What is dense-phase CO₂ and why is it used for pipeline transport?

Dense-phase CO₂ refers to CO₂ at conditions above its critical point (304.13 K, 73.77 bar) where it exhibits liquid-like density (~700-900 kg/m³) and gas-like viscosity (~50-80 µPa·s). Pipeline transport uses dense phase because the high density allows much smaller pipe diameters and lower pressure drop per unit mass than gas-phase transport, while avoiding the two-phase risk that would cause slug flow and damage equipment.

Why is the Peng-Robinson equation of state inadequate for final CO₂ pipeline design?

PR-EOS, even with Peneloux volume correction, has 3-5% density error in the dense phase region near the critical point. For final design, the Span-Wagner (1996) reference equation of state — implemented in NIST REFPROP — provides density accuracy better than 0.05% across the full fluid surface. PR is acceptable for screening and front-end engineering but should be replaced by SW96 for detailed design.

What is the DNV-RP-J202 velocity limit for CO₂ pipelines?

DNV-RP-J202 recommends keeping CO₂ pipeline velocity below approximately 4 m/s for dense-phase service to avoid erosion-corrosion issues at fittings, bends, and reducers. This limit is more restrictive than the API RP 14E erosion velocity (122/√ρ ≈ 4.5 m/s for dense CO₂), reflecting the higher density and momentum of dense-phase CO₂ vs natural gas.

What pipeline pressure should be maintained for dense-phase CO₂ transport?

For supercritical CO₂ (T > 31°C) pipeline operating pressure should remain above approximately 85 bara to keep the fluid in the dense phase and avoid the rapid density change near the critical point. For subcritical CO₂ (T < 31°C), pressure must remain above the saturation pressure at the operating temperature, with a typical 5-10 bar margin to prevent two-phase flow.

Why are CO₂ pipeline booster stations typically spaced 100-300 km apart?

Booster station spacing is set by the pressure drop budget — typical 100-300 km between stations for trunk-line dense-phase CO₂ service. Closer spacing (50-100 km) is used for high-throughput trunk lines (>500 t/h), while wider spacing (>300 km) is possible for low-throughput or large-diameter lines. The spacing is bounded by the inlet pressure (typically 150-180 bara) and the minimum suction pressure required to keep the fluid in dense phase (typically 85-100 bara above the critical point).

CO₂ Dense-Phase Transport Fundamentals

Q: Why are CO₂ pipeline booster stations typically spaced 100-300 km apart?

Booster station spacing is set by the pressure drop budget — typical 100-300 km between stations for trunk-line dense-phase CO₂ service. Closer spacing (50-100 km) is used for high-throughput trunk lines (>500 t/h), while wider spacing (>300 km) is possible for low-throughput or large-diameter lines. The spacing is bounded by the inlet pressure (typically 150-180 bara) and the minimum suction pressure required to keep the fluid in dense phase (typically 85-100 bara above the critical point).

1. Overview

Carbon Capture, Utilization, and Storage (CCUS) projects are scaling rapidly. The transport leg moves captured CO₂ from emission point to injection well — typically 10–500 km via pipeline. CO₂ pipeline engineering shares many tools with natural-gas hydraulics but differs in three important ways:

Phase region: CO₂ is transported in dense phase (supercritical or compressed liquid), not gas. The dense-phase density (~700–900 kg/m³) is 50–100× higher than natural gas at the same pressure, allowing much smaller pipe diameters.
EOS sensitivity: Density varies sharply near the critical point (304.13 K, 73.77 bar). Cubic equations of state (Peng-Robinson, SRK) have 3–5% error in this region; Span-Wagner (1996) is the reference for final design.
Phase boundary management: A pressure drop into the two-phase region produces slug flow that damages valves, pumps, and meters. Pipeline pressure must stay above the saturation curve (subcritical T) or the critical pressure (supercritical T) at every point.

