This is the third document in the Citane / BUNA library. Each plays a different role:
The framework. Explains the leaf's chemistry — reservoirs, activation events, energy systems, process domains, and emergent states — and why processing choices matter.
The manual. Seven core pathways from harvest to cup, with toolbox, batch sheet, troubleshooting, and the framework's observation-first philosophy.
The workbook. A broader catalogue of pathways, sensory directions, and aging protocols — written as structured experiments to run, compare, and record, not as specifications to follow.
If you haven't read Paper I and Paper II, the pathways below will still make sense — but the reasoning behind why they're framed as open questions rather than guarantees comes from those two documents.
The pathways, sensory directions, and protocols in this document are structured experiments, not guaranteed recipes. Each one is a starting point — a leaf choice, a sequence of steps, an illustrative timeline — designed to produce something worth tasting and comparing, not a specification with a fixed endpoint.
Where a pathway names a flavour direction (honey-like, floral, citrus, and so on), this describes what to listen for in the cup, not a promised result. Where a timeline or set of parameters is given, treat it as a first attempt to calibrate — your leaf, your equipment, and your conditions at Thumpassery will shape the outcome, and the first several batches of any pathway should be treated as that calibration.
Many of the most interesting Citane expressions may not come from any single pathway followed in isolation, but from combinations — an oxidative pathway followed by aging, a fermentation finished with light roasting, a sensory direction pursued through two different routes and compared. This document is best read as a landscape of possibilities to explore and record (see Paper II, Part 8, for the batch sheet), rather than a catalogue of fixed destinations to reproduce exactly.
Citane processing applies heat, fermentation, oxidation, and aging to fresh coffee leaves — transforming them from a perishable botanical into a shelf-stable beverage base. Different processing choices lead to different sensory territories.
Citane is not a single method. It is a landscape of approaches, several of which are explored here as named pathways. The goal of this document is to map that landscape broadly — more broadly than Paper II's seven core pathways — as a set of starting points for experimentation, not as a menu of confirmed products.
Unlike coffee roasting — which is well-documented globally — coffee leaf processing is only recently being studied systematically. What follows draws on three kinds of source, distinguished throughout this document: Documented Research Protocol findings adapted from peer-reviewed coffee leaf or analogous-plant research; Traditional Practice methods documented in coffee-growing communities, referenced as inspiration; and Experimental Citane Protocol / Exploratory Concept — pathways and ideas developed for this project that have not yet been validated through batches at Thumpassery.
Experimental Citane Protocol
The pathways below borrow names and structure from tea processing (oolong, black, white) and from pu-erh aging — vocabulary that describes the kind of transformation each pathway is modelled on, not a confirmed outcome for coffee leaf. Per Paper I, Section 24, coffee leaf's own chemistry — particularly its xanthone (mangiferin) pool — means the resulting cup may diverge from its tea analogue even under similar processing. Treat each description below as "this is the kind of thing this pathway tends toward," not "this is what this pathway produces."
This catalogue is broader than Paper II's seven core pathways. Oolong, Roasted, Black-Style, and Koji-Assisted correspond directly to Paper II Pathways 3, 5, 4, and 6. The remaining entries below — White, Lactic Fermented, Yeast-Fermented, Aged (Sheng-Style), Smoked, and Fire-Kissed — represent additional, more exploratory territory not yet given full pathway treatment in Paper II.
Cold Fresh Citane: Enzyme-preserved, minimal oxidation. A starting point for bright, clean, grassy character — used fresh or briefly dried. Closely related to Paper II Pathway 1 (Green Citane, State Alpha).
White Citane: Slow shade-wilting with controlled finishing. A direction toward honey-hay subtlety. Exploratory — not yet covered in Paper II.
Oolong Citane: Wilt + bruise + partial oxidation. A direction toward layered, floral-to-toasty complexity — the most versatile pathway in this catalogue. Corresponds to Paper II Pathway 3.
