Celeritas: Boosting LArTPC Simulations With Advanced Models

Hey everyone, let's talk about something super crucial for some of the biggest physics experiments out there: Liquid Argon Time Projection Chambers (LArTPCs). We're talking about giants like DUNE, ICARUS, SBND, and MicroBooNE – these guys are absolutely vital for exploring the mysteries of neutrinos and dark matter. To really get the most out of these cutting-edge detectors, our simulations need to be incredibly precise, especially when it comes to how scintillation light is produced. And that's exactly where Celeritas comes into play, and where we're looking to give it a major upgrade by integrating more sophisticated recombination and scintillation-quenching models.

Why is this so important? Well, accurate modeling of scintillation light isn't just a nice-to-have; it's fundamental. The amount of light detected tells us a ton about the energy deposited by particles. But here's the kicker: the relationship isn't always straightforward. Factors like the local electric field and the energy deposition density significantly influence how much light we actually see. Currently, while Celeritas is awesome at generating and propagating photons, it's missing some key physics that LArTPC experiments absolutely rely on. This enhancement isn't just about adding features; it's about pushing the boundaries of what's possible in high-energy physics simulations, ensuring that our virtual experiments perfectly mirror reality. Imagine being able to simulate particle interactions with unprecedented accuracy, directly impacting our understanding of fundamental physics. It’s a game-changer for the entire LArTPC community, paving the way for more robust data analysis and groundbreaking discoveries. Getting these models right in Celeritas means we can leverage its GPU acceleration for optical photon generation, which is a massive win for computational efficiency and simulation fidelity across the board. This isn't just a technical upgrade; it's a strategic move to empower the next generation of LArTPC research, making complex analyses faster and more reliable than ever before. We're talking about closing the gap between simplified models and the intricate reality of particle interactions in liquid argon.

Understanding Scintillation in LArTPCs: A Deeper Dive

When we talk about scintillation in LArTPCs, we're diving into the heart of how these detectors 'see' particles. Every time a charged particle passes through the liquid argon, it excites and ionizes the argon atoms, leading to the emission of tiny flashes of ultraviolet light – that's our scintillation light. This process is super important for detecting and reconstructing particle events. The intrinsic scintillation yield is roughly proportional to the energy deposited (dE/dx), but it's not the full story. There's also a crucial recombination factor, denoted as ${R( au, dE/dx)}$, which depends not only on the energy deposition but also on the local drift electric field (${| au|}$) and potentially temperature. This recombination factor is the unsung hero that determines how many of those excited argon atoms actually go on to produce scintillation photons, versus how many simply recombine to form free electrons that drift away to be collected as charge signals. The complex interplay between ionization and excitation, and how the electric field influences these pathways, leads to the experimentally observed anti-correlation between the amount of ionization charge collected and the amount of scintillation light produced. This anti-correlation is a cornerstone of LArTPC physics, providing crucial information for particle identification and energy measurement. Without properly accounting for this, our simulations would struggle to accurately reproduce real detector responses, leading to potential misinterpretations of experimental data. Our goal with Celeritas is to build out a robust framework that can handle these nuanced interactions, ensuring that the light production models are as close to reality as possible. Currently, Celeritas' ScintillationOffload mechanism calculates the mean number of optical photons based on a user-provided material property (the scintillation yield). While this is a great start, it doesn't account for two absolutely critical phenomena: first, recombination-based scintillation suppression, which is inherently dependent on the electric field, and second, Birks-like empirical saturation which is often applied to liquid argon and other scintillator materials to model non-linear light yield at high energy deposition densities. These omissions are significant because they prevent Celeritas from fully capturing the complex physics observed in LArTPC experiments, making it harder to integrate seamlessly into existing LArSoft-based simulations used by major collaborations. Filling these gaps will make Celeritas an even more powerful and versatile tool for the LArTPC community, allowing for more faithful simulations and, ultimately, better physics results. This isn't just about adding a few lines of code; it's about embedding a deeper understanding of the physics into our simulation tools. The current ad-hoc solutions used by experiments, while functional, often lack the computational efficiency and flexibility that native Celeritas support could provide, especially with GPU acceleration. We're talking about a significant leap forward in simulation fidelity and performance.

