Granular, Pellet or Sphere: Choosing an Activated Carbon Form

An extruded pellet bed runs about 1 inch of water column per foot of bed depth. A granular bed in the same duty runs about 2 — half the pressure drop, on the published odour-control rule of thumb. That ratio settles most granular-versus-pellet arguments before anyone talks about capacity. A sphere buys something different again, and it is not pressure drop: Suracsh’s SURSORB Ultra is specified at ball pan hardness min 99 and ash max 0.5 %, which is why spherical carbon ends up sewn into protective clothing rather than dumped into a cheap odour-control vessel.

The short version

• Odour-control practice budgets granular carbon at about 2 in. w.c. per foot of bed and extruded/pelletised carbon at about 1 in. w.c. per foot, within a typical total design allowance of 6 in. w.c.
• In the Ergun equation the viscous term scales as 1/d_p² and 1/φ² and the inertial term as 1/d_p and 1/φ — larger and more spherical means lower pressure drop. No published side-by-side dP curve for the three forms could be found.
• US Patent 4,228,037 reports its spherical product gave no detectable dust by gravimetric analysis, against 0.10–0.18 % for the comparative granular samples. That is the one hard published number here.
• Hardness is a lot-comparison index, not a durability rating. ASTM D3802 says so in its own scope, and AWWA B604 applies different abrasion tests to different raw materials.
• Suracsh publishes all three forms: granular SURSORB SA (hardness min 98), the pelletised SURSORB PAC family (hardness min 96–98), and the spheres SURSORB Ultra (BET min 1350 m²/g, hardness min 99, ash max 0.5 %) and SURSORB Neo (BET min 1100 m²/g).

How the three forms are made

Granular carbon is crushed and screened: a carbonised, activated feedstock — coconut shell, coal, wood — is broken down and sieved into mesh cuts such as 8×30 or 12×20. Particles are irregular and span a size band, not a single diameter.

Extruded pellets are formed, not broken: ground carbon or char is mixed with a binder, forced through a die, cut, then carbonised and activated. The result is a cylinder of essentially uniform diameter — 2, 3, 4 or 6 mm are common. That uniformity fixes the void fraction, and therefore the pressure drop.

Spheres are grown rather than shaped, from polymeric precursors in Suracsh’s case — a different route entirely from crushing a shell. Their ash reflects the purity of a synthetic feedstock rather than the mineral content of a plant or a coal seam, which is why SURSORB Ultra can be specified at ash max 0.5 % while a coconut-shell pellet sits an order of magnitude higher.

Why does particle shape change pressure drop?

Because pressure drop through a packed bed scales inversely with particle diameter and inversely with sphericity. In the Ergun equation the viscous term carries 1/d_p² and 1/φ² and the inertial term 1/d_p and 1/φ, so doubling particle diameter cuts the viscous contribution by four and raising sphericity towards 1 cuts both terms. Larger and more regular therefore means lower pressure drop, which is the whole granular-versus-pellet argument — roughly 2 in. w.c. per foot against roughly 1. The equation itself, its variables and the EPA’s fitted gas-phase correlation are set out in bed sizing, EBCT and contact time.

What we will not tell you is that spheres deliver a specific measured percentage less pressure drop than pellets. No published side-by-side dP curve for the three forms under the same conditions could be found. What exists is the Ergun dependence above, plus the case US Patent 4,228,037 makes for its spherical product — a regular, uniform particle that packs predictably, which is the same argument the Ergun exponents make. That wording is ours, not the patent’s; the patent’s own hard number is the dust result below. The sphere’s dP advantage is a derivation from a cited equation, not a published measurement.

What is actually published about dusting and attrition

One hard number exists. US Patent 4,228,037, Spherical activated carbon having low dusting property and high physical strength, reports a dust test in which the spherical product gave a level so low it was “difficult to measure the dust by the gravimetric analysis,” against 0.18 % and 0.10 % for the comparative granular samples. The same document gives its spheres 0.1–1.5 mm diameter and surface area 800–1,600 m²/g.

Manufacturer claims fill in the rest and should be read as manufacturer claims. Kureha describes its bead-shaped BAC as generating “less carbon dust” because of the bead formation process, as having “High strength and high wear resistance” that prevents carbon contamination, and as having high flowability giving uniform distribution in complex-shaped containers. That last point matters: an awkward cartridge geometry is where a poorly flowing medium leaves voids and channels.

There is no verified numeric attrition comparison across all three forms. Treat any you are shown as a single-supplier internal result until you see the method.

What does ball pan hardness really tell you?

Ball-pan hardness, measured under ASTM D3802, tells you whether two lots of the same grade are alike. It does not tell you how the carbon will survive your vessel.

ASTM attaches a blunt limitation to the result: the method, in its own scope, “does not actually measure in-service resistance to degradation” — it establishes the comparability of lots ostensibly of the same grade. The trade convention, as summarised in Water Conditioning & Purification, is that “The highest hardness number is 100 and most specifications call for 97 to 98.” The procedure behind the number is set out in how to read an activated carbon datasheet.

The second limitation matters more when comparing suppliers. ANSI/AWWA B604 does not apply D3802 universally. It prescribes the Ro-Tap test for bituminous- and coconut-based carbons and the stirring abrasion test for lignite- and wood-based carbons. A hardness number is therefore comparable within a raw-material family and much less comparable across families. Any claim tying a D3802 number directly to deep-bed head loss or fines generation is unverified, and we will not make one.

The three Suracsh forms, on published specification

The same base chemistry in three forms. Hardness and ash are where the sphere separates from the other two — not surface area alone.

