The Significance of Point Source Emission

Research Journal of Chemistry and Environment______________________________________Vol.15 (2) June (2011) Res.J.Chem.Environ

(1)

TThhee SSiiggnniiffiiccaannccee ooff PPooiinntt SSoouurrccee EEmmiissssiioonn ((NNOO22)) bbyy PPeettrroocchheemmiiccaall PPllaannttss aatt NNoorrtthh EEaasstt ooff PPeenniinnssuullaarr

MMaallaayyssiiaa Ibrahim Mohd Habir* 1, Abdullah Ahmad Makmom 2, Adam Nor Mariah 3 and Ismail Mohd Halim Shah 4

1. Kolej Universiti TATI, Jalan Panchor, Teluk Kalong, 24000 Kemaman, Terengganu, MALAYSIA 2. Department of Environmental Sciences, Faculty of Environmental Studies, Universiti Putra Malaysia, 43000 UPM Serdang, Selangor,

MALAYSIA 3. Department of Mechanical and Manufacturing Engineering, Universiti Putra Malaysia, 43000 UPM Serdang, Selangor, MALAYSIA 4. Department of Chemical and Environmental Engineering, Universiti Putra Malaysia, 43000 UPM Serdang, Selangor, MALAYSIA

*[email protected]

Abstract Petrochemical industry is one of the major

pollutant generators around the world. The same scenario is observed occurring at North East of Peninsular Malaysia. The ISCT3 air dispersion of Gaussain Transport Model was used to simulate the average predicted on ground concentration of NO2 emitted by point source from the petrochemicals plants. The simulation covers major input of geographical domain set up, NO2 generators specification and meteorological parameters. The geographical domain set up is at 20 km x 20 km covering area centre of the petrochemicals plants with 0.5 km receptors grid spacing. The NO2 emission rate was estimated through the use of ultimate fuel analysis method. The NO2 generated combined with 5 yearly meteorological data obtained was applied to perform the simulation at the optimum correlation of wind direction. The simulation performed discovered that the predicted monthly and yearly average of on ground NO2 concentration range from 13.97 to 20.43 ug/m3 and 6.91 to 8.49 ug/m3 respectively. The yearly predicted average concentration shows that the value is below the WHO guideline which is at 40 ug/m3. No benchmark could be performed at the monthly average since there is no standard guideline available.

Keywords: Meteorological modeling, correlation, NO2, ISCT3 Introduction

Globally, petrochemicals industry has been considered as the second major industrial emission sources of SO2, NOx and CO2 in Asia, after steel and iron industry2. At North East of Peninsular Malaysia, petrochemicals industry is capable of producing 6.9 million tons per year of various chemical products that is equivalent to 55 % of the total Malaysian chemicals product produced22. One of the petrochemicals industry located in this region is known as Kertih Petrochemical Industrial Area (Figure 1) There are 27,000 major and hundreds of thousands of minor sources of air pollution in the United States28. It was noted

by the United States Environmental Protection Agency (EPA) in 1994 that more than half of the six most important air pollutants (CO, Pb, NOx, VOCx, PM10 and SOx; except CO) are contributed by stationary sources of which mostly from petrochemical plants16. With the increasing trend of the recent petrochemical production volume from 4.3 to 9.2 ton per year from 2001 to 200926, there is an expectation that this industry might well contributed to more generation of air pollution.

Environmental concerns have raised public

awareness of environmental issues that lead to driving forces for regulations through out the world. The five commonly substances being monitored are CO, NO2, O3, PM10 and SO2, In particular, the permitted level of pollutant substances in Malaysia is controlled under the Environmental Quality Act 1974 that has specifically defined as the guideline standard of Recommended Malaysian Air Quality Guidelines (RMQG)8. The same enforcement is implemented in developed nation such as the United States. However, much evidence has recorded that the petrochemicals operator has released much higher pollution despite the above enforcement made25.

Much effort has been made by researchers in order to predict the air pollution. Among those substances, NO2 is considered one of the most popular substances used by researchers as research and indicator either involving acute or chronic cases4,23,24,27,29. One of the reason might be due to the usage of natural gas as the primary feedstock to the industry12. Recently practice also has indicated that petroleum refiner has switched away from heavy high sulfur oil towards the natural gas as to reduce SO2

30. The emissions of NO2 may also be significant in Malaysian scenario knowing that natural gas is also known as commonly used fuel by the petrochemicals industry7.