Why Standard Models Fall Short for Modern LArTPCs

Let's be real, guys, the standard scintillation models currently implemented in Celeritas, while good for general-purpose photon generation, fall short when it comes to the highly specialized environment of LArTPCs. The core issue is that ScintillationOffload primarily relies on a fixed, user-defined scintillation yield for a given material. This approach completely misses the dynamic nature of light production in liquid argon, which is fundamentally influenced by local conditions. Think about it: a particle traversing liquid argon doesn't just deposit energy uniformly. It creates ionization trails with varying densities, and these trails exist within a non-uniform electric field (especially near detector boundaries or wire planes). The number of scintillation optical photons per step in LAr isn't just a simple multiple of the energy deposited. Instead, it critically depends on both the intrinsic scintillation yield (which is related to ${dE/dx / W_{ph}}$, where $W_{ph} ext{ is roughly } 19.5 ext{ eV}$ per photon) and that all-important recombination factor ${R( au, dE/dx)}$. This factor is a beast, determined by the local drift field ${| au|}$ and the ionization density (${dE/dx}$). Why does this matter so much? Because these effects are absolutely essential for reproducing the experimentally observed anti-correlation between ionization charge and scintillation light signals. Without accurately modeling recombination, our simulations would predict far too much light in regions of high ionization or low electric field, leading to discrepancies with actual detector data. This isn't a minor detail; it's a fundamental aspect of how LArTPCs operate and extract physics information. The current limitation means Celeritas does not include the critical features needed by sophisticated LArTPC simulations: specifically, recombination-based scintillation suppression that is truly field-dependent, and the Birks-like empirical saturation which accounts for the reduction in light yield at very high energy deposition densities. For experiments like DUNE, ICARUS, SBND, and MicroBooNE, these models aren't optional extras; they are core components of their simulation frameworks. Integrating these directly into Celeritas would be a massive improvement, not only for the fidelity of simulations but also for computational efficiency, especially if we can leverage Celeritas' GPU acceleration for these complex calculations. It would mean a seamless, high-performance integration of optical physics into frameworks like LArSoft, addressing long-standing related issues like #1548 and #1344. We're talking about making our simulations smarter and faster, directly translating to better science. The difference between a simple yield and a dynamically calculated recombination is the difference between a rough sketch and a detailed, high-resolution portrait of particle interactions.

The Anti-Correlation Conundrum: Charge vs. Light

Alright, let's unpack one of the most fascinating and challenging aspects of LArTPC physics: the anti-correlation between ionization charge and scintillation light. Guys, this isn't just some academic curiosity; it's a fundamental signature that gives us invaluable information about the particles interacting in our detectors. Here's the deal: when a particle zips through liquid argon, it leaves behind a trail of both free electrons (which become our charge signal) and excited argon atoms (which decay to produce scintillation light). But these two signals aren't independent; they're locked in a delicate dance. When more energy is deposited in a smaller volume (high dE/dx), or when the local electric field is weak, the probability of an electron recombining with an ion before it can be drifted away increases significantly. This recombination process has a dual effect: it reduces the number of free electrons available for the charge signal (hence, less charge collected) and simultaneously reduces the number of scintillation photons produced because the recombination process often quenches the excited states that would otherwise radiate light. So, when recombination is high, you generally get less charge and less light. Wait, didn't I just say anti-correlation? Ah, here's the nuance: the anti-correlation refers to the relative yield of charge versus light for a given amount of deposited energy. If more energy goes into producing free electrons that don't recombine, you get more charge and less light from that specific energy deposition, because those electrons didn't lead to light production via recombination. Conversely, if more recombination occurs, you might lose both, but the ratio shifts. More accurately, the total deposited energy is split between producing prompt light, and producing ionization that either recombines (contributing to delayed light and reduced charge) or drifts away (contributing to charge signal). The models are designed to capture the fact that as the electric field increases, recombination decreases, leading to more charge collected and a relatively smaller fraction of energy going into prompt scintillation light (due to fewer recombination-induced light events, though direct excitation still produces light). This means that for a fixed energy deposit, a stronger electric field will tend to yield more charge and comparatively less light from recombination-related processes. This observed phenomenon is absolutely critical for distinguishing different types of particles (like electrons vs. alpha particles), which leave very different ionization tracks. Current Celeritas models, by not intrinsically modeling field-dependent recombination, cannot accurately reproduce this crucial anti-correlation. They simply apply a fixed scintillation yield, which means they can't capture how the ratio of charge to light changes based on the local electric field or the particle's stopping power. Without this, experiments might misidentify particles, mis-estimate their energies, or struggle with background rejection. This isn't just about getting the total light right; it's about getting the dynamics of light production and its relationship with charge collection spot on. By integrating these advanced models into Celeritas, we empower simulations to faithfully represent this complex interplay, leading to more robust data analysis and ultimately, more reliable physics conclusions. It’s about giving our LArTPC experiments the sharpest tools possible to uncover new physics, ensuring that the simulated universe accurately reflects the real one, especially in these delicate particle interactions. The ability to simulate this anti-correlation correctly is what separates a good LArTPC simulation from a truly excellent one.

