Codes & standards referenced
Defines five admixture families — accelerating, retarding, water-reducing, air-entraining, superplasticising — and sets performance requirements (water content, setting time, compressive strength, length change, bleeding) plus the uniformity tests (Cl. 5–6) that verify a delivered batch of the same product matches the originally-approved sample on specific gravity or dry-solids content, pH, air-entraining effect, and setting-time drift against a control. The uniformity checks are batch-to-batch checks on the admixture as delivered, not site-condition checks on the mixed concrete — they catch drift between successive consignments of the same brand.
Cited inside IS 9103's foreword as the BIS standard for integral waterproofing admixtures. Covers hydrophobic, water-reducing, air-entraining and pozzolanic waterproofing compounds — not plasticisers, superplasticisers or any retempering practice. Referenced here only to mark the boundary of what an admixture standard covers.
A test method, not an admixture product standard. Referenced here because admixture modifications to set time and early strength must be checked for their effect on reinforcement bond — a critical site-quality check that an accelerator can quietly compromise if dosed blind.
Cl. 5.2 (materials), Cl. 8.2.4.2 (permissible mineral admixtures), Cl. 5.6 (admixtures — chloride cap), and Cl. 16 (acceptance criteria). Every IS 9103 admixture has to coexist with IS 456's max w/c, min cement, and chloride-content limits.
The original task brief associated IS 2645 with "retempering" and IS 2770 with "accelerating admixtures." Both attributions are not what those standards cover. The IS 9103:1999 foreword itself clarifies that integral waterproofing admixtures are covered in IS 2645:2003; IS 2770 (Part 1):1967 is a pull-out bond test. Accelerating admixtures are evaluated under IS 9103 itself, not IS 2770. We keep the three codes in the citation set but describe each in its actual scope — to avoid passing a misleading framing forward.
Why this matters
Every other ingredient in the concrete mix is a noun — cement, sand, aggregate, water. The admixture is a verb: it acts on the cement grains. Get the chemistry wrong and the same nominal mix gives 25 MPa; get it right and the same mix gives 38 MPa at the same w/c, or 38 MPa at a lower w/c. That single decision routinely shifts the cement content of a 50 m³ pour by 500–1,500 kg, and the binder cost by ₹3,000–₹10,000.
The field confusion is real. "Plasticiser" is a generic word that covers five very different chemistries. The dose that saturates a lignosulphonate typically sits in the 0.20–0.30% solids-by-mass-of-cement range; a short-side-chain PCE saturates at 0.10–0.30% solids and a slump-retentive long-side-chain PCE at 0.30–0.50% solids — these saturation windows overlap; the saturating dose is a chemistry-by-cement match, not a fixed ratio. The dose that reaches saturation depends on the cement's C3A and SO3 content, the side-chain length, the grafting density and the mix temperature. IS 9103:1999 organises this by family and sets uniform performance criteria so that a manufacturer can't sell air-entrainer as a superplasticiser, but the standard doesn't pick the chemistry for you — and that's where projects go wrong.
w/c, w/cm, w/p — three ratios, three different denominators. The denominator matters as much as the number. w/c counts only the OPC and Portland-composite cements; w/cm adds the supplementary cementitious materials (fly ash, GGBS, silica fume) into the binder; w/p adds the full powder content including inert fillers such as limestone fines (the denominator SCC designs use). A PPC + 30% fly ash mix at 0.40 w/cm is roughly 0.52 w/c; the same mix at 0.40 w/c is roughly 0.28 w/cm and almost certainly under-hydrated. Spec writers who use w/c when they mean w/cm, or the reverse, routinely under-design or over-design the binder by 30–50 kg/m³. IS 456:2000 and IS 10262:2019 use w/c in the body of the standard; the SCM-aware w/cm is an ACI 211.1 / Indian-Practice convention that the spec should call out explicitly when SCMs are present.
A normal plasticiser (lignosulphonate) cuts water 5–12%; a PCE superplasticiser cuts it 20–35%. On the worked-example M30 mix below (400 kg/m³ cement, 0.42 w/c, mixing water 168 L/m³) that translates to ≈ 8–20 L/m³ saved by a lignosulphonate and ≈ 30–55 L/m³ saved by a PCE. The lower w/c that follows means lower permeability and higher strength at the same cement content — or lower cement content at the same strength.
