Tạp chí của tập đoàn dầu khí quốc gia Việt Nam - Petrovietnam - Số 10 - 2011 pdf

The primary goals of the project were an analysis of the hydrocarbon potential of the Song Hong basin by application of sequence stratigraphy and basin modelling, and geo-scientifi c tra

Petrovietnam and ONGC Videsh Limited signed an

agreement of cooperation

In India, both sides agreed to further enhance

high-level visits and meetings between the two countries At

the Vietnam - India Business Forum, President Truong Tan

Sang noted that there still remains considerable potential

for cooperation and agreed to strongly enhance the

comprehensive strategic partnership between the two

countries based on key pillars of politic, defense, security,

economic, cultural and human resource development

cooperation

On Oct.12, President Truong Tan Sang and Prime

Minister of the Republic of India Mamohan Singh

witnessed the signing of the agreement between

Petrovietnam and ONGC Videsh Limited (ONGC-VL)

The agreement was signed by Dr.Sc Phung Dinh Thuc,

Chairman of Petrovietnam and Mr D.K Sarraf, CEO and

Managing Director, ONGC-VL The agreement is intended for developing long term cooperation in oil and gas industry and shall be in force and eff ect for three years.Two sides will exchange information on the petroleum industry, exchange of working visits of authorities and specialists in various domains of the petroleum industry, new investments, expansion and operations of oil and gas exploration and production including refi ning, transportation and supply in Vietnam, India, and third countries according to the laws and regulations of their countries

In Sri Lanka, President Truong Tan Sang and Sri Lanka’s leaders witnessed the signing of memoranda

of understanding on bilateral political consultations, machinery manufacturing, investment promotion, cooperation in education, fi nance, defense and oil and gas…

Petrovietnam’s leaders accompanied President Truong Tan Sang visited the Asian countries

President Truong Tan Sang and Prime Minister of the Republic of India Mamohan Singh witnessed the signing of the agreement between

Petrovietnam and ONGC Videsh Limited (ONGC-VL) Photo: PVN

The mission of The Vietnam National Oil and Gas Group (Petrovietnam) led by Dr.Sc Phung Dinh Thuc - Chairman

of Petrovietnam accompanied President Truong Tan Sang, has visited and worked in India and Sri Lanka from Oct 11

to Oct 15 to promote cooperation in the fi eld of energy Before that, Dr Do Van Hau - President & CEO of Petrovietnam accompanied President Truong Tan Sang had visited and worked in Singapore and Malaysia from Sept 26 to Sept 30

PV Drilling V - the fi rst TAD

rig of Vietnam

On Sept 26, President

Truong Tan Sang, Deputy

Prime Minister Nguyen

Thien Nhan and a high-level

delegation visited the fi rst

TAD rig of Vietnam (named

PV Drilling V that was

constructed at the Keppel

FELS shipyard - Singapore)

According to Dr Do Van

Hau - President & CEO of

Petrovietnam, this is a large

project with investment

worth over VND 5,000 billion,

is designed with modern

technology and will be fi rst

used in Vietnam

Developed by Keppel Off shore & Marine’s (Keppel

O&M) Deepwater Technology Group (DTG), PV Drilling

V design has revolutionised the way in which drilling

tenders work, allowing them to be deployed next to

deepwater fl oating platforms for the fi rst time This ground

breaking series, with their enhanced capabilities, can

operate and maintain station position in deep waters of

up to 5,000ft alongside Spars and Tension Leg Platforms

Highly suitable for off shore Vietnam’s harsh environment,

PV Drilling V features an 8-point mooring system, which

allows the rig to maintain its position amidst currents of

up to 6 knots Under the plan, PV Drilling V will drill the

fi rst contract for the drilling campaign of the Bien Dong

Petroleum Operating Company (Bien Dong POC)

At the same time, a delegation comprising leaders

of member units of Petrovietnam (PVEP, PV Drilling,

PTSC, PV GAS…) led by the Dr Do Van Hau attended the

Vietnam - Singapore Business Forum be held on Sept 26

Establish a Joint Venture Company investing in a FSO

During his visit to Malaysia, on Sept 29, President

Truong Tan Sang visited Malaysia’s National Oil and Gas

group (Petronas), in which he appreciated the strict,

eff ective and practical cooperation between Petrovietnam

and Petronas He spoke highly of the close, eff ective

and practical co-operation between Petrovietnam and

Petronas over the past 20 years, which brought great economic benefi ts to each country The Vietnamese State leader suggested that the two groups strengthen co-operation, especially in exploration and production in new oil and gas fi elds in both Vietnam and Malaysia On the occasion, President Sang received USD 50,000 from Petronas for Vietnamese Agent Orange victims

On the same day, Petrovietnam Technical Services Corporation (PTSC) and Yinson Holdings Berhad signed

a joint-venture company investing in a Floating Storage

& Offl oading (FSO) Facility The signing was held at the Malaysia - Vietnam Business Forum under the witnessed

of President Truong Tan Sang, Deputy Prime Minister Nguyen Thien Nhan, Dr Do Van Hau - President & CEO of Petrovietnam and offi cials from the Malaysian Government

FSO “PTSC Bien Dong 01” is the fi rst condensate FSO which has the break-up mooring & subsea systems This project is totally performed by PTSC, ranging from engineering, new-building, to pre-commissioning, hook-

up, off shore installation, and operation and maintenance during the period of some 20 years The total investment

of the project is around 150 million USD

PV Drilling held a naming ceremony for PV Drilling V on Oct 1 Photo: PVD

Viet Ha

Gross revenue was now

equivalent to 98 percent of

Petrovietnam’s yearly target and would

likely reach its goal later this month

Petrovietnam estimates to gain a total

revenue of VND 678 trillion for the whole

year, surpassing its initial target set for

this year by VND 178 trillion The Group’s

contribution to the State budget totalled

VND 116 trillion so far this year, a surge of

14 percent over its target for the whole

year and 32 percent over the same

period last year

The Group also succeeded in

accelerating recoverable oil and gas

reserves in 2011, following the discovery

of three new oil and gas fi elds and

the signing of fi ve new oil and gas

agreements, including four domestic

deals and one overseas contract In the past nine months,

with oil prices ranging from USD 100 - 115 per barrel,

Petrovietnam’s oil sales reached USD 9.3 billion, a

year-on-year increase of 49 percent, beating the Group’s yearly

plan by 9 percent

Meanwhile, the oil and gas services contributed

28.6 percent to the Group’s total revenue, worth VND

141 trillion, up 16 percent against the same period

last year Petrovietnam’s profi t during the January to

September period reached VND 27 trillion Petrovietnam

had disbursed nearly VND 60 trillion and postponed and

rescheduled 56 projects worth VND 7.25 trillion In the

last quarter, Petrovietnam aimed to begin the tapping

of three new gas and oil fi elds, including two in Viet Nam

and one in Malaysia

Dr Do Van Hau, President & CEO of Petrovietnam said

that to help ensure national energy and food security,

the Group shall maintain safe operation of the oil and gas pipeline system, the Dung Quat Refi nery, the Power Plants Ca Mau 1 & 2, Nhon Trach 1 & 2 and the Phu My fertilizer plant The Group contributed 10.23 billion kWh to the national grid so far this year, up 3.5 percent year-on-year and equivalent to 83 percent of its yearly goal “Petrovietnam will aggressively carry out its power projects with the target of contributing up to 25 percent

of the country’s energy production by 2015”, Dr Hau said

Dr Do Van Hau also said the Group had co-operated with foreign partners in the exploration and production of oil and gas overseas, with projects deployed in 13 countries, including Russia, Nigeria, Malaysia and Venezuela With regard to the Group’s divestment from non-core business lines, Hau confi rmed Petrovietnam would continue its on-going investments in core businesses including oil and gas exploration and production, gas and power industry and oil and gas services, and re-consider other investments in minor industries

Petrovietnam earned total revenue of VND 493 trillion

In the fi rst nine months of 2011:

Mai Huong

On Oct 11, The Vietnam National Oil and Gas Group (Petrovietnam) held a press conference to announce business result of the third quater of 2011 Petrovietnam earned total revenue of VND 493 trillion (USD 23.48 billion) in the fi rst nine months of the year, an increase of 47 percent over the same period last year The Group had completed three business goals for the year ahead of schedule, which are revenue, contributions to the State budget and acceleration

of oil and gas reserves.

Dr Do Van Hau, President & CEO of Petrovietnam chaired the third quater press

conference Photo: Duc Chinh

On September 29, Prime Minister Nguyen Tan

Dung attended the Netherlands - Vietnam

Energy, Oil and Gas Cooperation Forum in the Netherlands

At the forum, the two sides touched upon issues helping

to ensure Vietnam’s energy security, including advanced

energy technologies, bilateral energy cooperation

mechanisms, facilitation of environmentally friendly

energy usage, development of energy-saving programs,

and sharing of information on oil supplies…

Speaking at the forum, Prime Minister Nguyen Tan

Dung stressed that Vietnam considers energy a key

economic sector which contributes to realising the

country’s Millennium Development Goals (MDGs) and

shortens the national industrialization and modernization

process He said that Vietnam’s total energy demand

for commercial purposes is forecast to increase by over

7 percent each year in the 2010 - 2025 period, with

electricity demand rising some 15% annually

The Vietnamese Government is actively taking various

measures to ensure energy security, such as putting forth

suitable policies to attract investment, especially foreign

investment in the energy sector, and create favorable conditions for implementing energy projects in the country Vietnam is also diversifying energy sources, giving priority to the development of renewable energy, liquefi ed natural gas-fuelled power and atomic energy Meanwhile, it is fostering energy-saving programs, building up energy price roadmaps

Dutch Deputy Prime Minister Maxime Verhagen said the Netherlands - Vietnam cooperative ties have developed vigorously in various spheres, including energy The Netherlands will continue to closely cooperate with Vietnam in technology transfer, scientifi c research and oil and gas training Deputy Prime Minister added that it will also encourage and support the two countries’ enterprises, especially those operating in the fi elds of energy and oil and gas

Vietnam and Netherlands are expected to further their energy cooperation after a series of agreements were signed At the forum, Prime Minister Nguyen Tan Dung witnessed the signing of memoranda of understanding (MoU) on energy cooperation along with an agreement

Boost energy cooperation

Vietnam and Netherlands are expected to further their energy cooperation after a series of agreements were signed Photo: Nhat Bac

Vietnam, Netherlands, Uzbekistan:

on oil and gas cooperation between Vietnam National Oil

and Gas Group (Petrovietnam) and the Netherland’s Royal

Dutch Shell Group A letter of intent between Petrovietnam

and companies and organisations which join in Advanced

Energy Technologies International (AETIN) programme

on promoting cooperation - investment - technology

transfer and training, and an MoU on training cooperation

between the Petrovietnam University (PVU) and the Delft

University of Technology were also signed

On the same day, Prime Minister Nguyen Tan Dung

received the leaders of the Royal Dutch Shell Group,

affi rming that the Vietnamese Government will support

and create favourable conditions for the Group to continue

doing business in Vietnam Prime Minister Nguyen Tan

Dung emphasized that, during 20 years of operations in

Vietnam, Shell has made great contributions to Vietnam’s

oil and gas sector, helping meet the country’s increasing

energy demand Prime Minister Nguyen Tan Dung

suggested the Group attaches importance to developing

new, safe and environmentally friendly energy resources

as it is a focus of the Vietnamese Government’s attention

to ensure the country’s sustainable growth

After the successful visit to Netherlands, Prime

Minister Nguyen Tan Dung continued his trip to Uzbekistan

from Oct 2 to Oct 3 Prime Minister Nguyen Tan Dung

discussed with his Uzbekistani counterpart, Shavkat Mirziyoyev, a wide range of measures to boost the multifaceted cooperation between the two countries, focusing on economic trade and investment ties in Tashkent on October 3 Both Prime Ministers spoke highly of the eff ective oil and gas exploration and production of Vietnam National Oil and Gas Group (Petrovietnam) in Uzbekistan, stressing that oil and gas is a spearhead cooperative area between the two nations

Prime Minister Nguyen Tan Dung proposed Uzbekistan soon fi nalise procedures for the signing of a contract between Petrovietnam and the Uzbekistani Oil and Gas National Holding Company in Bukhara Khiva, Uzbekistan The Vietnamese Government is ready to facilitate the Uzbekistani Oil and Gas National Holding Company’s involvement in oil and gas exploration and exploitation projects on Vietnam’s continental shelf

For his part, Prime Minister Shavkat Mirziyoyev affirmed that the Uzbekistani Government always supports and creates favourable conditions for oil and gas projects with Vietnam, which, he said, is an important cooperative fi eld The two PMs assigned leaders of ministries and branches to meet and implement specifi c cooperative measures

On the same day, Minister of Industry and Trade Vu Huy Hoang and Minister of Foreign Economic Relations, Investments and Trade of the Republic of Uzbekistan witnessed the signing of memoranda of understanding (MoU) on cooperation in scientifi c research (joint study

on the issues of the petroleum industry in Vietnam and Uzbekistan which are of interests for both parties), training activities to share expertise and technology transfer… between Vietnam Petroleum Institute (VPI) and Uzbekistan’s Institute of Geology and Exploration of Oil and Gas Fields; cooperative agreement between PV Engineering and a member unit of Uzbekneftegaz; a cooperative agreement between PVI Holdings and Alfa Invest…

