tpp kosovo b pollution mitigation

60R05429.01-Q070-011

February 6, 2006

European Agency for Reconstruction

Contract nr 04KOS01/03/009

Lot 2

PRE-FEASIBILITY STUDY FOR POLLUTION MITIGA-TION MEASURES AT KOSOVO B POWER PLANT

Draft final

Lot 2, Kosovo B February 6, 2006 Pollution mitigation Page 1 (73)

EXECUTIVE SUMMARY This pre-feasibility study of pollution mitigation measures at the Kosovo B thermal power plant, operated by Korporata Energjetike e Kosoves (KEK) in Kosovo, was prepared by Electrowatt-Ekono Oy, as commissioned by the European Agency for Reconstruction (EAR) as part of the project “Pre-feasibility Studies for the New Lignite Fired Power Plant and Pollution Miti-gation Measures at Kosovo B Power Plant”. This report contains the essential findings of the environmental audit of the Kosovo B power plant, descriptions of the proposed mitigation measures and an Environmental Action Plan (EAP) including cost estimates. The aim of the pre-feasibility study is im-provement of the environmental performance of Kosovo B and the environ-ment surrounding the site.

The study has been carried out with complete EU-compliance of Kosovo B as a prerequisite as the Law on Environmental Protection states that environ-mental protection in Kosovo shall be based on the gradual introduction of European Union standards. Possible transition periods for future EU compli-ance have not been taken into account.

It is assumed that Kosovo B will be operational at least for the next 15 years. Its load will be close to base load. If the Kosovo B units shall continue unlim-ited operation well into the future it will be necessary to improve the level of environmental mitigation measures of the plant significantly. SO2, dust and NOx emissions will have to be reduced, which will require investments in abatement technology.

The Kosovo B power plant's stack emissions do not constitute any major problem with regards to air quality, based on findings of the study. The air quality situation in the region is however alarming, mainly because of the dust emissions from the ash landfills, Kosovo A power plant and the adjoin-ing lignite drying plant. Compliance with the EU LCP-directive in the future and reduction of fugitive dust emissions from the ash disposal area will im-prove the situation. A visible improvement in Kosovo B (and especially for Kosovo A plant) might be a decisive factor in getting the public approval of the new power plant development.

Waste and wastewater management installations and practices will have to be improved to comply with Kosovan legislation, EU-Directives and best envi-ronmental practices. The current ash disposal site can not be utilized for much longer, and a new site has to be found. Old abandoned lignite quarries are the most logical and environmentally sustainable places for this, and the Mirash mine is to be used for this purpose in the near future. The disposal should be done in compliance with the EU directive 1999/31 on waste landfills regard-ing drainage layers and covering with soil and other required measures.

Aftercare of the current ash landfill is required to mitigate impacts on the river Sitnica, groundwater and ambient air quality.

The current overall energy efficiency of about 30 % must be improved for en-vironmental and economical reasons closer to the initial 39 - 40 % design per-formance.


A monitoring programme in line with the requirements of various EU direc-tives regarding air quality, water quality and emissions must be established to demonstrate compliance. Monitoring of emissions into the air from the power plant will require investments in continuous monitoring equipment in order for compliance with the monitoring requirements of the EU LCP-directive to be achieved.

The schedule for the proposed measures can not be determined at this stage. Kosovan legislation allows gradual implementation of EU regulations. Possi-ble EU accession negotiations in the future may also touch this subject, as has been the case in other accession countries.

The Environmental Action Plan has been formulated based on both what can be achieved in the short term without major investments, and what is neces-sary to implement in order to obtain compliance with Kosovan legislation and EU directives. The Environmental Action Plan summary is presented below:

Chapter Measure Investment needs (estimate)

Annual operating costs (estimate)

6 month follow-up program to de-termine which SO2 abatement tech-nique is sufficient

0,8 MEUR -

SO2 emissions abatement, alterna-tive methods:

• Dry lime injection into the boiler

• Semi-Dry flue gas desul-phurization. SDA

• Semi-Dry flue gas desul-phurization. Integrated desulphurization process

• Wet flue gas desulphuriza-tion

10 MEUR 50 MEUR 40 MEUR 110 MEUR

3,4 MEUR 6,3 MEUR 6,0 MEUR 9,6 MEUR

Dust emissions abatement 20 MEUR 1,2 MEUR

9.3.1 Air pollution abate-ment

NOx emissions abatement (Low-NOx-burners)

26 MEUR -

Waste water treatment improvement through establishment of appropri-ate treatment installations. Sanitary waste waters are assumed to be led to municipal sewerage network.

In the order of 0,4 - 0,6 MEUR.

Can not be esti-mated at this stage

9.3.2 Water pollution abatement

Recycling of waste water Can not be esti-mated as exact plans and methods are not known at this time

-

Preparation of a waste management plan

25 000 - 35 000 EUR

-

Utilisation study for fly ash and bottom ash

Low -

9.3.3 Waste management

Construction of a new ash disposal site

In the order of 3 - 5 MEUR in-cluding design, but not including land acquisition costs

-


Removal of all waste from the site in a sustainable way.

Low -

Handling and storage of hazardous waste according to regulations in bunded and roofed area.

Investment needs 10 000 - 20 000 EUR for repair of existing storage

Development and implementation of an aftercare program for the ash disposal site

In the order of 10 - 12 MEUR includ-ing design (re-moval cost in the range of 50 MEUR)


HFO storage tanks and oil pipes inspection and repair if necessary

Low Low 9.3.4 Fuel storage

Inspection and repair if necessary of secondary retaining basin of the HFO tank

Low Low

Soil and groundwater contamina-tion surveys

70 000 - 80 000 EUR

- 9.3.5 Soil and groundwa-ter contamination

Implementation of a remediation project of contaminated soil and groundwater


-

9.3.6 Monitoring Elaboration of monitoring system for emissions into air and water.

0,2 - 0,4 MEUR 20 000 - 40 000 EUR

Development of an Environmental Management System

50 000 - 75 000 EUR

- 9.3.7 Environmental management and aware-ness Implementation of an Environ-

mental education and awareness campaign

30 000 - 40 000 EUR

-


Contents 1 INTRODUCTION ................................................................................................................... 6 2 NATURE OF THE PROJECT TO BE SUPPORTED......................................................... 7

2.1 PROJECT OBJECTIVES.................................................................................................................. 7 2.2 BENEFICIARIES........................................................................................................................... 7

3 METHODS AND MATERIAL .............................................................................................. 8 4 BASIC INFORMATION ABOUT “KOSOVO B “ ............................................................ 11

4.1 FACILITY AND SITE HISTORY .................................................................................................... 11 4.2 DESCRIPTION OF PROCESSES, FACILITIES AND ASSETS.............................................................. 12 4.3 FACILITY LOCATION AND DESCRIPTION OF NATURAL ENVIRONMENT ....................................... 13 4.4 KEY ENVIRONMENTAL, HEALTH AND SAFETY ASPECTS OF THE FACILITY ................................. 15

5 KOSOVO B ENVIRONMENTAL, HEALTH AND SAFETY MANAGEMENT .......... 16 5.1 EHS POLICIES AND PRACTICE................................................................................................... 16 5.2 ORGANIZATION OF EHS MANAGEMENT ................................................................................... 16 5.3 CONTINGENCY PLANNING AND EMERGENCY PROCEDURES....................................................... 17 5.4 STAFF TRAINING AND SUPERVISION.......................................................................................... 17

6 KOSOVO B ENVIRONMENTAL PERFORMANCE ...................................................... 18 6.1 NATIONAL REGULATORY REQUIREMENTS ................................................................................ 18 6.2 APPLICABLE EU/OTHER REQUIREMENTS AND STANDARDS ...................................................... 18 6.3 INPUTS, PRODUCTS AND RELEASES ........................................................................................... 19 6.4 PROCESS EFFICIENCY................................................................................................................ 32 6.5 GENERAL HOUSEKEEPING ISSUES ............................................................................................. 33 6.6 HAZARDOUS MATERIALS MANAGEMENT .................................................................................. 34 6.7 WASTE WATER MANAGEMENT ................................................................................................. 35 6.8 WASTE MANAGEMENT............................................................................................................. 38 6.9 SOIL, SURFACE WATER AND GROUNDWATER CONTAMINATION ................................................ 41 6.10 COMPLAINTS ............................................................................................................................ 42 6.11 ENVIRONMENTAL CLAIMS ........................................................................................................ 43 6.12 CURRENT ENVIRONMENTAL EXPENDITURE............................................................................... 43

7 HEALTH AND SAFETY PERFORMANCE ..................................................................... 43 7.1 NATIONAL REGULATORY REQUIREMENTS ................................................................................ 43 7.2 APPLICABLE EU/INTERNATIONAL REQUIREMENTS AND STANDARDS ....................................... 43 7.3 KEY HEALTH AND SAFETY ISSUES ............................................................................................ 43 7.4 PREVIOUS EMERGENCY SITUATIONS AND ACCIDENTS............................................................... 45 7.5 CONTROL OF MAJOR ACCIDENT HAZARDS ................................................................................ 45 7.6 CURRENT HEALTH AND SAFETY MONITORING PRACTICE .......................................................... 45

8 SUMMARY OF REGULATORY COMPLIANCE STATUS........................................... 46 8.1 GENERAL ................................................................................................................................. 46 8.2 ENERGY EFFICIENCY ................................................................................................................ 46 8.3 RE-USE OF SUBSTANCES ........................................................................................................... 47 8.4 AIR PROTECTION ...................................................................................................................... 47 8.5 WATER PROTECTION ................................................................................................................ 53 8.6 SOIL PROTECTION..................................................................................................................... 54 8.7 WASTE MANAGEMENT ............................................................................................................. 54 8.8 NOISE PROTECTION .................................................................................................................. 55 8.9 HEALTH AND SAFETY ............................................................................................................... 56 8.10 PCB ......................................................................................................................................... 56 8.11 ENVIRONMENTAL MONITORING AND INFORMATION................................................................. 57

9 CONCLUSIONS AND RECOMMENDATIONS............................................................... 58


9.1 CONCLUSIONS .......................................................................................................................... 58 9.2 KEY RISKS AND LIABILITIES OF THE PROJECT ........................................................................... 59 9.3 PROCESS EFFICIENCY AND ENVIRONMENTAL OPPORTUNITIES .................................................. 59 9.4 ENVIRONMENTAL ACTION PLAN ............................................................................................... 60

10 APPENDICES ....................................................................................................................... 66 APPENDIX 1. SUPPORTING DOCUMENTATION. ............................................................................................... 66 APPENDIX 2. LAYOUT OF KOSOVO B............................................................................................................. 68 APPENDIX 3. AERIAL PHOTO OF KOSOVO B................................................................................................... 69 APPENDIX 4. TECHNICAL REPORT ON SO2, NOX AND DUST EMISSIONS MITIGATION MEASURES AT KOSOVO B

TPP. ........................................................................................................................................... 0 APPENDIX 5. DISPERSION OF EXHAUST GASES FROM KOSOVO B POWER PLANT IN OBILIC, KOSOVO. FINAL

REPORT....................................................................................................................................... 0 APPENDIX 6. PHOTO LOG................................................................................................................................. 0

ABBREVIATIONS EAR European Agency for Reconstruction EBRD European Bank for Reconstruction and Development EHS Environmental, Health and Safety EIA Environmental Impact Assessment ERO Energy Regulatory Office ESP Electrostatic precipitator EU European Union KEK Korporata Energjetike e Kosovës KTA Kosovo Trust Fund MEM Ministry of Environment and Mining MESP Ministry of Environment and Special Planning PGD Power Generation Division UNMIK United Nations Mission in Kosovo WHO World Health Organization Disclaimer

The work reported herein has been conducted in connection with a pre-feasibility study of pollution mitigation measures at the Kosovo B power plant in Kosovo. The work has been carried out by an experienced project team in accordance with our high standard of professional services and best professional judgement and competence of the team.

This report is based on documentation disclosed to us by the target entities, information obtained from public sources and orally disclosed information, all acquired in August - October 2005. Hence, the report reflects the situation and the information available at that time. We have endeavoured to confirm the accuracy of the information disclosed to us from alternative sources as far as practicable.

Except as set forth in the agreement with our client and in relation to him, no representation or warranty is included or implied by Electrowatt-Ekono Oy for the completeness, accuracy or reliability of the information contained in this report.


1 INTRODUCTION This pre-feasibility study of pollution mitigation measures at the Kosovo B thermal power plant, operated by Korporata Energjetike e Kosoves (KEK) in Kosovo, was prepared by Electrowatt-Ekono Oy, as commissioned by the European Agency for Reconstruction (EAR) as part of the project “Pre-feasibility Studies for the New Lignite Fired Power Plant and Pollution Miti-gation Measures at Kosovo B Power Plant” (Contract No 04KOS01/03/009). This report contains the essential findings of the environmental audit of the Kosovo B power plant, descriptions of the proposed mitigation measures and an Environmental Action Plan (EAP) including cost estimates. The aim of the pre-feasibility study is improvement of the environmental performance of Kosovo B and the environment surrounding the site.

The pre-feasibility study was prepared by Mr. Mika Pohjonen, Director and Mr. Thomas Bonn, Senior Consultant in Management Consulting at Electro-watt-Ekono Oy, Finland, and by Mr. Yrjö Riionheimo. Mr. Mustafa Neziri of Kosova Invest assisted and acted as interpreter during the interviews and site visits.

Contact: Electrowatt-Ekono Oy Mika Pohjonen or Thomas Bonn P.O. Box 93 (Tekniikantie 4 A) FI-02151 Espoo Finland Tel. +358 9 469 11 Fax +358 9 469 1981

E-mail: [email protected]


2 NATURE OF THE PROJECT TO BE SUPPORTED

2.1 Project objectives The objective of the Lot 2 “Pollution Mitigation Measures at Kosovo B Power Plant” project was to review the environmental pollution situation at the Kosovo B power plant and to propose and describe appropriate mitigation measures. This work also included the measurement of the flue gas emissions (dust, SO2 and NOx) of unit B1 and the evaluation of the impact of the flue gas emissions on ambient air quality by a mathematical dispersion model. Based on the findings concrete mitigation measures with cost estimates have been recommended in this report to fulfil the legal and other regulatory re-quirements and to improve the environmental performance of Kosovo B in general. These recommendations with their cost estimates have been formu-lated to be used for further investment decision making. The findings of this study can also be used to strengthen the overall environmental management and performance of Kosovo B and KEK.

The pre-feasibility study aims at improving the environmental performance of Kosovo B and thus providing a better environment for the people living in the areas which are affected by the emissions or discharges of Kosovo B.

This report gives information on the requirements of EU Directives on envi-ronmental performance level, and thus the project is in line with the accession policy of the region.

The project has been executed in close cooperation with the beneficiaries in order to transfer knowledge both about the execution of environmental audits and environmental aspects of power generation. The environmental mitiga-tion project for Kosovo B will also provide the Ministry of Energy and Min-ing (MEM) & Korporata Energjetike e Kosoves (KEK) with the necessary tools and methods for continued monitoring of the environmental perform-ance of the Kosovo B units.

2.2 Beneficiaries The direct beneficiaries of the project are the Ministry of Energy and Mining (MEM), Korporata Energjetike e Kosoves (KEK), the Energy Regulatory Of-fice (ERO), Ministry of Environment and Special Planning (MESP) and Kos-ovo Trust Agency (KTA).


3 METHODS AND MATERIAL Electrowatt-Ekono Oy has carried out the assignment as described in the Terms of Reference and our technical proposal. Electrowatt-Ekono Oy has been responsible for the administrative management of the project and quality assurance. Mr Tuomo Marjokorpi from Electrowatt-Ekono has acted as Pro-ject Manager. Mr Mika Pohjonen is the key expert for the environmental is-sues, Mr Yrjö Riionheimo is the key expert in flue gas cleaning and combus-tion of lignite related issues and Mr Thomas Bonn is the environmental ex-pert. During the execution of the studies, site surveys and emission measure-ments back-up and special expertise personnel from Electrowatt-Ekono Oy has been utilized as listed in this chapter. All these experts have been pre-sented to the Client for approval prior to their assignment for this particular project.

Emission measurements were carried out in accordance with the appropriate international standards by the accredited measurement laboratory of Electro-watt-Ekono Oy. The Flue gas dispersion modelling was carried out by the Finnish Institute of Meteorology, which is an internationally reputable insti-tute in dispersion modelling.

Mr. Mustafa Neziri of Kosova Invest is the local representative of Electro-watt-Ekono Oy. He and his company assisted in making contacts, arranging practical issues and translating the final deliverables into Albanian language.

MEM was the main coordinator of the activities in Kosovo and also provided an office room for the consultants.

The environmental evaluation of the operations and performance of Kosovo B is based on general information of the environmental legislation and current situation in Kosovo, on information regarding the operations of KEK as far as it has been accessible to the consultant, an on-site environmental audit of Kosovo B and discussions with the personnel of KEK, MEM and MESP. The environmental audit was carried out based on EBRD Environmental Proce-dures (28.7.2003) and Electrowatt-Ekono company methodology based on in-ternational standard ISO 14015:2001 and ASTM standards E1527-05 and E1903-97(2002).

A number of reports elaborated by other consultants in the period after the conflict have been reviewed and information contained in these reports has been a significant source for this study (see Chapter 10 Appendices).

The discussions and site visits took place 17-18.8.2005 and 19-23.9.2005.

Kosovo B, visit August 18th, 2005

Discussions were held with the following persons:

Mr. Shefqet Avdiu Manager of Power Generation Division, Korporata Energjetike e Kosovës Ms. Leonora Hysenaj Project Officer, Kosovo Trust Agency, Energy Division


Mr. Ibrahim Krasniqi Chief of Project, Strategy and Policy Division, Ministry of Energy and Mining Mr. Nail Reshidi Director, Strategy and Development Department, Ministry of Energy and Mining Mr. Naim Imeraj Head of Electricity Division, Ministry of Energy and Mining Mr. Baskhim Gjurgjeala Head of Analysis Services, Kosovo B Ms. Merita Osmani-Ejupi Officer, Environmental Protection Division, Ministry of Energy and Mining

From Electrowatt-Ekono the following persons attended the discussions:

Mr. Tuomo Marjokorpi Project manager - Electrowatt-Ekono Mr. Mika Pohjonen Environmental expert – Electrowatt-Ekono Ms. Katariina Simola Electricity market expert – Electrowatt-Ekono Mr. Johan Lindholm Financial expert – Electrowatt-Ekono Mr. Mustafa Neziri General Manager – Kosova Invest, Electrowatt-Ekono local rep.

Kosovo B, visit September 19th - 23rd, 2005 Discussions were held with the following persons:

Mr. Baskhim Gjurgjeala Head of Analysis Services, Kosovo B Mr. Agim Morina Engineer responsible for environment, Analysis Services, Kosovo B Mr. Isak Zhitia Executive Office for Industrial Safety and Fire Protection, Kosovo B Ms. Merita Osmani-Ejupi Officer, Ministry of Energy and Mining, Environmental protection division Mr. Baton Begolli Director, Ministry of Environment and Spatial Planning, Water Department Ms. Nezakete Hakaj Head od Division, Ministry of Environment and Spatial Planning, Environmental Protection Division Mr. Ahmet Ahmeti Chief of unit, Ministry of Environment and Spatial Planning, Unit for Industrial Pollution Mr. Behxhet Shala Ministry of Environment and Spatial Planning, Environmental Protection Division, EIA Sector

From Electrowatt-Ekono the following persons attended the discussions:

Mr. Tuomo Marjokorpi Project manager - Electrowatt-Ekono Mr. Thomas Bonn Senior Consultant - Electrowatt-Ekono Mr. Mustafa Neziri Gen. mgr. – Kosova Invest, Electrowatt- Ekono local rep.

At Kosovo B personnel from the Analysis Services department were inter-viewed about environmental, health and safety issues. After this a compre-hensive walk-through audit was made. A second meeting with personnel deal-


ing with environmental and health and safety issues was held the next day to clarify some issues. Unit B1 was in overhaul during the latter site visit, and was started again on October 7th, 2005.

The observations made during the walk-through audits are not reported here in great detail, but are rather used as a background to form a general view of the environmental performance of the company and order of importance of various environmental aspects related to its operations. A photo log has been attached to this report.

The key-expert on the boiler emissions, Mr. Yrjö Riionheimo, visited the plant in September 26th - 30th, 2005. Unit B1 was still not in operation and it was possible to inspect also the internal condition of the B1 electrostatic pre-cipitator. After the inspections the experts formulated the detailed programme for the flue gas emission measurements.

The flue gas emission measurements were carried out October 17th - 21st, but due to a breakdown of the slag conveyor of unit B2, resulting in a shutdown of 10 days, only emissions from unit B1 were measured. During the meas-urements fuel and ash samples were taken for detailed laboratory analysis.

The environmental team of the Consultant has presented preliminary ideas on the actions required at Kosovo B in the Environmental Action Plan included in this report. The EAP also includes preliminary cost estimates for the pro-posed actions.


4 BASIC INFORMATION ABOUT “KOSOVO B “

4.1 Facility and site history The Kosovo B power plant is located in the municipality of Obilic, about 3 km from the city limit of Prishtina. The average altitude is around 550 m above sea level.

Kosovo B is one of two power plants operated by KEK. The other power plant is Kosovo A, situated some 2 km to the SE of Kosovo B. In addition, the public enterprise KEK operates the lignite mines providing fuel for the power plants. Kosovo A and Kosovo B are the only thermal power plants in Kosovo. There is one small 35 MW hydropower station and four district heat-ing systems, with a combined capacity of 200 MW, which are fuelled with lignite and imported fuel oil.

KEK is a state owned company, but as the future status of the province of Kosovo is unclear, the future ownership is not fully clarified. Temporarily KEK belongs to the Kosovo Trust Agency, which is established to operate state properties until the situation is clarified.


FIGURE 4/1 Map of the area showing location of Kosovo B (Carl Bro Group 2003).

4.2 Description of processes, facilities and assets Kosovo B consists of two units, B1 and B2, with thermal capacity of about 850 MWth each. The original design was for a capacity of 339 MWe gross per unit. Currently the total net maximum capacity is about 285 MWe per unit. The Kosovo B power plant was commissioned in 1983-84. Both units have logged less than 100 000 operating hours since their commissioning. The Kosovo B power plant has the following main installations:

• Lignite coal storage area with a capacity of approximately 360 000 t, i.e. sufficient for 18 days at maximum daily consumption

• HFO storage tank • Belt conveyor for lignite transport to boilers • Plant for preparation of process water (mainly cooling tower make-up) • Two electrical precipitators (ESP's) for collection of fly ash at each unit

and temporary storage facilities (two fly ash silos) • One common 182 meters high stack for both units, originally designed for

220 meters • Collection and slurry transport systems for fly ash • Disposal site for ashes • Wastewater treatment plant

All installations are relatively old and have been badly maintained until re-cently. The situation has been improved in recent years and many new com-ponents have been installed to maintain operation, but only to a limited extent to improve the environmental performance. The EAR has spent € 175 million on the refurbishment of the plant since 1999.

Supply of heat for district heating in Prishtina was planned and the first transmission pipes constructed in the late 1990’s, but after the conflict this project seems to have been abandoned.

Kosovo B uses low-calorific lignite from the nearby surface mine Bardh for fuel. Lignite from another surface mine, Mirash, is only occasionally used at Kosovo B. The mines are located in the near proximity and they are operated by KEK. The existing mines can only supply these Kosovo A and B power plants up to 2008 and thereafter new sources have to be found. The develop-ment of lignite mining in the Sibovc deposit, just north of the existing mines is planned for the near future. Due to lack of financing and bad maintenance the current mined lignite is not of the expected quality. Lignite from the Bardh mine is slightly better in quality than from the Mirash mine.

The transportation of lignite from the mines to the coal yard is performed by open belt conveyors. No sorting, classification or homogenisation of the lig-nite is done either at the mine or at the power plant lignite yard. Lignite is then transported to the power plant by a closed belt conveyor and distributed by internal belt conveyor systems to lignite silos at each unit for drying and pulverizing before being injected into the boilers mixed with air. The daily


lignite supply to Kosovo B is normally between 5 000 - 20 000 t/d, while daily consumption is max 20 000 t/d with full capacity of both boilers. Typi-cally the plant is taking the lignite straight from the mine to the boilers by-passing the fuel yard.

Light fuel oil and heavy fuel oil is used for start-ups and heavy fuel oil for support firing of the boilers if the quality of lignite is low. Transportation of oils, chemicals and equipment is done by trucks.

The pollution abatement equipment is mainly from the time the units were erected. They are badly maintained and of insufficient capacity considering the higher quantities of fuel currently used. The electrostatic precipitators of unit B2 were rehabilitated by the Polish company Elvo in 2003. No other ma-jor investments in pollution abatement equipment are known to have been made. The ESP's of unit B1 are planned to be rebuilt in 2006.

4.3 Facility location and description of natural environment

4.3.1 General description The KEK industrial site is situated about 10 km from the city centre of Prishtina. The most important activities at this site from an environmental point of view are mining of lignite coal and power generation.

There have been activities at the site related to exploitation of the lignite re-sources for almost 100 years and large-scale operation as it is seen today with open cast mining and power generation has been going on for about 40 years without much concern for the impacts on the environment.

The impact of the previous activities is today seen as contaminated soil and infiltration of trace metals and salts in the soils resulting in contamination of rivers and groundwater. Hence, the current contamination of rivers and groundwater is not only related to the current activities and the contamination will continue for a long time even if the activities were stopped if the site is not thoroughly cleaned up.

There is no other heavy polluting industry in the area. The nearest industry is in Mitrovica about 30 km north-northwest of the site. The second main pol-luter in the area is traffic. Also pollution from small generators is currently significant.

According to literature from the former Yugoslavia the territory of Kosovo only received limited background pollution from other territories and it can thus be concluded that the current poor air quality in Kosovo originates from the activities in Kosovo and that implementation of appropriate measures in Kosovo can significantly mitigate the problems.

4.3.2 Air quality Kosovo A and B power plants and the mines are today, after the lead-smelter in Mitrovica has ceased operation, the biggest polluters of the air in Kosovo,


but the current poor air quality is also caused by traffic and lack of street cleaning.

In addition to fulfilling the requirements regarding specific emissions (mg/Nm3) from the plants, it is necessary also to fulfil the requirements re-garding ambient air quality (µg/m3).

The air quality in the area is measured by the Institute of Public Health and by INKOS but the measuring equipment is said not to be calibrated correctly. The published 2001 immission results from INKOS (an institute within KEK) are shown below (average values of air; measured location: industrial zone of KEK)(Kosovo - State of The Environment Report):

Polluter Jan. Feb. Mar. April May June July Aug. Sep. Oct. Nov. Dec. SO2, µg/m3

22,43 19,0 18,04 17 16,7 13,32 25,9 13,14 23 12 19 20

Smut, µg/m3

- 11,43 7,23 10,11 10,5 10,13 7,27 12,25 8,3 15 19,1 29,12

Air parti-cles, µg/m3

- - - - - 90 113,5 72,6 98,28 124,6 79,3 122,3

Sediments, mg/(m2d)

1137 1140 1722 3039 312 655 559 1096 1161 534,6 811 1457

Carl Bro Group (2003) reports the following values for particulate matter in January and February 2001:

The air quality limits set in EU directive 1999/30/EC are:

Dust

• 24 hour limit value: 50 µg/m3 from 2005 • Annual limit value: 40 µg/m3 from 2005 and 20µg/m3 from 2010

SO2

• Hourly limit value: 350 µg/m3 from 2005 • 24 hour limit value: 125 µg/m3 from 2005 • Annual limit value: 20 µg/m3 from 2001

No monthly average limits have been defined in the directive.

2001 January February Prishtina 148 µg/m3 138 µg/m3 Kodra e diellit 299 µg/m3 181 µg/m3 Sofalia 52 µg/m3 59 µg/m3 Tasligje 131 µg/m3 114 µg/m3 K e trimave 417 µg/m3 110 µg/m3 Qyteza Pejton 43 µg/m3 87 µg/m3 Obiliq (village next to Kosovo B)

148 µg/m3 376 µg/m3


The values measured by INKOS far exceed the values in EU directive 1999/30/EC but are not only related to pollution from the power plants, mines and ash disposals – proper street cleaning could reduce these values signifi-cantly.

It is safe to assume that no major improvements have happened in this regard since 2001.

4.3.3 Surface waters The river Sitnica runs close by Kosovo B and its ash disposal site. This river is considered to be the most polluted river in Kosovo, and the KEK power plants are major polluters of the river.

4.4 Key environmental, health and safety aspects of the facility Kosovo B is a significant source of pollution. The pollution abatement equipment in Kosovo B is mostly old, from the 1980’s. Requirements con-cerning environmental performance are presently much tighter than in the 1980's. The most significant environmental aspects of Kosovo B are emis-sions into the air and wastewater discharge into the river Sitnica. Additionally the groundwater and soil of the area are most likely heavily polluted due to past and present activities of both Kosovo A and Kosovo B.

Dust is a significant problem in the working environment of the power plant.

The risk of health impacts on the population, as well as on the employees, does exist. The concentration of industrial facilities along the Sitnica River has increased the number of inhabitants living next to the river. The density is some 300 inhabitants per km2.

Large ash disposal sites and excessive dust emissions form, together with wastewaters, the highest potential risk for the health of people. It is however very difficult to establish a direct correlation between health problems and emitted pollutants. To make such conclusions it would be necessary to carry out extensive environmental, socio-economic and medical researches.

The workplaces most affected by ash dust in Kosovo B are in the vicinity of the boilers and the lignite transporters.

Ash (dust) is very harmful for the human body as it causes lung fibrosis and sight deterioration.


5 KOSOVO B ENVIRONMENTAL, HEALTH AND SAFETY MAN-AGEMENT

5.1 EHS policies and practice According to Carl Bro Group (2003) a general concern about the environ-mental impact of mining and power generation is strongly expressed by the Management of KEK and it is fully understood that the condition for future operation is that the environmental performance is improved and that the plants must fulfil international requirements as stated in the new Law on En-vironmental Protection in Kosovo.

However, KEK is still to elaborate clear environmental policies, objectives and targets and implement these.

5.2 Organization of EHS management The generation department of KEK employs about 1640 persons, 110 in engi-neering, 440 in maintenance, 650 in Kosovo A and 450 in Kosovo B. KEK has been under reconstruction and a new organisational framework is under development.

The Strategic Development Office (SDO) of KEK has an Environmental Unit with three employees. In the Production Division environmental protection is a part of Analysis services in the engineering department. This Analysis ser-vices team stationed at Kosovo B is comprised of 4 - 5 persons, most of them dealing with other matters as well. This team has only existed a few years. The Analysis services team appears to have inadequate resources and influ-ence. The head of the team is on the third organizational level below the plant manager. However, according to information received during discussions with the personnel of the Analysis services team of Kosovo B, the Head of the Analysis Services will in near future begin to report directly to the Gen-eral Manager of the power plant.

Additionally there is another environmental organisation within KEK, namely the INKOS institute, which performs monitoring at the site. INKOS is said to have insufficient means and equipment for this task and the monitoring re-sults are mostly considered unreliable due to lack of calibration of equipment and experience.

An environmental co-ordination group has been established involving all per-sons working with environmental issues in KEK.

Health and safety issues are managed by the Executive Office for Industrial Safety and Fire Protection. This office has 15 - 17 employees, 5 of which are directly involved in power generation.

The Ministry of Environment and Spatial Planning (MESP) has environ-mental inspectors and it has regularly inspected Kosovo B in the past. A co-operation group with members from KEK and MESP was established in


2004, but the group has not functioned in the last year or so despite MESP ef-forts. MESP has not performed any inspections during this period, and ac-cording to MESP personnel there is currently no real co-operation or sharing of information between KEK and MESP.

The Analysis services team at Kosovo B is fully aware of the fact that the cur-rent situation must be improved and consider the environmental problems of high importance. So far few concrete environmental mitigation measures have been implemented due to lack of funding.

KEK is not a holder of any ISO or other similar quality or environmental cer-tification. A future development might be to establish an ISO 14001-compatible environmental management system to secure the systematic han-dling of environmental issues in all operations as well as continuous im-provement of environmental performance.

Based on the material obtained and interviews carried out, the responsible persons at the Analysis services team are well aware of and know the envi-ronmental legislation and regulations concerning the operations of the power plant. They are aware of the environmental aspects of their operations and also of the most important development needs. However, they lack the neces-sary authority and means to implement these measures. As KEK relations to the supervising authority MESP are not very good, MESP does at present not have practically any influence on the operations of Kosovo B.

5.3 Contingency planning and emergency procedures According to personnel of the Executive Office for Industrial Safety and Fire Protection a loan has been approved for development of safety procedures and training at KEK, and this has led to the development of a policy for reducing risks at the power plants. It is unclear what the financing institution is.

The plants have their own fire brigade and the only emergency plans found are related to fires. There are hand held extinguishers in the premises of the power plant, and water posts at several points outside the plant.

5.4 Staff training and supervision A training course in environmental issues for the Analysis services unit per-sonnel at Kosovo B has been held recently by a German consultant.

Raised environmental and health and safety awareness of all employees should be promoted through regular training and supervision. At this point for instance noise protection gear in areas with high noise levels is not worn by all personnel.


6 KOSOVO B ENVIRONMENTAL PERFORMANCE

6.1 National regulatory requirements In the former Yugoslavia the environmental policy was the responsibility of each republic, but no specific policy was implemented in Kosovo. After the conflict the main focus has been on generation of power and the development of environmental policies has started only a couple of years ago with the pass-ing of the Law on Environmental Protection (No. 2002/8) in April 2003. The Law on Environmental Protection was elaborated with assistance from UN-MIK and UNMIK has also established a Ministry of Environment and Spatial planning in the interim administration. Since then a few other environmental laws have been passed or are currently under preparation, such as:

• Law on Air Protection (No. 2004/30) • Law on Water (approved by Parliament, not yet issued) • Law on Waste (approved by Parliament, not yet issued)

The operation of Kosovo B has so far not required any environmental per-mits.

6.2 Applicable EU/other requirements and standards The Law on Environmental Protection and subsequent environmental laws implement EU-directives for the territory of Kosovo. Relevant directives and guidelines are:

• Directive no 99/30/EC relating to limit values for sulphur dioxide, nitrogen oxides and oxides of nitrogen, particular matter and lead in ambient air

- The first Daughter Directive (1999/30/EC), as amended by Commission Decision 2001/744/EC • Directive 2001/80/EC of the European Parliament and of the Council

on the limitation of emissions of certain pollutants into the air from large combustion plants

• Council Directive 98/83/EC on quality of water intended for human consumption

• Council Directive 75/442/EEC on waste - as amended by Directive 91/156/EEC - as amended by Directive 91/692/EEC - as amended by Commission Decision 96/350/EC • Directive 91/689/EEC on hazadous waste, as amended • Decision 2000/532/EC establishing a list of waste, as amended • Directive 75/439/EEC on the disposal of waste oils, as amended • Directive 91/157/EEC on batteries and accumulators containing

dangerous substances, as amended • Directive 96/59/EC on the disposal of PCBs and PCTs • Council Directive 87/217/EEC on the prevention and reduction of

environmental pollution by asbestos


• Council Directive 1999/31/EC on the landfill of waste. • BREF (BAT Reference Document) “Large Combustion Plants”

(LCP)/05.2005 • BREF “Cooling Systems/12.2001 • BREF “Emissions from storage of bulk or dangerous

materials”/01.2005 • BREF “Waste treatments”/08.2005 • BREF “Monitoring systems”/07.2003

6.3 Inputs, products and releases

6.3.1 Fuels The main fuel of Kosovo B is lignite. Kosovo has one of the largest reserves of lignite in Europe and the quality is relatively good. However, due to lack of investment and bad maintenance the currently mined lignite is not of the ex-pected quality. This is of environmental importance, as the power units then must operate on lignite with higher ash content and lower heating value than designed giving incomplete combustion, low efficiencies and high emissions.