PCE is the only chemistry that holds 200 mm slump for 90–120 minutes in transit, which is why every long-haul RMC truck in India runs on a PCE blend. Lignosulphonate loses slump in 30–45 minutes — fine for site-batched, fatal for transit.
PCE compatibility is sensitive to the cement's C₃A, SO₃, and alkali content. The same PCE that flows at 220 mm in one OPC may slump-collapse at 100 mm in another, or vice versa. IS 9103 §6.1.1 specifically calls for evaluation "for specific use with the materials and mix proportions to be used on the work".
Specification walkthrough
1 — IS 9103:1999: the five families and their pass marks
IS 9103:1999 (Reaffirmed 2018) defines admixtures as "a material other than water, aggregates, hydraulic cement and additives like pozzolana or slag and fibre reinforcement… added to the batch immediately before or during its mixing to modify one or more of the properties of concrete in the plastic or hardened state." The standard covers five chemical families:
- Accelerating admixtures (CaCl₂, calcium nitrate, triethanolamine blends)
- Retarding admixtures (sugars, gluconates, citric acid)
- Water-reducing admixtures (lignosulphonates, hydroxycarboxylic acids)
- Air-entraining admixtures (Vinsol resin, synthetic surfactants)
- Superplasticising admixtures (sulphated melamine formaldehyde SMF, sulphonated naphthalene formaldehyde SNF, polycarboxylate ethers PCE)
Each family must clear Table 1A performance requirements against a control mix of the same cement, aggregates and w/c. The headline numbers:
| Property vs control | Accelerator | Retarder | Water-reducer | Superplasticiser |
|---|---|---|---|---|
| Water content (max) | — | — | 95% | 80% |
| 3-day compressive strength (min) | 125% | 90% | 110% | 125% |
| 7-day compressive strength (min) | 100% | 90% | 110% | 125% |
| 28-day compressive strength (min) | 100% | 90% | 110% | 115% |
| Bleeding (max increase) | 5% | 5% | 5% | 5% |
| Length change (max increase) | 0.010% | 0.010% | 0.010% | 0.010% |
Source: IS 9103:1999 (Reaffirmed 2018) Table 1A (Cl. 4). The "—" entries mean IS 9103 doesn't set a pass/fail at that age because the family is not designed to act there (e.g. an accelerator isn't required to alter 28-day strength).
2 — Uniformity tests, the bit that protects the buyer
The most-missed part of IS 9103 is the uniformity test (Cl. 5–6). The standard requires the manufacturer to declare chloride content (Cl. 1.3) and to run a full set of qualification tests on the first consignment. Every subsequent batch has to show uniformity against that first-consignment reference — same specific gravity (liquid) or same dry-solids content (powder), same pH, same air-entraining effect, same setting-time drift against a control cement. These are batch-to-batch checks on the admixture as delivered, not site-condition checks on the mixed concrete. They tell you whether the admixture in the consignment matches the originally-approved sample — not whether the mix is correct for site conditions. If a PCE brand you've used for 18 months suddenly starts retarding the mix by an hour, the uniformity data should have caught it before the cube results did; if the cube results are the first signal, the uniformity file was not being reviewed.
3 — IS 2645:2003: the waterproofing boundary
IS 9103 itself excludes waterproofing admixtures and points the reader to IS 2645:2003 (Specification for Integral Cement Waterproofing Compounds). IS 2645 covers four waterproofing families — water-reducing/plasticising, air-entraining, pozzolanic, and hydrophobic — and prescribes permeability tests on mortar and concrete. It is not a standard for retempering, nor for slump-restoration at the pour site. The two are sometimes confused because some waterproofing admixtures in the market are sold as "plasticisers" — a labelling practice IS 2645 warns against, since the performance criteria are different. When you need to restore a lost slump, the standard solution is a PCE dose at the slump-loss window (≤30 min before discharge), not a waterproofing admixture.
4 — IS 2770 (Part 1):1967: bond, not admixture
IS 2770 (Part 1):1967 is the pull-out test for the bond between reinforcement and concrete. It belongs in the admixture chapter because set-accelerating admixtures change the bond envelope — early strength goes up, but the 7- to 28-day bond development can be weaker than an unaccelerated control. Any site using calcium-chloride-based accelerators with heavily-reinforced sections should run a pull-out test per IS 2770 alongside the cube tests. The standard itself does not classify or evaluate admixtures; it only provides the bond test method.
