Scaling Down & Powering Up: Driving Efficiency in Medicinal Chemistry Through Miniaturization
01/07/2026
Share this blog
01/07/2026
In early-stage drug discovery, library synthesis is a critical engine for driving Structure-Activity Relationship (SAR) exploration. However, traditional workflows are frequently bottlenecked by material waste, manual compound handling, and rigid IT infrastructure.
In a recent SPT Labtech webinar, Matthias Schmid (Senior Scientist at Boehringer Ingelheim) and Cory Tiller (Product Manager at SPT Labtech) broke down how microscale synthesis and smart automated sample storage are radically transforming medicinal chemistry workflows.
Here, we explore their key insights on transitioning from macro to microscale workflows, overcoming solubility hurdles, and implementing automated storage on any budget.
For years, a semi-preparative parallel chemistry workflow at a 100 µm scale was the industry norm, utilizing 8 mm glass tubes and requiring up to half a liter of acetonitrile/water solvent mix per purification run. But looking closely at the data reveals an unsustainable disparity:
When I looked at roughly 6,000 compounds from 2020, still a few years later only 1.3 milligrams had been consumed out of the 15 milligrams registered per compound.
Matthias Schmid - Boehringer Ingelheim
Because early-stage biochemical assays require very little material (often 16 aliquots of just 50 nanoliters each to generate an 8-point IC50 curve), over 99% of synthesized materials sit unused in collections indefinitely.
By scaling down from a 100 µmol scale to a 10 µmol scale—a factor of 10—laboratories can massively reduce consumption of solvents, building blocks, and energy without losing an ounce of chemical capability.
Q: What is the most effective way for an organization to learn about and evaluate automated solutions in this space?
Matthias Schmid: A strong peer-to-peer network is essential. We stay closely connected with technology groups at other pharma and crop science companies. Because these conversations focus strictly on technology rather than proprietary drug structures, the discussions can be remarkably open and collaborative. Beyond that, major conferences like SLAS and smaller vendor exhibitions are vital for getting a hands-on look at what’s available on the market.
Cory Tiller: I completely agree. While webinars provide great initial exposure, nothing beats conversations with industry peers to understand what is truly practical to implement inside a working laboratory.
Q: How can a laboratory implement microscale synthesis or micropurification on a tighter capital budget?
Matthias Schmid: Full medicinal chemistry automation—incorporating glove boxes, automated liquid handlers, and integrated storage—can easily approach seven-figure costs. However, a lower-budget setup is entirely feasible if you start with the absolute essentials: miniaturized tubes or vials, a reliable storage layout, and a rigorous tracking system. You could technically begin with a standard laboratory fridge and an internal inventory management system, picking samples manually. The absolute priority is the layout: knowing exactly where your samples are and being able to retrieve them quickly. Tube pickers and automated units can always be phased in later.
Cory Tiller: Miniaturizing the physical container—moving away from traditional glass tubes to an automation-compatible vial format—is the critical first step. Once your consumables are standardized for automation, you have successfully future-proofed your lab's workflow for whenever capital budget opens up for hardware additions.
Q: How will AI-driven reaction optimization impact miniaturized traditional chemistry pipelines in the near future?
Matthias Schmid: AI will undoubtedly play a massive role, but its output is strictly dependent on data quality. Many public datasets currently suffer from historical bias. At Boehringer Ingelheim, our current focus is generating structured, high-quality internal data to train future models. For now, we lean on standardized catalyst and reaction screening templates, which serve our current workflows efficiently until external databases mature.
Q: Are stock solutions stored directly in DMSO, and is the solvent removed before chemical synthesis?
Matthias Schmid: No, we do not remove the DMSO. In our microscale workflow, the stock solution solvent typically makes up about 25% of the final reaction mixture. During our early method development, we evaluated whether standard chemistries could tolerate that threshold. While some reactions (such as specific cross-couplings) traditionally struggle in pure DMSO, we found that many function perfectly fine when it is restricted to 25% of a secondary solvent mix (like dioxane).
That said, flexibility is key. We don't rely solely on DMSO; we actively store solutions in alternative solvents including DMA, dioxane, ethanol, NMP, DMPU, and DMI depending on what the chemistry requires and ensuring it doesn't degrade the vial material.
Q: How do you handle compound solubility limitations at these micro-concentrations?
Matthias Schmid: We don't actively alter the compounds to optimize solubility. Instead, we established standard storage concentrations at 0.1 M or 0.2 M. Because our standard reaction environment operates at 0.05 M, storing at 0.2 M allows us to aliquot directly into reactions without a time-consuming reconcentration step.