The huge lignite resource found in Kosovo can be characterized with the fol-lowing analysis of monthly composite samples, as analyzed by INKOS (as % AR):

Year H2O Ash Total carbon

Fixed carbon Volatile Com-

bustibleLHV

kJ/kg Total S Inorg S Org S

2002 avg 43,8 16,6 30,6 14,0 25,7 39,6 8 044 1,08 0,69 0,39 2003 avg 42,9 19,2 31,9 12,6 25,2 37,8 7 478 0,89 0,62 0,28 2004 avg 43,6 18,1 31,4 13,4 25,0 38,4 7 649 0,91 0,71 0,20 2002-2004 Average 43,3 18,3 31,4 13,2 25,3 38,4 7 660 0,94 0,66 0,27

Monthly Min 39,6 14,7 28,5 10,0 21,8 34,5 6 446 0,70 0,41 0,12

Monthly Max 46,5 24,4 35,0 16,2 28,0 41,1 8 509 1,43 0,90 0,58

This Kosovo lignite can be characterized as having a relatively low ash con-tent, low combustible sulphur, most of the sulphur as sulphate/sulphite form and as having ample amounts of calcium. The lignite also contains a substan-tial amount of chlorine. The ash softening and melting temperatures are low and cause problems in conventional combustion process if not properly con-sidered at the design.

The total consumption of lignite coal in Kosovo B according to official data has been (Carl Bro Group 2003 and KEK):

Year Consumption, t/a 2000 1 782 948 2001 2 207 504 2002 3 039 845 2003 2 500 000


2004 3 500 000 2005 3 500 000

The quantities of lignite used for power generation were calculated by Carl Bro Group in 2003 based on data found in different reports and as informed by KEK, and also considering the operation conditions with many starts and stops of the units. According to INKOS the annual consumption of Kosovo A and B together was about 5,2 million tons with an average heat content of 7,7 GJ/t in 2002. Carl Bro Group concluded, that this cannot be considered reli-able as only the best quality of lignite coal should then have been mined in year 2002 and the units should have operated close to their design values. Thus the yearly quantities of lignite used in 2000 - 2002 above are probably underestimated by some 20 %.

Heavy fuel oil is used for start-up of boilers and for support firing of the boil-ers if the quality of lignite is low. Diesel and gasoline is used for trucks, bull-dozers, excavators and cars etc.

The total consumption of fuel oil at Kosovo A and B together in 2002 was:

Heavy fuel

oil Light fuel oil/diesel Gasoline

t Power plants 6 294 4 731

No information on the fuel oil consumption in Kosovo B alone was available.

The characteristics of the liquid fuels are not known and no information is available about the consumption of diesel and gasoline. The HFO is typical Eastern European "Mazut" which typically has relatively high sulphur content (well over 1 % permitted in EU).

6.3.2 Water source and consumption Raw water is supplied by Iber Lepenc from Quazivodo lake some 40 km from the plant to the north, by channels and pipelines. There is no reservoir for raw water at the plant, only for treated water for 2 - 3 hours’ operation of the plant.

Raw water is used for cooling tower make-up, boiler water preparation, wash-ing water and household potable water.

The water for boilers at the Kosovo B power plant is subjected to the usual treatment procedures (filtration, flocculation, and demineralization). The dif-ferent phases of the treatment are currently operated manually, although the system is originally designed for automatic operation.

The reported total consumption of raw water in 2002 at Kosovo B, about 7 000 000 m3, is high. For modern plants with evaporative cooling towers wa-ter consumption is typically below 2,0 m3/MWh and for older rehabilitated plants around 2,5 m3/MWh. The corresponding value was 3,6 m3/MWh for Kosovo B in 2002.


6.3.3 Other materials

Mineral oils Oils are mainly use for lubrication of machinery and for cooling. Kosovo B used an estimated amount of about 7 m3 mineral oils in the year 2002.

Chemicals The power plant uses relatively small amounts of chemicals considered haz-ardous.

The chemicals are used mainly for water preparation. The following con-sumptions in the year 2002 were reported:

2002 hydrochloric acid (HCL) 30% 268,1 t sodium hydroxide (NaOH) 40% 211,6 t calcium hydroxide (CA(OH)2) 90% 121,5 t ferric sulphate (Fe2(SO4)3) 50% 43,6 t ammonium hydroxide (NH4OH) 25% 15,4 t hydrazine (N2H4) 15% 12,0 t ammonia (N2H3) Aktifos 645 44,4 t Total 716,6 t

Besides that, hydrazine and ammonium hydroxide are used in the water-steam system in order to remove oxygen and regulate pH.

This consumption is somewhat high. According to Carl Bro Group (2003) a thumb rule from EU-countries is a consumption of chemicals on the level of 0,1 – 0,2 kg per MWh. For Kosovo B the consumption is 0,37 kg.

Monthly records of chemical usage are kept.

6.3.4 Air emission amounts and quality The main pollutants emitted from fossil fuel combustion are:

• Solid particles, "dust"; in other words ash in the flue gases from the boilers)

• SO2 and NOx (sulphur and nitrogen oxides) • Products from incomplete combustion of the fuel (soot, carbon-monoxide

and hydrocarbons) • CO2

(carbon dioxide), a green house gas

The boilers at Kosovo B are equipped with electrostatic precipitators, but not with FGD (flue gas desulphurisation) or de-NOx installations. Flue gases from the boilers are discharged to the atmosphere through a common 182 meters high stack.

Dust

Power plant

The reason for dust emission from the power plant is the content of ash in the lignite and the ash released in the pulverized combustion process.


Fly ash can be removed from the flue gas by different means of filtering, but electrical precipitators (ESP's) are the most commonly used means of pollu-tion control and they are also used at Kosovo B power plant.

The original Kosovo B ESP's were designed for a maximum emission of par-ticulates of 260 mg/Nm3. Earlier measurement results are described in appen-dix 4.

During the site visit the continuous measurements showed values of about 510 - 520 mg/m3 in the control room, but the newly installed measuring equipment at Kosovo B unit B2 was said not to be calibrated and to be wrongly installed (vibrations).

The results of the dust emission measurements of unit B1 in October 2005 were 526 - 577 mg/Nm3 (dry gasses, standard conditions, 6 % O2). The dust emissions from unit B2 could not be measured as the unit was being repaired during the measuring period. As the ESP of unit B2 has been reconstructed recently it is likely that the dust emissions from this unit are at least somewhat lower.

With the calculated consumption of lignite and air the emissions of dust from Kosovo B in the year 2002 Carl Bro Group (2003) calculated the dust emis-sions as follows:

Kosovo B Unit B1 Unit B2 Total

2 218 t 2 109 t 4 327 t 250 mg/m3 250 mg/m3 n/a

Much lower values were reported by KEK. As the dust emissions of unit B1 were measured to be more than double that of the estimated emissions in the table above it can be concluded, that the total dust emissions of Kosovo B are currently most likely significantly higher than the amounts calculated by Carl Bro Group in 2003.

The annual quantity particulate matter emissions from Kosovo B are signifi-cant and dust also forms a significant health problem for the workers at the plant.

The particle concentration in flue gases from unit B1 measured in October 2005 exceeds the EU limit value (50 mg/Nm3) more than tenfold.

Handling, transport and disposal of ashes

The dust problems from ashes start at the boilers where large quantities of ashes are found because the pulverizers, precipitators and pipe connections are leaking large quantities of ash inside and on the ground outside the boiler buildings. This is firstly a local health problem for the workers but when the ashes are mechanically removed and transported to the disposal areas it also becomes an external problem.

Collected fly ash is transported dry to temporary storage silos from where it is transported as slurry in a pipeline to the disposal site. The transport of wetted fly ash as slurry causes no dust problems, but large quantities are lost on the


way from the plant to the disposal areas and when the disposal area dries out it causes dusting in the surroundings.

No serious actions are seen taken to limit the dust problems caused by ash management.

It has been estimated that the handling and transport of ashes causes about 19 500 tonnes, and the disposal areas about 50 000 tonnes of dust emissions annually.

Sulphur dioxide, SO2 The reason for SO2 emissions is the sulphur content in the lignite. However, to an extent depending on the burning technology, a natural desulphurisation takes place as the lignite has a high content of Calcium. This matter is further discussed in Appendix 4.

The sulphur content of the lignite is not high (about 0,7 - 1 %) but taking into account the low heating value of lignite (about 7-8 GJ/t) and the fuel con-sumption the annual sulphur dioxide emissions are relatively big.

Earlier SO2 measurement results are described in appendix 4.

With the actual sulphur content, content of CaO and operation conditions the SO2 emissions from Kosovo B in the year 2002 were estimated at (Carl Bro 2003):

Kosovo B Unit 1 Unit 2 Total

35 204 t 7 533 t 68 683 t 894 mg/m3 894 mg/m3 n/a

The calculation was based on a sulphur content of 1% and a desulphurisation of 77,5% due to the presence of Calcium in the fuel.

Much lower values were reported by KEK.

The results of the SO2 emission measurements of unit B1 in October 2005 were surprisingly low, only 142 - 332 mg/Nm3 (dry gasses, standard condi-tions, 6 % O2). The SO2 emission value after ESP was rather unstable during the measurement campaign, which refers to some changes either in boiler sul-phur capture or in lignite quality. As the fluctuations were significant and ir-regular, additional continuous measurements over a longer period are required to establish the actual SO2 emissions, which may be higher. The SO2 emis-sions from unit B2 could not be measured as the unit failed during the meas-uring period. It can be assumed that the SO2 emissions of unit B2 are in line with those of unit B1 due to the same combustion technology and fuel.

As the exact SO2 emissions from units 1 and 2 are not known it is impossible at this stage to determine if LCP-compliance requires additional desulphurisa-tion.

Nitrogen oxides, NOx The reason for NOx emission is the presence of nitrogen in the lignite and air availability in the combustion process. The quantities developed are related to the combustion process and they increase with increased efficiency.


Earlier NOx measurement results are described in appendix 4.

With the current content of N in the lignite coal and with the current combus-tion processes the emissions of NOx in the year 2002 were calculated to be as shown in the table below (Carl Bro Group 2003):

Kosovo B Unit 1 Unit 2 Total

6 513 t 6 194 t 12 706 t 735 mg/m3 735 mg/m3 n/a

KEK reported values of the same magnitude as calculated.

The results of the NOx emission measurements of unit B1 in October 2005 were 661 - 713 mg/Nm3 (dry gasses, standard conditions, 6 % O2). The NOx emissions from unit B2 could not be measured as the unit failed during the measuring period. It can be assumed that the NOx emissions of unit B2 are similar as the same combustion technology and fuel are used.

Based on the measurement results from October 2005 it can be assumed that some measures will be necessary to comply with the EU requirement of maximum 500 mg/Nm3. The nitrogen oxide concentration in flue gases measured in 2005 exceeds the EU limit value by at least 30 %. After the year 2015 the EU limit value for NOx emissions will be 200 mg/Nm3. Burner ad-justments will be sufficient until the year 2015 according to Alstom, but the stricter limit from 1.1.2016 onwards will require air staging and installation of low-NOx -burners.

CO The problem of CO emissions due to incomplete combustion is only marginal at Kosovo B, compared to Kosovo A. Measurements of emissions from unit B1 in October 2005 showed, that the CO emissions are 73 - 113 mg/Nm3 (dry gasses, standard conditions, 6 % O2). This can be considered to be on the high side but is still acceptable.

GHG Emission of carbon dioxide is not directly harmful to people and ecosystems, but has an accelerating impact on global warming. The emission of CO2 is proportional to the content of carbon in the fuel and the quantity of fuel com-busted.

The combustion at the Kosovo A and B plants in the year 2002 developed about 8,9 million t CO2 (Carl Bro Group 2003).

In 2005 KEK reported the CO2 concentration in flue gasses as 335 840 mg/Nm3 for Kosovo B.

There is no commercial method available to remove the CO2 from the flue gas and the best options to reduce the CO2 emission are to improve the effi-ciency and burn less fuel or change to fuel with a lower content of carbon (for example from coal to natural gas or biomass).


Improved efficiency is obviously a possibility at Kosovo B power plant while change of fuels on the plants would appear to be out of scope.

Other emissions In addition to the above mentioned emissions there are heavy metal emissions from the power plant. Data on heavy metals concentrations in flue gasses does not exist as measurements have never been carried out.

Monitoring The plant personnel stated that the last reliable emission measurements before 2005 were made in 1994. At present, the plant does not have reliable enough means to measure the emissions. A continuous measurement system of dust emissions based on optical measuring has been installed in conjunction with the rehabilitation of the ESP of unit B2, and the results are shown continu-ously in the control room of the plant. The measurements of this system are, however, not entirely reliable due to problems with the functionality of the equipment.

INKOS monitors the concentration of SO2, dust and small particles in the ambient air at the KEK site and in the village nearby, although the results are said not to be very reliable because of lack of proper equipment and experi-ence. The published 2001 results from the KEK site shows concentrations of SO2, fly ash (smut) and air particles exceeding the WHO norms by 100 %, whereas the suspended particle concentrations exceed the WHO norms by more than 30 times. It can be determined, that the EU norms were exceeded as well, although no monthly limit values have been set. The situation has not improved significantly since 2001.

The impact of the present emissions on the air quality of the surroundings of the Kosovo B power plant can be estimated to be significant, although Kos-ovo A is a much greater polluter.

Because reliable air quality measuring results are not available at present, dis-persion modelling of the flue gases from Kosovo B was performed in Decem-ber 2005 by the Finnish Institute of Meteorology based on the emission measurement results from October 2005. The results of this dispersion model-ling are presented in chapter 8.4.

EU legislation compliance of Kosovo B

Emissions into the air

The particulate matter (dust), sulphur dioxide and nitrogen oxides emissions are significant and one of the major environmental concerns at the site.

The emission limits set for existing boilers over 50 MWth in the EU LCP-directive are shown in table 6 – 1. Emission limits for CO are not defined in the LCP-directive.


TABLE 6 – 1 The emission limits for solid fuel fired plants set in the EU LCP-directive (for boilers over 50 MWth capacity).

1) Plants licensed before 1.7.1987 must either comply to these limit values by 1.1.2008, or they can be subject to a national emission reduction plan for existing plants. 2) Without prejudice to Directives 96/61/EC and 96/62/EC, existing plants may be exempted from compliance with the emission limit values referred to in Article 4(3) and from their inclusion in the national emission reduction plan on the following conditions: (a) the operator of an existing plant undertakes, in a written declaration [submitted by 30 June 2004 at the latest] to the competent authority, not to operate the plant for more than 20,000 operational hours starting from 1 January 2008 and ending no later than 31 December 2015; (b) the operator is required to submit each year to the competent authority a record of the used and unused time allowed for the plants’ remaining operational life.

3) Where the emission limit values above cannot be met due to the characteristics of the fuel, a rate of desulphurisation of at least 60 % shall be achieved in the case of plants with a rated thermal input of less than or equal to 100 MWth, 75 % for plants greater than 100 MWth and less than or equal to 300 MWth and 90 % for plants greater than 300 MWth. For plants greater than 500 MWth, a desulphurisation rate of at least 94 % shall apply or of at least 92 % where a contract for the fitting of flue gas desulphurisa-tion or lime injection equipment has been entered into, and work on its installation has commenced, be-fore 1 January 2001. 4) Plants, of a rated thermal input equal to or greater than 400 MW, which do not operate more than the following numbers of hours a year (rolling average over a period of five years), - until 31 December 2015, 2000 hours; - from 1 January 2016, 1500 hours; shall be subject to a limit value for sulphur dioxide emissions of 800 mg/Nm3.

5) Until 31 December 2015 plants of a rated thermal input greater than 500 MW, which from 2008 on-wards do not operate more than 2000 hours a year (rolling average over a period of five years), shall: - In the case of plant licensed in accordance with Article 4(3)(a), be subject to a limit value for nitrogen oxide emissions (measured as NO2) of 600 mg/Nm3; - In the case of plant subject to a national plan under Article 4(6), have their contribution to the national plan assessed on the basis of a limit value of 600 mg/Nm3.

EU Council Directive 2001/80/EC on the limitation of emissions of certain pollutants into the air from large combustion plants (LCP-directive) Limits for "pre-1987" plants to be applied from 1.1.2008 onwards:1)2)

Pollutant Monthly average: SO2 for solid fuels (02 content 6 %):3)4) >500 MWth: 400 mg/Nm3 NOx for solid fuels (02 content 6 %):5)6)

>500 MWth: 500 mg/Nm3 From 1 Jan 2016:

>500 MWth: 200 mg/Nm3

Solid sub-stances

for solid fuels (02 content 6 %):

≥500 MWth: 50 mg/Nm3 7)


From 1 January 2016 such plants, which do not operate more than 1500 hours a year (rolling average over a period of five years), shall be subject to a limit value for nitrogen oxide emissions (measured as NO2) of 450 mg/Nm3.

6) Until 1 January 2018 in the case of plants that in the 12 month period ending on 1 January 2001 oper-ated on, and continue to operate on, solid fuels whose volatile content is less than 10 %, 1200 mg/Nm3 shall apply.

7) A limit value of 100 mg/Nm3 may be applied to plants licensed pursuant to Article 4(3) with a rated thermal input greater than or equal to 500 MWth burning solid fuel with a heat content of less than 5800 kJ/kg (net calorific value), a moisture content greater than 45 % by weight, a combined moisture and ash content greater than 60 % by weight and a calcium oxide content greater than 10 %.

----------

Plants, which have been granted the original construction or operation license before 1.7.1987, like Kosovo B, have either to comply with the limits by 1.1.2008 or:

a) they can be subject to a so called "national emission reduction plan for existing plants", which every Member State can compile, or

b) the operator of such a plant can submit a written declaration to the competent authority, confirming that the plant will not be operated for more than 20 000 operational hours starting from 1 January 2008 and ending no later than 31 December 2015.

The LCP-directive also sets requirements for the monitoring equipment and practices of emissions from large combustion plants with a thermal input more than 100 MW.

As a conclusion, until 1.1.2008 Kosovo B is compliant with EU air emission regulations as implemented in Kosovo legislation.

After 1.1.2008 Kosovo B in its current configuration will not be compliant with the LCP-directive. Also according to the Law on Environmental Protec-tion emission limits for the discharge and emission of pollutants into the air shall be developed in subsidiary acts (not yet issued). The Government shall have the authority to adjust downward such permissible maximum levels over time to bring them gradually into compliance with the prescribed levels of the EU (in the LCP-directive) in a manner that is both realistically affordable by the authorities, persons and undertakings and consistent with the sustainable economic development of Kosovo.

6.3.5 Waste water amounts and quality Waste waters from Kosovo B include the following:

• Cooling water (evaporation & concentrated output from the system) • Process waste waters • Sanitary waste waters • Waste waters from washing (oily waste waters etc.) • Rainwater • Waters from ash disposal systems and area


Wastewater from Kosovo B and the ash disposal area flow to nearby rivers and infiltrate to the groundwater. Water is not recycled and consequently large quantities are discharged to the rivers or infiltrate the groundwater as overflow or drainage waters. No records of water consumption are available but an estimation was made in 2003 (Carl Bro Group) that the quantities of wastewater discharged from Kosovo A and B combined were as follows:

To air (evapo-rated)

To wastewater treatment To rivers To ground-

water

000 m3

Power and drying plants 7 830 9 598 3 200 1 Water preparation plants - - 0 0 Water treatment plants 0 (9 598) 9 598 0 Ash and slag handling 0 - 30 3 Ash and slag disposals 793 - 988 243 Total 8 623 - 13 816 247

It should be noted, that Carl Bro Group apparently used an ash to water ratio of 1:10 for the ash transportation water for both Kosovo A and B when calcu-lating the amounts, and therefore they may be overestimated for Kosovo B. The ratio is said to be 1:10 only at Kosovo A. No information regarding how large the Kosovo B share of these amounts was could be found in the records, but the amounts of waste waters discharged from Kosovo B to the river were said to be 600 m3/h when both units are in operation. This amounts to some 4 750 000 m3/year.

The main pollutants in the wastewaters from Kosovo B are:

• Ash • Oil and chemical residuals • Fecal matter and microbes

These pollutants are harmful for humans and the ecosystem mainly for the following reasons:

• Combustion residuals (ashes) • Suspended solids lead to increased turbidity, sedimentation and increased

oxygen consumption which has many harmful impacts on aquatic ecosystems

• Excess nutrients (P and N) leads to eutrophication • Oils have toxic impacts • If the pH of the effluents differs significantly from the pH of the recipient

the effluents will have local harmful effects on the ecosystem.


Additionally waste waters from desalination and preparation of process water are either acid or alkali.

6.3.6 Waste amounts and characteristics Formation of waste in mining and power generation by combustion of lignite coal cannot be avoided. However it can be reduced and taken care of in an environmentally appropriate way.

The main sources of solid waste at Kosovo B are:

• Combustion residuals (ashes) • Worn out equipment and materials • Residuals from water treatment (sludges)

Typical wastes from combustion equipment are the following:

• Sludge from the bottom of tanks for petroleum products • Hydrochloric acid • Other organic solvents, for flushing • Liquids and mother liquors • Waste paints and varnishes containing organic solvents or other

substances • Waste glues, sealing materials containing organic solvents • Non-chlorinated mineral engine, gearbox and lubrication oils • Waste isolation or heat/transferring oils containing PCB • Sludge from oil separators • Sludge from oil traps • Sludge from impurity traps • Package containing remains of impurities • Absorption agents, filtration materials, including oil filters, contaminated

cleaning textiles • Solid fractions from sand traps and oil separators • Lead accumulator batteries • Wastes containing petroleum products • Brickwork and refractory materials containing hazardous substances • Soil and stones containing hazardous substances • Other insulation materials that are dangerous or that contain hazardous

substances like asbestos • Construction and demolition wastes • Fluorescent lamps and other waste containing mercury • Decommissioned equipment containing hydrogen fluorides and chlorides • Dye, printing ink, glues and resins containing hazardous substances

Some wastes are harmless for human beings and eco-systems and might even be valuable as recycling products (different kind of metals and ashes) while others are harmful and must be handled with care.

Some waste types are further discussed below.

Ashes


Ash is produced in the boilers as a residual from the combustion process as bottom ash (slag) and fly ash.

The quantities produced depend mainly on the content of non-combustible materials (typically overburden) in the lignite. The current level of none-combustible materials in the lignite coal is about 30 %. Coal ash is classified as non-hazardous in the European Waste List (2000/532/EC) under code 10 01 02. However, if unmanaged as in the case of Kosovo B it can give rise to environmental problems.

According to official data Kosovo B generated about 850 000 tonnes of fly and bottom ash in 2002. Carl Bro Group (2003) estimated the amount to be much higher in reality. An estimate of 2 100 000 tonnes of ash generated from Kosovo A and B together in 2001 was given. As the precipitators at Kosovo B are far more efficient than the ones at Kosovo A it can be assumed that the main portion of this total ash amount was generated at Kosovo B, al-though no separate estimates for Kosovo A and B were given.

During the emission measurements performed by Electrowatt-Ekono Oy as part of this study, samples of lignite, fly ash and bottom ash were collected by KEK personnel and analyzed in Finland by a certified laboratory. The charac-teristics of the fly and bottom ashes, according to these analyses, are as fol-lows:

Flyash Bottomash Amount t/h 63,6 2,3 % 96,5 3,5 H2O content % 0,3 49,7 Cl % DB 0,02 0,07 UBC % DB 0,9 10,7 UBC by INKOS % DB 3,1 14,9 SiO2 % DB 36 51 Al2O3 % DB 12 13 Fe2O3 % DB 7,6 6,4 CaO % DB 27 8,0 MgO % DB 3,5 2,2 SO3 % DB 7,0 K2O % DB 0,97 15 MnO2 % DB 0,24 0,21 Na2O % DB 1,0 0,74 P2O5 % DB 0,17 0,12 TiO2 0,42 0,55 As mg/kg DB 31 10 Ba mg/kg DB 610 420 Be mg/kg DB < 2,5 < 2,6 B mg/kg DB 580 230 Cd mg/kg DB < 0,5 < 0,51 Co mg/kg DB 17 18 Cr mg/kg DB 160 180 Cu mg/kg DB 47 43 Hg mg/kg DB 0,11 < 0,046 Mo mg/kg DB 3,5 2,6


Ni mg/kg DB 190 150 Pb mg/kg DB 18 16 Sb mg/kg DB 2,6 < 2,6 Sn mg/kg DB < 2,5 < 2,6 V mg/kg DB 95 90 Zn mg/kg DB 68 84 S mg/kg DB 28100 5000 % DB 2,8 0,5 SO3 % DB 7,0 1,3 As total S in fuel % DB 0,83 -- % AR 0,45

In addition one fly ash sample was analyzed for active Ca(OH)2. This is deter-mined with acid titration method and it represents the amount of free reactive CaO present in flyash. This portion of total Ca has been calcined in the boiler from CaCO3 to CaO but not reacted with SO2 and is therefore available for sulphur removal. The results are shown below:

Flyash Bottomash H2O % 0,3 45 UBC % BD 2,0 17,5 Total S % BD 3,1 0,5 Cl % BD 0,07 Ca(OH)2 % BD 10,2 0,5

Based on above results and assuming that all sulphur in fly ash is present as CaSO4 the Ca/S mass balance can be calculated as shown below:

In lignite In flyash

% g/kgf % g/kgf % from in lignite

Ash 16,7 167,0 96,5 161,2 Total S 0,86 8,6 2,8 4,5 52 Inorg S 0,38 3,8 3,8 100 Org S 0,48 4,8 0,7 15 CaO 5,2 51,8 27,0 43,5 84 MgO 0,53 5,3 3,5 5,6 106 Ca(OH)2 10,2 Ca/S 5,3 % of Ca Total Ca 3,7 37,0 19,3 31,1 100 Ca as CaCO3 31,6 10,9 35 Ca as dolomite 5,3 5,6 18 Ca as active CaO 5,5 8,9 29 Bound as CaSO4 5,6 18

Asbestos According to information available asbestos has not been used in large amounts when Kosovo B was constructed. Asbestos was reportedly used in


the cooling towers, but has since been replaced. The current presence of some asbestos can not be excluded according to plant personnel.

PCB Taking into account the age of the equipment and the lack of regular mainte-nance, it is almost certain that certain pieces of equipment, for instance trans-formers, contain oils with polycarbonated biphenyls (PCB). No study of PCB presence in Kosovo B equipment has been carried out.

6.3.7 Noise and vibrations Noise can be a serious problem for the workers health and safety. In some places of the plant the noise level most likely exceeds 85 dB(A).

Noise insulation is applied around the noisiest equipment. The workers are in-structed to use protection when working in the noisiest areas of the plant.

Factors such as wind direction and speed, atmospheric absorption, and air humidity affect the distribution of noise. The residents of the villages in the neighbourhood of Kosovo B have reportedly not made complaints regarding noise. Thus no measuring and noise mapping have been considered necessary. During night time the noise in the nearest village Obiliq some 500 meters from Kosovo B is probably above the level for residential areas according to Carl Bro Group (2003). With lack of noise mapping the noise levels were es-timated as follows in 2003:

Location Distance Local noise Estimated

noise level Obiliq 500m 55-60 63

Typical limit values for noise immission at the nearest settlements that are applied in EU countries are 55 dB(A) in daytime and 40 - 50 dB(A) in night time.

Mitigation measures for noise should be provided for health and safety pro-tection of the workers and population in the vicinity. In order to direct the measures required noise measurements should be carried out at the work places of Kosovo B and at the nearest settlements. Based on these results the required measures (noise insulation, protective gear, noise damping, new equipment etc.) can be planned. After the measures have been carried out noise measurements should again be carried out in order to assess the effects of the measures. These measures at Kosovo B are however not considered es-pecially urgent by the consultant, considering the other significant impacts of the plant and mitigation measures required.

6.4 Process efficiency The efficiency of a power plant is normally expressed as a percentage (pro-duced energy / consumed energy x 100) or as specific consumption of fuel


(GJ/MWh) where a consumption of 3,6 GJ/MWh corresponds to an efficiency of 100%.

The technology used in coal fired power plants and physical limitations limit the maximum obtainable efficiency to about 45 % or about 8 GJ/MWh, but obviously old units operate far below this efficiency. However, rehabilitated and retrofitted units of the same type in other countries have demonstrated ef-ficiencies in the level of 35 to 38 % or a consumption of about 11 – 12 GJ/MWh.

The concept of improved efficiency to reduce the environmental impact is further developed in CHP production where efficiencies in the order of 90 % can be obtained. The EU also promotes this concept and has issued a directive on the promotion of cogeneration (Directive 2004/8/EC).

With the fuel consumption reported by KEK of 1,6 t/MWh for Kosovo B and an assumed heating value of 8,0 GJ/t the current efficiency is about 30 % at Kosovo B according to Carl Bro Group (2003). The total plant gross effi-ciency MWe/MWth may in reality currently be below 30 %.

The lower efficiency is due to losses in turbines, feedwater quality, by-passing of HP preheaters, pulverizing problems, slagging and high exhaust temperature, continuous sootblowing due to slagging and high O2 concentra-tion due to leakages in the ESP's. Official data for the plant gives much higher efficiencies.

The in-house consumption at the power plant, 361 GWh/a, seems high and probably also includes some consumption for coal handling outside the power plant.

6.5 General housekeeping issues Special teams are reported to have been allocated to clean-up and store equipment and materials where they belong.

After the conflict the Kosovo B plant was found littered with garbage and worn-out equipment spread all over the site. A significant cleaning-up process has been implemented but is still not completed.

A lot of new tools and equipment has been provided to the plant after the con-flict and it seems that these are well maintained and regularly cleaned. Dis-carded equipment is stored randomly at various locations on the grounds of the power plant.

General housekeeping is said to have improved greatly in recent years, but there is still room for improvement. For example, there is lot of dust in the boiler room, causing health and safety risks. This matter can, however, proba-bly not be solved until the ESP leakages have been fixed.


6.6 Hazardous materials management

6.6.1 Oils

Heavy fuel oil Heavy fuel oil is stored in one aboveground storage tank (5000 m3) which is equipped with a concrete secondary retaining containment of sufficient vol-ume. The bottom of the secondary containment is made of concrete plates which are not hermetically sealed.

No drainage pit or valve for the release of condensates potentially containing heavy fuel oil exists. The water filtrates into the ground through the bottom seams of the containment thus providing a risk for soil contamination in case of leakages.

The HFO pumping building is equipped with a containment basin. Where possible leaks are led from this basin is not known. The equipment in the pumping house is old and in general badly maintained.

There is an old smaller HFO storage tank which is leaking near the disused waste water treatment plant. This HFO is used for production of steam in the heating station when the Kosovo B units are started after overhauls.

Heavy fuel oil is transported by railway or by tank-trucks to the pump station where it is loaded into the storage tank. Spillages are collected by adsorbers. These are discharged of together with other wastes or stored in barrels.

Lubricating oil New mineral lubricating oil is stored in metallic and plastic barrels in a locked warehouse together with various chemicals and equipment. The ware-house is equipped with a concrete floor, but this floor is cracked. The roof lets rainwater through in several places and there is water inside the warehouse. No drains or thresholds could be detected, so the oils could easily get to the surrounding soil and groundwater in case of a leakage. There are hundreds of litres of oils in the warehouse.

No leakages are said to have happened, and no signs of such leakages could be visually detected.

6.6.2 Chemicals Chemicals used are stored and managed in a separate locked warehouse some 100 meters from the plant together with mineral oils and various equipments. Some chemicals are stored in or by the water treatment plant. HCl for regen-eration of ion exchange materials is stored in outdoor storage tanks (4 tanks, each 50 m3) with sufficient containment pits. Smaller quantities of chemicals used are stored in tanks equipped with secondary containment in the plant it-self.

The warehouse is equipped with a concrete floor, but this floor is cracked. The roof lets rainwater through and there is water inside the warehouse at


several places. No drains or thresholds could be detected, so the chemicals could easily get to the surrounding soil and groundwater in case of a leakage.

No leakages are said to have happened, and no signs of such leakages could be visually detected.

Hydrogen for the cooling of the electrical generator is stored in a tank in an open-air place.

6.6.3 Ozone depleting substances (ODS) and radioactive materials There are no ozone depleting substances at the site. No radioactive materials are used at the location. Smoke detectors, if installed, may contain little amounts of radioactive substances, but no smoke detectors were observed.

6.6.4 Welding gases Gases used for welding, argon and acetylene, are stored inside the power plant in locked places.

6.7 Waste water management

6.7.1 Waste water from power plant Kosovo B has waste water treatment installations for technical waters and sanitary waste waters, but they are out of date and not operating properly. Furthermore, the water consumption at Kosovo B, and consequently the wastewater amount, is currently much higher than was designed, which re-duces the efficiency of the waste water treatment.

Large quantities (said to be 550 m3/h) are evaporated in the cooling process. This emission is normally considered harmless. However, it may cause risks of spreading of legionellosis and of icing of roads and other structures in the wintertime. No indication of chemical treatment of cooling waters to prevent legionellosis or testing of the presence of legionella bacteria at Kosovo B was found.

Acid and alkaline wastewaters are generated during the ion-exchangers re-generation. They are collected into a small settling pond. A fraction of this water is used for the transport of ash to the disposal site. Some of the waste water in this pond is discharged into the river Sitnica. These wastewaters are either acid or alkali and with high concentration of salt. The amount of waste waters arising from process water preparation is currently large, and the proc-ess water treatment could be much more efficient. The problem here is said to be the removal of lime sludge from water treatment tanks, which currently can not be done efficiently enough.

Traces of heavy fuel oil from the separation of excess water from the HFO is currently also led to the pond from just outside the oil containment basin. Oily wastewaters are also generated due to the oil spills during normal operations as well as maintenance. Mineral oil leakages are contained in sumps or barrels situated outdoors. Parts of these are not equipped with lids, which leads to


them filling with rain water, thus leading to spillages when they overflow. Spilled mineral oil and heavy fuel oil are mechanically collected with adsorb-ers like sand. No oil booms or other preventive measures are applied for oil spillages into the river and no oil spillage combating equipment could be found on site.

The waste waters from bottom ash cooling and other technical waters (less than 125 m3/h) are led to a small concrete sedimentation basin. Sanitary wa-ters are currently also led to this sedimentation basin as the existing separate sewage treatment plant has not been operational for years. From this sedimen-tation basin the waste waters are discharged directly into the river Sitnica without any further treatment.

The pH-value of waste waters is said to reach 9 - 10 and no neutralisation is done before discharge at either discharge point.

No monitoring of the wastewater discharged from the power plant itself is carried out.

Plans exist to direct the waste waters back to the power plant to be used as process water after treatment in order to reduce the amount of raw water that has to be purchased. This would also result in environmental benefits as the amount of partly untreated waste water discharged to the river would be sig-nificantly reduced, perhaps even by some 30 - 40 %. The implementation of this plan is said not to require significant investments or external expertise. The Consultant is of the opinion, that the costs and benefits of different meth-ods should be thoroughly explored to obtain the most cost-efficient method of reaching this goal of preparing process water of adequate purity from waste waters.

6.7.2 Waste water from ash disposal site The main sources of water to the ash disposal site are rainfall and water used to transport the ash. The suspension of water and ash is hydraulically trans-ported to the ash landfill where mechanical deposition of ash happens and overflow and drainage wastewaters are discharged through a canal into the river Sitnica without any other treatment than settling. The ratio of ash to wa-ter in the transport system is said to be as low as 1:1, because the pipelines are washed after ash has been pumped. However, the ratio is more likely to be at least somewhere between 1:2 and 1:5 to enable the ash transportation. Water is not recycled, and no precautions are taken to avoid infiltration of rainwater into the ash landfill.

No protective pumping around the ash disposal site to create a hydraulic bar-rier and thus protect the groundwater of the area is carried out.

The quality of waste water discharged is monitored regularly at eight points around Kosovo B and its ash disposal site. The waste water discharged ex-ceeds limit values for a recipient like Sitnica for several parameters - conduc-tivity, oxygen saturation, suspended substances and sulphates among others. Values for salts are very high while no monitoring is carried out for trace elements, because INKOS has no equipment for this.


The pH value of waste waters from the ash disposal area is 8 - 8,5. Hence, in large concentrations, it is harmful to fish and other animals living in rivers and lakes and humans using the water for household purposes.

Permanent monitoring upstream and downstream of the discharge point is not carried out and it is thus difficult to estimate the impact of Kosovo B on the river Sitnica. However, based on site inspection where small and large creeks are seen running to the river from the disposal area (see photo log) it can be assumed that a substantial impact exists.

6.7.3 Conclusions The following conclusions can be made:

Large quantities of wastewater are discharged into the river Sitnica from Kosovo B and its ash landfill

The river Sitnica is the most polluted river in Kosovo, and the part nearest to the KEK site is the most polluted

Groundwater is probably polluted although no monitoring results are available.