The Urgent Need for Advanced Models in Celeritas

Let's be blunt: there's an urgent need for advanced models in Celeritas because the stakes are incredibly high for LArTPC experiments. These aren't just small university projects; we're talking about multi-billion-dollar international collaborations pushing the boundaries of human knowledge. For DUNE, ICARUS, SBND, and MicroBooNE, precise simulations aren't a luxury; they're a necessity for robust data interpretation, detector optimization, and ultimately, making groundbreaking discoveries. The existing ScintillationOffload in Celeritas is fantastic for many applications, but it simply doesn't cut it for the highly specific and complex requirements of LArTPCs. The main keywords here are electric-field-dependent recombination, anti-correlated ionization and scintillation signals, suppression of scintillation light yield, and optional Birks-law–type saturation. These aren't just technical terms; they represent fundamental physical phenomena that must be accurately modeled to unlock the full potential of LArTPC data.

Addressing Key Experimental Requirements

First up, we absolutely need to implement electric-field–dependent recombination. Imagine a particle track winding through the liquid argon. The electric field isn't perfectly uniform across the entire detector volume; it has variations, especially near wires or detector boundaries. The amount of recombination – where electrons re-attach to ions instead of drifting away – is directly influenced by the strength and direction of this local electric field. A stronger field pulls electrons away faster, reducing recombination and thus potentially altering the light yield. LArTPC experiments require models that use a parameterized field map, ${ au(x,y,z)}$, to accurately calculate the recombination probability at every point along a particle's track. This level of detail is critical for precise event reconstruction and energy measurement. Without it, our simulations would predict a uniform light yield where a non-uniform one actually exists, leading to systematic errors in our analyses. This isn't just theory, guys; it's what's happening inside the detectors every second.

Next, the models must support anti-correlated ionization (${Ar^{+}}$) and scintillation (${Ar^{*}}$) signals. We touched on this, but it bears repeating because it's that important. The total energy deposited by a particle is split between producing electrons/ions (charge) and exciting argon atoms (light). Due to recombination, these two signals are intrinsically linked in an anti-correlated way. If more energy goes into ionization that recombines, you get less charge signal collected, but potentially more light from the recombination process itself. Conversely, if recombination is suppressed by a strong electric field, you collect more charge, and the light yield from recombination might decrease proportionally (though direct excitation still contributes). Capturing this precise balance is crucial for particle identification, especially when distinguishing between different types of particles (like minimum ionizing particles vs. heavily ionizing alpha particles) that leave very different charge/light ratios. Current Celeritas implementations simply can't do this natively.

We also need to implement the suppression of scintillation light yield as a function of both dE/dx and ${ au}$. This is another two-factor punch! It’s not just the electric field, nor just the energy deposition density (${dE/dx}$), that dictates the light output. It’s their combined effect. At very high ${dE/dx}$ (like from a stopping proton or an alpha particle), the ionization density is so high that many ions and electrons are packed close together. This enhances recombination and other quenching mechanisms, leading to a suppression of the light yield – you don't get as much light as a simple linear model would predict. The electric field then modulates this suppression. Getting this right is vital for accurate energy calibration and for identifying heavily ionizing particles, which are often backgrounds or crucial signals in searches for physics beyond the Standard Model. Without this combined dependence, our simulations would consistently overestimate light output for dense tracks, leading to biased results.