With the above concerned, it is the objective of this paper to predict the significant level of the average on ground concentration of NO2 emitted by point source of petrochemicals plants within the Kertih Petrochemicals Industrial Area. Method - Meteorological Modeling

This analysis applying simulation using air modeling software, the Industrial Source Complex Term


(2)

Version 3 (ISCT3) which is recommended by United States Environmental Agency to perform atmospheric dispersion modeling. This software in cooperates the steady state Gaussian Plume Model and capable to evaluate the pollutant concentration. The simulation covers major input of geographical domain set up, NO2 generators specification and meteorological parameters. Geographical domain set up: The geographical domain is set up covering 20 km x 20 km area surrounding the petrochemicals plants with 1 km grid spacing as dummy receptors. The source location coordinates are identified by matching the longitude and latitude coordinate to AutoCAD coordinate. The average NO2 concentration would be projected to receptors accordingly to the NO2 generator parameters and meteorological parameter. NO2 generator specification; the petrochemical plants: There are several different methods of estimating the emission rate. These are; source sampling, source emission model, surveying, material balance, emission factors and extrapolation. The source sampling is considered the most reliable method6,18. Alternatively, emission rates may be calculated from manufacturer’s specification or directly using process knowledge. Depending on the data availability (Table II), estimation of emissions rate calculated in this research is according to actual sampling data and material balance method of ultimate fuel analysis. With 2 different compositions of data obtained, the point source emission rates of NO2 are estimated accordingly by using the following formulae. 1. NO2 emission rate calculated using Composition 1 based on actual sampling data; the conversion of volumetric flow rate of g/m3 to emission rate of g/s. ER (g/s) = C (g/m3) x A (m2) x V (m/s) Where; ER = Emission rate pollutant C = Concentration of pollutant A = Area of point emission based on

internal diameter V = Velocity

2. NO2 emission rate calculated using Composition 2 based on fuel flow rate using the ultimate fuel analysis19.

ER (g/s) = Qf (kg/hr) x PCf (%) x (MWp/MWf) x (1000/60 x 60)

Where; ER = Emission rate pollutant Qf = Fuel flow rate PCf = Pollutant concentration in fuel N2

= 0.9 %10 MWp = Molecular weight of pollutant

emitted (g/g-mole) NO2 = 46

MWf = Molecular weight of pollutant in fuel (g/g-mole) N2 = 28 The NO2 emission rates estimated are at the

average since the emission concentration (C) and the fuel flow rate (Qf ) are obtained at average value. This analysis assumes that all emission concentration rate and other parameters are constant throughout this study. There are 28 point source emissions included in this analysis. The overview input into the ISCT3 includes the result of emission rate of NO2 calculated using the above method which lead to minimum and maximum input into the ISCT3 (Table III). Meteorological Parameter: Five years of meteorological data (2004 to 2008) from Kuala Terengganu Meteorological Station would be processed into the ISCT3 Software. Kuala Terengganu Meteorological Station located at latitude 05o 23’ N and longitude of 103o 06’ E, which is approximately 95 km north of the study site (Figure 2). This station is selected since no local meteorology station is available beside of its proximity and has similarity meteorological and geography parameters to the study site. As an example, research site is located approximately 2 km away from coastal line which is approximately the same distance of Kuala Terengganu Meteorology Station to coastal line. Both of these sites are influenced by sea and land breeze circulation that create turbulent diffusion5,20 which would influence the concentration of air pollutant15.

The similarity parameters as described would be practical for Kuala Terengganu meteorological data to be used as an input to this study. Malaysian experiences relatively uniform temperature throughout the year with the mean temperature ranging between 26 Co to 20 Co 1,21. Beside that, previous pollutant prediction using air modeling has applied remote meteorological data as far as 500 km off the research site9.