Huy Hoang

Minister of Industry and Trade Vu Huy Hoang and Minister of Foreign Economic Relations,

Investments and Trade of the Republic of Uzbekistan witnessed the signing of MoU between

Vietnam Petroleum Institute and Uzbekistan’s Institute of Geology and Exploration of Oil and

Gas Fields Photo: VPI

The 2011 Science - Technology

Conference titled “The 50th

Anniversary of the Traditional Day

of the Vietnam Petroleum Industry:

Achievements and Development

Strategy” will take place on Oct 28,

2011 on the 4th fl oor, VPI Tower, 173

Trung Kinh, Hanoi The Conference

is a forum for introducing the

achievements and the development

orientation of the core business of

Petrovietnam including exploration

and production, refi neries and

petro-chemistry, gas and power industries,

advanced petroleum technical

services as well as EOR solutions,

maintaining petroleum productivity

and oversea investments

In the Conference, 13 reports will

present the successful achievements

of the last 50 years and major

development orientations for

successfully realizing the Strategy

for Accelerating the Developments

to 2015 and towards 2025 of

Petrovietnam including the topics

such as exploration and production

in Vietnam and overseas, refi neries,

processing and petro chemistry,

gas and power industries, market

development for petroleum products,

advanced petroleum technical

services, infrastructure construction,

The 50th Anniversary of the Traditional Day of the Vietnam Petroleum Industry:

Achievements and Development Strategy

Scientifi c and technological activities always play an important role in the development strategy of the Vietnam National Oil and Gas Group The 2011 Science - Technology Conference titled “The 50 th Anniversary of the Traditional Day of the Vietnam Petroleum Industry: Achievements and Development Strategy” is a forum to introduce achievements and development orientation of the core business of Petrovietnam Besides the Conference, the 9 th

“Vietnam Oil and Gas Expo 2011” will be one of the meaningful activities marking the contribution and determination

of the Vietnam petroleum industry in fulfi lling the aspirations of respectful President Ho Chi Minh.

Fig.2 Revenue and contribution to state budget (source: PVN) Fig.1 Oil and gas production and crude oll export (source: PVN)

Ngoc Linh

scientifi c - technological and human resource development

measures

Scientifi c and technological activities always play

an important role in the development strategy of the

Vietnam National Oil and Gas Group Petrovietnam’s policy

is to intensively invest on scientifi c and technological

R&D, and innovation As a result, The Vietnam Petroleum

Science and Technology has remarkably contributed

to Petrovietnam’s achievements sofar on its 50 years

of development Among the commendable scientifi c

and technological achievements, the scientifi c complex

on prospecting discovering and eff ectively exploiting

oil reservoirs in the pre-tertiary granitoid basement of

Cuu Long basin in the continental shelf of Vietnam is an

outstanding one With a series of the patents granted by

the National Offi ce of Intellectual Property of Vietnam

(NOIP) and the Intellectual Property Agency of the Russian

Federation, this scientifi c complex had changed the

traditional concept on hydrocarbon exploration, enriched

the petroleum system theory and made a turning-point

for the development of the Vietnam Petroleum Industry

Besides the above mentioned Conference, the 9th

“Vietnam Oil and Gas Expo 2011” will be held on Oct

27 - Oct 29, 2011 in Giang Vo Exhibition Center (Hanoi)

Participating in the Expo are 5 on pavilions of France, Russia,

Singapore, Norway and Vietnam and about 190 exhibitors

from 19 countries and territories around the world The 9th

Vietnam Oil and Gas Expo will be the largest one so far The

expo takes place at a critical time when the whole industry

is implementing the strategy of accelerating developments,

making great eff orts in exploration and production in the

overall continental shelf of Vietnam as well as in other

parts of the world Advanced technologies, equipment and technical services that are displayed at the expo by diff erent international and local groups and companies can

be applied to the implementation of investment projects in Vietnam in the coming years

Compared to the previous ones, the expo this year will appeal even more to the interests of international and local investors and companies in the Vietnam petroleum industry This is represented by the increasing number of exhibitors and larger space required by national pavilions and companies; particularly by the presence of some big companies marking their return after a period of absence Being a biennial event from 1991 to date, Vietnam Oil and Gas EXPO is growing in size and gradually becoming

an exhibition brand name in the region, attracting international and domestic oil and gas companies, equipment manufacturers, services suppliers and product retails in the petroleum and energy sector

With the theme “Vietnam petroleum industry - 50 years

of fulfi lling the aspiration of President Ho Chi Minh” the booth of Petrovietnam - the mother company - focuses on milestones in all its operations during the last 50 years of development, particularly the exploration and production success in fractured basement, one of the meaningful achievements of Vietnam’s petroleum industry; adding 17 separated booths of member corporations display their most outstanding successes on their development The expo will be one of the meaningful activities marking the contribution and determination of the Vietnam petroleum industry in fulfi lling the aspiration of respectful President

Ho Chi Minh

Dung Quat Refi nery Photo: BSR

I Introduction

This paper briefl y describes the history of the

long-lasting co-operation between VPI and GEUS It presents

highlights of the geo-scientifi c results of this outstanding

co-operation and provides a comprehensive reference list

of written communications that document the project activities and results that have contributed to the success

of the co-operation

The initiation of the co-operation between VPI and GEUS can be traced back to a fact-fi nding mission to

ENRECAProjectPhaseI,II,IIIfrom2001to2011:contributions to geological understanding and petroleum potential assessmentofsedimentarybasinsinVietnam

N.A Duc, N.T Huyen, N.T Dau, L.C Mai

Vietnam Petroleum Institute

L.H Nielsen, I Abatzis, M.B.W Fyhn, H.I Petersen, J.A Bojesen-Koefoed

Geological Survey of Denmark and Greenland

Abstract

The Vietnam Petroleum Institute (VPI) and the Geological Survey of Denmark and Greenland (GEUS) have developed co-operation within geo-scientifi c research and institutional capacity building since 1995 Two projects have focused on the hydrocarbon potential of the Cenozoic Song Hong basin The co-operation is continuing within the frame of a new, long-term project, started in 2001, with the aim of enhancement of research capacity (ENRECA) within basin analysis and modelling in Vietnam The fi rst phase of the project is directed towards the Phu Khanh basin off shore central Vietnam The second phase is focused on the Malay - Tho Chu and Phu Quoc basins in the Southwestern continental shelf of Vietnam The third phase, directed to the Northern part of Song Hong basin and onshore Hanoi trough, is at the preparation stage Many fruitful results of the co-operation are proving benefi cial to our twinning institutions in terms of scientifi c research, technology transfer, advanced training and post-graduate education.

The origin and the petroleum potential of the Song Hong, Phu Khanh, Malay - Tho Chu and the Phu Quoc basins have been assessed by the group of Vietnamese and Danish geoscientists and new models are proposed A technology transfer, project integrated with the geo-scientifi c projects, has provided VPI with modern facilities for interactive interpretation of geophysical and geological data Extensive training of Vietnamese geo-scientists in system management, seismic and logging data interpretation, geochemical modelling, basin modelling and paleontological interpretation has been carried out via in-house training at VPI and “On-the-job-training” programs at GEUS The facilities and their great potential for advanced and eff ective interpretation have been demonstrated to a broad audience through a number of workshops Co-operation between VPI and GEUS has signifi cantly enhanced the human resources of VPI; four VPI project members have received their master certifi cation and four VPI geoscientists achieved their doctoral certifi cation during implementation of cooperative research projects

GEUS in 1993 by CCOP-representatives (Co-ordinating

Committee for Geoscience Programmes in East and

South East Asia) A project proposal from GEUS based on

this initial contact was later approved and funded by the

Danish Energy Authority (DEA), and the fi rst joint project

between VPI and GEUS was started in 1995 The primary

goals of the project were an analysis of the hydrocarbon

potential of the Song Hong basin by application of

sequence stratigraphy and basin modelling, and

geo-scientifi c training of the involved VPI staff Needs for

modern interpretation facilities were identifi ed during

the progress of the project, and support for a technology

transfer project was obtained from Danida (Danish

International Development Assistance) These initial

projects were accompanied by CCOP workshops on

interactive geo-scientifi c interpretations and mapping

held at VPI, which was, thereby, established as a regional

training centre Several VPI geo-scientists participated

also in “On-the-Job-Training” (OJT) programmes at GEUS

The Royal Danish Embassy in Hanoi and the Fellowship

Programme at Danida funded these activities

Based on the successful conclusion of the fi rst

geo-scientifi c project, DEA approved the funding for a second

phase of the project focused on the petroleum systems

of the Song Hong basin Parallel with this project several

geo-scientists from VPI participated in “OJT” programmes

at GEUS, and CCOP-workshops conducted by GEUS

instructors were held at VPI with the participation of

geo-scientists from the CCOP-member countries The second

geo-scientifi c project was concluded in 1998 (Andersen &

Dien 1998; Andersen 1999)

The implementation of these projects set the

foundations for further co-operation between the two

sister research institutes, and revealed the fundamental

need for enhancement of the research skills of VPI

geo-scientists as part of an institutional capacity building

at VPI It was therefore decided by VPI and GEUS to

investigate the possibilities for carrying out a long-term

project

Already in November 1996 when the Board of Danida

visited VPI, members of the Board draw the attention of

VPI and GEUS to the ENRECA-programme (Enhancement

of Research Capacity) at Danida In 1999, VPI and GEUS

decided to investigate the possibility of continuing and

expanding the fruitful co-operation between the two

“sister-organisations” within the frame of this programme Danida

approved an initiative grant, and fact-fi nding missions conducted in Hanoi and Copenhagen concluded with the submission of an application for a long-term geo-scientifi c research capacity building project in late 2000 The project proposal entitled “Integrated analysis and modelling of geological basins in Vietnam and an assessment of their hydrocarbon potential” was submitted to Danida The general development objective of the project is to improve the capabilities of researchers and research institutions in Vietnam to analyse and model their geological basins in

an eff ective, integrated and comprehensive way in order

to establish a good geo-scientifi c background on which political and/or technical decisions concerning exploration and exploitation of hydrocarbons can be based Each phase

of the project has a duration of three years, while the term co-operation can be extended for up to four phases The fi rst two phases were approved in 2001 and 2005 focusing on Phu Khanh basin, Malay - Tho Chu and Phu Quoc basins respectively The third phase directed to Northern part of Song Hong basin and onshore Hanoi trough is at the

long-preparation stage

A joint research group from the Geological Survey

of Denmark and Greenland (GEUS), Vietnam Petroleum Institute (VPI) and universities in Denmark and Vietnam has since 1995 worked to assess the geology and petroleum potential of the Vietnamese basins based on analysis of vast amounts of seismic and gravimetric data, basin modelling and analysis of well data, source rocks and information from onshore outcrops, core holes, and seep oils This study throws light on the structural and stratigraphic development of the Vietnamese margin addressing the regional tectonic mechanism driving the evolution of the Song Hong, the Phu Khanh, the Malay - Tho Chu and the Phu Quoc basins The hydrocarbon potential of these underexplored basins outlining the margin is similarly addressed

II Results

The results obtained during the co-operation between VPI and GEUS since 1995 span numerous areas including, among others, scientifi c results, transferred technology, provision of tailor-made advanced training courses, fi eld-trips, and post-graduate education The joint studies resulted in new scientifi c knowledge and to the establishment of close personal friendships among the researchers in association with mutually valuable experiences

1 Geo-scientifi c results

1.1 Sedimentary depositional development

The Vietnamese margin is fl oored by a number

of sedimentary basins with a considerable petroleum

potential (Fig 1) Most of these basins are in an early state

of exploration and the overall understanding of their

development is still limited

Outcrop studies of Cretaceous strata on Phu Quoc

island and onshore Cambodia complemented by analysis

of the fully cored 500m deep ENRECA-2 well on Phu Quoc

island indicate a prevalence of sandstones in the Phu Quoc

basin (Fyhn et al 2010) Alluvial sandstones with an average

of ca 10% rhyolite-dominated lithic fragments make

up the primary content of the up to ca 4km thick Upper Jurassic - Cretaceous sediments fi lling the basin Only

a few thin shallow marine sandstone beds have been encountered in the otherwise terrestrial succession The sandstone dominated succession intercalates with subordinate alluvial plain and lacustrine silt- and mudstone intervals Coal fragments are abundant at specifi c stratigraphic levels, but do not posses any source potential The Cenozoic rift basins along the Vietnamese margin are fi lled by thick and varied sedimentary successions Seismic facies analysis supported by well data indicates the presence of a broad range of sediment types in the basins, signifying changing Cenozoic depositional systems in the region (Figs 2, 3)