It can be concluded that most likely both surface water and groundwater are contaminated and should not be used as drinking water, for irrigation or other household purposes.

Contamination from runoff water and wastewaters from the power plant can be stopped and the rivers could again be suitable for providing household wa-ter, but the infiltration to groundwater from the ash disposal site will continue for hundreds of years even if no new quantities are added if no protective pumping is carried out to create a hydraulic barrier and thus protect the groundwater around the area. The infiltration from the ash disposal site can be significantly reduced also through the implementation of proper drainage sys-tems and covering of the site with soil and vegetation. The situation is com-plicated by the fact that the ash disposal site is located in an area prone to flooding.

The waste water amounts need to be reduced and the waste water treatment needs to be made more efficient. The following waste water treatment meas-ures, whenever needed for different waste water fractions should be imple-mented at Kosovo B:

• Mechanical separation of sludge / sand • Aeration of oxygen depleting waters • Biological treatment of nutrient-rich waters, chemical precipitation of

phosphorous if needed • Neutralisation of acid / alkaline waters through mixing in neutralisation

basin or through addition of acids/alkali • Oil separation wells need to be installed at discharge points • Sewage waters need to be led to the municipal sewage network, if

available. Alternatively the sewage treatment facility of Kosovo B needs to be rehabilitated or reconstructed.


• The current condition and sufficiency of the sewerage network of the power plant site needs to be determined to enable more exact formulation of the required measures to achieve an acceptable level of waste water management.

The degree of pollution of the waste waters also depends to a very high extent on management and housekeeping, which can still be improved significantly at Kosovo B.

Other possible means of reducing the impacts of the waste waters discharged from Kosovo B include:

• Fuel and ash residuals in the wastewater can be minimised by avoiding losses on site and by re-circulation of wastewater (re-use of water).

• Oil and chemical residuals from losses (dropping) to the drainage system can be avoided by appropriate handling procedures and instructions. Recycling of used oils and chemicals should be mandatory.

• Chemical residuals from processes can be reduced by use of appropriate and as harmless as possible chemicals, ie. hydrazine can be substituted by a more modern deoxygenating chemical (however hydratzine is more a health risk than environmental one) and by avoiding use of surplus of chemicals.

The inspectors from the Water department of the Ministry of Environment and Spatial Planning have offered assistance to KEK in the waste water treat-ment issues of Kosovo A and B, but with no results.

6.8 Waste Management Procedures for waste management have been introduced but not yet fully im-plemented.

KEK has a department for centralised management of waste materials.

6.8.1 Ash Bottom ash and fly ash have until now been disposed of in the easiest and cheapest way without environmental concerns. KEK has established a dis-posal site for bottom ash and fly ash conveniently just outside the Kosovo B power plant area where the heap formed by ash creates a visually very domi-nant element in the landscape. The ash landfill has exceeded the originally in-tended volume threefold by now. The originally intended maximum height of the landfill was 15 meters, but today the height of the landfill is at least about 30 - 40 meters. The landfill occupies an area of approximately 55 - 60 ha. It is said to be "geometrically out of control" which can lead to instability and other problems and risks.

The first priority should be to secure the physical stability of the landfill if it is to remain where it is now in the future. The impacts of the ash landfill on surface and groundwater should thereafter be minimised through measures described earlier (see chapter 6.7.2), dusting should be minimised and land-scaping of the landfill should be carried out. Some measures, like for instance


the covering of the landfill with soil and establishment of vegetation on it, have several benefits; in this case inhibition of rainwater infiltration and adap-tation to the landscape.

Fly ash is hydraulically transported as a suspension of water and ash to the ash landfill where mechanical deposition of ash happens on "plateaus" equipped with only very narrow embankments. Bottom ash is transported to the landfill with trucks and conveyors.

The quantities of combusted lignite and residuals are much higher than neces-sary due to low combustion efficiency and insufficient quality of the lignite received from the mines. Both can be improved in an economically sustain-able way, and at the same time the environmental impact of power generation and disposal will be reduced.

The ashes contain salts and trace metals of which some quantities, when in contact with water, sooner or later will be released. When trace elements and salts have leached from the ash they are discharged with the infiltrating water to the river or to the groundwater. Easily leachable trace elements and salts will leach already during the wet transport whereas the remaining substances will leach over a longer period of time.

Large quantities of dissolved salts will end up in the river due to the discharge of transporting waters and runoff waters to the river.

Ash spreading from the ash landfill due to winds is a significant environ-mental problem. No sprinkler system is installed at the ash landfill to prevent wind erosion. In situations with dry air and windy weather episodic air pollu-tion with ash particles occurs. The ash particles can spread quite far as the ash landfill is high.

Ashes can to some extent be used in cement production, in the construction industry and for road construction or rehabilitation, depending on certain properties of the ash. With the construction and road rehabilitation activities currently ongoing and planned for the future in Kosovo it should be possible to utilise large quantities of ashes, but so far only a small quantity of fly ash from Kosovo B has been utilised in the cement industry. A higher degree of utilisation for cement production, road construction and the construction in-dustry is considered, but this is reported to currently not be attractive for the industries. Whether this is for cost or quality reasons is not known.

The most environmentally friendly way to manage ashes is to use them as re-placement for sand, gravel and other natural materials for the cement industry, the construction industry and for roads. The second best solution is to return them to the mines. Considering the large quantities, even an effective utilisa-tion will not solve all the problems, and large quantities will still have to be disposed of. Testing of disposal of Kosovo B ashes in the Mirash mine was said to begin in October 2005. If it proves to be feasible, all ashes would thereafter be disposed of in the Mirash mine and the existing ash disposal area would not be used anymore. A detailed closure programme for the existing disposal site must however be developed and implemented to minimize envi-ronmental risks and impacts. The transport of all the ashes from the current


disposal site to some other disposal location, be it the Mirash mine or some other location, has also been discussed.

6.8.2 Waste oils Waste oil and chemical residuals are reported to be recycled according to the producer's instructions. Waste oils are stored outdoors in barrels in several places. Some of the barrels have no lid or cover thus enabling overflowing due to filling with rainwater. The barrels are mostly stored on bare soil with no water or oil containment structures.

The present storage platform for hazardous wastes, made of concrete but not covered, was not utilised at the time of the site visits. Waste oils are sold, but it is not clear to whom and for what purpose.

6.8.3 PCB Only very limited quantities of PCB's or substances containing PCB's are re-portedly used currently and the use is, if possible, avoided.

Since there is no special care of waste oils discarded into drums without spe-cific labels there is no labelling of oil, which may or may not contain PCB. There are no procedures or instructions for waste oil control or if there are any, they have obviously not been implemented.

Larger quantities were used before the conflict and were probably disposed of in the ash disposal site, but exactly where is not known.

6.8.4 Other waste Waste at Kosovo B is collected into containers. No specifically designated waste containers exist as there is no system for recycling of waste in Kosovo. Hazardous waste is however separated. Waste is classified and separated on location by the employees.

Solid waste residuals are the result of the water treatment processes and they contain trace metals and salts. The waste is removed as lime sludge in quanti-ties of an estimated 3 m3/d from Kosovo B. The waste is not analysed but the content of large quantities of salts and trace elements is obvious. This waste is disposed of together with ash products or separately at various places at the power plant.

Worn out equipment and materials removed from the site are reported to be recycled. Based on the site visits it would, however, seem that large amounts of such materials are stored outdoors at the power plant. This includes equip-ment possibly containing hazardous substances, such as old transformers con-taining oils. This equipment forms a potential risk for soil and water contami-nation. Vast amounts of worn out equipment and other waste is stored be-tween Kosovo B and the ash disposal site in an open field without any surface paving or other protection. The waste stored here includes waste oils in bar-rels and probably other hazardous waste as well.


Contaminated soil, dropped lignite, ashes and other solid products are col-lected and periodically disposed of in the ash landfills. There are no records of quantity or quality of these fractions.

6.8.5 Waste records The records of Kosovo B include tables registering the types and quantities of individual wastes from the plants during the year. No information regarding different kinds of wastes dumped in the ash landfills exists. It is likely that waste containing PCB's, asbestos and other hazardous substances has been disposed of this way at least earlier.

6.8.6 Conclusions A new disposal area for ash needs to be constructed in accordance with rele-vant regulations. Whether the planned new disposal area, the abandoned Mirash mine, fulfils the requirements set by the regulations is not known, but it is unlikely that this is the case.

Waste generated at Kosovo B is temporarily stored in several places at the plant site. Different solid waste types are mixed, except metal. There is no waste disposal area inside Kosovo B arranged in compliance with appropriate regulations and practices.

The waste storage area between Kosovo B and the ash landfill is unsustain-able and the use of this area must not be continued. Alternatively this area should be paved and equipped with drainage systems and roofed storage fa-cilities. The waste must be stored appropriately according to regulations de-termined in the Waste law, which has been approved by the Parliament, but not released by the Secretary General yet.

Hazardous wastes should be moved to a roofed warehouse with concrete floor and adequate containment basin in order to eliminate the risk of contamina-tion of soil and groundwater.

Solid waste residuals from the water treatment processes are disposed of to-gether with ash products or separately at various places at the power plant in an unsustainable way. These residuals need to be disposed of in a sustainable way.

Waste handling practices are said to have been improved, and vast amounts of waste have been removed from the site in recent years. Where they have been disposed of is not known.

One additional problem related to waste management is the lack of hazardous waste treatment facilities in Kosovo. The hazardous waste must be kept in storage or shipped abroad which is expensive.

6.9 Soil, surface water and groundwater contamination Both ground water and soil are most likely contaminated at the Kosovo B site.


A special problem is the gasification plant that was in operation at the site of Kosovo A until about 10 years ago. No monitoring results were available, but the groundwater in the area is most likely contaminated with phenols from this plant.

Management of the ground water resources must be improved and further contamination avoided. This should especially be a concern when new ash disposal sites are constructed.

The contamination of soil is visible all over the site but no actions have been taken so far to verify the extent of contamination. A detailed mapping of the site, systematic soil sampling and analyses should be carried out to create a picture of the situation.

No oil containment structures or equipment are available and no effort is made to stop the release of HFO into the environment from the place where excess water from the heavy fuel oil is discharged.

Monitoring of the water quality of the river is carried out by INKOS and monthly results are available, but this is far from sufficient to determine the true extent of contamination. The monitoring does not include all necessary parameters needed to evaluate the water quality.

The monitoring programme is insufficient also in terms of points and regular-ity of monitoring. The quality is measured appropriately upstream (before Kosovo B) but only measured several kilometres downstream from the Kos-ovo B after small rivers and creeks have added substantial quantities of dilut-ing non-contaminated water.

The monitoring plan should thus be improved to better reflect the impact of Kosovo B.

Because the ash disposal system is wet and the disposal area does not have any structural protection against pollution of surface or ground water it is ob-vious that harmful substances infiltrate to the groundwater and/or flow to the river Sitnica.

INKOS has newly established monitoring points around the disposal sites, but monitoring results are only available from one point for the whole year. Hence, no firm conclusions can be drawn, but assessment of the results from a point upstream and a point downstream in October 2002 indicate heavy con-tamination with pH values rising from 7,4 to 9,2 and the content of suspended materials and sulphate rising from 1435 mg/l to 2590 mg/l and from 335 mg/l to 1533 mg/l, respectively.

The content of trace metals is not analysed because of lack of suitable equip-ment.

6.10 Complaints Possible complaints have not been systematically filed at Kosovo B.


6.11 Environmental claims No environmental claims are known to have been filed.

6.12 Current environmental expenditure The EAR has spent € 175 million on the refurbishment of the plant since 1999. Many new components have been installed to maintain operation but it seems that absolutely no attention has been paid to even the simplest and cheapest measures to improve the environmental performance of the plant. It is evident that environmental issues have not been a priority in this case.

7 HEALTH AND SAFETY PERFORMANCE

7.1 National regulatory requirements In late 2003 UNMIK published safety regulations for Kosovo - Occupational Safety, Health & The Working Environment Regulation (No. 2003/33).

7.2 Applicable EU/international requirements and standards See chapter 6.2.

7.3 Key health and safety issues

Noise Although no measurements have been made, it can be estimated that the noise level may exceed the permissible levels at many working areas like work-shop, boiler room, engine room etc. A number of technical and organizational measures have been adopted to mitigate the burden on the personnel (noise resistant rooms, available protective aids etc.) but without measurements the benefit of these remains undefined.

No noise measurements have been carried out at the nearest residential areas, either. Thus it remains unclear, what the noise impacts are at these areas.

Lighting Lighting is quite poor in the boiler house and other areas of the power plant. This forms a health and safety risk for the workers, especially combined with poor housekeeping and insufficient protective measures. This is especially true in the boiler house and in the stairways.

History The recent development of occupational health and safety issues was also briefly reviewed in this study. The results are presented below.

Injuries during the years 2004 - 2005 have occurred as follows:


Kosovo B 2004 jan-sept. 2005 jan-sept.

nr. of injuries 28 14

some general statistics: 2004

scale of injury 5 % serious 95 % light by location: boiler house 47 %machine house 15 %ash and coal system 15 %water treatment plant 2,5 %work house 2,5 %other 18 % by part of body: head 9 %eyes 16 %hands 10 %fingers 12 %legs 35 %other 18 %

No other information was available.

Instructions and training Instructions for the field of safety and health protection have been issued by the management. Regular training in the field of occupational safety and health and fire protection is organized.


Personal protective equipment is allocated according to the needs of opera-tion. The usage of this equipment is, however, still not exemplary everywhere in Kosovo B.

7.4 Previous emergency situations and accidents Previous accidents or incidents include the cases described below.

Fire caused by lightning in 2002 A fire broke out because of lightning stroke the transformer of unit 1 in July 2002. The fire destroyed the control room and several other structures of the plant, and the personnel had to be evacuated by helicopter from the roof of the boiler house. One person is reported to have died in the fire.

The personnel of the site consider the environmental impacts of normal op-eration to be a bigger problem than risks of accidents or incidents.

7.5 Control of major accident hazards The fire fighting system is fed by three electrically powered pumps. The suf-ficiency of the capacity of the pumps and the capacity of the water reservoirs should be reviewed, as well as the functionality of this kind of system.

The fire fighting system should be checked regularly in its entirety. Fire ex-tinguishers are present inside Kosovo B. No automatic sprinkler or CO2-based extinguishing system exists. The control room, electrical rooms and cable spaces around should have a CO2-based extinguishing system.

Kosovo B has a number of full time fire prevention/fighting staff.

Protection from rotating machinery and open manholes should be improved through safety rails, better lighting and better housekeeping coupled with re-placement of missing covers or lids and implementation of safety regulations and other protective measures.

7.6 Current health and safety monitoring practice The Managing Director has set up a safety committee consisting of the Ex-ecutive Directors of KEK. This committee meets on a monthly basis to re-view safety issues. As the safety management systems are developed the committee will monitor their implementation in each division.

KEK has an Executive Office for Industrial Safety and Fire Protection, the primary objective of which is to inspect safety aspects in KEK and provide guidance on safety issues. It is not clear to whom the manager of this office reports. If possible the manager should report directly to the Managing Direc-tor of KEK.

Draft safety regulations have been developed in 2004. These should be final-ized in line with legal requirements and posted in all work locations.


Inspections should be carried out on a regular basis, and monthly reports de-tailing measures implemented to address issues detailed in the inspection re-ports should be produced by staff of Kosovo B.

Statistics of accidents and incidents at Kosovo B are compiled. These statis-tics include detailed information regarding type of injury and length of sick leave.

Several measures need to be implemented for Kosovo B to comply with The Occupational Safety, Health and Working Environment Regulation (No. 2003/33). These measures include among others:

• Improvement of prevention of risks through ensurance of equipment and machinery safety and through training of staff.

• Risk assessments - detailed risk assessments for each workplace should be carried out and mitigation measures should be formulated.

• Improvement of incident response through provision of equipment and training.

• Improvement of rest rooms, changing rooms, lighting • Improvement of electrical systems to match European standards • Improvement of noise and vibration abatement to comply with minimum

working standards.

8 SUMMARY OF REGULATORY COMPLIANCE STATUS

8.1 General In this section the compliance with laws, directives and standards is evaluated with the following priority order:

1. Kosovo legislation in force and EU-Directives. 2. International standards and norms, especially from the UN organisation

and/or EU countries, where no provisions are found in Kosovo legislation or in EU-Directives or recommendations.

Former Yugoslavian legislation that might still be applicable has not been re-viewed here.

According to the Law on Environmental Protection (No. 2002/8) environ-mental protection in Kosovo shall be based on the gradual introduction of European Union standards aimed at ultimately providing individuals with a healthy environment and the principle of using the most appropriate practices, accepted in the scientific community, for improving the environment.

Possible transition periods regarding compliance with EU regulations will be agreed upon later during EU accession negotiations.

8.2 Energy efficiency

• EU-Resolution 98/C 394/01


The resolution on energy efficiency in the European Community requests in-creased use of combined heat and power (CHP), including district heating and cooling, where appropriate.

This measure is also proposed in the World Bank – ESTAP report but it has not been implemented and no actions, like feasibility studies, technical speci-fications or tender documents, have so far been seen.

8.3 Re-use of substances

• The Law on Environmental Protection (No. 2002/8)

The use of substances that can be re-used, recycled, and divided biologically shall be a priority over the use of raw materials.

Kosovo B is currently in non-compliance regarding:

Water used for transport of ash as slurry can be recycled and the con-sumption of raw water reduced

Ashes can be reused in the cement industry, road construction etc.

8.4 Air protection

Dust emissions

• Law on Air Protection (No. 2004/30)

It is the duty of all natural, legal, local and international individuals to keep the air clean and to protect it during the activities they conduct in the territory of Kosovo, and hence they are obliged to:

a. monitor emissions b. minimize polluting emissions and unpleasant smells c. not to exceed limited values of emissions

The Government, on proposal of MESP, shall approve standards for discharge into the air, based on EU and WHO standards. The permitted limited values are not yet fixed but it is assumed that Kosovo B will be in non-compliance once they are fixed.

For existing air pollution sources, operating after the entry of this Law into force, in cases when the technical and technological level does not allow ap-plication of discharge standards in accordance with this Law, MESP shall au-thorize establishment of temporary standards. The temporary standards and the methods for their establishment shall be approved by the Government at the proposal of the MESP.

In cases, when after implementation of temporary standards, the pollution source causes damage to human health and the environment, MESP shall or-der closure of such source.

Discharging equipment should have such a level, that the level of the envi-ronmental indicator of air quality does not exceed the admissible limit for an


indicated pollutant. The level of discharge requirements of equipment shall be determined on a case basis in the environmental permit.

Natural and legal persons, local or foreign ones are obliged to acquire an en-vironmental permit, issued by the MESP in cooperation with the local author-ity in charge for, among others:

the constructing and using of pollution sources, and for later modifica-tions of such sources

changing technical and operational parameters and organizational technical measures of the pollution source

Facilities that cause air pollution are obliged to organize internal monitoring in order to follow the pollution levels of the air.

Air pollution by pollution source operators, which violate discharge norms, norms for special protection zones, smog and emergency regulations, techni-cal conditions and other requirements specified in the environmental permit, big and medium pollution sources shall have imposed on them penalties start-ing from € 1 000 to € 10 000 depending on the type and amount of discharge. Their activity shall be banned until the operator brings their discharge under the allowed limits.


The Government, after receiving a proposal from the Minister, shall have the authority to issue subsidiary normative acts establishing the acceptable limit levels for the discharge and emission of pollutants – including solid, liquid, gaseous pollutants – and hazardous substances into the air, water and soil.

Where it deems it to be in the public interest to do so, the Government may suspend, in whole or in part, the applicability of a subsidiary normative act to an existing polluter if such existing polluter enters into an administrative con-tract with the Government requiring such existing polluter to comply with a specific set of ever more strict emission standards over time, in accordance with a specified set of progressive deadlines. Any such contract shall require the concerned existing polluter to pay on a regular basis a reasonable fee, which shall become part of the Kosovo Consolidated Fund. Each such con-tract shall be submitted to the Assembly for review, and no such contract shall become effective unless and until it is ratified by the Assembly.

Where it deems it in the public interest to do so, the Government may enter into an administrative contract with an existing polluter requiring such pol-luter to comply with stricter emission standards than those established by the relevant subsidiary normative acts issued pursuant to this law. In order to provide existing polluters with sufficient incentives to enter into such an ad-ministrative contract, the Government shall develop proposals, for considera-tion by the Assembly, recommending that the Assembly establish certain in-centives that the Government may offer to existing polluters. Any such con-tract that grants an incentive to an existing polluter shall be submitted to the Assembly for review, and no such contract shall become effective unless and until it is ratified by the Assembly.


The Government shall issue a subsidiary normative act establishing pre-scribed permissible maximum levels for the discharge and emission of pollut-ants into the air. The Government shall ensure that the levels established are consistent with the ability of Kosovo to comply at a reasonable cost. The Government shall have the authority to adjust downward such permissible maximum levels over time to bring them gradually into compliance with the prescribed levels of the EU.

• EU-Directive 2001/80

The dust emission limit value for Kosovo B is 50 mg/Nm3.

The results of the dust emission measurements of unit B1 in October 2005 were 526 - 577 mg/Nm3 (dry gasses, standard conditions, 6 % O2). The dust emissions from unit B2 could not be measured as the unit had a forced shut-down and was repaired during the measuring period. As the ESP's of unit B2 have been reconstructed recently it is likely that the dust emissions from this unit are at least somewhat lower.

The particle concentration in flue gases from unit B1 measured in October 2005 exceeds the EU limit value (50 mg/Nm3) more than tenfold.

SO2 emissions


See above under "Dust emissions".




The SO2 emission limit value for Kosovo B is 400 mg/Nm3 or a desulphurisa-tion rate of at least 94 %.

The results of the SO2 emission measurements of unit B1 in October 2005 were surprisingly low, only 142 - 332 mg/Nm3 (dry gasses, standard condi-tions, 6 % O2). The SO2 emission value after ESP was rather unstable during the measurement campaign, which refers to some changes either in boiler sul-phur capture or in lignite quality. As the fluctuations were significant and ir-regular additional continuous measurements over a longer period are required to establish the actual SO2 emissions, which may be higher.

The SO2 emissions from unit B2 could not be measured as the unit had a forced shutdown. It can be assumed that the SO2 emissions of unit B2 are in line with those of unit B1 due to the same technology and fuel.

As the exact long term SO2 emission values from units 1 and 2 are not known it is difficult at this stage to determine if LCP-compliance requires additional desulphurisation measures to reach the EU limit value.

NOx emissions







Assuming that the operation of Kosovo B is planned to be more than 2 000 hours between 2008 and 2016 the emission limit value for Kosovo B is 500 mg/m3 until the end of year 2015, and 200 mg/m3 from the beginning of the year 2016.

The results of the NOx emission measurements of unit B1 in October 2005 were 661 - 713 mg/Nm3 (dry gasses, standard conditions, 6 % O2). The NOx emissions from unit B2 could not be measured for reasons stated above. It can be assumed that the NOx emissions of unit B2 are in line with this thanks to the same combustion technology and fuel.

Based on the measurement results from October 2005 it can be assumed that some measures will be necessary to comply with the EU requirement of maximum 500 mg/Nm3. The nitrogen oxide concentration in flue gases meas-ured in 2005 exceeds the EU limit value by at least 30 %. After the year 2015 the EU limit value for NOx emissions will be 200 mg/Nm3. Burner adjust-ments will be sufficient until the year 2015 according to Alstom (the current manufacturer of the applied tangential burner system), but the stricter limit from 1.1.2016 onwards will require air staging and installation of special low-NOx -burners.

Air quality


Facilities that cause air pollution are obliged to organize internal monitoring in order to follow the pollution levels of the air.

Limited areas with high air pollution levels, upon the proposal of the MESP and the Health Ministry, shall be proclaimed by the Government as areas that require special air protection. For specially protected areas, MESP, the Health Ministry and relevant local authorities shall instruct particular measures for protection of air quality.

• EU-Directive 1999/30 on Air Quality

The air quality limits are:

Dust

24 hour limit value: 50 µg/m3 from 2005 4) Annual limit value: 40 µg/m3 from 2005 and 20 µg/m3 from 2010

SO2

Hourly limit value: 350 µg/m3 from 2005 2) 24 hour limit value: 125 µg/m3 from 2005 1) Annual limit value: 20 µg/m3 from 2001

NOx


Hourly limit value: 200 µg/m3 from 2010 3) 24 hour limit value: 40 µg/m3 from 2010 Annual limit value: 30 µg/m3 from 2001

1) Not to be exceeded more than 3 times during a calendar year 2) Not to be exceeded more than 24 times during a calendar year 3) Not to be exceeded more than 18 times during a calendar year 4) Not to be exceeded more than 35 times during a calendar year (stage 1) 7 times during a calendar year (stage 2, to be met by 1.1.2010)

---

The impacts of the emissions of Kosovo B on the ambient air quality have been evaluated by the Finnish Institute of Meteorology in December 2005 (see appendix 5 for the full report). The dispersion of sulphur dioxide, nitro-gen oxides and particulate matter emitted from the power plant was simulated by the Finnish Meteorological Institute (FMI) using the operational local scale air pollution dispersion model. The computed concentrations were com-pared with EU limit values.

Based on the results it can be concluded that around the stack there is zone of several hundred meters with quite low concentrations which is typical for stack releases (‘stack shading’ effect). The distribution patterns suggest that the main dispersion sectors are southwest or south from the emission source which agrees well with the wind distribution in the used meteorological data.

From the distributions of the highest sulphur dioxide concentrations it can be seen that the maximum of the annual mean (0,3 µg/m3) is situated about 2 kilometres to the southwest of the plant site. The maximum of the highest daily mean (3,2 µg/m3) is situated about 2,7 kilometres towards southwest and the maximum of the highest 1-hour mean (21 µg/m3) is situated about 1 kilometre to the south of the plant site.

From the distributions of the NO2-concentrations it can be seen that the maxima are situated further away from the plant site than the maxima of SO2.

The main reason for this is that the dispersion model takes into account the chemical transformation of nitrogen oxides during the dispersion process. The maximum of annual mean NO2-concentration (0,15 µg/m3) is situated about 4,2 kilometres to the southwest from the Kosovo B plant site and the maxi-mum of the highest 1-hour mean (12,3 µg/m3) is situated about 6 kilometres to the southwest from plant site.

The maximum of the annual mean (0,4 µg/m3) and the maximum of the high-est daily mean (1,4 µg/m3) of particle concentration is situated about 2 kilo-metres to the southwest of the plant site.

The comparison of the computed maximum SO2, NO2 and particulate matter (PM) concentrations to the respective EC air quality limit values is presented below in figure 8/1. All concentration values used in the comparison are those corresponding to the proper statistical definitions of the limits of the EC.


FIGURE 8/1 Comparison of the calculated maximum sulphur dioxide, nitrogen dioxide and particle concentrations with the EC limit values. The maximum computed 1-hour mean SO2-concentration (21 µg/m3) is 6,0 % of the limit value (350 µg/m3). The maximum of the computed 24-hour mean SO2-concentration (3,2 µg/m3) is 2,5 % of the limit value (125 µg/m3). The maximum computed 1-hour mean NO2-concentration is 6,1 % and the maxi-mum annual mean is 0,4 % of the respective limit values. According to the re-sults of the dispersion simulations the maximum 24-hour mean particle con-centration is 2,9 % and the maximum annual mean is 1,1 % of the respective stage 1 limit values for PM10.

The computed maxim annual mean nitrogen oxide (NOx) concentration is 2,1 % of the respective limit value and the maximum annual mean sulphur diox-ide (SO2) concentration is 1,5 % of the respective limit value. These annual limit values for NOx and SO2 are given for the protection of vegetation and their application area is limited. The comparison is made for the absolute maximum concentrations of the study area. All over the rest of the study area the concentrations are lower than those used in the comparison.

According to the dispersion model study the stack used (182 meters) is high enough to assure that the concentrations due to emissions of the Kosovo B power plant are sufficiently low on the ground surface to clearly meet the air quality standards of the European Communities. It has also to be recognized that all these maximum values fall onto the areas that are within the existing or future mines and there will not be any population centres there.

Based on these results the poor air quality in the area, especially with regards to dust, is mainly caused by dusting from the ash disposal site. Emissions from Kosovo A also contribute more to the poor air quality than Kosovo B.

Green house gasses


The impact of the implementation of the Emission Trading Scheme (ETS) of EU on Kosovo B can not be assessed at this point as Kosovo has no official position as far as the Kyoto protocol is concerned.

The other required emission reduction measures, such as optimisation of boiler performance, and the measures to increase energy efficiency of the plant that most likely will be formulated and implemented sooner or later will lead to significant reductions in CO2 emissions as well.

Conclusions Kosovo B is currently in non-compliance regarding several of the EU regula-tions concerning emissions into the air, but not regarding air quality. The Law on Environmental Protection however apparently allows the Government to only gradually implement these EU regulations.

8.5 Water protection


According to the law discharge of solid, liquid or gaseous non-toxic materials into water that causes change in the turbidity, sedimentation, flavour or other changes in the quality of water is prohibited.

The Government shall issue a normative act establishing prescribed permissi-ble maximum levels for the discharge and emission of pollutants into the wa-ter. The Government shall ensure that the levels established are consistent with the ability of Kosovo to comply at a reasonable cost. The Government shall have the authority to gradually adjust such levels over time to bring them gradually into compliance with the prescribed levels of the EU.

Without prejudice to standards, rules and prescribed limit levels of pollution, discharge of solid, liquid or gaseous non-toxic materials, including organisms and energy into waters, that directly or indirectly cause a change in the turbid-ity, sedimentation, flavour or other quality of the water is prohibited.

The subsidiary normative act authorized by this law may require industrial and other large scale undertakings and public authorities that discharge pol-lutants into the water and public sewage systems to, where necessary, install the equipment necessary to measure and/or record the type and quantity of such discharges and to regularly provide such information to the ministry.

Kosovo B is in non-compliance as wastewater is discharged untreated or in-sufficiently treated and as no precautions are taken to reduce the water infil-tration of disposals and contaminated soil and consequently the groundwater is contaminated.

• The Water Law (approved by Parliament, not yet issued)

According to the law it is prohibited to use waters for waste water discharge in a way that damages ecological or chemical conditions of natural lakes, fish ponds or other watercourses or groundwater.

The Government, through a subsidiary act, will define the limit values for ef-fluents discharged into water recipients and into the public sewage network.


The holder of the water permit required in the near future for discharge of waste waters is obliged to install and maintain equipment for measurement of quantity of wastewater discharge.

According to the law a water permit will be needed. This water permit shall be applied for from MESP within 18 months after the Water law enters into force. This permit will stipulate waste water discharge limit values to be de-fined. All water activities shall be harmonized with conditions and procedures according to the Water law within a period of 36 months.

Kosovo B is in non-compliance as wastewater is discharged practically un-treated and unmeasured to the river Sitnica and as no precautions are taken to reduce the water infiltration from disposal areas and contaminated soil and as consequently the groundwater is contaminated.

8.6 Soil protection


According to the law any discharge, disposal or storage into the soil of harm-ful or hazardous substances (including solids and liquids) beyond permitted limit values, as required by special legal acts, is prohibited. The Government shall ensure that the levels established are consistent with the ability of Kos-ovo to comply at a reasonable cost. The Government shall have the authority to gradually adjust such levels over time to bring them gradually into compli-ance with the prescribed levels of the EU.

The permitted limited values are not yet fixed but it is assumed that Kosovo B will be in non-compliance once they are fixed at least regarding hydrocarbons as extensive soil contamination most likely occurs because of past and current practices of oil storage and handling. The need for actual decontamination measures is however difficult to predict based on the material available.

According to the law a legal person or natural persons who explores or ex-ploits natural resources, deposits waste, ashes, slag or other materials must re-store the surface of the soil, the previous water regime and the geo-mechanical stability. The restoration must be conducted during the operation and after termination of such activities, in accordance with approved envi-ronmental protection and re-cultivation projects.

Kosovo B is in non-compliance as no environmental protection and re-cultivation project has been elaborated and as no major restoring is conducted during operation.

8.7 Waste management

• Waste Law (approved by Parliament, not yet issued)

The law deals with wastes and waste management, landfills and treatment and discharge of liquid effluents from facilities for holding, recovery, treatment or disposal of waste.


A waste management license will be required. This license ensures that the conditions for waste management are met. Facilities which require an envi-ronmental permit according to the Law on Environmental Protection will have to compile a waste management plan. All operators that require a waste management license shall also compile an operational waste management plan, the content of which is defined in the law.

Waste shall be collected and sorted according to the treatment it will go through. Waste shall also be stored safely utilising precautionary measures for prevention or reduction of negative impacts on the environment and human health. Additionally waste management records shall be kept.

According to the law landfills also include internal landfills, where a producer of waste is carrying out his own waste disposal at the place of production, and places used for temporary storage of waste for more than one year.

The operator of landfill facilities shall compile an operational waste manage-ment plan for the landfill and ensure its implementation, dispose of waste in accordance with the conditions set out under the waste management license, apply measures which guarantee environmental protection in accordance with the provisions in force, and maintain records according to the provisions of the Waste law.

Waste shall be stored in technically prepared places for temporary storage of waste. The period of time for temporary waste storage should be no longer than three years.

Kosovo B is in non-compliance in several respects, as no waste management plan that would fulfil the provisions of this law exists, as different wastes are not sorted and stored temporary in adequate facilities, as liquid effluents from the ash landfill are not treated properly and as adequate precautionary meas-ures to prevent or reduce environmental damage are not applied.

• EU-Directive 99/31 on landfills of waste

The EU directive on landfill of wastes prescribes, among others, drainage layer for collection of infiltrated wastewater and surface layer of soil to pro-tect against infiltration.

Kosovo B is in non-compliance as no drainage layer has been established in the ash disposal site and as no surface cultivation, apart from some demon-stration projects, has been performed.

Disposal of ash in the Mirash mine should be carried out in accordance with the Law on waste and EU-regulations. Whether these regulations have been taken into account when preparing to test ash disposal in the Mirash mine, is not known.

8.8 Noise protection


The Ministry shall issue provisions which regulate permitted noise limit lev-els and measuring of noise.


No provisions according to the law are implemented at Kosovo B.

• EU-Directives

None of the directives specify limits for noise.

• Other

The immission (reference) point targets, equivalent sound pressure levels LAeq, vary from country to country, but generally they can be expressed as:

Daytime 50 - 55 dB(A) Nighttime 40 - 45 dB(A)

Occupational exposure limits specify the maximum sound pressure levels and exposure times to which nearly all workers may be repeatedly exposed with-out adverse effect on their ability to hear and understand normal speech. Ac-cording to WHO an occupational exposure limit of 85 dB for 8 hours should protect most people against a permanent hearing impairment induced by noise after 40 years of occupational exposure. Machinery which cannot fulfil this requirement shall be noise insulated with protection walls or similar. Warning signs are required at all entrances to such rooms and ear protection in these areas should be mandatory.

Kosovo B is most likely in non-compliance with regards to internal noise at several working places of the plant, although no measurement results were obtained.

8.9 Health and safety

• Occupational Safety, Health & The Working Environment Regulation (No. 2003/33)

The regulation sets requirements regarding health and safety issues in the working environment.

Judging by the current situation regarding lighting, dusting and other issues Kosovo B is most likely not in compliance with this regulation.

• EU-Directive 90/384 (Amendment to 89/391)

The directive sets requirements regarding protection of workers from the risk related to carcinogens at work (among others protection against asbestos).

No measures accordingly have been taken at Kosovo B.

8.10 PCB


Any undertaking disposing of waste oil must obtain a permit from the compe-tent authority. Any undertaking collecting waste oils must be registered and adequately supervised. Regenerated oils may not contain more than 50 parts per million (ppm) of polychlorinated biphenyls and terphenyls (PCB/PCT).



This directive aims at the elimination of polychlorinated biphenyls and poly-chlorinated terphenyls (PCB's) and at the decontamination of equipment con-taining them. Equipment containing PCB's not yet decontaminated must be kept in good working order to avoid leaks.

Undertakings disposing of PCB's must be licensed in accordance with Direc-tive 75/442/EEC. Incineration as a means of disposal must meet the standards set in Directive 94/67/EEC.

