa.

    [e] As indicated in user manual provided with monitor at time of purchase. NS = not specified in manual.

    [f] Requires additional proprietary hardware or software not included with monitor purchase.

Experimental Setup

The monitors were placed in a sealed chamber (0.047 m3 internal volume, submersible enclosure, Part No. 5376K312, McMaster-Carr, Elmhurst, Ill.) and exposed to low concentrations of H2S over extended periods to replicate storage and use conditions in livestock production operations (fig. 1). Ports in the chamber accommodated power supply, gas delivery, and pressure release. Two monitors of each model were placed inside the chamber: one was fully activated throughout the study (primary), and the other was placed in standby mode (if available) or turned off throughout the study (secondary). The secondary monitor was used to develop storage recommendations for an unused monitor.

Hydrogen sulfide gas (25 ppm H2S, Praxair, Inc., Danbury, Conn.) was delivered through a port in the top of the chamber (fig. 1). A fan was placed inside the chamber to ensure well-mixed air. Power to the fan and the ToxiRAE reference monitor was supplied through additional ports on the right side of the chamber. The vent port on the left side of the chamber was opened only during chamber charging to prevent pressurization.

The reference H2S concentration in the chamber was measured with a ToxiRAE Pro EC (RAE Systems by Honeywell International, Morris Plains, N.J.). One-minute averaged concentrations were logged continuously throughout each chamber test. The cumulative concentration (ppm-time) in the chamber was computed from the ToxiRAE readings. To quantify any sensor drift of this reference monitor, the chamber H2S concentra-

Figure 1. Experimental setup of test chamber. The primary test monitors that were continuously operating are at the bottom, and the secondary monitors in standby mode are behind them. The ToxiRAE reference monitor logged H2S concentrations throughout each chamber test, and the probes for the external VRAE and Altair 4X reference monitors were inserted initially and periodically throughout each test to confirm the ToxiRAE readings of the chamber H2S concentration.

tions were checked at the start and end of each chamber test and daily with two additional reference monitors: an Altair 4X (MSA Safety, Cranberry Township, Pa.) and a VRAE (RAE Systems by Honeywell International, Morris Plains, N.J.). All three reference monitors were calibrated at the beginning of each chamber test (zero, 20 ppm) using calibration gas (34L-428-20, Gasco Precision Calibration Mixtures, Oldsmar, Fla.).

Test Protocol

The test monitors were exposed to H2S in the test chamber over 24 sequential short-term tests at concentrations above typical concentrations measured in production buildings, but below alarm level (10 ppm), in order to test the drift and longevity of the monitors in a shorter period while in a controlled environment. This method is similar to aerosol monitor tests (Vanderpool et al. 2001; Kenny et al., 2004) used to assess monitor stability and response time (NIOSH, 2012).

The H2S concentration and exposure duration were established to approximate the environment that a monitor might encounter over a one-year period in a livestock operation. Specifically, the protocol assumed that these monitors might be stored in livestock buildings, where the electrochemical sensors would be exposed to low concentrations of H2S throughout the year even if not in use, which could degrade the sensors over time. To convert the chamber test conditions to equivalent barn exposure, the cumulative annual exposure of an H2S monitor in use in a livestock operation was estimated. This assumed a low background concentration typical inside swine production buildings (1 ppm) and added H2S contributions anticipated from pumping and weekly bump testing, in which a known concentration of gas is applied to the sensor to ensure that it is responding to the gas (table 2). These short-term peaks contribute little to the annual cumulative concentration relative to the continuous, low background exposures in a livestock building. The annual cumulative exposure for a monitor used in the field was estimated to be just over 367 ppm-day, yielding an average daily H2S exposure of 1.005 ppm over 365 days. The test monitors were subjected to 24 sequential chamber tests to achieve exposure equivalent to the estimated one-year cumulative concentration in the field.