Finally, the request includes optional Birks-law–type saturation for more general scintillator materials. While LArTPCs are our primary focus here, Celeritas is a versatile tool. Birks' Law is a well-established empirical model for light quenching in many scintillators, not just liquid argon. It describes how the light yield saturates at high energy deposition densities. Adding this as an optional feature would make Celeritas even more broadly applicable, allowing it to accurately simulate other types of scintillators that might be used in related experiments or detector R&D. This adds flexibility and future-proofs Celeritas as a general-purpose simulation engine.

Currently, guys, these essential effects are handled through ad-hoc, parameterized optical-photon models that are built directly into LArTPC simulations like LArSoft. While these work, they are often less computationally efficient and harder to maintain than a native solution. Native support in Celeritas would enable full, offloading GPU-accelerated optical photon generation and propagation within LArSoft. This means faster simulations, better physics fidelity, and a more streamlined workflow for experimentalists. We’re talking about a massive leap in performance and accuracy that directly impacts our ability to analyze data and uncover new physics. This isn't just a convenience; it's a strategic necessity to keep LArTPC experiments at the cutting edge, making simulations not just faster, but fundamentally more correct in their physical representation.

Embracing Standard LArTPC Recombination Models

To truly integrate Celeritas into the LArTPC ecosystem and make it indispensable, we need to embrace the standard recombination models that are already tried-and-true in the community. These aren't just arbitrary formulas; they are carefully tuned and validated against experimental data, representing years of research and development. The goal is to introduce built-in support for these models, parameterized by ${dE/dx}$ (energy deposition density) and the local drift field ${| au|}$. By natively supporting these, Celeritas can become the go-to tool for highly accurate LArTPC optical simulations, leveraging its GPU acceleration for computations that are currently bottlenecks in CPU-based frameworks. Let's dive into the two major players:

Birks-like Recombination Model: The ICARUS Approach

First up, we have the Birks-like recombination model, which has been a workhorse for experiments like ICARUS and was also used in earlier DUNE simulations. This model provides a pragmatic and effective way to account for light suppression due to both ionization density and the electric field. The formula might look a bit intimidating at first, but it's quite elegant in its simplicity:

${R( au, dE/dx) = \frac{A_B}{1 + k_B * \frac{dE/dx}{| au|}}}$

Here, guys, $A_B$ is a normalization parameter, essentially scaling the overall light yield, and $k_B$ is the Birks constant. This $k_B$ is the crucial parameter that quantifies the degree of quenching; a higher $k_B$ means more significant light suppression for a given ionization density and electric field. What's really neat about this model is how it incorporates suppression from both ionization density and the electric field. If your ${dE/dx}$ is high (lots of energy deposited in a small space), or if your electric field ${| au|}$ is low (electrons aren't pulled away quickly), the denominator becomes larger, making the recombination factor $R$ smaller. A smaller $R$ means fewer scintillation photons produced for a given amount of deposited energy, accurately reflecting the observed quenching. This model has proven its value in reproducing the overall trends of scintillation light yield in liquid argon. It's relatively straightforward to implement and tune, which is why it's been a popular choice. While it's a simplification of the underlying microphysics, it provides a powerful empirical description that has served the community well. For Celeritas, integrating this model means we can directly support the simulation needs of experiments that rely on this formulation, ensuring consistency and accuracy across their entire analysis chain. It also provides a robust baseline for comparison with more complex models. The beauty of the Birks-like model lies in its ability to capture the essence of quenching with just a couple of parameters, making it accessible and widely applicable. This means we can simulate particle interactions more realistically, especially in scenarios where energy deposition is dense, and electric fields vary, which is a common occurrence in LArTPC detectors. The ability to correctly model this effect ensures that reconstructed particle energies and types are more accurate, which directly impacts the physics results of these experiments. It’s a foundational piece of the puzzle for simulating liquid argon detectors correctly and efficiently, and Celeritas is perfectly poised to bring this capability to the GPU-accelerated world.

