High-Pressure Hydrogenation: Capability Questions to Ask

Selecting a high-pressure hydrogenation CRO is one of the few outsourcing decisions where the wrong answer is invisible until material is already committed. Hydrogenation is the most common reduction in pharmaceutical synthesis. It appears in nitro reductions on the way to anilines, in alkene and alkyne saturations, in benzyl and Cbz deprotections, in nitrile reductions, in reductive aminations, and in the aromatic ring reductions that convert pyridines to piperidines. Because it is so common, every synthesis provider claims it. Because the equipment tiers differ by an order of magnitude in capital cost and safety burden, most of those claims mean balloon chemistry.

The gap matters commercially. If you send a supplier a route with a 60 bar ring reduction and they own only a Parr shaker, one of two things happens. They subcontract the step, which adds time, adds a supply chain node you did not approve, and often adds a second analytical package to reconcile. Or they attempt it at 4 bar, report low conversion after two weeks, and propose a route change that was never necessary.

This post gives you the questions that reveal actual capability, the catalyst and selectivity issues that cause the most expensive surprises, and the residual metal requirements that follow every hydrogenation into the regulatory filing. It sits alongside our CRO buyer’s guide, which covers the broader selection framework.

What Hydrogenation Actually Does, and Which Substrates You Will Meet

Catalytic hydrogenation adds molecular hydrogen across a bond or removes a group, using a transition metal catalyst to split the hydrogen molecule into reactive surface-bound atoms. The substrate binds to the same metal surface, receives hydrogen, and desorbs. Everything that makes hydrogenation easy or difficult follows from that shared surface: whatever else in your molecule can bind to the metal will compete, and whatever else can accept hydrogen may do so.

The Seven Transformations You Will Encounter

Nitro to amine. The workhorse. An aromatic nitro group reduces to an aniline in a six-electron process, and it is fast, high-yielding, and generally selective. Palladium on carbon at 1 to 5 bar converts most substrates in two to eight hours. This is the transformation most often demonstrated when a supplier says they do hydrogenation, because it works under almost any conditions.

Alkene and alkyne reduction. Alkenes saturate readily at low pressure. Alkynes can be stopped at the alkene using a poisoned catalyst such as Lindlar palladium, which is a genuine selectivity skill rather than a default outcome. If a supplier cannot describe how they would stop an alkyne at the cis-alkene, they have not done it.

Benzyl and Cbz deprotection. Hydrogenolysis cleaves benzyl ethers, benzyl esters, benzylamines, and carbamates. Palladium is the catalyst of choice precisely because it cleaves these bonds, which makes it a feature when deprotection is the goal and a serious problem when it is not.

Nitrile to primary amine. Requires more forcing conditions, typically Raney nickel or rhodium at 20 to 60 bar, often with ammonia added to suppress secondary amine formation. The competing dimerization pathway is the main yield killer and the main reason this step benefits from a real autoclave rather than a shaker.

Reductive amination. A carbonyl and an amine condense to an imine, which is then reduced in situ. Convenient and atom-efficient, but the equilibrium and the reduction rate have to be balanced, and over-reduction of the carbonyl to the alcohol competes directly.

Aromatic ring reduction. Pyridines to piperidines, benzenes to cyclohexanes, and the partial reductions in between. These are the transformations that genuinely require pressure. Rhodium or ruthenium at 50 to 150 bar and elevated temperature is typical. This is where balloon-chemistry suppliers fail outright, and it is also where a large fraction of modern medicinal chemistry lives, because saturated nitrogen heterocycles are everywhere in current drug candidates. Our piperidines and piperazines category exists largely because of the demand for these saturated cores.

Carbonyl reduction. Ketones and aldehydes to alcohols. Usually done with hydride reagents in the lab, but hydrogenation becomes attractive at scale because it avoids the aqueous workup and the metal waste of a borohydride quench. It typically wants higher pressure and a ruthenium or platinum catalyst.

Why the Substrate Class Determines the Equipment

Reading the list above, note the pressure requirements. Nitro reduction, alkene saturation, and benzyl deprotection all run comfortably below 5 bar. Nitrile reduction, aromatic ring reduction, and carbonyl hydrogenation want 20 bar minimum and frequently far more. That single split is the most useful thing to know when you evaluate a supplier, because it maps directly onto the equipment tiers.

The Four Pressure Tiers and Why They Are Not Interchangeable

Hydrogenation equipment falls into four distinct classes, and moving between them is a capital purchase rather than a procedure change. When a CRO says they run hydrogenations, ask which tier.