Kosovo B is in non-compliance as no inventory of equipment containing PCB has been done, and as the transformer which was destroyed by lightning a few years ago which most likely contains PCB's is still stored outside on plain ground without any decontamination measures carried out.

8.11 Environmental monitoring and information


The law specifies detailed “Discharge and Emission Monitoring and Envi-ronmental Record Keeping” and “Environmental Information System” to be established at Kosovo B.

The current monitoring conducted by INKOS is far from sufficient to comply with the law. INKOS has prepared a “Strategy Proposal for Environmental Monitoring in KEK, April 2002”. However, this proposal will only improve the current monitoring but not fully comply with the provisions of the law.

• EU-Directives

All relevant EU-Directives have provisions on how monitoring shall be per-formed and on the regularity and numbers of monitoring. Monitoring is in most directives required on hourly, daily, monthly and annual basis and limit values are issued accordingly. Further, audit on monitoring is required.

Kosovo B is in non-compliance with these requirements both regarding moni-toring methods and equipment as well as regularity and numbers of monitor-ing.

No regular audits on Kosovo B’s monitoring methods and equipment are im-plemented.


9 CONCLUSIONS AND RECOMMENDATIONS

9.1 Conclusions This pre-feasibility study has been carried out with complete EU-compliance of Kosovo B as a prerequisite as the Law on Environmental Protection states that environmental protection in Kosovo shall be based on the gradual intro-duction of European Union standards. Possible transition periods for future EU compliance have not been taken into account.

It is assumed that the plant will be operational at least for the next 15 years. Its load will be close to base load as the Kosovan lignite offers very competi-tive fuel resource if compared with the other power producers in the Balkan region. If the Kosovo B units shall continue unlimited operation well into the future it will be necessary to improve the level of environmental mitigation measures of the plant significantly. SO2, dust and NOx emissions will have to be reduced, which will require investments in abatement technology.

The Kosovo B power plant's stack emissions do not constitute a major prob-lem with regards to air quality. The air quality situation in the region is how-ever alarming, mainly because of the dust emissions from the ash landfills, Kosovo A power plant and the adjoining lignite drying plant. Compliance with the EU LCP-directive in the future and reduction of fugitive dust emis-sions from the ash disposal area will improve the situation. A visible im-provement in Kosovo B (and especially for Kosovo A plant) might be a deci-sive factor in getting the public approval of the new power plant develop-ment.

Waste and wastewater management installations and practices will have to be improved to comply with Kosovan legislation, EU-Directives and best envi-ronmental practices. This will also require investments. If a part of the fly ash and the bottom ash can be utilized the disposed quantities may decrease some. On the other hand required more efficient dust removal will counteract this and may result in larger quantities of ash to be disposed of. This all depends on the amounts of ash that can be utilized. The current ash disposal site can not be utilized for much longer, and a new site will have to be found. Old abandoned lignite quarries are the most logical and environmentally sustain-able places for this, and the Mirash mine is to be used for this purpose in the near future. The disposal should be done in compliance with the EU directive 1999/31 on waste landfills regarding drainage layers and covering with soil and other required measures. Whether this has been complied with at Mirash is not known.

Aftercare of the current ash landfill is required to mitigate impacts on the river Sitnica, groundwater and ambient air quality. This will require signifi-cant investments.

The current overall energy efficiency is less than 30 % and it must be im-proved for environmental and economical reasons closer to the initial 39 - 40 % design performance.


A monitoring programme in line with the requirements of various EU direc-tives regarding air quality, water quality and emissions must be established to demonstrate compliance. Monitoring of emissions into the air from the power plant will require investments in continuous monitoring equipment in order for compliance with the monitoring requirements of the EU LCP-directive to be achieved.

The schedule for these measures can not be determined at this stage. Kosovan legislation allows gradual implementation of EU regulations. Possible EU ac-cession negotiations in the future may also touch this subject, as has been the case in other accession countries.

9.2 Key risks and liabilities of the project Not applicable in this assessment project.

9.3 Process efficiency and environmental opportunities The design nominal rated power of the boiler unit is 339 MWe as gross elec-trical generator output. The design fuel consumption is 1,13 t/MWhe giving a plant heat rate of 8990 kJth/kWhe and gross efficiency of 40,0 % calculated as MWe gross/MWth. This performance has been demonstrated in delivery tests in 1984.

In the present conditions the maximum obtained power is 285 MWe. The in-dicated fuel consumption is in the range of 1,4 - 1,5 t/MWhe giving the plant a heat rate of 11 000 - 12 000 kJth/kWhe and a gross efficiency of 30 - 33 %.

It shall be noted that the specific plant emissions of CO2, SOx, NOx and dust calculated as mass flow in t/MWh are directly proportional to plant fuel con-sumption, i.e. heat rate and efficiency. The total pollutant mass flow is deter-mining the effect stack emissions to the ambient air quality i.e. immission values as ground level concentrations and environmental impact on soil by the pollutant deposits as well as long range pollutant transportation.

Therefore it is of utmost importance to bring the boiler efficiency to the range of the design value.

Energy conservation seems to be neglected in KEK’s strategy and in reports assessed during elaboration of the study. Energy conservation is recognised by the EU as probably the best method to establish sustainable energy con-sumption and at the same time it saves money.

The energy conservation potential regarding Kosovo B is probably huge as no measures have been implemented so far.

A specific measure to increase the process efficiency could be the implemen-tation of district heating in Prishtina, as combined heat and power production is environmentally far more efficient than power production alone.


9.4 Environmental action plan The following Environmental Action Plan has been formulated based on both what can be achieved in the short term without major investments, and what is necessary to implement in order to obtain compliance with Kosovan legis-lation and EU directives.

As very few environmental protection measures have been implemented at Kosovo B so far it is necessary to implement measures not needing major capital expenditure in order to show that the management of KEK takes envi-ronmental issues seriously. These improvements are more related to the pri-oritisation of environmental issues, the improvements in current procedures, the raised awareness of all employees and the systematic management of en-vironmental, health and safety issues as a part of power plant operations. Such measures can be found in the field of waste management, water recycling, general housekeeping issues and environmental education and awareness.

The environmental benefits of these short term measures can be significant, and every effort should be made to implement these measures without delay.

Recommendations regarding the KEK organisation have not been given here. The recommendations of ESB International (2004) regarding this matter are referred to here.

No actual implementation plan has been prepared as this was out of scope in this assignment. The proposed measures are listed below as logical groups.

9.4.1 Air pollution abatement

SO2 emissions Desulphurization. As stated earlier the measurement results cannot be

considered to be representative for long-term emission consideration. There was a strong fluctuation in measured SO2 emission indicating either variations in lignite quality or variations in boiler sulphur cap-ture. The fuel and fly ash Ca/S mass balance does not support the un-derstanding of he actual sulphur capture in the boiler. In earlier meas-urements by INKOS the sulphur capture in the boiler was in range of 60 %. In similar boilers with similar fuels with ample Calcium (Esto-nian oil shale) sulphur capture in the range of 70…80 % has been measured. To analyse and clarify the actual boiler behaviour in re-gards to sulphur capture in the boiler due to fuel alkalinity in various operating conditions and fuel qualities we recommend the following:

Install a permanent SO2 analyser after at least one boiler unit, preferably for both boilers.

Launch a 6 month research program/measurement campaign to study the behaviour of the SO2 emission in different operating conditions and variable fuels coming from the mines.

Include in the research program detailed laboratory analysis of the fuel and fly ash parameters to determine the organic sul-phur analyzing method, limestone calcination degree in the


boiler and Ca/S balance. This Follow-up Program is required to establish reliable design parameters to the measures required to bring Kosovo B TPP SO2 emissions to the level required by LCP directive. The selected method for SO2 reduction dictates also the requirements for the ESP enlargement.

The total cost for implementation of these recommendations is estimated to be 0,8 MEUR.

To reduce the plant SO2 emissions the following SO2 reduction tech-nologies could be considered:

Classification of lignite by sulphur content and firing low sul-phur fuel in Kosovo B. Based on the present understanding this is only a theoretical concept and is not possible to imple-ment at Kosovo B.

Dry lime injection into the boiler. Additional SO2 removal po-tential in Kosovo B boilers with lime/limestone injection is rather low but not completely outruled. Investment needs 8 - 10 MEUR + operating costs 3,4 MEUR/a. ESP enlargement re-quirement must be clarified.

Installing a Flue Gas Desulphurization system after Kosovo B boiler plant. The following technologies should be considered:

o Semi-Dry flue gas desulphurization. SDA: Investment needs 50 MEUR + operating costs 6,3 MEUR/a. Inte-grated desulphurization process: Investment needs 40 MEUR + operating costs 6,0 MEUR/a. ESP enlarge-ment is not required in either of these alternative meth-ods.

o Wet flue gas desulphurization, WFGD. Investment needs 100 - 110 MEUR + operating costs 9,6 MEUR/a. The need for ESP enlargement shall be clarified based on the technical solution.

Dust emissions Dust emissions will need to be reduced in the future to comply with

regulations. To meet EU LCP directive requirements an electrostatic precipitator with five electrical fields and a relative size of 2,5 com-pared to existing precipitators is required. ESP investment needs 20 MEUR + operating costs 1,2 MEUR/a.

NOx emissions Nitrogen oxide emissions will need to be reduced in the future to com-

ply with regulations. It can be assumed, that it is possible to reach the LCP directive NOx level for the period 1.1.2008 - 31.12.2015 by tun-ing up and optimizing the existing combustion system. This requires retuning the burner and combustion air distribution systems. The bot-tom burners can be optimized to operate at lower combustion sto-


chiometry that the top burners. Also the total excess air rates hall be optimised. At the same time the operation and maintenance practices of the lignite pulverizing system has to be optimised to prevent the in-crease of fly ash UBC content. It is important to keep the fineness of the pulverized lignite close to the original design values. Total costs 0,6 - 1 MEUR.

The level 200 mg/Nm3 6 % O2, to be applied from 1.1.2016 onwards, can be reached by installing new Low-NOx burners and a staged com-bustion air system with OFA nozzles. Low-NOx-burner investment 20 - 26 MEUR.

Action plan for air pollution abatement Because the measures required to decrease the dust emissions are very much dependent on the final selection of SO2 emissions reduction method, it is im-portant to clarify the SO2 issue first. Therefore the following action plan is proposed:

1 Clarify the SO2 emission characteristics of the boilers = Installation of emission analyz-ers and implementing the Follow-up Program

2 Establish the design criteria for SO2 reduction technology

3 Issue the tender for SO2 reduction technology

4 Select the suitable FGD reduction technology

5 Establish the design criteria for dust reduction technology depending on the selected SO2 reduction method as follows:

• If Lime/limestone injection is selected, determine the design criteria for ESP modification

• If Semi Dry FGD technology is selected, no ESP modification is required

• If Wet FGD technology is selected, determine the need and degree of ESP modification depending on the process requirements and whether waste gyp-sum or commercial gypsum is to be produced.

6 Execute the required modifications for ESP

7 Carry out the boiler combustion system optimization for NOx reduction

8 Carry out the measures to increase the boiler efficiency

A summary of costs and schedules is presented below:

COST SUMMARY ESP Lime Inject SDA ISDA WFGD Low-NOx

burners Investments M€ 20 10 50 40 110 26


Investment costs M€/a 1,5 0,8 3,8 3,0 8,4 2,0 Operating costs M€/a 1,2 3,4 6,3 6,0 9,6 Total costs M€/a 2,7 4,2 10,1 9,0 18,0 2,0 Specific total costs €/MWh 0,6 0,9 2,2 2,0 3,9 0,4 €/t dust 0,27 €/t SO2 0,35 0,31 0,28 0,55 €/t NO2 0,33 Delivery time Months 18-22 6-8 22-24 22-24 30-32 26 Erection time Months 10 Plant shut-down time New flue gas line Weeks 4-6 6-10 6-10 8-12 Modify existing units Months 3-5 1-2 4

For more in depth details please refer to appendix 4.

9.4.2 Water pollution abatement Waste water treatment improvement through establishment of appro-

priate treatment installations. Sanitary waste waters are assumed to be led to municipal sewerage network. Investment needs in the order of 0,4 - 0,6 MEUR.

Recycling of waste water in accordance with existing plans. Plans for recycling of process water have been discussed at the plant. This will reduce the amount of waste water discharged into the river Sitnica significantly if implemented. Re-circulation and utilisation of waste-water especially for the ash handling process should be introduced as this would significantly reduce the water consumption and discharged wastewater amounts at the site. However, discharging harmful waste waters into the ash transport water is not a sustainable solution and must not be encouraged as a substitute for appropriate waste water treatment. Investment needs can not be estimated as exact plans and methods are not known.

Development of drainage system in order to eliminate rainfall runoff from current ash landfill into the river Sitnica. See 9.3.3 aftercare pro-gram for ash landfill.

Discharge of oily waters from HFO storage tank has to be stopped, oily waters are to be collected and treated according to best practices and regulations. To prevent pollution of river Sitnica with mineral oil and heavy fuel oil, it is necessary to supplement and improve the ex-isting preventive and combating protection measures and equipment. Part of waste water treatment system improvement, see above.


9.4.3 Waste management A waste management plan must be prepared and implemented. In-

vestment needs 25 000 - 35 000 EUR for preparation of plan.

Utilisation of fly ash and bottom ash should be explored. Residuals from combustion can replace use of natural recourses as sand and gravel etc. No investment needs.

A new ash disposal site will have to be constructed (Mirash mine may already be in use for this purpose). EU compliance will require in-vestments in drainage structures, bottom isolation layers etc. if these measures haven't been implemented at Mirash. Investment needs in the order of 3 - 5 MEUR including design, not including land acquisition costs.

Removal of all waste from the site. Waste removed should be dis-posed of in a sustainable way. Investment needs low.

Handling and storage of hazardous waste according to regulations in bunded and roofed area. Investment needs 10 000 - 20 000 EUR for repair of existing storage.

Development and implementation of an aftercare program for the ash disposal site based on experience gained from Kosovo A ash disposal site. The first priority should be to secure the physical stability of the landfill if it is to remain where it is now in the future. The infiltration from the ash disposal site should be significantly reduced through the implementation of proper drainage systems and covering of the site with soil and vegetation. Dusting should be minimised and landscap-ing of the landfill should be carried out. Plans for transportation of the deposited ash to old open pit mines have also been discussed. Invest-ment needs in the order of 10 - 12 MEUR including design (Removal cost in the range of 50 MEUR).

9.4.4 Fuel storage HFO storage tanks and oil pipes are to be inspected and repaired if

needed. The HFO tank of the heating plant is to be decommissioned and removed if found unnecessary. Investment needs low.

The secondary retaining basin of the HFO tank has to be inspected and repaired if necessary. Investment needs low.

9.4.5 Soil and groundwater contamination Soil and groundwater contamination surveys to map the true extent of

contamination need to be carried out. Based on the findings a remedia-tion project comprising of soil and ground water remediation and pro-tection measures should be elaborated. Investment needs in the order of 70 000 - 80 000 EUR.


Implementation of a remediation project of contaminated soil and groundwater should be carried out. Investment needs probably high, in the order of hundreds of thousands EUR depending on magnitude of contamination.

9.4.6 Monitoring A reliable monitoring system for emissions into air and water urgently

needs to be elaborated. Comprehensive monitoring programmes are needed to verify the environmental impact of future operation. The performed monitoring of emissions and air quality as well as waste water and quality of water in rivers and lakes is insufficient to estab-lish the true picture of the extent of pollution. Calibration of current equipment and procurement of new equipment to comply with, among others, the requirements of the LCP-directive is needed. Investment needs 0,2 - 0,4 MEUR and annual maintenance & calibrations 20 000 - 40 000 EUR.

9.4.7 Environmental management and awareness Development of an Environmental Management System for (EMS) for

Kosovo B. Investment needs in the order of 50 000 - 75 000 EUR.

Implementation of an Environmental education and awareness cam-paign for management and all employees of KEK/Kosovo B. Invest-ment needs 30 000 - 40 000 (for consultancy if needed).

TABLE 9 - 1 Environmental Action Plan summary.

Chapter Measure Investment needs (estimate)

Annual operating costs (estimate)

6 month follow-up program to de-termine which SO2 abatement tech-nique is sufficient

0,8 MEUR -

SO2 emissions abatement, alterna-tive methods:

• Dry lime injection into the boiler

• Semi-Dry flue gas desul-phurization. SDA

• Semi-Dry flue gas desul-phurization. Integrated desulphurization process

• Wet flue gas desulphuriza-tion

10 MEUR 50 MEUR 40 MEUR 110 MEUR

3,4 MEUR 6,3 MEUR 6,0 MEUR 9,6 MEUR

Dust emissions abatement 20 MEUR 1,2 MEUR

9.3.1 Air pollution abate-ment

NOx emissions abatement Phase 1: 1 MEUR Phase 2 (Low-Nox-burners, by 2016): 26 MEUR

-

9.3.2 Water pollution Waste water treatment improvement In the order of 0,4 - Can not be esti-


10 APPENDICES

Appendix 1. Supporting documentation. The EA has been elaborated based on site observations, information obtained at meetings and during discussions and the following reports/documents:

through establishment of appropri-ate treatment installations. Sanitary waste waters are assumed to be led to municipal sewerage network.

0,6 MEUR.

mated at this stage abatement

Recycling of waste water Can not be esti-mated as exact plans and methods are not known at this time

-

Preparation of a waste management plan

25 000 - 35 000 EUR

-

Utilisation study for fly ash and bottom ash

Low -

Construction of a new ash disposal site

In the order of 3 - 5 MEUR in-cluding design, but not including land acquisition costs

-

Removal of all waste from the site in a sustainable way.

Low -

Handling and storage of hazardous waste according to regulations in bunded and roofed area.

Investment needs 10 000 - 20 000 EUR for repair of existing storage

9.3.3 Waste management

Development and implementation of an aftercare program for the ash disposal site

In the order of 10 - 12 MEUR includ-ing design (re-moval cost in the range of 50 MEUR)


HFO storage tanks and oil pipes inspection and repair if necessary

Low Low 9.3.4 Fuel storage

Inspection and repair if necessary of secondary retaining basin of the HFO tank

Low Low

Soil and groundwater contamina-tion surveys

70 000 - 80 000 EUR

- 9.3.5 Soil and groundwa-ter contamination

Implementation of a remediation project of contaminated soil and groundwater


-

9.3.6 Monitoring Elaboration of monitoring system for emissions into air and water.

0,2 - 0,4 MEUR 20 000 - 40 000 EUR

Development of an Environmental Management System

50 000 - 75 000 EUR

- 9.3.7 Environmental management and aware-ness Implementation of an Environ-

mental education and awareness campaign

30 000 - 40 000 EUR

-


1. Review of Safety & Environmental Issues in KEK. June 2004. ESBI Consultants Ltd.

2. Environmental Impact Assessment and Action Plan for Kosovo A and B Power Plants and Coal Mines. Environmental Audit. May 2003. Carl Bro Group.

3. Environmental Impact Assessment and Action Plan for Kosovo A and B Power Plants and Coal Mines. Environmental Impact Assessment. June 2003. Carl Bro Group.

4. Environmental Impact Assessment and Action Plan for Kosovo A and B Power Plants and Coal Mines. Environmental Action Plan. June 2003. Carl Bro Group.

5. Environmental Impact Assessment and Action Plan for Kosovo A and B Power Plants and Coal Mines. Environmental Action Plan. Logical Framework Analysis. June 2003. Carl Bro Group.

6. Environmental Impact Assessment and Action Plan for Kosovo A and B Power Plants and Coal Mines. Monitoring Plan. June 2003. Carl Bro Group.

7. Law on Environmental Protection (No. 2002/8). 8. Law on Water, draft. Ministry of Environment and Spatial Planning. 9. Law on Air Protection (No. 2004/30) 10. Waste Law, draft. Ministry of Environment and Spatial Planning. 11. Kosovo - State of The Environment Report. - UNMIK. Provosional

Institutions of self-government. April 26, 2003. http://enrin.grida.no/htmls/kosovo/Kosovo_SOE_part1.pdf


Appendix 2. Layout of Kosovo B.


Appendix 3. Aerial photo of Kosovo B.

Appendix 4. Technical report on SO2, NOx and dust emissions mitigation measures at Kosovo B TPP.

Appendix 5. Dispersion of exhaust gases from Kosovo B power plant in Obilic, Kos-ovo. Final report.

Appendix 6. Photo log.

60R05429.01-Q070-011

February 6, 2006

European Agency for Reconstruction

Contract nr 04KOS01/03/009

LOT 2

APPENDIX 4, FLUE GAS CLEANING

Draft final

Lot 2, Appendix 4 February 6, 2006 Flue gas cleaning Page 2 (50) Contents

Lot 2, Appendix 4 February 6, 2006 Flue gas cleaning Page 3 (50)

1 INTRODUCTION................................................................................................................ 4

2 KOSOVO B TPP .................................................................................................................. 4

2.1 Boiler plant............................................................................................................................. 4 2.2 Fuel......................................................................................................................................... 5 2.3 Electrostatic precipitators ..................................................................................................... 11 2.4 Plant efficiency..................................................................................................................... 13

3 EARLIER PLANT EMISSION MEASUREMENTS..................................................... 14

3.1 Precipitator performance measurements 1984 ..................................................................... 14 3.2 Precipitator performance measurements 2003 ..................................................................... 15 3.3 Emission measurements by INKOS 1994............................................................................ 15

4 EMISSION MEASUREMENTS BY ELEKTROWATT-EKONO............................... 17

4.1 Measurement Methodology.................................................................................................. 17 4.2 Measurement results............................................................................................................. 18 4.3 Results analysis .................................................................................................................... 19

5 SUMMARY AND RECOMMENDATIONS FOR MEASURES TO MEET LCP DIRECTIVE EMISSIONS................................................................................................................ 30

5.1 General ................................................................................................................................. 30 5.2 Plant efficiency..................................................................................................................... 31 5.3 Dust emissions...................................................................................................................... 31 5.4 SO2 emission ........................................................................................................................ 33 5.5 NOx emissions ...................................................................................................................... 44

6 ACTION PLAN.................................................................................................................. 48

7 COST SUMMARY............................................................................................................. 49


1 INTRODUCTION This draft report prepared by Elektrowatt Ekono (EE) contains the tasks included in Lot 2, “Prefeasibility study for pollution mitigation measures at Kosovo B TPP”, The target of the measures presented in this report is to bring the stack emissions from Kosovo B TPP to the level required by European Directive 2001/80/EC, “Limitations of emissions of certain pollutants into the air from large combustion plants”, referred as “LCP Directive” later in this report. The limit values to be considered for Kosovo B TPP are as follows: Pollutant Limit value Due date Dust mg/Nm3 dry 6 % O2 50 Jan 1, 2008 SO2 mg/Nm3 dry 6 % O2 400 Jan 1, 2008 NOx mg/Nm3 dry 6 % O2 500

200 Jan 1, 2008 Jan 1, 2016

2 KOSOVO B TPP

2.1 Boiler plant Kosovo B power plant consists of two lignite fired boilers B1 and B2. The main design data is presented in Table 1. In this report the plant heat rate and gross efficiency are referred as ratio of fuel thermal input kJth to generator gross electrical output kWhe:

Table 1, Kosovo B1 and B2 main data

Kosovo B power plant Boiler manufacturer Stein Industrie Turbine Manufacturer MAN Generator manufacturer Alstom Fuel Lignite Nominal rated Power MWe 339

Steam flow t/h 1 000 Steam temperature oC 543 Steam pressure Bar 186 Lignite feed t/h 385 t/MWhe 1,13 Plant heat rate kJth/kWhe 8 990 Plant gross efficiency MWe gross/MWth % 40,0 Minimum load without fuel oil MWe 189 Max load ramp rate MWe/min 5 Number of cold starts/year Forecast 10 Number of hot trips /year Forecast 10 Base load operating hours H/year 8 200 Lignite: design ash content % AR 12-18 Lignite: design water content % AR 42-50 Lignite: design lower heating value kJ/kg 7940 Year of commissioning 1983-84


The boilers are fired tangentially with eight vertical burner registers. Fuel to each burner register is supplied by one of the eight beater wheel type lignite pulverizing mills, which are located symmetrically around the boiler. The lignite is dried in the mills with hot flue gases withdrawn from the top of the furnace and mixed with hot combustion air recirculation flue gases. The dried and pulverized lignite is taken through a mechanical classifier to the burners carried by drying gases and evaporated water vapor. See Enclosure 1.

Both boilers have two separate flue gas lines after the combustion chamber and have two rotary Air Preheaters, two Electrostatic Precipitators and two ID fans as well as two FD fans, see Enclosure 2.

The bottom ash is conveyed to a silo with mechanical conveying system. The flyash from the bottom of the precipitators is collected to an intermediate bin with air slide conveyor system and further with pneumatic conveying system to two fly ash silos. From the silos the flyash is conveyed with hydraulic conveying system the final deposit site. Flue gases are discharged to the atmosphere through common 182 m high concrete stack. The plant electrical and C/I systems as well as control room equipment have been completely rebuilt after fire damage in 2003. The electrostatic precipitators on boiler B2 have been rebuilt on 2003. Boiler B1 precipitators will be rebuilt on 2006.

The boilers have approx. 90.000h operating hours and are expected to be in operation for further 15 years at least.

The reported reliability of Kosovo B TPP has been reported to be over 90 % after boiler overhauls in 2001 and Fire Damage Repairs in 2003.

2.2 Fuel Lignite fuel for Kosovo B TPP is received mainly from the nearby Bardhi mine while lignite for Kosovo A is supplied from Mirashi mine. Only occasionally when some problems are encountered, lignite from Mirashi mine is fired at Kosovo B. Lignite from Bardhi mine is slightly better in quality than from Mirashi mine. Lignite is transported from the mine to the power plant storage yard with two belt conveyors OPB 02/03 with transport capacity of 1 400 t/h (33 000 t/day) per conveyor. The daily lignite supply to Kosovo B TPP is normally between 5 000-20 000t/day while daily consumption is max 20 000 t/d with full capacity of two boilers. The lignite is stored at the power plant in open yard with storage capacity of approx. 360 000 t, i.e. 18 days at maximum daily consumption. No sorting, classification or homogenization of lignite is done either at mine or at power plant lignite yard. Typically the plant is by-passing the fuel yard and takes the lignite straight from the mine to the boilers.


Samples from the lignite are taken daily and analyzed at KEK/Inkos laboratory for ash and water contents. Based on these values the lower calorific value is calculated as follows: LHV = (100 – Ash % – H2O %) x 314 – 4369 kJ/kg

Main characteristics of the lignite during 2002-2004 are presented in Figures 1 and 2. The data is based on the daily analysis results carried out by Inkos.

Figure 1, Lignite daily sample analysis for LHV during 2002-2004

Lower heating value and weight-% of combustible matter

0

2 000

4 000

6 000

8 000

10 000

12 000

Jan-

02Fe

b-02

Mar

-02

Apr-

02M

ay-0

2Ju

n-02

Jul-0

2Au

g-02

Sep-

02O

ct-0

2N

ov-0

2D

ec-0

2Ja

n-03

Feb-

03M

ar-0

3Ap

r-03

May

-03

Jun-

03Ju

l-03

Aug-

03Se

p-03

Oct

-03

Nov

-03

Dec

-03

Jan-

04Fe

b-04

Mar

-04

Apr-

04M

ay-0

4Ju

n-04

Jul-0

4Au

g-04

Sep-

04O

ct-0

4N

ov-0

4D

ec-0

4

Low

er h

eatin

g va

lue

kJ/

kg

0

10

20

30

40

50

60

Com

bust

ible

% A

R LHV Max

LHV Average

LHV Min

Comb max

Comb average

Comb min

2002 2003 2004


Figure 2, Lignite daily sample analysis of ash and water contents during 2002-2004

Ash and water content

0

10

20

30

40

50

60Ja

n-02

Feb-

02M

ar-0

2A

pr-0

2M

ay-0

2Ju

n-02

Jul-0

2A

ug-0

2S

ep-0

2O

ct-0

2N

ov-0

2D

ec-0

2Ja

n-03

Feb-

03M

ar-0

3A

pr-0

3M

ay-0

3Ju

n-03

Jul-0

3A

ug-0

3S

ep-0

3O

ct-0

3N

ov-0

3D

ec-0

3Ja

n-04

Feb-

04M

ar-0

4A

pr-0

4M

ay-0

4Ju

n-04

Jul-0

4A

ug-0

4S

ep-0

4O

ct-0

4N

ov-0

4D

ec-0

4

Wei

ght -

% A

R H2O max

H2OaverageH2O min

Ash max

Ash average

Ash min

2002 2003 2004

The average data as well as minimum and maximum values of each individual parameter of daily lignite samples are presented in Table 2.

Table 2, Lignite daily sample results 2002-2004 as % AR

Year H2O Ash Combustible LHV kJ/kg2 002 44,9 15,4 39,8 8 092 2 003 43,7 18,3 37,9 7 518 2 004 43,9 17,6 38,3 7 706 Average 44,1 17,3 38,5 7 728 Min 35,7 8,2 25,2 3 509 Max 50,7 32,3 50,4 11 427

There are rather large variations in the fuel quality, especially during summer 2003. During 2004 the lignite quality has been more stable.

In addition to the above daily samples, also one monthly composite sample has been analyzed as presented in Table 3. In the table the annual average values as well as monthly min/max values of each individual parameter are presented.


Table 3, Lignite monthly composite sample results2002-2004 as % AR

Year H2O Ash TotalC

Fixed C

Volatile

Combustible

LHV kJ/kg

Total S

Inorg S

Org S

Stot/ Sorg

2 002 avg 43,8 16,6 30,6 14,0 25,7 39,6 8 044 1,08 0,69 0,39 3,4 2 003 avg 42,9 19,2 31,9 12,6 25,2 37,8 7 478 0,89 0,62 0,28 3,7 2 004 avg 43,6 18,1 31,4 13,4 25,0 38,4 7 649 0,91 0,71 0,20 4,7 2002-2004 Average 43,3 18,3 31,4 13,2 25,3 38,4 7 660 0,94 0,66 0,27 4,0

Monthly Min 39,6 14,7 28,5 10,0 21,8 34,5 6 446 0,70 0,41 0,12 2,0

Monthly Max 46,5 24,4 35,0 16,2 28,0 41,1 8 509 1,43 0,90 0,58 7,3

There is much larger variation of data between the results presented in Tables 2 and 3, especially in LHV value. The data in Table 2 is analyzed from daily lignite samples and LHV value is calculated as presented above. This data represents the actual daily variations in fuel quality. The possible irregularities in taking and analyzing the daily fuel samples increase naturally the variation range of the data. In Table 3 the data is obtained from monthly composite samples, which are prepared from the daily samples. This data gives better picture of the average fuel quality. The actual performance and emissions of the boiler plant depend strongly on the actual momentary fuel quality and variations in fuel characteristics.

In the analysis carried out by Inkos the lignite sulphur content is determined as organic and inorganic sulphur. According to the available information the organic sulphur content is determined in the calorimeter in connection with measuring the LHV. This is the portion of sulphur burning to SOx in the calorimeter conditions and is also assumed to be producing SO2 emissions after boiler. The inorganic sulphur content is determined from the fuel ash and represents the portion of sulphur bound as sulphates and other stable products in the ash and thus not producing SO2 emissions. These values and proportion between organic and inorganic sulphur are presented in Table 3 and in Figure 3.


Figure 3, Lignite monthly composite sample analysis for sulphur content

Lignite Sulphur content

0,0

0,2

0,4

0,6

0,8

1,0

1,2

1,4

1,6

1,8

2,0Ja

n-02

Feb-

02M

ar-0

2A

pr-0

2M

ay-0

2Ju

n-02

Jul-0

2A

ug-0

2S

ep-0

2O

ct-0

2N

ov-0

2D

ec-0

2Ja

n-03

Feb-

03M

ar-0

3A

pr-0

3M

ay-0

3Ju

n-03

Jul-0

3A

ug-0

3S

ep-0

3O

ct-0

3N

ov-0

3D

ec-0

3Ja

n-04

Feb-

04M

ar-0

4A

pr-0

4M

ay-0

4Ju

n-04

Jul-0

4A

ug-0

4S

ep-0

4O

ct-0

4N

ov-0

4D

ec-0

4

Lign

ite S

-con

tent

% A

R

-12,0

-10,0

-8,0

-6,0

-4,0

-2,0

0,0

2,0

4,0

6,0

8,0

S-tot

S-inorg

S-org

Stot/Sorg

2002 2003 2004

Stot

/Sor

gLi

gnite

Sco

nten

t % D

B

2,0

0,0

1,5

1,0

0,5

As seen the proportion between organic and inorganic sulphur varies in rather big range from 2,0 to 7,3 being as average of 4,0.

From the monthly composite samples the analysis of fuel ash has also been carried out by Inkos. The results are presented in Table 4.

Table 4, Lignite monthly composite sample ash analysis 2002-2004 as % DB

Year SiO2 Al2O3 Fe2O3 CaO MgO SO3 Ca/Sorg Ca/Stot

2 002 28,7 11,4 7,5 37,5 4,9 9,2 10,9 2,9 2 003 33,0 14,0 7,3 32,5 3,7 7,7 12,5 3,4 2 004 32,3 12,5 7,2 34,6 3,7 7,5 16,0 3,4 Average 31,8 12,9 7,3 34,3 4,0 8,0 13,4 3,3 Min 25,0 9,0 6,0 27,6 2,6 6,2 4,3 1,8 Max 37,0 16,7 10,1 42,0 5,8 10,8 28,0 4,5

The average SO3 content of the fuel ash is 8,0 %. This value represents the inorganic portion of sulphur. It corresponds to a value of 0,58 % as S in fuel. This value can be compared to average analyzed inorganic S-content of 0,66 % in fuel shown in Table 3. The difference can be assumed to reflect the problemacy of determining the organic and inorganic S-contents in the fuel and related combustion and analyzing temperature and O2 content. Te variation in SO3 content is of the same order of magnitude as in inorganic S content. The average content of CaO in the fuel ash is 34,3 % and MgO 4,0 %. It can be assumed that Mg is present as dolomite CaMg(CO3)2 therefore the binding portion of total CaO, which does not participate in the sulphur capture (5,6 % -units, equals to 16 % of total


Ca). If the rest 28,7 % of CaO is present as CaCO3 the corresponding free CaCO3 content is 51,2 % from fuel ash. The fuel available Ca/S ratio can be calculated as ratio of total available Ca (total Ca deducted with dolomite bound Ca) vs. fuel sulphur content. This value presents the amount of calcium contained in the fuel available for sulphur removal in the boiler provided that all CaCO3 is calcinated to CaO. In Figure 4 the Ca/S ratio is presented as calculated based on both organic and total sulphur contents of lignite. Normally fuel total sulphur content is used.

Figure 4, Calculated lignite Ca/S ratio

Ca/S ratio

0

10

20

30

40

Jan-

02Fe

b-02

Mar

-02

Apr

-02

May

-02

Jun-

02Ju

l-02

Aug

-02

Sep

-02

Oct

-02

Nov

-02

Dec

-02

Jan-

03Fe

b-03

Mar

-03

Apr

-03

May

-03

Jun-

03Ju

l-03

Aug

-03

Sep

-03

Oct

-03

Nov

-03

Dec

-03

Jan-

04Fe

b-04

Mar

-04

Apr

-04

May

-04

Jun-

04Ju

l-04

Aug

-04

Sep

-04

Oct

-04

Nov

-04

Dec

-04

Ca/

S-ra

tio

0

5

10

15

20

25

30

35

40

45

CaO

con

tent

% D

B

Ca/Sorg

Ca/Stot

CaO

2002 2003 2004

The average Ca/Sorg ratio during 2002-2004 has been 13,4. There is, however, a very big variation in this value, minimum of 4,3 and maximum of 28. The average Ca/Stot ratio during 2002-2004 has been 3,3. There is also a big variation in this value, minimum of 1,8 and maximum of 5,5. This means that also the natural sulphur capture in the boiler can vary depending of the quality of fired lignite and actual boiler combustion conditions.

The method of analyzing the fuel sulphur content and SOx producing portion as well as Ca/S ratio is further discussed in Chapter 4.3.5.


2.3 Electrostatic precipitators Both boilers B1 and B2 are equipped with two electrostatic precipitators (ESP) for dust removal. Each ESP handles therefore 50 % of the total flue gas flow of one boiler.

