Sustained Deep-Work Stack for Knowledge Workers: 5 Daily Supplements That Actually Hold Focus

Here's the short version for anyone who just wants to work two hard blocks a day without falling apart at 2pm: you need five things. That's it.Stack-kit editorial

If you're a knowledge worker trying to hold two real deep-work blocks across a 6-10 hour day without the 2pm collapse, the protocol is shorter than the supplement aisle suggests. Five daily items, ~$95-145/month if you buy everything. Built to compose with ADHD medication, not replace it.

Most people get this exactly backwards. They go wide — twelve-ingredient "brain blends," a new powder every month — when the thing that actually moves a workday is narrow and slightly boring. We'll show you what earns a slot and, just as important, what we cut and why.

Quick answer — the protocol in 30 seconds

Daily, every workday:

1. Caffeine 100mg + L-theanine 200mg — taken together, 30 minutes before your first deep-work block
2. Creatine monohydrate 5g — any time, with or without food
3. Magnesium L-threonate 2g (Magtein) — evening, 1-2 hours before bed
4. Methylated B-complex — one capsule with breakfast
5. Electrolytes (~1000mg sodium + 200mg potassium + 60mg magnesium) — mid-morning to early afternoon

Key caveat. This stack composes with ADHD medication; it does not substitute for it. If you're prescribed Adderall, Vyvanse, Concerta, Ritalin, Strattera, or equivalent, keep taking it as prescribed and drop item #1's caffeine layer. Full medication-complement notes below.

Effect size, stated honestly. The caffeine + L-theanine pairing is the best-evidenced item here, and its effect is real but small-to-moderate in well-rested people — concentrated in the second hour after intake, not a step-change. The biggest, cleanest gains show up in impaired states (sleep deprivation). We flag this up front because the aisle sells transformation and the evidence supports a reliable, modest nudge.

The Protocol — Detailed

1. Caffeine + L-theanine — the foundation pairing

In plain terms: coffee makes you alert but wired; the theanine takes the wired part off so you get the focus without the jitter.

Caffeine works by blocking adenosine receptors — adenosine is the molecule that builds up through the day and makes you feel tired, so blocking it is what "waking up" chemically feels like. It also raises norepinephrine and dopamine and stretches out how long you can stay sharp on a cognitive task before fatigue sets in. The cost of that is familiar to anyone who's had a cup too many: jitter, anxiety, the rebound crash. L-theanine — an amino acid that shows up naturally in tea — raises alpha-wave activity in the brain (the relaxed-but-alert state) and files down caffeine's sympathetic edge. Of all the nootropic combinations people chase, this is the one with the most replications behind it. Alertness, minus the wired feeling.

So: 100mg caffeine + 200mg L-theanine, taken together 30 minutes before your first deep-work block (a 2:1 theanine-to-caffeine ratio, the one most studied). You can run an optional second dose in the early afternoon — 50mg caffeine + 100mg theanine — but never later than 6 hours before bedtime. Food or no food is fine; eating first slows onset by about 20 minutes but is easier on the stomach.

On timing and half-life. Caffeine's half-life is roughly 5 hours, so a 100mg afternoon dose still leaves ~50mg in circulation near a 10pm bedtime — keep doses before early afternoon if you protect sleep. L-theanine clears far faster (half-life ~1 hour), which is why the "calm" fades while the alertness lasts; if a long block runs past the first 90 minutes, re-dose the theanine, not the caffeine.

For the theanine, the bottle we'd actually buy is Momentous L-Theanine — NSF Certified for Sport, 200mg capsules, ~$26/60 servings. Look for Suntheanine (the studied, patented L-isomer) on the label. For the caffeine half, skip the pill and use coffee or matcha from a verified single-origin source; the dose is identical and the ritual earns its keep. If you genuinely need pill-form for timing precision, the Bulletproof Caffeine + L-Theanine combo capsules cover it — but only buy a pre-combined product if the label states both milligram amounts (you can't hit the studied dose if it hides behind a "proprietary blend").

The evidence, stated honestly. The traceable acute evidence is smaller and more useful than a single cinematic number: Owen et al. 2008 (n=27, double-blind crossover) and Giesbrecht et al. 2010 (n=44) both found the caffeine + L-theanine pairing improved attention/task-switching measures versus placebo or caffeine alone. Camfield et al. 2014, a Nutrition Reviews systematic review/meta-analysis of tea constituents, found the strongest signal for L-theanine + caffeine on attention and task switching in the first hour after dosing, with small trials and heterogeneous tasks. That is enough to earn the slot; the credible claim is modest acute attention support, not a guaranteed electrophysiology outcome. Note what the evidence does not support: L-theanine alone as a focus supplement. Theanine earns its slot by smoothing caffeine and improving the quality of the stimulant signal, not by acting like a standalone attention drug.