Modified Box Model: DUNE, SBND, and MicroBooNE's Choice

Now, let's talk about the Modified Box model, which is the preferred choice for some of the biggest names in LArTPC research today, including DUNE, SBND, and MicroBooNE. This model takes things a step further, offering a more refined and often more accurate description of quenching across the full dE/dx spectrum. It’s a bit more complex than the Birks-like model, but for good reason: it aims to reproduce measured anti-correlation between light and charge with higher fidelity. The formula for the recombination factor $R$ in the Modified Box model is given by:

${R( au, dE/dx) = \frac{\ln(\alpha + \beta * \frac{dE/dx}{| au|}) }{ \beta * \frac{dE/dx}{| au|}}}$

Here, guys, $\alpha$ and $\beta$ are the tuned parameters. These parameters are derived from extensive experimental data and microscopic models, and they allow the Modified Box model to capture the nuances of recombination in liquid argon more precisely. What makes this model particularly powerful is its ability to produce more accurate quenching across the full dE/dx spectrum. Unlike simpler models that might only be accurate in certain energy regimes, the Modified Box model excels at handling both low-energy and high-energy depositions, which is critical for the diverse range of particles encountered in LArTPC experiments. Furthermore, it has been shown to reproduce measured anti-correlation between light and charge with exceptional accuracy. This is a huge deal! As we discussed, the anti-correlation is a defining feature of LArTPCs, providing key information for particle identification. By getting this right, DUNE, SBND, and MicroBooNE can perform more reliable particle identification (e.g., distinguishing between electrons and gamma rays, or between protons and muons) and more accurate energy reconstruction, which are paramount for their physics goals, particularly in neutrino oscillations and rare event searches. Integrating the Modified Box model into Celeritas means providing these flagship experiments with a native, GPU-accelerated solution for their primary optical simulation needs. This not only speeds up their simulations significantly but also ensures that the fundamental physics of light production is handled consistently and accurately within the high-performance Celeritas framework. For Celeritas users, it means access to a state-of-the-art model that is proven in the LArTPC community, enhancing the realism and trustworthiness of their simulation results. This kind of native support simplifies the simulation workflow, reduces the need for external, ad-hoc corrections, and ultimately frees up experimentalists to focus more on the physics and less on the intricacies of simulation plumbing. The adoption of this model by major experiments underscores its importance and validity, making its inclusion in Celeritas a critical step forward for supporting the broader LArTPC community. This is about delivering precision and performance where it matters most, directly influencing the quality of data analysis and the potential for new scientific discoveries. It’s a commitment to providing the most advanced tools for the most demanding experiments.

The Power of Customization: User-Supplied Models

While supporting the Birks-like and Modified Box models is absolutely crucial, we know that the world of LArTPC experiments and scintillator development is always evolving. That’s why we also need to build in the power of customization through user-supplied/configurable models. This isn't just a nice-to-have feature; it’s a strategic necessity to ensure Celeritas remains flexible, adaptable, and future-proof. We can't predict every single recombination model or quenching parameter set that future experiments might develop or require. Therefore, providing a mechanism for experiments to load their own parameters, field maps, or even functional forms is paramount. This level of flexibility ensures that Celeritas can support a broad range of detector configurations, R&D efforts, and specific analysis requirements.

