IOP Conference Series: Earth and Environmental Science

PAPER • OPEN ACCESS

The ammonia release hazard and risk assessment: A case study of urea
fertilizer industry in Indonesia
To cite this article: A S Danasa et al 2019 IOP Conf. Ser.: Earth Environ. Sci. 399 012087

View the article online for updates and enhancements.

This content was downloaded from IP address 193.31.73.212 on 01/01/2020 at 01:08

ISenREM 2019 IOP Publishing
IOP Conf. Series: Earth and Environmental Science 399 (2019) 012087 doi:10.1088/1755-1315/399/1/012087

The ammonia release hazard and risk assessment: A case study

of urea fertilizer industry in Indonesia

A S Danasa1, T E B Soesilo1,3, D N Martono1, A Sodri1, A S Hadi2, and G T

Chandrasa2
1
School of Environmental Science, Universitas Indonesia, Jakarta 10430, Indonesia.
2
Agency for Meteorology, Climatology, and Geophysics, Jakarta 10720, Indonesia.

E-mail: [email protected]

Abstract. Ammonia is a raw material of fertilizer, refrigerant, and other commercial cleaning
products which commonly stored in a large capacity. The higher storage capacity, the higher risk
possibly occurs impacted to the population and environment. The study aims to provide a
modeling of ammonia release scenarios, escape from the storage facility, Urea Fertilizer Industry
X, located in Indonesia. This model is utilizing Area Location Hazardous Atmosphere (ALOHA)
software to forecast the threat zone of ammonia release scenario and QGIS to overlay and spatial
analysis. The result shows that the incident causes a significant impact on the area of 41.7 km2
and potentially threatens a massive scale of the inhabitant with higher evacuation factor (Ef),
lower affected population (Ap). The risk determined by estimated probability and consequence
considered as high risk, therefore, besides the main aim of establishing an emergency response
plan, this study could also be used as a reference in risk evaluation of chemical release.
Keywords: ammonia release, gas dispersion, risk assessment, toxic.

1. Background
The most common alkaline gas on the atmosphere, called ammonia, is significantly contributing to the
neutralization of various processes that occur in the atmosphere [1]. Ammonia has used in many sectors
of industries (Anhydrous Ammonia/synthetic). It is a toxic and corrosive agent that could threaten the
health of humans and the environment [2], where the air, water, and soil possibly threatened in case of
uncontrolled ammonia release occur. Ammonia threats the atmosphere as ammonia vapor instigated by
an organic material decomposition, cattle manure, soil-applied fertilizer; gas ventilation, leakage or spill
during the synthetic industrial process, production, transportation, failure in cooling/refrigeration
equipment; liquid waste; coal and wood firing; and as result of volcanic activity. Ammonia contaminates
the water through the leftover of waste treatment facility, waste of process, runoff of fertilizer pitch, and
livestock area. Ammonia defects the soil through the application of either natural or synthetic fertilizer,
degradation of cattle manure, decaying of organic materials, and by indirectly fixation of nitrogen [3].
Ammonia has the exposure limit of Time Weighted Average (TWA) 25 ppm (18 mg/m3) and Short
Term Exposure Limit (STEL) 35 ppm (27 mg/m3); it is a colorless gas having an overpowering smell
and suffocating effect; causes health effect if exposed to human such as irritation to eye, nose and throat;
dyspnoea (breathing difficulty), wheezing, chest pain; pulmonary edema; frothy pink sputum; skin
burns, and vesiculation; the exposure occur from liquid ammonia is likely to cause a freeze