Tier	Typical pressure	Equipment	Typical scale	What it can do
Atmospheric	1 bar	Balloon or gas bag over a flask	10 mg to 10 g	Nitro reduction, easy alkenes, Cbz removal
Low pressure	1 to 5 bar	Parr shaker, glass or steel bottle	100 mg to 100 g	Most of the above, faster and more reliably
Medium pressure	5 to 50 bar	Stirred autoclave, Hastelloy or stainless	1 g to 20 kg	Nitriles, harder alkenes, some ring reductions
High pressure	50 to 200 bar	Certified rated autoclave with gas booster	1 g to multi-kg	Aromatic rings, carbonyls, forcing asymmetric work

What Balloon Chemistry Cannot Do

A hydrogen balloon over a flask delivers about 1 bar and a finite quantity of gas. It is fine for a 500 mg nitro reduction on a discovery timeline. It cannot maintain constant pressure as gas is consumed, it cannot be heated safely above solvent reflux, and it offers no meaningful mass transfer beyond what a stir bar provides. Many CROs describing hydrogenation capability mean exactly this. It is legitimate chemistry at the milligram scale and irrelevant to a kilogram delivery.

The Parr Shaker Ceiling

The Parr shaker is the standard low-pressure workhorse and covers a genuinely useful range: a heavy-walled bottle, mechanical shaking for gas-liquid contact, and a pressure gauge. Its practical ceiling is around 4 to 5 bar with a glass bottle and somewhat higher with a steel one. It is inexpensive, and a supplier owning three of them can screen conditions in parallel, which is valuable. It is not a substitute for an autoclave, and glass bottles have a well-known failure mode when scratched.

Medium and High Pressure Autoclaves

A stirred autoclave is a different animal: a rated pressure vessel with a mechanical or magnetically coupled stirrer, a heating jacket, internal thermocouple, rupture disc, and usually a gas mass flow meter so hydrogen uptake can be tracked as a reaction progress signal. Uptake curves are one of the most useful diagnostics in hydrogenation, and a supplier who cannot give you an uptake plot is running blind.

Above roughly 50 bar you need a gas booster to charge the vessel, heavier wall construction, and a more rigorous inspection regime. The step from 50 to 100 bar is where most CRO capability stops. Ask for the actual maximum working pressure on the vessel nameplate, not the burst rating and not the aspiration.

Catalyst Selection: The Decision That Sets Everything Else

Catalyst choice determines rate, selectivity, and whether your protecting groups and halogens survive. It is the single most consequential decision in the step, and it is cheap to screen and expensive to guess.

Catalyst	Best for	Watch out for
Pd/C	Nitro reduction, alkenes, benzyl and Cbz removal	Cleaves benzyl groups, strips aryl halides, sulfur-poisoned
Pt/C	Nitro reduction when benzyl or halide must survive	Slower on alkenes, more expensive
PtO2 (Adams)	Nitro reduction, oximes, halide-tolerant work	Reduces in situ to platinum black, higher loading cost
Raney nickel	Nitriles, large-scale nitro reduction, desulfurization	Pyrophoric, removes sulfur unintentionally, needs pressure
Rh/C or Rh/Al2O3	Aromatic ring reduction, nitriles without dimerization	Expensive, needs pressure
Ru catalysts	Carbonyls, aromatic rings, asymmetric ketone reduction	Needs high pressure and often base or acid additive

How Chemoselectivity Actually Behaves

Palladium is the most active metal for hydrogenolysis, which means it cleaves benzylic carbon to heteroatom bonds and carbon to halogen bonds readily. If you want a Cbz off, this is exactly right. If your substrate is a benzyl-protected alcohol that must survive a nitro reduction, palladium is the wrong choice and platinum is the right one. Platinum reduces nitro groups efficiently while leaving benzyl ethers and aryl halides largely intact, which is why platinum oxide remains in constant use despite costing more per gram.

Rhodium and ruthenium sit at the forcing end. They reduce aromatic rings that palladium will not touch, and ruthenium is the metal behind most asymmetric ketone hydrogenation. Both want pressure, and both are expensive enough that loading optimization and metal recovery are real line items rather than academic concerns.

Poisoning: Sulfur, and the Heterocycles Nobody Warns You About

Catalyst poisoning is when a species binds the metal surface strongly enough to block substrate access. Divalent sulfur is the classic poison: thiols, thioethers, and thiophenes will kill palladium at very low levels, sometimes at loadings below one mole percent. If your molecule contains sulfur in any reduced form, tell the CRO in the enquiry. The workarounds are higher catalyst loading, sulfur-tolerant catalysts such as sulfided platinum, or a route change.