Roasted Citane: Post-dry roasting for deeper flavour development — a direction toward comfort, warmth, and cereal body. Corresponds to Paper II Pathway 5.
Black-Style Citane: Heavy bruising, deep oxidation — a direction toward malty, dried-fruit richness and full body. Corresponds to Paper II Pathway 4.
Lactic Fermented Citane: A direction toward tangy-sweet character, bridging fresh and aged profiles. Exploratory — related to Paper II Part 6 (LAB as a tool) but not yet a named pathway there.
Koji-Assisted Citane: A direction toward umami depth and a savoury, novel profile. Corresponds to Paper II Pathway 6.
Yeast-Fermented Citane: A direction toward fruity esters and cider-like complexity. Exploratory — related to Paper II Part 6 (yeast as a tool) but not yet a named pathway there.
Aged Citane (Sheng-Style): Compressed and aged over months to years — a direction toward deep, earthy complexity, modelled on Sheng pu-erh. See the dedicated section below. Exploratory.
Smoked Citane: Light or heavy smoke exposure — a direction toward resinous or char-caramel notes. Exploratory.
Fire-Kissed Citane: Quick charcoal exposure — a direction toward mild char and roasted depth. Exploratory.
Each pathway can be adapted further through leaf age and position (Paper II, Part 2), harvest timing, equipment, and finishing choices. Together, they form a ladder of expression: from core styles, to premium directions, to experimental reserves — explored in more detail in "Citane Expression: From Core to Reserve" below.
Exploratory Concept
Rather than following a single pathway, Citane production can be question-driven: pick a sensory direction you're curious about, and use the starting points below as a first attempt at moving toward it.
Each direction below names a volatile compound that has been documented, in published research on coffee leaf or related plant material, as associated with the sensory character named (honey, floral, citrus, and so on). The OAV (Odour Activity Value) figures shown indicate the range over which that compound has been reported to be perceptible in some matrix — they are not measurements of Chandragiri leaf, and they are not a guarantee that processing this leaf in the way described will produce that compound at that level, or that the compound will be perceptible in the cup at all.
The leaf choice, processing, and "illustrative sequence" given for each direction are first attempts at calibration, not specifications. Treat the first several batches toward any direction as exploratory: follow the sequence, brew, taste, and record (Paper II, Part 8). What to look for is described as a possible signal to taste for — not a pass/fail marker. If it isn't there, that's a finding too, and may point toward a different leaf choice, a longer step, or a different direction entirely.
As with Paper I's Reservoir K (Volatile Precursors) — the most speculative reservoir in that framework — the connection between any of these compounds and Citane's actual chemistry remains unconfirmed. These six directions are a map of where to start looking, not a map of where you will arrive.
Experimental Citane Protocol
These five protocols are starting points adapted from general food-fermentation and tea-processing principles, structured as a comparison set: Protocol 1 is a baseline, and Protocols 2–5 each change one or two variables relative to it. Run them side by side where possible — the comparison is often more informative than any single protocol on its own.
"Mature coffee leaves (3–6 months old)" below is a starting description and doesn't map precisely onto Paper II's leaf-position system (L1–L6, Part 2). As a rough starting point, "3–6 months old, mature" is closer to Paper II's L2–L4 (mid-branch) than to terminal flush or senesced leaf — but this hasn't been cross-calibrated. Record the leaf position you actually use alongside the age, per the batch sheet (Paper II, Part 8).
Before running any fermentation-based protocol (2–5): Paper II, Section 21 (Fermentation Failures) describes specific signs that mean stop and discard — ammonia smell, unexpected mould colour, sliminess, or an off/chemical smell. None of these signs are repeated in the steps below; check Section 21 before your first attempt at any of these protocols, and again if anything about a batch seems off.
Purpose: Establish a baseline green leaf character with minimal oxidation — the reference point every other protocol here is compared against.
Associated direction: green/vegetal notes, mild bitterness, grassy aroma — related to Direction 6 above and Paper II Pathway 1 (Green Citane).
Brewing: 2g dried leaf in 200ml water @ 95°C; steep 7–8 minutes; strain.