When to leave it alone: if you're on stimulant ADHD medication, the caffeine layer is redundant and stacking it raises blood pressure and worsens the medication crash — take the L-theanine alone at 200mg instead. Skip the caffeine layer if you have a history of arrhythmia, palpitations, uncontrolled hypertension, panic/anxiety disorder, or if a clinician has told you to limit stimulants. If you're pregnant or trying to become pregnant, keep total daily caffeine under 200mg unless your obstetric clinician gives different guidance.

2. Creatine monohydrate — the underrated cognition layer

Quick version: the gym supplement also feeds your brain, and it matters most if you don't eat much meat or you're short on sleep.

Creatine's job is to buffer ATP regeneration — ATP being the cell's basic energy currency — in tissues that burn through a lot of it fast. Muscle is the obvious one. The brain is the one nobody markets. The cognitive evidence arrived later than the athletic evidence but it's gotten steadily harder to wave off: creatine cuts mental fatigue under sleep deprivation, heavy cognitive load, and low-oxygen conditions. The biggest responders are vegetarians and vegans, for the simple reason that dietary creatine lives in meat.

Dose is 5g/day, any time, with or without food. Ignore the loading-phase advice you've seen — that's an athletic-performance ritual and it does nothing for everyday cognition. Stir it into morning coffee, water, or a smoothie. The effect builds over 2-4 weeks, so don't go looking for a day-one difference; you won't find one.

We'd reach for Pure Encapsulations Creatine Monohydrate — micronized, no fillers, third-party tested, ~$22/30 servings. Thorne Creatine is the same tier of quality if that's what's in front of you. Prefer Creapure-sourced or NSF Certified for Sport monohydrate; skip "HCl," "buffered," and flavored creatine blends — they cost more without better evidence and the flavoring usually drags in sucralose and artificial colors.

The numbers, honestly framed: a 2023 Nutrition Reviews meta-analysis (Prokopidis et al.) found creatine improved memory overall (SMD ~0.29–0.31), with the clearest benefit in older adults (66–76y) and a weaker signal in younger people; the acute working-memory effect appears tied to higher short-term dosing (~20g/day for 5–7 days) rather than the 5g maintenance dose. A separate 2024 Scientific Reports trial (Gordji-Nejad et al., n=15) showed a single ~0.35 g/kg dose improving processing speed and short-term memory during 21 hours of sleep deprivation — but that high single dose is a research protocol, not a daily habit. Earlier work (Rae et al. 2003, n=45 vegetarians) reported improved performance on Raven's Progressive Matrices after 6 weeks at 5g/day. Bottom line: creatine for focus is a sleep-debt and aging-brain tool, not a daily focus pill for a rested 25-year-old.

Skip it if you're already eating 6+ ounces of red meat or fish daily — your baseline stores are near saturation and supplementing on top buys a marginal cognitive effect at best. And if you have diagnosed kidney disease, sit this one out: creatine is safe at 5g/day in healthy adults across studies running 5+ years, but if your kidneys are compromised, that's a conversation for your physician first. (Creatine modestly raises serum creatinine — a lab marker, usually not a sign of harm — so flag your use before any kidney panel.)

3. Magnesium L-threonate — the sleep-supported cognition layer

The short story: this is the premium magnesium form with human cognitive data. Use it as a long-horizon memory/sleep-support layer, not as a same-day deep-work engine.

Magnesium is a cofactor in something like 600 enzymatic reactions, but here's the catch most labels won't tell you — most forms of it barely cross the blood-brain barrier, the filter that decides what's allowed into brain tissue. L-threonate is the form with the strongest brain-delivery rationale. In animal models it raises synaptic density in the prefrontal cortex and hippocampus; in humans, the evidence points to working-memory and executive-function support in older adults with cognitive complaints. That is useful, but bounded: the study is not a knowledge-worker productivity trial, and it does not justify treating Magtein as the thing that creates your deep-work block.

Take 2g magnesium L-threonate (delivering ~144mg elemental magnesium) in the evening, 1-2 hours before bed. Splitting it — 1g in the afternoon, 1g in the evening — works just as well and can smooth out focus through the back half of the workday.