How to apply
1 — Read the dosage-vs-slump curve, not the data sheet
Every admixture has a saturation point — a dose beyond which the chemistry has adsorbed onto every available cement grain and further dose adds nothing but cost. For lignosulphonate plasticiser the curve flattens at about 0.30% solids by mass of cement, and the maximum achievable slump gain is roughly 80 mm above the control. For PCE superplasticiser the curve climbs past 200 mm and the saturation plateau arrives much later, typically at 0.20–0.30% solids for short-side-chain PCEs and 0.30–0.50% for long-side-chain, slump-retentive PCEs. Dose beyond saturation and the mix will segregate — the admixture is repelling cement grains so effectively that the paste viscosity drops below what the aggregate skeleton needs to stay suspended.
2 — Match the chemistry to the cement
All five IS 9103 families work predictably. PCE is the modern default for M30+ structural concrete. C₃A > 8% OPCs (typical 53-grade) consume more PCE — the early C₃A hydration preferentially adsorbs the polymer, leaving less to disperse C₃S. Expect 10–15% higher dose than for 43-grade.
Fly ash in the blend adsorbs the PCE — usually 15–25% higher PCE dose than OPC at the same workability. The carbon content of the fly ash (LOI) matters: a high-LOI fly ash (> 3%) pulls even more PCE out of solution. Always pre-test the actual cement + fly ash combination.
GGBS-blend cements are the easiest for PCE — the smooth, glassy GGBS surface adsorbs less polymer than fly ash or high-C₃A clinker. PCE doses run 10–20% below equivalent OPC mixes, and slump retention is generally better. Watch the set retardation on high-GGBS PSC — initial set can be 1–2 hours later than OPC, which is sometimes a feature, sometimes a defect.
Sulphate-Resistant Cement (SRC, IS 12330) has low C₃A — PCE dose drops further. Be careful with the lignosulphonate plasticisers here: the sulphonate groups compete with the cement's own sulphate for adsorption sites, and an overdose can shift set time by 2–3 hours.
3 — Cost per m³ worked example (M30, transit-mix)
Take a 400 kg/m³ cement, 0.42 w/c, 20 mm crushed-aggregate, 100 mm target slump, 45-minute transit time. Two options:
| Option | Admixture | Dose | Water reduction | Net admixture cost (₹/m³) |
|---|---|---|---|---|
| A | Lignosulphonate plasticiser | 0.30% by mass of cement | ~8% | ~₹25 (1.2 L @ ₹20/L) |
| B | PCE superplasticiser | 0.20% by mass of cement | ~28% | ~₹80 (0.8 L @ ₹100/L) |
| C | PCE + 0.05% accelerator (winter) | 0.20% + 0.05% | ~28% | ~₹100 |
Option A is cheapest but limits the slump gain to ~80 mm and doesn't survive 45 min in transit without retempering. Option B costs ₹55 more per m³ but saves ≈ 47 L of water (28% of the 168 L mixing water) — this 28% water reduction converts to a cement-content saving only when the design is re-targeted (e.g. dropping from 400 to ~360 kg/m³ while still hitting M30, worth ~₹280/m³ in cement cost). Option B also holds slump for 90+ minutes. The PCE option is a net saving of ~₹225/m³. On a 10,000 m³ project, that is ₹22.5 lakh — before counting the better surface finish, fewer blowholes, and lower permeability.
Indicative retail prices, July 2026, Pune. Manufacturer-specific prices vary ±30%; the cement saving figure assumes the 28% water reduction maps 1:1 to cement reduction at fixed w/c. Always run a project-specific trial mix per IS 10262 before fixing the design.
What goes wrong
Four failure modes account for the majority of admixture-related site incidents. All are avoidable; all are expensive to discover after the pour.
Above the saturation dose, the mix doesn't get "more workable" — it segregates. Coarse aggregate rolls to the bottom, paste rises to the top, and the result is a surface that looks fine but has honeycombed aggregate skeletons in the bulk. The temptation to add "just a little more" to chase a slump number is the most common single cause of segregation at the boom-end of the pump.
"Driver added two buckets of water at site" is the most expensive sentence in Indian QA/QC. Every 10 L of added water at site raises w/c by 0.025 and drops 28-day strength by roughly 5 MPa at M30. The correct action on a stiff load is to re-dose a small quantity of the same admixture, not to add water. IS 456 Cl. 9.2 prohibits adding water once the mix has left the plant unless the re-mix is fully re-tested.