Non-marine depositional environments with estuarine interludes prevailed during the Palaeogene syn-rift period due

to the immature development

of the East Sea The syn-rift succession is therefore dominated

by alluvial, fl uvial and lacustrine deposits, of which carbonaceous lake successions and humic coals constitute the primary source-rock type in the area Restricted marine incursions occur within the syn-rift interval suggesting periodic connections with either the proto-East Sea or the youngest East Sea that initiated during the early Oligocene

A pronounced transgression occurred during the earliest Miocene as the East Sea expanded and gradually approached the Vietnamese margin Widespread subaerially exposed areas became inundated, which promoted carbonate platform growth from the central Song Hong basin to the central Phu Khanh basin, while terrigenous alluvial and shallow marine deposition prevailed in the Northern Song Hong, Southern Phu

Fig 1 Map showing the sedimentary basins on the continental shelf of Vietnam

Hoang Sa Basin

Phu Quoc Basin

Tu Vung May Basin

Chinh-Truong Sa Basin

Khanh, the Nam Con Son, the Cuu Long and the Malay

basins located farther to the South from the initial East

Sea (Fig 2b)

Local uplift in part of the North-Western Phu Khanh basin in the Middle miocene caused subaerial exposure

of lower Miocene-lowest middle Miocene platform

carbonates Consequently, carbonate growth retreated northward and was replaced by terrigenous deposition During the same period, the continued opening of the East Sea introduced open-marine conditions

in the Southern most part of the Phu Khanh basin, which instigated the growth and deposition of carbonates like in the Nam Con Son basin Carbonate growth in this area was interrupted due to the end-middle Miocene uplift probably associated with the termination of seafl oor spreading (Fig 2c) (Fyhn et al 2009).Carbonate deposition was re-established subsequently on the Northern Con Son swell bordering the Southern Phu Khanh basin, whereas deep marine siliciclastic deposition came to prevail farther off shore in the Phu Khanh basin similar to the situation in the central Nam Con Son basin (Fig 2d) At the same time alluvial and shallow marine deposition dominated in the Malay - Cho Thu basin which, like the Cuu Long basin, was located farther away from the open-marine part of the East Sea

The depositional pattern along the East Vietnamese margin changed considerably as sediment supply increased around late Miocene time in response to the South East Indochinese uplift (Fig 2e) Carbonate deposition was impeded by subaerial exposure of the Phan Rang carbonate platform that covered the Northern Con Son swell Platform growth only re-established patchily during the subsequent transgression

as the input of terrigenous matter and inorganic nutrients to the area

Fig 2 Facies maps of part of the central and south Vietnamese margin a) Palaeogene

fa-cies map illustrating the dominance of non-marine to restricted marine deposits situated

in fault confi ned depressions b) Early Miocene facies map mirroring the early Neogene

transgression triggering widespread carbonate accumulations in the North and alluvial

to shallow marine siliciclastics farther South c) Middle Miocene facies map

Increas-ingly open marine conditions in the South-Eastern part of the area promoted carbonate

growth Shallow marine siliciclastic sedimentation prevailed farther landward and to the

North where carbonate deposition retreated due to magmatism and local uplift d) Latest

middle to late Miocene facies map A transgression followed in the wake of a late middle

Miocene uplift in the South-Eastern part of the area This resulted in widespread carbonate

deposition across the Northern Con Son swell and deeper marine deposition in the Eastern

part of the area Volcanism in the Phu Khanh basin and alluvial to shallow marine

deposi-tion in the Cuu Long basin continued during the period e) Latest Miocene-Recent facies

map Siliciclastic supply increased during the most recent time of the basin evolution This

promoted the build up of a prominent shelf slope and inhibited carbonate production in

the region Magmatism in the Phu Khanh basin dropped during the period whereas

vol-canism was initiated to the South The Dam Thi Nai area is indicated Modifi ed from Fyhn

et al (2009)

increased signifi cantly (Fyhn

et al 2009) Nutrifi cation of

the surface waters along the

Con Son swell was controlled

mainly by intense onshore

erosion and an orographic

induced change of the summer

monsoon that triggered

seasonal upwelling along the

Con Son swell Consequently,

carbonate platforms drowned

off shore South and central

Vietnam throughout the latest

Miocene and early Pliocene

times Siliciclastic dominated

deposition subsequently took

over in previously carbonate

dominated areas This led to the

establishment of a prominent

shelf slope along the central

and South Vietnamese

margin that prograded tens of

kilometres East wards during

the remaining part of the

Neogene and characterizes

the modern outline of the

margin

1.2 Petroleum geology

Cenozoic lacustrine

mudstones and coals/coaly

mudstones are the principal

source rocks in the Vietnamese

and adjacent Chinese basins

(Todd et al 1997; Petersen

et al 2004; Andersen et al

2005; Bojesen-Koefoed et

al., 2005; 2009) Potential

source-rock analogues occur

onshore Total organic carbon

(TOC) content and hydrogen

Index (HI) values of immature

Cenozoic lacustrine mudstone

analogues from the Dong Ho

area and from the ENRECA-1

well drilled in the onshore Song

Ba trough mainly range from

Fig 3 Simplifi ed stratigraphic columns for the basins along the Vietnamese margin with main

regional tectonic events indicated N CSS = North Con Son swell, NE CLB = Northeast Cuu Long basin, NE MB = Northeast Malay basin (Malay - Tho Chu basin), N NCSB = North Nam Con Son basin, PKB = Phu Khanh basin, SHB = Song Hong basin After Fyhn et al (2009)

Table 1 Source rock parameters

4 - 20wt% and from 300 - 700mg HC/g TOC, respectively,

indicating that the organic matter largly is composed of

algal-rich kerogen (type I/II) and is comparable to the

lacustrine source rocks encountered in off shore wells

(Table 1) (Petersen et al 2001; 2004; 2005; 2010; Nielsen

et al 2007) Onshore humic coals display HI values up

to 350mg HC/g TOC, compatible with those sampled in

off shore wells suggesting a potential for oil generation

Data from these onshore source-rock analogues thus

emphasize that mature Cenozoic lacustrine mudstones

and coals/coaly mudstones provide excellent source

rocks for oil and gas generation in the region These source

rocks are interpreted to be abundant in the Palaeogene

syn-rift of the Song Hong, Phu Khanh and Malay - Tho

Chu basins, and sporadically present in Miocene deposits

based on well data and seismic interpretation (Matthews

et al 1997; Lee & Watkins, 1998; Nielsen et al 1999; 2007;

Lee et al 2001; Andersen et al 2005; Petersen et al 2004;

2009; 2010; Fyhn et al 2009; 2010)

• The Phu Khanh basin

Oil from Cenozoic marly source rocks is the most

common seep oil in the Dam Thi Nai lagoon but lacustrine

seep oils, comparable to oils produced from fi elds in the

Cuu Long basin and oils encountered in wells in the Song

Hong basin, were sampled as well by the ENRECA Group

along the lagoonal coast (Traynor & Sladen, 1997;

Bojesen-Koefoed et al., 2005; Fyhn et al 2009) Biological marker

distribution of the prevailing Dam Thi Nai oil exhibits

characteristics resembling extracts of Miocene marly

source rocks in the Nam Con Son basin deposited near

reefal and intra-reefal settings (Traynor & Sladen, 1997;

Bojesen-Koefoed et al., 2005) A compatible early Miocene

fore-reef setting is interpreted immediately off shore

from the Dam Thi Nai area based on seismic data in the

Phu Khanh basin (Fig 2b) (Fyhn et al 2009) The “marly”

Dam Thi Nai Oil may therefore have originated from

lower Miocene fore-reef marls deposited in the narrow

depression along the trace of the EVBFZ in the Northern

half of the Phu Khanh basin, which would require a fairly

simple 40 - 50km up-dip migration pathway for the seep

oils (Fig 2b)

2-D hydrocarbon modelling was carried out to give a

fi rst assessment of the maturation and the hydrocarbon

generation history of the successions potentially

sourcing the oil seeps in the Dam Thi Nai lagoon as well

as to illuminate the timing and control of hydrocarbon

generation and migration in the Phu Khanh basin (Fyhn et

al 2009) Seismic interpretation and gravimetric modelling were used to constrain lithology, ages, structures, crustal thickness, and heat fl ow, and pre-defi ned standard PetroMod physical rock parameters were assigned in the absence of well data

Modelling indicates that part of the syn-rift succession entered the oil window during the Palaeogene During the early Neogene the level of maturation in widespread areas only increased moderately However, as the sediment accumulation rate increased during the late Neogene, the potential syn-rift and early Miocene source intervals were deeply buried by prograding deposits, which forced the main potential source intervals through the main oil window and caused the majority of the Palaeogene syn-rift farther seawards to be situated in the oil window The late Neogene is therefore interpreted as the single-most important period for oil and gas generation in the Phu Khanh basin, although magmatic activity may have infl uenced source maturation locally in the basin

The Dam Thi Nai oil seeps and the recent White Shark oil discovery in the central part of the Phu Khanh basin indicate working petroleum systems within the basin This

is substantiated by numerous potential direct hydrocarbon indicators (DHI), such as gas seeps, amplitude anomalies,

fl at spots and chimney-like features, mostly situated

in various carbonate and sand-prone intervals (Lee & Watkins 1998; Fyhn et al 2009) The ENRECA study has thrown light on a series of structural and stratigraphic trap types situated in favourable positions relative to potential source rocks The traps mainly formed before or during early Neogene time, preceding the late Neogene main oil generation The study further indicates that potential reservoir rocks are composed of Miocene carbonates, diverse sand-prone depositional facies ranging from non-marine fl uvial deposits to deep marine turbidite sequences and fractured basement highs in the western half of the basin sealed by carbonate drowning sequences, transgressive shales and lacustrine mudstones A series of promising hydrocarbon plays thus exist in the basin, many

of which are located in shallow water (Fig 4)

• The Song Hong basinAnalysis of the Song Hong basin carried out during the initial phase of the ENRECA project suggested the presence of working petroleum systems in the basin Oil-

source correlations suggest the presence of a Miocene

coaly source-rock and a lacustrine mudstone source

rock (Nielsen et al 1999; Andersen et al 2005) Miocene

intervals containing thick coal seams encountered in wells

were mapped out seismically across a larger region of the

basin Similarly, Palaeogene lacustrine mudstones with

excellent source potential crop out on Bach Long Vi and

onshore in the Dong Ho area Refl ector intervals situated

within the syn-rift of the Song Hong basin composed

of continuous, low-frequency, amplitude refl ectors interpreted as thick dominantly lacustrine mudstone successions occur regionally along the rim of the Song Hong basin and have been mapped out Early modelling of source-rock maturity and petroleum generation indicated the likeliness of active petroleum systems in the NE Song Hong basin (Nielsen et al 1999; Andersen et al 2005) Modelling further suggested late timing of maturation

high-of syn-rift source rocks along the basin margin and of post-rift coals situated

in the central part of the basin This has allowed for extended periods of time for post-rift structures to form and to be sealed prior to hydrocarbon expulsion and migration A number of subsequent discoveries made in recent years in blocks 102, 103, 106 and 107 have confi rmed the existence of these Cenozoic petroleum systems in the area and the fi ndings of the initial modelling eff ort in the basin Even so, a signifi cant gab in the understanding of the geology and the petroleum systems of the Song Hong basins exists The ENRECA group is therefore in the process of revisiting the Song Hong basin and plan to drill a fully cored well in the syn-rift succession of the basin as part of our activities

• The Malay - Tho Chu basin

In the Malay - Tho Chu basin petroleum exploration began during the early 1970’ encouraged by the successful exploration activities immediately South

of Vietnamese territory The fi rst well was drilled in 1994, and since then, signifi cant gas, condensate and oil discoveries have been made in several wells drilled

in the Malay - Tho Chu basin, but only a few discoveries are as yet considered commercial A re-evaluation of the tested exploration strategies is therefore necessary in order to optimize and focus future exploration

Exploration has mainly aimed at lower to middle Miocene fl uviodeltaic sand reservoirs with late Neogene

Fig 5 (a) Yükler-modelled Present-day maturation level for type I kerogene at the base

of the interpreted Palaeogene syn-rift source interval in the North-Eastern Song Hong

basin (b) Present-day modelled maturation level for type III kerogene at the base of the

interpreted Palaeogene syn-rift source interval After Andersen et al (2005)

Fig 4 Schematic diagram summarising potential hydrocarbon play-types in the Phu

Khanh basin The potential plays are based on source rocks primarily composed of

Pal-aeogene lacustrine mudstones and coals and lower Miocene marly mudstones Various

sand-, carbonate- and basement-reservoir types are outlined relying on both structural

and stratigraphic trapping mechanisms Modifi ed after Fyhn et al (2009)

structural trapping mechanisms Potential source rocks

have been interpreted to be alginite-bearing lacustrine

shales and humic coals situated in the Palaeogene syn-rift

and in the lowermost post-rift successions

Only few potential source-rock levels have been

penetrated by wells, but those that have, have suppressed

vitrinite refl ectance (VR) values compared to VR values

obtained from overlying Neogene coals Suppressed VR

values may occur in alginite-rich rocks and VR suppression

is therefore particularly common in lacustrine shales with

high HI values The maturity trends of such VR datasets

may not be well-constrained and produce abnormally

low thermal maturity gradients Thus Fluorescence

Alteration of Multiple Macerals (FAMM) was applied in

order to obtain reliable thermal maturity trends in rocks

containing vitrinite with suppressed and enhanced VR

values By combining conventional VR measurements and

FAMM data a revised and more accurate thermal maturity gradient has been established (Fig 6) (Petersen et al 2009) 2-D modelling of the maturation history of the basin was carried out based on the revised thermal maturity gradient, detailed seismic mapping, well information and custom kinetics for bulk petroleum generation; the latter determined from outcrop samples of lacustrine source rock analogues and a terrestrially infl uenced mudstone collected from wells (Petersen et al 2010)