The brand is Momentous Magnesium L-Threonate (Magtein, the patented form used in clinical trials), ~$48/30 servings. This one isn't interchangeable: Magtein is the only L-threonate with human RCT data behind it, and generic L-threonate may or may not deliver the same bioavailability — meaning how much of it your body actually absorbs and puts to use.

The study people cite is Liu et al. 2016 (n=44, double-blind RCT, 12 weeks): a Magtein-based formula at roughly the 2g/day magnesium L-threonate dose improved a composite cognitive score including executive function, working memory, episodic memory, and attention in adults 50-70 with cognitive complaints. Treat that result as promising, not settled: it is one small 12-week study, not a replicated deep-work outcome, and the most dramatic framing of the effect size came from the authors.

Don't bother switching if you're already on magnesium glycinate, citrate, or malate above 300mg elemental for sleep or muscle recovery. Magtein runs about 4x the price of glycinate, and the marginal cognitive bump may not clear that gap. Stay with what you've got and revisit only if you specifically want the prefrontal-cortex effect.

4. Methylated B-complex — the foundational cofactor layer

In plain English: B vitamins are the raw material your brain uses to build the chemicals that run your mood and focus, and a meaningful share of people can't process the cheap form properly.

B vitamins are cofactors for neurotransmitter synthesis — B6 feeds serotonin and dopamine, while B9 (folate) and B12 drive the methylation cycles that shape mood and cognition. The word "methylated" on the label is doing real work: a substantial share of people carry MTHFR polymorphisms — common genetic variations that can blunt the body's ability to convert folic acid into active folate — and methylfolate sidesteps that bottleneck. Knowledge workers running on bad diets, heavy alcohol, or chronic stress are frequently depleted in these without ever tripping the classic deficiency markers.

It's one capsule of a methylated B-complex with breakfast. Single dose — B vitamins are water-soluble with short half-lives (a half-life being how long it takes the body to clear half a dose), so morning timing parks the cofactor peak right where the workday needs it.

The pick is Thorne Basic B Complex — methylfolate (5-MTHF), methylcobalamin (B12), P-5-P (B6), pantethine (B5). Third-party tested, no fillers, ~$24/60 servings. Avoid any B-complex built on folic acid and cyanocobalamin; those are the cheap, unmethylated forms.

Evidence: Kennedy et al. 2010 (n=215, double-blind RCT) found high-dose B-complex supplementation in healthy young adults reduced subjective stress and improved cognitive-test performance after 28 days. The effect was moderate, and the biggest gains came from subjects who started out low — which is precisely why we file this as a foundational layer rather than something you'll feel acutely.

Skip it if you eat a diverse diet — regular eggs, leafy greens, salmon, grass-fed beef — and recent bloodwork puts your serum B12 above 500 pg/mL with red blood cell folate in the upper third of the reference range. At that point you're saturated, and the B-complex turns into expensive urine. Test before stacking if you can.

5. Electrolytes — the boring layer that fixes the afternoon crash

Plainest possible version: a lot of what feels like the 2pm slump is just dehydration. Salt and water fix more of it than another coffee will.

The classic mid-afternoon cognitive slump is, more often than not, dehydration and electrolyte loss rather than a stimulant shortfall. Knowledge workers under-drink water, over-drink coffee (which is mildly diuretic — it nudges you toward losing fluid), and quietly shed sodium in air-conditioned rooms and skipped meals. Replacing those electrolytes is the single highest-leverage afternoon move, and almost nobody makes it.

One serving (~1000mg sodium, 200mg potassium, 60mg magnesium) mixed in 16-20 oz water, sipped between 11am and 2pm. Morning trainers should take it post-training in place of the mid-morning coffee.

For brands: LMNT (1000mg sodium per stick) on high sweat-rate or post-exercise days. Athletic Greens AG1 if you want the multivitamin layer folded in — but be clear-eyed that AG1 is a whole stack, not a single-purpose product; buy it for what it actually is, not for one headline claim. The cheapest route that still works is DIY: 1g salt + ¼ tsp NoSalt (potassium chloride) + 1 magnesium glycinate capsule in 20oz water.

The supporting data: Edmonds et al. 2013 (n=58, double-blind) found that drinking 500mL of water before a cognitive battery improved reaction time by 14% in mildly dehydrated subjects (mild defined as >1% body mass deficit). The electrolyte layer stretches that same effect across longer windows and holds up against the diuretic drag of coffee.