3
To whom any correspondence should be addressed ([email protected])

Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution
of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Published under licence by IOP Publishing Ltd 1
ISenREM 2019 IOP Publishing
IOP Conf. Series: Earth and Environmental Science 399 (2019) 012087 doi:10.1088/1755-1315/399/1/012087

burn/frostbite; it is corrosive to copper and galvanic [4]. This gas could liquefy under pressure, where
it commonly sent to the consumer in the form of liquified compression ammonia gas. It also liquified
by reducing its temperature down to -33°C, particularly for raw material of fertilizer and refrigerant cold
storage industry.
As a raw material, anhydrous ammonia commonly stored in a vast volume; consequently, this could
increase the risk of chemical release where it could initiate by a human error, lack of
training, substandard production, and lack of maintenance and inspection program [5]. The incident is
also possible to suffer the surrounding area with a certain level of toxicity, depending on its level of
dispersion where it determined by a calculation of the release rated [6]. Henceforth, the risk could be
escalated to be a catastrophic event that the impact could be as much as the population, location
characteristic, meteorological factors, and geographical features [5]. Because of its soluble characteristic
in water as well as toxic to aquatic biota, the exposure of ammonia release could cause extensive
ecological damage [7].
This study develops a model by extending Orozco et al., 2019 [8], which calculated the effect of an
ammonia release incident in order to estimating the consequence as well as to predict the number of
people being affected; while, this paper expands it in determining the level of risk (low, moderate and
high) of an ammonia release scenario. The risk is acceptable if it is in an acceptable area; the risk must
be reduced if it is in an unacceptable area; procedures must be complied with if the risk in the ALARP
(As Low as Reasonably Practicable) region, which represents a moderate risk area. By calculating the
consequence, it expected that the prevention and mitigation of chemical release emergency conditions
could be established appropriately to minimize the impact of such a disaster.

2. Method
The law of chemicals dispersion determined by its physical properties, the stability of the atmosphere,
and the conditions of the release source [9]. Ammonia could behave like a heavy gas that has a deadly
higher risk since it is longer exposure duration compared to other lighter gases [10]. The highest risk of
ammonia release incident is a toxic vapor cloud, while it does not represent a serious flammability
hazard (mixtures of air and ammonia containing from 15% to 28% ammonia vapor by volume will ignite
when sparked, or exposed to temperatures exceeding 651°C) [8]. This study utilizes ALOHA Software
to forecast the toxic threat zones and QGIS to overlay and spatial analysis; 6 stages of research were
used (figure 1). By utilizing the threat zone area and the population density, the affected population can
be estimated as the following form:
Ap = (Tz x Ds) - (Ef x Tz x Ds) (1)
Ap: Affected Population
Tz: The AEGL-3/Red Zone
Ds: Density
Ef: Evacuation factor

The parameter of evacuation factor that is a level of emergency preparedness procedure of the
inhabitants in the event of a real disaster occur, includes: (1) Personnel and Organisational Structure;
(2) Standard Operation Procedure during emergency situation; (3) Emergency Drills and Exercises; (4)
Emergency Equipment; (5) Communication System. They weighted by utilizing the scoring method.
At the end stage of the research, the risk shall determine by calculating the probability and
consequence arise by the worst scenario of the ammonia release incident. The risk metric provides
guidelines for prioritizing and assessing the acceptability of risk concerning risk criteria. The general
form of the risk equation shall be as follows:
Risk = Probability × Consequence (2)

2
ISenREM 2019 IOP Publishing
IOP Conf. Series: Earth and Environmental Science 399 (2019) 012087 doi:10.1088/1755-1315/399/1/012087

Identification of Location Characteristic

Identification of Atmospheric Characteristic

Determining the Worst-Case Scenario

Modeling and Spatial Analysis

Human Health Risk Assessment

Determining the Risk Level

Figure 1. Stage of Research

2.1. Identification of Location Characteristic

The subject of study is an ammonia storage facility located in Urea Fertilizer Industry X, Industrial Area
Y, Indonesia. The industry employs 1,133 workers with an occupied area of 510 Ha, adjacent to 8 other
industries (most of them are petrochemical), housing area (4 villages), educational facilities, and other
public facilities. Based on statistical data, this area has a density of 1,321 inhabitants/km 2 [11]. This
research location is one of the the-eight rice barns in Indonesia; therefore, maintaining the sustainability
of this facility means to assure national food security.

