Thermodynamics in Gas Processing –Phase Envelope Predictions and Process Design

Efstathios (Stathis) Skouras, Principal Researcher

Gas Processing, StatoilHydro Research Centre Trondheim

1st Trondheim Gas Technology Conference, 21-22 October 2009

2

Presentation outline

• Importance of correct phase envelope predictions

• Hydrocarbon dew points for real gases

– Success factors

– Predictions with thermodynamic models

• Phase envelope predictions: Impact on process design

• Conclusions

0102030405060708090

100110120

-10 -5 0 5 10 15 20 25Temperature [°C]

Pres

sure

[bar

a]

Individual dew points

3

Phase envelope of a typical natural gas

0

20

40

60

80

100

120

-50 -40 -30 -20 -10 0 10 20 30 40

Temperature/°C

Crit P

EOS = SRK Penelou

Dew point lineDew point line

GasGas

Two Two phasesphases

Bubble point lineBubble point line

CricondenbarCricondenbar

LiquidLiquid

CricondenthermCricondentherm

Dense phaseDense phase

4

Importance of correct phase envelope predictions• In offshore processing the cricondenbar

specification must be fulfilled to avoid condensation in the pipelines

• Rich gas transported to onshore terminals in dense phase in pipelines up to 830 km long

• In onshore processing the cricondenthermspecification must be fulfilled to achieve desired gas quality for the sales gas

Correct phase envelope predictions

•• Efficient operation/separation

• Optimise pipeline capacity

• Reduce design margins

5

Success factors

• Gas sampling and conditioning

• Chromatographic gas analysis

• Characterisation of C7+ components

• Thermodynamic models

Dew points for real gases: Success factors

6

Success factors: Gas sampling and conditioning

Sampling at places where the gas is at:

• High temperature (one phase area)

• High pressure (in dense phase)

1st stage separator

2nd stage scrubber

1st stage scrubber

Glycol contactor

3rd stage scrubber

Rich gas (RG)

Right place!

Wrong place!

7

Success factors: Chromatographic gas analysis• How detailed should the gas analysis be?

• Enough with GC-analysis with a C6+, C7+ fraction?

• Does it really matter?

Example: Rich gas with C6+ fraction of 0.38 mol%

• Deviations up to 20°C at the cricondentherm

• Deviations up to 10 bar at the cricondenbar

• The leaner the gas, the more effect the heavy ends have in the dew point

• Detection limit for lean gases 0.1 ppm for the heavy ends (ISO 23874)

0

10

20

30

40

50

60

70

80

90

100

110

120

-35 -25 -15 -5 5 15 25 35

Temperature [°C]

Pre

ssur

e [b

ar]

GC-analysis up to C5 (C6+ fraction)

GC-analysis up to C6 (C7+ fraction)

GC-analysis up to C9

8

Success factors: Characterisation of C7+ components

• Is it enough to characterise C7+ components as n-alkanes?

• Is it important to distinguish between paraffinic (P), napthenic (N) and aromatic (A) components?

• Shall we characterise them as pseudo-components (C7*, C8*, etc)?

• How should we assign physical properties (mol. weight, density) and model parameters (Tc, Pc, ω) to the pseudo-components?

• Does it really matter?

Effect of characterisation of C7+ components

• Deviations up to 15°C at the cricondentherm

• Deviations up to 7-8 bar at the cricondenbar

0

10

20

30

40

50

60

70

80

90

100

110

120

-15 -10 -5 0 5 10 15 20 25 30 35

Temperature [ºC]

Pres

sure

[bar

a]

C7+ components as n-alkanes

PNA characterisation

C7+ components as pseudo-components

C7+ as pseudo-components

• Mol.weight and density by detailed GC-analysis

• Tc, Pc and ω from generalised correlations as function of mol.weight and density *

* Tc, Pc from Riazi-Daubert (1987) and ω from Kesler-Lee (1976)

9

• Characterisation with PNA analysis gives better results than assuming C7+ as n-alkanes

• Characterisation with pseudo-components (utilising correct mol.weight and density and generalised correlations for Tc, Pc and ω) provides the correct shape of the phase envelope

• Cricondenbar underestimated. No model can predict both the cricondenbar and the cricondentherm

• Experimental dew point measurements still required to decide the correct phase envelope

Dew points of real gases: Results for a rich gas

0

10

20

30

40

50

60

70

80

90

100

110

120

-15 -10 -5 0 5 10 15 20 25 30 35

Temperature [ºC]

Pres

sure

[bar

a]