Why dense phase? A dense-phase CO₂ pipeline at 150 bara, 35 °C carrying 500 t/h needs only a 12-inch line. The same flow as gas at 50 bara would require a 36-inch line. The density advantage is decisive.

Standards governing CO₂ pipeline design

Standard	Scope
ASME B31.4	Pipeline Transportation Systems for Liquids and Slurries — covers CO₂ in dense / supercritical phase
ISO 27913:2016	CO₂ capture, transportation and geological storage — Pipeline transportation systems
DNV-RP-J202 (2017)	Design and Operation of CO₂ Pipelines — velocity limits, fracture arrest, materials
API RP 1186	Recommended Practice for CO₂ pipeline operations (recently issued)
EU CCS Directive	2009/31/EC — regulatory framework for CO₂ transport in Europe

2. CO₂ Phase Behavior

Pure CO₂ has a triple point (216.59 K, 5.18 bar) where solid, liquid, and vapor coexist, and a critical point at 304.1282 K (30.98 °C) and 7.3773 MPa (73.77 bar). Above the critical point CO₂ is a single supercritical phase; below, it can exist as solid (rare in pipelines), liquid, or vapor.

Wagner-form saturation pressure (Span & Wagner 1996)

The vapor pressure curve from triple point to critical point is given by:

ln(P_sat / P_c) = (T_c/T) · [a₁τ + a₂τ^1.5 + a₃τ² + a₄τ⁴] τ = 1 − T/T_c a₁ = −7.0602087 a₂ = 1.9391218 a₃ = −1.6463597 a₄ = −3.2995634 T_c = 304.1282 K P_c = 7.3773 MPa

Sample values from this equation:

T (°C)	P_sat (bara)	ρ_{liq sat} (kg/m³)	ρ_{vap sat} (kg/m³)
−40	10.1	1117	26.6
−20	19.7	1032	52.9
0	34.9	928	97.6
10	45.0	861	135
20	57.3	773	194
25	64.3	713	240
30	71.9	594	344
30.978 (T_c)	73.77	467.6	467.6

Phase regions for pipeline operation

For dense-phase pipeline transport, the operating point must remain in one of two single-phase regions:

Supercritical (T > T_c): Single fluid phase regardless of pressure. Density varies smoothly with pressure but rapidly near the critical point. Typical operation: 35–50 °C at 100–200 bara.
Subcritical compressed liquid (T < T_c, P > P_sat(T)): Dense liquid phase. Used in colder climates. Operating pressure must be at least 5–10 bar above P_sat(T) to maintain margin.

Two-phase risk: Crossing the saturation curve produces vapor bubbles in the line. The resulting slug flow damages valves and pumps, causes meter inaccuracy, and creates J-T cooling that can drop steel below its brittle-to-ductile transition. Must be avoided at every operating point AND every transient.

Sublimation curve

Below the triple point (T < 216.59 K), CO₂ can sublime directly from solid to vapor. The sublimation pressure equation (Span-Wagner) is:

ln(P_sub / P_t) = (T_t/T) · [b₁τ + b₂τ^1.9 + b₃τ^2.9] τ = 1 − T/T_t b₁ = −14.740846 b₂ = 2.4327015 b₃ = −5.3061778 T_t = 216.59 K P_t = 5.18 bar

This region matters for emergency depressurization analysis (calc A6) — at atmospheric pressure CO₂ sublimes at −78.5 °C, meaning rapid blowdown can produce dry ice and very low metal temperatures.

3. Density & Viscosity Correlations

Density: Peng-Robinson + Peneloux (screening) vs Span-Wagner (final)

The Peng-Robinson cubic equation of state (Peng & Robinson 1976) provides screening-quality density estimates:

P = RT/(V−b) − a(T)/[V(V+b) + b(V−b)] a(T) = 0.45724 · R²·T_c² / P_c · α(T) b = 0.07780 · R · T_c / P_c α(T) = [1 + κ(1 − √(T/T_c))]² κ = 0.37464 + 1.54226·ω − 0.26992·ω² For CO₂: ω = 0.22394

The Peneloux volume translation (Peneloux et al. 1982) corrects PR's systematic density bias:

V_corrected = V_PR − c c = 0.40768 · R·T_c/P_c · (0.29441 − Z_RA) Z_RA = 0.2722 for CO₂ (Spencer-Adler)

This combination achieves ~3–5% density accuracy in the dense phase region — adequate for screening and front-end design.