2.2. Identification of Atmospheric Characteristic

The significant factors affecting the dispersion of gas clouds are the speed and direction of the wind and
atmospheric turbulence [12]. Stability is the property of the low-lying atmosphere that governs the
vertical movement of air, the tendency of the atmosphere to resist or enhance vertical motion, and thus
turbulence. ALOHA uses the Pasquill-Gifford-Turner scheme consisting of six stability classes based
on five surface wind speed categories, three types of daytime solar radiation, and two types of nighttime
cloud cover [13], as shown in table 1 as follows:

Table 1. Solar Radiation and Stability Class.

Dayb Night
Wind Speeda
Solar Radiation Cloud Cover
(at 10 m) (m/s)
Strongc Moderated Slighte >50% <50%
<2 A A-B B E F
2-3 A-B B C E F
3-5 B B-C C D E
5-6 C C-D D D D
>6 C D D D D
a
Wind speed measured at 10 m above ground
b
A, very unstable; B, moderately unstable; C, slightly unstable; D, neutral; E, slightly stable; F, stable
c
Solar altitude greater than 60° on a clear day
d
Solar altitude between 60° and 35° on a clear day
e
Solar altitude between 15° and 35° on a clear day

Solar radiation plays a large role in atmospheric stability. Therefore two different scenarios (day and
night) were considered in this study to model the dispersion of ammonia. The weather data refers to the
ERA5 Reanalysis Datasets from European Centre for Medium-Range Weather Forecast (ECMWF) [14].
This dataset describing the recent history of the atmosphere from 2008 until 2018 of the location, with
the summary of data, as shown in table 2 as follows:

3
ISenREM 2019 IOP Publishing
IOP Conf. Series: Earth and Environmental Science 399 (2019) 012087 doi:10.1088/1755-1315/399/1/012087

Table 2. Atmospheric Data.

Weather Data Day Night
(2008-2018) (1.00 p.m.) (1.00 a.m.)
Wind Speeda 2.7 m/s 1.6 m/s
Wind is fromb NE NNW
Humiditya 66 % 88 %
Air Temperaturea 30.7°C 24.8°C
Stability Class A E
a
The average of data
b
The highest percentage of data

2.3. Determining the Worst-Case Scenario

Based on the Failure Rate and Event Data for use within Risk Assessments of The Chemicals,
Explosives and Microbiological Hazardous Division 5 (CEMHD5), Health & Safety Executive (HSE),
the highest failure rate of large vessels is 2.5 x 10-3 with a leak hole diameter of 300 mm where the
leakage occurs due to deteriorated of component integrity, corrosion or lack of maintenance and
inspection [15]. The release source scenario is shown in table 3 as follows:

Table 3. Release Source Scenario.

The leak from a hole in a vertical cylindrical tank Hole at bottom
Tank Capacity 10,000 MT
Tank Diameter 32.004 meters
Tank Length 18.313 meters
State of Chemical Tank Contains Liquid
Internal Storage Temperature -33°C
Chemical Mass in Tank 60%
Circular Opening Diameter 300 mm
Opening is Bottom of Tank
Source: Secondary Data of Urea Fertilizer Industry X

3. Result and Discussion

The results of the modeling are ammonia escaped from the tank (not burning) as a mixture of gas and
aerosol (two-phase flow). Ammonia is one of the pollutants that is lighter than air, but it behaves as a
dense gas because it could form an aerosol and requires a longer dispersion time compared to other
pollutants so that it tends to be closer to the ground due to the effect of gravity [10].

