   UNITED STATES ENVIRONMENTAL PROTECTION AGENCY

        ANN ARBOR, MI 48105

September 12, 2007

                                                                        
                                        	      			                    
OFFICE OF 	                                                             
                                                                        
                       	               									             AIR AND
RADIATION

MEMORANDUM

SUBJECT:	Estimation of Particulate Matter Emission Factors for Diesel
Engines on Ocean-Going Vessels

FROM:	Mike Samulski

		Assessment and Standards Division

TO:		Docket EPA-HQ-OAR-2007-0121

	This memorandum describes an analysis of exhaust particulate matter
emission rates for propulsion and auxiliary diesel engines on
ocean-going vessels.  Basic emission rates were determined from in-use
test data collected on a number of ships.  In addition, this memorandum
describes adjustments to the emission rates for fuel sulfur content.

I. Background

	As part of its ongoing efforts to reduce emissions from mobile sources,
EPA is considering new exhaust emission standards for Category 3 marine
engines.  Category 3 engines refer to the large engines (>30
liters/cylinder) typically used on ocean-going vessels.  Currently,
these engines are subject to Tier 1 emission standards for oxides of
nitrogen (NOx).  These Tier 1 standards, which went into effect in 2004,
essentially adopt the NOx standards established through the
International Maritime Organization (IMO).

Earlier this year, the U.S. government submitted a paper to IMO
recommending more stringent NOx standards for ocean-going vessels as
well as new standards for particulate matter (PM) and sulfur oxides
(SOx).  EPA is also developing an Advanced Notice of Proposed Rulemaking
which describes potential new national emission standards for Category 3
marine engines.

	An important part of this rulemaking effort is to calculate the
emissions contribution from Category 3 marine engines.  Since the Tier 1
rulemaking, EPA and others have made significant efforts to better
quantify exhaust emissions from ocean-going vessels.  One area of this
work has been the development of exhaust emission factors for PM.  We
are updating our estimates of PM emission factors for engines on
ocean-going vessels using an expanded set of emission test data.

II. Methodology

Through a literature search, we were able to collect PM test data on
several marine engines operating both on distillate fuel and on residual
fuel.  These data and the associated reference sources are included in
Appendices A and B to this memo.

All of the PM test data were collected under steady-state operation. 
However, the engine operation (load and speed) varied from test to test.
 For several of the engines, emission test data were collected at
multiple load points.  In this analysis, we averaged the modal test data
to determine a single PM emission rate for each engine.  We then
averaged the emission rates for the engines to develop PM emission
factors.

In some cases, engines were tested at very low power operation.  A
characteristic of reporting emissions on a brake-specific basis is that
the value approaches infinity as power approaches zero.  For this
reason, test modes at very low loads were excluded from this analysis. 
A total of four test modes were excluded for this reason.  Fuel sulfur
concentration has a significant effect on PM emissions.  Therefore, test
data were excluded where the fuel sulfur was not reported or was
reported as abnormally low for a residual fuel.  A total of five engines
were excluded for this reason.

For two of the test reports, PM emissions were reported in fuel-specific
terms rather than brake-specific terms.  In these cases, we used an
assumed brake-specific fuel consumption rate of 200 g/kW-hr to convert
the emission factors to units of g/kW-hr.  This is the same fuel
consumption rate that was used by the IMO in developing a brake-specific
SO2 limit equivalent to the 1.5% fuel sulfur requirement in Sulfur
Emission Control Areas.  Where appropriate, the appendix presents the
emission data both in terms of g/tonne of fuel consumed and g/kW-hr.

The test data is divided between tests on distillate fuel and tests on
residual fuel.  Higher PM emission rates were expected for residual fuel
compared to distillate fuel for two reasons.  First, residual fuel tends
to have much higher fuel sulfur than distillate fuel.  A portion of the
fuel sulfur forms sulfate PM in the exhaust.  This is discussed in more
detail below.  The second reason is related to general fuel quality. 
Residual fuel has a higher density, poorer ignition quality, and
contaminants such as ash and heavy metals that can increase PM
emissions.  One study measured PM on a marine engine operating on low
sulfur residual and distillate fuel.  This study showed much higher
non-sulfate PM on the residual fuel than the distillate fuel.