Skip it if you have diagnosed hypertension and a physician has put you on a sodium-restricted diet — 1000mg per stick is half a typical daily ceiling, so use plain water + magnesium instead. Skip it too if you're already taking in 4+ liters of electrolyte-replete fluids a day, which covers most endurance athletes in season.

What to cut — and why

The cognitive-enhancement aisle is crowded with products that don't earn a place in a serious knowledge-worker stack. Here are the ones we name explicitly — and the reasoning, because "trust us" isn't an argument.

Proprietary "focus" / "unlock your brain" blends with no per-ingredient doses

This is the single biggest waste. U.S. labeling law lets a company list a "proprietary blend" with only the total weight — so a label can name L-theanine, citicoline, and Bacopa while including fairy-dust amounts of each. You cannot hit the 200mg theanine / 100mg caffeine that the evidence actually uses if the label won't tell you the milligrams. Rule: if it doesn't disclose dose per ingredient, you're buying a guess.

Nootropic "stacks" with 12+ ingredients

Alpha Brain, Qualia Mind, Mind Lab Pro, Onnit-class blends. The kitchen-sink approach is engineered to sound rigorous and lands as the opposite. Every ingredient shows up at a dose well below where it has any independent evidence, the combinations have never been validated as combinations, and the monthly cost runs 3-5x what a focused 5-item stack delivers. The rule is simple: if the per-ingredient dose isn't at study levels, the ingredient isn't doing the work.

Modafinil / armodafinil purchased online

Modafinil is a prescription drug — Schedule IV in the US. Whether off-prescription use makes sense for you is a medical decision between you and a physician, not a stack decision. And we don't point people at supply chains we can't verify; prescription drugs from online pharmacies fall into that bucket by default.

Lion's mane for acute focus

Lion's mane has genuinely interesting evidence for long-arc neurogenesis and nerve growth factor expression. For same-day focus it has effectively none. People stack it expecting this afternoon's deep-work block to sharpen — wrong tool, wrong timescale. If long-horizon cognitive longevity is what you're after, lion's mane has a home in a different protocol (sk:longevity). For today's workday, it's filler.

Phenylpiracetam, noopept, racetams

Poorly regulated, thin long-term safety data, and schedule-controlled in several jurisdictions including the UK and parts of the EU. The cognitive-enhancement case rests on small Russian trials from the 1980s-90s that haven't been independently replicated at modern methodological standards. For something you'd take daily, the risk-to-evidence ratio is simply pointed the wrong way.

Pre-workout powders used as "focus" supplements

Built for muscle output, not cognition. They typically pile 200-400mg caffeine on top of beta-alanine, citrulline, and proprietary blends, all dosed for a 60-90 minute training window — not a 4-hour deep-work block. The crash is engineered to land after your workout, which means it lands mid-afternoon if you've repurposed it as a cognitive tool. Wrong tool, and the failure mode is predictable.

Medication-complement notes

This protocol is designed to compose with, not replace, prescription medication.

If you take ADHD stimulant medication (Adderall, Vyvanse, Concerta, Ritalin, dexmethylphenidate, amphetamine salts): Keep taking your medication as prescribed. Drop item #1's caffeine layer (keep the L-theanine — it smooths stimulant edge the same way it smooths caffeine). Items #2-5 compose cleanly.

If you take SSRI / SNRI antidepressants: Items #1-5 compose without known interaction. If caffeine worsens your anxiety or sleep, drop it and keep the L-theanine.

If you take blood pressure medication: Drop the caffeine in item #1; reduce LMNT sodium dose in item #5 (use the DIY alternative at half-strength). Discuss with your prescribing physician before starting item #1.

If you take blood thinners (warfarin, apixaban, rivaroxaban): The B-complex can affect INR if you're on warfarin. Test more frequently for the first 4 weeks. No issue with apixaban or rivaroxaban.

This protocol is not a substitute for diagnosed ADHD treatment, depression treatment, or any other condition for which you have an active prescription. If you're considering replacing prescribed medication with supplements, that is a medical decision between you and your prescriber, not a stack decision. We will not recommend that path.

Frequently asked questions

How long until I notice anything?

It depends entirely on the item. Caffeine + L-theanine works day one, inside an hour. Electrolytes hit the same day if you were under-hydrated. Creatine takes 2-4 weeks for the cumulative cognitive effect — faster if you're vegetarian. Magnesium L-threonate needs 2-4 weeks for the working-memory effects, though sleep quality may pick up inside the first week. B-complex runs about 4 weeks if you were depleted, and produces no detectable effect if you weren't.

Can I stack this with ADHD medication?