Basic nitrogen heterocycles are the poison people forget. Pyridines, imidazoles, and free amines coordinate to metal surfaces and slow reactions substantially. The standard fix is to run the reduction on the hydrochloride or acetate salt, or to add a stoichiometric acid, which protonates the nitrogen and removes it from competition. Compounds like 3-Cyanopyrrolidine HCl (CAS 1187930-86-4) are supplied as salts partly for exactly this reason, and a nitrile reduction on the free base of that scaffold behaves quite differently from the same reaction on the salt.

Catalyst Loading Economics

Standard loadings run 5 to 10 weight percent of a 5 or 10 percent metal-on-carbon catalyst, which puts actual metal at roughly 0.25 to 1 mole percent. At gram scale nobody cares. At 10 kilograms with rhodium, catalyst is a five-figure line item and loading optimization pays for itself in a single batch. Ask whether the CRO screens loading as a matter of course, whether they recover and return spent catalyst, and how they account for the metal value. Some return spent catalyst to a refiner and credit you; most do not unless you ask.

Asymmetric Hydrogenation: Enormous Value, Honest Framing

Asymmetric hydrogenation uses a chiral ligand bound to the metal to deliver hydrogen to one face of a prochiral substrate, producing a single enantiomer directly. When it works, it is the best chiral technology available: high enantiomeric excess in one step, no fifty percent yield ceiling from a resolution, catalytic quantities of the chiral information, and clean scale-up.

What Substrates Are Realistic

The best-established asymmetric hydrogenations are enamides, dehydroamino acids, unsaturated carboxylic acids, and certain ketones. Rhodium with a chiral bisphosphine handles most alkene cases; ruthenium with a BINAP-type ligand handles ketones. Substrates carrying a coordinating group near the reacting bond do best, because that group anchors the substrate in a defined orientation on the metal. Substrates with no such handle are much harder and often not worth attempting.

Ligand Screening Is a Project, Not a Step

This is where buyers get disappointed. There is no reliable way to predict which chiral ligand will work on a novel substrate. The practical approach is a screen: twenty to sixty combinations of metal precursor, ligand, solvent, additive, pressure, and temperature, run in parallel in small vessels, with chiral HPLC on every well. Realistic timeline is four to twelve weeks depending on how many hits appear early. Budget accordingly, and treat it as contract R&D rather than as a synthesis step with a fixed price.

Ask whether the CRO has a parallel pressure reactor block, how many ligands they hold in-house, and whether they have licenses for the ligand families they propose. Proprietary ligands carry license terms that matter at commercial scale, and finding that out after a successful screen is a bad time to find out.

Measuring Enantiomeric Excess Properly

Enantiomeric excess is only as good as the method that measured it. Chiral HPLC with a validated separation and a confirmed racemic standard is the baseline. A method that has never been challenged with authentic racemate cannot demonstrate that both enantiomers actually resolve, and a co-eluting minor enantiomer reads as 99 percent ee when the true value is much lower. Insist on the racemate chromatogram alongside the sample. Our analytical services group runs chiral HPLC method development for exactly this verification step, and the same discipline applies to any supplier’s data.

The Selectivity Problems That Cost Real Money

Four failure modes account for most hydrogenation surprises. All four are predictable from the structure, and all four are cheaper to screen for than to discover.

Dehalogenation: The Expensive Classic

Palladium on carbon strips aryl halides under hydrogen. The rate order is iodide much faster than bromide, bromide faster than chloride, with aryl fluorides essentially stable. This is the most common expensive surprise in the whole field, because halogenated aromatics are everywhere in medicinal chemistry as cross-coupling handles, and the halogen is frequently installed several steps earlier at real cost.

If you are hydrogenating a nitro group on a bromoindole scaffold such as 4-Bromoindole (CAS 52488-36-5), palladium will remove the bromide alongside the intended reduction, and you will get a mixture that is difficult and wasteful to separate. The fixes: switch to platinum oxide or platinum on carbon, run cold and at the lowest pressure that converts, add a mild acid, or use a chemical reduction such as iron in acetic acid or tin chloride instead of hydrogenation entirely. Aryl iodides are the worst case; a substrate like 4-Iodobenzonitrile (CAS 3058-39-7) will lose iodine under palladium hydrogenation very quickly indeed, so a nitrile reduction on that scaffold has to be planned around the halogen from the start. The broader halogenated compounds category is full of substrates where this question arises.

