🔬 physics

Modeling of In-Situ Nitrogen Alloying in SS316 during Laser Directed Energy Deposition

This study presents a coupled thermal–thermodynamic–kinetic model that predicts the extent of in-situ nitrogen alloying in AISI 316 during laser directed energy deposition, revealing that nitrogen uptake is kinetically limited by surface absorption dynamics rather than thermodynamic equilibrium and is strongly influenced by process parameters and dissolved oxygen.

Original authors: Md Al Amin Meia¹, Shafin Ahmed¹, Md Shahriar Forhad¹, Monsuru Ramoni¹, Jianzhi Li¹

Published 2026-10-29

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Md Al Amin Meia¹, Shafin Ahmed¹, Md Shahriar Forhad¹, Monsuru Ramoni¹, Jianzhi Li¹

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Stainless steel is everywhere, from the sinks in our kitchens to the surgical tools in hospitals, prized for its ability to resist rust and hold its shape. But for engineers designing parts that face extreme stress or harsh chemicals, standard stainless steel often needs a boost. One of the most effective ways to strengthen this material is to mix nitrogen into it while it is still molten. Nitrogen acts like a hidden reinforcement, making the metal harder and more resistant to corrosion without making it brittle. The challenge lies in how to get that nitrogen inside. Traditional methods require buying expensive, pre-alloyed powder or treating the metal after it is made, both of which are costly and inflexible. A newer manufacturing technique called laser directed energy deposition offers a different path: it melts metal powder with a focused beam of light and builds parts layer by layer. Theoretically, one could simply swap some of the protective argon gas used in the process with nitrogen, allowing the metal to absorb the element as it melts. However, the physics of this absorption is tricky. The molten metal exists for only a fraction of a second before it cools and solidifies, and the nitrogen molecules in the air must break apart and dissolve into the liquid fast enough to make a difference. If the process is too slow, the nitrogen escapes, and the metal remains unchanged.

Researchers at The University of Texas Rio Grande Valley set out to understand exactly how this nitrogen absorption happens during the laser process, moving beyond simple guesses about what might occur. They built a detailed computer model that simulates the entire event, tracking how heat moves through the metal, how the molten pool behaves, and how nitrogen molecules interact with the liquid surface. Their work reveals that the process is governed by a race between time and chemistry. The model shows that for the standard conditions used in these machines, the molten pool does not live long enough to reach a state of perfect balance where it absorbs as much nitrogen as physics allows. Instead, the process is limited by how fast the nitrogen can actually dissolve into the metal. This kinetic limitation means that simply increasing the amount of nitrogen gas in the chamber does not lead to a proportional increase in the nitrogen inside the metal, as one might expect from basic chemistry. Instead, the amount of nitrogen picked up rises almost directly with the pressure of the nitrogen gas, a behavior that only occurs when the reaction is too slow to catch up with the changing conditions.

The team discovered that the temperature of the molten pool plays a counterintuitive role. In many chemical reactions, higher heat speeds things up, but in this specific scenario, making the pool hotter actually lowers the maximum amount of nitrogen the metal can hold. While the heat does help the nitrogen dissolve faster, it simultaneously reduces the ceiling of how much can stay dissolved. For the laser settings they studied, these two effects nearly cancel each other out, making the final nitrogen content surprisingly insensitive to small changes in laser power. What matters far more is how long the metal stays molten. Slowing down the laser's travel speed gives the metal more time to absorb nitrogen, leading to significantly stronger results. The researchers also found that the presence of oxygen acts as a major barrier. Even tiny amounts of dissolved oxygen can block the surface sites where nitrogen needs to attach, drastically reducing the amount of nitrogen that gets into the final part. This suggests that using ultra-pure powder and maintaining a tight seal against air leaks are just as critical as the laser settings themselves.

Using their model, the researchers mapped out the potential outcomes for a wide range of laser powers, travel speeds, and gas mixtures. They found that under typical operating conditions, the process can reliably add between 470 and 1,000 parts per million of nitrogen to the steel. This is a substantial improvement over the starting material, yet it remains safely below the threshold where gas bubbles would form and ruin the part's structure. The model predicts that if the process were run with pure argon, the metal would actually lose nitrogen, dropping below its original levels. This confirms that the nitrogen is not just sitting there; it is actively exchanging with the gas environment. The study also highlights that the current understanding of the reaction speed relies on a single adjustable factor that has not yet been measured directly in this specific laser setup. The researchers propose a clear path forward: by running a simple series of tests with different gas mixtures, engineers can calibrate this factor and then use the model to predict exactly how much nitrogen will end up in a part before they even begin building it.

The work provides a clear, physics-based explanation for why nitrogen alloying in laser manufacturing has historically been inconsistent. It rules out the idea that the metal simply reaches a natural limit based on the gas pressure alone, showing instead that the speed of the reaction is the bottleneck. The findings suggest that to get the most out of this technique, manufacturers should focus on slowing the laser down to extend the melting time and rigorously controlling oxygen levels, rather than just cranking up the power or the gas pressure. By treating the process as a timed chemical exchange rather than a simple mixing event, this research offers a roadmap for creating stronger, more durable stainless steel parts on demand, turning a complex physical challenge into a controllable engineering variable.