The 2-D models suggest that most of the syn-rift succession in the Vietnamese Malay basin is located in or has passed through the main oil and gas windows Syn-rift source rocks have therefore produced and expelled signifi cant quantities of hydrocarbons, however, the main oil generation generally took place during the early and middle Miocene prior to formation of structural traps in late Neogene time

2-D modelling of the hydrocarbon generation therefore suggests that the main risks in the tested play types are 1) the timing of petroleum generation relative to trap formation completed

in the late Neogene, 2) pervasive Neogene faulting, which may have complicated petroleum migration to the structures and breached charged traps, and 3) the distribution and amount of matured source rocks in smaller grabens Based on the abovementioned and the presence of DHI’s,

an untested alternative play type is proposed relying on syn-rift sandstones located up-dip from and near source-rock intervals with Palaeogene structural and stratigraphic trapping mechanisms that did not experience subsequent Neogene deformation

Fig 6 Mean average VR values from a well in the South-Eastern part of

the Malay - Tho Chu basin and FAMM-derived EqVR values plotted

accord-ing to depth Five of the VR values defi ne a trend that has a very high

cor-relation coeffi cient The two deepest samples are omitted due to

compli-cations caused by VR suppression and possibly cavings The curve defi ned

by the EqVR values includes all seven samples From Petersen et al (2009)

geophysical and petrophysical interpretations of digital

data Comprehensive training in data loading and system

management were provided to VPI geo-scientists as an

integrated part of the technology project The necessary

training comprised both in-house training at VPI instructed

by GEUS experts and long-term OJT programmes at GEUS

The technology project was fully integrated with the

on-going geo-scientifi c project, which benefi ted greatly from

the new powerful tools as the interpretation of seismic data

and well-log data were carried out on the workstations

In addition, the experiences harvested by utilising the new facilities and the applied methodologies were disseminated to geo-scientists from all CCOP-member countries during open workshops held at VPI and instructed by a team of GEUS and VPI experts

The introduction of interactive interpretation

of geophysical and geological data at VPI and the establishment of a Vietnamese group of capable geo-scientists has further stimulated VPI in its eff orts to create

a modern public research institute The equipment

provides the necessary technological background for eff ective and modern interpretations, and with the appropriate project data bases, the geo-scientists at VPI are able to provide research-based advice to the energy authorities in Vietnam and to perform consulting work for the hydrocarbon industry active in Vietnam

3 On-the-job-training (OJT), ing courses, seminars and workshops, post-graduate education

train-The OJT programmes at GEUS for geo-scientists from VPI, HUMG and HUS have all been tailored towards specifi c needs identifi ed during the progress

of the research projects The needs for training have been related to data and system management, and interactive interpretation of seismic, geological and petrophysical well data on workstations,

as well as various aspects within sedimentology, biostratigraphy, organic geochemistry, organic petrography and basin modelling

Selected topics including stratigraphic principles, seismic interpretation, sequence stratigraphic methodology and source rock geology have been introduced

to a larger audience at VPI in Hanoi at seminars and classes typically of one-week duration These courses have been open for participants not only from VPI and Petrovietnam in Hanoi but also for their staff from the offi ces in Ho Chi Minh city, and for students and teachers from the local universities in Hanoi Four workshops

Fig 7 Drilling of ENRECA-2 well on the Phu Quoc Island

Fig 8 Transboundary Vietnam - Cambodia fi eld trip organized by the ENRECA project

with the support from CCOP

open for geo-scientist from the CCOP-member countries

have been instructed on the interpretation facilities at

VPI covering various aspects from seismic interpretation

and mapping to the construction of geological models

for exploitation of hydrocarbons or groundwater

reservoirs Manuals and selected literature have been

provided to the course and workshop participants for

further studies

The projects have also provided funds for the

participation of several VPI geo-scientists in international

conferences giving them an opportunity to be exposed

to the international research community and to establish

scientifi c contacts with other researchers and discuss with

them scientifi c matters of common interest The projects

also provided training to the involved VPI researchers for

enhancing their capabilities in scientifi c speaking and

report writing

One of the fruitful results of cooperation between VPI

and GEUS is enhancement of human resources of VPI Four

VPI project members received their master certifi cation, and

four VPI geoscientists got their doctoral certifi cation in the

implmentation process of cooperation research projects

They play a more important role contributing to the scientifi c

research programme of VPI

III Summary and conclusions

The long-lasting co-operation between VPI and

GEUS has been successful and benefi cial to all parties

The close geo-scientifi c relationship between the two

sister-organisations is accompanied by many personal

friendships and rewarding experiences

Many Vietnamese project participants have been

introduced to, and trained, methods that are commonly

applied within the international oil and gas industry, and

a large number of the participants have been exposed

to the international scientifi c community At the same

time, many GEUS geo-scientists have been exposed

to stimulating and exciting geological problems and

have gained invaluable personal experiences with the

challenges met during project management, conduction

of scientifi c research and provision of training courses in a

diff erent working environment

Besides the numerous cultural and personal gains,

the scientifi c research conducted jointly by Vietnamese

and Danish researchers has resulted in many encouraging

and important conclusions regarding the geology and hydrocarbon potential of Song Hong, Phu Khanh, Malay - Tho Chu and Phu Quoc basins

Non-marine to restricted marine deposition prevailed along the margin from the Eocene to the Oligocene, and lacustrine mudstones and humic coals with source rock potential were deposited during the period The marine infl uence increased during latest Oligocene and Neogene time as the East Sea approached its present outline Carbonates therefore constitute a signifi cant part of the Miocene along the East Vietnamese margin, whereas clastic deposition prevailed in the northern Song Hong and the Malay - Tho Chu basins situated near entry points of terrestrial input and farther from open sea areas Depositional rates increased signifi cantly during the late Neogene in response to uplift and denudation

of Southern Indochina The uplift was associated with

an intensifi cation of volcanism in the region, which was initiated off shore during the early Neogene and subsequently broadened

Maturation modelling of the Song Hong basin suggests a regional petroleum potential in the basin Miocene oil-prone coals encountered in wells are presently at a mature state in the distal parts of the basin Eocene - Oligocene syn-rift source rocks crop out

on the Bach Long Vi and at Dong Ho and are expected

to be abundant along the basin margin based on seismic facies mapping Maturation modelling indicate that these syn-rift source rocks are presently oil to gas generating

in widespread areas along the basin margin, which has allowed considerable time for traps to have formed and been sealed

The Phu Khanh basin contains active petroleum systems as indicated by the Dam Thi Nai oil seeps and oil tested recently in the central part of the basin Signifi cant source rock intervals may be present in the basin and include thick Palaeogene syn-rift sequences interpreted

to contain abundant lacustrine and coaly intervals, and lower Miocene carbonaceous fore-reef marls

In the Malay - Tho Chu basin FAMM analyses of coals and carbonaceous mudstones has led to a revised and steeper maturation gradient in the area Basin modelling incorporating the revised maturation gradient and new custom kinetic data for coals and lacustrine source rocks indicate that the main risks for the tested Neogene play types are 1) the timing of petroleum generation relative

to trap formation completed in the late Neogene, 2)

the pervasive faulting, which may have complicated

petroleum migration to the structures and breached

charged traps, and 3) the distribution and amount of

matured source rocks in smaller grabens An alternative

syn-rift play type is therefore suggested, relying on

sand reservoirs located next to source-rock intervals,

and Palaeogene trapping mechanisms unaff ected by

subsequent Neogene structuring

The group of Vietnamese and Danish geoscientists

has evaluated the geology and petroleum potential of

the Song Hong, Phu Khanh, Malay - Tho Chu and the Phu

Quoc basins along with the training of MSc and Ph.D

students in hydrocarbon related geology/geophysics In

the coming phase of the ENRECA project the Song Hong

basin will be revisited As part of the activities a fully cored

stratigraphic well is considered to be drilled through the

Palaeogene syn-rift succession in the Song Hong basin

in order to test and improve knowledge of, for example,

source potential, source-rock deposition and maturation,

Palaeogene biostratigraphy and age of rifting, overall

syn-rift sedimentology, petrography, provenance areas and

recognition of source-rock intervals from seismic data The

stratigraphic well is combined with and complemented

by structural and stratigraphic studies based on seismic

interpretation, basin modelling and outcrop analysis

A strong foundation for further co-operation has

been established It is the sincere hope that the fruitful

co-operation laid down during the previous activities will

continue in the future on a long-term basis

Acknowledgements

The focus in this contribution has been on the results of

the co-operation between VPI and GEUS The results would

not have been achieved without the great support from a

large number of individuals at Petrovietnam, VPI, GEUS,

CCOP, Institute of Geography and Geology (IGG) - University

of Copenhagen, Hanoi University of Mining and Geology

(HUMG), Hanoi University of Science (HUS) - Hanoi National

University All colleagues and friends at these institutes

are thanked for their contributions, encouragement and

support Various Vietnamese Authorities, including the

Ministry of Planning and Investment and the Vietnamese

Embassy in Copenhagen are thanked for valuable advice

and administrative support The Royal Danish Embassies

in Hanoi, Danida Fellowship Centre, and Danida are

thanked for the fi nancial support Vietnam Petroleum Institute, Petrovietnam and GEUS are acknowledged for the institutional support, provision of manpower, access

to samples and data, and permissions to publish the scientifi c results

geo-References

1 Andersen, C., Mathiesen, A., Nielsen, L.H., Tiem,

P.V., Petersen, H.I., Dien, P.T., 2005 Evaluation of petroleum systems in the northern part of the Cenozoic Song Hong basin (Gulf of Tonkin), Vietnam Journal of Petroleum

Geology, 28, 167-184

2 Andersen, C & Dien, P.T., 1999 Geoscientifi c operation between GEUS and Petrovietnam/VPI Basin analysis and modelling of the Cenozoic Song Hong basin, Vietnam (phase II) CCOP Newsletter Vol 24, No 3.

co-3 Andersen, C & Dien, P.T (Eds), 1998 Basin analysis and modelling of the Cenozoic Song Hong basin, Vietnam

Final report to phase II, Hanoi Confi dential

4 Bojesen-Koefoed, J.A., Nytoft, H.P & Dau, N.T.,

2009 Petroleum composition in the Cuu Long basin (Mekong Basin) off shore Southern Vietnam Marine & Petroleum

28, 3-18

6 Fyhn, M.B.W., Boldreel, L.O & Nielsen, L.H.,

2009 Geological development of the central and South Vietnamese margin: Implications for the establishment of the East Sea, Indochinese escape tectonics and Cenozoic volcanism Tectonophysics, 478, 184-204.

7 Fyhn, M.B.W., Boldreel, L.O & Nielsen, L.H., 2009

Tectonic and climatic control on growth and demise of the Phanh Rang Carbonate Platform off shore South Vietnam

Basin Research, 21, 225-251

8 Fyhn, M.B.W., Boldreel, L.O & Nielsen, L.H., 2010

Escape tectonism in the Gulf of Thailand: Paleogene lateral pull-apart rifting in the Vietnamese part of the Malay basin Tectonophysics, 483, 365-376.

left-9 Fyhn, M.B.W., Nielsen, L.H., Boldreel, L.O., Thang, L.D., Bojesen-Koefoed, J., Petersen, H.I., Huyen, N.T., Duc,

N.A., Dau, N.T., Mathiesen, A., Reid, I., Huong, D.T., Tuan,

H.A., Hien, L.V., Nytoft, H.P & Abatzis, I 2009 Geological

evolution, regional perspectives and hydrocarbon potential

of the northwest Phu Khanh basin, off shore Central Vietnam

Marine & Petroleum Geology, 26, 1-24

10 Fyhn, M.B.W., Pedersen, S.A.S., Boldreel, L.O.,

Nielsen, L.H., Green, P.F., Dien, P.T., Huyen, L.T & Frei, D.,

2010 Palaeocene-early Eocene inversion of the Phu Quoc -

Kampot Som basin: SE Asian deformation associated with

the suturing of Luconia Journal of the Geological Society

of London, in press

11 Lee, G.H & Watkins, J.S., 1998 Seismic stratigraphy

and hydrocarbon potential of the Phu Khan Basin, off shore

Central Vietnam, East Sea AAPG Bulletin, 82, 1711-1735

12 Matthews, S.J., Fraser, A.J., Lowe, S., Todd, S.P

& Peel, F.J., 1997 Structure, stratigraphy and petroleum

geology of the SE Nam Con Son Basin, off shore Vietnam In:

Fraser, A.J., Matthews, S.J & Murphy, R.W (eds.): Petroleum

Geology of SouthEast Asia Geological Society, London,

Special Publication, 126, 89-106.