3.1. Human Health Risk Assessment

The consequences of the toxic effect referred to the Acute Exposure Guideline Level (AEGL), which
represent threshold exposure limits for the general public; has multiple exposure limit from 10 minutes
(min) to 8 hours (h). Three levels – AEGL-1, AEGL-2, and AEGL-3 – are developed for each of five
exposure periods (10 min, 30 min, 1 h, 4 h, and 8 h) and are distinguished by varying degrees of severity
of toxic effects. By referring to Summary of AEGL Values for Ammonia in Acute Exposure Guideline
Levels for Selected Airborne Chemicals: Volume 6 (table 4), people who are exposed to AEGL-3
potentially suffer lethality; AEGL-2 potentially suffer disabling (irreversible or other serious, long-
lasting adverse health effects or an impaired ability to escape); AEGL-1 potentially suffer non-disabling
(notable discomfort, irritation, or specific asymptomatic non-sensory effects) [16]. Ammonia causes
discomfort at a general level with a concentration of 150-200 ppm; significant irritation at a

4
ISenREM 2019 IOP Publishing
IOP Conf. Series: Earth and Environmental Science 399 (2019) 012087 doi:10.1088/1755-1315/399/1/012087

concentration of 400-700 ppm; serious health effects at 500 ppm; and death within a few minutes at
2000 ppm [17].

Table 4. Summary of AEGL Values for Ammonia.

Classification 10 min 30 min 1h 4h 8h End Point (Reference)
AEGL-1 30 ppm 30 30 ppm 30 ppm 30 ppm Mild irritation
(nondisabling) (21 ppm (21 (21 (21 (MacEwen et al.,1970)
mg/m3) (21 mg/m3) mg/m3) mg/m3)
mg/m3)
AEGL-2 220 ppm 220 160 ppm 110 110 Irritation: eyes and
(disabling) (154 ppm (112 ppm ppm throat; urge to cough
mg/m3) (154 mg/m3) (77 (77 (Verberk, 1977)
mg/m3) mg/m3) mg/m3)
AEGL-3 2,700 1,600 1,100 550 390 Lethality
(lethal) ppm ppm ppm ppm ppm (Kapeghian et al., 1982;
(1,888 (1,119 (769 (385 (273 MacEwen & Vernot,
mg/m3) mg/m3) mg/m3) mg/m3) mg/m3) 1972)
Source: National Academy of Sciences, 2007

Figure 2(a) display threat zone on a daytime; figure 2(b) display threat zone on nighttime; table 5
shows the range of dispersion for each time. The biggest impact/farthest exposure radius occurs on
nighttime, with AEGL-3 zone: 6.1 km; AEGL-2 and AEGL-1 zone are greater than 10 km. On daytime,
AEGL-3 zone: 3.1 km; AEGL-2 zone: 8.9 km; and AEGL-1 zone is greater than 10 km. This result
parallel with Inanloo (2016) that releases during the daytime would have relatively smaller impact areas
in comparison with those that occur at nighttime [18]. In the daytime, strong solar radiation warms the
ground and the low-lying air, and then the warm air rises to generate eddies and high-level turbulence.
This unstable atmosphere enhances vertical mixing and turbulence so that the gas cloud will be dispersed
quickly and homogenous. In contrast with the nighttime, when the air temperature increases with height,
buoyancy forces act to inhibit vertical mixing creating a highly stable atmosphere.

Figure 2(a). Threat Zone on daytime Figure 2(b). Threat Zone on nighttime

5
ISenREM 2019 IOP Publishing
IOP Conf. Series: Earth and Environmental Science 399 (2019) 012087 doi:10.1088/1755-1315/399/1/012087

Table 5. Range of Dispersion

Daytime Nighttime
Dispersion Model Heavy Gas Heavy Gas
AEGL-3/Red Zone [1100 ppm; 60 min] 3.1 km 6.1 km
AEGL-2/Orange Zone [160 ppm; 60 min] 8.9 km greater than 10 km
AEGL-1/Yellow Zone [30 ppm; 60 min] greater than 10 km greater than 10 km

Reflecting figure 3(a), 3(b), 3(c), and 3(d) (QGIS-overlay), the incident potentially impacts 41.7 km2
area if it occurs at nighttime and 6.8 km2 if occur in the daytime.

Figure 3(a). Based on

QGIS-overlay, the
dispersion of ammonia
on daytime potentially
impacts to 4 villages
(6.8 km2 area).