Table 2. Estimated annual cumulative H2S exposure for a monitor stored in a livestock building.
Typical Exposure SourceConcentration
(ppm)
Duration and
Frequency Assumptions
Cumulative
ppm-day
Background concentration in building1Constant, 365 days365
High concentration events2515 min, 4 per year1.04
Weekly bump testing201 min, 52 per year0.72
Annual total:366.76

The procedure for each chamber test was to charge the chamber with H2S gas, allow the H2S concentration to decay, and then remove the monitors for performance tests. To charge the test chamber, 25 ppm of H2S was delivered into the chamber until the concentration was just below the low alarm level (10 ppm). During charging, the chamber concentration was recorded at 2 min intervals with the two test monitors with displays (Pac 3500 and T40 Rattler) and the ToxiRAE reference monitor. When the chamber concentration reached approximately 9.5 ppm, the H2S flow was stopped. The initial chamber concentration was recorded using all three reference monitors (ToxiRAE, VRAE, and Altair 4X), after which all chamber ports were closed.

After charging, the H2S concentration in the chamber decreased over subsequent days. In phase 1 of this study, the chamber concentration during each chamber test was allowed to decay to ~1 ppm prior to opening the chamber to evaluate monitor performance. Phase 1 lasted 16 weeks, during which a total of 21 chamber tests achieved a cumulative exposure of 273 equivalent barn days (274 ppm-days) for the test monitors. For the first chamber test in phase 1, the chamber was recharged midway through the 5.8-day period (18.5 equivalent barn days); all other phase 1 tests had a single chamber charge. After 21 chamber tests, the test monitors began to show declines in performance, and testing was accelerated. In phase 2, the last three chamber tests included chamber recharging, without chamber opening, when the H2S concentration decayed to ~5 ppm in order to maintain higher mean concentrations over time. Phase 2 lasted two weeks to achieve an additional 100 equivalent barn days of cumulative exposure.

For each of the 24 chamber tests, when the chamber H2S concentration decayed to ~1 ppm, the chamber was opened, the monitor performances were checked (bump tested), and the reference monitors were fully calibrated. Each individual chamber test lasted from 3 to 7 days.

Before each chamber test, all primary test monitors were bump tested by applying 20 ppm of H2S calibration gas to the sensor. Performance data were collected at this time, including time to respond to low and high alarms (all monitors) and concentration readings (T40 Rattler and Pac 3500 only). In addition, the logged data from the ToxiRAE reference monitor were downloaded to compute the cumulative exposure of the monitors in the chamber. The secondary test monitors were not bump tested after each chamber test; instead, they were bump tested only at the end of the study.

After testing began, Honeywell released a BW Clip Real Time (RT) monitor with concentration display that used the same electrochemical sensor as the BW Clip monitor. A BW Clip RT monitor was added to the study at chamber test 14.

Performance Analysis

The cumulative H2S exposure (ppm-day) of the test monitors was computed from the activation of the test monitor using the ToxiRAE logged concentration data. The cumulative concentration at the end of each chamber test was used to mark changes in test monitor performance. The cumulative concentrations were converted to equivalent time in a livestock building by dividing the chamber concentrations by 1.005 ppm H2S per day in a barn to compute the equivalent barn days associated with the laboratory tests. Failures were reported in equivalent barn days to indicate when a failure would be anticipated in a typical livestock operation. The equivalent days of barn exposure were identified for each monitor when the high alarm failed to signal at either 60 s (manufacturer’s criteria) or 15 s (field recommendation criteria) during bump testing (Wanek, 2011). For the quantitative test monitors (Pac 3500 and T40 Rattler), failure was also identified when the monitor drifted below the calibration gas concentration during bump tests (20 ppm) or differed from the reference monitor by 3 ppm (15% difference).

For the secondary test monitors, which were in standby mode or off throughout the study, these same criteria were used at the end of the study to determine if the concentration had drifted by more than 3 ppm or if the alarm response had failed.

Results

The test monitors were challenged in the chamber for 373 equivalent barn days. The bump testing of the primary test monitors over the 24 chamber tests was equivalent to field bump testing every two weeks over a one-year period. Figure 2 illustrates the patterns of gas concentration generated to expose the test monitors across the chamber tests. Figure 2a shows the pattern over the entire test period, with phases 1 and 2 delineated. Figure 2b illustrates the concentrations in chamber test 1 (phase 1), with a quick concentration increase to just below the low alarm level (10 ppm) and then a decay to 1 ppm. This first test included a second charging at day 10 (equivalent barn day). Figure 2c illustrates the charging pattern in phase 2, using chamber test 22, to illustrate how higher chamber concentrations were maintained in the last three tests, in which an initial charge to ~9.5 ppm was followed by two additional chamber charges when the concentration decayed to 5 ppm on equivalent barn days 294 and 299.