13 Nielsen, L.H., Mathiesen, A., Bidstrup, T., Vejbæk,

O.V., Dien, P.T., Tiem, P.V., 1999 Modelling the hydrocarbon

generation in the Cenozoic Song Hong basin, Vietnam: a

highly prospective basin Journal of Asian Earth Sciences,

17, 269-294

14 Nielsen, L.H., Petersen, H.I., Thai, N.D., Duc, N.A.,

Fyhn, M.B.W., Boldreel, L.O., Tuan, H.A., Lindstrøm, S &

Hien, L.V., 2007 A Middle-Upper Miocene fl uvial-lacustrine

rift sequence in the Song Ba Rift, Vietnam: an analogue to

oil-prone, small-scale continental rift basins Petroleum

Geoscience, 13, 145-168

15 Nielsen, L.H (ed.) 1997 Analysis of geological

basins in Vietnam using sequence stratigraphy and

basin modelling EFP-94: Ens nr 1313/94-0011: Ulande,

Oliegeologisk forskningssamarbejde, Vietnam GEUS

rapport 1997/111 Confi dential

16 Nielsen, L H., 1996 Offi cial inauguration of

workstations at Vietnam Petroleum Institute CCOP

Newsletter, vol 21 no 1, p 14.

17 Nielsen, L.H & Dien, P.T (Eds) 1997 Basin analysis

and modelling of the Cenozoic Song Hong basin, Vietnam

Final report to phase one, Vol 1 & 2: 12 technical reports

incl encl Confi dential

18 Petersen, H.I., Andersen, C., Anh, P.H., Koefoed, J.A., Nielsen, L.H., Nytoft, H.P., Rosenberg, P &

Bojesen-Thanh, L., 2001 Petroleum potential of Oligocene lacustrine mudstones and coals at Dong Ho, Vietnam - an outcrop analogue to terrestrial source rocks in the greater Song Hong basin Journal of Asian Earth Sciences, 19, 135–154.

19 Petersen, H.I., Mathiesen, A., Fyhn, M.B.W., Dau,

N.T., Bojesen-Koefoed, J.A., Nielsen, L.H., 2010 Modeling of petroleum generation in the Vietnamese Northeastern part

of the Malay basin (Gulf of Thailand/East Sea) using custom kinetics for generation of bulk petroleum AAPG bulletin, in

press

20 Petersen, H.I., Nytoft, H.P & Nielsen, L.H., 2004

Characterisation of oil and potential source rocks in the northEastern Song Hong basin, Vietnam: indications of

a lacustrine-coal sourced petroleum system Organic

Geology, 26, 319-332

22 Petersen, H.I., Tru, V., Nielsen, L.H., Duc, N.A

& Nytoft, H.P., 2005 Source rock properties of lacustrine mustones and coals (Oligocene Dong Ho Formation), onshore Song Hong basin, Northern Vietnam Journal of Petroleum Geology, 28, 19-38.

23 Todd, S.P., Dunn, M.E & Barwise, A.J.G., 1997

Characterizing petroleum charge systems in the Tertiary of

SE Asia In: Fraser, A.J., Matthews, S.J., Murphy, R.W (Eds.), Petroleum Geology of Southeast Asia Geological Society,

London, Special Publication, 126, pp 25-47

24 Traynor, J.J & Sladen, C 1997 Seepages in Vietnam - onshore and off shore examples Marine &

Petroleum Geology, 14, 345-362

25 Vejbæk, O.V., Madsen, L & Tiem, P.V., 1997

Technical report 8 - Seismic interpretation of the Song Hong basin, Vietnam In: Nielsen, L.H., Dien, P.T (Eds.) Basin

analysis and modelling of the Cenozoic Song Hong basin, Vietnam In-house confi dential report

Description of the data set

The passive seismic survey carried out in April, 2007

at Voitsdorf, Austria includes six lines running

North-South across two reservoir sections of the fi eld Each line

consisted of ten ultra-sensitive, 3 component portable

broadband seismometers deployed at 1km spacing More

detailed description of the geology and acquisition can

be found in Lambert et al (2008)

On April 8th, at 5.42am local time an earthquake

of magnitude 6.5 richter originated from Northern

Afghanistan at a depth of around 200km arrived in the area

and was detected by one of the survey lines The location

of the line relative to the reservoir is presented in Fig 1a

Fig 1b shows 100 minutes (5 - 6.40am) of the vertical components of the data containing the earthquake signals that were extracted for analysis In this Fig the earthquake can be clearly identifi ed as two main events (P-wave and S-wave arrivals) and several minutes of codas We divided the data set into three intervals for comparison: (1) before the earthquake, the portion of data up to the

fi rst detectable arrival of the earthquake; (2) during the earthquake, the interval from the fi rst earthquake arrival until the average amplitudes return to their normal values and (3) after the earthquake, the rest of the data set These time intervals are marked as dotted rectangles in Fig 1b

Earthquakeinfluenceonlow-frequencyseismic wavefieldoverahydrocarbonreservoirin

Voitsdorf,Austria

Nguyen Thanh Tung

Vietnam Petroleum Institute

Phan Thien Huong

Hanoi Mining University 

Introduction

Several studies at various oil and gas fi elds worldwide have shown low-frequency spectral anomalies in the passive surface particle velocity signals which have a good correlation with the location of hydrocarbon (HC) reservoirs (e.g., Dangel et al., 2003; Holzner et al., 2005; Birialtsev et al., 2006; Mastrigt and Al-Dulaijan, 2008) The main observation

is an increase of spectral amplitudes in the low frequency band of passive seismic data approximately between 1 and 6Hz above a hydrocarbon reservoir The anomalous signatures on a data spectrum can be quantifi ed by some passive seismic attributes to aid the interpretation (Lambert et al 2008).

Steiner et al (2008) applied time reverse modelling to show that the a reservoir zone is the origin of the frequency energy The physical mechanism of this phenomenon is not yet fully understood, however it is usually suggested that the ever-present seismic background noise of the earth (e.g., Berger et al., 2004) acts as the driving force for the generation of the anomalous signals with underlying possible rock-physical mechanisms discussed in Graf et al (2007) This raises an interesting question; if HC reservoirs react to such low energy sources, how will they respond to strong earthquakes? In this paper we describe an investigation of the impact of earthquakes on passive seismic data around oil and gas reservoirs.

low-Three diff erent techniques were applied for the investigation using signals from a major earthquake that was recorded during a passive seismic survey at a known oil and gas fi eld near Voitsdorf, Austria: (a) spectral analysis technique that is commonly used to identify passive seismic anomalies, (b) empirical mode decomposition to extract intrinsic modes of the data and analyse their properties and (c) time reverse modelling of the data in time domain to localize possible low frequency sources in the subsurface The objective is to compare the properties of three portions

of data: before the earthquake, during the earthquake and after the earthquake.

Compared to the overall quality of the survey the quality

of the data set is rather poor with signals from stations 41,

42, 43, and 47 being strongly contaminated with surface

ambient noise mainly caused by a pump and a highway

nearby An analysis also found that the background noise

level increased with time for this data set

Spectral analysis

The spectral analysis of passive seismic data is the

most common way to identify anomalous locations Some

quantitative attributes of the spectrum can also be used

to aid the interpretation (Lambert et al., 2008) Usually the

quietest time intervals within several hours of recordings

are selected and transformed into frequency domain for

analysis Unfortunately the optimal (i.e quietest) time

interval cannot be chosen for the data set used in this

study because it is constrained by the earthquake arrival

To reduce the eff ects of higher level of contaminated

artifi cial noise an additional processing procedure was

implemented as following:

- Segment the data into non-overlapping 10 second

time windows

- Compute the amplitude standard deviations (STD)

of each time window

- Reject all time windows with STD exceeding triple

of STD of the whole data length

The example in Fig 2 shows a comparison of the

data at station 41 before and after the processing It can

be seen that most strong noisy transients have been

eff ectively removed by the procedure

In the next step, the power spectral density (PSD)

was computed for each of the remaining 10 second

time window The arithmetic mean had been removed

and the Hanning window function had been applied

before the transformation to reduce the spectral leakage

Representative spectra of the three time intervals,

before the earthquake, during the earthquake and after

the earthquake were then generated for each station

simply by averaging the PSD values of all 10 second time

windows within the corresponding interval

In Fig 3 the representative spectra of the data from

all stations of the survey line in Fig 1 are plotted side

by sided for comparison Obviously, despite the extra

processing steps, broadband noise still dominates at

station 42 above reservoir 2 and obscures possible useful

information from it We therefore mainly focused on signatures from reservoir 1 whose location suff ered less from surface noise Only the frequencies between 1Hz-10Hz are considered because the signals above 10Hz are mostly noise while the signals below 1Hz are fully dominated by the earthquake main event

It can be seen in Fig 3a that before the earthquake there is a small amplitude increase around 3Hz at station

51 above reservoir 1 but it can hardly be defi ned as an anomaly without priory knowledge of the reservoir location During the earthquake, frequency components

of up to 5Hz, especially the earthquake dominant frequency range (below 2Hz), signifi cantly increased, as seen in Fig 3b As the earthquake passed, the amplitudes

of its dominant frequency range returned to their normal values but the frequency components between 2 - 5Hz remain high even after the main earthquake events disappeared This could be due to the earthquake but could also be the background noise that increased with time However, the most interesting observation in this part is that the reservoir section at stations 51, which hardly shows any anomalous value before the earthquake (Fig 2a) becomes clearly highlighted after the earthquake

(marked by the ellipse in Fig 2c).

The earthquake eff ects on two attributes of the data spectrum that are commonly used as hydrocarbon indicators: the PSD-IZ attribute (Saenger et al 2007) and the V/H ratio (Lambert et al 2007) were also investigated

Fig 4 shows the PSD-IZ values computed for the data

before and after the earthquake (a) and their diff erences (b) The values during the earthquake could not be determined because the spectra have no local minimum

in the 1-1.7Hz range Before the earthquake, the

PSD-IZ does not show any anomalous signature over the reservoir at station 51 After the earthquake, the PSD-IZ value above the reservoir strongly increased and becomes

a local maximum But the best indicator for the reservoir location

in this case is the diff erence of PSD-IZ values before and after the earthquake Fig 4b shows that the reservoir is clearly singled out by the PSD-IZ diff erence

V/H ratio

Lambert et al (2007) suggested that the dominant amplitude peak of the ratio of vertical to horizontal components (V/H) in the low frequency range (i.e 1 - 6Hz) may be related to hydrocarbon presence Fig 5 shows the V/H peak amplitudes computed for the data before, during, and after the earthquake One can immediately notice that unlike the PSD-IZ values the V/H peak amplitudes in this case are not sensitive to the earthquake They have very similar values for all three time intervals probably because the horizontal and vertical components of the signals are aff ected by the earthquake in a similar way The spectral ratio of the two components therefore has eff ectively deconvolved the earthquake signals The profi le in Fig 5 does not show anomalous V/H peaks over the reservoirs using data from any of the 3 time windows This is consistent with the observations reported by Lambert et al (2008), who carried out a thorough analysis of the whole survey

Empirical mode decomposition

Empirical Mode Decomposition (EMD) is a signal processing technique which allows non-stationary and nonlinear time-series to be examined (Huang et al.,

Fig 3 Data spectra before, during, and after the earthquake Above the reservoir (station 51),

the amplitudes around 3Hz increase strongly after the earthquake (marked by the ellipse)

Fig 1 Locations of the stations relative to the reservoirs (a) and the earthquake signals picked

up by them (b) The dotted rectangles mark the three time intervals used for analysis

Fig 2 Strong transient noise contaminated in the original data (blue) have been eff ectively

removed after the application of the removal tool (red)

1998) The technique has been proven to be remarkably

eff ective in several areas of data analysis (see Huang and

Shen, 2005; Sole et al., 2007) According to Huang et al

(1998) the EMD method is implemented using a recursive

“sifting” algorithm, which consists of two loops, an outer

loop and an inner recursive loop The outer loop runs

from lower to higher mode functions which are created

by the inner recursive loop Given a time series zt with

equal spacing, the eff ective algorithm of EMD can be

summarized as follows (see also McDonald et al 2007):

- Identify all extrema of the time series (zt).

- Interpolate between minima (maxima) using cubic

spline method, ending up with an upper and a lower

envelope (eminz, emaxz).

- Compute the mean of the two envelopes

mn=(eminz + emaxz)/2.

- Extract the residual/detail dz = zt - mn.