The ESP’s are originally delivered by Lurgi. On 2003 both ESP’s on boiler B2 were rebuilt by polish company ELWO. All the internal parts, collecting and emitting systems were completely replaced with new parts and the damaged or corroded parts of the flue gas ducting and ESP casings were repaired. Also the high voltage transformer control system was replaced with modern thyristor controller, supplied by Merlin Gerin. The old transformers remained in use.

A similar rebuilding project for boiler B1 precipitators in planned to be carried out in 2006.

The design basics of the original ESP’s is shown in Table 5.

Table 5, Electrostatic precipitator technical data

Kosovo B1 and B2 electrostatic precipitators Flue gas flow, total Nm3/h 1 910 000 Flue gas flow/ESP Nm3/h 955 000 Flue gas flow/ESP m3/s 444 Temperature oC 170 Static pressure Pa - 3 000 Inlet dust content g/Nm3 30 Outlet dust content mg/Nm3 260 Efficiency % 99,14 Number of electrical fields in direction of gas flow 2 Number of electrical parallel fields 2 Total number of electrical fields/ESP 4 Effective length of each field m 4,8 Effective height of each field m 13,0 Number of gas passages/field 34 Number of gas passages/ESP 2x34=68 Width of gas passage mm 300 Effective width of each field m 10,2 Effective width of each ESP m 20,4 Effective collecting plate area/field m2 4 243 Effective collecting plate area/ESP m2 16 972 Gas velocity m/s 1,67 Specific collecting plate area m2/m3/s 38,2 wk factor by Andersson formula 59,2 w factor by Deutsch formula m/s 12,5

In some documents an effective electrode height of 13,75 m is given. With this value the calculated effective collecting area would be 17 952 m2, specific collecting plate area 40,4 m2/m3/s, gas velocity 1,58 m/s and wk factor 56,0. Also in some documents the design gas


velocity of 1,8 m/s is specified. This velocity will be valid in ID fan design Point 2 with total gas flow of 2 024 000 Nm3/h.

The collecting plate area of 19 448 m2 given in design document is probably the total plate area including the profile additions which cannot be calculated to be effective for electrical collection activity.

For comparison boiler flue gas fan design operating points are shown in Table 6. There are two 50 % ID fans per boiler, one for each ESP line.

Table 6, ID fan design data

Kosovo B1 and B2 ID fans Point 1 Point 2 Design Flue gas flow/fan m3/s 452 468,3 542,5 Flue gas flow/fan Nm3/h 977.000 1.012.000 1.172.000 Flue gas flow/boiler Nm3/h 1 954 000 2 024 000 2 344 000 Total pressure rise Pa 3580 3780 4630 Density kg/m3 0,75 75 75 Temperature oC 165 165 165 Power consumption kW 1990 2150 3200 Efficiency % 81,5 82,5 79,0 Speed rpm 600 600 600 Motor power kW 3020

The ESP plant has been sized for rather low inlet dust concentration of 30 g/Nm3 and rather high outlet emission of 260 g/Nm3, giving collection efficiency of 99,14 %. The design gas velocity in the ESP, dictated by the physical size if the filter casing, is 1,7 m/s and is clearly higher than velocity of 1,2…1,3 m/s commonly used in modern high efficiency ESP’s, The high velocity increases the possibility of high dust emissions during rapping of collecting electrodes (rapping losses) and in occasion, when the inlet dust content is increased, like during soot blowing and during operation with lignite with high ash content. Higher than the design flue gas flow due to high O2 content, higher specific lignite consumption (lower plant efficiency) or poor lignite quality have also the same effect. The design inlet dust concentration of 30 g/Nm3 is also low and corresponds to lignite ash content of approx. 14 %. In Table 7 the effect of ash content to the expected performance of existing ESP with the design gas flow and design wk is shown. The increase of wk with increasing inlet dust content is taken into account as shown. The amount of bottom ash is assumed to be 3,5 % as measured during the tests in October 2005. The average ash content of monthly composite samples during 2002…2004 was 18,3 % and maximum 24,4. The maximum ash content of daily lignite samples during the same period was 32,3 % as shown in Tables 2 and 3.


Table 7, Effect of lignite ash content to dust emission with existing ESP

Lignite ash content % AR Design14,0 18,3 24,4 32,3

Inlet dust content g/Nm3 wet 30 40 60 80

Inlet dust content g/Nm3 dry 6 % O2 38 50 75 95

Outlet dust content mg/Nm3 wet 260 290 360 390

Outlet dust content mg/Nm3 dry 6 % O2 330 370 460 500 Specific collecting plate area m2/m3/s 38,2 38,2 38,2 38,2 wk factor 59,2 62,7 68,0 72,0 Efficiency % 99,14 99,25 99,39 99,47

The expected stack emission is in range of 330…500 mg/Nm3 dry 6 % O2 with existing lignite quality and analysis variations. The size of the ESP, i.e. specific collecting plate vs. design efficiency i.e. wk factor is in line with common sizing rules with the used design data and design efficiency. The performance of ESP can be judged based on wk factor (migration velocity). This factor is process and dust specific and describes the collectability of flyash. In this report wk factor is calculated based on Andersson formula as follows:

Q100Awk

1 ××

−

=− eη

η ollection efficiency/100 k igration velocity

A Collecting plate area, m2 Q Gas flow m3/s

2.4 Plant efficiency The design nominal rated power of the boiler unit is 339 MWe as gross electrical generator output. The design fuel consumption is 1,13 t/MWhe giving the plant heat rate of 8990 kJth/kWhe and gross efficiency of 40,0 % calculated as MWe gross/MWth. This performance has been demonstrated in delivery tests in 1884. In the present conditions the maximum obtained power is 285 MWe. The indicated fuel consumption is in range of 1,3 - 1,5 t/MWhe giving the plant heat rate of 10 000 - 12 000 kJth/kWhe and gross efficiency of 30 - 35 %. The following main reasons for low unit efficiency can be pointed out: Lowered turbine efficiency Lowered feed water quality By-Passing of HP Preheaters Increased flue gas outlet temperature, 170 – 180 oC High flue gas O2 content Slagging of boiler heat surfaces


Continuous soot blowing and high steam consumption Coarse lignite grinding result after mills. Original design is 50 % > 90 µm and max 5 % >1mm, actual performance is 70 % > 90 µm and 5-10 % > 1 mm High fly ash unburnt carbon content, UBS, in range of 6 % It shall be noted that the specific plant emissions of CO2, SOx, NOx and Dust calculated as mass flow in t/MWh are directly proportional to plant fuel consumption, i.e. heat rate and efficiency. The total pollutant mass flow is determining the effect of stack emissions to the ambient air quality i.e. immision values as ground level concentrations and environmental impact on soil by the pollutant deposits as well as long range pollutant transportation. Therefore it is of utmost importance to bring the boiler efficiency to the range of the design value.

3 EARLIER PLANT EMISSION MEASUREMENTS

3.1 Precipitator performance measurements 1984 In 1984 the ESP on boiler B1 was measured during commissioning tests. The results are shown in Table 8.

Table 8, Kosovo B1 ESP measurements 1984

Kosovo B1 electrostatic precipitator Boiler load MWe 340 Lignite feed t/h 381 t/MWhe 1,12 Boiler heat rate kJth/kWhe 8893 Plant gross efficiency % 40,5 Lignite H2O content % 45,7 Lignite ash content % 15,7 Lignite LHV kJ/kg 7940 Flue gas flow, total Nm3/h dry 1 236 000 Approx. Nm3/h wet 1 600 000 Flue gas flow/ESP Nm3/h 800 000 Flue gas flow/ESP m3/s 350 Temperature oC 155 O2 content % 5,1 Inlet dust load g/Nm3 46,6 Outlet dust load mg/Nm3 143 wk factor 68,8 Efficiency % 99,69

The indicated flue gas flow is almost 20 % lower than the design gas flow. During these tests the plant was operating close to or slightly better than the contractual conditions in terms of lignite consumption and plant heat rate. In these tests the electrostatic precipitators were performing slightly better than the design value based on measured wk factor 68,8 vs. design value of 59,2.

Lot 2, Appendix 4 February 6, 2006 Flue gas cleaning Page 15 (50) 3.2 Precipitator performance measurements 2003

In 2003 the ESP on boiler B2 was measured after the rebuilding project. The results are shown in Table 9.

Table 9, Kosovo B2 ESP measurements 2003 after rebuilding

Kosovo B2 Esp 1 Esp 2 Boiler total Boiler load MWe 250 Flue gas flow Nm3/h 1 051 000 882 000 1 933 000 m3/s 476 398 Temperature oC 172 170 O2 content % 7,5 7,0 Outlet dust load mg/Nm3 210 200 wk factor 78,7 65,8 Efficiency approx. % 99,5 99,5

No fuel consumption or fuel analysis data is available from these tests.

During these tests the measured flue gas O2 content is clearly higher than in test in 1984. Also the flue gas flow is high compared to boiler load even considering the high O2 content of 7,0…7,5 % (assumed as dry) at the ESP outlet. The measured flue gas flow refers to very high lignite consumption in range of 1,7 kg/MWhe and boiler heat rate over 13 000 kJth/MWhe corresponding to total plant gross efficiency MWe/MWth of clearly below 30 %,

In these tests the performance of electrostatic precipitators was slightly better compared to design value based on measured wk factor.

3.3 Emission measurements by INKOS 1994 The following data shown in Table 10 is available from Kosovo B2 boiler emission measurements carried out by INKOS in 1994 .


Table 10, Kosovo B2 emission measurements 1994 by INKOS

Kosovo B2 Boiler load MW 285 Lignite feed t/h 429 t/h/MWe 1,51 Boiler heat rate kJth/kWhe 12 300 Total gross efficiency % 29,2 Lignite H2O content % 48 Lignite ash content % 12,0 Organic S-content, Sd % 0,3 Total S-content, St % 0,9 Lignite LHV calculated as KEK kJ/kg 8 190 Flue gas flow Nm3/h 1 740 000 Excess air ratio 1,82 O2 content after ESP, calculated % dry 9,5 SO2 content, assumed act O2 mg/m3 dry 626 SO2 content based on above mg/Nm3 dry 6 % O2 816 NOx content, assumed act O2 mg/m3 dry 810 NOx content based on above mg/Nm3 dry 6 % O2 1056 CO content, assumed act O2 mg/m3 dry 106 Outlet dust content mg/m3 dry 280 Dust content based on above mg/Nm3 dry 6 % O2 365 ESP efficiency % 99,45 wk factor 61,1 Inlet dust content, calculated from efficiency g/Nm3 dry 6 % O2 66,4

Inlet dust content, calculated from lignite ash content g/Nm3 dry 6 % O2 34,2

In the measurement report it is not clearly stated, if the measured values are expressed as mg/m3 or mg/Nm3 or have been corrected to 6 % O2, Also in conversions from kg/h to concentrations there are uncertainties. The flue gas quantity is not in accordance with lignite feed rate and excess air ratio (correct value would be in range of 2 100 000 Nm3/h, 1 740 000 Nm3/h would refer to O2 content of 6 % dry))

During these tests the ESP performance has been in accordance with design values in terms of wk factor assumed that the indicated flue gas flow is correct. The low ash content of lignite would indicate lower inlet dust load, in range of 34 g/Nm3 dry 6 % O2, which would also give lower efficiency and lower wk factor. The indicated efficiency refers to inlet dust concentration of 66,4 g/Nm3 dry 6 % O2 , which corresponds to lignite ash content of 19,5 % AR. The lignite consumption during the tests was high, 1,51 kg/MWe giving boiler heat rate of 12300 kJth/MWhe and plant gross efficiency of 29,2 %.


Based on the organic S-content of lignite 0,3 % AR, the theoretical SO2 concentration without any sulphur capture in the boiler would be 1860 mg/Nm3 dry 6 % O2 equals to 1460 mg/Nm3 dry at O2 9,5 % dry. This would indicate that the boiler sulphur capture is in range of 60 %.

4 EMISSION MEASUREMENTS BY ELEKTROWATT-EKONO

Elektrowatt-Ekono carried out emissions measurements on Kosovo B1 boiler on October 18, 2005. Both ESP lines 1 and 2 were measured. Boiler B2 could not be measured due to a failure of the slag conveyor, which resulted in an immediate forced shutdown of 10 days for repairs.

During the tests samples of lignite, flyash and bottom ash were collected by KEK personnel and analyzed in Finland by certified laboratory. The boiler plant operating data was obtained from boiler data collection system.

4.1 Measurement Methodology The methods used in the measurements are explained in the following Table 11 Table 11, Summary of equipment and standards

Compound Standard or method

Analyzer / method

Temperature and flow of stack gas

ISO 9096:1992 L-pitot tube method for flow determination Calibrated K-type thermoelement

Moisture content

DIN 1942:1994 Calculation from the combustion balance

Particulates ISO 9096:1992 SF dust sampling system, out-stack method

O2 EPA 3A Horiba PG-250/paramagnetic

NOx EPA 7E Horiba PG-250/chemiluminesence

SO2 EPA 6C Horiba PG-250/NDIR (non-dispersive infrared)

CO EPA 10 Horiba PG-250/NDIR (non-dispersive infrared)


4.2 Measurement results The results of EEO emission measurements are shown in Table12.

Table 12, Kosovo B1 emission measurement results Oct 2005 by EEO

Unit 1 plant operating data Line 1 Line 2 Power output MWe 283 283 Steam flow t/h 803 803 Steam temperature oC 539 540 Steam pressure bar 157 158 Lignite feed t/h 391 397 t/MWhe 1,38 1,40 Plant heat rate kJth/kWhe 10 790 10 950 Plant gross efficiency MWth/MWe

% 33,4 32,9

Thermal load MWth 848 861 No of mills in use 6 6 Opacity line mg/Nm3 790 301 Flue gas temperature oC 154 159 O2 before AP % wet 5,2 4,4 Unit 1 measured data Line 1 Line 2 Flue gas Temperature oC 161 153 Static pressure Pa -3 430 - 3760 Flow Nm3/s, wet 256 253 m3/s 421 410 Nm3/h wet 921 600 910 00 Flow total for boiler Nm3/h wet 1 852 000 Calculated from fuel data Nm3/h wet 1 883 000 O2 % dry 8,6 9,5 % wet 7,0 7,8 H2O v-% 17 16 Emissions Particulates mg/Nm3, wet 396 339 Particulates mg/Nm3, dry 477 403 Particulates (6 % O2) mg/Nm3dry 6% O2 577 526 SO2 ppm, dry 40 87 SO2 mg/Nm3, dry 117 255 SO2 (6 % O2) mg/Nm3dry 6% O2 142 332 NOx ppm, dry 266 266 NOx as NO2 mg/Nm3, dry 546 546 NOx as NO2 (6 % O2) mg/Nm3dry 6% O2 661 713 CO ppm 48 69 CO mg/Nm3, dry 60 87 CO (6 % O2) mg/Nm3dry 6% O2 73 113

Lot 2, Appendix 4 February 6, 2006 Flue gas cleaning Page 19 (50) 4.3 Results analysis

Operating conditions During the tests the boiler was run in rather normal present full load conditions 283 MW and with lignite feed of 1,4 t/MWhe. Based on available data the present max load of the plant is approx 285 MW and the lignite feed in range of 1,4…1,5 t/MWhe which corresponds to heat rate of 11 000…12 000 kJth/kWhe and gross efficiency of 30…33 %. This is clearly lower than the original plant design value. The boiler was operated with 6 mills during the tests. The measured UBC content of flyash was 0,9…3,1 see results in Tables 17 and 18. According to the Inkos measurements the average UBC value is 6 % and maximum 7…8 %. The measured UBC content of bottom ash 10,7…17,5 % see Tables 16 and 17. A mass flow of 2,3 t/h corresponding to 3,5 % of total ash flow was measured by KEK during the tests.

Fuel The results of the fuel analysis are in the Table 13. INKOS results of ash and moisture contents and LHV are also shown in the same Table. The complete independent fuel analysis result are shown in Enclosure 3.

Table 13, Analysis of lignite used during EEO measurements October 2005

Lignite EE 18.10. -05

INKOS 18.10. -05

2002-2004 average

Contract tests 15.12.1984

H2O content % 45,9 43,4 43,3 45,7 Ash content % AR 16,7 20,8 18,3 15,7 Combustible % AR 37,4 35,8 38,4 38,76 LHV MJ/kg 7806 6845 7660 7940 Volatile % AR 26,2 25,3 Fixed carbon % AR 11,2 13,2 Fuel Ratio FR 0,42 0,52 C % AR 23,9 23,2 H % AR 1,57 1,78 N % AR 0,4 0,57 O % AR 10,6 12,0 Cl % AR < 001 S-org % AR 0,48* 0,27 0,60 S-inorg % AR 0,38* 0,66 S-total % AR 0,86 0,94 Lignite ash SiO2 % DB 24,0 31,8 Al2O3 % DB 8,9 12,9 Fe2O3 % DB 8,4 7,3 CaO % DB 31,0 34,3 MgO % DB 3,2 4,0 SO3 % DB 8,8 8,0 K2O % DB 0,52


MnO2 % DB 0,17 Na2O % DB 1,0 P2O5 % DB 0,18 TiO2 0,7 Ca/S-org 5,3 13,4 As mg/kg DB 46 Ba mg/kg DB 650 Be mg/kg DB < 2,5 B mg/kg DB 590 Cd mg/kg DB < 10 Co mg/kg DB 12 Cr mg/kg DB 150 Cu mg/kg DB 50 Hg mg/kg DB < 0,045 Mo mg/kg DB 7,2 Ni mg/kg DB 190 Pb mg/kg DB 15 Sb mg/kg DB 2,8 Sn mg/kg DB < 2,5 V mg/kg DB 120 Zn mg/kg DB 110 S mg/kg DB 35 000 % DB 3,5 SO3 % DB 8,8 As inorg S in lignite

% DB 1,08

% AR 0,58 * see explanations below

The data for 2002-2004 average values is taken from monthly composite samples shown in Table 3.

There is rather big difference in ash content and calorific value between the external laboratories and INKOS results on the same samples. The INKOS data is taken from Enclosure 4, positions OPB02/03. The high ash content explains the low LHV, which is calculated based on ash and water contents. The external analysis data is used on later calculations. As overall the fuel used in these tests was in rather good conformity with the average fuel used at the plant during 2002-2004. The lignite sulphur content shown in the table and the proportion of organic and inorganic sulphur has to be elucidated more in details as follows.

In the external laboratory analysis the lignite sulphur content is analyzed according to ASTM-D 4239 C. First the raw lignite sample is dried at 35 oC for 16 hours to determine the surface moisture followed by drying at 105 oC until no weight change is measured. Based on these the total moisture is determined. Then the dried lignite sample is burned in induction oven in normal ambient atmosphere at 1360 oC and the released sulphur is measured with LTRA analyzer, which is very similar with a better known LECO analyzer.


The resulting sulphur content of 0,86 % AR (1,6 % DB) is therefore the amount of total sulphur emitted in these analyzing conditions, i.e. oxygen of 21 % and temperature of 1360 oC.

The lignite sulphur content was also analyzed with the same method by burning the dried lignite sample in lower temperatures. As shown in Figure 5 the amount of released sulphur depends of the combustion temperature.

Figure 5, Analyzed lignite S-content as function of combustion temperature

Lignite S content as function of combustion temperature

0,00

0,20

0,40

0,60

0,80

1,00

1,20

1,40

1,60

800 900 1000 1100 1200 1300 1400

Combustion temperature oC

Lign

ite S

con

tent

%

AR

DB

The lignite ash sulphur content (inorganic S) was measured according to EN13656 by first burning the dried lignite sample at 850 oC for 16 hours. The remaining fuel ash with was then analyzed as above by combusting the ash sample at the same temperature of 1360 oC but using ferrosulpate as accelerator to release all the sulphur. The resulting ash S-content was 35 000 mg S/kg fuel DB i.e. 3,5 % DB in ash corresponding to 0,58 % AR (1,08 % DB) sulphur in lignite. Furthermore the fuel ash was burned in 1650 oC with same catalyst giving ash S-content result of 4,1 % DB in ash corresponding to 0,68 % AR (1,25 % DB) in fuel. As seen, the sulphur release from lignite is strongly dependent on its combustion temperature. In addition combustion zone O2 content and stochiometric conditions in different parts of flame can effect to the amount of emitted SO2. Therefore the analysis results have to be interpreted carefully keeping in mind the actual conditions in the boiler. If the ashing temperature of 850 oC, as used in fuel ash analysis is used as basis for lignite organic sulphur content (compare Inkos procedure), it can be concluded as shown in Table 14.


Table 14, Lignite sulphur analysis results, bold numbers analyzed

Analysis T oC Organic S % AR Inorganic S % AR Total S % AR 850 0,48 0,48 1360 0,48 0,38 0,86 1360 + CAT 0,48 0,58 1,06 1650 + CAT 0,48 0,68 1,10

• Organic sulphur content of lignite is 0,48 % AR (0,89 % DB) based on Figure 5 • Total sulphur content as analyzed in 1360 oC without catalyst is 0,86 % AR (1,6 %

DB) giving inorganic sulphur to be 0,38 % (0,71 DB) • Inorganic sulphur content as analyzed in 1360 oC with catalyst is 0,58 % AR (1,08 %

DB) giving total sulphur of 1,06 % AR (1,97 % DB) • Inorganic sulphur content as analyzed in 1650 oC with catalyst is 0,68 % AR (1,25 %

DB) giving total sulphur of 1,16 % AR (2,14 DB)

The use of catalyst in the analyzing method increases inorganic sulphur content result by approx, 0,2 % AR and increased analyzing temperature additional 0,1 % AR. Later in this report lignite organic S-content of 0,48 % AR, inorganic S content of 0,38 % and total-content of 0,86 % AR has been used.

Flue gas flow and O2 content The measured flue gas flow of 1 852 000 Nm3/h corresponds very well with value of 1 883 000 Nm3/h calculated from fuel data. Flue gas O2 content after ESP is very high 8,6…9,5 % dry corresponding to excess air ratio of 1,76. Flue gas O2 contents before and after air preheaters were measured on 19.10.2005 with results shown in Table 15.

Table 15, O2 content measurements

Unit B1 Unit B2 Line 1 Line 2 Line 1 Line 2 Plant measurement % wet 4,3 5,0 Broken 7,4 % dry 5,2 6,1 9,0 Before LUVO % dry 5,8 6,0 5,9 6,1 After LUVO % dry 7,0 7,8 7,0 7,4 O2 increase in LUVO % dry 1,2 1,8 1,1 1,3 LUVO gas leakage % 9 14 8 10

The plant O2 measurements on line 1 were indicating rather correct values. On Unit B2 one measurement was not in service and another showed 2,9% dry, a too high value. Both units were running at high excess air ratio. O2 increase in air preheaters were1,1…1,8 % dry corresponding to air inleakage to flue gas of 8…14 %. The other values are rather normal, 14 % is a little high value. The total O2 increase from the boiler to ID fan inlet was in average of 2,7 % dry corresponding to total air inleakage of 20 % out of which 8…14 % is coming from


LUVO and the rest is coming from the flue gas ducting and ESP. One part of this is the flyash transport air, which is introduced to flue gas duct just in front of the ESP.

Dust content and ESP performance The operation of ESP’s during the tests is presented in Table16.

Table 16, ESP performance

Unit 1 ESP performance Line 1 Line 2 Flue gas flow m3/s 421 410 O2 content % wet 8,6 9,5 % dry 7,0 7,8 H2O content 17 16 Gas velocity m/s 1,59 1,55 Effective collecting plate area/ESP m2 16 972 16 972 Specific collecting plate area m2/m3/s 40,3 41,4 Lignite ash content % AR 16,7 167 Inlet dust load, calculated g/Nm3 wet 35,0 32,9 g/Nm3 6 % O2 50,0 50,0 Plant opacity measurement mg/Nm3 790 301 Outlet dust content mg/Nm3 wet 396 339 mg/Nm3 6 % O2 577 526 ESP efficiency % 98,4 98,95 wk factor 49,4 50,1

During the test both ESP’s were operating with lower than their design performance based on wk factor. The gas flow was slightly lower than the design value. The plant opacity meter on Line 2 shows near the measured dust content in units mg/Nm3 wet. On line 1 the opacity meter showed too high values, see Table 12. In the internal inspection of the system on September 28, 2005 the opacity meter lenses were found to be dirty on line 2, and the lenses were cleaned. Line 1 the opacity meter was not inspected. The opacity meters require regular cleaning. If the ESP had been operating with desig wk, the outlet dust emission would have been 360…380 mf/Nm3 dry 6 % O2.

The outlet dust content of both ESP’s was very uneven on different sides of the outlet duct as shown in Table 17. The dust concentration was measured in three horizontal cross-sections across the whole outlet duct.

Table 17, Dust emission distribution across the outlet duct

Unit 1 ESP performance Line 1 Line 2 Left Center Right Left Center Right Dust content mg/Nm3 wet 415 572 199 565 244 207

This very uneven distribution refers to either very different electric operation on different fields or very uneven gas velocity/inlet dust content distribution across the ESP cross-sectional area. Also uneven gas temperature/H2O content/O2 content distribution due to high air inleakage can contribute to this result.


The ESP operating parameters during the measurements are shown in the following Figure 6.

Figure 6, ESP operating data

Electrical performance of precipitator unit B1 for 18.10.05 /7:30/

field 3 field 4 field 7 field 8 amperes in secundar 1,5 1,2 1,5 1volts in secundar 67,8 67,8 72,32 76,84watt in secundar 101,7 81,36 108,48 76,84volts primer 300 300 320 340

field 1 field 2 field 5 field 6amperes in secundar 1,0 1,0 0,2 0,2volts in secundar 67,8 67,8 22,6 27,12watt in secundar 67,8 67,8 4,52 5,424volts primer 300 300 100 120

total el.power (KW)514

Electrical performance of precipitator unit B2 for 18.10.05 /7:30/

field 3 field 4 field 7 field 8 amperes in secundar 1,5 1,2 1,5 1,6volts in secundar 35 40 42 50watt in secundar 52,5 48 63 80volts primer

field 1 field 2 field 5 field 6amperes in secundar 0,8 0,2 0,7 0,2volts in secundar 40 42 40 40watt in secundar 32 8,4 28 8volts primer

total el.power (KW)320

line 1 line 2

line 1 line 2

Nga : Bashkim Gjurgjeala, udhëheqës I analizave Tel: 561 113/ lokal – 358 Fax : 561314;E mail

During the tests the electrical input on fields 5 and 6 on line 2 ESP were very low, an indication of some operational problems.

To determine the electrical characteristics of the flyash in regards to its collectability in electrostatic precipitator, one flyash sample was analyzed for resistivity in Alstom laboratory in Sweden. The result is shown in following Figure 7.


Figure 7, Flyash resistivity analysis

According to the results the characteristics of the flyash are in good range considering the electrical operation of the ESP. Generally if the resistivity is over 1,0E+12, the flyash is considered to be of high resistivity. In this case the size of ESP has to be increased and modern pulsing type of high voltage transformer controls have to be applied. ESP inspection During the data collection visit the ESP on B1 was partially inspected. On line 1 all the fields were entered, on line 2 only two fields were entered. Before the inspection the “Final Report on the visual inspection of the electrostatic precipitator plant at Unit B1 of Kosovo B Power Plant” prepared by Vattenfall Europe in 2003 was reviewed. The findings were in good conformity to the above document. Repairs of some collection electrodes had been carried out but numerous worn-out electrodes were found. The voltage distances were within reasonable tolerances compared to normal criteria. The findings and recommendation done in the above document can be agreed.

SO2 content and Ca/S ratio The measured SO2 content after the ESP was in average of 240 mg/Nm3 dry 6 % O2. This means that the plant was operating below the SO2 emission of 400 mg/Nm3 dry 6 % O2 required by EU LCP Directive.


This value is clearly lower than measured by INKOS 2004 shown in Table 10. The theoretical SO2 concentration with lignite organic sulphur content of 0,48 % AR is 3100 mg/Nm3 dry 6 % O2. The measured SO2 emission value corresponds to 92 % sulphur capture in the boiler. With lignite sulphur content of 0,86 % AR the corresponding values are SO2 content of 5500 mg/Nm3 dry 6 % O2 and boiler sulphur capture of 96 %. The SO2 emission value after ESP was rather unstable during the measurement campaign, which refers to some changes either in boiler sulphur capture or in lignite quality. NOx and CO contents were clearly more stable. There were some small fluctuations in O2 value during the tests. O2 profile was checked in the ESP outlet duct before the emission measurements and it was found to be quite even To closer analyze the boiler plant Ca/S balance flyash samples were taken during the tests analyzed with results shown in Table 18 and in Enclosure 5 and 6.


Table 18, Flyash and bottom ash analysis results

Flyash Bottomash Amount t/h 63,6 2,3 % 96,5 3,5 H2O content % 0,3 49,7 Cl % DB 0,02 0,07 UBC % DB 0,9 10,7 UBC by INKOS % DB 3,1 14,9 SiO2 % DB 36 51 Al2O3 % DB 12 13 Fe2O3 % DB 7,6 6,4 CaO % DB 27 8,0 MgO % DB 3,5 2,2 SO3 % DB 7,0 K2O % DB 0,97 15 MnO2 % DB 0,24 0,21 Na2O % DB 1,0 0,74 P2O5 % DB 0,17 0,12 TiO2 0,42 0,55 As mg/kg DB 31 10 Ba mg/kg DB 610 420 Be mg/kg DB < 2,5 < 2,6 B mg/kg DB 580 230 Cd mg/kg DB < 0,5 < 0,51 Co mg/kg DB 17 18 Cr mg/kg DB 160 180 Cu mg/kg DB 47 43 Hg mg/kg DB 0,11 < 0,046 Mo mg/kg DB 3,5 2,6 Ni mg/kg DB 190 150 Pb mg/kg DB 18 16 Sb mg/kg DB 2,6 < 2,6 Sn mg/kg DB < 2,5 < 2,6 V mg/kg DB 95 90 Zn mg/kg DB 68 84 S mg/kg DB 28100 5000 % DB 2,8 0,5 SO3 % DB 7,0 1,3 As total S in fuel % DB 0,83 -- % AR 0,45

The bottom ash flow was measured by KEK during the tests.

In addition one flyash sample was analyzed for active Ca(OH)2. This is determined with acid titration method and it represents the amount of free reactive CaO present in flyash. This portion of total Ca has been calcined in the boiler from CaCO3 to CaO but not reacted with SO2 and is therefore available for sulphur removal. The results are shown in the Table 19.


Table 19, Flyash free Ca analysis results

Flyash Bottomash H2O % 0,3 45 UBC % BD 2,0 17,5 Total S % BD 3,1 0,5 Cl % BD 0,07 Ca(OH)2 % BD 10,2 0,5

UBC content of flyash is quite low compared to the plant history. Measured UBC was 0,9...3,1% in different samples. Total sulphur contents are close to same value in both flyash samples.

Based on above results and assuming that all sulphur in flyash is present as CaSO4, the Ca/S mass balance can be calculated as shown in Table 20.

Table 20, Ca/S balance, % AR and g/kgfuel AR

In lignite In flyash

% g/kgf % g/kgf % from in lignite

Ash 16,7 167,0 96,5 161,2 Total S 0,86 8,6 2,8 4,5 52 Inorg S 0,38 3,8 3,8 100 Org S 0,48 4,8 0,7 15 CaO 5,2 51,8 27,0 43,5 84 MgO 0,53 5,3 3,5 5,6 106 Ca(OH)2 10,2 Ca/S 5,3 % of Ca Total Ca 3,7 37,0 19,3 31,1 100 Ca as CaCO3 31,6 10,9 35 Ca as dolomite 5,3 5,6 18 Ca as active CaO 5,5 8,9 29 Bound as CaSO4 5,6 18

Based on this the following conclusions can be done: • 52 % of total sulphur contained in fuel can be found in flyash, 4,5 vs. 8,6 g/kgf.

Assuming all 3,8 g/kgf inorganic sulphur is present in flyash, then 0,7 g/kgf organic sulphur is found in flyash. This equals to 15 % SO2 capture in the boiler

• 84 % of total Ca contained in fuel can be found in flyash. 31,1 vs, 37,0 g/kgf.

• All Mg/dolomite contained in fuel can be found in flyash 5,6 vs. 5,3 g/kgf.


• 29 % of the total Ca in flyash is present as active CaO. This corresponds to Ca/S-org ratio of 1,5. This means that there is free reactive Ca left in flue gases to react with remaining SO2.

• 35 % of the total Ca in flyash is present as unreactive CaCO3, which has not been

calcined in the boiler.

• 18 % of the total Ca in flyash is present as sulphur compounds.’

As a summary of SO2 measurements it can be stated, that: • There are some uncertainties in analysis results of lignite sulphur content of EEO and

Inkos. The lignite total S-content during the tests was slightly lower and the organic S-content slightly higher than the average result analyzed by INKOS during 2002-2004. The analyzing method of organic sulphur content should be clarified,

• The fuel-flyash Ca-S mass balance does not explain the above uncertainties or the

measured low SO2 emissions. There is enough free CaO in flue gases to react with remaining SO2

• It shall be emphasized, that the total duration of measurement and fuel/flyash

sampling campaign was very short to give fully representative understanding of the boiler sulphur capture and Ca/S balance.

NOx content The measured NOx content of 661…713 mg/Nm3dry 6% O2 is higher than the value of 500 mg/Nm3 dry 6 % O2 required by EU LCP directive to be met on January 1, 2008. This value is lower than measured by INKOS 2004 shown in Table 10. The measured value is somewhat higher than expected by the boiler supplier.


5 SUMMARY AND RECOMMENDATIONS FOR MEASURES TO MEET LCP DIRECTIVE EMISSIONS

5.1 General The cost calculations shown in the below chapters are based on the assumptions shown in Table 21:

Table 21, Calculation data for one boiler unit

BOILER DATA Boiler load MW 285 Operating time h/a 8 000 Total produced power MWh/a 2 280 000 Lignite LHV MJ/kg 7 660 Lignite S-tot % AR 0,94 Lignite consumption t/MWh 1,40 t/h 400 Milj t/a 3,20 Heat rate kJ/kWh 10 720 Gross efficiency % 33,6 ESP DATA Dust after existing ESP mg/Nm3 dry 6 % O2 550 kg/h 700 Dust after new ESP mg/Nm3 dry 6 % O2 50 kg/h 64 Separated additional dust kg/h 636 SO2 DATA Lime injection SDA and WFGD SO2 after boiler mg/Nm3 dry 6 % O2 1 000 2000 kg/h 1 270 2 540SO2 after FGD mg/Nm3 dry 6 % O2 400 400 kg/h 510 510Separated SO2 kg/h 760 2 030FGD efficiency % 60 80COST DATA Investments holding time a 15 Interest % 10 Annuity factor 13,15 Electricity €/MWh 40,0 Ca(OH)2 €/t 58 CaO €/t 44 CaCO3 €/t 30 End product €/t 3,5 Water €/m3 0,05 Personnel €/man-year 7 000

Lot 2, Appendix 4 February 6, 2006 Flue gas cleaning Page 31 (50) 5.2 Plant efficiency

As stated earlier Kosovo B TPP is at the present operating at greatly reduced efficiency compared to the original design. As the demonstrated fuel gross efficiency in the early stage was 40 % the present actual efficiency is in range of 30 – 35 %. This means that the plant total pollutant mass flows (per produced electricity unit) are 20 – 30 % higher compared to the original operation. The limited plant maximum power capacity offsets some of this issue. It is recommended to take all possible action to bring the plant efficiency more close to the original design value. This mainly requires optimization of the plant operation and maintenance practices.

5.3 Dust emissions Summary The measured average dust emission was 550 mg/Nm3 dry 6% O2.

By EU LCP directive the dust emission value of 50 mg/Nm3 dry 6% O2 has to be reached before January 1, 2008.

As explained above, the existing ESP size in very small compared to modern high efficiency ESP’s. The following factors effect to the performance of existing precipitators: • High design gas velocity of 1,7 m/s. High efficiency filters are normally designed for

1,2…1,3 m/s • Low design inlet dust load 30 g/Nm3 and low design collection efficiency 99,14 %.

• High gas velocity combined with higher than design inlet dust load will increase

drastically outlet emissions, especially rapping losses.

• The dust content is even higher during boiler soot blowing, which, based on present practice, is in continuous operation to keep the boiler operational.

