Quantifying Salt Precipitation During CO2 Injection: How Flow Rate, Temperature, and Phase State Control Near-Wellbore Crystallization

Imagine you are trying to pump a giant amount of carbon dioxide (CO₂) deep underground into a salty, water-filled rock layer to keep it out of our atmosphere. This is a key strategy to fight climate change. But there's a catch: when this dry CO₂ hits the salty water (brine) in the rocks, it acts like a hairdryer blowing on a wet sponge. It sucks the water out, leaving the salt behind.

If too much salt dries out and turns into solid crystals (like rock salt), it can clog the tiny holes in the rock, like a clogged drain. This stops the CO₂ from being pumped in, creates dangerous pressure, and could ruin the whole project.

This paper is like a high-speed, microscopic detective story where the scientists built a tiny, transparent "rock" (a microfluidic chip) to watch exactly how, when, and why these salt crystals form. They wanted to figure out how to keep the "drain" from getting clogged.

Here is what they discovered, explained with some everyday analogies:

1. The "Hairdryer" Effect: Phase Matters

The scientists tested the CO₂ in three different states: Liquid (like water in a bottle), Gas (like air in a balloon), and Supercritical (a weird, super-dense state that acts like both a liquid and a gas).

The Analogy: Imagine trying to dry a wet towel.
- Liquid CO₂ is like gently wiping the towel with a damp cloth. It's slow and leaves a lot of moisture behind.
- Gas CO₂ is like blowing warm air on it. It dries faster.
- Supercritical CO₂ is like a high-powered industrial heat gun. It blasts the water away instantly and evenly.

The Finding: The Supercritical state was the clear winner. It pushed the water out of the rock pores most efficiently, leaving the least amount of "wet spots" behind. This is great news because it means the CO₂ spreads out better, but it also means the water evaporates so fast that salt crystals can form almost instantly.

2. Speed Kills (The Crystals)

The faster the CO₂ moves, the faster the water evaporates, and the faster the salt turns into solid rock.

The Analogy: Think of making rock candy. If you let a sugar syrup sit slowly, you get a few big, beautiful crystals. If you blast it with heat and wind, you get a million tiny, crunchy crystals instantly.
The Finding: When the CO₂ moved slowly (low flow rate), it took nearly an hour for the first salt crystal to appear. When they cranked up the speed (high flow rate), crystals appeared in less than one minute.
The Temperature Twist: Heat acts like a turbocharger. At 20°C (cool), it took 57 minutes to start crystallizing. At 60°C (hot), it took just 1 minute.

3. The "Traffic Jam" vs. The "Highway"

The scientists looked at how the CO₂ moved through the tiny holes in the rock.

Slow movement (Diffusion): The CO₂ has to wiggle through the water molecule by molecule. It's like a traffic jam where cars move inch by inch. This is slow and inefficient.
Fast movement (Convection): The CO₂ sweeps through like a highway, carrying the water vapor away with it. This is much faster.

The Finding: The speed of the CO₂ (flow rate) was the most important factor. When the CO₂ moved fast, it created a "highway" effect that dried out the rock so quickly that salt precipitation exploded.

4. The "Surprise Party" of Crystals

One of the most interesting things they found is that you can't predict exactly where the first crystal will pop up.

The Analogy: It's like a surprise party. You know the party is happening (salt will form), and you know roughly when (based on speed and heat), but you can't predict which specific guest (pore in the rock) will be the first to arrive.
The Finding: Even though the start is random, the end result is surprisingly even. By the time the rock is completely dry, the salt crystals are spread out fairly evenly across the whole area, not just stuck at the entrance.

Why Does This Matter?

This research gives engineers a "rulebook" for building carbon capture plants.

Don't go too fast: If you pump CO₂ in too fast, you create a "salt clog" immediately.
Watch the temperature: Hotter rocks mean faster clogging.
Supercritical is best (but tricky): While the supercritical state moves CO₂ through the rock best, it also dries the rock so fast that it creates the most salt. Engineers need to balance this speed to get the CO₂ in without clogging the well.