Under the current modeling, different PM emission factors are used for
medium-speed and slow-speed engines.  Therefore, we considered that
medium-speed and slow-speed engines may have significantly different PM
emission factors.  However, based on the data in Appendix A, no
significant difference was observed.  Consistent with the existing
analysis, we use the same emission factors for auxiliary and propulsion
engines.  Therefore, the test data were combined to calculate a single
emission factor for each fuel type.  Table 1 presents the average PM
emission rate and fuel sulfur level.  Descriptive statistics are
presented in Appendix C.

Table 1:  95 Percent Confidence Range

for CI Marine Fuel Sulfur and PM Data

	Distillate	Residual

Sulfur %	0.24 +/- 0.20	2.46 +/- 0.27

PM g/kW-hr	0.23 +/- 0.08	1.35 +/- 0.24



III. Fuel Sulfur Adjustments

	As mentioned above, the sulfur content of the fuel affects PM emissions
through the formation of sulfate PM in the exhaust.  To estimate the
relationship between PM emissions from marine vessels, we developed an
equation that corrects PM emissions for changes in fuel sulfur level. 
We use the nominal PM emission rates and fuel sulfur levels in Table 1
and correct these values for different fuel sulfur levels using the
following equation:

PMEF = PMNom + [(SAct – SNom) × BSFC × FSC × MWR × 0.0001]

	where:

		PMEF 	= PM emission factor adjusted for fuel sulfur

		PMNom	= PM emission rate at nominal fuel sulfur level

			= 0.23 g/kW-hr for distillate fuel, 1.35 g/kW-hr for residual fuel

		SAct	= Actual fuel sulfur level (weight percent)

		SNom	= nominal fuel sulfur level (weight percent)

			= 0.24 for distillate fuel, 2.46 for residual fuel

		BSFC	= fuel consumption in g/kW-hr

= 200 g/kW-hr used for this analysis

		FSC	= percentage of sulfur in fuel that is converted to direct sulfate
PM

			= 2.247% used for this analysis

		MWR	= molecular weight ratio of sulfate PM to sulfur

= 224/32 = 7 used for this analysis

	Rates of sulfur oxides and sulfate PM emitted from marine engines are
directly related to the amount of sulfur in the fuel.  Although the
majority of the sulfur in the fuel forms sulfur oxides in the exhaust
(primarily SO2), a small but significant fraction forms direct sulfate
PM.  In the NONROAD model, EPA uses a factor of 2.247 percent to
represent the fraction of fuel sulfur that forms direct sulfate PM in
the exhaust of diesel engines.  This factor was developed empirically
based on test data from several nonroad engines operating on distillate
fuel with varied sulfur levels.  One study showed similar fractions of
sulfur from the fuel found on particles in the exhaust for marine
engines operating on high sulfur residual fuel.  Therefore, we use the
established sulfur to sulfate conversion factor for this analysis.

	In engine exhaust, sulfate is bound with water to form sulfuric acid
(H2SO4).  Other than in a chemical laboratory, sulfuric acid does not
exist without a further bond with water.  The usually accepted amount of
water binding to sulfuric acid in diesel exhaust is seven water
molecules per molecule of sulfuric acid.,,  The actual ratio can vary
depending on what else is present with the PM and also depending on
humidity.  H2SO4 has a molecular weight of 98 and water has a molecular
weight of 18.  Using a ratio of seven water molecules per molecule of
sulfuric acid, we calculate that the molecular weight ratio of direct
sulfate PM to sulfur from the fuel is (98 + 7 × 18)/32 = 224/32 = 7.