The ToxiRAE reference monitor, which was used throughout the chamber tests, showed no significant drift over the study period. Linear regression between between the ToxiRAE and both the VRAE and the Altair 4X produced slopes of 1 and intercepts indicating that the ToxiRAE read 0.2 ppm lower, on average, than the other two reference monitors (R2 = 0.97 to 0.98).

Figure 2. Pattern of chamber H2S concentration during (a) entire duration of testing (24 chamber tests), (b) chamber test 1 (phase 1) in which the concentration was allowed to decay to 1 ppm, and (c) chamber test 22 (phase 2) in which the concentration was maintained with recharging. Concentrations were from the data-logging ToxiRAE reference monitor.

Performance of Primary Monitors

The times at which notable performance failures occurred are shown in figure 3. Over the 18 weeks of chamber testing, the T40 Rattler required battery changes four times. While not considered failures under the study’s criteria, these events were noted to determine the higher maintenance requirements of the T40 Rattler. The Altair failed to activate the high alarm nine times during bump tests. The BW Clip experienced a high alarm failure only once during testing after 324 equivalent barn days. These high alarm failures were considered critical because the monitor failed to detect the gas at the high alarm setting within 1 min of gas exposure. The Pac 3500 had 17 events in which the monitor reading was at least 3 ppm (15%) lower than the 20 ppm calibration gas. This 15% underestimation at low concentrations may not be critical, but it could be important at higher, more hazardous concentrations.

Figure 3. Test monitor failures during bump testing over the chamber test period by equivalent barn days and by cumulative concentration of monitor exposure (ppm-days).

The concentrations displayed during bump tests of the quantitative monitors are provided in figure 4. The T40 Rattler read consistently higher than the bump test gas (mean = +2.3 ppm). On equivalent barn day 203, the T40 Rattler read below the 20 ppm challenge concentration, but the battery charge read low soon after this calibration. The reported low concentration with low battery indicates that battery management is critical for using the T40 Rattler in the field. The Pac 3500 consistently read low (mean = -3.4 ppm), with 17 of 22 tests low by more than 15% (3 ppm). The mean difference of 3.4 ppm is equivalent to a drift of 17% from the bump test gas, which might be anticipated after 17 months of field use with the maximum concentration drift reported for the Pac 3500; however, this discrepancy was observed almost as soon as the monitor was turned on. The BW Clip RT, added at chamber test 14, was challenged to only 206 equivalent barn days of H2S. No alarm failures were noted, and the concentration display did not drop below the 15% criterion (3 ppm) during the bump tests.

Figure 4. Concentrations displayed during bump tests of primary quantitative monitors using 20 ppm H2S challenge gas.

Table 3 details the response times required for the primary qualitative test monitors to signal the low (10 ppm) and high (15 ppm) alarms during bump testing. During phase 1,

Table 3. Alarm response of primary test monitors using 20 ppm H2S calibration gas. “Pass” indicates quick alarm activation, a number indicates the seconds required to signal the alarm when the calibration gas was applied to the sensor, and “Fail” indicates that the alarm did not activate within 60 s.
Equivalent
Barn Days
MSA AltairBW Clip
10 ppm15 ppm10 ppm15 ppm
Phase 10PassPassPassPass
30PassPassPassPass
36PassPassPassPass
46PassPassPassPass
56PassPassPassPass
64PassPassPassPass
78PassPassPassPass
88PassPassPassPass
98Pass31PassPass
10612FailPassPass
1151149PassPass
1251240PassPass
13912Fail912
15514Fail1013
16712Fail913
18511Fail1018
19413491019
20311421018
21510Fail1020
24313491020
Phase 227310Fail1023
28912Fail11Fail
37316Fail1220

both monitors signaled the low and high alarms quickly, but by equivalent barn day 98, the Altair required 31 s to signal the high alarm. By equivalent barn day 273, the Altair did not activate the high alarm when challenged with 20 ppm of H2S for more than 60 s. The BW Clip continued to activate the high alarm until day 289, with a recovery on the next test. A noticeably longer response time was clearly developing in the BW Clip at the end of one equivalent year of use.