P (bara)	T (°C)	ρ Span-Wagner (kg/m³)	ρ PR-Peneloux (kg/m³)	Δ%
100	30	768	754	−1.8%
100	40	631	608	−3.6%
150	30	848	838	−1.2%
150	40	783	761	−2.8%
200	50	841	815	−3.1%

For final design: Use NIST REFPROP or a process simulator (HYSYS, ProMax) with the SW96 EOS. The error band of PR-Peneloux is acceptable for screening but can mistake a single-phase point for two-phase near the saturation curve.

Viscosity: Fenghour-Vesovic-Wakeham (1998)

The reference viscosity correlation for CO₂ is from Fenghour, Vesovic & Wakeham (J. Phys. Chem. Ref. Data 27, 1998, p. 31):

η(ρ,T) = η₀(T) + Δη(ρ,T) [µPa·s] Zero-density limit (gas): η₀(T) = 1.00697 · √T / Ψ*(T) Ψ*(T) = exp(Σ aᵢ·(ln T*)ⁱ) T* = T / 251.196 K a₀ = 0.235156 a₁ = −0.491266 a₂ = 5.2112e−2 a₃ = 5.3479e−2 a₄ = −1.5371e−2 Excess viscosity (high density): Δη(ρ,T) = d₁₁·ρ + d₂₁·ρ² + d₆₄·ρ⁶/T_r³ + d₈₁·ρ⁸ + d₈₂·ρ⁸/T_r T_r = T / 251.196 K d₁₁ = 4.0711e−3 d₂₁ = 7.1980e−5 d₆₄ = 2.4117e−16 d₈₁ = 2.9711e−22 d₈₂ = −1.6279e−22

FVW achieves ~1% accuracy across the full fluid surface and is implemented exactly in NIST REFPROP. Sample values:

P (bara)	T (°C)	ρ (kg/m³)	µ (cP)
100	30	768	0.072
150	35	836	0.067
150	50	746	0.063
200	40	872	0.078

Note that dense-phase CO₂ viscosity is roughly 1/15 that of water (1 cP) — very low — which is why high Reynolds numbers (10⁶–10⁷) are typical even in trunk pipelines.

4. Hydraulics & Pressure Drop

Darcy-Weisbach equation

The basic incompressible-flow pressure drop equation applies (with care to track density variation along the pipe):

ΔP = f · (L/D) · (ρ V² / 2) f = friction factor (Colebrook-White, see below) L = pipe length (m) D = pipe inside diameter (m) ρ = fluid density (kg/m³) V = mean velocity (m/s) = ṁ / (ρ · A) A = π D² / 4

Colebrook-White friction factor

For turbulent flow (Re > 4000) the friction factor is solved iteratively from:

1/√f = −2 · log₁₀(ε/(3.7·D) + 2.51/(Re·√f)) Re = ρ V D / µ ε = pipe absolute roughness (default 0.0457 mm for commercial steel)

The Swamee-Jain explicit form is sometimes used as a starting estimate but Colebrook should be iterated to convergence (typically 3–5 iterations).

Why segment the integration

For a long dense-phase CO₂ pipeline, density varies significantly between inlet and outlet (e.g., 836 kg/m³ at 150 bara vs 681 kg/m³ at 100 bara). Using a constant inlet density gives a pressure drop estimate that is too low (V² grows as ρ drops). Industry practice is to integrate Darcy-Weisbach over N segments:

Divide pipeline length L into N equal segments (typically N=20).
For each segment, evaluate ρ(P,T) at the segment inlet.
Compute ΔP for that segment with local ρ and V.
Propagate outlet pressure to the next segment.

