1.Introduction
For many years, Lake Kivu's methane gas reserves only interested scientists because of the uniqueness of the lake water composition. Primarily because of the country's energy crisis, Rwanda has, in recent years, increased efforts to establish the first large scale exploitation of this dissolved methane gas.

Last year, the Rwandan Ministry of Infrastructure (MININFRA) granted the Industrial Development Group (IDG) Consortium of South Africa a concessionⁱ to extract 100 millionm m³_s/annum of methane from the lake for processing to produce much-needed liquid fuels. Appointed by the IDG, Added Value Engineering Consultants (Pty) Ltd (AVEC) recently completed a feasibility study for an integrated process scheme to extract the dissolved gases from Lake Kivu, recover the methane and the further processing thereof in a GTL facility to produce liquid fuel conforming to Euro-5 specifications. (This project was financed by the Industrial Development Corporation (IDC)).

2.Energy Crisis in Rwanda
In the recent years Rwanda has experienced an energy crisis - attributable to both electricity and fuel shortages. In 2004 electrical shortages were evident when the rate of economic development significantly exceeded electrical supply capacity. At this time, most of the electricity produced in Rwanda was generated through hydroelectric power. However, low rainfall and declining hydroelectric dam levels have aggravated electrical supply limitations. In general, the highly variable climate of Central Africa and man-made disturbances (primarily deforestation and agricultural development too close to water supplies) to the hydrological cycle resulted in these power systems becoming less reliableⁱⁱ.

In response to a rapidly growing problem that could impede economic development, thermal power plants were installed, most notably Jabana II. Currently, 40.3 MW of electrical power generated in Rwanda is from thermal power plants that use diesel as the fuel sourceⁱⁱⁱ.

In addition to electricity shortages, Rwanda has also been hit by fuel shortages with some petrol stations even running dry. The country is dependent on importing fuel and the routes are vulnerable to interruptions of supply. Moreover, Rwanda is extremely vulnerable to price fluctuations in global trading prices.

Petroleum products account for 11% of the primary energy use with 80-90% of this consumed as diesel or petrol for transportation^iv. The use of petroleum products is expected to grow at an annual average rate of 3-5%/annum^v. The combination of high prices of fuel on the international market and the very long and expensive supply routes result in high prices for petroleum products in Rwanda.

Figure 1 - Lake Kivu - Proposed gas extraction site for CH4 recovery

3.Methane Resource in Lake Kivu
As part of the western branch of the African rift valley, Lake Kivu was formed about 500,000 years ago in the course of the emergence of the Virungu volcano chain. The lake reaches depths of more than 450 m, comprises part of the border between Rwanda and the Democratic Republic of Congo (DRC) and covers a surface of about 2,400 km², distributed almost equally between the two countries. The lake is unique because its deeper waters contain an enormous quantity of dissolved gases: estimates are 250 billion m³_s of CO2, 60-65 billion m³_s of CH₄^vi and lesser amounts of N₂ and H₂S.

Although not fully understood, the lake's CO₂ seems to be partly of volcanic origin as well as formed by the decomposition and fermentation of organic material by anaerobic bacteria accumulating in the bottom sediment. CH₄ is generated by fermentation processes and by the reduction of volcanic CO₂ by the same bacteria. Between 100-150 million m³_s of CH₄ is generated annually in the lake.

CH₄ is available in Lake Kivu and if this gas can be extracted economically, it can be used as a source of energy. Based on consumption predictions, the stipulated CH₄ extraction volume from the lake can produce between 500 and 860 bbl/day of liquid fuel for 40+ years. This would cater for between 11 and 19% of the current Rwandan national fuel requirements.

4.Gas Extraction - CH4 Recovery
The study was carried out on the basis of using Methane Hydrates Limited (MHL) technology because AVEC was associated with the process engineering of this technology as well as the operation of the test rig in November 2003. The MHL technology provides a more modern solution than its counterparts and complies with the recommendations of the International Expert Working Group for Lake Kivu Development regarding the extraction of dissolved gases from the lake^vii. The technology is based on a multi-stage gas separation and washing system from as deep as 60m, using a stimulated, controlled limnic eruption of dissolved gas.

The MHL gas gathering platforms are of a modular design to allow for the capacity of the gas gathering process to be easily increased as well as to simplify the logistics and commissioning operations.

Each gas extraction module comprises a fully submerged processing plant which has to extraction stages in series to separate methane gas. Each stage consists of two separation vessels (gas-liquid disengagement), a wash tower (CH₄ - CO₂ separation) and associated piping systems. Two-stage, submerged extraction allows a higher recovery of pressurized and better quality CH₄ than from a single-stage plant. The gas extraction plant requires services such as control systems, wash water and compressed air to carry out normal start-up and operations. These services are housed on a floating platform anchored above the gas extraction plant, from which the modules are operated and controlled. The entire off-shore platform is located where the water depth is greater than 370m.