Over-Reduction

Stopping a reaction at the intended oxidation level requires either an intrinsically slower second step or active monitoring. Alkyne to alkene needs a poisoned catalyst. Reductive amination should not proceed to the alcohol. A pyridine reduction targeted at the tetrahydropyridine should not run through to the piperidine. Hydrogen uptake monitoring is the practical control: when the theoretical number of equivalents has been consumed, stop. A supplier without uptake measurement is relying on time-point sampling, which is workable but slower and less precise.

Epimerization

Stereocenters alpha to a carbonyl or alpha to a nitrogen can epimerize on a metal surface, particularly under basic conditions and elevated temperature. If you are hydrogenating near an existing stereocenter, ask for a chiral purity check on the product rather than assuming the center survived. A protected ketopiperazine like 1-BOC-3-oxopiperazine (CAS 76003-29-7) is a useful example of the geometry where this question arises, and it also illustrates the Boc-stability question, since Boc survives neutral hydrogenation but not acidic conditions.

Unintended Ring Reduction

Electron-rich heterocycles can reduce under conditions chosen for something else. Indoles reduce to indolines, pyrroles to pyrrolidines, and furans to tetrahydrofurans, particularly under platinum or rhodium and particularly in acid. If your intended reduction is elsewhere in the molecule, the heterocycle is a liability at forcing conditions. Conversely, when the reduced ring is the target, the same reactivity is exactly what you want — 2-Phenylpyrrolidine (CAS 1006-64-0) type scaffolds are commonly made this way from the corresponding pyrrole or via reductive cyclization.

Safety and Engineering: What Separates a Real Hydrogenation Lab

Hydrogenation carries two independent hazards that compound each other: a flammable gas with an unusually wide explosive range, and catalysts that ignite in air. A lab that handles both properly looks visibly different from one that does not.

Hydrogen Itself

Hydrogen is flammable in air from roughly 4 to 75 percent by volume, one of the widest ranges of any common gas, and its minimum ignition energy is low enough that static discharge suffices. It is also the smallest molecule, so it leaks through fittings that hold other gases, and it burns with a nearly invisible flame. Design responses include hydrogen detection with interlocks, no-spark electrical fittings, adequate ventilation with attention to ceiling accumulation since hydrogen rises, and cylinder storage outside the working area. The OSHA hydrogen standard sets the baseline requirements in the United States.

Pyrophoric Catalyst Handling

Dry palladium on carbon ignites on contact with air, especially when wet with a flammable solvent. So does Raney nickel, which must never be allowed to dry. Correct handling means water-wet catalyst wherever possible, weighing under inert atmosphere or with a wetting solvent, adding catalyst to a vessel that has already been inerted, and keeping spent catalyst filter cake wet until disposal. Filter cake fires are among the most common hydrogenation incidents, and they happen almost exclusively during workup rather than during the reaction.

Inerting and Purge Cycles

Every hydrogenation begins and ends with purge cycles. The vessel is evacuated and refilled with nitrogen several times to remove oxygen, then purged with hydrogen to remove nitrogen, and the reverse at the end before opening. Three cycles is a common minimum. Ask a CRO to describe their purge procedure. The answer tells you whether hydrogenation is routine there or occasional.

Pressure Vessel Certification

This is the question buyers skip and should not. A pressure vessel has a rated maximum allowable working pressure established at manufacture, a rupture disc or relief valve sized to that rating, and a periodic inspection requirement. In the United States, vessels above certain size and pressure thresholds fall under ASME Section VIII, and the ASME boiler and pressure vessel code is the governing standard. Ask for: the nameplate maximum allowable working pressure, the date of the last inspection, the relief device rating and its last test date, and who performs the inspections. A supplier who can answer these in one email has a real program. A supplier who has to go and look has an autoclave someone bought and nobody owns.

Mass Transfer Is the Real Scale-Up Constraint

Here is the single most important engineering fact in hydrogenation, and the one most often missed in a technology transfer. At laboratory scale the reaction rate is usually limited by chemical kinetics. At scale it is almost always limited by how fast hydrogen dissolves from the headspace into the liquid.

Hydrogen has low solubility in most organic solvents. The catalyst can only react with dissolved hydrogen, so if consumption at the catalyst surface exceeds the rate of dissolution, the reaction runs at whatever the mass transfer rate allows, and adding more catalyst does nothing. The transfer rate depends on interfacial area, which depends on agitation type and speed, on gas induction if the impeller is designed for it, on vessel geometry, and on headspace pressure.