- Iterate on dz through steps 1 to 4 until the mean of

dz satisfi es a pre-defi ned near-zero stopping criteria The

recursive inner loop then is fi nished

After step 5, a component of the data called the intrinsic mode function (IMF) is received The IMF is then subtracted from the data series producing a residual time series The fi rst pass of the outer loop then is completed and the inner loop (steps 1 to 5) is repeated for the residual The process continues until all the variability

in the data series (zt) is decomposed into a number of

IMFs and a background trend, which is the last residual It should be noted that adding all IMFs and the background trend will recover exactly the initial time series

As demonstrated by several authors (Huang et al 1998; Rilling et al 2003, McDonald et al 2007; amongst others) one of the main advantages of EMD method is that after the decomposition, signals of diff erent natures tend to reside in diff erent IMFs even if they overlap in both time and frequency

Since the data set used in this study is contaminated

Fig 4 The PSD-IZ values computed for time intervals before and

after the earthquake (a) and their diff erence (b) Reservoir sections

are located under stations 42 and 51

Fig.5 V/H peak amplitudes computed for three time intervals The

values are stable throughout the earthquake with no signifi cant

dif-ference between data intervals

Fig 6 EMD modes of a 10 second time intervals Modes 1 and 2

mostly consist of artifi cial noise while modes 5, 6, and 7 contain the low frequency components such as OWP

with strong artifi cial broadband transients an application

of EMD can be useful to separate the noise from the

signals of interest This will help to preserve a signifi cant

amount of data that otherwise would be rejected by the

processing steps described in previous section

While working with EMD, we noticed that it is a slow

process compared to other spectral decomposition

techniques Moreover the computational time seemed

to exponentially increase with the data length making it

impractical to implement EMD on the whole data set in

one go A solution was to split the data into 10-second

time windows and apply EMD to each of them The

practice saves the computing cost with a trade off : the

properties of the same mode can slightly vary between

diff erent time windows However, this should not be a

problem for this study because the main purpose of using

EMD here is simply to reduce apparent transients

The example in Fig 6 shows seven intrinsic modes of

10 seconds of data recorded at station 41 as well as their

spectra It can be seen that modes 1 and 2 mainly consist

of broadband transients or high frequency signals, which

most probably are artifi cial noise caused by a pump and

the highway nearby the survey line Modes 5, 6 and 7 are

dominated by low frequency components with central

frequencies below 1Hz corresponding to famous

ever-present ocean wave peaks These modes were removed

from the data leaving only the combination of modes

3 and 4, which are expected to contain useful signals

within the characteristic frequency range of 1-6Hz The

next steps of the analysis were carried out similarly to the

common spectral analysis approach PSD of the 10 second

time windows were grouped into the three intervals:

before, during and after the earthquake then averaged to

generate the representative PSD for these intervals

Fig 7 shows a comparison of representative PSD values

of the three time intervals computed for combination

of EMD modes 3 and 4 at all ten stations Before the

earthquake, the amplitude values at station 51 above

reservoir 1 are noticeably higher than for the adjacent

stations However, it is still diffi cult to identify them as

anomalous signatures because the stations at both ends

of the line also show comparable (or even higher) values

During the earthquake, the spectral amplitudes at

station 51 are signifi cantly increased and it becomes the

most highlighted along the profi le (Fig 7b) After the

earthquake the amplitude values at station 51 slightly decrease but remain anomalously higher than the values

at other stations of the survey line (Fig 7c)

As mentioned before, data from station 42 above reservoir 2 suff ered from extremely high level of surface noise The clear anomaly observed at this station throughout all three time intervals therefore is regarded

as the eff ect of residual noise that EMD could not fully remove and has not been taken into account

Time reverse modeling

Steiner et al (2008) demonstrated that time reverse modelling (TRM) can be used to locate the origin of low frequency tremor signals in the subsurface With the observation that spectral amplitudes at low frequency range over the reservoir increase after the earthquake found in the previous sections, a natural next step is

to apply TRM technique to locate the sources of the anomalous signals

As described in Steiner et al (2008), TRM is implemented by using a fi nite diff erence algorithm for two-dimensional (2D) elastic wave propagation Physical parameters of the media are defi ned by a velocity model

of the profi le Recorded passive seismic data in time domain are band-pass fi ltered to 1-6Hz, reversed in time and used as input signals for numerical wave propagation The maximum absolute particle velocities at every numerical grid point of the model is stored throughout the time reverse process The fi nal output is a grid of data points, for which locations of possible low-frequency seismic sources in the subsurface are indicated by areas of high amplitude values More detailed description of the algorithm can be found in Steiner et al (2008)

Fig 8 shows the velocity model of the subsurface along the survey line with two known reservoirs lying on top of the high velocity basement (marked by white ellipses) The actual velocity information is only available for the left hand side of the model The values in the shaded area

at the right end are horizontally extrapolated It should

be noted that for TRM the contaminated transient noise cannot be eff ectively removed because the technique requires several minutes of continuous and synchronized signals of both vertical and horizontal components Therefore an experiment with ambient noise was carried out to understand the noise eff ects on the TRM result before the application to real data

Noise analysis

In the experiment three minutes

of white noise being scaled and

fi ltered to match the mean amplitudes and frequency content of the real data were used as the input for the TRM The result in Fig 9 shows that ambient noise apparently causes three high focusing areas (marked as A, B and C) on the TRM result of this profi le The cloud A is obviously caused by the excessive noise level at station

42 The other high amplitude areas B and C within a low velocity layer are probably the focusing eff ect caused

by the high velocity lens above them These spurious anomalous areas must

be taken into account when TRM is implemented for the real data

Time reverse modelling of real data

For the real data set we again applied TRM to the three time intervals: before, during and after the earthquake (Fig 1b) for comparison The results are shown in Fig 10

It can be seen that before the earthquake the TRM section (Fig 10a) has very similar properties to the noise test section in Fig 9 with three spurious areas being highlighted

We conclude that ambient noise has fully dominated this time interval overprinting all possible signals from the reservoirs During the earthquake

it is interesting to observe that signals from reservoir 1 had exceeded the background noise and the reservoir location has been brought out by high focusing amplitudes as shown

in Fig 10b As the earthquake passed, reservoir 1 became dimmer on TRM section with only part of it being highlighted (Fig 10c) In all three cases reservoir 2 cannot be clearly distinguished from the strong noisy cloud above it

Fig 7 PSD of EMD modes 3 and 4 computed for three time intervals before, during, and

after the earthquake For station 51 above reservoir 1, the amplitudes around 3Hz increase

strongly during and after the earthquake The anomalous amplitudes at station 42 above

reservoir 2 are probably caused by noise

Fig 8 Velocity model of the passive seismic line in Fig 1a

Fig 9 TRM sections generated for three time intervals: (a)before, (b)during and (c)after

the earthquake The dashed ellipses indicate the location of the known reservoirs

To identify the areas on the TRM section that changed

most because of the earthquake we subtracted the

section before earthquake in Fig 10a from the sections

during (Fig 10b) and after earthquake (Fig 10c) The

diff erences are shown in Fig 11 Apparently, the zone of

reservoir 1 was most aff ected by the earthquake Patches

of noise scattering all over the section in Fig 10b have

disappeared in the diff erences (Fig 11a) singling out the

reservoir during the earthquake After the earthquake the

anomalous signals decreased and the reservoir is partly

highlighted (Fig 11b) The diff erence section however

does not help to highlight reservoir 2 which is always

masked by the cloud of strong noise under station 42

Discussion

Comparing the results of three analysis techniques

we can see that they are consistent for two time intervals:

before and after the earthquake Before the earthquake,

all three methods cannot clearly identify the reservoir

locations but after the earthquake they all show anomalous values at the location

of reservoir 1 For the time interval during the earthquake the EMD and TRM show the best results with reservoir

1 obviously highlighted but the PSD spectra generated by a common spectral analysis method hardly exhibit any anomalous signature A possible reason

is that the noise removal process used

by this method was too aggressive for this time interval due to high variation of the data caused by the earthquake As a consequence, useful strong signals have been removed by the algorithm together with the noise

The empirical mode decomposition technique was fi rst applied for this type

of data and shows promising results Artifi cial transients and low frequency background are eff ectively separated from the data leaving only components within the frequency range of interest The anomalous values found by this method match with reservoir 1 location and are consistent with the TRM results.Despite a high level of noise contamination in the data the TRM technique has been successfully applied The result shows that during the earthquake the zone

of reservoir 1 appears to have the strongest amplitude increase on the TRM sections An experiment with random noise input revealed that the intrinsic properties of the velocity model in combination with noisy input data may also create areas with higher focusing amplitudes on TRM sections Possible spurious locations however can be identifi ed by a simple test with random noise input

The increase of PSD in the low frequency range (around 3Hz) at the surface above the reservoir and the increase of focusing amplitudes at the location of reservoir

1 at depth on the TRM section after the earthquake arrival suggest that the reservoir may have been stimulated by the earthquake The physical mechanism of the phenomenon

is still under research A possible explanation could be the modelling study by Frehner et al (2007) who demonstrated that micro-scale pore fl uid oscillations are able to change the frequency content of the large-scale elastic wave in

Fig 10 The diff erences caused by the earthquake (a) during the earthquake, (b) after

the earthquake Reservoir 1 is apparently highlighted during the earthquake

Fig 11 TRM of random noise input showing spurious high amplitude areas caused

by the heterogeneous velocity model

the low frequency range due to a coupling between pore

fl uid oscillations and the elastic wave propagation

Conclusions

The analysis of earthquake signals recorded at two

Voitsdorf oil and gas fi elds suggests that earthquakes can

modify the low-frequency wave fi eld around hydrocarbon

reservoirs These modifi cations are observable in the

spectra of passive seismic data in the form of some

increase of power spectral density within in the low

frequency range (around 3Hz) above the reservoir

Compared to the background of ambient noise,

the anomalous signatures are subtle and require

careful processing to be brought out Empirical mode

decomposition was found to be a promising tool for this

purpose

An application of time reverse modelling technique to

the data indicates that the reservoir zone is the source of

signals which create the low frequency energy anomalies

observed at the surface A physical mechanism is needed

to explain the phenomenon and more earthquake data

are desirable to confi rm the stability of the observations

found in this study

References

1 Berger, J., Davis, P., and Ekstrom, G 2004 Ambient

earth noise: A survey of the global seismographic network

Journal of Geophysical Research, 109, B11307

2 Birialtsev, E V., Plotnikova, I N., Khabibulin, I R., and

Shabalin, N Y., 2006 The analysis of microseisms spectrum

at prospecting of oil reservoir on Republic Tatarstan EAGE

Conference, Saint Petersburg, Russia, B016

3 Dangel, S., Schaepman, M E., Stoll, E P., Carniel,

R., Barzandji, O., Rode, E.-D., and Singer, J M., 2003

Phenomenology of tremor-like signals observed over

hydrocarbon reservoirs Journal of Volcanology and

Geothermal Research, 128(1), 135-158

4 Frehner M., Schmalholz S.M and Podladchikov

Y., 2007 Interaction of seismic background noise with

oscillating pore fl uids causes spectral modifi cations of

passive seismic measurements at low frequencies SEG

International Exposition and 77th Annual Meeting, San

Antonio, USA, Expanded Abstracts, 1307-1311

5 Graf, R., Schmalholz, S M., Podladchikov, Y., and

Saenger, E H., 2007 Passive low frequency spectral analysis:

Exploring a new fi eld in geophysics World Oil, 228(1), 47-52.

6 Holzner R., Eschle P., Zürcher H., Lambert M., Graf R.,

Dangel S and Meier P.F., 2005 Applying microtremor analysis

to identify hydrocarbon reservoirs First Break 23, 41-46

7 Huang, N E., Shen, Z., Long, S R., Wu, M L C., Shih, H H.,

Zheng, Q N., Yen, N C., Tung, C C., and Liu, H H The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis Proceedings of the

Royal Society of London Series a-Mathematical Physical and Engineering Sciences, 454, 903-995

8 Huang, N E and Shen, S S., 2005 The Huang transform and its applications Interdisciplinary

Hilbert-mathematical sciences; 5, World Scientifi c, New Jersey, 2005

9 Lambert, M., Schmalholz, S M., Saenger, E H and

Podladchikov, Y Y., 2007 Low-frequency anomalies in spectral ratios of single-station microtremor measurements: Observations across an oil and gas fi eld in Austria SEG

Expanded Abstracts, 26, 1352

10 Lambert, M., Schmalholz, S M., Saenger, Steiner,

B., 2008 Low-frequency microtremor anomalies over an oil and gas fi eld in Voitsdorf, Austria Geophysical Prospecting,

57, 1365-2478

11 Mastrigt, P., Al-Dulaijan, A., 2008 seismic Spectroscopy Using Amplifi ed 3C geophones EAGE 70th Conference & Exhibition, Rome, Italy, Expanded Abstracts, B047

12 McDonald, A J., Baumgaertne, A J., Fraser, G

J., George, S E., and Marsh, S., 2007 Empirical Mode Decomposition of the atmospheric wave fi eld Annales

Geophysicae, 25, 375-384

13 Saenger, E H., Torres, A., Rentsch, S Lambert, M.,

Schmalholz, S M and Mendez-Hermandez, E., 2007 A hydrocarbon microtremor survey over a gas fi eld: Identifi cation

of seismic attributes SEG Expanded Abstracts, 26, 1277.