• The inlet dust content and flue gas flow are dependent of lignite analysis, which varies

in rather wide range causing fluctuations in the ESP performance.

• The high O2 content of flue gases increases the gas flow and velocity.

• Air inleakages including exhaust air from flyash transport system can cause uneven temperature, O2 and fluegas moisture distribution deteriorating performance. Generally gas velocity distribution is a very important factor for proper ESP operation.

• High UBC content of flyash deteriorates the performance of ESP.


• Low boiler efficiency and heat rate increases ESP loading.

The effect of lignite ash content to ESP inlet dust load and performance requirement is presented in Table 22. The average ash content of monthly composite samples during 2002-2004 was 16,7 % and maximum 24,4 % The maximum ash content in daily samples during the same time period was 32,3 %, see Tables 2 and 3. Relative ESP size with present design gas flow to reach LCP directive requirement is also shown in the table 22.

Table 22, ESP size comparison

Lignite ash content % AR 14,0 16,7 24,4 32,2 Inlet dust content g/Nm3 wet 30 38 60 95

g/Nm3 dry 6 % O2 38 48 76 93

Outlet dust content mg/Nm3 wet 260

mg/Nm3 dry 6 % O2 330 50 50 50 Specific collecting plate area m2/m3/s 38,2 79,7 90,7 95,7 Efficiency % 99,14 99,90 99,93 99,95 wk factor 59,2 62,7 68,0 72,0 Relative ESP size 1,0 2,0 2,1 2,1

In modern power plants ESP is often designed to meet the required emission value while one electrical section or field is out of service and running on the worst fuel expected. This gives also more flexibility to the operators in case o disturbances. In modern precipitators the voltage distance is commonly 400 mm instead of old design with 300 mm. This requires higher secondary voltage in transformer-rectifiers, typically 100 kV. With 400 mm voltage distance generally with same casing size, i.e. with same gas flow and velocity, the same efficiency can be reached with less collecting plate area, thus decreasing the costs. Also wk factor will be proportionally higher Recommendations Based on above to meet EU LCP directive requirements an electrostatic precipitator with five electrical field and relative size of 2,5 compared to existing precipitators is required. With average lignite quality the required outlet emission can be met with four electrical fields, having one spare field in these conditions. With lower lignite quality all five fields are used The possibility to extend the existing precipitator with three fields should be studied. In this case the ID fans shall be relocated. The benefit of this solution is, that the pressure drop of the ESP plant increases only marginally caused by ducting changes. The existing ID fans can be used in this case.

Costs

Investments/boiler 10 Milj €

Operating costs/boiler

Power consumption 1000 kW Maintenance 3 %/a 0,3 Milj €/a


Total costs Kosovo B1+B2

Investments costs 1,5 Milj €/a Operating costs 1,2 Milj €/a Total costs 2,7 Milj €/a

0,6 €/MWh 0,27 €/removed t of dust

Space requirement: 20 m after existing ESP’s. There is space available

Delivery time: 18-22 months Erection time 10 months Plant shut-down time: If new separate ESP:

4 – 6 weeks If extending existing ESP’s: 3 – 5 months

5.4 SO2 emission

Summary The measured average dust emission was 240 mg/Nm3 dry 6% O2.

By EU LCP directive the SO2 emission value of 400 mg/Nm3 dry 6% O2 has to be reached before January 1, 2008. As stated above the measurement results cannot be considered to be representative for long-term emission consideration. There was strong fluctuation in measured SO2 emission indicating either variations in lignite quality or variations in boiler sulphur capture. The fuel and flyash Ca/S mass balance does not support the understanding of the actual sulphur capture in the boiler. In earlier measurements by INKOS the sulphur capture in the boiler was in range of 60 %. In similar boilers with similar fuels with ample Calcium (Estonian oil shale) sulphur capture in range of 70…80 % has been measured. Also due to nature of sulphur contained in Kosovo lignite there are some unclear issues in analyzing and determining the proportion of fuel sulphur which actually is burning in the boiler to produce SO2 emissions. In the following Figure 8 the theoretical SO2 concentration after indicated sulphur capture % in the boiler is shown as function of lignite sulphur content.


Figure 8, SO2 emission depending of fuel sulphur and boiler capture

0,12

0,27

0,58

0

500

1 000

1 500

2 000

2 500

3 000

3 500

4 000

0,0 0 ,1 0 ,2 0,3 0,4 0 ,5 0 ,6 0 ,7 0 ,8 0 ,9 1 ,0

L ignite S -o rg content, % AR

SO2

emis

sion

mg/

Nm

3 dr

y 6

% O

2

0

90

Req

uire

d ad

dtio

nal S

O2

rem

oval

% fo

r 400

mg/

Nm

3

B o iler su lphur cap ture % 40%

50%

80%

70%

60%

90%

N orm al lign ite S -org range

M in 0,12 % M ax 0,58 % A vg 0,27 %

89

87

80

84

73

60

20

Assuming 70 % sulphur capture in the boiler the required SO2 emission by LCP can be met up to lignite organic sulphur content of 0,2% AR. An additional sulphur capture of 25 % is required with 2002-2004 average lignite sulphur content of 0,27 % and 65 % with maximum sulphur content of 0,58 %. If the sulphur capture in the boiler is lower, higher additional SO2 removal is required. Also correspondingly if higher portion of lignite total sulphur is burning to SO2 higher additional SO2 removal is required. Recommendations

Follow-up Program To analyse and clarify the actual boiler behaviour in regards to sulphur capture in the boiler due to fuel alkalinity in various operating conditions and fuel qualities we recommend: • Install a permanent SO2 analyser after at least one boiler unit, preferably for both

boilers. • Launch a research program/measurement campaign to study the behaviour of the SO2

emission in different operating conditions and variable fuels coming from the mines.

• Include in the research program detailed laboratory analysis of the fuel and flyash parameters to determine the organic sulphur analyzing method, limestone calcination degree in the boiler and Ca/S balance

The Follow-up Program is required to establish reliable design parameters to the measures required to bring Kosovo B TPP SO2 emissions to the level required by LCP


directive. The selected method for SO2 reduction dictates also the requirements for the ESP enlargement.

To reduce the plant SO2 emissions the following SO2 reduction technologies could be considered:

• Classification of lignite by sulphur content and firing low sulphur fuel in Kosovo B

plant and higher sulphur fuel in new Power Plant • Dry lime injection to the boiler

• Installing Flue Gas Desulphurization system after Kosovo B boiler plant. The

following technologies should be considered

• Semi-Dry FGD process

• Wet FGD process Classification if lignite by sulphur content Based on the present understanding this is only theoretical concept and is not possible to implement at Kosovo B TPP Dry lime injection to the boiler In this technology dry limestone or hydrated lime is injected to the upper part of the boiler where it reacts with SO2 in the flue gases see Figure 9. The reagent should be injected to the boiler in proper temperature range and it should be distributed and mixed evenly with the flue gases across the whole passage area. This requires a pneumatic conveying system with numerous injection nozzles. Good quality fine ground limestone CaCO3 or hydrated lime Ca(OH)2 can be used. If hydrated lime is used, the proper temperature range is 750…1200 oC. If limestone is used temperature has to be over 900 oC, which is the lower dissociation temperature of CaCO3 to CaO and CO2. Hydrated lime decomposition temperature from Ca(OH)2 to CaO and H2O is 580 oC . Flue gas temperature over 1200 oC can result in sintering of the surface of sorbent destroying the structure of the pores and reducing the reactive surface area. The reaction of CaO with SO2 and O2 produces CaSO4. The reaction occurs as long as the temperature is over 750 oC and the transformation occurs quickly within 1-2 sec. Below 750 oC the reaction practically stops. With hydrated lime SO2 removal efficiency up to 50 % has been reported with Ca/S ratio of 2. Currently Kosovo B uses hydrated lime for water treatment at a cost of 60 EUR/ton. If limestone is used, lower efficiencies and/or higher Ca/S ratio is expected. For proper injection locations and temperature zones in the boiler it is recommended consult/involve the boiler manufacturer.


Figure 9, Lime injection to the boiler

The system requires lime silo, lime dosing feeder and pneumatic conveying system with distribution piping and boiler injection nozzles. The injection nozzles may require changes in boiler pressure parts. Because extra lime is injected to the flue gases this will increase the dust load to the ESP. This has to be taken into account in sizing the possible ESP extension. In 1987-1988 two sets of tests were performed in one of the boilers to assess the natural desulphurization. The following data is available of these tests:

Table 23, Natural desulphurization measurement 1987-1988

Lignite LHV kJ/kg 5860 – 9145 H2O content % 40,4 – 45,5 Ash content % 11,6 – 24,5 S content % 0,8 – 1,6 Ca/S 1,5 – 3,8 Boiler SO2 capture % 85 – 95

The conclusion of the tests was that “The natural desulphurization efficiency in Kosovo B boiler was extremely high even at relatively low Ca/S ratio. Contemporary investigations are contemplated to explain this exceptional behaviour, which may be related to quite conservative furnace design.” In table 23 obviously lignite total sulphur content value was used to calculate the SO2 capture % and Ca/S ratio. Based on above it can be concluded that additional SO2 removal potential in Kosovo B boiler with lime/limestone injection is rather low but not completely outruled. Extra capture could be achieved by careful optimization of lime or limestone injection system. However it is imperative to understand the natural desulphurization behavior in Kosovo B boiler and used fuel qualities to determine the additional needed desulphurization degree.


Lime or limestone injection to the boiler will increase the outlet dust load from the boiler and will affect the performance of the ESP. This has to be taken into account in designing the possible ESP enlargement. Semi-Dry Flue Gas Desulphurization There are two commercial semi-dry processes available today, the traditional semi-dry process, (SDA, Spray Dry Absorption) and new integrated desulphurization process. In both processes quicklime CaO or hydrated lime Ca(OH)2 is used as sorbent. In SDA process, see Figure 10, the sorbent CaO is first slaked to Ca(OH)2 in a lime slaker, producing “lime milk” slurry in solids concentration of 15..20 %. The lime milk is then sprayed to a reactor either with mechanical rotary atomizer or with air atomizing nozzles. In rotary atomizers the fine slurry droplets are produced with centrifugal energy in an atomizing disc with rotational speed of 10 000 …12 000 rpm. In air atomized nozzle system the fine droplets are produced with nozzle pressure energy with slurry and compressed air pressure of 8...10 bar.

Figure 10, SDA process

The physical design of the reactor depends on the atomizing principle. With rotary atomizers the reactor works with turbulent rotating mixing pattern, With nozzle system the reactor operates with laminar plug flow pattern. In both cases the reactor is rather big in size, the gas velocity is in range of 2-3 m/s and residence time in range of 10 sec. The temperature after the reactor is regulated with amount of atomized slurry (amount of water for cooling) so that the temperature after reactor is at least 20 oC above the adiabatic saturation temperature (wet bulb temperature). Therefore the reactors and other flue gas system can be manufactured from normal mild steel. Surface treatment


requirements are mainly dictated by the amount of chlorides contained in the fuel or in process water. In the reactor the slurry droplet are completely dried and the dust after the reactor is completely dry. The dry dust is then separated either in electrostatic precipitator or in fabric filter. Portion of the collected dust can be recirculated back to the process to improve lime utilization. The portion of dust corresponding to reaction products is conveyed to the end product silo. The SO2 separation takes place mainly in the reactor. If fabric filter is used as end-collector additional SO2 capture takes place on the surface of filter bags. Integrated Desulphurization process. The integrated desulphurization process, see Figure 11, operates basically with the same chemical and temperature principal as SDA process. The main difference is that the water for gas cooling, instead of spraying to flue gases as slurry as in SDA, is first mixed with large amount of dust in a special mixer unit and the moisturized ”dry” dust is then mixed with flue gases in a duct reactor.

Figure 11, Integrated desulphurization process

The moisture of the humidified duet is in range of 3…5% (instead of 80…85 % in SDA) and can be thus handled and conveyed as dry dust. The temperature after the reactor is regulated with mount of dust feed to the rector and is of the same range as in SDA process, i.e. approx. 20 oC above the adiabatic saturation temperature. In this technology the reactor is practically a vertical flue gas duct between the mixer and end collector. Much higher flue gas velocities and shorter residence times are used compared to SDA process. Therefore the separate big absorption reactor is not needed as in SDA reducing space requirement and costs. The dry dust is separated after the reactor in an end-collector, either electrostatic precipitator or fabric filter. The main portion of the collected dust is recirculated back to


the mixer unit. The portion of dust corresponding to reaction products is conveyed to the end product silo. Either CaO or Ca(OH)2 is used as sorbent. In case of CaO it is first converted to Ca(OH)2 in a integrated dry slaker. The dry sorbent is then mixed with the recirculated dust in the mixer unit. One special feature of this technology is that if the flyash entering the process contains adequate amount of free CaO, it can be activated to sorbent Ca(OH)2 in the mixer unit. In this case no extra sorbent is required. There are some references of this technology with this operating principle. However to determine the suitability of this possibility each case has to be studied carefully in details including pilot testing in appropriate laboratory. In both technologies the pressure drop of the Semi-Dry FGD plant using fabric filter as end-collector, is in range of 3000...4000 Pa depending of complexity of flue gas ducting. In case of ESP as end-collector the pressure drop is 1000...1500 Pa lower. An additional booster fan for FGD plant is therefore required. In Kosovo B case the flue gases to the FGD system can be taken after the existing ID fans. SDA process can accept rather high inlet dust concentrations and therefore the existing ESP’s can operate as pre-collectors without any need for extensions. After FGD plant treatment the flue gases are taken back to duct before the stack and then to existing stack. The old duct will be equipped with by-pass damper. The main portion of the reaction products in both technologies is rather soluble CaSO3. In most cases it is first mixed with flyash for stabilization and then deposited in environmental sustainable manner. There are limited amount of utilization found to the SDA end-product In case of adapting integrated FGD process without extra sorbent by utilizing the fly ash alkalinity, adequate amount of reactive flyash has to enter the FGR process, either with flue gases or as separate dust stream from flyash collected in the existing ESP’s. The amount should be adjusted based on the free alkalinity of the flyash. Approx. 10 % of free Ca(OH)2 was analysed in the flyash during EE tests, With maximum lignite sulphur content of 0,58 % AR and 70 % boiler sulphur capture in the boiler the amount of flyash required to reach LCP SO2 emission limit of 400 mg/Nm3 dry 6 % O2 is with Ca/S ratio is approx. 20 t/h. This is about 30 % of the total collected flyash in the ESP’s. Wet Flue Gas Desulphurization, WFGD In WFGD system the flue gases are treated in a wet scrubber type absorber, see Figure 12. The slurry from the bottom sump of the reactor is recirculated with recalculation pumps to the upper part of the absorber, where it is distributed to the flue gases either with low pressure spray headers and nozzle system or with grid/packing system to absorb with SO2 from the flue gases. The slurry then returns to the absorber sump.


Figure 12, WFGD process

The slurry pH in the sump is adjusted in range of 5,5…6,0 by adding normally fine ground limestone or other suitable sorbent. Low pH is required to dissolve CaCO3 to free Ca to react with SO2 in flue gases and maintain the pH value of the scrubbing liquor.

In the flue gases SO2 is absorbed to the scrubbing liquor to form first CaSO3, which is then force oxidized to gypsum CaSO4 * 2 H2O by blowing air to the slurry. In the sump the gypsum crystals grow slowly and are the removed from the slurry based on slurry solids content, which is normally in range of 15-20 %. The gypsum crystals are then removed either in vacuum belt filter or in centrifuge. The removed water is then recirculated back to the process.

Normally pebble type limestone in size of 1-5 mm is used as raw material. Limestone is then ground at the FGD plant to limestone slurry with wet ball mills. Fresh limestone slurry is the used as sorbent.

Due to low pH and often high chloride content the absorbers have to be lined with corrosion resistant lining as rubber or high nickel alloy lining like C 276 to control corrosion. Fibreglass scrubbers are also in operation.

Wastewater treatment usually is required in WFGD plant to remove chlorides and fine particulates from the systems. With low chloride fuels a closed loop system without wastewater treatment can be used. In this case the blowdown of chlorides and fines takes place with the water contained in the gypsum.

The end product of WFGD process is gypsum, which can be utilized in construction industry.

Because the flue gases after WFGD absorber are fully saturated, they have to be reheated before entering the dry stack. Normally rotary type Gas-Gas Heaters (GGH) is used for this purpose, Typically in lignite fired plants the flue gas temperature after absorber is 60…65 oC and it is normally reheated to 95...105 oC before entering the stack. The energy is taken from the flue gases before the absorber by cooling them typically from 170 oC to 120 oC .


Higher temperature in the stack is required for corrosion standpoint and also to disperse the stack plume higher to the atmosphere better dispersion and to prevent the flue gas and condensate droplet fall down near by the plant.

In some cases a wet stack without reheating can be used. However this solution requires a special new acid brick lined or other corrosion resistant stack.

One common modern solution is to take the saturated flue gases to the cooling tower, where they are dispersed and exited with ambient air. In this case no GGH and flue gas stack is required. In retrofit cases however the type and material of cooling tower elements have to be checked for suitability.

Wet scrubbers can achieve removal efficiencies up to 98…99%. The removal requirement normally dictated the number of recirculation pumps and proper L/G (Liquid to Gas) ratio. Costs The following is the summary of costs calculations for each alternative: FOLLOW-UP PROGRAM

Investments: Two environmental analyser systems 0,4 Milj €

Manpower costs: 0,3 Milj € Laboratory costs etc. 0,1 Milj € Total costs: 0,8 milj € Execution time: 6 months

LIME INJECTION

Investments costs/boiler: 4 – 5 Milj, € Operating costs/boiler Electrical power consumption 150 kW Ca(OH)2 3000kg/h, 3-4 kg/kg removed SO2 Maintenance 3 %/v 0,2 Milj €/v Operating personnel 0,5 operator/shift Total costs Kosovo B1+B2


0,9 €/MWh 0,35 €/ t of removed SO2

Space requirement: L 20 m, W 20 m, H 20 m Delivery time: 6 – 8 months Plant shut-down time: 1 – 2 months


Note: ESP enlargement requirement shall be clarified. SDA Investments costs/boiler: 25 Milj €

Operating costs/boiler:

Electrical power consumption Booster fan 2 500 kW Others 1 500 kW Total 4 000 kW

Water 80 m3/h CaO 2 400 kg/h,

1,2 kg/kg removed SO2 End product 5 500 kg/h,

2,7 kg/kg removed SO2 Maintenance 3 %/v 0,75 Milj €/v Operating personnel 2 operators per shift




Space requirement: L 80 m, W 50 m, H 40 m Delivery time: 22 – 24 months Plant shut-down time: 6 – 10 weeks (duct connections)

Note: ESP enlargement is not required

INTEGRATED SDA

Investment costs/boiler: 20 Milj € Operating costs:

Electrical power consumption Booster fan 2500 kW Others 800 kW Total 3 300 kW

Water 80 m3/h CaO 3 100 kg/h

1,5 kg/kg removed SO2 End product 6 500 kg/h,

3,2 kg/kg removed SO2 Maintenance 3 %/v 0,6 Milj €/v Personnel, 2 operators per shift Total costs Kosovo B1+B2

Investments costs 3,0 Milj €/a Operating costs 6,0 Milj €/a


Total costs 9,0 Milj €/a 2,0 €/MWh 0,28 €/ t of removed SO2

Space requirement: L 60 m, W 40 m, H 25 m Delivery time: 22 – 24 months Plant shut-down time: 6 – 10 weeks (duct connections)

Note: ESP enlargement is not required

WFGD

Investments costs/boiler: Without GGH 50 Milj € With GGH 55 Milj €

Operating costs/boiler:

Electrical power consumption Booster fan 1500 kW no GGH

Booster fan 2500 kW with GGH Others 4 000kW

Total 5 500 or 6 500 kW Water 120 m3/h CaCO3 3 500 kg/h,

1,7 kg/kg removed SO2 Gypsum 7 700 kg/h,

3,8 kg/kg removed SO2 Maintenance 3 %/v 1,5 Milj €/v Personnel, 3 operators per shift




Space requirement: L 100 m, W 60 m, H 40 m Delivery time: 30 – 32 months Plant shut-down time: 8 – 12 weeks with GGH

(duct connections)

Note: The need for ESP enlargement shall be clarified based on the technical solution.

Lot 2, Appendix 4 February 6, 2006 Flue gas cleaning Page 44 (50) 5.5 NOx emissions

Summary The measured average dust emission was 660 – 710 mg/Nm3 dry 6% O2.

By EU LCP directive the NOx emission value of 500 mg/Nm3 dry 6% O2 has to be reached before January 1, 2008 and emission of 200 mg/Nm3 dry 6% O2 before January 1, 2016.

The NOx emissions are produced in a combustion process mainly by two mechanisms. The “Thermal NOx” is produced in the flame by oxidizing the combustion air nitrogen in high temperature to NOx. The “Fuel NOx” is produced by oxidizing the organic nitrogen contained in the fuel to NOx. Small portion of fuel nitrogen in bound as inorganic compounds in fuel ash and carbon residue but this contribution is very minor to the total NOx emissions. The amount of “Thermal NOx” is mainly dependent on combustion temperature and residence time in high temperature. The amount of Fuel NOx is mainly dependent on fuel nitrogen content, the Fuel Ratio (FR=Fixed Carbon/Volatiles) and amount of the excess air in the main flame, i.e. stochiometric conditions in the primary combustion zone. Therefore the main measures to minimize the formation of NOx emissions are to decrease the combustion temperature and decrease the excess air ratio in primary flame. Lignite fired boilers are always rather large in size, twice as large compared to PC boilers. Therefore the combustion temperature is naturally low. Also the high water content of the lignite lowers the combustion temperature. Therefore the amount of thermal NOx is comparatively low compared to hardcoal fired boilers. The NOx emissions are mostly effected by the design of burner-combustion air system and optimization of the operational parameters. The Low-NOx combustion system consists normally of Low-NOx burners and staged combustion air system. In lignite fired boilers two stage OFA (Over Fire Air) system is normally used, see Figure 13. Low NOx combustion system requires operation at low total excess air ratio, typically in range of 1,15. In the primary combustion zone the combustion stochiometry is below 1,0, typically 0,9 – 0,95.


Figure 13, Low-NOx combustion system

In Figure 14 the details are shown of modification of Kosovo B type combustion system to Low NOx firing system.

Figure 14, Low-NOx combustion system modification

Generally the following NOx emission values can be reached with modern Low-NOx combustion systems:


• Anthracite, FR=10, NOx <1000 mg/Nm3 6 % O2 • South African, FR=2,2, NOx <400 mg/Nm3 6 % O2 • Indonesian Coal, FR=1,2, NOx <250 – 300 mg/Nm3 6 % O2 • Lignite, FR=0,6, NOx <200 mg/Nm3 6 % O2 The FR of Kosovo B lignite is average of 0,5. Therefore the modern Low-NOx combustion systems offer an attractive low-cost primary method to control the formation of NOx emissions. The staged combustion system has also positive an effect in avoiding slagging of the boiler surfaces. Due to low excess air ratio the Low-NOx combustion system tends to increase the unburnt content of the flyash. Therefore the performance of the lignite pulverizing system has to be optimised to keep the UBC content below 5 %. Secondary methods i.e. SCR catalytic systems are not used in lignite fired boilers.

Recommendations In phase one the NOx emissions have to be reduced to 500 mg/Nm3 6 % O2.

It can be assumed, that it is possible to reach this NOx level by tuning up and optimizing the existing combustion system. This requires retuning the burner and combustion air distribution systems. The bottom burners can be optimized to operate at lower combustion stochiometry that the top burners. Also the total excess air ratio shall be optimised. At the same time the operation and maintenance practices of the lignite pulverizing system has to be optimised to prevent the increase of flyash UBC content. It is important to keep the fineness of the pulverized lignite close to the original design values. In phase one the NOx emissions have to be reduced to 200 mg/Nm3 6 % O2. This level can be reached by installing new Low NOx burners and staged combustion air system with OFA nozzles. Costs

Phase 1, optimizing the combustion system:

Total costs: 0,3 – 0,5 Milj € Delivery time: 2 – 3 months

Phase 2, Low NOx burners and OFA system

Investment costs/boiler: 10 – 13 Milj € Operating costs: No change


Total costs Kosovo B1+B2 Total costs 2,0 Milj €/a

0,4 €/MWh 0,33 €/ t of removed NOx

Delivery time: 26 months Plant shut-down time: 4 months


6 ACTION PLAN

Because the measures required to decrease the dust emissions are very much dependent on the final selection of SO2 emissions reduction method, it is important to clarify the SO2 issue at first. Therefore the following action plan can be proposed:

1 Clarify the SO2 emission characteristics of the boilers = Installation of emission analyzers and implementing the Follow-up Program

2 Establish the design criteria for SO2 reduction technology

3 Issue the tender for SO2 reduction technology

4 Select the suitable FGD reduction technology

5 Establish the design criteria for dust reduction technology depending on the selected SO2 reduction method as follows:

• If Lime/limestone injection is selected, determine the design criteria for ESP modification

• If Semi Dry FGD technology is selected, no ESP modification is required

• If Wet FGD technology is selected, determine the need and degree of ESP modification depending on the process requirements and whether waste gypsum or commercial gypsum is to be produced.

6 Execute the required modifications for ESP

7 Carry out the boiler combustion system optimization for NOx reduction

8 Carry out the measures to increase the boiler efficiency

Lot 2, Appendix 4 February 6, 2006 Flue gas cleaning Page 49 (50) 7 COST SUMMARY

The basis for cost calculations is shown in Tables 24…27.

Table 24, Cost calculations, Plant operating data

Plant operating data Power MW 285Operating time h/a 8 000Lignite LHV MJ/kg 7 660Lignite H2O % 43,3Lignite ash % AR 18,3Lignite S-tot % AR 0,94Lignite consumption t/MWh 1,40 t/h 399Gross efficiency % 33,6

Table 25, Cost calculations, Plant emission data

Emission data Dust SO2 NOx Lime Inj. DeSOx After boiler mg/Nm3 dry 6 % O2 55 500 1 000 2000 500 t/h 70,5 1 270 2 540 640 After existing ESP mg/Nm3 dry 6 % O2 550 kg/h 700 Emission mg/Nm3 dry 6 % O2 50 400 400 200 kg/h 60 510 510 260 Separated in old ESP kg/h 69 800 Separated mass flow kg/h 640 760 2 030 380 Efficiency % 99,91 60 80 60

Table 26, Cost calculations, Unit costs

Unit costs Investments holding time a 15Interest % 10Annuity factor 13,15 Electricity €/MWh 40,0Ca(OH)2 €/t 58CaO €/t 44CaCO3 €/t 30Endproduct €/t 3,5Water €/m3 0,05Personel €/manyear 7 000


Table 27, Cost calculations, Consumption data

Consumption Data/Boiler ESP Lime Inject. SDA ISDA WFGD

Electricity kW 1 000 200 4 000 3 300 6 500 CaCO3 kg/h 3 451 Ca(OH)2 kg/h 3 040 CaO kg/h 2 440 3 050 Water m3/h 80 80 120 Endproduct kg/h 5 480 6 500 7 720 Personell (5 shifts) 3 10 10 15 Space requirement Length m 20 10 80 60 100 Width m 30 10 50 40 60 m2 100 4000 2400 6000 Height m 25 25 40 25 40

The cost summary is shown in the following Table 28

Table 28, Cost summary

COST SUMMARY ESP Lime Inject SDA ISDA WFGD

Low-NOx

burners Investments M€ 20 10 50 40 110 26 Investment costs M€/a 1,5 0,8 3,8 3,0 8,4 2,0 Operating costs M€/a 1,2 3,4 6,3 6,0 9,6 Total costs M€/a 2,7 4,2 10,1 9,0 18,0 2,0 Specific total costs €/MWh 0,6 0,9 2,2 2,0 3,9 0,4 €/t dust 0,27 €/t SO2 0,35 0,31 0,28 0,55 €/t NO2 0,33 Delivery time Months 18-22 6-8 22-24 22-24 30-32 26 Erection time Months 10 Plant shut-down time New flue gas line Weeks 4-6 6-10 6-10 8-12 Modify existing units Months 3-5 1-2 4

FIP000085-05, B1 Fuel, Kosovo, 18.10.05Sample Identity

MoistureAsh cont.Ash cont. as rec.Volatile MatterVolatile Matter as rec.C-fix (calc.)C-fix as rec. (calc.)Sulphur SSulphur S as rec.Chlorine ClChlorine Cl as rec.Carbon CCarbon C as rec.Hydrogen HHydrogen H as rec.Nitrogen NNitrogen N as rec.Oxygen O (calc.)Oxygen O as rec. (calc.)Emission factor (calculated)

Analysis%% db%% db%% db%% db%% db%% db%% db%% db%% db%ton CO2/TJ

UnitsISO 589ISO 1171/ASTM-D 5142 modISO 1171/ASTM-D 5142 modISO 562/ASTM-D 5142 modISO 562/ASTM-D 5142 modISO 562/ASTM-D 5142 modISO 562/ASTM-D 5142 modASTM-D 4239 CASTM-D 4239 CASTM-D 4208ASTM-D 4208ASTM-D 5373ASTM-D 5373ASTM-D 5373ASTM-D 5373ASTM-D 5373ASTM-D 5373ASTM-D 5373ASTM-D 5373

45.930.916.748.426.220.711.21.60.86<0.01<0.0144.223.92.96.70.80.419.651.499.1

Result

8447276-832275Customer umberEnergy FuelsSample Type

Electrowatt-Ekono OyMika KoivunenPL 93

FINLAND

+

Certificate of Analysis

02151 Espoo

Metod/ref

***

LLLLLLLLLLLLLLLLLLLL

Rapport utfärdad av

Site

Report issued by

ackrediterat laboratorium

Accredited Laboratory

2005-11-02Arrival dateReport printed 2005-11-18

Lidköping

Acc.±±±±±±±±±±±±±±±±±

%%%%%%%%%%%%%%%%%

21515557710102525101010103030

Sample Number BE000914-05

Explanation to abbreviations and *, see reverse side..

**

*****

****

±±±±±±±±±±±±±±±±

%%%%%%%%%%%%%%%%

4.78.946650<2.55902231<1012150505.98.4<0.0450.43

% db% dbmg/kg TSmg/kg TSmg/kg Tsmg/kg TS% db% dbmg/kg Tsmg/kg TSmg/kg TSmg/kg TS% db% dbmg/kg Ts% db

Aluminium AlAluminiumoxide Al2O3Arsenic AsBarium BaBeryllium BeBoron BCalcium CaCalciumoxide CaOCadmium CdCobalt CoChromium CrCopper CuIron FeIronoxide Fe2O3Mercury HgPotassium K

20202520202020203030303020202520

EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.SS028150-2SS028150-2SS028150-2EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.SS028150-2EN 13656 mod.

LLLLLLLLLLLLLLLL

Page 1 (2)

MJ/kg

Analysis UnitsResult

8447276-832275Customer umberEnergy FuelsSample Type

Bengt Axelsson

Metals analyzed from pre-ashed sampel.V2O4=0.020 % ; SO3=8.75 %


Metod/ref


Site

Report issued by



Lidköping

Acc.


Explanation to abbreviations and *, see reverse side..

***

***

*********

±±±±±±±±±±±±±±±±±±±±±

%%%%%%%%%%%%%%%%%%%%%

0.521.93.20.110.177.20.761.0190150.0800.182.81124<2.5350000.160.27120110

% db% db% db% db% dbmg/kg Ts% db% dbmg/kg TSmg/kg TS% db% dbmg/kg TS% db% dbmg/kg TSmg/kg TS% db% dbmg/kg Tsmg/kg TS

Potassiumoxide K2OMagnesium MgMagnesiumoxide MgOMangan MnManganoxid MnO2Molybdenum MoSodium NaSodiumoxide Na2ONickel NiLead PbPhosphorus PPhosphorusoxide P2O5Antimony SbSilicon SiSiliconoxide SiO2Tin SnSulfur STitanium TiTitaniumoxide TiO2Vanadium VZinc Zn

202020303020202030202020202020202020202530

B

B

EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.

LLLLLLLLLLLLLLLLLLLLL

Kcal/kg MWh/ton BTU/lb

16.975 4057 4.714 72989.184 2195 2.550 3948

7.806 1866 2.168 335616.381 3915 4.549 704223.706 5666 6.583 101917.724 1846 2.145 332116.347 3907 4.540 702823.657 5654 6.570 10170

Calorific value

Net calorificvalue

as receiveddry basis

Const volume - dry basisConst volume - as received

Const pressure - as receivedConst pressure - dry basis

Const volume - db, ashfree

Const pressure - db, ashfree

Calorific Value SS-ISO 1928

Page 2 (2)

FIP000085-05, B1 Fule, Kosovo, 18.10.05, tilläggsanalysProvets märkning

Analysnamn EnhetResultat

8447276-843556Bränsle, energiProvtyp

Analysansvarig

Bengt Axelsson

Sulphur 1200 C = 1.35 % ; Sulphur 1000 C = 1.06 % ; Sulphur 800 C = 0.85 %.


FINLAND

Analysrapport

02151 Espoo

2005-11-30Provtagningsdatum

Metod/ref


Ort

Report issued by



2005-11-02Provet ankomAnalysrapport klar 2005-12-01

Lidköping

Mäto.

KundnrJournalnr BE001047-05

Förklaring till förkortningar och *, se omstående sida.

Sida 1 (1)

FIP000083-05, B1 Fly ash, Kosovo, 18.10.05Sample Identity

MoistureCont. incombustible prod.Chlorine ClSilicon SiSiliconoxide SiO2Calcium CaCalciumoxide CaOAluminium AlAluminiumoxide Al2O3Iron FeIronoxide Fe2O3Potassium KPotassiumoxide K2OMagnesium MgMagnesiumoxide MgOMangan MnManganoxid MnO2Sodium NaSodiumoxide Na2OPhosphorus PPhosphorusoxide P2O5Titanium TiTitaniumoxide TiO2Antimony SbArsenic AsLead PbBarium BaBeryllium BeBoron BCadmium CdCobalt CoCopper CuMercury HgChromium CrMolybdenum MoNickel Ni

Analysis%% d.m% d.m% d.m% d.m% d.m% d.m% d.m% d.m% d.m% d.m% d.m% d.m% d.m% d.m% d.m% d.m% d.m% d.m% d.m% d.m% d.m% d.mmg/kg d.mmg/kg d.mmg/kg d.mmg/kg d.mmg/kg d.mmg/kg d.mmg/kg d.mmg/kg d.mmg/kg d.mmg/kg d.mmg/kg d.mmg/kg d.mmg/kg d.m

UnitsSS 187170SS 187187ASTM-D 4208EN 13656 mod.EN 13656 mod.SS028150-2SS028150-2EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.SS028150-2EN 13656 mod.EN 13656 mod.EN 13656 mod.SS028150-2EN 13656 mod.EN 13656 mod.EN 13656 mod.

0.30.90.02173619276.2125.37.60.800.972.13.50.150.240.771.00.0750.170.250.422.63118610<2.5580<0.5017470.111603.5190

Result

8447276-832275Customer NumberFuel EnergiSample Type


FINLAND


02151 Espoo

2005-11-08Sampled Date

MethodLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLL

Site

2005-11-02Arrival DateReport printed 2005-11-18

Acc.±

±±±±±±±±±±±±±±±±±±±±±±±±±±±±±±±±±±

%

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

6

25202020202020202020202020303020202020202020252020202030303025302030


Explanation to abbreviations and *, see reverse side.

B

Page 1 (2)

Vanadium VTin SnZinc ZnSulfur S

Analysismg/kg d.mmg/kg d.mmg/kg d.mmg/kg d.m

UnitsEN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.