Granular — SURSORB SA. Coconut shell, steam activated, no impregnant. BET 1000–1500 m²/g, bulk density 0.500 ± 0.05 g/cc, ball pan hardness min 98, mesh cuts 8×30, 12×30, 12×20 and 7×16 (other granular grades run down to 14×35). The moisture and ash lines on the current SA sheet carry source-typo flags in our product data and are being corrected at the factory, so we are not quoting them.

Pellet — the SURSORB PAC family. Cylindrical pellets from coconut shell charcoal, no impregnant. In this range PAC denotes pelletised, not powdered — a real and frequent source of confusion, since PAC almost everywhere else means powdered activated carbon.

Those are the lines the form comparison turns on; the CTC and iodine figures for the same three grades, and what they do and do not tell you, are in what CTC activity tells you. That is one carbon at three activation levels, and it shows what activation costs: apparent density falls from 480 to 380 g/L while hardness and crushing strength give ground, 96 and 3.5 kg on PAC-1250 against 98 and 4 kg on the other two.

Spheres — SURSORB Ultra and SURSORB Neo. Both from polymeric precursors, no impregnant.

The Neo hardness figure carries a source-typo flag in our product data; treat it as provisional.

The separation is not surface area — SA reaches 1500 m²/g at the top of its range and Ultra is specified at min 1350. It is ash max 0.5 % against max 8 % on the pellets, hardness min 99, and a per-sphere compressive strength figure that granular carbon cannot meaningfully be given, because an irregular particle has no defined geometry to compress.

Where spheres actually win: NBC protective clothing

The clearest published case for spherical carbon is not a vessel. It is a garment.

Blücher GmbH’s US Patent 7,465,490 B2, Protective clothing providing NBC protection, states that spherocarbon has the decisive advantage of being “enormously abrasion-resistant and very hard,” and ties that directly to the wear properties of the finished clothing. The specification it claims is precise: bursting pressure per sphere generally not less than about 5 newtons, especially not less than about 10 newtons, up to about 20 newtons; average diameter below 1.0 mm, preferably below 0.35 mm and not below 0.1 mm; areal loading 5–500 g/m², with 50–100 g/m² preferred; surface area not less than 1000 m²/g BET, up to 1500 m²/g; and roughly 40 g/m² of moisture storable during perspiration.

The engineering argument underneath is straightforward. An impermeable NBC suit protects by excluding everything, water vapour included, so metabolic heat has nowhere to go and the tolerable wear period in hot conditions collapses. The alternative is an air-permeable fabric that passes air and perspiration while an adsorbent layer holds the agent — which only works if the adsorbent is small enough to disperse through a textile, hard enough to survive folding and abrasion for the life of the garment, and clean enough not to shed dust onto skin. Those requirements select for spheres.

Field data supports the concept. DRDO’s paper in Defence Science Journal on activated-carbon-sphere-based permeable protective clothing against sulfur mustard reports more than 24 hours of protection against HD, air permeability above 30 cm³/s/cm², adsorbent capacity of 1029 mg/g in terms of iodine, suit mass below 2.75 kg in XL, water vapour resistance below 9.6 m²Pa/W, and a thermal burden of 34 °C on the WBGT index.

One correction to a claim that circulates widely: spheres are not the norm in respirator cartridges. Chemviron describes respirator filters as using a finer mesh carbon and vehicle or collective-protection filters a coarser bed carbon, “typically in granular or pellet forms.” It does not name spherical carbon for respirators. Suracsh’s own EN 14387 respiratory grades are granular.

Suracsh Filters Pvt Ltd has published a case study on developing a commercial adsorption medium for NBC filter fabric in sphere form. What it establishes is the scale of the development work: “more than 200 lab scale trials and pilot trials”, the case study says, across more than 25 types of raw material. The same document states that Suracsh has become the single largest supplier of activated carbon spheres to the Indian defence forces — that is a company claim, and we present it as one. No independent source verifies it, and the case study publishes no figures at all. Every specification here for the spherical grades comes from the published datasheets instead.

Export control. Spherical carbon for NBC clothing, and the CBRN grades and filter assemblies in the Suracsh range generally, are export-controlled. Nothing here is a capability claim beyond the published tables. Any enquiry touching this material requires an identified end user and end use before a quotation can be issued.

Choosing between the three

Pressure drop appears once, at the top, and it is the pellet’s argument, not the sphere’s.

The full specification for every grade named here is published in HTML at suracsh.us, and every datasheet downloads without a form or an NDA. The contaminant-to-grade mapping is at /selection-guide/.

Sources

• PDH Academy course 437 — Activated Carbon Odor Control Systems
• Ergun equation, form and constants
• US EPA — Carbon Adsorbers chapter, pressure drop correlation (Eq. 1.30)
• US EPA Air Pollution Control Cost Manual — Carbon Adsorbers, 7th edition
• Knaebel — Adsorber design notes, Ergun equation and flow restriction
• US Patent 4,228,037 — Spherical activated carbon having low dusting property and high physical strength
• Kureha — Bead-shaped Activated Carbon (BAC)
• ASTM D3802 — Ball-Pan Hardness of Activated Carbon
• Gilson — ASTM D3802 ball pan hardness test set
• Water Conditioning & Purification — Testing tips for activated carbon users, part 1
• ANSI/AWWA B604-18 — Granular Activated Carbon, preview pages
• Blücher GmbH, US 7,465,490 B2 — Protective clothing providing NBC protection
• DRDO, Defence Science Journal — Chemical protection studies of activated carbon spheres based permeable protective clothing against sulfur mustard
• Chemviron — Respirator protection with activated carbon
• Böhringer et al., Polymer-based spherical activated carbons, Chemie Ingenieur Technik 2011
• Progress on fabrication and application of activated carbon sphere, J. Ind. Eng. Chem.

Questions