A PCE that works flawlessly with one OPC can slump-collapse with a different brand or even a different lot of the same brand — usually a change in C₃A, SO₃, alkali or limestone content. The first three loads of any new cement + PCE combination should be treated as a compatibility trial, not as production. IS 9103 §6.1.1 explicitly requires evaluation "for specific use with the materials and mix proportions to be used on the work."
Calcium-chloride-based accelerators are cheap and effective, but the chloride cap from IS 456 Cl. 5.6 (0.4% by mass of cement for plain concrete, 0.2% for RCC) is real. Exceed it and the reinforcement corrosion that follows is invisible for 2–3 years, then catastrophic. Triethanolamine or calcium-nitrate accelerators avoid the chloride issue but are 3–5× the cost; the right choice depends on cover, exposure, and project lifetime.
Field practitioner blocks
The blocks below translate the IS 9103:1999 performance and uniformity tests, and the dosage-vs-slump curve above, into what an engineer or QC supervisor actually does on the day of the pour: questions to answer before approving the admixture, items to check on the delivery ticket, and three worked-out failure scenarios with technically qualified outcomes.
Eight questions to answer before approving the admixture
- Which IS 9103:1999 family is this product (water-reducing, superplasticising, retarding, accelerating, air-entraining), and does the delivery challan match the originally-approved brand and grade?
- Does the manufacturer hold a valid IS 9103:1999 test certificate covering the performance requirements (Table 1A, 1B) and the uniformity tests (Cl. 5–6)? When was the last certificate issued?
- What is the chloride content of this consignment (IS 9103 Cl. 1.3), and is it below the IS 456 Cl. 5.6 cap (0.2% by mass of cement for RCC, 0.4% for plain concrete)?
- Has a PCE–cement compatibility trial been run for this specific cement + admixture combination on this project? IS 9103 Cl. 6.1.1 requires evaluation "for specific use with the materials and mix proportions to be used on the work."
- What is the dosage range (PCE typically 0.6–1.2% by binder mass for SCC, 0.2–0.5% for CVC; lignosulphonate 0.2–0.3% solids), and where on the dosage-vs-slump curve is the trial mix?
- What is the expected water reduction (lignosulphonate 5–12%, PCE 20–35%) and the expected 28-day strength gain over the control (≥ 110% for water-reducer, ≥ 115% for superplasticiser)?
- What is the slump-retention window of this admixture at the expected mix temperature (PCE short-side-chain: 30–60 min; long-side-chain slump-retentive: 90–120 min)?
- What is the maximum permitted dosage at site without re-trial, and what is the escalation path if the target slump is missed (PCE re-dose at the slump-loss window, never water)?
Pre-batch and per-delivery checklist
- IS 9103:1999 certificate on file — current, valid for the consignment, covering Table 1A performance and Cl. 5–6 uniformity tests.
- Manufacturer's batch COA — specific gravity or dry-solids content within ± the tolerance stated on the data sheet; pH within range; air-entraining effect on the control mix matches the original sample.
- Chloride content on the COA — must be below 0.2% by mass of cement for RCC.
- Storage condition — liquid admixtures stored above 5°C (PCE freezes and re-disperses unevenly); tank agitated for ≥ 10 minutes before dosing; stock rotation first-in-first-out.
- Compatibility trial signed off — first three loads of any new cement + PCE combination treated as a compatibility trial, not production; record the actual dose needed to hit the target slump-flow.
- Dosage log — actual dose in mL or kg per batch, against the target dose; deviations > ± 5% trigger a flag for review.
- Site retempering rule — no water added to the truck at site, ever (IS 456 Cl. 9.2). Lost slump is recovered with a small PCE dose at the slump-loss window, with the re-mixed concrete re-tested before placement.
- Admix-cement incompatibility watch — first sign of a sudden change in water demand, setting time, or air content on a previously-stable mix → check uniformity data on the new consignment before continuing.