For ΔP < 20% of inlet pressure, N=20 segments converges well. For larger ΔP, increase N to 50–100 or use a streamline integration (RK4).

Engineering rule of thumb: Dense-phase CO₂ pipelines typically operate at 0.5–2 bar/km pressure drop. Below that, the pipe is over-sized; above 3 bar/km, booster spacing becomes uneconomic.

Elevation change

Hydrostatic pressure adds or subtracts depending on flow direction:

ΔP_elevation = ρ · g · Δz Δz = elevation change (outlet − inlet, m) g = 9.80665 m/s²

For dense-phase CO₂, hydrostatic effects are significant: ρ·g·Δz ≈ 8 bar per 100 m of elevation rise at 800 kg/m³.

5. Velocity & Diameter Sizing

DNV-RP-J202 erosion velocity

DNV-RP-J202 recommends ≤ 4 m/s for dense-phase CO₂ pipelines, more restrictive than the API RP 14E criterion of v_e = C/√ρ (with C=100 metric, giving ~4.5 m/s for 800 kg/m³ CO₂). The DNV limit reflects observed erosion-corrosion at fittings, bends, and reducers under dense-phase momentum.

Velocity range	Application	Notes
1–2 m/s	Trunk lines, conservative design	Comfortable margin to erosion limit; lowest pressure drop
2–3 m/s	Typical design point	Optimum balance of pipe cost vs ΔP
3–4 m/s	High-throughput trunks, distribution headers	At DNV limit; verify fitting selection
> 4 m/s	Not recommended for dense-phase CO₂	Re-size or accept higher integrity inspection

Diameter sizing approach

The diameter sizing problem is: given mass flow ṁ, length L, inlet pressure P₁, minimum outlet pressure P₂, and temperature T, find the smallest standard NPS that meets:

Outlet pressure ≥ P₂_min (single-phase margin)
Maximum velocity ≤ 4 m/s (DNV-RP-J202)

The standard approach iterates over candidate IDs. Pipeline calc A3 (CO₂ Pipeline Diameter Sizing) automates this for NPS 6"–48".

NPS	OD (mm)	ID @ STD wall (mm)	Typical max ṁ at 4 m/s, ρ=800 kg/m³ (kg/h)
6"	168.3	154.1	215,000
8"	219.1	202.7	372,000
10"	273.0	254.5	586,000
12"	323.85	304.8	843,000
16"	406.4	387.4	1,360,000
20"	508.0	489.0	2,166,000
24"	609.6	590.6	3,160,000
30"	762.0	736.6	4,920,000

6. Booster Station Spacing

Long CO₂ trunk pipelines (> 100 km) need intermediate booster stations to maintain dense-phase pressure. The number and spacing of stations is set by the inlet pressure (typically the discharge of upstream compression at 150–180 bara) and the minimum acceptable suction pressure for the next booster (typically 85–100 bara to keep the fluid above the critical pressure).

Industry practice

Pipeline	Length	Diameter	Throughput	Boosters
Cortez (Texas–Colorado, 1984)	808 km	30"	23 Mtpa	5 boosters
Sheep Mountain (Colorado, 1972)	660 km	20"/24"	9 Mtpa	3 boosters
Weyburn (USA–Canada, 2000)	330 km	14"/16"	5 Mtpa	2 boosters
Snøhvit (Norway, 2008)	153 km offshore	8"	0.7 Mtpa	0 (single-stage)
Quest (Alberta, 2015)	65 km	12"	1.2 Mtpa	0 (single-stage)

Typical spacing is 100–300 km for dense-phase trunk lines, set by:

Allowable pressure drop window (P_discharge − P_{min suction}): typically 80–100 bar
Pipeline ΔP/km (set by diameter, flow, and velocity limit)
Topography (large elevation rises require more booster stations)

7. Worked Example

Problem: A CCUS project requires transporting 500 t/h of pure CO₂ over 50 km. Inlet conditions are 150 bara at 35 °C. Available pipe sizes are NPS 10", 12", 14", 16". Determine pressure drop and recommended pipe size.