Performance of Secondary Monitors

The performance of the secondary test monitors was examined at the end of the test period, with bump tests occurring before the chamber studies began and at 273 and 373 equivalent barn days of chamber exposure. The secondary T40 Rattler, which was turned off during the study, passed the alarm tests and displayed 20 ppm, matching the bump test gas. The secondary Pac 3500, also turned off for the chamber tests, reported 2.5 ppm below the calibration gas target, similar to the primary Pac 3500 monitor. The secondary BW Clip was put into hibernation mode at the start of the study using the manufacturer’s hibernation case. The display of time remaining confirmed that it had been hibernating, and the monitor responded adequately to the bump tests (alarm activation and speed). The secondary Altair was incapable of being shut off or hibernating once activated. This secondary test monitor was active throughout the entire chamber test period but was not bump tested between each chamber test. Like the primary Altair, the secondary Altair failed the high alarm test when it was challenged at equivalent barn day 273.

Discussion

Each of the four models of test monitors exhibited a decline in performance over the study. These results confirmed that long-term exposure to even low levels of H2S might jeopardize the reliability of direct-reading monitors. While the user instructions provided with these monitors indicated that little or no maintenance is needed, bump tests are critical to ensure that alarms function and concentrations are accurate. Based on performance, qualitative observations, features, and manufacturer’s instructions, the ranked order of the recommended monitors is: (1) Honeywell BW Clip, (2) Industrial Scientific T40 Rattler, (3) Dräger Pac 3500, and (4) MSA Altair.

The Honeywell BW Clip monitors (primary and secondary) responded consistently to bump tests for most of the test period. Low alarms (10 ppm) occurred within 10 s for every bump test. This monitor required the least amount of maintenance and exhibited the fewest failures. This monitor also has an ancillary hibernation case (~$15) that extends the life of the monitor, with evidence of this protection demonstrated by the secondary monitor. While the BW Clip RT was only available midway through the study and was tested for only 204 equivalent barn days, it had no failures. This unit uses the same hibernation case and has identical sensor technology as the BW Clip that completed the full test period. Because 89% of recently surveyed livestock producers indicated that concentration display was an important feature of H2S monitors (Anthony, 2017), the BW Clip RT may be highly attractive for agricultural operations.

The Industrial Scientific T40 Rattler required more maintenance than the BW Clip due to the number of battery changes, but otherwise showed no failures. Although the test gas readings were not accurate, the primary test monitor overestimated the concentrations, erring on the side of worker protection. However, a concern is that the secondary T40 Rattler displayed low concentrations at the end of study. Users can fully calibrate this monitor, allowing adjustment of the reported concentrations to match calibration gas. Due to its vulnerability during low battery charge, new batteries should be installed prior to using this monitor for activities with high risk of H2S exposure. It is unclear how long the batteries will last when the monitor signals alarms for long periods in the field.

Both Dräger Pac 3500 test monitors displayed concentrations lower than the 20 ppm calibration gas during bump tests at monitor activation and consistently throughout the chamber tests. Under-reporting of the actual concentration is problematic because a hazardous environment might be identified as safe. For industries in which chronic low exposures are monitored and regulated, inaccurately low readings in the range of 20 ppm would be unacceptable. More importantly, this monitor includes a menu-driven bump test feature, not used during the study, that may be problematic in the field because an unsuccessful bump test using the device menu would disable the device, leaving a producer with no functioning monitor should the device read low.

The MSA Altair monitors cannot be turned off after initial activation. According to the manufacturer, the alarm should respond to bump testing within 60 s of gas application, although response in 15 s is recommended (Wanek, 2011). The primary monitor failed to signal an alarm at 20 ppm at 106 equivalent barn days of exposure. Without bump tests to verify its performance, the Altair provides no other indication of failure to the user, which could put a worker in serious danger of exposure. Due to the high failure rate of the Altair in this experiment, this monitor is not recommended for agricultural operations with limited monitor calibration and maintenance programs.