Yes, with one modification — drop the caffeine layer in item #1. Keep the L-theanine; it smooths stimulant edge the same way it smooths caffeine. Items #2-5 compose cleanly with all common ADHD stimulants and non-stimulants.

Do I need all five daily items?

No. The high-leverage core is caffeine + L-theanine (item #1) and electrolytes (item #5) — if you're picking two to start, start there. Add creatine (item #2) next if you don't eat much meat or you're regularly sleep-deprived. Magnesium L-threonate and B-complex are foundational layers that take weeks to show themselves, so bring them in once you've confirmed the core works for you.

Why no lion's mane, ashwagandha, or rhodiola?

Each one is a timescale or category mismatch, not a quality knock. Lion's mane's evidence is for long-arc neurogenesis, not same-day focus — wrong tool for this protocol. Ashwagandha has stronger evidence for cortisol modulation and sleep than for daytime cognition. Rhodiola shows interesting fatigue-resistance data but replicates inconsistently at standard doses. All three may earn spots in adjacent protocols (sk:longevity, sk:sleep, sk:stress); none of them displaces anything in this stack.

What about bloodwork before starting?

Useful, not required. Two markers actually change the recommendation: serum B12 (above 500 pg/mL and you can probably skip item #4) and ferritin (low ferritin can mimic cognitive fatigue, and it's worth addressing head-on rather than papering over with stimulants). A basic metabolic panel and lipid panel are reasonable baseline reads if you're committing to a long-term stack.

How much does the full stack cost per month?

~$95-145/month if you buy everything at the recommended brands and don't chase bulk discounts. The cheapest sensible swap — generic creatine and DIY electrolytes — knocks off about $30/month with no quality loss. The priciest single item is Magtein magnesium L-threonate at ~$48/month. If budget is the constraint, prioritize items #1, #2, and #5; they return the most focus per dollar.

Will this work for ADHD without medication?

If you've got diagnosed ADHD and you're unmedicated, this protocol might take some edge off — but it is not a treatment. The evidence base for supplement-only ADHD management is thin, and the effect sizes are small next to first-line medications. If your ADHD is interfering with work or quality of life, the answer is a prescriber, not a stack. We will not recommend supplements as a substitute for diagnosed-ADHD treatment.

Evidence — key citations

1. Owen GN et al., Nutritional Neuroscience (2008) — 100mg L-theanine + 50mg caffeine improved speed and accuracy on an attention-switching task at 60 min vs placebo.
2. Giesbrecht T et al. (2010, n=44) — replicated the caffeine + L-theanine attention benefit on rapid visual information processing.
3. Camfield DA et al., Nutrition Reviews (2014) — systematic review/meta-analysis of tea constituents: L-theanine + caffeine shows small acute attention/task-switching benefits, strongest around the first hour; trial sizes are small and heterogeneous.
4. EFSA NDA Panel (2011) — the evidence submitted for L-theanine-alone cognitive-function claims was insufficient for a causal claim.
5. Caffeine safety guidance: FDA / ACOG — broad adult caffeine ceiling around 400mg/day for most healthy adults; pregnancy guidance commonly caps intake at 200mg/day.
6. Prokopidis K et al., Nutrition Reviews (2023) — creatine improved memory (SMD ~0.29–0.31), strongest in older adults (66–76y); acute working-memory benefit linked to ~20g/day short-term dosing.
7. Gordji-Nejad A et al., Scientific Reports (2024, n=15) — single ~0.35 g/kg creatine dose improved processing speed and short-term memory during 21h sleep deprivation (research protocol, not a daily dose).
8. Liu G et al. (2016, n=44, 12-week RCT) — Magtein-based magnesium L-threonate formula improved composite cognitive measures in adults 50-70 with cognitive complaints; single small study, not a direct deep-work trial.

Affiliate disclosure

Stack-kit earns affiliate commission on the brand recommendations above. We do not sell our own SKUs. We do not have a house brand, a premium tier, or a founder's discount. The brands we recommend earn their slots by evidence and third-party testing, not by paying for placement.

If a better brand emerges next quarter, we'll switch the recommendation and publish the reason. If a recommended brand changes its formulation, sourcing, or third-party verification status in a way that lowers our confidence, we'll update or remove it. Our incentive is to be right about the protocol, not loyal to any one manufacturer.

Every product link above goes to a one-time-purchase option. We do not push subscriptions. You shouldn't "subscribe and save" on a supplement until you've validated it works for you across 4+ weeks of one-time purchases.