Figure 3(b). By
considering the highest
percentage of wind is
from the northeast
(NE) to southwest
(SW), then the
potential threatened
areas are Employee
Housing of Urea
Fertilizer Industry X;
Public Housing 1 & 2;
Educational Facility;
Industrial Area Z; and
Highway.

6
ISenREM 2019 IOP Publishing
IOP Conf. Series: Earth and Environmental Science 399 (2019) 012087 doi:10.1088/1755-1315/399/1/012087

Figure 3(c). Based on

QGIS-overlay, the
dispersion of ammonia
on nighttime
potentially impacts to
16 villages (41.7 km2
area).

Figure 3(d). By
considering the highest
percentage of wind is
from north-northwest
(NNW) to south-
southeast (SSE), then
the potential
threatened areas are
most of the Urea
Fertilizer Industry X
(Industrial Area Y)
and Industrial Area Z;
Public Housing 3, 4, &
5; and Highway.

After acquiring the affected area, the next stage is calculating the evacuation factor and affected
people. The affected population is estimated by 55,086 - (55,086 x Ef) inhabitants during the nighttime
and 8,983 - (8,983 x Ef) inhabitants during the daytime (table 6). By considering the evacuation factors,
the affected population determined, as seen in table 7. A higher evacuation factor results in the lower
affected population.

7
ISenREM 2019 IOP Publishing
IOP Conf. Series: Earth and Environmental Science 399 (2019) 012087 doi:10.1088/1755-1315/399/1/012087

Table 6. Affected Area and Population.

Time Daytime Nighttime Unit
2
Red Zone Area 6.8 41.7 km
Density 1,321 inhabitants/km2
Possibly Affected Population 8,983 55,086 inhabitants
Evacuation Factor Ef %
Evacuated Population 8,983 x Ef 55,086 x Ef inhabitants
Affected Population 8,983-(8,983 x Ef) 55,086 - (55,086 x Ef) inhabitants

Table 7. Affected Population.

Affected Population Affected Population
Evacuation Factor
on Daytime (inhabitants) on Nighttime (inhabitants)
100% 0 0
75% 2,246 13,771
50% 4,491 27,543
25% 6,737 41,314
0% 8,983 55,086

3.2 Determining the Risk Level

The Probability (which is estimated by the highest failure rate of large vessels is 2.5 x 10-3 (minor type)
with a leak hole diameter of 300 mm) expressed in terms of frequency as several events occurring during
a specific timeframe. This frequency of 2.5 x 10-3 will be expressed as 0.0025 failures per year. The
event frequency is associated with probability by following table 8. It means that the frequency of 2.5 x
10-3 can be classified as a moderate level with a statistical range of 0.001 to 0.01 or 1 x 10-3 to 1 x 10-2.
The consequence is expressed as a numerical value or characterized by a consequence category
associated with the severity of potential fatality, serious injury, medical treatment or first aid that may
result from such an ammonia release event. The consequence criteria showed in table 9.

Table 8. Probability Criteria Table 9. Consequence Criteria

Possible Annual Failure Description Health Consequence
Qualitative Rank Probability or Catastrophic A large number of
Frequency fatalities
Very high >0.1 Major Single fatality
High 0.01 to 0.1 Serious Serious injuries
Moderate 0.001 to 0.01 Significant Minor injuries
Low 0.0001 to 0.001 Minor/ First-aid injuries/
Very low/Remote <0.0001 insignificant No significant
Source: API 580 Source: API 580

The calculated risk assigned to the boxes on the risk matrix as high, medium, and low (figure 4); the
risk level, as shown in Table 10, is calculated as high risk (3E), which represent the moderate level of
probability and very high level of consequence. The risk at these criteria shall be reduced to the ALARP
region.