The meteorological parameters were converted to

hourly surface data file (SCRAM MET 144) and daily am and pm mixing height file (SCRAM). Inputs of the hourly surface data file to ISCT3 would provide an overview of the wind blow direction. The maximum am and pm mixing height was applied according to local standard time. Estimation of mixing height value was obtained following Venkatram method under SBL as follows3: h = 2300U*

1.5

Where; h = mixing height U* = friction velocity = kUref / In(Zref/Zo) k = Von Karman constant = 0.4 Uref = wind speed at reference height Zref = reference height for wind = 14 meter Zo = roughness height = 0.15 meter


(3)

The commonly friction velocity applied11,13,14 is the combination of Von Karman constant (k), wind speed (Uref,), reference height of wind (Zref) and surface roughness (Zo). The Zo at 0.15 meter was selected considering the study site topography to be between “Tree Covered” and “Low-density Residential” which is at 0.1 and 0.2 respectively17. These values are also recommended by the U.S. EPA. The Zref at 14 meter refers to the anemometer height for the Kuala Terengganu Meteorology Station.

Combination of hourly surface data file and daily mixing height data file would generate the final meteorological data file as an input to this study. The meteorological data file consists of hourly random flow vector, wind speed, ambient temperature, winds class stability, rural and urban mixing height of the particular year. Sensitivity Analysis and Verification: The final stage of this modeling process would be the verification of the average predicted against the actual NO2 measured which utilizes actual hourly data of NO2 from year 2004 to 2008. The actual NO2 measured locally by the Malaysian DOE located at the Paka Station, which is approximately less than 2 km center of the petrochemical industry. Since no meteorological data available locally, the optimum correlation of the predicted versus the measured NO2 are identified by performing the sensitivity analysis simulations of meteorological data file with the incremental of 5o rotation of wind direction range from 0o to +/-75o. The correlation coefficient formula is defined as follows:

Where; A = actual data P = predicted data σ = standard deviation N = number of data Result and Discussion The wind rotation angle versus correlation coefficient: The following plots (Figure 3) shows the wind rotation angle versus the correlation coefficient obtained with respect to verification of the predicted to the actual data of NO2. It is noted the optimum correlation coefficient achieved at wind rotation angle range from -30o to 45o with respect to year 2008 and 2004 at the value of 0.910 and 0.785. Further detail analyses at the incremental of 1o angle are performed. The wind direction rotation angle with respect to the optimum achievable correlation coefficient maximum value of NO2 and location are summarized accordingly (Table IV). It is noted that wind direction

angles range from -29o to 45o accordingly to year 2008 and 2004 with correlation coefficient ranging from 0.773 to 0.955 according to year 2007 and 2006. The optimum conditions lead to further prediction of the maximum magnitude of the average concentration of on ground NO2. The optimum wind condition are applied accordingly resulting to the NO2 distribution contours within the domain set up of 20 km x 20 km (Figure 4). The analysis revealed that the average maximum predicted monthly NO2 concentration released by point source from petrochemical industry varies from 13.97 ug/m3 to 20.43 ug/m3 with respect to year 2008 and 2007. The annual average range from 6.91 ug/m3 to 8.49 ug/m3 (of year 2008 and 2005) which is lower than either the WHO standard (Table V). The analysis also obtained the predicted location of NO2 to receptors. The distribution of the maximum monthly and annually varies accordingly shows that the yearly coverage is larger than the monthly which is approximately at 5 km2 and 3 km2 respectively (Figure 5). Conclusion

Since the maximum average predicted of NO2 is below the benchmark of the WHO standard, this study concludes that there is no significant effect to pose appreciable likelihood of adverse health effects to the local receptor due to point source emissions contributed by petrochemical industry. References 1. A.D.B., Asia Development Bank and the clean Air Initiatives for Asia Cities (CAI-Asia) Center. Country Synthesis Report Urban air Quality Management – Malaysia (2006) 2. Akimoto H. and Narita H., Distribution of SO2, NOx and CO2 emissions from fuel combustion and industrial activities in Asia with 1° × 1° resolution Atmospheric Environment, 28 (2) (1994) 3. Baklanov, A., Joffre, S.M., Piringer, M., Deserti, M., Middleton, D.R., Tombrou, M, Karppinen, A., Emeis, S., Prior V., Rotach, M.W., Bonafè, G., Baumann-Stanzer K. and Kuchin A., DMi Ministry of Transport and Energy (Copenhagen), Scientific Report 06-06 Towards estimating the mixing height in urban areas. Recent experimental and modeling results from the COST-715 Action and FUMAPEX project (2006) 4. Beelen R., Hoek G., Fischer P., Brandt P.A.V.D. and Brunekreef B., Estimated long-term outdoor air pollution concentrations in a cohort study Atmospheric Environment, 41 (7), 1343-1358 (2007) 5. Carbon B., Good Practice Guide for Atmospheric Dispersion Modeling, Prepared by the National Institute of