Step 1: Confirm dense-phase regime. T = 35 °C > T_c (30.98 °C), so we are supercritical. Inlet P = 150 bara >> P_c (73.77 bara) — comfortably in the dense supercritical region.

Step 2: Estimate density and viscosity at inlet.

From PR-Peneloux at 150 bara, 35 °C: ρ ≈ 836 kg/m³ From FVW at ρ=836, T=308 K: µ ≈ 0.067 cP = 6.7×10⁻⁵ Pa·s

Step 3: Try 12" NPS (ID = 304.8 mm).

A = π(0.3048)²/4 = 0.0729 m² ṁ = 500,000 / 3600 = 139 kg/s V = 139 / (836 × 0.0729) = 2.28 m/s ✓ (≤ 4 m/s) Re = 836 × 2.28 × 0.3048 / 6.7e−5 = 8.7 × 10⁶ Colebrook with ε/D = 4.57e−5/0.3048 = 1.5e−4: f ≈ 0.0131 ΔP per km (with ρ=836, constant): = f × (1000/D) × (ρV²/2) = 0.0131 × (1000/0.3048) × (836 × 2.28² / 2) = 0.0131 × 3281 × 2173 = 93,400 Pa/km = 0.93 bar/km

Step 4: Total pressure drop over 50 km, with density variation. Using segmented integration (20 segments) accounting for ρ dropping from 836 to ~681 kg/m³:

Total ΔP ≈ 50.3 bar Outlet P = 150 − 50.3 = 99.7 bara ✓ (above P_c) ΔP/km = 1.01 bar/km

Step 5: Margin checks.

Outlet 99.7 bara > 1.1·P_c (81.2 bara) — single-phase margin ✓
v_max = 2.79 m/s ≤ 4 m/s — DNV velocity limit ✓
Re = 8.7×10⁶ > 4000 — Colebrook valid ✓
ΔP/P₁ = 33% — increase segments to N=50 for higher accuracy

Result: 12" NPS pipe is adequate for this duty. ΔP = 1.01 bar/km is in the typical range. Smaller pipe (10") would push velocity above 3.5 m/s and ΔP > 2 bar/km — borderline. Larger pipe (14"/16") gives more margin but at higher capital cost.

8. Standards & References

ASME B31.4-2022, Pipeline Transportation Systems for Liquids and Slurries (CO₂ service)
ISO 27913:2016, Carbon dioxide capture, transportation and geological storage — Pipeline transportation systems
DNV-RP-J202 (2017), Design and Operation of CO₂ Pipelines
API RP 1186 (2024), Recommended Practice for CO₂ pipeline operations
Span, R., Wagner, W. (1996). "A New Equation of State for Carbon Dioxide Covering the Fluid Region from the Triple-Point Temperature to 1100 K at Pressures up to 800 MPa," J. Phys. Chem. Ref. Data 25(6), 1509–1596.
Fenghour, A., Wakeham, W.A., Vesovic, V. (1998). "The Viscosity of Carbon Dioxide," J. Phys. Chem. Ref. Data 27(1), 31.
Peng, D.-Y., Robinson, D.B. (1976). "A New Two-Constant Equation of State," Ind. Eng. Chem. Fundam. 15(1), 59–64.
Peneloux, A., Rauzy, E., Freze, R. (1982). "A Consistent Correction for Redlich-Kwong-Soave Volumes," Fluid Phase Equilibria 8, 7–23.
API RP 14E (2007), Design and Installation of Offshore Production Platform Piping Systems (erosion velocity)
NIST REFPROP, NIST Standard Reference Database 23, Version 10.0

CO₂ Dense-Phase Transport for CCUS Pipelines

304.13 K · 73.77 bar

700–900 kg/m³

≤ 4 m/s

Contents

4 calculators in this topic