95<2.56828100

Result


AnalysansvarigBengt Axelsson

V2O4=0.015 % ; SO3=7.0 %


MethodLLLL

SiteAcc.±±±±

%%%%

25203020



B

Page 2 (2)

FIP000084-05, B1 Bottom ash, Kosovo, 18.10.05Sample Identity

MoistureCont. incombustible prod.Chlorine ClSilicon SiSiliconoxide SiO2Calcium CaCalciumoxide CaOAluminium AlAluminiumoxide Al2O3Iron FeIronoxide Fe2O3Potassium KPotassiumoxide K2OMagnesium MgMagnesiumoxide MgOMangan MnManganoxid MnO2Sodium NaSodiumoxide Na2OPhosphorus PPhosphorusoxide P2O5Titanium TiTitaniumoxide TiO2Antimony SbArsenic AsLead PbBarium BaBeryllium BeBoron BCadmium CdCobalt CoCopper CuMercury HgChromium CrMolybdenum MoNickel Ni

Analysis%% d.m% d.m% d.m% d.m% d.m% d.m% d.m% d.m% d.m% d.m% d.m% d.m% d.m% d.m% d.m% d.m% d.m% d.m% d.m% d.m% d.m% d.mmg/kg d.mmg/kg d.mmg/kg d.mmg/kg d.mmg/kg d.mmg/kg d.mmg/kg d.mmg/kg d.mmg/kg d.mmg/kg d.mmg/kg d.mmg/kg d.mmg/kg d.m

UnitsSS 187170SS 187187ASTM-D 4208EN 13656 mod.EN 13656 mod.SS028150-2SS028150-2EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.SS028150-2EN 13656 mod.EN 13656 mod.EN 13656 mod.SS028150-2EN 13656 mod.EN 13656 mod.EN 13656 mod.

49.710.70.0724515.78.07.1134.56.41.21.51.32.20.130.210.550.740.0530.120.330.55<2.61016420<2.6230<0.511843<0.0461802.6150

Result



FINLAND

+


02151 Espoo

2005-11-08Sampled Date

MethodLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLL

Site

2005-11-02Arrival DateReport printed 2005-11-18

Acc.±

±±±±±±±±±±±±±±±±±±±±±±±±±±±±±±±±±±

%

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

6

25202020202020202020202020303020202020202020252020202030303025302030



B

Page 1 (2)

Vanadium VTin SnZinc ZnSulfur S

Analysismg/kg d.mmg/kg d.mmg/kg d.mmg/kg d.m

UnitsEN 13656 mod.EN 13656 mod.EN 13656 mod.EN 13656 mod.

90<2.6845000

Result


AnalysansvarigBengt Axelsson

V2O4=0.015 % ; SO3=1.25 %


MethodLLLL

SiteAcc.±±±±

%%%%

25203020



B

Page 2 (2)

DISPERSION OF EXHAUST GASES FROM

KOSOVO B POWER PLANT IN OBILIC, KOSOVO

Risto Varjoranta

Harri P ietar i la

DISPERSION OF EXHAUST GASES FROM KOSOVO B POWER PLANT IN OBILIC, KOSOVO

Risto Varjoranta Harri Pietarila

FINNISH METEOROLOGICAL INSTITUTE AIR QUALITY EXPERT SERVICES

Helsinki 20.12.2005

CONTENTS

1 INTRODUCTION .............................................................................................. 3

2 COMPUTATIONAL METHODS...................................................................... 3

2.1 Local scale dispersion model .................................................................... 3

2.2 Meteorological methods and data.............................................................. 5

3 EMISSIONS AND SOURCE CHARACTERISTICS........................................ 7

4 RESULTS……. .................................................................................................. 7

4.1 General…. ................................................................................................. 7

4.2 Concentrations........................................................................................... 9

5 DISCUSSION…............................................................................................... 10

5.1 The ambient air quality limits in the EC ................................................. 10

5.2 Maximum concentrations compared to the EC air quality limits............ 12

6 CONCLUSION ................................................................................................ 14

REFERENCES .................................................................................................. 15

FIGURES

3

1 INTRODUCTION This report discusses the effects on local air quality of the emissions from Kosovo B power plant situated in Obilic about 10 kilometers westwards from the city of Pristina in Kosovo. The dispersion of sulphur dioxide, nitrogen oxides and particulate matter emit-ted from the power plant was simulated by the Finnish Meteorological Institute (FMI) using the operational local scale air pollution dispersion model. The computed concen-trations were compared with the EU limit values. The meteorological data used in the simulations was extracted from the archives of the European Centre for Medium Range Weather Forecast. The data covering the years 1996─1998 was interpolated from the available observations of synoptic weather sta-tions located nearest the plant site. The source term and the emission data were supplied by Electrowatt-Ekono Oy. The emission data supplied was based on emission meas-urements performed in October 2005. 2 COMPUTATIONAL METHODS

2.1 Local scale dispersion model In this study the impact of the power plant on air quality and the pollution levels of sul-phur dioxide, nitrogen oxides and particles were simulated with the urban dispersion modelling system of the Finnish Meteorological Institute (UDM-FMI). The dispersion model is based on Gaussian plume equations for various source categories. The Gaus-sian plume equations can be mathematically derived from the atmospheric diffusion equation in case of homogenous and stationary turbulence. We address the solutions for a point, a volume, an area and a line source. The dispersion parameters are modelled in a form, which facilitates the use of the mete-orological pre-processor. Parametrizations are therefore written as a function of the Monin-Obukhov length, the friction velocity and the mixing height (instead of the Pas-quill classes). The model can be adjusted to the site-specific conditions by changing few parameters of surface and terrain characteristics. The roughness length for various surfaces (buildings, forests, open field etc.) can be adjusted for different directions from the emission source. When the radiative and conductive energy fluxes considered in the model are

4

resolved are the site-specific conditions taken into account by adjusting the parameter for surface albedo. The surface elevation can be given separately for each receptor point. Difference in roughness of various surfaces (e.g. city, rural) gives only minor dif-ference (at maximum few tens of percents) in ground level concentrations. The differ-ence in meteorological conditions in different locations has clearly bigger effect on the concentrations than the differences in surface roughness or albedo. It is most vital that the meteorological data used in the dispersion calculations should represent the actual meteorology of the plant site. The plume rise and several removal processes are included into the model framework. The plume rise comprise the aerodynamic plumerise and the plume rise due to buoy-ancy. The model allows also the downwash due to the buildings and the stack. The re-moval processes contain three phenomena: dry deposition, wet deposition and gravita-tional settling of particles. The chemical transformation for NO oxidation to NO2 is taken into account in the model to evaluate the actual NO2-concentrations in ambient air. Nitrogen oxides (NOx) emitted in burning of fossil fuels consists from a larger part of nitric oxide (NO, 90–95 %) and from a smaller part of nitrogen dioxide (NO2, 5–10 %). After exhaust from flue the percentage of NO2 in air starts to increase through oxidation of NO mainly by at-mospheric ozone (O3). In the Finnish local scale model the initial fraction of NO2 in the emission can be indi-vidually set for each source. In this study the initial fraction of NO2 in the emission is assumed to be 5 % of the volumic rate of NOx-emission. NO2 produced by oxidation processes is approximated via empirical relation

NONO

A ex

X2 1= − −( )α (1)

where X is distance from the stack. Equation 1 as well as numerical values for variables A and α are based upon 10 years of measurements of NO2/NOx ratios in the stack plumes in the Netherlands (JANSSEN et al., 1988). The most important parameters in determining A and α are atmospheric conditions (e.g. wind speed) and seasonal varia-tion of ambient ozone concentrations. Diurnal variation, specified for each month of the year in the model, of background ozone concentrations is based in this study on the val-ues from the ozone measurements during 4 years period of 2001–2004 in Iskrba in Slo-

5

venia. The ozone data was fetched from the WMO’s World Data Centre for Greenhouse Gases. The Finnish operational air quality model is based on the steady-state hypothesis, where emission rates and meteorological conditions are assumed to remain constant during each model time-step (one hour). The one-hour air concentrations for each pollutant in emission inventory are computed in a rectangular area for the whole period of meteoro-logical data. The 1-hour mean values form the basis for statistical analyses for other av-eraging periods. In normal practice the daily (24-hour), monthly (30-day) and annual statistical means are computed. The results can be presented either as isolines on map or in tabular form. The isolines give an instant indication of those areas, where the expected air concentration levels of the pollutants exceed some prescribed value(s). By suitable choice of isoline separation these plots can be compared to national air quality standards or regulatory limits. The modelling system is described in more detail by Karppinen et al. (1997). 2.2 Meteorological methods and data The basic meteorological parameters relevant for dispersion simulations are wind, am-bient air temperature, boundary layer stability and mixing height. Wind determines the speed and direction of dispersion. Stability gives indication of the turbulent mixing rate inside the boundary layer. Turbulent mixing is the most important factor for pollutant dilution during transport. Mixing height describes the vertical extent of the plume. Turbulence data and boundary layer height are not available from any routine base measurements. Indirect methods have therefore been introduced to calculate these pa-rameters. The meteorological pre-processing model MPP-FMI developed in the Finnish Meteorological Institute (KARPPINEN et al., 1997 & 2000) has been utilised in this study. This pre-processing model is based on a slightly modified version of the energy budget method of van Ulden and Holtslag (1985). This method evaluates the turbulent heat and momentum fluxes in the atmospheric boundary layer (ABL) by utilising the routinely available synoptic weather observations. The parameterisation of the ABL height is based on the boundary layer scaling and meteorological sounding data. In ab-sence of sounding data the mixing height is estimated analytically (ARYA, 1981, KITAIGORODSKII & JOFFRE, 1988). The output of the pre-processor consists of es-timates of the hourly time series of the relevant atmospheric parameters (the Monin-

6

Obukov length scale, the friction velocity and the convective velocity scale) as well as the boundary layer height. Meteorological time series for dispersion simulations are compiled by interpolating the weather data to the site of application (city or factory location) with a straightforward distance-weighted interpolation. Several synoptic stations can be included in the inter-polation. Time series normally cover from 1 to 3 years of data, depending on applica-tion. For model applications outside the Finnish borders, the pre-processor incorporates a routine to fetch synoptic data from the data archives of the European Centre for Me-dium Range Weather Forecasts (ECMWF) at Reading, the UK. For this application the weather observations from the weather stations of Pristina (42°39′ N, 21°09′ W), Prizren (42°13′ N, 20°44′ W) and Pec (42°40′ N, 20°18′ W) were used. The plant site is situated about 10 kilometers to the west of Pristina, about 60 kilometers to the southwest of Prizren and abot 70 kilometers to the east of Pec. The synoptic weather observations of the years 1996─1998 were used. This period was cho-sen because for the more recent years the availability and quality of local meteorologi-cal data was too poor for dispersion simulations. The wind speed and direction statistics of the whole period of 1996─1998 is presented as a wind rose in figure 1. The percent values on the circles give the proportion of each direction (or main sector) in the whole data. For each direction the wind speed distribu-tion is presented in five categories (percentage scale in each sector in the figure). The most frequent wind direction is northeast (18 %) followed by north (17 %). The propor-tion of light winds (speed category under 2 m/s) is 35─55 % depending on the direction sector. The proportion of fresher winds (wind speed over 5 m/s) is greatest in the south west sector. The estimated mixing heights and the model computed boundary layer mixing rate, computed as a function of inverse of the Monin-Obukhov length are given in figure 2. The monthly variation of these variables indicates that poor mixing and low mixing heights corresponding obviously to the nocturnal conditions are more common during winter than during summer. In daytime the solar heating is intensive enough to cause unstable conditions with stronger mixing and quite high mixing heights through the year.

7

3 EMISSIONS AND SOURCE CHARACTERISTICS The Kosovo B power plant consists of two blocks each having power generation capac-

ity up to 280 MWe. Main fuel of the power plant is lignite. The relevant source charac-

teristics and emission rates as supplied by the Electrowatt-Ekono are as follows:

− the stack height 182 meters − the inner diameter of stack 9,6 meters − the height of the power house 90 meters − the exhaust gas volume flow 3 665 000 m3n/h − the exhaust gas temperature 150 ºC The used emission rates corresponding to maximum power production are:

− sulphur dioxide (SO2) 260 g/s − nitrogen oxides (NOx) 556 g/s − particulate matter (PM ) 376 g/s

The emission data supplied was based on emission measurements performed by Elec-trowatt-Ekono Oy in October 2005. Only the emissions from Kosovo B unit 1 were measured as unit 2 was not in operation at the time. To be on the safe side it was as-sumed that the emissions from unit 2 are the same, and that both units are operating with maximum capacity in the modelled situation. In the dispersion simulations it has been assumed that the plant is running at full load 6500 hours per year and that the running hours are randomly distributed through the year. Then the approximate mean annual emissions would be: sulphur dioxide 6100 t/a, nitrogen oxides 13000 t/a and particles 8800 t/a. 4 RESULTS 4.1 General Three air pollutants were considered, namely sulphur dioxide (SO2), nitrogen oxides (NOx) and particles. The fraction of NOx converted to nitrogen dioxide (NO2) was computed as a separate component.

8

The pollutant concentrations were computed in a rectangular area of 10 000 m x 10 000 m with the plant located approximately in the centre of this area. The grid mesh was 100 m around the plant in a square of 4 000 m x 4 000 m and 500 m over the rest of the area, resulting in total 2041 grid points. Over these grids the 1-hour air concentrations (in µg/m3) for each pollutant and for all hours of considered 3 years period (26 304 hours) were computed. Based on 1-hour val-ues the daily, monthly and annual mean concentration fields were accumulated. Isolines for each quantity were then analysed over the grid points. The basic quantities are de-fined as follows: − highest annual mean is the largest true mean value for each grid point over the

whole simulation period (3 years) − highest daily (24-hour) mean is the largest true mean value averaged for each grid

point of the simulation period i.e. each value represents the worst pollution situation of the period in this grid point

− 4th highest daily (24-hour) mean is the largest value for each grid point of the 4th highest daily mean value calculated for each year of the simulation period

− 36th highest daily (24-hour) mean is the largest value for each grid point of the 36th highest daily mean value calculated for each year of the simulation period

− highest 1-hour mean is the largest value for each grid point of the simulation period i.e. the absolute maximum value for each grid point

− 19th highest 1-hour mean is the largest value for each grid point of the 19th highest daily mean value calculated for each year of the simulation period

− 25th highest 1-hour mean is the largest value for each grid point of the 25th highest daily mean value calculated for each year of the simulation period

It should be noted that neither the 24-hour nor the 1-hour mean refer to any single con-tinuous pollution situation or single reference time. Rather they represent the worst case over the whole period, which can (and usually does) happen at different times in each grid point. Results of dispersion calculations are presented as isopleths in figures 3–10. Quantities presented in the isopleths presentations are those comparable to the air quality standards of the EC. The figures cover the nearest vicinity of the plant (10 000 m x 10 000 m). Locations of maximum are depicted with yellow stars.

9

In general the distributions of concentrations follow closely the wind rose (figure 1). However it should be kept in mind that the maximum do not refer to any single mete-orological situation or reference time. The changes in the 1-hour wind direction cause the plume to be transported to different direction from one hour to the other. Conse-quently the isopleths should be interpreted as a probability distribution, giving the most probable dispersion directions present in meteorological data considered. During any other period the distribution may be different. Numerical values of maximum depend upon the emission rate assumptions and the meteorological data used. If the plant is not operated continuously and on constant load or the used meteorological data is not valid for the plant site, the maximum differ from those presented here.

4.2 Concentrations The maximum computed concentrations of the considered pollutants due the emissions of the Kosovo B power plant are presented in table 1. The figures given in the table in-clude all the values comparable to the EC limits values. The spatial distributions of the concentrations are presented as isopleths in figures 3─10. Table 1. The maximum computed air concentrations (µg/m3) of sulphur dioxide (SO2), nitrogen

dioxide (NO2), nitrogen oxides (NOx) and particulate matter (PM10) due to the emissions of Kosovo B power plant. The values presented in bold italic correspond to the EC limit values.

Averaging time Concentrations [µg/m3] SO2 NO2 NOx PM10

Annual mean 0,3 0,15 0,6 0,4

Highest daily mean 6,1 4,3 12,9 8,8

4th highest daily mean 3,2 2,1 6,8 4,6

36th highest daily mean 1,0 0,5 2,1 1,4

Highest hourly mean 54,5 36,7 116,5 78,8

19th highest hourly mean 22,0 12,3 47,0 31,8

25th highest hourly mean 21,0 10,3 44,9 30,4

From the isopleth presentations it can be seen that around the stack there is zone of sev-eral hundred meters with quite low concentrations which is typical for stack releases (‘stack shading’ effect). The distribution patterns suggest that main dispersion sectors

10

are southwest or south from the emission source which agrees well with the wind distri-bution in the used meteorological data. From he distributions of the highest sulphur dioxide concentrations (figures 3─5) it can be seen that the maximum of the annual mean (0,3 µg/m3) is situated about 2 kilometres southwest from the plant site. The maximum of the highest daily mean (3,2 µg/m3) is situated about 2,7 kilometres towards southwest and the maximum of the highest 1-hour mean (21 µg/m3) is situated about 1 kilometre towards south from the plant site. From the distributions of the NO2-concentrations (figures 7─8) it can be seen that the maxima are situated more far away from the plant site than are the maxima of SO2. The main reason for this is that the dispersion model takes into account the chemical transformation of the nitrogen oxides during the dispersion process. The maximum of annual mean NO2-concentration (0,15 µg/m3) is situated about 4,2 kilometres to the southwest from plant site and the maximum of the highest 1-hour mean (12,3 µg/m3) is situated about 6 kilometres to the southwest from plant site. The maximum of the annual mean (0,4 µg/m3) and the maximum of the highest daily mean (1,4 µg/m3) of particle concentration is situated about 2 kilometres to the south-west of the plant site (figures 9─10). 5 DISCUSSION 5.1 The ambient air quality limits in the EC

The European Communities (EC) has given common limit values for the ambient air concentrations to be attained within a given period and not be exceeded once attained in the member states. The aim of limit values is to avoid, prevent or reduce harmful effects on human health and/or the environment as a whole. The objective is to maintain ambi-ent air quality where it is good and improve it in other cases. The assessment of the con-centration in ambient air is carried out on the basis of common methods and criteria in the whole EC. The values of the EC limit values for sulphur dioxide (SO2), nitrogen oxides (NO2/NOx) and particulate matter (PM10) are given in table 2. The statistical definition for different air quality reference values is given on the footnotes in the same table. The EC hourly limit value for SO2 is not allowed to exceed more than 24 times and the daily limit value

11

more than 3 times during a calendar year. Respectively the hourly limit value for NO2

is not allowed to exceed more than 18 times and the daily limit value of PM10 more than 35 times during a calendar year. The PM10 limit values are to be tightened in stage 2 (to be met by 1.1.2010). The EC annual limit values do not allow any exceedances. Table 2. The EC limit values concerning sulphur dioxide, nitrogen oxides and particulate

matter.

Pollutant EC limit values [µg/m3 ]

Protection of Human Health

Sulphur Dioxide (SO2)

Daily mean 1251

Hourly mean 3502

Nitrogen Dioxide (NO2)

Annual mean 40

Hourly mean 2003

Particles, PM10

Annual mean 40/20

Daily mean 504

Protection of Ecosystems and Vegetation

Sulphur Dioxide (SO2)

Annual mean 20

Nitrogen Oxides (NOx)

Annual mean 30

1) Not to be exceeded more than 3 times during a calendar year 2) Not to be exceeded more than 24 times during a calendar year 3) Not to be exceeded more than 18 times during a calendar year 4) Not to be exceeded more than 35 times during a calendar year (stage 1) 7 times during a calendar year (stage 2, to be met by 1.1.2010)

12

5.2 Maximum concentrations compared to the EC air quality limits

The comparison of the maximum computed SO2, NO2 and PM concentrations to the re-spective EC air quality limit values is presented in figure A. All concentration values used in the comparison are those corresponding to the proper statistical definitions of the limits of the EC.

6,0 %2,5 %

6,1 %0,4 % 2,9 % 1,1 %

0 %

20 %

40 %

60 %

80 %

100 %

120 %

so2-hour so2-day no2-hour no2-year pm10-day pm10-year

so2no2pm10

Figure A. Comparison of the calculated maximum sulphur dioxide, nitrogen dioxide and particle concentrations to the EC limit values. The maximum computed 1-hour mean SO2-concentration (21 µg/m3) is 6,0 % of the limit value (350 µg/m3). The maximum of the computed 24-hour mean SO2-concentration (3,2 µg/m3) is 2,5 % of the limit value (125 µg/m3). The maximum com-puted 1-hour mean NO2-concentration is 6,1 % and the maximum annual mean is 0,4 % of the respective limit values. According to the results of the dispersion simulations the maximum 24-hour mean particle concentration is 2,9 % and the maximum annual mean is 1,1 % of the respective stage 1 limit values for PM10. The computed maxim annual mean nitrogen oxide (NOx) concentration is 2,1 % of the respective limit value and the maximum annual mean sulphur dioxide (SO2) concentra-tion is 1,5 % of the respective limit value. These annual limit values for NOx and SO2

13

are given for the protection of vegetation and their application area is limited. The com-parison is made for the absolute maximum concentrations of the study area. All over the rest of the study area the concentrations are lower than those used in the comparison.

14

6 CONCLUSIONS The local air concentrations of sulphur dioxide, nitrogen oxides and particles emitted from a 2×280 MWe power plant of Kosovo B situated in Obilic, Kosovo have been computed by the Finnish Meteorological Institute using an operational local scale air pollution dispersion model. The simulations were made with the meteorological data from available observations of weather stations relatively near the actual plant site. The chemical transformation for NO oxidation to NO2 was taken into account in the model to evaluate the actual NO2 and NO concentrations in ambient air. Diurnal variation, specified for each month of the year used in the model, of background ozone concentra-tions is based in this study on the values from the ozone measurements at nearest station of WMO’s network in Iskrba, Slovenia. The dispersion simulations were made with emission data corresponding to maximum power of the plant and supposing the plant is running 6500 hours per year. The results indicate that the maximum concentrations of sulphur dioxide, nitrogen dioxide, nitrogen oxides and particles would not exceed the European Communities (EC) Ambient Air Quality Limits which has been used as reference values in this study. According to the results the maximum sulphur dioxide and nitrogen dioxide concentra-tions caused by the emission of the considered power plant are about 6 % of the corre-sponding limit values given for protection of human health. The maximum particle concentrations will be lower than 5 % of the corresponding limit values. The annual mean concentrations of sulphur dioxide and nitrogen oxides due to the exhaust gases will be as most about 2 % of the corresponding limit values which are given for the pro-tection of vegetation. As the stack of the power plant is high (182 meters) the concentra-tions on the ground in the vicinity of the plant will be quite low and the concentration maxima will be situated from 500 meters to several kilometres from stack position. Ac-cording to the dispersion simulations the most probable direction of the highest ground surface concentrations would be to the southwest of the plant location. According to the dispersion model investigation the used stack (182 meters) is high enough to assure that the concentrations due to emissions of the power plant are suffi-ciently low on the ground surface to meet clearly the air quality standards of the Euro-pean Communities.

15

REFERENCES ARYA, S.P.S., 1981. Parametrizing the height of the stable atmospheric boundary layer. J. Appl. Meteor. 20, p. 1192-1202. BUSINGER, J.A., WYNGAARD, J.C. IZUMI, Y. & BRADLEY, E.F., 1971. Flux-profile relations in the atmospheric surface layer. J. Atmos. Sci. 28, p. 181 - 189. DEARDORFF, J. W., 1972. Numerical investigation of neutral and unstable planetary boundary layers. J. Atmos. Sci., 29, 91-115. JANSSEN, L.H.J.M., van WAKEREN, J.H.A., van DUUREN, H. & ELSHOUT, A.J., 1988. A classification of NO oxidation rates in power plant plumes based on atmosphe-ric conditions. Atmos. Environ. 22:1, p. 43 - 53. KARPPINEN, A., KUKKONEN, J., NORDLUND, G., RANTAKRANS, E., VALKAMA, I., 1997. A dispersion modelling system for urban air pollution. Finnish Meteorological Institute. KARPPINEN, A., JOFFRE, S.M., KUKKONEN, J., 2000. The refinement of a mete-orological preprocessor for urban environment. International Journal of Environment and Pollution 14, p. 565–572. KITAIGORODSKII, S.A. & JOFFRE, S.M., 1988. In search of a simple scaling for the height of the stratified atmospheric boundary layer. Tellus 40A, p. 419-433. OLESEN, H.R., 1995. Datasets and protocol for model validation. Internal Journal of Environmental and Pollution, Vol. 5, Nos. 4-6, pp. 693-701. TENNEKES, H., 1973. A model for the dynamics of the inversion above a convective boundary layer. J. Atmos. Sci., 30, 558-567. ULDEN, A. P. van and HOLTSLAG A. A. M., 1985. Estimation of Atmospheric Boundary Layer Parameters for Diffusion Applications. J. Clim. Appl. Meteor., 24, 1196-1207.

FIGURES

Figure 1. Distribution of the wind direction and speed (wind rose) on the Obilic area in 1996-1998.

0 %

10 %

20 %

30 %

40 %

50 %

60 %

70 %

80 %

90 %

100 %

1 2 3 4 5 6 7 8 9 10 11 12

month

Poor mixing

Moderate mixing

Strong mixing

0 %

10 %

20 %

30 %

40 %

50 %

60 %

70 %

80 %

90 %

100 %

1 2 3 4 5

Figure 2. Monthly percenta (above) and estim on the Obilic area

> 500 m

200-500 m

< 200 m

6 7 8 9 10 11 12

month

ges of mixing rate categories ated mixing heights (below) in 1996-98.

0 1 2

kilometers

Leskovcic Obilic

Krucevac

Crkvena Vodica

SimicaAde

Milosevo

Priluzje

Bivoljak

Plenetina

Annual mean [µg/m³]EC limit = 20 µg/m³

> 0,250,20 to 0,250,15 to 0,200,10 to 0,15< 0,10

Finnish Meteorological Institute 2005= stack position of the power plant= maximum = 0.3 µg/m³

KOSOVO B POWER PLANTOBILIC, KOSOVO

Figure 3. Distribution of the annual mean sulphur dioxide concentration [µg/m³].

EC limit = 20 µg/m³

0 1 2

kilometers

Leskovcic Obilic

Krucevac

Crkvena Vodica

SimicaAde

Milosevo

Priluzje

Bivoljak

Plenetina

Daily mean [µg/m³]EC limit = 125 µg/m³

> 2,52,0 to 2,51,5 to 2,01,0 to 1,5< 1,0

Finnish Meteorological Institute 2005= stack position of the power plant= maximum = 3,2 µg/m³


Figure 4. Distribution of the highest daily mean sulphur dioxide concentration [µg/m³].


0 1 2

kilometers

Leskovcic Obilic

Krucevac

Crkvena Vodica

SimicaAde

Milosevo

Priluzje

Bivoljak

Plenetina

Hourly mean [µg/m³]EC limit = 350 µg/m³

> 2015 to 2010 to 155 to 10

< 5

Finnish Meteorological Institute 2005= stack position of the power plant= maximum = 21 µg/m³


Figure 5. Distribution of the highest hourly mean sulphur dioxide concentration [µg/m³].


0 1 2

kilometers

Leskovcic Obilic

Krucevac

Crkvena Vodica

SimicaAde

Milosevo

Priluzje

Bivoljak

Plenetina


> 0,40,3 to 0,40,2 to 0,30,1 to 0,2< 0,1



Figure 6. Distribution of the annual mean nitrogen oxides concentration [µg/m³].


0 1 2

kilometers

Leskovcic Obilic

Krucevac

Crkvena Vodica

SimicaAde

Milosevo

Priluzje

Bivoljak

Plenetina


> 0,140,10 to 0,140,06 to 0,100,02 to 0,06< 0,02



Figure 7. Distribution of the annual mean nitrogen dioxide concentration [µg/m³].


0 1 2

kilometers

Leskovcic Obilic

Krucevac

Crkvena Vodica

SimicaAde

Milosevo

Priluzje

Bivoljak

Plenetina

Hourly mean [µg/m³]EC limit = 200 µg/m³

> 106 to 104 to 62 to 4< 2



Figure 8. Distribution of the highest hourly mean nitrogen dioxide concentration [µg/m³].


0 1 2

kilometers

Leskovcic Obilic

Krucevac

Crkvena Vodica

SimicaAde

Milosevo

Priluzje

Bivoljak

Plenetina

Annual mean [µg/m³]EC limit µg/m³

> 0,40,3 to 0,40,2 to 0,30,1 to 0,2< 0,1



Figure 9. Distribution of the annual mean PM-concentration [µg/m³].


0 1 2

kilometers

Leskovcic Obilic

Krucevac

Crkvena Vodica

SimicaAde

Milosevo

Priluzje

Bivoljak

Plenetina

Daily mean [µg/m³]EC limit = 50 µg/m³

> 1,31 to 1,30,7 to 10,5 to 0,7< 0,5



Figure 10. Distribution of the highest daily mean PM-concentration [µg/m³].


I lmanlaadun asiantunti japalvelut

PL 503

00101 HELSINKI

puh. (09) 19291

ilmanlaatupalvelut@fmi. f i

Air qual i ty expert services

P.O.Box 503

FIN-000101 HELSINKI

tel . +358 9 19291

airquali ty.services@fmi. f i

www.fmi. f i

I L M AT I E T E E N L A I TO SF I N N I S H M E T E O R O L O G I CA L I N S T I T U T E

Kosovo B photo log (September 2005):

44444444444444444444444444444444444444444444444444444444444444444EMISSION MEASUREMENT TECHNICAL INFORMATION CENTER

NSPS TEST METHOD44444444444444444444444444444444444444444444444444444444444444444

Method 3A - Determination of Oxygen and Carbon Dioxide Concentrationsin Emissions from Stationary Sources(Instrumental Analyzer Procedure)

1. APPLICABILITY AND PRINCIPLE

1.1 Applicability. This method is applicable to the determination ofoxygen (O ) and carbon dioxide (CO ) concentrations in emissions from2 2

stationary sources only when specified within the regulations.

1.2 Principle. A sample is continuously extracted from the effluentstream: a portion of the sample stream is conveyed to an instrumentalanalyzer(s) for determination of O and CO concentration(s). Performance2 2

specifications and test procedures are provided to ensure reliable data.

2. RANGE AND SENSITIVITY

Same as in Method 6C, Sections 2.1 and 2.2, except that the span of themonitoring system shall be selected such that the average O or C02 2

concentration is not less than 20 percent of the span.

3. DEFINITIONS

3.1 Measurement System. The total equipment required for thedetermination of the O or CO concentration. The measurement system2 2

consists of the same major subsystems as defined in Method 6C, Sections3.1.1, 3.1.2, and 3.1.3.

3.2 Span, Calibration Gas, Analyzer Calibration Error, Sampling SystemBias, Zero Drift, Calibration Drift, Response Time, and CalibrationCurve. Same as in Method 6C, Sections 3.2 through 3.8, and 3.10.

3.3 Interference Response. The output response of the measurementsystem to a component in the sample gas, other than the gas componentbeing measured.

4. MEASUREMENT SYSTEM PERFORMANCE SPECIFICATIONS

Same as in Method 6C, Sections 4.1 through 4.4.

——————————————————————————————————————————————————————————————————————————————Prepared by Emission Measurement Branch EMTIC TM-003A

Technical Support Division, OAQPS, EPA May 6, 1989——————————————————————————————————————————————————————————————————————————————44444444444444444444444444444444444444444444444444444444444444444EMTIC TM-003A NSPS TEST METHOD Page 244444444444444444444444444444444444444444444444444444444444444444

5. APPARATUS AND REAGENTS

5.1 Measurement System. Any measurement system for O or CO that meets2 2

the specifications of this method. A schematic of an acceptablemeasurement system is shown in Figure 6C-1 of Method 6C. The essentialcomponents of the measurement system are described below:

5.1.1 Sample Probe. A leak-free probe of sufficient length to traversethe sample points.

5.1.2 Sample Line. Tubing to transport the sample gas from the probe tothe moisture removal system. A heated sample line is not required forsystems that measure the 0 or C0 concentration on a dry basis, or2 2

transport dry gases.

5.1.3 Sample Transport Line, Calibration Valve Assembly, MoistureRemoval System, Particulate Filter, Sample Pump, Sample Flow RateControl, Sample Gas Manifold, and Data Recorder. Same as in Method 6C,Sections 5.1.3 through 5.1.9, and 5.1.11, except that the requirements touse stainless steel, Teflon, and nonreactive glass filters do not apply.

5.1.4 Gas Analyzer. An analyzer to determine continuously the O or CO2 2

concentration in the sample gas stream. The analyzer must meet theapplicable performance specifications of Section 4. A means ofcontrolling the analyzer flow rate and a device for determining propersample flow rate (e.g., precision rotameter, pressure gauge downstream ofall flow controls, etc.) shall be provided at the analyzer. Therequirements for measuring and controlling the analyzer for measuring andcontrolling the analyzer flow rate are not applicable if data arepresented that demonstrate the analyzer is insensitive to flow variationsover the range encountered during the test.

5.2 Calibration Gases. The calibration gases for CO analyzers shall be2

CO in N or CO in air. Alternatively, CO /SO , O /SO , or O /CO /SO gas2 2 2 2 2 2 2 2 2 2

mixtures in N may be used. Three calibration gases, as specified in2

Sections 5.3.1 through 5.3.4 of Method 6C, shall be used. For O monitors2

that cannot analyze zero gas, a calibration gas concentration equivalentto less than 10 percent of the span may be used in place of zero gas.

6. MEASUREMENT SYSTEM PERFORMANCE TEST PROCEDURES

Perform the following procedures before measurement of emissions (Section7).

6.1 Calibration Concentration Verification. Follow Section 6.1 ofMethod 6C, except if calibration gas analysis is required, use Method 3and change the acceptance criteria for agreement among Method 3 resultsto 5 percent (or 0.2 percent by volume, whichever is greater).

6.2 Interference Response. Conduct an interference response test of theanalyzer prior to its initial use in the field. Thereafter, recheck themeasurement system if changes are made in the instrumentation that couldalter 44444444444444444444444444444444444444444444444444444444444444444EMTIC TM-003A NSPS TEST METHOD Page 344444444444444444444444444444444444444444444444444444444444444444

the interference response (e.g., changes in the type of gas detector).Conduct the interference response in accordance with Section 5.4 ofMethod 20.

6.3 Measurement System Preparation, Analyzer Calibration Error, ResponseTime, and Sampling System Bias Check. Follow Sections 6.2 through 6.4 ofMethod 6C.

7. EMISSION TEST PROCEDURE

7.1 Selection of Sampling Site and Sampling Points. Select ameasurement site and sampling points using the same criteria that areapplicable to tests performed using Method 3.

7.2 Sample Collection. Position the sampling probe at the firstmeasurement point, and begin sampling at the same rate as that usedduring the response time test. Maintain constant rate sampling (i.e.,±10 percent) during the entire run. The sampling time per run shall bethe same as for tests conducted using Method 3 plus twice the averagesystem response time. For each run, use only those measurements obtainedafter twice the response time of the measurement system has elapsed todetermine the average effluent concentration.

7.3 Zero and Calibration Drift Test. Follow Section 7.4 of Method 6C.

8. QUALITY CONTROL PROCEDURES

The following quality control procedures are recommended when the resultsof this method are used for an emission rate correction factor, or excessair determination. The tester should select one of the following optionsfor validating measurement results:

8.1 If both O and CO are measured using Method 3A, the procedures2 2

described in Section 4.4 of Method 3 should be followed to validate the O2and CO measurement results.2

8.2 If only O is measured using Method 3A, measurements of the sample2

stream CO concentration should be obtained at the sample by-pass vent2

discharge using an Orsat or Fyrite analyzer, or equivalent. Duplicatesamples should be obtained concurrent with at least one run. Average theduplicate Orsat or Fyrite analysis results for each run. Use the averageCO values for comparison with the O measurements in accordance with the2 2

procedures described in Section 4.4 of Method 3.

8.3 If only CO is measured using Method 3A, concurrent measurements of2

the sample stream CO concentration should be obtained using an Orsat or2

Fyrite analyzer as described in Section 8.2. For each run, differencesgreater than O.5 percent between the Method 3A results and the average of

the duplicate Fyrite analysis should be investigated.

44444444444444444444444444444444444444444444444444444444444444444EMTIC TM-003A NSPS TEST METHOD Page 444444444444444444444444444444444444444444444444444444444444444444

9. EMISSION CALCULATION

9.1 For all C0 analyzers, and for O analyzers that can be calibrated2 2

with zero gas, follow Section 8 of Method 6C, except express allconcentrations as percent, rather than ppm.

9.2 For O analyzers that use a low-level calibration gas in place of a2

zero gas, calculate the effluent gas concentration using Equation 3A-1.