Use case: Make this batch first, before any of Protocols 2–5. Everything else in this section is compared against it.
Purpose: Explore whether controlled fermentation moves the cup toward fruity, honey-like, or floral character relative to Protocol 1.
Associated direction: honey-like, fruity, floral; softer green notes than the non-fermented control — related to Directions 1, 2, and 4 above.
Brewing: 2g dried fermented leaf in 200ml water @ 95°C; steep 6–7 minutes; strain.
Use case: Compare directly against Protocol 1, same leaf source and brewing where possible. The difference between the two cups is the effect of this fermentation step — record it.
Purpose: Explore a middle ground between the green/vegetal and fruity/fermented directions — a cup with gentle acidity rather than either extreme.
Associated direction: honey-like, fruity, with mild lactic or citrus-like acidity — intended to be refreshing rather than sharp.
Brewing: 2–3g dried leaf in 200ml hot water (95°C); steep 6–8 minutes; strain.
Use case: Run side-by-side against Protocol 1 (non-fermented) and Protocol 2 (full fermentation) to see where this lands between them — and which of the three you and your tasters prefer.
Purpose: Explore whether a wither step, before bruising and fermentation, produces a softer, more rounded cup from tougher mature leaves — one half of a paired comparison with Protocol 5.
Best for: Older or more fibrous leaves.
Associated direction: a softer infusion, more rounded and full-bodied, with fruity/honey notes.
Brewing: 2–3g dried leaf in 200ml water (95°C); steep 6–8 minutes; strain.
Use case: This protocol and Protocol 5 use the same leaf, the same fermentation, and the same drying — the only difference is the cell-disruption method (withering + rolling here, vs. freeze-thaw + rolling in Protocol 5). Run both from a single split batch and compare directly; the difference in the cup is the effect of that one variable.
Purpose: Explore whether ice-crystal disruption of cell walls, as an alternative to withering, produces a cleaner or brighter cup — the paired comparison with Protocol 4.
Best for: Tougher mature leaves, as a direct comparison against Protocol 4.
Associated direction: potentially cleaner fruity/floral notes with enhanced extractability, possibly preserving more sweetness than the withering route — untested.
Note on this method: freeze-thaw disruption is not discussed in Paper I's Activation Events (Section 15, Mechanical Disruption), which covers rolling, wringing, bruising, and crushing. Its effects on the leaf's reservoirs haven't been mapped alongside those methods. This protocol is included as a documented alternative worth comparing — not because its chemistry is better understood than the others.
Brewing: 2–3g dried leaf in 200ml water (95°C); steep 6–8 minutes; strain.
Use case: Compare directly against Protocol 4 (same leaf split into two, same fermentation and drying). Record which you and your tasters prefer, and whether the difference is large enough to justify the extra freeze-thaw step.
Traditional Practice (Reference) Exploratory Concept (Timeline)
This protocol is modelled on Sheng pu'erh tea: leaves are made shelf-stable (steamed, pressed into cakes, dried to low moisture), then aged for months to years under controlled storage, with slow non-enzymatic changes (oxidation, polymerization) rather than active fermentation producing a progressively different cup over time.
What Citane borrows from this tradition is the process — press, dry fully, wrap breathably, store, monitor. What is not known is whether coffee leaf, with its different compound pool (notably the mangiferin/xanthone reservoir — see Paper I, Reservoir B), ages along a similar trajectory, a different one, or barely changes at all on the timescales pu'erh does. This is a genuinely open question — not an assumption this protocol can rely on.
Leaves are made shelf-stable first (sanitized, dried to low moisture), then stored under controlled conditions to observe how they change over time.
No Thumpassery aging records yet exist for any of the intervals below. The descriptions are hypotheses — drawn loosely from the general pu'erh aging arc — about what might be found, framed as things to taste for and record, not as a guaranteed sequence of results. Each interval is a checkpoint: open a sample, taste it, write down what's actually there, and compare to both the previous checkpoint and to these hypotheses. Over several aged batches, replace these hypothesis descriptions with your own observed pattern.