Technical Summary: Modeling of In-Situ Nitrogen Alloying in SS316 during Laser Directed Energy Deposition

Problem Statement
Nitrogen is a critical interstitial element for enhancing the yield strength, austenite stability, and pitting corrosion resistance of austenitic stainless steels like AISI 316. While wrought 316LN grades typically contain 0.10–0.16 wt% nitrogen, achieving comparable nitrogen levels in additively manufactured parts via Laser Directed Energy Deposition (LDED) remains challenging. Existing methods, such as using high-nitrogen powder or gas-nitrided feedstock, introduce cost or process complexity. In-situ alloying, where nitrogen is introduced via the shielding or carrier gas, offers a viable alternative. However, previous studies have reported highly variable nitrogen pickup levels. The core problem addressed is that equilibrium thermodynamics alone cannot predict nitrogen uptake in LDED because the melt pool lifetime is often too short to reach equilibrium. Furthermore, standard arc-welding models, which assume plasma-dissociated atomic nitrogen, are inappropriate for LDED, where the gas phase typically remains molecular $N_2$ .

Methodology
The authors developed a coupled thermal–thermodynamic–kinetic model to simulate in-situ nitrogen alloying. The framework integrates four distinct components:

Thermal Model: An Eagar–Tsai moving Gaussian heat source solution calculates the melt pool geometry (surface area $A_s$ , volume $V$ , length $L$ ) and thermal state (residence time $\tau$ , area-averaged surface temperature $\bar{T}_s$ ). The model assumes quasi-steady conduction with effective thermal conductivity to approximate Marangoni convection.
Thermodynamics: Equilibrium solubility is determined using Sieverts' law, incorporating temperature-scaled Wagner interaction parameters to account for the specific chemistry of AISI 316 (Cr, Ni, Mo, etc.). The model accounts for the fact that solubility decreases with increasing melt superheat due to Cr–N interactions.
Kinetics: Absorption is modeled as a reversible dissociative exchange at the free surface ( $N_2(g) \rightleftharpoons 2[N]$ ), treating the reverse reaction as second-order in dissolved nitrogen. The rate law includes site-blocking effects from surface-active elements (Sulfur and Oxygen). This differential equation yields an exact analytical solution involving a hyperbolic tangent function, governed by a Damköhler number ( $Da = \beta N_{eq} \tau$ ).
Process Integration: The model sequentially calculates nitrogen pickup in three stages:
- Powder-in-flight: Absorption by particles before they enter the melt pool.
- Melt-pool stage: Absorption during the residence time.
- Layer accumulation: Repeated remelting of previous layers, which re-exposes material to the gas.
- Porosity Screening: A Scheil-based criterion estimates the critical nitrogen concentration at which $N_2$ porosity forms during solidification.

The model was implemented in Python and verified against analytical Rosenthal solutions and numerical integrations.

Key Contributions
The paper presents four primary contributions:

Closed-Form Kinetic Solution: A derivation of an exact hyperbolic-tangent solution for reversible dissociative nitrogen absorption, reduced to a single dimensionless Damköhler number ( $Da$ ) that combines laser power, travel speed, $N_2$ partial pressure, and alloy chemistry.
Regime Prediction: The prediction that uncalibrated laser DED of 316 operates in a kinetically limited regime ( $Da \approx 0.1–0.6$ ), explaining why experimental nitrogen pickups are often modest compared to equilibrium predictions.
Temperature Sensitivity Reversal: The identification of a sign change in temperature sensitivity. In the kinetic regime ( $Da \ll 1$ ), hotter pools increase pickup (due to faster kinetics), whereas in the equilibrium regime ( $Da \gg 1$ ), hotter pools decrease pickup (due to lower solubility).
Process Maps and Sensitivity: The generation of process maps over a range of powers (600–1000 W), speeds (5–15 mm/s), and $N_2$ pressures (0.25–1 atm), alongside a sensitivity analysis identifying dissolved oxygen and the remelt fraction as critical variables.

Results
For a baseline condition (800 W, 10 mm/s, 0.5 atm $N_2$ , uncalibrated rate multiplier $\chi=1$ ):

Thermal State: The mean pool surface temperature is 2897 K, with a residence time of 0.17 s.
Kinetic State: The Damköhler number is 0.22, indicating the process is far from equilibrium. The equilibrium ceiling is 0.085 wt%, but the steady-state deposit reaches only 0.061 wt% (607 ppm).
Efficiency: The deposit closes only 42% of the gap between the feedstock nitrogen (0.043 wt%) and the equilibrium ceiling. Powder-in-flight absorption contributes ~10% of the total pickup, while layer remelting amplifies the first-pass pickup by a factor of 1.33.
Scaling Laws: In this kinetic regime, nitrogen pickup scales almost linearly with $N_2$ partial pressure (rather than the square root predicted by Sieverts' law) and increases with melt superheat (apparent activation energy of 50.5 kJ/mol).
Sensitivity: Dissolved oxygen acts as a strong inhibitor; increasing dissolved oxygen from 0 to 100 ppm reduces pickup by up to 108 ppm. Travel speed is a more effective control lever than laser power.
Porosity: The predicted nitrogen levels (up to ~1000 ppm) remain below the calculated Scheil-based porosity threshold of 1410 ppm for the baseline parameters. Pure argon deposition is predicted to result in nitrogen loss relative to the feedstock.

Significance and Claims
The authors claim that this work provides a necessary framework for understanding why in-situ nitrogen alloying in LDED often yields sub-equilibrium results. By distinguishing between kinetic and thermodynamic control, the model explains the variability in reported literature values. The paper asserts that the rate multiplier $\chi$ is the sole adjustable parameter, which can be calibrated via a single gas-composition series experiment. Once calibrated, the model becomes predictive. The study emphasizes that for the uncalibrated baseline, the process is kinetically limited, and that optimizing nitrogen pickup requires managing residence time (via travel speed) and minimizing dissolved oxygen, rather than simply increasing laser power. The paper concludes that the model offers five experimentally testable predictions to discriminate between kinetic and equilibrium regimes in future experimental campaigns.

More like this