In a nutshell: This paper tells us that if we want to store CO₂ underground safely, we have to be careful not to "dry out" the rock too quickly. We need to find the perfect "Goldilocks" speed—fast enough to store the gas, but slow enough to keep the salt from turning the rock into a solid brick.

Here is a detailed technical summary of the paper "Quantifying Salt Precipitation During CO2 Injection: How Flow Rate, Temperature, and Phase State Control Near-Wellbore Crystallization."

1. Problem Statement

The injection of dry or undersaturated CO₂ into saline aquifers for geological carbon storage (CCS) induces the evaporation of formation brine. This process leads to the progressive concentration of dissolved salts (primarily NaCl/halite) until supersaturation is reached, causing salt precipitation within the pore space.

Consequences: This phenomenon, known as "dry-out" or "salting-out," rapidly clogs pore throats, reduces permeability, lowers injection rates, and causes excessive pressure buildup near the wellbore.
Knowledge Gap: While field observations and core-scale experiments confirm permeability decline, they lack the spatial and temporal resolution to resolve the underlying pore-scale mechanisms. Specifically, the quantitative interplay between CO₂ phase states (liquid, gas, supercritical), temperature, and hydrodynamic transport regimes on crystallization kinetics remains poorly characterized. This hinders the development of predictive models for reservoir-scale injectivity impairment.

2. Methodology

The authors employed high-resolution microfluidic experiments to systematically quantify halite crystallization dynamics under controlled thermodynamic conditions.

Experimental Setup:
- Device: A 2D borosilicate glass micromodel (20 mm × 10 mm × 20 µm) with rock-like pore geometry derived from natural sandstone scans (mean pore size ~240 µm, porosity 0.48).
- Conditions: Experiments covered a parameter space representative of shallow to intermediate-depth CCS projects:
  - Pressure: 50–80 bar.
  - Temperature: 20–60°C.
  - Flow Rates: 100 and 1000 mL/min (ambient).
  - Phases: Liquid, Gaseous, and Supercritical CO₂ (scCO₂).
- Fluids: Dry CO₂ injected into brine-saturated (5.5 mol/L NaCl) micromodels.
Imaging & Analysis:
- High-speed digital microscopy (1 fps) captured the displacement, evaporation fronts, and crystal growth.
- Image Processing: Custom MATLAB pipelines used intensity thresholding and image subtraction to segment phases (brine, CO₂, crystals) and quantify metrics like brine saturation ( $S_w$ ) and crystal fraction ( $X_c$ ).
Dimensionless & Kinetic Analysis:
- Transport: Characterized by Reynolds ( $Re$ ), Péclet ( $Pe$ ), and Schmidt ( $Sc$ ) numbers to define flow regimes and mass transport dominance.
- Evaporation: Quantified via the Sherwood number ( $Sh$ ) to measure convective mass transfer enhancement.
- Crystallization Kinetics: Modeled using the Avrami equation ( $X_c(t) = 1 - \exp(-Kt^n)$ ) to extract rate constants ( $K$ ) and exponents ( $n$ ).
- Temperature Dependence: Analyzed via Arrhenius plots to determine apparent activation energy ( $E_a$ ).
- Displacement Efficiency: Assessed using fractal dimension ( $D$ ) analysis of CO₂ invasion patterns.

3. Key Contributions

Systematic Quantification: Provided the first high-resolution, systematic quantification of how CO₂ phase state and temperature jointly control halite crystallization kinetics in porous media.
Transport-Controlled Mechanism: Demonstrated that crystallization is predominantly transport-controlled rather than kinetically limited, with convective transport dominating diffusion.
Kinetic Parameters: Established quantitative relationships between dimensionless parameters ( $Pe$ , $Sh$ ) and kinetic constants ( $K$ , $E_a$ ), providing critical benchmarks for validating pore-scale models.
Phase State Impact: Revealed that supercritical CO₂ offers superior displacement efficiency and evaporation rates compared to liquid or gaseous phases, fundamentally altering precipitation dynamics.