Three worked scenarios with technically qualified outcomes
The temptation is to add water at the chute. Do not. IS 456 Cl. 9.2 prohibits adding water once the mix has left the plant unless the re-mix is fully re-tested. On a 400 kg/m³ cement, 0.42 w/c M30 mix, every 10 L of site water raises w/c by ≈ 0.025 and drops 28-day strength by ≈ 5 MPa. Qualified outcome: the correct response is to add a small PCE dose at the slump-loss window (≤ 30 min before discharge), re-mix the drum at full speed for ≥ 90 s, and re-test the slump. If the slump cannot be recovered within the permitted re-dose limit, reject the load — do not "fix" it with water. The cost of rejecting one truck is one cube set; the cost of accepting the water-retempered pour is a strength-investigation on the whole element.
Past the knee of the dosage-vs-slump curve, additional PCE does not buy more workability — it buys segregation, bleed, and a loss of strength equivalent to retempering with water. Qualified outcome: stop. Re-check the dosage-vs-slump curve for this specific cement + PCE combination (C3A and SO3 shifts change the saturation knee). If the mix is past the knee, the correct response is to dilute back with the next batch (a partial-load interleave) and re-dose at the correct knee, not to add more PCE. A VSI 2–3 reading at the spread is the diagnostic: severe bleed halo, aggregate pile-up at the centre, paste migration to the edges.
The IS 9103:1999 uniformity tests (Cl. 5–6) are designed to catch this before it reaches site. A drift in setting time on the same brand signals one of three things: (a) the manufacturer changed the formulation without notification, (b) the new consignment is from a different production batch with drift outside the uniformity tolerance, or (c) the cement source has changed. Qualified outcome: hold the pour; request the manufacturer's batch uniformity data (specific gravity, pH, setting-time drift against the control cement) for the consignment in use; cross-check the cement source against the trial-mix approval. Do not accept the load until the drift is explained in writing and either confirmed within tolerance or replaced with a verified consignment. A PCE that drifts outside uniformity is a non-conforming product, not a site problem.
References & further reading
- IS 9103:1999 (Reaffirmed 2018) — Specification for Concrete Admixtures. Bureau of Indian Standards. Defines admixture families, performance requirements (Table 1A, 1B), uniformity tests (Cl. 5–6).
- IS 2645:2003 (reaff 2014) — Specification for Integral Cement Waterproofing Compounds. Bureau of Indian Standards. Waterproofing-admixture scope only; cited here to mark the boundary with IS 9103.
- IS 2770 (Part 1):1967 — Methods of Testing Bond in Reinforced Concrete — Part 1: Pull-Out Test. Bureau of Indian Standards. Bond test method; used to verify admixture effect on reinforcement bond.
- IS 456:2000 (Amend. 6:2024) — Plain and Reinforced Concrete — Code of Practice (Fourth Revision). Bureau of Indian Standards. Cl. 5.2 materials, Cl. 5.6 admixture chloride cap, Cl. 8.2.4.2 mineral admixtures, Cl. 9.2 placing & retempering, Cl. 16 acceptance.
- IS 10262:2019 — Concrete Mix Proportioning — Guidelines (Second Revision). Bureau of Indian Standards. Mix-design context for admixture use.
- ACI 212.3R-16 — Report on Chemical Admixtures for Concrete. American Concrete Institute. Cross-reference for PCE chemistry, dosage ranges, and compatibility discussion.
- ASTM C494 / C494M-19 — Standard Specification for Chemical Admixtures for Concrete. ASTM International. The classification system (Type A, B, C, D, E, F, G) that IS 9103 was developed to align with.
Year notes: IS 9103:1999 confirmed as the current edition (periodically reaffirmed, most recently documented in 2018); the site's earlier "2018 edition" framing was a citation error, corrected here. IS 2645:2003 (reaff 2014) confirmed from BIS public records. IS 2770 (Part 1):1967 — title verified; reaffirmation year not independently confirmed in this draft (flagged for verification against the BIS catalogue before tendering).
Frequently Asked Questions
Related articles
IS 10262 Mix Design — Step-by-Step (M30)
The five-step mix-design procedure — target strength, water content, aggregate ratio, admixture dose, and trial mix — with worked M30 numbers.
Water-Cement Ratio Explained
Why a 0.42 w/c at the plant can become 0.55 by the time the truck arrives, and what Abrams' law really says about strength.
Acceptance Criteria — IS 456 Table 11
The fck + 0.825σ / fck + 3 MPa rules, the 4-sample and 1-sample exceptions, and what to do when a single cube fails.
Amit Haridas
Founder & Proprietor, ConcreteInfo. 25+ years of experience in concrete technology, RMC plant operations, construction quality, consulting and technical training across India.