Maintenance Recommendations

The sensors used in the tested monitors have a finite life. Over time, the sensor performance may be reduced due to loss of electrolyte from reactions with H2S and other gases and from deposition of particles on the diffusion surfaces (Warburton et al., 1998; Woodfin, 1994; NIOSH, 2012; Pandley et al., 2012). Loss of electrolyte results in decreased response to a fixed gas concentration, reducing the ability of the monitor to detect the gas accurately. For this reason, gas monitors should be routinely challenged with a known concentration of test gas to ensure that any changes in the electrical signal over time result in accurate reporting of the gas concentration and trigger appropriate alarms.

According to the manufacturers’ documentation for these monitors, none of the monitors require calibration at the time of purchase. Two manufacturers (MSA and Dräger) recommend bump testing their monitors, as performed in this study. The manufacturers recommended that bump testing or other basic checks be performed “according to industry standards” or “according to local protocols” in the documentation. However, the agricultural community has not formally established recommendations for the frequency of these checks. Based on this study’s findings, routine (at least monthly) bump testing is recommended for these low-cost monitors during activities of low risk as well as prior to conducting high-risk operations (e.g., manure pumping, pressure washing of swine rooms, testing confined space air quality). Bump testing is recommended well in advance of high-risk activities to verify that the sensor is performing well so that there will be time to calibrate, repair, or replace the monitor if necessary. In addition, due to the performance differences among these monitors when taken out of the box, bump testing is recommended upon first activation of the monitor.

The cost of supplies and expertise needed to test these monitors is not trivial. While each monitor can be obtained for less than $250, calibration gas ($100 and up) and a regulator ($100 to $200) are also needed to perform bump tests. Calibration gas for H2S typically has a shelf-life of up to 18 months. In small and medium-size operations, the calibration gas may likely expire before it is consumed. Feasible systems for providing technical expertise to bump test and calibrate the monitors, and possibly resources to purchase and maintain calibration systems, may be needed for local farming communities. The need to perform checks of the sensor response must be communicated to the workers who rely on these monitors.

When a monitor is purchased, storage in areas free of H2S is recommended. While the bump test concentrations reported by the quantitative monitors changed throughout the study period, the readings at the end of the study were the same as at the start. However, the electrolytic solutions may decay in the sensors of the alarm-only monitors studied here. Because manufacturers can change sensor components over time, all sensors need to be protected from H2S exposure when not in use to maximize their life.

The Altair and BW Clip have internal batteries that are not designed for user changing. Thus, when the battery is dead or the two-year activation period has been reached, the monitor must be replaced. If the alarm was not activated, these monitors would theoretically last the duration of the two-year warranty period. The Pac 3500 has a replaceable lithium battery, and the T40 Rattler uses a single AA alkaline battery. At first glance, the AA battery was identified by farmers as a favorable feature (Anthony, 2017). However, the number of battery changes required for the T40 Rattler during this study could indicate a vulnerability, and particular attention should be given to the battery level to ensure that the monitor remains responsive during use.

Study Limitations

This study did not account for humidity, temperature extremes, or other coexisting contaminants in livestock operations, which could further decrease the accuracy and useful life of the electrochemical H2S sensors. Temperatures above 25°C have been shown to increase sensor readings by approximately 0.5% to 1.0% per °C (Chou, 2000). Similarly, extremely high humidity can cause condensation to form on the electrode, interfering with the accuracy and operation of the sensor (Chou, 2000). Particle deposition on the sensor or its external filter may also affect the monitor response and should be included in field performance assessments of H2S monitors. Field use may require replacement of the filter, which might not be possible for all low-cost monitors. Thus, long-term evaluations of these monitors in agricultural environments would provide realistic feedback on how temperature, humidity, dust, and co-contaminants affect the sensors over time.

Although these monitors were challenged at concentrations higher than typical background levels in livestock buildings, the study used equivalent barn exposure assumptions to simulate the performance of each monitor over approximately one year in a livestock barn with a background concentration of ~1 ppm. Under these conditions, both alarm-only monitors showed signs of alarm failure at the high alarm setting and began to show effects at the low alarm level. If the background concentration in a livestock building is higher than 1 ppm, these monitors may show signs of failure even earlier than reported here. This study did not examine how well these monitors perform over time when exposed to concentrations that are immediately dangerous to life and health (100 ppm), both in terms of monitor response and useful life. It is reasonable to recommend bump testing the monitors after exposure to field concentrations that result in alarms.