14 Sole, J., Turiel, A., and J E Llebot J E., 2007 Using empirical mode decomposition to correlate paleoclimatic time-series Natural Hazards and Earth System Sciences, 7,

299-307

15 Steiner, B, Saenger, E H and Schmalholz, S M.,

2008 Time reverse modeling of low-frequency microtremors: Application to hydrocarbon reservoir localization

Geophysical Research Letters, 35, L03307

16 Turuntaev, S B., Burchik, V N and Turuntaev, D S.,

2006 Microseismic background study for gas fi eld exploration

EAGE Conference, Saint Petersburg, Russia, P245

1 Introduction

Predicting the natural gas dew point is of primary

importance in reservoir engineering, where liquid dropout

diminishes production from gas condensate wells It is

also equally important to consider it in the design of gas

transportation pipelines and custody transfer Natural

gas is an important energy and feedstock source, and it

appears poised to become the fastest growing source

of energy in the near future From an environmental

viewpoint, it is the least off ensive fossil fuel because it

burns clean and because it produces less greenhouse

gases than coal or oil The importance of natural gases

requires an accurate knowledge of their thermodynamic

properties over wide ranges of pressure and temperature

in order to optimize their production, transportation, and

utilization These properties can be measured with high

accuracy using state of the art techniques, and this is the

most reliable and accurate approach Nevertheless, the composition and thus properties of the natural gases can vary widely depending upon the reservoir from which the fl uid comes Hence, it is almost impossible, because of economical and time constraints, to measure all the properties for all the possible mixtures over the wide temperature and pressure ranges required The measurement of the required volumetric properties and vapor-liquid equilibrium in multicomponent mixtures, especially for high-pressure conditions such as those for deep reservoir gases, is expensive and diffi cult Measurements require considerable experimental skill and complex equipment, with the attendant fi nancial and time constraints [3] Thus, the demand for better knowledge of the properties requires reliable models that can both correlate and predict the properties with the required accuracy

Predictionofphaseequilibriumofnaturalgases mixtureswiththecubicandpolarPC-SAFT

equationsofstate

Nguyen Huynh Dong, Tran Thi Kim Siem

Petrovietnam Manpower Training College

Nguyen Huynh Duong

Petrovietnam Gas

Abstract

In this work, the authors tested the recently developed generalized method for correlating k

ij using only pure compound parameters combining with the Polar PC-SAFT EoS to predict thermodynamic properties of complex natural gas mixtures, using the simple one-fl uid mixing rule The correlation of k

ij is inspired by London’s theory of dispersive interactions, and uses “pseudo-ionization energies” [1] Phase envelopes of natural gas mixtures calculated with the equation of state are compared to highly accurate published data to check the validity of the model We compared the performance of the PR, SRK, PSRK and Polar PC-SAFT EoS in the calculation of the phase behavior of natural gases, such as the PR or SRK EoS routinely used by the oil and gas industry for the design of recovery and processing operations of gases at high pressures Nonetheless, many non-cubic equations of state, derived from statistical thermodynamics and with a solid theoretical basis, have been proposed in recent years [2] Among them, EoS of the SAFT-family are possibly those with the widest acceptance.

Calculations were performed for 16 natural gases for which experimental data are available in the literature For most mixtures, the prediction results of the Polar PC-SAFT EOS are in better (good) agreement with the experimental data.

Natural gases consist of many components: nitrogen

(N

2), carbon dioxide (CO

2), hydrogen sulfi de (H

2S) are usually the non-hydrocarbon components Methane

(CH4), ethane (C2H6), and other hydrocarbons (up to C40

or even C50) might be present in a natural gas mixture

Nitrogen and methane are always supercritical under

reservoir conditions, whereas the heavy hydrocarbons,

which cause condensation, are present under subcritical

conditions Also, new technologies allow exploration and

extraction of nonconventional, hyperbaric reservoirs in

which temperatures and pressures are extremely high [4]

Thus, the development of reliable and accurate models

is not simple From an historical viewpoint, equations

of state (EoS) have played a pivotal role in modeling the

thermodynamic properties of complex mixtures [5], and

they have been able to predict with some success phase

equilibrium for hydrocarbon reservoir fl uids [6]

The selected models are the cubic EoS which is a

simple but frequently applied equation for industrial

purposes, and the Polar perturbed-chain SAFT

(PPC-SAFT) model, which is among the most recent

modifi cations to the successful SAFT theory Although

this model has predicted accurately thermodynamic

properties [2], it is complex compared to simple cubic EoS

commonly used in the natural gas industry Cubic EoSs

are common choices in the chemical and gas industries

to model complex phase behavior and to design

chemical processes because of their computational

simplicity and the relatively good results obtained [5, 7]

EoSs are plentiful, the Redlich and Kwong [8] (RK) EoS, a

modifi ed version of it by Soave [9] (SRK), and the Peng

and Robinson [10] (PR) EoS are usually prescribed for

calculating natural gas dew points [5] However, many

EoSs, including some variants of the popular PR and SRK

EoSs, have been developed, and this trend seems to be

continuing Although cubic EoS are used widely for phase

equilibrium calculations in hydrocarbon fl uids, they

have drawbacks and limitations, such as low accuracy

for volumetric properties, that reduce their applicability

for complex mixtures such as natural gas [11] The raises

the question of whether the SRK and PR EoSs are still

the best for hydrocarbon processing? To answer this

question, the newly developed EoSs should be taken into

consideration Clearly, evaluation of all developed EoSs

may not be practical; however, it is likely that, based on

some criteria, one can select the appropriate EoSs and

study them to calculate natural gas dew points

The purpose of this study is to evaluate and recommend suitable EoSs by comparing the quality

of predictions of a simple EoS to a complex based EoS We use the Peng-Robinson and Soave Redlich and Kwong [9] EoS which is possibly the cubic EoS most widely used by the oil industry, and the Polar PC-SAFT EoS, which is a non-cubic EoS and one of the most recent modifi cations of the successful SAFT theory [12] Furthermore, EoS based on the SAFT approach have the potential advantage of being capable of modeling polar systems by including an association and/or polar contribution to the Helmholtz free energy of the system

molecular-In this way, an EoS such as PPC-SAFT may be useful to model operations such as dehydration of natural gases with glycols and their sweetening with amines

In a previous paper [13], diff erent kind of EoSs were used to predict the supercritical behavior of various binary and ternary mixtures, in terms of phase behavior, for wide ranges of pressure and temperature The behavior of components in a mixture is clearly diff erent than that in a pure state; however, these comparisons can still give good indications for the suitability of EoSs in predicting natural gas phase envelopes Afterwards, the accuracies of the EoSs are compared to predict the dew points of some natural gases that were reported recently

by diff erent authors [14-18]

There are accurate correlations/equations of state (EoSs) for calculating natural gas properties Except the SRK and PR EoSs, the other models are either complex or require many pure compound constants and/or binary parameters [19, 20] The SRK and PR EoSs are often used

in the gas industry as predictive tools In addition, these two EoSs can be used for predicting the properties of natural gas mixtures containing heavy fractions [11] The evaluation of EoS in predictions of natural gas phase behavior has been the subject of some recent publications Voulgaris et al [21] compare the performance of the PR and Simplifi ed-Perturbed-Hard-Chain Theory EOS in predictions of the condensation behavior of natural gases; Nasrifar et al [11, 22] compare cubic EOS in predictions of thermodynamic properties of natural gases; Martinez and Hall [23], Mørcha et al [24] compare three EOS: generalized three-parameter Redlich-Kwong/Peng-Robinson, Patel-Teja and PC-SAFT in the evaluation of thermodynamic properties of synthetic natural gas mixtures Louli et al [25] use UMR-PRU for predicting of the phase behavior of synthetic natural gas and oil systems…

2 The Equations of State

2.1 Soave-Redlich-Kwong EoS [9]

In 1972, Soave [9] proposed a new improved version

of Wilson’s idea He redefi ned the R(TR Z) function and

kept the Redlich-Kwong volume functionality, giving what

became, in a short period, one of the most popular EoS in

the hydrocarbon industry Computer simulation packages

became popular during the late 1960s, and there was

great need for a simple, generalized, and reasonably

accurate EoS for the many repetitive calculations required

in process simulations Soave’s EoS, commonly known as

the SRK equation, satisfi ed the need at that time Also,

in the mid-1970s, optimization of industrial processes

became extremely important, because of the so-called oil

crisis, and Soave’s equation was again of great help for the

type of calculations required in optimization algorithms

The SRK equation can be summarized as follows:

2.2 Peng-Robinson EoS [10]

After Soave’s proposal, many modifi cations were

presented in the literature for improving predictions of

one or another property [5] These SRK were not limited

to proposing new temperature models for R(TR, Z), but

also considered modifi cations of the volume dependence

of the attractive pressure term The most popular of all

of these modifi cations is the one proposed by Peng and

Robinson in 1976 [10] Peng and Robinson improved upon

Soave’s equation by recalculating the R(TR, Z) function and

by modifying the volume dependency of the attractive

term These changes allowed them to obtain better

results for liquid volumes and better representations of

vapor-liquid equilibrium (VLE) for many mixtures Peng

and Robinson’s equation is:

The SRK and the PR equations are the most popular cubic equations used in research, simulations and optimizations in which thermodynamic and VLE properties are required These two equations have been considered for all types of calculations, from simple estimations of pure-fl uid volumetric properties and vapor pressures

to descriptions of complex multicomponent systems Most modern computer process simulation packages (e.g ChemCAD, AspenPlus, Hysis, PRO/II, Prosim) include the SRK and PR equations among the thermodynamic options

Both SRK and PR EoS require three parameters for pure

compounds, namely, Tc, Pc, and ω (critical temperature, critical pressure and acentric factor)

2.3 Perturbed-Chain Statistical Associating Fluid Theory EoS [26]

PC-SAFT (perturbed-chain statistical associating

fl uid theory) is a novel equation of state, proposed by Gross and Sadowski [26] This is an equation of state based on a perturbation theory for chain molecules PC-SAFT is a variant of the original SAFT [12] EoS based

on Wertheim’s theory, and has already found several successful applications (especially for polymer and co-polymer solutions at both low and high pressures, for highly asymmetric, associating and/or polar mixtures, e.g alkanols, water, CO

2…); including vapor-liquid equilibria, liquid-liquid equilibria and gas solubilities for both single-solvent and mixed-solvents (ternary) systems In recent years, PC-SAFT [26] has been widely used to investigate the phase equilibria for both pure fl uids and mixtures [2,

27, 28] Compared with other EoS like SAFT or Cubic EoS, PC-SAFT is more precise for correlation of experimental data and more predictive when applied to mixtures.The detailed expression of PC-SAFT EoS is not recalled here (see the original papers for more details [26])

When polar compounds such as nitrogen or carbon dioxide are present in the considered systems, the use of

(1)

(2)

(3)

(4)(5)

(6)

(7)

(8)(9)

(10)

additional polar terms in the EoS improves

its prediction capability In terms of molar

free energy:

a = anon-polar + aQQ

Where a non-polar refers to a

PC-SAFT EoS and aQQ is a term accounting for

quadrupole - quadrupole interactions As

shown recently by diff erent authors

[29-31] an additional polar term as a

quadrupolar-quadrupolar does allow to improve signifi cantly

the representation of the phase diagram

of mixtures containing polar fl uids

Furthermore, a more predictive model

(k

ij = l

ij = 0) may be thus obtained as

demonstrated by Nguyenhuynh et al

[29], Dominik et al [31] and Tumakaka et

al [32] Indeed, Jog et al [33] argued that,

if the polar terms are neglected in the

EoS, the polar interactions are eff ectively

included in the Van der Waals attractions,

leading to overestimations of the energy

parameter Hij values As a consequence,

artifi cially large values of binary

interaction parameter must be needed for

representing accurately phase equilibria

of mixtures In that case, k

ij values may

be diffi cult to estimate, and the approach

is not suitable for predictive purposes

such as ours It is then preferable to take

into account explicitly the

quadrupolar-quadrupolar interactions

The term aQQ used here is obtained

by extending to chain molecules

Gubbins and Twu’s theory [34] originally

developed for spherical molecules The

extension here closely follows the

so-called ‘segment-approach’ proposed

by Jog et al [33] In this approach, the

multipole moments are assumed to be

well localized on certain segments of the

chain From this point of view, multipolar

interactions between chain molecules

are viewed as interactions between

polar spherical segments The extension

of Twu and Gubbins [34] theory is then

straightforward

The quadrupolar-quadrupolar contribution aQQ is obtained by perturbation techniques to the 3rd order involving only second and third order terms a2 and a3 (fi rst order term is zero) Since the perturbation series is slowly converging, it was re-summed as a Padé approximant to get a better agreement with Monte Carlo simulation [35] One reads:

The agreement between Monte Carlo simulation and expression (12)

is within a few percent for small and moderate polar moment (Twu and Gubbins [34]), but its quality decreases (5% and more deviation) for large values of the polar moment (Verlet and Weis [36])

The a2 & a3 terms in equation (12) may be written as follows:

In these expressions, it is assumed that a given molecule has one or several identical quadrupole moments Also, the charge distribution in the molecule is assumed to be highly symmetrical so that the quadrupole moment reduces to a scalar, while in fact it is a tensor