The recovered gas is piped directly to shore and this piping is suspended 10m below the lake surface via a buoy and counter-weight system, with anchor lines.

AVEC had previously created simulation models using Aspen software and these models were based on earlier lake data/information. For this most recent project it was necessary to create a new model ion ChemCAD V6.3 which was based on previous work, but used the most recent dissolved gas data. In addition, improvements to the latest MHL design and published data for gas solubility in water were taken into consideration. The thermodynamic methodology used gave an acceptable match with published data for CH₄, CO₂, N₂ and H₂S solubility in water over the expected operating pressure and temperature range of the gas extraction modules.

Various design aspects on the lower-pressure extraction stage were optimized for methane yield and purity at a sufficient operating pressure to ensure that the gas could be routed to the on-shore GTL processing facility.

5.GTL Facility
In general, the various GTL commercial offerings follow two broad chemical "paths"; a direct conversion to liquids from gas and an indirect conversion via the production of synthesis gas.

Indirect conversion refers to converting natural gas to synthesis gas by reforming and the synthesis gas is used to produce synthetic crude or other intermediate products. The indirect conversion processes require synthetic crude upgrading to synthetic fuel.

Direct conversion of the CH4 in natural gas produces intermediate or final products such as methanol, formaldehyde, ethyl alcohol, petrol, diesel and jet fuel via oxidation without intermittent synthesis gas production. This technology can produce synthetic fuel which can be directly used in motor vehicles and needs no further processing.

The choice of GTL process for further development depended on the following key points :

• Technology supplier - GTL technology is a relatively young technology, its development is still largely in transition between development and commercialization and the supplier technology offering might not necessarily be suited to small-scale production or even suitable for the specific gas composition.

• Capital costs - Because the expected output is only a fraction of a typical refinery production, there is no benefit from economies of scale. A capital cost which is too high would obviously render the project infeasible.

• Technology offering - This includes technical aspects such as the level of complexity of the technology, extent of intermediate products, utility requirements, operability, waste streams, whether the technology produces synthetic crude or final products, flexibility and suitability for the recovered Lake Kivu CH₄ gas.

• Product type and yield - The value and market of the products as well as efficiency and yield affect the project viability.

• Land requirements - The area is generally mountainous and undulating thus requiring a high degree of site preparation. Therefore, technology offerings with smaller footprints are preferable.

Taking the above into consideration, the most suitable technology offering for a GTL facility in Rwanda is using proprietary technology for homogenous, direct oxidation of CH₄ into methanol and synthesis of high-grade petrol. Given the gas composition, only petrol can be produced with direct oxidation. The process was optimized to utilize all the waste thermal energy from the reactors for internal electricity generation, ensuring an energy efficient plant.

6.Project Financials
The proposed gas extraction GTL project will produce approximately 820 bbl/day of petrol based on 8,000 h/annum operation. The petrol conforms to Euro-5 specification and meets the minimum RON of 95.

The CAPEX of the proposed project comprises the following main items:

• Gas Extraction Platform

• GTL Facility

• Storage Depot (tank farm)

• Land Preparation Costs

The financial model inputs and sensitivities were tested using Monte Carlo simulations. The model showed that a reasonable return on investment could be achieved, based on favourable negotiations with the Government of Rwanda (GOR).

As an alternative to liquid fuel production, the amount of CH₄ gas extracted is the equivalent of a 50 MW electrical power plant. A separate financial model was set up on the assumption of only electrical power being produced and sold to the Rwandan national power grid on a take-or-pay basis. The results of this model showed that the returns for liquid fuel production are in the same as for electrical power production.

7.Conclusion
There are both technical and financial risks to this project. The technical risks are to be mitigated by means of gas extraction demonstration module and collaboration with the technology suppliers. Financial risks are related to oil trading prices and this raises the need for negotiation with the Rwandan Ministry of Trade and Industry (MINICOM) in respect of pricing strategy. The project financials show that a reasonable return on investment can be achieved based on the negotiation of favourable investment incentives. At this stage, both the technical and financial risks are deemed manageable and the decision to proceed to the next phase of the project is expected in early 2011.

Implementation of this project will provide employment and wealth creation opportunities in the Gisenyi district and will also raise both the technical and industrial profile of Rwanda. Investment in this project will not only provide stabilization of Rwanda's petrol supply by up to 60%, but will also have the additional effect of reducing an uncontrolled gas release from the lake which could have potentially catastrophic consequences.