Flexibility for Diverse Experiments and Future Innovation

Think about the sheer flexibility for diverse experiments this would unlock. Different LArTPC experiments might have unique detector geometries, slightly different argon purity levels, or even explore novel operating conditions (like varying temperatures or different dopants). Each of these factors can influence the recombination and scintillation quenching processes in subtle ways, potentially requiring a slightly tweaked or entirely new recombination model. For example, a small-scale R&D experiment might be testing a novel field cage design that produces a highly complex electric field map, or they might be investigating the effects of a new wavelength-shifting dopant. In such scenarios, relying solely on the pre-programmed Birks or Modified Box models, while excellent, might not be sufficient to capture the specific nuances of their setup. By allowing users to provide their own recombination functions, either as compiled code, lookup tables, or parameterized forms that Celeritas can interpret, we make the framework incredibly versatile. This means experiments aren't locked into a fixed set of models; instead, they can tailor Celeritas' optical physics to perfectly match their specific detector characteristics or research questions. Furthermore, this approach directly supports future innovation. The field of LArTPC technology is continuously advancing. New theoretical models for electron-ion recombination might emerge, or experimental measurements might refine our understanding of quenching parameters under extreme conditions. With a user-configurable interface, Celeritas can immediately adopt these new insights without requiring a major software release or core code changes. Experimentalists could simply load their updated parameters or even an entirely new functional form to reflect the latest scientific understanding. This iterative development is crucial for staying at the cutting edge. It also caters to specific detector configurations. Imagine an experiment with a highly non-uniform electric field due to unique internal structures. Instead of trying to approximate this with a generic model, they could supply a detailed electric field map and a corresponding recombination function tailored to that field. This bespoke approach ensures the highest possible accuracy for their simulations. In essence, permitting experiments to load custom parameters, field maps, or functional forms for recombination and quenching models transforms Celeritas from a powerful tool with fixed capabilities into an incredibly adaptable, extensible platform for optical simulations. This not only enhances its immediate utility for current LArTPC experiments but also ensures its relevance and value for the next generation of particle physics research and detector development. It's about empowering the community to innovate and conduct the most precise simulations possible, ensuring that Celeritas remains at the forefront of high-energy physics simulation tools. It truly means opening up Celeritas to an even wider array of scientific inquiries and allowing it to evolve hand-in-hand with the experimental community, proving its long-term value and commitment to cutting-edge research. This is how we build a truly collaborative and forward-thinking simulation ecosystem, allowing researchers to push boundaries without being constrained by rigid software limitations. We want Celeritas to be a partner in discovery, not a hurdle.

Conclusion: A Brighter Future for LArTPC Simulations with Celeritas

Alright, guys, let's wrap this up. It's abundantly clear that enhancing Celeritas with advanced recombination and scintillation-quenching models isn't just a technical upgrade; it's a fundamental leap forward for the entire LArTPC community. We’re talking about a future where simulations for giants like DUNE, ICARUS, SBND, and MicroBooNE are not only faster but also significantly more accurate, mirroring the complex reality inside these incredible detectors. The current reliance on fixed scintillation yields and external, ad-hoc models simply doesn’t capture the nuanced, field-dependent, and density-dependent physics of light production in liquid argon, which is absolutely vital for making sense of their data.

By integrating native support for electric-field–dependent recombination, properly accounting for the anti-correlation between ionization charge and scintillation light, and implementing robust light yield suppression models (like the Birks-like and Modified Box models), Celeritas will deliver unparalleled simulation fidelity. This means more precise particle identification, more accurate energy reconstruction, and ultimately, more reliable scientific conclusions from the groundbreaking experiments designed to explore neutrino oscillations, dark matter, and other exotic physics phenomena. Think about the impact: better background rejection, clearer signal identification, and reduced systematic uncertainties – all leading to stronger physics results.

Furthermore, providing the crucial capability for user-supplied/configurable models ensures that Celeritas remains flexible and adaptable, ready to support new detector designs, evolving physics models, and future R&D efforts. This future-proofs the software, making it a truly versatile tool for a dynamic scientific field. This level of customization is what transforms a powerful tool into an indispensable partner for innovation.

But here's the kicker: all these sophisticated physics models will be GPU-accelerated natively within Celeritas. This is a massive win! It means significantly faster simulations, reducing the computational burden on experiments and allowing them to generate the enormous datasets needed for modern analyses. This enhanced performance, combined with improved accuracy, creates a synergistic effect that will accelerate discovery. The seamless integration of these advanced optical physics models into frameworks like LArSoft through Celeritas will streamline workflows, minimize inconsistencies, and empower experimentalists to focus more on the physics and less on the intricacies of simulation development.

In essence, this is about unlocking the full potential of LArTPC experiments. A brighter future for LArTPC simulations with Celeritas means a clearer path to understanding the universe's most elusive particles. It’s an exciting time, guys, and Celeritas is poised to play a pivotal role in the next wave of scientific breakthroughs. This isn't just about code; it's about pushing the boundaries of what we can discover. We're setting the stage for decades of high-impact research, ensuring that our virtual experiments are as powerful and insightful as our real-world detectors. The investment in these advanced models will pay dividends in scientific discovery, cementing Celeritas' role as a vital tool for the entire particle physics community. It's a testament to the collaborative spirit of science, building better tools to achieve bigger goals.