The practical consequences are direct. A reaction that finishes in two hours in a shaken 100 mL bottle can take a full day in a 20 L autoclave with a poorly chosen impeller, at identical pressure, temperature, and catalyst loading. Selectivity changes too, because an intermediate that would be rapidly consumed under hydrogen-rich conditions accumulates under hydrogen-starved conditions and finds other reactions to undergo. This is why hydrogenation scale-up demands more than a linear scaling of quantities, and it connects to the broader issues covered in scale-up challenges in custom synthesis and in process chemistry optimization from lab to pilot.

Ask what impeller type the CRO uses, whether they have gas-inducing agitators, and whether they characterize the volumetric mass transfer coefficient for their vessels. Most will not have measured it. Some will know the number for their standard configuration, and those are the ones whose scale-up predictions you can trust.

When Continuous Flow Is the Better Answer

Continuous hydrogenation in a packed catalyst bed sidesteps several of these problems at once. The catalyst is immobilized, so filtration and pyrophoric handling largely disappear. Hydrogen inventory at any instant is small, which improves the safety case dramatically. Mass transfer is excellent because the liquid film over the catalyst is thin. The tradeoffs are catalyst bed lifetime, sensitivity to particulates, and the fact that not every substrate tolerates the residence-time distribution. The full comparison sits in flow chemistry versus batch, and the decision usually turns on whether the campaign is large enough and repetitive enough to justify the setup.

Residual Metal: The Requirement That Follows Every Hydrogenation

Every hydrogenation leaves metal in the product, and every pharmaceutical intermediate has to answer for it. This is not optional and it is not a downstream problem.

What the Limits Are

ICH Q3D classifies elemental impurities by toxicity and route of administration. Palladium, platinum, rhodium, ruthenium, and nickel all appear with defined permitted daily exposures. Palladium and platinum sit in Class 2B, with oral permitted daily exposures around 100 micrograms per day and parenteral limits roughly ten times lower. Nickel is more restrictive still. Translated into a concentration specification at typical daily doses, most oral APIs need palladium at or below roughly 10 parts per million, and parenteral products need considerably less. The ICH Q3D guideline is the authoritative text, and the FDA guidance on elemental impurities mirrors it for US filings.

How Metal Actually Gets Removed

Filtration through a pad of filter aid removes the bulk catalyst and typically leaves single-digit to low tens of parts per million of dissolved and colloidal metal. That is often not enough. The usual sequence is:

1. Hot filtration through celite or a sintered pad to remove bulk catalyst while the product stays in solution
2. Activated carbon treatment, cheap and effective for colloidal metal but costs yield through adsorption of product
3. A functionalized scavenger resin — thiol, thiourea, or trimercaptotriazine-based — which is the reliable route to single-digit parts per million but adds cost and a contact-time step
4. Crystallization, which frequently rejects metal well and is worth trying before buying resin
5. ICP-OES or ICP-MS confirmation on the isolated solid, because none of the above can be assumed

Ask the CRO which of these they run as standard, whether metal analysis is included in the release package or quoted separately, and what levels they typically achieve on a substrate like yours. A supplier who quotes hydrogenation without mentioning metal removal has not thought about your filing. The detail sits in our post on elemental impurities and ICH Q3D for API buyers.

The Capability Questions to Ask, In Order

Run this list on any prospective supplier. It takes one email and it sorts the field quickly.

1. What is your maximum working pressure, and what is the nameplate rating and inspection date on the vessel that delivers it? The two halves of the question matter equally. An uncertified vessel rated by hope is a liability, not a capability.
2. What vessel sizes do you have, and what is the working volume range at that pressure? Capability at 100 bar in a 100 mL vessel does not scale to a kilogram delivery. Ask for the ladder: screening volume, development volume, delivery volume.
3. Can you screen catalysts in parallel, and how many conditions at once? A parallel pressure reactor block turns a three-week sequential optimization into a week. Without one, screening is serial and expensive.
4. Do you measure hydrogen uptake? Uptake curves give reaction progress, endpoint detection, and a diagnostic for mass transfer limitation. Their absence is not disqualifying but it slows everything.
5. What is your catalyst filtration and metal removal capability, and can you demonstrate ICH Q3D compliant levels? Ask for a representative ICP result on a prior project, redacted as needed.
6. How is hydrogen supplied and handled? Cylinder, bundle, or generator; detection and interlocks; storage location; who is trained.
7. What have you actually run above 20 bar in the last twelve months? This is the question that cannot be bluffed. Ask for substrate classes and scales, not names. A supplier with real capability will answer easily and specifically.
8. What is your position on sulfur-containing and heterocycle-rich substrates? Their answer reveals whether they have hit poisoning problems and learned from them.
9. Do you subcontract any part of this? Not disqualifying, but you need to know, and you need it in writing before the purchase order.