C7+ as n-alkanes/SRK

PNA characterisation/SRK

C7+ as pseudo-components/SRK

C7+ as pseudo-components/PR

Experimental

10

Phase envelope predictions: Impact on process design

Testseparator

1st stage separator

2nd stage scrubber

1st stage scrubber

Glycol contactor

3rd stage scrubber

Rich gas (RG)

• The rich gas (RG) has to meet a specification of 105 barg spec

• The quality of the gas is decided at the 2nd stage scrubber (operating point: 40 barg, 20°C)

• The model predicts a cricondenbar of 100 barg

• Design margin of 5 bar to cricondenbar spec0

102030405060708090

100110120

-20 -10 0 10 20 30Temperature [°C]

Pres

sure

[bar

a]

Calculated dew point curve 2nd stage scrubber outlet

Operating point 2nd stage scrubber

EOS = SRK

Rich Gas Cricondenbar

11

Phase envelope predictions: Impact on process design (cont.)

Testseparator

• Experimental measurements show give a cricondenbar of 106 barg. Gas is off spec!

• Design margin of 5 bar not sufficient. Higher design margin needed (10 bar)

• Improve accuracy of the model in order to reduce the design margin

30

40

50

60

70

80

90

100

110

-20 -15 -10 -5 0 5 10 15 20 25Temperature [°C]

Pres

sure

[bar

a]

Calculated dew point curve 2nd stage scrubber outlet

Measured dew points

Scrubber operating point

CCB prediction 6 bartoo low

EOS = SRK

12

Conclusions• Focus on sample chain is decisive (sample taking,

conditioning and GC-analysis)

• Characterisation of the heavy ends (C7+) is crucial

• The thermodynamic models are not capable to model sufficiently the whole phase envelope

• The models underpredict the cricondenbar

• Experimental dew point measurements are still needed to verify the model predictions

• Focus on thermodynamic models in order to achieve good process designs, reduce design margins and ensure product quality

Source: www.statoilhydro.com

13

Thank you for your attention!

14

Back-up slides

15

Gas in

Gas out

Cooling media inCooling media in

Cooling media outCooling media out

Mirror

Chilled mirror optical apparatus from

Chandler Engineering

0102030405060708090

100110120

-10 -5 0 5 10 15 20 25Temperature [°C]

Pres

sure

[bar

a]

Individual dew points

Dew points for real gases: Experimental measurements

16

Dew point predictions: Pure components (ethane)• Agreement between experimental dew points and predictions with SRK and PR EoS

Dew points of ethane

10

15

20

25

30

35

40

45

50

-20 -15 -10 -5 0 5 10 15 20 25 30 35

Temperature [°C]

Pres

sure

[bar

]

Simulated SRK EoS (Pro/II)

Simulated PR EoS (Pro/II)

Experimental measurements(Rotvoll)

17

• Both cricondenbar and cricondentherm is under-predicted

• Deviations up to 5°C in cricondentherm and 15 bar in cricondenbar

• Good accordance at low pressure

• Shape of the experimental dew point line is different from predicted by the model

SNG1

0

10

20

30

40

50

60

70

80

90

100

110

-30 -25 -20 -15Temperature [°C]

Pres

sure

[bar

a] PR

SRK

PC-SAFT

Gerg

Lab

Dew point predictions: Simple synthetic gases (up to C5)

Component SNG1

Methane 93.505

Ethane 2.972

Propane 1.008

i-Butane 1.050

n-Butane 1.465

n-Pentane -

18

SNG6 & SNG7 & SNG8 - Effect of PNA

0

10

20

30

40

50

60

70

80

90

100

110

-20 -15 -10 -5 0 5 10 15 20Temperature [°C]

Pres

sure

[bar

a]

ParaffinLab ParaffinNaphteneLab NaphteneAromaticLab Aromatic

• Cricondenbar still under-predicted, but cricondentherm over-predicted

• Aromatic and naphtene compounds give significant steeper dew point line than paraffins

• PNA characterization important for phase envelope prediction

Synthetic gases with a selected C7 component

Component SNG7 SNG8 SNG9

Methane 93.121 93.176 83.940

Ethane 3.048 3.064 10.016

Propane 0.994 1.014 4.109

i-Butane 1.032 1.027 0.601

n-Butane 1.510 1.521 1.031

i-Pentane - - -

n-Pentane - - -

n-Hexane - - -

Benzene (A) 0.295 - -

N-Heptane (P) - 0.198 -

Cyclo-Hexane (N) - - 0.302

19

Sales gas

Onshore processing

Rich gas

Offshore processing

Industry

Household

Power plant

Other

Terminal

Market

Cricondenbar spec

Cricondentherm spec

Correct phase envelope predictions

•• Efficient operation/separation

• Optimise pipeline capacity

• Reduce design margins

Importance of correct phase envelope predictions