Finally, this study tested a very small sample of monitors. Only four models were tested, with only two monitors of each model. While this work informs how monitors might perform over a year, it is impossible to determine that any model is significantly better- or worse for agricultural use without additional testing, as sensors and monitors can vary between manufacturer lots. Bump testing initially and throughout the monitor’s useful life is standard practice in other industries and, based on findings of this study, is recommended to appropriately inform users of H2S risks in agricultural operations.

Conclusion

Each of the low-cost H2S monitors that were studied exhibited performance characteristics indicating that maintenance is needed throughout the monitor’s life. Monitors should be challenged with a test gas (bump tests) to ensure that the sensor can detect H2S concentrations and signal alarms as designed. As shown in this study, bump tests are more important later in the life of the monitor, when the likelihood of failure increases. However, if tested and maintained appropriately, these monitors have potential to warn agricultural workers of potentially deadly H2S exposures. Future work is needed to determine if environmental factors within a barn would contribute to sensor failure even earlier than identified in these chamber tests. The results of this study can be used in agricultural educational programs aimed at increasing awareness of H2S exposure risks, where discussions of gas monitor selection and use can incorporate both the benefits and limitations of direct-reading monitors.

Acknowledgements

This research was funded in part by CDC/NIOSH Great Plains Center for Agricultural Health (Grant No. U54 OH007548).

References

ACGIH. (2017). TLVs and BEIs based on the documentation of the threshold limit values for chemical substances and physical agents and biological exposure indices. Proc. American Conf. of Governmental Industrial Hygienists. Cincinnati, OH: ACGIH.

Adekoya, N., & Myers, J. R. (1999). Fatal harmful substances or environmental exposures in agriculture, 1992 to 1996. J. Occup. Environ. Med., 41(8), 699-705. https://doi.org/10.1097/00043764-199908000-00013

Anthony, R. (2017). Gas monitors on the farm. Farm Families Alive & Well, 23(2), 1-2. Iowa City, IA: Great Plains Center for Agricultural Health. Retrieved from https://www.public-health.uiowa.edu/icash/wp-content/uploads/2017/03/Alive-Well-Newsletter-March-2017.pdf

Anthony, T. R., Altmaier, R., Jones, S., Gassman, R., Park, J. H., & Peters, T. M. (2015). Use of recirculating ventilation with dust filtration to improve wintertime air quality in a swine farrowing room. J. Occup. Environ. Hyg., 12(9), 635-646. https://doi.org/10.1080/15459624.2015.1029616

ASABE. (2011). EP470.1: Manure storage safety. St. Joseph, MI: ASABE.

Beaver, R. L., & Field, W. E. (2007). Summary of documented fatalities in livestock manure storage and handling facilities: 1975-2004. J. Agromed., 12(2), 3-23. https://doi.org/10.1300/J096v12n02_02

Chou, J. (2000). Hazardous gas monitors: A practical guide to selection, operation, and applications. New York, NY: McGraw-Hill.

Donham, K. J., Knapp, L. W., Monson, R., & Gustafson, K. (1982). Acute toxic exposure to gases from liquid manure. J. Occup. Environ. Med., 24(2), 142-145.

Fabian-Wheeler, E. E., Hile, M. L., Murphy, D. J., Hill, D. E., Meinen, R., Brandt, R. C., ... Hofstetter, D. (2017). Operator exposure to hydrogen sulfide from dairy manure storages containing gypsum bedding. J. Agric. Saf. Health, 23(1), 9-22. https://doi.org/10.13031/jash.11563

Guarrasi, J., Trask, C., & Kirychuk, S. (2015). A systematic review of occupational exposure to hydrogen sulfide in livestock operations. J. Agromed., 20(2), 225-236. https://doi.org/10.1080/1059924X.2015.1009667

Hendrickson, R. G., Chang, A., & Hamilton, R. J. (2004). Co-worker fatalities from hydrogen sulfide. American J. Ind. Med., 45(4), 346-350. https://doi.org/10.1002/ajim.10355

ISU. (2016). Hydrogen sulfide safety. Ames, IA: Iowa State University Extension and Outreach. Retrieved from http://www.agronext.iastate.edu/immag/info/H2S%20Monitors.pdf