4. Key Results

A. Displacement Efficiency and Phase State

Supercritical CO₂ (scCO₂) achieved the highest displacement efficiency, resulting in the lowest residual brine saturation ( $S_w \approx 0.22–0.36$ ) and the highest fractal dimensions ( $D \approx 1.79–1.82$ ), indicating more uniform, space-filling invasion.
Liquid CO₂ showed the poorest performance with high residual saturation ( $S_w \approx 0.40–0.62$ ) and lower fractal dimensions ( $D \approx 1.70–1.80$ ), leading to heterogeneous trapping.
Gaseous CO₂ fell between the two but approached scCO₂ performance at higher flow rates.

B. Evaporation and Nucleation Kinetics

Nucleation Time: Highly sensitive to phase and temperature.
- Liquid/ScCO₂: Nucleation occurred in <1 minute (at 40–60°C).
- Gas: Nucleation took ~20 minutes.
- Liquid (20°C): Nucleation was delayed to 57 minutes.
Evaporation Rates: ScCO₂ and gas phases exhibited significantly faster evaporation (Sherwood numbers 2–3× higher than liquid phase) due to higher diffusivity and favorable viscosity ratios.
Total Dryout Time: Reduced from ~300 min (gas at 20°C) to ~90 min (scCO₂ at 40°C).

C. Crystallization Kinetics (Avrami & Arrhenius)

Transport Control: The Avrami rate constant ( $K$ ) increased by two orders of magnitude as the Péclet number ( $Pe$ ) increased from 50 to 1440. This confirms that higher flow velocities (convection) accelerate supersaturation and crystal growth.
Temperature Sensitivity: The process follows Arrhenius behavior with an activation energy $E_a = 58.6 \text{ kJ/mol}$ . This intermediate value suggests crystallization is governed by a combination of interfacial attachment kinetics and transport-mediated supersaturation delivery.
Crystal Morphology:
- Gas Phase: Produced fewer, larger crystals (mean diameter ~200 µm) due to slower evaporation allowing growth.
- Liquid/ScCO₂: Produced many smaller, uniform crystals (~120–130 µm) due to rapid supersaturation favoring high nucleation frequency.
Spatial Distribution: Despite probabilistic nucleation, final crystal distributions were spatially uniform with no systematic inlet-outlet bias in the microfluidic chip.

D. Final Crystal Fractions

Crystal fractions increased 10-fold from liquid conditions (0.008) to gas/supercritical conditions (0.08–0.12).
This confirms that convective transport and phase state dominate over diffusion-limited mechanisms in determining the total amount of salt precipitated.

5. Significance and Implications

Model Validation: The derived quantitative relationships ( $Pe$ - $Sh$ - $K$ - $E_a$ ) serve as essential benchmarks for validating pore-network models, lattice-Boltzmann simulations, and reactive transport codes used in CCS reservoir simulation.
Operational Strategy: The study suggests that controlling injection velocities is a critical lever for managing salt accumulation. High injection rates (high $Pe$ ) accelerate crystallization and permeability loss, necessitating mitigation strategies (e.g., brine injection, temperature control) in high-flow zones.
Reservoir Prediction: The findings indicate that in saline and hypersaline aquifers, near-wellbore permeability impairment will be most severe under conditions favoring high convective transport (high flow rates, supercritical/gas phases, elevated temperatures).
Upscaling: While the 2D microfluidic geometry limits direct extrapolation to 3D reservoirs, the mechanistic insights and scaling laws provide a robust foundation for upscaling efforts to predict injectivity decline in real-world CCS projects.

In summary, this work shifts the understanding of salt precipitation from a purely thermodynamic phenomenon to a transport-dominated process, highlighting that flow rate and CO₂ phase state are the primary controls on the speed and extent of near-wellbore clogging.