C - Cma oa

C = ————————— (C̄ - C ) + C Eq. 3A-1gas m ma

C - Cm o

Where:

C = Effluent gas concentration, dry basis, percent.gas

C = Actual concentration of the upscale calibration gas, percent.ma

C = Actual concentration of the low-level calibration gas,oa

percent.

C = Average of initial and final system calibration bias check m

responses for the upscale calibration gas, percent.

C = Average of initial and final system calibration bias check o

responses for the low level gas, percent.

C̄ = Average gas concentration indicated by the gas analyzer, drybasis, percent.

10. BIBLIOGRAPHY

Same as in Bibliography of Method 6C.

444444444444444444444444444444444444444444444444444444444444444444444444444444EMISSION MEASUREMENT TECHNICAL INFORMATION CENTER

NSPS TEST METHOD44444444444444444444444444444444444444444444444444444444444444444444444444444

——————————————————————————————————————————————————————————————————————————————Prepared by Emission Measurement Branch EMTIC TM-006CTechnical Support Division, OAQPS, EPA March 14, 1990——————————————————————————————————————————————————————————————————————————————

Method 6C - Determination of Sulfur Dioxide Emissionsfrom Stationary Sources

(Instrumental Analyzer Procedure)


1.1 Applicability. This method is applicable to the determination of sulfurdioxide (SO ) concentrations in controlled and uncontrolled emissions from2

stationary sources only when specified within the regulations.

1.2 Principle. A gas sample is continuously extracted from a stack, and aportion of the sample is conveyed to an instrumental analyzer for determinationof SO gas concentration using an ultraviolet (UV), nondispersive infrared2

(NDIR), or fluorescence analyzer. Performance specifications and test proceduresare provided to ensure reliable data.


2.1 Analytical Range. The analytical range is determined by the instrumentaldesign. For this method, a portion of the analytical range is selected bychoosing the span of the monitoring system. The span of the monitoring systemshall be selected such that the pollutant gas concentration equivalent to theemission standard is not less than 30 percent of the span. If at any time duringa run the measured gas concentration exceeds the span, the run shall beconsidered invalid.

2.2 Sensitivity. The minimum detectable limit depends on the analytical range,span, and signal-to-noise ratio of the measurement system. For a well designedsystem, the minimum detectable limit should be less than 2 percent of the span.

3. DEFINITIONS

3.1 Measurement System. The total equipment required for the determination ofgas concentration. The measurement system consists of the following majorsubsystems:

3.1.1 Sample Interface. That portion of a system used for one or more of thefollowing: sample acquisition, sample transport, sample conditioning, orprotection of the analyzers from the effects of the stack effluent.

3.1.2 Gas Analyzer. That portion of the system that senses the gas to bemeasured and generates an output proportional to its concentration.

3.1.3 Data Recorder. A strip chart recorder, analog computer, or digitalrecorder for recording measurement data from the analyzer output.

3.2 Span. The upper limit of the gas concentration measurement range displayedon the data recorder.

444444444444444444444444444444444444444444444444444444444444444444444444444444EMTIC TM-006C EMTIC NSPS TEST METHOD Page 244444444444444444444444444444444444444444444444444444444444444444444444444444

3.3 Calibration Gas. A known concentration of a gas in an appropriate diluentgas.

3.4 Analyzer Calibration Error. The difference between the gas concentrationexhibited by the gas analyzer and the known concentration of the calibration gaswhen the calibration gas is introduced directly to the analyzer.

3.5 Sampling System Bias. The difference between the gas concentrationsexhibited by the measurement system when a known concentration gas is introducedat the outlet of the sampling probe and when the same gas is introduced directlyto the analyzer.

3.6 Zero Drift. The difference in the measurement system output reading fromthe initial calibration response at the zero concentration level after a statedperiod of operation during which no unscheduled maintenance, repair, oradjustment took place.

3.7 Calibration Drift. The difference in the measurement system output readingfrom the initial calibration response at a mid-range calibration value after astated period of operation during which no unscheduled maintenance, repair, oradjustment took place.

3.8 Response Time. The amount of time required for the measurement system todisplay 95 percent of a step change in gas concentration on the data recorder.

3.9 Interference Check. A method for detecting analytical interferences andexcessive biases through direct comparison of gas concentrations provided by themeasurement system and by a modified Method 6 procedure. For this check, themodified Method 6 samples are acquired at the sample by-pass discharge vent.

3.10 Calibration Curve. A graph or other systematic method of establishing therelationship between the analyzer response and the actual gas concentrationintroduced to the analyzer.


4.1 Analyzer Calibration Error. Less than ±2 percent of the span for the zero,mid-range, and high-range calibration gases.

4.2 Sampling System Bias. Less than ±5 percent of the span for the zero andmid-range calibration gases.

4.3 Zero Drift. Less than ±3 percent of the span over the period of each run.

4.4 Calibration Drift. Less than ±3 percent of the span over the period of eachrun.

4.5 Interference Check. Less than ±7 percent of the modified Method 6 resultfor each run.


5.1 Measurement System. Use any measurement system for SO that meets the2

specifications of this method. A schematic of an acceptable measurement systemis shown in Figure 6C-1. The essential components of the measurement system aredescribed below:


5.1.1 Sample Probe. Glass, stainless steel, or equivalent, of sufficient lengthto traverse the sample points. The sampling probe shall be heated to preventcondensation.

5.1.2 Sample Line. Heated (sufficient to prevent condensation) stainless steelor Teflon tubing, to transport the sample gas to the moisture removal system.

5.1.3 Sample Transport Lines. Stainless steel or Teflon tubing, to transportthe sample from the moisture removal system to the sample pump, sample flow ratecontrol, and sample gas manifold.

5.1.4 Calibration Valve Assembly. A three-way valve assembly, or equivalent,for blocking the sample gas flow and introducing calibration gases to themeasurement system at the outlet of the sampling probe when in the calibrationmode.

5.1.5 Moisture Removal System. A refrigerator-type condenser or similar device(e.g., permeation dryer), to remove condensate continuously from the sample gaswhile maintaining minimal contact between the condensate and the sample gas. Themoisture removal system is not necessary for analyzers that can measure gasconcentrations on a wet basis; for these analyzers, (1) heat the sample line andall interface components up to the inlet of the analyzer sufficiently to preventcondensation, and (2) determine the moisture content and correct the measured gasconcentrations to a dry basis using appropriate methods, subject to the approvalof the Administrator. The determination of sample moisture content is notnecessary for pollutant analyzers that measure concentrations on a wet basis when(1) a wet basis CO analyzer operated according to Method 3A is used to obtain2

simultaneous measurements, and (2) the pollutant/CO measurements are used to2

determine emissions in units of the standard.

5.1.6 Particulate Filter. An in-stack or heated (sufficient to prevent watercondensation) out-of-stack filter. The filter shall be borosilicate or quartzglass wool, or glass fiber mat. Additional filters at the inlet or outlet of themoisture removal system and inlet of the analyzer may be used to preventaccumulation of particulate material in the measurement system and extend theuseful life of the components. All filters shall be fabricated of materials thatare nonreactive to the gas being sampled.

5.1.7 Sample Pump. A leak-free pump, to pull the sample gas through the systemat a flow rate sufficient to minimize the response time of the measurementsystem. The pump may be constructed of any material that is nonreactive to thegas being sampled.

5.1.8 Sample Flow Rate Control. A sample flow rate control valve and rotameter,or equivalent, to maintain a constant sampling rate within 10 percent.(Note: The tester may elect to install a back-pressure regulator to maintain thesample gas manifold at a constant pressure in order to protect the analyzer(s)from overpressurization, and to minimize the need for flow rate adjustments.)

5.1.9 Sample Gas Manifold. A sample gas manifold, to divert a portion of thesample gas stream to the analyzer and the remainder to the by-pass dischargevent. The sample gas manifold should also include provisions for introducingcalibration gases directly to the analyzer. The manifold may be constructed ofany material that is nonreactive to the gas being sampled.

5.1.10 Gas Analyzer. A UV or NDIR absorption or fluorescence analyzer, todetermine continuously the SO concentration in the sample gas stream. The2

analyzer shall meet the applicable performance specifications of Section 4. A


means of controlling the analyzer flow rate and a device for determining propersample flow rate (e.g., precision rotameter, pressure gauge downstream of allflow controls, etc.) shall be provided at the analyzer. (Note: Housing theanalyzer(s) in a clean, thermally-stable, vibration-free environment willminimize drift in the analyzer calibration.)

5.1.11 Data Recorder. A strip chart recorder, analog computer, or digitalrecorder, for recording measurement data. The data recorder resolution (i.e.,readability) shall be 0.5 percent of span. Alternatively, a digital or analogmeter having a resolution of 0.5 percent of span may be used to obtain theanalyzer responses and the readings may be recorded manually. If thisalternative is used, the readings shall be obtained at equally spaced intervalsover the duration of the sampling run. For sampling run durations of less than1 hour, measurements at 1-minute intervals or a minimum of 30 measurements,whichever is less restrictive, shall be obtained. For sampling run durationsgreater than 1 hour, measurements at 2-minute intervals or a minimum of96 measurements, whichever is less restrictive, shall be obtained.

5.2 Method 6 Apparatus and Reagents. The apparatus and reagents described inMethod 6 and shown by the schematic of the sampling train in Figure 6C-2 are usedto conduct the interference check.

5.3 SO Calibration Gases. The calibration gases for the gas analyzer shall be2

SO in N or SO in air. Alternatively, SO /CO , SO /O , or SO /CO /O gas mixtures2 2 2 2 2 2 2 2 2 2

in N may be used. For fluorescence-based analyzers, the O and CO2 2 2

concentrations of the calibration gases as introduced to the analyzer shall bewithin 1 percent (absolute) O and 1 percent (absolute) CO of the O and CO2 2 2 2

concentrations of the effluent samples introduced to the analyzer.Alternatively, for fluorescence-based analyzers, use calibration blends of SO2

in air and the nomographs provided by the vendor to determine the quenching correction factor (the effluent O and CO concentrations must be known). Use2 2

three calibration gases as specified below:

5.3.1 High-Range Gas. Concentration equivalent to 80 to 100 percent of thespan.

5.3.2 Mid-Range Gas. Concentration equivalent to 40 to 60 percent of the span.

5.3.3 Zero Gas. Concentration of less than O.25 percent of the span. Purifiedambient air may be used for the zero gas by passing air through a charcoal filteror through one or more impingers containing a solution of 3 percent H O .2 2


Perform the following procedures before measurement of emissions (Section 7).

6.1 Calibration Gas Concentration Verification. There are two alternatives forestablishing the concentrations of calibration gases. Alternative No. 1 ispreferred.

6.1.1 Alternative No. 1--Use of calibration gases that are analyzed followingthe Environmental Protection Agency Traceability Protocol No. 1 (see Citation 1in Bibliography). Obtain a certification from the gas manufacturer that ProtocolNo. 1 was followed.

6.1.2 Alternative No. 2--Use of calibration gases not prepared according toProtocol No. 1. If this alternative is chosen, obtain gas mixtures with amanufacturer's tolerance not to exceed ±2 percent of the tag value. Within


6 months before the emission test, analyze each of the calibration gases intriplicate using Method 6. Citation 2 in the Bibliography describes proceduresand techniques that may be used for this analysis. Record the results on a datasheet (example is shown in Figure 6C-3). Each of the individual SO analytical2

results for each calibration gas shall be within 5 percent (or 5 ppm, whicheveris greater) of the triplicate set average; otherwise, discard the entire set andrepeat the triplicate analyses. If the average of the triplicate analyses iswithin 5 percent of the calibration gas manufacturer's cylinder tag value, usethe tag value; otherwise, conduct at least three additional analyses until theresults of six consecutive runs agree within 5 percent (or 5 ppm, whichever isgreater) of the average. Then use this average for the cylinder value.

6.2 Measurement System Preparation. Assemble the measurement system byfollowing the manufacturer's written instructions for preparing andpreconditioning the gas analyzer and, as applicable, the other system components.Introduce the calibration gases in any sequence, and make all necessaryadjustments to calibrate the analyzer and the data recorder. Adjust systemcomponents to achieve correct sampling rates.

6.3 Analyzer Calibration Error. Conduct the analyzer calibration error checkby introducing calibration gases to the measurement system at any point upstreamof the gas analyzer as follows:

6.3.1 After the measurement system has been prepared for use, introduce thezero, mid-range, and high-range gases to the analyzer. During this check, makeno adjustments to the system except those necessary to achieve the correctcalibration gas flow rate at the analyzer. Record the analyzer responses to eachcalibration gas on a form similar to Figure 6C-4. Note: A calibration curveestablished prior to the analyzer calibration error check may be used to convertthe analyzer response to the equivalent gas concentration introduced to theanalyzer. However, the same correction procedure shall be used for all effluentand calibration measurements obtained during the test.

6.3.2 The analyzer calibration error check shall be considered invalid if thegas concentration displayed by the analyzer exceeds ±2 percent of the span forany of the calibration gases. If an invalid calibration is exhibited, takecorrective action and repeat the analyzer calibration error check untilacceptable performance is achieved.

6.4 Sampling System Bias Check. Perform the sampling system bias check byintroducing calibration gases at the calibration valve installed at the outletof the sampling probe. A zero gas and either the mid-range or high-range gas,whichever most closely approximates the effluent concentrations, shall be usedfor this check as follows:

6.4.1 Introduce the upscale calibration gas, and record the gas concentrationdisplayed by the analyzer on a form similar to Figure 6C-5. Then introduce zerogas, and record the gas concentration displayed by the analyzer. During thesampling system bias check, operate the system at the normal sampling rate, andmake no adjustments to the measurement system other than those necessary toachieve proper calibration gas flow rates at the analyzer. Alternately introducethe zero and upscale gases until a stable response is achieved. The tester shalldetermine the measurement system response time by observing the times requiredto achieve a stable response for both the zero and upscale gases. Note thelonger of the two times as the response time.

6.4.2 The sampling system bias check shall be considered invalid if thedifference between the gas concentrations displayed by the measurement system for


the analyzer calibration error check and for the sampling system bias checkexceeds ±5 percent of the span for either the zero or upscale calibration gases.If an invalid calibration is exhibited, take corrective action, and repeat thesampling system bias check until acceptable performance is achieved. Ifadjustment to the analyzer is required, first repeat the analyzer calibrationerror check, then repeat the sampling system bias check.


7.1 Selection of Sampling Site and Sampling Points. Select a measurement siteand sampling points using the same criteria that are applicable to Method 6.

7.2 Interference Check Preparation. For each individual analyzer, conduct aninterference check for at least three runs per during the initial field test ona particular source category. Retain the results, and report them with each testperformed on that source category. If an interference check is being performed,assemble the modified Method 6 train (flow control valve, two midget impingerscontaining 3 percent H O , and dry gas meter) as shown in Figure 6C—2. Install2 2

the sampling train to obtain a sample at the measurement system sample by-passdischarge vent. Record the initial dry gas meter reading.

7.3 Sample Collection. Position the sampling probe at the first measurementpoint, and begin sampling at the same rate as used during the sampling systembias check. Maintain constant rate sampling (i.e., ±10 percent) during theentire run. The sampling time per run shall be the same as for Method 6 plustwice the average system response time. For each run, use only thosemeasurements obtained after twice the response time of the measurement system haselapsed to determine the average effluent concentration. If an interferencecheck is being performed, open the flow control valve on the modified Method 6train concurrent with the initiation of the sampling period, and adjust the flowto 1 liter per minute (±10 percent). (Note: If a pump is not used in the modifiedMethod 6 train, caution should be exercised in adjusting the flow rate sinceover-pressurization of the impingers may cause leakage in the impinger train,resulting in positively biased results).

7.4 Zero and Calibration Drift Tests. Immediately preceding and following eachrun, or if adjustments are necessary for the measurement system during the run,repeat the sampling system bias check procedure described in Section 6.4. (Makeno adjustments to the measurement system until after the drift checks arecompleted.) Record the analyzer's responses on a form similar to Figure 6C—5.

7.4.1 If either the zero or upscale calibration value exceeds the samplingsystem bias specification, then the run is considered invalid. Repeat both theanalyzer calibration error check procedure (Section 6.3) and the sampling systembias check procedure (Section 6.4) before repeating the run.

7.4.2 If both the zero and upscale calibration values are within the samplingsystem bias specification, then use the average of the initial and final biascheck values to calculate the gas concentration for the run. If the zero orupscale calibration drift value exceeds the drift limits, based on the differencebetween the sampling system bias check responses immediately before and after therun, repeat both the analyzer calibration error check procedure (Section 6.3) andthe sampling system bias check procedure (Section 6.4) before conductingadditional runs.

7.5 Interference Check (if performed). After completing the run, record thefinal dry gas meter reading, meter temperature, and barometric pressure. Recoverand analyze the contents of the midget impingers, and determine the SO gas2

Cgas ' (C&Co)Cma

Cm&Co


Eq. 6C-1

concentration using the procedures of Method 6. (It is not necessary to analyzeEPA performance audit samples for Method 6.) Determine the average gasconcentration exhibited by the analyzer for the run. If the gas concentrationsprovided by the analyzer and the modified Method 6 differ by more than 7 percentof the modified Method 6 result, the run is invalidated.


The average gas effluent concentration is determined from the average gasconcentration displayed by the gas analyzer and is adjusted for the zero andupscale sampling system bias checks, as determined in accordance withSection 7.4. The average gas concentration displayed by the analyzer may bedetermined by integration of the area under the curve for chart recorders, or byaveraging all of the effluent measurements. Alternatively, the average may becalculated from measurements recorded at equally spaced intervals over the entireduration of the run. For sampling run durations of less than 1 hour,measurements at 1-minute intervals or a minimum of 30 measurements, whichever isless restrictive, shall be used. For sampling run durations greater than 1 hour,measurements at 2-minute intervals or a minimum of 96 measurements, whichever isless restrictive, shall be used. Calculate the effluent gas concentration usingEquation 6C-1.


Where:

C = Effluent gas concentration, dry basis, ppm.gas

C = Average gas concentration indicated by gas analyzer, dry basis, ppm.avg

C = Average of initial and final system calibration bias check responseso

for the zero gas, ppm.

C = Average of initial and final system calibration bias check responsesm

for the upscale calibration gas, ppm.

C = Actual concentration of the upscale calibration gas, ppm.ma

BIBLIOGRAPHY

1. Traceability Protocol for Establishing True Concentrations of Gases Usedfor Calibrations and Audits of Continuous Source Emission Monitors:Protocol Number 1. U. S. Environmental Protection Agency, QualityAssurance Division. Research Triangle Park, N.C. June 1978.

2. Westlin, Peter R. and John W. Brown. Methods for Collecting and AnalyzingGas Cylinder Samples. Source Evaluation Society Newsletter. 3(3):5-15.September 1978.


Figure 6C-1. Measurement System Schematic.

IceWaterBath

MidgetImpingers

Dry GasMeter

Rotameter

DryingTube

NeedleValve

3% H O(15 ml each)

2 2

ExcessSample Vent

SampleBy-pass Vent


Figure 6C-2. Interference Check Sampling Train.


Figure 6C-3. Analysis of Calibration Gases.

Date

Analytic Method Used

Gas Concentration (indicate units)

Zero Mid- High-a

Range range b c

Sample Run:

1

2

3

Average

Maximum Percent Deviation

Average must be less than 0.25 percent of span.a

Average must be 50 to 60 percent of span.b

Average must be 80 to 90 percent of span.c

Figure 6C-4. Analyzer Calibration Data.

Source Identification: Runs:

Test Personnel: Span:

Date:

Analyzer Calibration Data for Sampling

Cylinder Analyzer Absolute DifferenceValue Calibration Difference (percent

(indicate Response (indicate of span)units) (indicate units)

units)

Zero Gas

Mid-Range Gas

High-Range Gas

SystemCalibrationBias' SystemCal.Response&AnalyzerCal.ResponseSpan

×100

Drift' FinalSystemCal.Response&InitialSystemCal.ResponseSpan

×100


Figure 6C-5. System Calibration Bias and Drift Data.

Source Identification: Run Number:

Test Personnel: Span:

Date:

Initial Values Final Values

Analyzer System System System System DriftCalibrat Calibrat cal. bias Calibrat Cal. (percen

ion ion (percent ion Bias t ofResponse Response of span) Response (percen span)

t ofspan)

Zero Gas

Upscale Gas

444444444444444444444444444444444444444444444444444444444444444444444444444444

EMISSION MEASUREMENT TECHNICAL INFORMATION CENTERNSPS TEST METHOD

444444444444444444444444444444444444444444444444444444444444444444444444444444

Method 7E - Determination of Nitrogen Oxides Emissionsfrom Stationary Sources

(Instrumental Analyzer Procedure)


1.1 Applicability. This method is applicable to the determination ofnitrogen oxides (NO ) concentrations in emissions from stationary sourcesx

only when specified within the regulations.

1.2 Principle. A sample is continuously extracted from the effluentstream; a portion of the sample stream is conveyed to an instrumentalchemiluminescent analyzer for determination of NO concentration.x

Performance specifications and test procedures are provided to ensurereliable data.


Same as in Method 6C, Sections 2.1 and 2.2.

3. DEFINITIONS

3.1 Measurement System. The total equipment required for thedetermination of NO concentration. The measurement system consists of thex

following major subsystems:

3.1.1 Sample Interface, Gas Analyzer, and Data Recorder. Same as inMethod 6C, Sections 3.1.1, 3.1.2, and 3.1.3.

3.1.2 NO to NO Converter. A device that converts the nitrogen dioxide2

(NO ) in the sample gas to nitrogen oxide (NO).2

3.2 Span, Calibration Gas, Analyzer Calibration Error, Sampling SystemBias, Zero Drift, Calibration Drift, and Response Time. Same as in Method6C, Sections 3.2 through 3.8.

3.3 Interference Response. The output response of the measurement systemto a component in the sample gas, other than the gas component beingmeasured.


Same as in Method 6C, Sections 4.1 through 4.4.

————————————————————————————————————————————————————————————————————————

——————Prepared by Emission Measurement Branch EMTIC TM-007ETechnical Support Division, OAQPS, EPA May 7, 1990——————————————————————————————————————————————————————————————————————————————444444444444444444444444444444444444444444444444444444444444444444444444444444EMTIC TM-007E EMTIC NSPS TEST METHOD Page 2444444444444444444444444444444444444444444444444444444444444444444444444444444


5.1 Measurement System. Use any measurement system for NO that meets thex

specifications of this method. A schematic of an acceptable measurementsystem is shown in Figure 6C-1 of Method 6C. The essential components ofthe measurement system are described below:

5.1.1 Sample Probe, Sample Line, Calibration Valve Assembly, MoistureRemoval System, Particulate Filter, Sample Pump, Sample Flow Rate Control,Sample Gas Manifold, and Data Recorder. Same as in Method 6C, Sections5.1.1 through 5.1.9, and 5.1.11.

5.1.2 NO to NO Converter. That portion of the system that converts NO2 2

in the sample gas to NO. A NO to NO converter is not necessary if the NO2 2

portion of the exhaust gas is less than 5 percent of the total NOxconcentration.

5.1.3 NO Analyzer. An analyzer based on the principles ofx

chemiluminescence to determine continuously the NO concentration in thex

sample gas stream. The analyzer must meet the applicable performancespecifications of Section 4. A means of controlling the analyzer flow rateand a device for determining proper sample flow rate (e.g., precisionrotameter, pressure gauge downstream of all flow controls, etc.) must beprovided at the analyzer.

5.2 NO Calibration Gases. The calibration gases for the NO analyzerx x

shall be NO in N . Use four calibration gases as specified in Method 6C,2

Sections 5.3.1 through 5.3.3. Ambient air may be used for the zero gas.


Perform the following procedures before measurement of emissions (Section7).

6.1 Calibration Gas Concentration Verification. Same as in Method 6C,Section 6.1, except if calibration gas analysis is required, use Method 7,and change all 5 percent performance values to 10 percent (or 10 ppm,whichever is greater).

6.2 Interference Response. Conduct an interference response test of theanalyzer prior to its initial use in the field. Thereafter, recheck themeasurement system if changes are made in the instrumentation that couldalter the interference response (e.g., changes in the gas detector).Conduct the interference response in accordance with Section 5.4 of Method20.

6.3 Measurement System Preparation, Analyzer Calibration Error, ResponseTime, and Sample System Bias Check. Same as in Method 6C, Sections 6.2through 6.4.

6.4 NO to NO Conversion Efficiency. If the NO concentration within the2 2

sample stream is greater than 5 percent of the NO concentration, conductx

an NO to NO conversion efficiency test in accordance with Section 5.6 of2

Method 20.

444444444444444444444444444444444444444444444444444444444444444444444444444444EMTIC TM-007E EMTIC NSPS TEST METHOD Page 3444444444444444444444444444444444444444444444444444444444444444444444444444444


7.1 Selection of Sampling Site and Sampling Points. Select a measurementsite and sampling points using the same criteria that are applicable totests performed using Method 7.

7.2 Sample Collection. Position the sampling probe at the firstmeasurement point, and begin sampling at the same rate as used during theresponse time test. Maintain constant rate sampling (i.e., ±10 percent)during the entire run. The sampling time per run shall be the same as thetotal time required to perform a run using Method 7 plus twice the averagesystem response time. For each run, use only those measurements obtainedafter twice the response time of the measurement system has elapsed todetermine the average effluent concentration.

7.3 Zero and Calibration Drift Test. Same as in Method 6C, Section 7.4.


Same as in Method 6C, Section 8.

BIBLIOGRAPHY

Same as the bibliography of Method 6C.

EMISSION MEASUREMENT TECHNICAL INFORMATION CENTERNSPS TEST METHOD

Method 10 - Determination of Carbon Monoxide Emissionsfrom Stationary Sources

1. PRINCIPLE AND APPLICABILITY

1.1 Principle. An integrated or continuous gas sample isextracted from a sampling point and analyzed for carbon monoxide(CO) content using a Luft-type nondispersive infrared analyzer(NDIR) or equivalent.

1.2 Applicability. This method is applicable for thedetermination of carbon monoxide emissions from stationary sourcesonly when specified by the test procedures for determiningcompliance with new source performance standards. The testprocedure will indicate whether a continuous or an integratedsample is to be used.


2.1 Range. 0 to 1000 ppm.

2.2 Sensitivity. Minimum detectable concentration is 20 ppm fora 0- to 1000—ppm span.

3. INTERFERENCES

Any substance having a strong absorption of infrared energy willinterfere to some extent. For example, discrimination ratios forwater (H2O) and carbon dioxide (CO2) are 3.5 percent H2O per 7 ppmCO and 10 percent CO2 per 10 ppm CO, respectively, for devicesmeasuring in the 1500- to 3000-ppm range. For devices measuring inthe 0- to 100-ppm range, interference ratios can be as high as3.5 percent H2O per 25 ppm CO and 10 percent CO2 per 50 ppm CO. Theuse of silica gel and ascarite traps will alleviate the majorinterference problems. The measured gas volume must be correctedif these traps are used.

4. PRECISION AND ACCURACY

4.1 Precision. The precision of most NDIR analyzers isapproximately ±2 percent of span.

4.2 Accuracy. The accuracy of most NDIR analyzers isapproximately ±5 percent of span after calibration.

5. APPARATUS

Note: Mention of trade names or specific products does notconstitute endorsement by the Environmental Protection Agency.

EMTIC TM-010 EMTIC NSPS TEST METHOD Page 2

5.1 Continuous Sample (Figure 10-1).

5.1.1 Probe. Stainless steel or sheathed Pyrex glass, equippedwith a filter to remove particulate matter.

5.1.2 Air-Cooled Condenser or Equivalent. To remove any excessmoisture.

5.2 Integrated Sample (Figure 10-2).

5.2.1 Probe. Same as in Section 5.1.1.

5.2.2 Air-Cooled Condenser or Equivalent. Same as in Section5.1.2.

5.2.3 Valve. Needle valve, or equivalent, to adjust flow rate.

5.2.4 Pump. Leak-free diaphragm type, or equivalent, to transportgas.

5.2.5 Rate Meter. Rotameter, or equivalent, to measure a flowrange from0 to 1.0 liter per minute (0 to 0.035 cfm).

5.2.6 Flexible Bag. Tedlar, or equivalent, with a capacity of 60to 90 liters (2 to 3 ft3). Leak-test the bag in the laboratorybefore using by evacuating bag with a pump followed by a dry gasmeter. When evacuation is complete, there should be no flowthrough the meter.

5.2.7 Pitot Tube. Type S, or equivalent, attached to the probe sothat the sampling rate can be regulated proportional to the stackgas velocity when velocity is varying with time or a sampletraverse is conducted.

5.3 Analysis (Figure 10-3).

5.3.1 Carbon Monoxide Analyzer. Nondispersive infraredspectrometer, or equivalent. This instrument should bedemonstrated, preferably by the manufacturer, to meet or exceedmanufacturer's specifications and those described in this method.

5.3.2 Drying Tube. To contain approximately 200 g of silica gel.

5.3.3 Calibration Gas. Refer to Section 6.1.

5.3.4 Filter. As recommended by NDIR manufacturer.

5.3.5 CO2 Removal Tube. To contain approximately 500 g ofascarite.

5.3.6 Ice Water Bath. For ascarite and silica gel tubes.


5.3.7 Valve. Needle valve, or equivalent, to adjust flow rate.

5.3.8 Rate Meter. Rotameter, or equivalent, to measure gas flowrate of 0 to 1.0 liter/min (0 to 0.035 cfm) through NDIR.

5.3.9 Recorder (Optional). To provide permanent record of NDIRreadings.

6. REAGENTS

6.1 Calibration Gases. Known concentration of CO in nitrogen (N2)for instrument span, prepurified grade of N2 for zero, and twoadditional concentrations corresponding approximately to 60 percentand 30 percent of span. The span concentration shall not exceed1.5 times the applicable source performance standard. Thecalibration gases shall be certified by the manufacturer to bewithin 2 percent of the specified concentration.

6.2 Silica Gel. Indicating type, 6- to 16-mesh, dried at 175°C(347°F) for 2 hours.

6.3 Ascarite. Commercially available.

7. PROCEDURE

7.1 Sampling.

7.1.1 Continuous Sampling. Set up the equipment as shown inFigure 10-1 making sure all connections are leak free. Place theprobe in the stack at a sampling point, and purge the samplingline. Connect the analyzer, and begin drawing sample into theanalyzer. Allow 5 minutes for the system to stabilize, then recordthe analyzer reading as required by the test procedure. (SeeSections 7.2 and 8). CO2 content of the gas may be determined byusing the Method 3 integrated sampling procedure, or by weighingthe ascarite CO2 removal tube and computing CO2 concentration fromthe gas volume sampled and the weight gain of the tube.

7.1.2 Integrated Sampling. Evacuate the flexible bag. Set up theequipment as shown in Figure 10-2 with the bag disconnected. Placethe probe in the stack, and purge the sampling line. Connect thebag, making sure that all connections are leak free. Sample at arate proportional to the stack velocity. CO2 content of the gasmay be determined by using the Method 3 integrated sampleprocedures, or by weighing the ascarite CO2 concentration from thegas volume sampled and the weight gain of the tube.

7.2 CO Analysis. Assemble the apparatus as shown in Figure 10-3,calibrate the instrument, and perform other required operations asdescribed in Section 8. Purge analyzer with N2 prior tointroduction of each sample. Direct the sample stream through theinstrument for the test period, recording the readings. Check thezero and the span again after the test to assure that any drift or


Eq. 10-1

malfunction is detected. Record the sample data on Table 10-1.

8. CALIBRATION

Assemble the apparatus according to Figure 10-3. Generally aninstrument requires a warm-up period before stability is obtained.Follow the manufacturer's instructions for specific procedure.Allow a minimum time of 1 hour for warmup. During this time checkthe sample conditioning apparatus, i.e., filter, condenser, dryingtube, and CO2 removal tube, to ensure that each component is ingood operating condition. Zero and calibrate the instrumentaccording to the manufacturer's procedures using, respectively, N2and the calibration gases.

TABLE 10-1 - FIELD DATA

Location: Date:

Test: Operator:

Clock Time Rotameter Readingliters/min (cfm)

Comments

9. CALCULATION--CONCENTRATION OF CARBON MONOXIDE

Calculate the concentration of carbon monoxide in the stack usingEquation 10—1.

where:

CCO(stack) = Concentration of CO in stack, ppm by volume, drybasis.

CCO(NDIR) = Concentration of CO measured by NDIR analyzer, ppm byvolume, dry basis.

FCO2 = Volume fraction of CO2 in sample, i.e., percent CO2from Orsat analysis divided by 100.


Figure 10-1. Continuous Sampling Train.

10. ALTERNATIVE PROCEDURE--INTERFERENCE TRAP

The sample conditioning system described in Method 101A, Sections2.1.2 and 4.2, may be used as an alternative to the silica gel andascarite traps.

BIBLIOGRAPHY

1. McElroy, Frank. The Intertech NDIR-CO Analyzer. Presented at11th Methods Conference on Air Pollution, University ofCalifornia, Berkeley, CA. April 1, 1970.

2. Jacobs, M.B., et al. Continuous Determination of CarbonMonoxide Infrared Analyzer. J. Air Pollution ControlAssociation. 9(2):110-114. August 1959.

3. Mine Safety Appliance Co. MSA LIRA Infrared Gas and LiquidAnalyzer Instruction Book. Technical Products Division,Pittsburgh, PA.

4. Beckman Instruments, Inc. Models 215A, 315A, and 415A InfraredAnalyzers. Beckman Instructions 1635-B, Fullerton, CA.October 1967.

5. Intertech Corporation. Continuous CO Monitoring System, ModelA5611. Princeton, NJ.

6. Bendix Corp. UNOR Infrared Gas Analyzers. Ronceverte, WV.


Figure 10-2. Integrated Gas Sampling Train.

Figure 10-3. Analytical Equipment.


ADDENDA

A. Performance Specifications for NDIR Carbon Monoxide Analyzers.

TABLE A-1. Performance Specifications for NDIR CO Analyzers——————————————————————————————————————————————————————————————————

Range (minimum) 0-1000 ppm

Output (minimum) 0-10 mV

Minimum detectable sensitivity 20 ppm

Rise time, 90 percent (maximum) 30 seconds

Fall time, 90 percent (maximum) 30 seconds

Zero drift (maximum) 10% in 8 hours

Span drift (maximum) 10% in 8 hours

Precision (maximum) ±2% of full scale

Noise (maximum) ±1% of full scale

Linearity (maximum deviation) 2% of full scale

Interference rejection ratio CO2 - 1000:1; H2O -500:1——————————————————————————————————————————————————————————————————

B. Definitions of Performance Specifications.

1. Range - The minimum and maximum measurement limits.

2. Output - Electrical signal which is proportional to themeasurement; intended for connection to readout or dataprocessing devices. Usually expressed as millivolts ormilliamps full scale at a given impedance.

3. Full Scale - The maximum measuring limit for a given range.

4. Minimum Detectable Sensitivity - The smallest amount of inputconcentration that can be detected as the concentrationapproaches zero.

5. Accuracy - The degree of agreement between a measured value andthe true value; usually expressed as ± percent of full scale.

6. Time to 90 Percent Response - The time interval from a stepchange in the input concentration at the instrument inlet toa reading of 90 percent of the ultimate recorded concentration.

7. Rise Time (90 Percent) - The interval between initial response


time and time to 90 percent response after a step increase inthe inlet concentration.

8. Fall Time (90 Percent) - The interval between initial responsetime and time to 90 percent response after a step decrease inthe inlet concentration.

9. Zero Drift - The change in instrument output over a stated timeperiod, usually 24 hours, of unadjusted continuous operationwhen the input concentration is zero; usually expressed aspercent full scale.

10. Span Drift - The change in instrument output over a stated timeperiod, usually 24 hours, of unadjusted continuous operationwhen the input concentration is a stated upscale value; usuallyexpressed as percent full scale.

11. Precision - The degree of agreement between repeatedmeasurements of the same concentration, expressed as theaverage deviation of the single results from the mean.

12. Noise - Spontaneous deviations from a mean output not causedby input concentration changes.

13. Linearity - The maximum deviation between an actual instrumentreading and the reading predicted by a straight line drawnbetween upper and lower calibration points.

tpp kosovo b pollution mitigation

Documents

kosovo b thermal power

kosovo b units

new power plant development

required measures

proposed mitigation

eu directive

current ash disposal

pollution mitigation