8
ISenREM 2019 IOP Publishing
IOP Conf. Series: Earth and Environmental Science 399 (2019) 012087 doi:10.1088/1755-1315/399/1/012087

PROBABILI Table 10. Risk Level

TY Probability 2.5 x 10-3
Very high >0.1 5 (Moderate-
0.01 to 3)
High 4
0.1 Consequen Large
0.001 to ce number of
Moderate 3
0.01 fatalities
0.0001 to (Catastrophi
Low 2 c-E)
0.001
Very low/ Risk 3E
< 0.0001 1
Remote
Minor /

Catastrop
insignific

Significa

Serious

Major

hic
ant

nt
A B C D E
CONSEQUEN
CE
Figure 4. Risk Matrix

4. Conclusion
Based on ALOHA modeling, the worst-case scenario occurs in the nighttime, because a highly stable
atmosphere inhibits vertical mixing of air and pollutant cloud. The risk considered as high risk; it
expected that the prevention and mitigation of chemical release emergency conditions could be
established appropriately to reduce the risk of this scenario.

5. References
[1] Felix E P, Cardoso A A 2004 Atmospheric Ammonia: Sources, Transformation, Sinks, and
Methods of Analysis Quim 27(1) 123–30
[2] Kwak D, Lei Y, Maric R 2019 Ammonia gas sensors: A comprehensive review
Nov;204(March):713–30
[3] Agency for Toxic Substances and Disease Registry 2004 Toxicological Profile for Ammonia
(September):269
[4] NIOSH 2018 NIOSH Pocket Guide to Chemical Hazards
[5] Anjana N S, Amarnath A, Harindranathan Nair M V 2018 Toxic Hazards of Ammonia Release
and Population Vulnerability Assessment using Geographical Information System J Environ
Manage Mar;210:201–9
[6] Namboothiri N V, Soman A 2018 Consequence assessment of anhydrous ammonia release using
CFD-probit analysis. Process Saf Prog Mar 24
[7] US EPA 2001 Hazards of Ammonia Releases at Ammonia Refrigeration Facilities (Update) 1–8
[8] Orozco J L, Van Caneghem J, Hens L, González L, Lugo R, Díaz S, et al 2019 Assessment of an
ammonia incident in the industrial area of Matanzas J Clean Prod Jun;222:934–41
[9] Tan W, Du H, Liu L, Su T, Liu X 2017 Experimental and numerical study of ammonia leakage
and dispersion in a food factory. J Loss Prev Process Ind May;47:129–39
[10] Khan F I, Husain T, Abbasi S A 2001 Safety Weighted Hazard Index (Swehi). A New, User-
Friendly Tool for Swift Yet Comprehensive Hazard Identification and Safety Evaluation in
Chemical Process Industries. Process Saf Environ Prot. 79(2):65–80
[11] Regency B-S of 2018 Karawang Regency in Figures
[12] Inanloo B, Tansel B 2015 Explosion Impacts During Transport of Hazardous Cargo : GIS-Based
Characterization of Overpressure Impacts and Delineation of Fl Ammable Zones for Ammonia

9
ISenREM 2019 IOP Publishing
IOP Conf. Series: Earth and Environmental Science 399 (2019) 012087 doi:10.1088/1755-1315/399/1/012087

J Environ Manage 156:1–9

[13] NOAA 2013 ALOHA Technical Documentation Seattle
[14] ECMWF 2019 Reanalysis Datasets
[15] HSE health and safety executive 2019 Failure Rate and Event Data for use within Risk
Assessments
[16] National Academy of Sciences 2007 Acute Exposure Guideline Levels for Selected Airborne
Chemicals: Volume 6
[17] Tran, Pascaline H, Lapadula, Gerard D, Liu X 2008 Ammonia oxidation catalyst for the coal-fired
utilities 2(12)
[18] Inanloo B, Tansel B, Jin X, Bernardo-Bricker A 2016 Cargo-Specific Accidental Release Impact
Zones for Hazardous Materials: Risk and Consequence Comparison for Ammonia and
Hydrogen Fluoride Environ Syst Decis

Acknowledgment
This work financially supported by the Universitas Indonesia through PITMA B (International Indexed
Publication for Master Students) funding program in the 2019 fiscal year.

10