In these equations, Uis the total number density of molecules, mα is the chain length of moleculeDVDE is the arithmetic average of segments diameters VD and VE QD are the quadrupole moments of the polar segments in the chain compound D

In the equations above refer to the fractions of quadrupolar segments in the chain of component Dand should not be confused with

xD, the mole fraction of component D Note that the theory of Twu and Gubbins has been extended by recognizing that xpD.mD is simply the number of polar spherical segments in the molecule D

J is an integral over two-body interactions These integrals were fi tted

to reduced density and temperature using an empirical function Theses expressions may be found in [34]

In the computations, three molecular parameters are needed to obtain the thermodynamic properties of pure substances, namely, the segment size (V , dispersive energy (H/k), and chain length (m) Besides m V, and H/k, there is a predictable binary interaction parameter kij(predicted using “pseudo ionization energy” [1]) for each binary system

PRO/II and Aspentech07 process simulation packages

included the SAFT [12] and PC-SAFT [26] equations

among the thermodynamic options; Polar PC-SAFT [29]

will be included in PROSIM [37]

3 Mixing rules and binary Interaction Parameters

Van der Waals mixing rules with a geometric mean

average combining rule for the attractive parameter are

used to extend the EoSs to multicomponent mixtures In

most cases, the van der Waals mixing rules adequately

describe natural gas mixtures However, because

these mixing rules are not perfect, binary interaction

parameters (BIPs) are usually incorporated to correct the

interaction between binary molecules; these corrections

are represented by the parameter k ij in the EoSs Using

binary interaction parameters with the EoS provides more

fl exibility and reliability for multicomponent mixture

calculations [38] Usually, these parameters result from

least-squares minimization of experimental binary VLE

data

Description of the k ij correlation – prediction Method [1,

39, 40]:

PC-SAFT applies to mixtures using the van der Waals

one fl uid model [26] and modifi ed Lorentz-Berthelot

mixing rules that relate the potential parameters ε

ij and σ

ij

between segments of molecules i and j.

With kij and lij are binary interaction parameters

Phase equilibria of mixtures that do not contain

small compounds can generally be predicted without

use of any binary interaction parameters i.e k

ij = l

ij = 0 [41, 42] However, when a mixture involves such a small

assumed that l

ij = 0 (binary parameter for diameters) and

that k

ij (binary parameter for energy) may be predicted

using the following relation inspired by the theory of

London for dispersive interactions and modifi cation of

equation proposed by Hudson and McCoubrey [45]

London (1930) obtained an expression of the

dispersive interaction potential between molecule i and

molecule j, by solving the Schrödinger equation under

simplifying assumptions and obtained the following expansion: [46, 47]

This expression is often approximated by truncating

at the fi rst term:

with

Where D represents polarizability, h is Planck’s

constant, and ν0refers to the characteristic frequency of

a molecule in its unperturbed state, corresponding to its “zero-point energy” [48] The parameter ν0 is related

to an energy transition ΔE between electronic quantum states by the Bohr relationship ΔE = hν0 The term ΔE is

approximately equal to the fi rst ionization potentialI for a

molecule, so that C

6 can be estimated from equation (20)

by replacing hν0by I It may easily be shown, after small

rearrangements (Hildebrand et al [48]), that:

Given a theoretical expression for the deviation from the geometric rule, in terms of intermolecular potential The expression is dependent on the exact defi nition of the energy parameter ε ij in the potential u ij If

one assumes that ε ij is simply proportional to C6,ij, then [1]:

In view of making useful predictions for engineering applications using relation (21), it is necessary to propose

a method for computing pseudo-ionization energies

A geometric average inspired by the original Berthelot combination rule for the energy parameter is proposed [1]:

Lorentz-Where J

k is the contribution of group k to the

pseudo-ionization of the segments in the considered molecule [1]

4 Methodology

The fugacities of each component i in the vapor

and liquid phases should be equal in the dew and

bubble points:

Where x and y are the mole fractions in the liquid

and vapor phases, respectively Construction of the P-T

diagram requires a series of bubble point and a series of

dew point calculations for a mixture of fi xed composition

The computational procedure is devised in such a way

that it automatically selects whether the pressure or the

temperature should be specifi ed, by tracking the value of

the derivative pressure with respect to temperature along

the saturation line In this way, impossible specifi cations

such as a temperature larger than the cricondentherm or

a pressure larger than the cricondenbar are avoided

In the case of cubic EoS, the binary interaction

parameters, critical parameters and acentric factors of

the EoS were set equal to that of recommended values existing in HYSYS process software version 2006 in order to compare the EoS as predictive tools, instead

of evaluating their ability to correlate experimental behavior based on fi tted parameters Parameters for the Polar PC-SAFT EoS were taken from the work of NguyenHuynh et al [49, 50] The global phase stability

of all calculated bubble, dew points was verifi ed by the tangent plane criterion [51]

5 Results and Discussions

In the current work, we compare phase envelopes for multicomponent mixtures to the EoS predictions We have considered literature NG mixtures with hydrocarbons no higher than n-octane (light NG mixtures) Tables 1 and 2 contain the compositions of 16 NG mixtures considered

in this work

(23)

Table 1 Natural gas mixture compositions (% molar) and code names Experimental data are taken from refs [16, 17]

Table 2 Natural gas mixture compositions (% molar) and code names Experimental data are taken from refs [14, 15, 18]

Prediction of phase envelopes in multicomponent

mixtures such as natural gas is a stringent test for any

theoretical model It is a common practice in the chemical

and gas industries to calculate phase envelopes using

cubic EoS such as PR or SRK Most software for process

design also uses these simple models An analysis of the

scarce, highly accurate experimental phase envelopes

available in the literature shows that these EoS do not

produce accurate predictions, with errors up to 5 K at the

cricondentherm and up to 3 MPa at the cricondenbar [11,

22] These errors are too high for most practical purposes,

and many industrial designs are overcompensated In our

opinion, it is not legitimate to apply these EoS to such

systems Here, we test the PR, SRK, PSRK phase envelope

predictions, comparing them to those obtained from the

PPC-SAFT

The behavior of components in a multicomponent

mixture can be signifi cantly diff erent than in a pure state,

depending on the diff erence in size, shape, and polarity of

components

The natural gases used in this study are considerably

lean in heavy hydrocarbons A real natural gas is usually

rich in heavy hydrocarbons All heavy hydrocarbons

are usually grouped together and reported as C6+ We

selected 16 natural gases NG in which all components

are well characterized substances, i.e., without lumped

heavy fractions such as C

6+ This eliminates the eff ect the lumping procedure might have in the EoS comparison

In their respective original references, these NG are

formulated in such a way that properties such as average

mole mass, density, and composition, are similar to those

of actual natural gases This subject is under studied

Figs 1 - 4 shows the phase envelopes calculated from

diff erent EoS for NG1 to NG4 We have calculated the

full phase envelope, from low to high temperatures to

illustrate the diff erent behavior of the three EoS

Figs 5 - 8 shows the dew points of mixture NG5 to

NG8, as predicted by the PR and PPC-SAFT EoSs and

experimental values are also shown Clearly, the PPC-SAFT

EoSs are in better agreement with experimental data than

those of the PR

The experimental data show that NG5 has a peculiar

behavior compared to the other systems studied: its

cricondenbar and cricondentherm are similar, and none

of the two families EoS could predict this behavior

satisfactorily

The phase diagram of NG6, NG7, NG9, NG10, NG16

is in good agreement with experimental dew point data from the low pressure region up to the cricondentherm Deviations become larger as the pressure increases The PPC-SAFT EoS performs slightly better than the PR or SRK or PSRK EoS, but the opposite happens in NG5, NG8, NG11, NG13 with all models giving poor predictions of the critical coordinates

The phase envelope of NG10, Fig 10, reveals huge diff erences from the three EoS Because no literature data exist for the bubble point curve, we cannot deduce which

of the EoS is more suitable for this sample, although it seems that PPC-SAFT produce better results

The predictions for NG11 and NG13, Figs 11 & 13, are not very good; the results for the dew point curve are essentially equal from the three EoS When the pressure increases and approaches the cricondenbar, huge diff erences appear with all the EoS, although the predictions obtained with PPC-SAFT and SRK are slightly better than those obtained with the PR

In Fig 15, the dew points of a natural gas mixture NG15 are predicted with and without BIPs by the PR EoSs and compared to experimental data The BIPs have been taken from the HYSYS software version 2006 As Fig 15 shows, the use of BIPs signifi cantly improves the prediction accuracy of the EoSs for the dew points of mixture NG15 Also in this Fig., it indicates that PPC-SAFT gives the best results among the three EoS While

PR (using all recommended kij value) and SRK produce low/high deviations for the dew point curve, when the pressure increases to the cricodenbar, they have large deviations

Results for NG12, NG14 and NG16 plotted in Figs

12, 14 and 16 shows that all the EoS predict the phase envelope well Once again, however, higher deviations appear at the cricodenbar, but for these samples SRK and PPC-SAFT are not too far removed from the experimental results For the low-temperature bubble points, the predictions from all four EoS are almost identical The PPC-SAFT EoS is in good agreement with the experimental data and considerably superior with respect to the others However, the SRK EoS is also in agreement with the measured data (Figs 13, 14, 16)

The behavior of NG16 is very well predicted by the PC-SAFT EOS

Overall, the PPC-SAFT predicts the phase envelops

of natural gas systems quite accurately The SRK and PR

EoSs are ranked second and third, respectively From the

analysis of phase envelope predictions for NG1-NG16

using the EoS, we arrive at several conclusions:

- PPC-SAFT EoS results are in better agreement with

experimental data than those of the PR, SRK or PSRK EoS

in most cases

- All EOS provide very similar predictions of low

pressure dew points up to the cricondentherm, even

though the results of the PPC-SAFT EoS are generally

closer to experimental data

- Bubble point curves predicted by both kind of EoS

families are usually very similar Experimental bubble

points were available only for mixtures NG1 to NG4

Given its good performance in predictions of natural

gas phase behavior and its potential ability to model

polar systems accurately, the PPC-SAFT EoS may be a

good model to help design dehydration and sweetening

operations [52, 53], but its ability to do so remains to be

investigated

- In most of cases, PPC-SAFT has lower deviations,

and PR provides the worst prediction for these systems

(including cricondenbar point) All EoS predict the

low-temperature bubble point curve equally

- PPC-SAFT appears to be superior compared to

PR for natural gas systems containing nitrogen and not carbon dioxide

- Cricodenbars are usually underestimated by all the EoS, but PPC-SAFT predicts cricodenbars with the lowest errors

- Predictions for the cricodentherm with SRK, PR or PSRK for very rich methane containing system (NG10) are worse than for the leaner samples Except for the PPC-SAFT EoS, excellent results of prediction was obtained comparing to experimental data

- Predictions for the low-temperature bubble points are almost the same with the four EoS except for sample NG10 for which cubic EoS show huge diff erences from the measured data

- The presence of non hydrocarbon compounds, e.g nitrogen or carbon dioxide in samples have a signifi cant eff ect upon the phase envelope of the mixtures and decrease the accuracy of predictions with the studied models

Based on the results obtained in this study, we believe that PPC-SAFT EoS exhibits clear superiority over the other cubic EoS studied, although its cricodenbar high pressure predictions are not stellar

Fig 2 Experimental data and predicted phase envelopes

for the natural gas mixture NG2

Fig 1 Experimental data and predicted phase envelopes

for the natural gas mixture NG1

Fig 7 Experimental data and predicted phase envelopes

for the natural gas mixture NG7

Fig 3 Experimental data and predicted phase envelopes

for the natural gas mixture NG3

Fig 4 Experimental data and predicted phase envelopes

for the natural gas mixture NG4

Fig 5 Experimental data and predicted phase envelopes

for the natural gas mixture NG5

Fig 6 Experimental data and predicted phase envelopes for

the natural gas mixture NG6

Fig 8 Experimental data and predicted phase envelopes

for the natural gas mixture NG8

Fig 9 Experimental data and predicted phase envelopes

for the natural gas mixture NG9

Fig 11 Experimental data and predicted phase envelopes

for the natural gas mixture NG11

Fig 12 Experimental data and predicted phase envelopes

for the natural gas mixture NG12

Fig 10 Experimental data and predicted phase envelopes

for the natural gas mixture NG10

Fig 13 Experimental data and predicted phase envelopes

for the natural gas mixture NG13

Fig 14 Experimental data and predicted phase envelopes

for the natural gas mixture NG14

PRO/II and Aspentech07 process simulation packages

included the SAFT [12] and PC-SAFT [26] equations

among the thermodynamic options; Polar PC-SAFT [29]... three EoS

Figs - shows the dew points of mixture NG5 to

NG8, as predicted by the PR and PPC-SAFT EoSs and

experimental values are also shown Clearly, the PPC-SAFT

EoSs... samples SRK and PPC-SAFT are not too far removed from the experimental results For the low-temperature bubble points, the predictions from all four EoS are almost identical The PPC-SAFT EoS is in