(4)

Water and Atmospheric Research, Aurora Pacific Limited and Earth Tech Incorporated for the Ministry for the Environment (2004) 6. C.C.P.A., Canadian Chemical Producers’ Association. Source Characterization Guidelines. Primary Particulate Matter and Particulate Precursor Emission Estimation Methodologies for Chemical Production Facilities (2004) 7. Dark T., Auckland J. and White J., Fuel. The John Zink Combustion Handbook. John Zink Company LLC. (pp.157-188) ISBN Number 0-8493-2337-1 (2001) 8. D.O.E., Department of Environment Malaysia. A guide to Air Pollution Index (API) in Malaysia (2000) 9. Elbir T., Comparison of model predictions with the data of an urban air quality monitoring network in Izmir, Turkey. Atmospheric Environment 37 2149–2157 (2003) 10. EPA U.S., Environmental Protection Agency, U S. Preliminary Characterization Study. Traditional Fuels and Key Derivatives - Advanced Notice of Proposed Rule Making, Identification of Nonhazardous Material That Are Solid Waste (2010) 11. Fatogoma O. and Jacko R.B., A Model to Estimate Mixing Height and its Effects on Ozone Modeling. Atmospheric Environment 36 3699–3708 (2002) 12. G.C.M.,. Gulf Coast Midwest Energy Partners, LLC. Natural Gas Basics (2008)

13. Guan Dexin, Zhu Tingyao, Han Shijie, Friction Velocity U* and Roughness Length Z0 of Atmospheric Surface Boundary Layer in Sparse-tree Land 1 Journal of Forestry Research, 10 (4) (1999) 14. Hsu S.A., Estimating Overwater Friction Velocity and Exponent of Power-Law Wind Profile from Gust Factor during Storm. Journal of Waterway, Port, Coastal, and Ocean Engineering, 129 (4) (2003) 15. Jennings A.A. and Kuhlman S.J., An air pollution transport teaching module based on GAUSSIAN MODELS 1.1 Environmental Modeling & Software, 12 (2-3) 151-160, 1997 (1997) 16. Masters G.M.. Air Pollution. Introduction to Environmental Engineering and Science 2nd Edition International Edition, pp 331. Prentice Hall International Inc New Jersey (1998) 17. McRae, G. J., Goodin, W.R. and Seinfeld, J.H., Development of a second-generation mathematical model for urban air pollution. Model formulation. Atmospheric Environment, 16 679-696. (1982) 18. M.E.I.P.M. Volume III, Mexico Emissions Inventory Program Manual 1996, Volume III – Basic Emission Estimating Techniques, DCN 96-670-017-01, RCN 670-017-20-04. Western Governers’ Association Denver, Colorado and Radian International Sacramento, California (1996)

Fig.1: Research location – Kertih, A major petrochemical

industry site at North-East of Peninsular Malaysia located

Fig.2: Kuala Terengganu Station, 9 5 km north of study site.


(5)

Fig. 3: Trend of correlation coefficient (r) with respect to the incremental of 5o wind rotation angles respectively for year 2004 to 2008 show at ‘0’ and at the optimum condition angle.

19. M..E.I.P.M. Volume IV, Mexico Emissions Inventory Program Manual 1996, Volume IV– Point Source Inventory Development, DCN 96-017-21-05, RCN 670-017-20-04. Western Governers’ Association Denver, Colorado and Radian International Sacramento, California (1996) 20. Melas, D., Ziomas, I. C. and. Zerofos, C. S., Boundary Layer Dynamics in an Urban Coastal Environment Sea Breeze Condition. Atmospheric Environment 29 (24) 3605-3617 (1995)

21. M.M.D.. Malaysian Meteorological Department, General Climate of Malaysia (2010) 22. M.P.A, Malaysian Petrochemicals Association, Asia Petrochemical Industry Conference 2003 Country Report – Malaysia (2003)


(6)

Year 2004 Year2005

Year 2006 Year 2007

Year 2008

Fig. 4: Distribution contours of monthly average on ground NO2 concentration at the optimum wind direction indicate the approximate maximum location.