	A sample calculation is presented below for adjusting the nominal PM
emission rate for changes in fuel sulfur.  In this example, PM is
calculated for an engine operating on residual fuel with a fuel sulfur
content of 2.7 percent:

PMEF 	= PMNom + [(SAct – SNom) × BSFC × FSC × MWR × 0.0001]

		= 1.35 g/kW-hr + [(2.7% - 2.46%) × 200 g/kW-hr × 2.247% × 7 ×
0.0001]

		= 1.35 g/kW-hr + 0.08 g/kW-hr

		= 1.43 g/kW-hr

IV. PM Emission Rates

	As described above, nominal PM emission rates were determined
separately for distillate and residual fuel based on average measured PM
from several engines.  The nominal fuel sulfur rates associated with the
PM data were determined using averages from the same data set.  Because
the sulfate portion of the PM is directly related to the fuel sulfur
content, an equation was developed to adjust the nominal PM emission
rates based on fuel sulfur.  These PM emission rates are presented in
Figure 1, for distillate and residual fuel as a function of fuel sulfur
content.  In addition, this figure presents the actual PM test data used
to develop the nominal PM emission rates.

Figure 1:  Marine Engine PM Data versus Fuel Sulfur Concentration

		

APPENDIX A:  PM Emission Test Data

Table A-1:  PM Emission Data for CI Marine Engines Operating on
Distillate Fuel

Table A-2:  PM Emission Data for CI Marine Engines Operating on
Residual Fuel

APPENDIX B:  PM Emission Data Sources

Cooper, D., “Exhaust Emissions from High Speed Passenger Ferries,”
Atmospheric Environment 35 (2001) 3817-3830, March 8, 2001.

Cooper, D., “Exhaust Emissions from Ships at Berth,” Atmospheric
Environment 37 (2003) 3817-3830, May 12, 2003.

Entec UK Limited, “Quantification of Emissions from Ships Associated
with Ship Movements

between Ports in the European Community,” July 2002.

Environment Canada, “Port of Vancouver Marine Emissions Test
Project,” ERMD Report #97-04, 1997.

Fleischer, F., Ulrich, E., Krapp, R., Grundmann, W., “Comments on
Particulate Emissions from Diesel Engines when Burning Heavy Fuels,”
International Council on Combustion Engines Congress, 1998.

Maeda, K., Takasaki, K., Masuda, K., Tsuda, M., Yasunari, M.,
Measurement of PM Emission from Marine Diesel Engines,” CIMAC Paper
107, International Council on Combustion Engines Congress, 2004.

MAN B&W, “Emission Measurement Results; A.P. Moeller – Sine
Maersk,” February, 2004.

Miller, et. al., “Measurements of NOx and Particulate Matter from the
Large Propulsion Engines on Ocean-going Vessels,” presentation for
Clean Ships: Advanced Technology for Clean Air, February 7-9, 2007.

Wright, A., “Marine Diesel Engine Particulate Emissions,” Trans
IMarE, Vol. 109, Part 4, pp. 345-366, May 20, 1997.

APPENDIX C:  Descriptive Statistics

Table C-1:  Descriptive Statistics for Data Collected on Distillate and
Residual Fuel

 Kasper, A., Aufdenblatten, S., Forss, A., Mohr, M., Burtscher, H.,
“Particulate Emissions from a Low-Speed Marine Diesel Engine,”
Aerosol Science and Technology, 41:24–32, 2007.

 Environmental Protection Agency, “Exhaust and Crankcase Emission
Factors for Nonroad Engine Modeling--Compression-Ignition,”
EPA420-P-04-009, NR-009c, April 2004.

 Lyyränen, J., Jokiniemi, J., Kauppinen, E., Joutsensaari, J.,
“Aerosol Characterisation in Medium-Speed Diesel Engines Operating
with Heavy Fuel Oils,” Journal of Aerosol Science, Vol. 30, No. 6, pp.
771-784, 1999.

 Fujii, T., Ikezawa, H., Kotani, Y., “A Study of the Analysis of PM
Components with CR-DPF,” Society of Automotive Engineers Paper
2002-01-1686, 2002.

 Environmental Protection Agency, “Size Specific Total Particulate
Emission Factors for Mobile Sources,” EPA Report 460/3-85-005, August
1985.

 Environmental Protection Agency, “Emission of Sulfur-Bearing
Compounds from Motor Vehicle and Aircraft Engines, A Report to
Congress,” EPA Report 600/9-78-028, August 1978.

 PAGE   

 PAGE   9 