Crucially, our automated workflow is designed to handle suspensions. If a compound isn’t fully soluble, we verify if it can form a stable suspension via rapid pipetting or magnetic stirring. If it can, we store the tube with an integrated magnetic stir bar sealed inside for its lifetime. When that sample is processed by our liquid handler, it sits on a magnetic stirring block and undergoes active tip-mixing during aliquoting to guarantee an accurate volumetric transfer.
Q: Do you ever bypass micropurification and apply micro-synthesized compounds directly to biological assays?
Matthias Schmid: We have actively evaluated direct-to-biology (D2B) workflows, and we expect their usage to expand. However, our primary workflow relies on micropurification. Pure compounds remain our priority because D2B is highly restricted to "traceless" chemistries where the reagents and byproducts won't interfere with cellular or biochemical assays. If you are running a reaction involving a heavy metal like palladium, you absolutely cannot apply that crude mixture directly to cells without risking false positives or toxicity.
Q: Is the comPOUND® automated storage system restricted exclusively to Azenta tubes?
Cory Tiller: Not at all. The comPOUND® system features open compatibility with 96-format labware from all major global vendors, including Micronic, Matrix, and FluidX by Azenta. While certain cap profiles and tube heights handle pneumatically better than others, labs have massive flexibility across brands.
Q: What environmental configurations can the comPOUND® system support?
Cory Tiller: The storage core is highly versatile, operating anywhere from -20°C up to ambient conditions. Dry air is standard, a 4°C climate is available, and nitrogen gas purging is a frequent option for labs looking to preserve moisture-sensitive environments. Regardless of the internal atmosphere, standard compressed air is always used externally to safely power the mechanical pneumatics.
.png?width=300&height=400&name=comPOUND_center_banner%20(2).png)
Q: How does comPOUND® prevent 2D matrix barcodes from getting scratched or damaged over thousands of cycles?
Cory Tiller: The internal design of the comPOUND® relies on unique pneumatic carousels where tubes travel safely—one tube per position—without direct physical impact. When tubes are double-stacked to maximize footprint density, the structural crown of the lower cap provides a protective buffer that prevents face-to-face friction against the barcode. In the rare event a barcode becomes unreadable, the system can dynamically assign a temporary tracking ID or flag the sample to be outputted for manual inspection.
Matthias Schmid: Barcode resilience was an essential factor when selecting our consumables. Standard 2D matrix codes feature redundant data encoding; even if a portion of the barcode is physically marred, standard optical readers can still successfully decode the 10-digit identifier.
Q: Are there speed or performance trade-offs when choosing a double-stacked tube configuration over single-depth storage?
Cory Tiller: The primary trade-off is velocity. When operating as a single-depth archive, comPOUND® acts with high-speed simplicity. Introducing a double-stacked configuration requires the system to lift both tubes together, clamp the top sample, and return or separate them. This shifts the average processing time from roughly 10 minutes per rack to approximately 15 minutes. However, the payoff is immense: you achieve exactly twice the storage capacity within the exact same laboratory footprint.
Q: Does cap color play a role when setting up an automated sample archive?
Cory Tiller: Historically, yes. The comPOUND® utilizes a precise optical laser sensor to validate cap presence in its input/output vestibule. Dark cap colors can absorb light rather than reflect it, creating a minor risk of false misreads. Brighter color spectrums are always preferred for workflow simplicity.
Matthias Schmid: We originally intended to color-code our caps by chemical class, but quickly shifted entirely to a standardized bright orange cap. Once a system is fully optimized, visual tracking becomes irrelevant—the automation handles everything.
Q: Did you have a finalized data management framework in place before deploying this microscale workflow?
Matthias Schmid: We did not have a complete software solution ready when the hardware arrived. We deployed an agile, interim data solution with internal IT support to bridge the gap. Data modernization is a continuous journey, but it’s vital to recognize that your physical workflow can successfully progress even while your software infrastructure evolves around it.
Cory Tiller: We see both sides from a vendor perspective—some organizations solve software on day one, while others address it at the tail end of implementation. Software is the vital digital backbone of any microscale repository. Ensuring it links cleanly with your instrumentation prevents substantial operational friction down the road.
Q: How do you sync data between the comPOUND® archive, your internal LIMS, and Electronic Lab Notebooks (ELNs)?
Matthias Schmid: The integration was remarkably straightforward because the comPOUND® system communicates via simple XML file exchanges over a network directory.
Cory Tiller: While flat XML file transfers might seem less modern than a fully real-time REST API, file-based communication is exceptionally stable and simple to program. Laboratories routinely configure their LIMS with directory watchers. The moment comPOUND® finishes a cherry-picking run, it exports an XML report file, which the LIMS instantly sweeps, parses, and logs to update sample records automatically.
Transitioning to a 10 µmol microscale synthesis platform yields massive dividends: it slashes material costs, eliminates downstream purification waste, and shortens the cycle time between library design and initial assay data from weeks to mere hours.