(7)

Fig.5: Distribution of maximum monthly and yearly coverage of on ground NO2 concentration contributed by point sources of the petrochemicals plants.

Table I

Emissions of the most important six air pollutants showing mobile sources and break down of stationary sources based on U.S. EPA, 1994b, (Masters, 1998)

Table II

Different composition of technical data available from petrochemical plants

Composition 1 Composition 2

(source sampling) (fuel flow rate)

*Emission rate (g/s) x x Gas flow rate (g/m3) √ x

*Release height (m) √ √

*Temperature (oK) √ √

*Exit velocity (m/s) √ √

*Inside diameter (m) √ √

Type of fuel √ √

Fuel flow rate (kg/hr) x √ √ = Data available x = Data not available

*Parameters required for ISCT3 simulation

Pollutants Mobile Stationary Total Transport Fuel Industrial Misc

Combustion Processes

C0 77% 6% 7% 10% 100% Pb 33% 10% 57% - 100%

NOx 45% 50% 4% 1% 100% VOCx 36% 3% 58% 4% 100% PM10 22% 46% 32% - 100% SOx 3% 88% 9% - 100%


(8)

Table III

Overview input range into the ISCT3

Minimum Maximum *Emission rate (g/s) 0.02 8.17 *Release height (m) 15.00 87.00

*Temperature (oK) 310.15 600.15 *Exit velocity (m/s) 4.13 28.00 *Inside diameter (m) 0.52 2.41

able IV

Summary of wind rotation angle with respect to correlation coefficient

Wind

Rotation Angle (o)

Correlation Coefficient (r)

Optimum Year

Predicted

2004 45 0.785 2005 31 0.887 2006 -14 0.955 2007 -18 0.773 2008 -29 0.914

Table V

Summary of maximum predicted NO2 to receptors against standard at the optimum condition

Monthly (ug/m3)

Location at (x,y) 1000m (south west = 0,0 )

Annually (ug/m3)

Location at (x,y) 1000m (south west = 0,0 )

WHO Standard n/a 40 Year Predicted

2004 17.36 16.00,10.00 8.19 15.5,12.0 2005 14.33 15.50, 9.50 8.49 16.0,11.0 2006 16.73 15.50,10.00 8.15 17.0,11.0 2007 20.43 15.50,10.00 7.70 17.0,11.5 2008 13.97 15.00,10.00 6.91 17.0,11.5

23. Neuberger M., Michael G., Schimek M.G., Horak F., Moshammer H., Kundi, M. Frischer T., Gomiscek B., Puxbaum H. and Hauck H., Acute effects of particulate matter on respiratory diseases, symptoms and functions:: epidemiological results of the Austrian Project on Health Effects of Particulate Matter (AUPHEP) Atmospheric Environment, 38 (24) 3971-3981 (2004) 24. Paatero P., Aalto P., Picciotto.S., Bellander T., Castaño G., Cattani G., Josef Cyrys J., Markku Kulmala M, Timo Lanki T. and Fredrik Nyberg F., Estimating time series of aerosol particle number concentrations in the five HEAPSS cities on the basis of measured air pollution and meteorological variables Atmospheric Environment, 39 (12) 2261-2273 (2005) 25. Perin M., Pollution problems offset petrochemical prosperity, studies show, Houston Business Journal, (2003)

26. PETRONAS, PETRONAS Annual Report 2002- 2009 (2009) 27. Scoggins A., Kjellstrom T., Fisher G., Connor J. and Gimson N., Spatial analysis of annual air pollution exposure and mortality Science of The Total Environment, 321 (1-3) 71-85 (2004) 28. Tietenberg T., Stationary Source Local Air Pollution In Environmental and Natural Resource Economic. Pp 366 Addison Wesley (2005) 29. Wong C.M., Ou C.Q., Thach T.Q., Chau Y.K., Chan K.P., Ho S.Y., Chung R.Y., Lam T.H. and Hedley A.J., Does regular exercise protect against air pollution-associated mortality? Preventive Medicine, 44 (5) 386-392 (2007) 30. Zanganeh K.E., Shafeen A. and Thambimuthu K., A Comparative study of Refinery Fuel Oxy-Fuel Combustion Options for CO2 Capture Using Simulation Process Data. Greenhouse Control Technologies, Volume II. (2005)

Documents

The Significance of Point Source Emission