Reading the Answers

The strongest signal is specificity. A supplier who answers question seven with “we ran a pyridine reduction at 80 bar on 400 grams in March, rhodium on carbon, and we had to add acetic acid to get conversion” is telling you something a brochure cannot. A supplier who answers “yes, we do high-pressure hydrogenation regularly” is telling you nothing.

The second signal is what they ask you. A capable hydrogenation group will come back asking about halogens, sulfur, existing stereocenters, protecting groups, and target metal specification before quoting. That reverse questioning is the clearest evidence that they have run these reactions and been burned by them.

Where Domestic Capability Sits

Hydrogenation capability in the United States is unevenly distributed, and pressure capability above 50 bar is genuinely scarce at CRO scale. Our US-based custom synthesis capability map covers the landscape, and the same scarcity logic that applies to cryogenic chemistry applies here: specialized equipment concentrates in a small number of facilities, and the ones that have it are booked further out.

ChemContract Research operates high-pressure hydrogenation as a standing capability within US-based custom synthesis from milligram to multi-ton, supported by contract R&D for catalyst and ligand screening and analytical services including ICP-OES for residual metal and chiral HPLC for enantiomeric purity. If you have a reduction that needs pressure, a substrate carrying a halogen or a sulfur atom you are worried about, or a metal specification you need to hit, send us the structure and the target and we will return a technical assessment and a quote within 24 hours.

Frequently Asked Questions

What pressure counts as high-pressure hydrogenation?

In practice, anything above roughly 50 bar. Balloon chemistry runs at atmospheric pressure, a Parr shaker typically to 4 or 5 bar, a standard stirred autoclave to about 50 bar, and high-pressure work runs 100 bar and above. Each tier is a distinct equipment class with its own vessel certification requirements, so a supplier who owns a Parr shaker cannot run your 80 bar aromatic ring reduction.

Why does Pd/C remove my aryl bromide during hydrogenation?

Palladium on carbon catalyzes hydrodehalogenation, cleaving carbon to halogen bonds under hydrogen. Aryl iodides go fastest, then bromides, then chlorides, with fluorides largely safe. If your substrate carries a halogen you need to keep, switch to platinum oxide or a platinum on carbon catalyst, lower the pressure and temperature, and consider adding a mild acid or a poison modifier. Screen this before committing material.

Which catalyst should I use for a nitro to amine reduction?

Palladium on carbon at 1 to 5 bar is the default and usually works in hours. Choose platinum oxide or platinum on carbon instead when the molecule carries a benzyl group, a halogen, or an alkene you want to preserve. Raney nickel is a cheap alternative at scale but requires higher pressure and careful pyrophoric handling. Screen two or three options in parallel rather than optimizing one.

How much residual palladium is acceptable in a pharmaceutical intermediate?

ICH Q3D places palladium in Class 2B with a parenteral permitted daily exposure of 10 micrograms per day and an oral limit of 100 micrograms per day. Translated to concentration at typical dosing, most specifications land around 10 parts per million or lower for an oral API. Catalyst filtration alone rarely gets there, so plan for a scavenger resin or activated carbon polish plus ICP-OES confirmation.

Why does hydrogenation slow down when scaled up?

Because hydrogenation at scale is limited by gas to liquid mass transfer rather than by reaction kinetics. Hydrogen must dissolve into the liquid phase before it reaches the catalyst surface, and dissolution rate depends on interfacial area, agitation, and headspace pressure. A reaction that finishes in two hours in a 100 mL Parr bottle can take twenty hours in a poorly agitated 20 L autoclave at identical pressure and temperature.

Is asymmetric hydrogenation worth attempting for a chiral intermediate?

It is worth evaluating when the transformation will run repeatedly at scale, because a working chiral ligand system delivers high enantiomeric excess in one step with no yield loss to resolution. Treat ligand screening as a project rather than a step. Budget four to twelve weeks and expect to test twenty or more ligand and metal combinations before you know whether a viable system exists.