The protocols and pathways above form a ladder of expression for Citane — not a ladder of certainty, but a rough guide to how much exploration each tier represents:
Core Styles: The non-fermented control and its direct comparisons (Protocols 1–3). These are the most accessible starting points — closest to Paper II's core pathways, and the place to begin if your equipment or leaf is new to you.
Premium Directions: The six sensory directions (the Citane Compass) and Protocols 4–5. These require more technical attention and comparison work, and the relationship between process and outcome is less established — expect more calibration batches before results stabilise.
Experimental Reserves: Aged (Sheng-style), smoked, koji-fermented, and yeast-fermented directions. These represent the least-mapped territory in this document — in the aged case, by design, results may not be visible for months or years.
Each direction that proves interesting can be developed into its own product story — but "proves interesting" means tasted, compared, and recorded, not assumed from this document alone.
This workbook is for estate producers, home practitioners, and anyone exploring Citane processing who wants a broader set of starting points than Paper II's seven core pathways — with the understanding that "broader" means "less validated," not "more advanced."
Start with Protocol 1 (the non-fermented control). This gives you a baseline understanding of your leaf material and equipment, and a reference point for everything else in this document.
Then run a paired comparison. Protocol 2 against Protocol 1, or Protocol 4 against Protocol 5 — same leaf source, one variable changed. The difference between two cups, tasted side by side, is more informative than either cup alone.
Pick a sensory direction from the Compass, not a guaranteed outcome. Use the Citane Compass to choose a direction you're curious about (e.g., toward honey-like sweetness), follow the illustrative sequence as a first attempt, and taste for what's described under "what to look for" — treating its absence as useful information, not a failed batch.
Document everything. Leaf age and position, harvest date, every processing step with duration and temperature, drying method, brewing parameters, and tasting notes — using the batch sheet in Paper II, Part 8. Patterns will emerge from records, not from memory.
Observe combinations, not just single pathways. Many of the most interesting Citane expressions may come from combining transformations rather than following one pathway in isolation — an oxidative pathway carried into aging, a fermentation finished with light roasting, a sensory direction pursued via two different routes and compared. Per Paper II, Part 9, the interaction between two steps can be as informative as either step alone. When you try a combination, record it as its own entry — not as a variant of whichever single pathway it most resembles.
Iterate. Each batch teaches you something, whether or not it matches the hypothesis you started with. Refine your next attempt based on what you actually found.
The aim is not to arrive at a fixed set of "finished" Citane products, but to build, batch by batch, an understanding of how this leaf, at this estate, responds — and a record specific enough that useful results can be repeated, and surprising ones can be followed up.
Not every useful observation comes from a planned experiment. One of the early observations that shaped the Citane approach came from an attempt to understand Engere on its own terms — not to create a new beverage, but to understand an existing one.
A prepared batch of Engere-style decoction, made in larger quantity than could be consumed immediately, was refrigerated simply for storage. Revisited over the following days, it seemed more integrated and balanced than when first prepared — certain harsher edges seemed reduced, and dilution sometimes seemed to make aspects of the beverage easier to perceive rather than weaker.
No claim is made about what caused this — settling, slow chemical change during cold holding, sensory adaptation, dilution, serving temperature, or some combination, all remain open. The observation itself is the point: processing may not necessarily end when heat is removed from the vessel. Storage, holding time, dilution, and serving temperature may all be worth treating as variables in their own right, not just as serving logistics — and worth recording on the batch sheet alongside the processing steps themselves.
More generally: useful observations sometimes arise from practical necessity, accident, repetition, or curiosity rather than from a pathway followed as planned. The same applies to pathways themselves — while following one pathway at a time makes it easier to see what each variable does, combinations, intersections, and permutations of pathways may produce outcomes that no single pathway produces alone (see "Observe combinations," above). Recording such interactions — planned or not — may be as valuable as recording the pathways themselves.