Kenny, L. C., Merrifield, T., Mark, D., Gussman, R., & Thorpe, A. (2004). The development and designation testing of a new USEPA-approved fine particle inlet: A study of the USEPA designation process. Aerosol Sci. Tech., 39(S2), 15-22. https://doi.org/10.1080/027868290502290

Murphy, D. J., & Manbeck, H. B. (2014). Confined space manure storage and facilities safety assessment. J. Agric. Saf. Health, 20(3), 199-201. https://doi.org/10.13031/jash.20.10377

NRC. (2010). Hydrogen sulfide acute exposure guideline levels. In Acute exposure guideline levels for selected airborne chemicals (Vol. 9, pp. 173-218). Washington, DC: National Research Council. https://doi.org/10.17226/12978

NIOSH. (1990). NIOSH alert: Preventing deaths of farm workers in manure pits. Publication 90-103. Washington, DC: National Institute for Occupational Safety and Health. Retrieved from https://www.cdc.gov/niosh/docs/90-103/

NIOSH. (2007). NIOSH pocket guide to chemical hazards. Publication 2005-149. Washington, DC: National Institute for Occupational Safety and Health. Retrieved from https://www.cdc.gov/niosh/docs/2005-149/pdfs/2005-149.pdf

NIOSH. (2012). Components for evaluation of direct-reading monitors for gases and vapors. Publication 2012-162. Washington, DC: National Institute for Occupational Safety and Health. Retrieved from https://www.cdc.gov/niosh/docs/2012-162/pdfs/2012-162.pdf

OSHA. (2017). Table Z-2: Air contaminants. 29 CFR 1910.1000, amendment 2017. Washington, DC: Occupational Safety and Health Administration. Retrieved from https://www.osha.gov/pls/oshaweb/owadisp.show_document?p_table=STANDARDS&p_id=9993

Pandey, S. K., Kim, K.-H., & Tang, K.-T. (2012). A review of sensor-based methods for monitoring hydrogen sulfide. Trends Anal. Chem., 32, 87-99. https://doi.org/10.1016/j.trac.2011.08.008

Poppendorf, W. (1991). Gases, vapors, liquids, and drugs. Proc. Surgeon General’s Conf. on Agricultural Safety and Health. Washington, DC: National Institute for Occupational Safety and Health. Retrieved from https://profiles.nlm.nih.gov/ps/access/NNBBWK.pdf

Reeve, K. A., Peters, T. M., & Anthony, T. R. (2013). Wintertime factors affecting contaminant distribution in a swine farrowing room. J. Occup. Environ. Hyg., 10(6), 287-296. https://doi.org/10.1080/15459624.2013.777303

Riedel, S. M., & Field, W. E. (2013). Summation of the frequency, severity, and primary causative factors associated with injuries and fatalities involving confined spaces in agriculture. J. Agric. Saf. Health, 19(2), 83-100. https://doi.org/http://dx.doi.org/10.13031/jash.19.9326

Swestka, R. J. (2010). Hydrogen sulfide spatial distribution and exposure in deep-pit swine housing. MS thesis. Ames, IA: Iowa State University, Department of Agricultural Engineering. Retrieved from http://lib.dr.iastate.edu/cgi/viewcontent.cgi?article=2440&context=etd

Vanderpool, R. W., Peters, T. M., Natarajan, S., Gemmill, D. B., & Wiener, R. W. (2001). Evaluation of the loading characteristics of the EPA WINS PM2.5 separator. Aerosol Sci. Tech., 34(5), 444-456. https://doi.org/10.1080/02786820117739

Wanek, R. (2011). Monitoring H2S to meet new exposure standards. Dallas, TX: Occupational Health and Safety. Retrieved from https://ohsonline.com/articles/2011/09/01/monitoring-h2s-to-meet-new-exposure-standards.aspx

Warburton, P. R., Pagano, M. P., Hoover, R., Logman, M., Crytzer, K., & Warburton, Y. J. (1998). Amperometric gas sensor response times. Anal. Chem., 70(5), 998-1006. https://doi.org/10.1021/ac970644y

Woodfin, W. J. (1994). Chapter H: Portable electrochemical sensor methods. In NIOSH manual of analytic methods (4th ed.). Publication 94-113. Washington, DC: National Institute for Occupational Safety and Health. Retrieved from https://www.cdc.gov/niosh/docs/2003-154/pdfs/chapter-h.pdf