Blog

Dispersion performance of EPEG polycarboxylate superplasticizer using response surface methodology

Polycarboxylate superplasticizers are indispensable concrete admixtures that enable high workability, low water‑to‑cement ratios, and improved durability. Among the various PCE types, those based on EPEG macromonomers have gained attention due to their unique molecular structure and excellent dispersion properties. However, the performance of EPEG‑based PCEs depends heavily on the synthesis recipe – multiple factors, such as acid‑ether ratio, temperature, chain transfer agent dosage, and functional monomer dosage, interact in complex ways.

Why use EPEG and Why Response Surface Methodology

EPEG (ethylene‑glycol mono‑vinyl polyoxyethylene glycol ether) is a polyether macromonomer with a reactive vinyl end. Compared to traditional MPEG or TPEG, EPEG offers good reactivity and allows the incorporation of functional groups. In this study, the researchers used:

EPEG 3000
Acrylic acid (AA) – provides carboxyl groups for adsorption
Hydrogen peroxide‑ferrous sulfate (H₂O₂‑FeSO₄) / ascorbyl glucoside – redox initiator system
Mercaptoethanol (ME) – chain transfer agent
A phosphorus‑containing functional monomer (FX‑513) introduces phosphate ester groups for enhanced adsorption and dispersion

Conventional single‑factor optimization cannot capture the interactions between variables. Response surface methodology (RSM) allows simultaneous evaluation of multiple factors, their interactions, and quadratic effects, leading to a truly optimal synthesis condition.

Experimental Design:Dispersion performance of EPEG polycarboxylate superplasticizer using RSM

Single‑Factor Pre‑tests

Before RSM, single‑factor experiments identified the rough range for each variable:

Acid‑ether ratio (AA : EPEG molar ratio): 3.80–4.75 – optimal near 4.51
Reaction temperature: 15–35 °C – optimal near 20 °C
Chain transfer agent (ME) dosage: 0.32–0.56% (on EPEG mass) – optimal near 0.44%
Reducer (ascorbyl glucoside) dosage: 0.08–0.14% – further optimized via RSM
Phosphorus functional monomer dosage: 0.39–0.72% – further optimized via RSM

Box‑Behnken Design (RSM)

Five independent variables (A–E) were studied at three levels each:

Factor	Variable	Low (−1)	Medium (0)	High (+1)
A	Acid‑ether ratio	4.28	4.51	4.75
B	Temperature ( °C)	15	20	25
C	Phosphorus monomer dosage (%)	0.39	0.56	0.72
D	Reducer (ascorbyl glucoside) dosage (%)	0.08	0.11	0.14
E	Chain transfer agent (ME) dosage (%)	0.32	0.44	0.56

Responses: initial cement paste flow (mm) and 1‑hour flow (mm) – measured at water/cement = 0.29 and PCE solid content 0.23% by cement mass.

A total of 46 runs (40 design points + 6 replicates) were performed.

Optimal Synthesis Conditions

Using the desirability function in Design‑Expert, the optimum parameters for maximizing both initial and 1‑hour flow were determined:

Parameter	Optimal value
Acid‑ether ratio (AA:EPEG)	4.41
Reaction temperature	19 °C
Reducer (ascorbyl glucoside)	0.10% (on EPEG mass)
Phosphorus functional monomer (FX‑513)	0.50% (on EPEG mass)
Chain transfer agent (ME)	0.41% (on EPEG mass)

Experimental Validation

Three replicate syntheses under the optimal conditions gave:

Average initial flow: 221 mm (predicted 230 mm, error 3.91%)
Average 1‑hour flow: 275 mm (predicted 266 mm, error 3.38%)

Both errors are <5%, confirming the model’s predictive accuracy.

Concrete Performance Validation

A C30 concrete mix (cement 320 kg/m³, sand 820 kg/m³, aggregates 1015 kg/m³, water 165 kg/m³, PCE dosage 6 kg/m³) was used to compare the optimized EPEG‑PCE with a commercial reference.

Sample	Slump (mm) initial / 2h	Spread (mm) initial / 2h	Compressive strength (MPa) 3d / 7d / 28d
Commercial	220/200	540/480	27.3 / 35.8 / 42.2
PCE‑1 (optimized)	225/205	570/490	27.5 / 36.4 / 42.3
PCE‑2	220/210	575/490	27.6 / 36.4 / 42.1
PCE‑3	220/205	570/495	27.1 / 36.2 / 42.5

The optimized EPEG‑PCE showed improved initial spread (30–35 mm higher) and excellent 2‑hour retention, with no negative impact on strength.

Structural Characterization

FT‑IR Spectrum

Key absorption bands confirmed the expected functional groups:

3431 cm⁻¹ – O–H stretching (hydroxyl groups)
1728 cm⁻¹ – C=O stretching (ester / carboxyl)
1281, 1240 cm⁻¹ – C–O–C or –COO asymmetric stretching (ether/ester bonds)

The presence of hydroxyl, carboxyl, and ester groups matches the designed molecular structure.

Gel Permeation Chromatography (GPC)

Parameter	Value
Number‑average molecular mass (Mn)	27,281 g/mol
Weight‑average molecular mass (Mw)	46,242 g/mol
Peak molecular mass	38,387 g/mol
Polydispersity index (PDI)	1.71 (narrow distribution)
Monomer conversion	84.33%
Retention time	16.533 min

The relatively narrow PDI and high conversion indicate a well‑controlled polymerization.

Discussion: Why the Optimized Recipe Works

Acid‑ether ratio 4.41 provides sufficient carboxyl groups for strong adsorption onto cement particles, but avoids excessive side‑chain crowding that would reduce steric hindrance.
Low temperature (19 °C) balances radical generation and chain propagation, preventing unwanted side reactions (e.g., crosslinking or self‑polymerization of acrylic acid).
The phosphorus-functional monomer (0.50%) introduces phosphate groups with a high affinity for calcium ions on cement surfaces, thereby enhancing anchoring and early‑age dispersion.
The chain transfer agent (0.41%) controls molecular weight within an optimal range (Mw ~46 kDa), ensuring both good adsorption and sufficient side‑chain length for steric stabilization.
A reduced dosage (0.10%) of H₂O₂‑FeSO₄ forms an efficient redox pair that works even at ambient temperature, eliminating the need for heating.

Conclusion

A five‑factor, three‑level Box‑Behnken design successfully modeled the dispersing performance of an EPEG‑based polycarboxylate superplasticizer containing a phosphorus‑functional monomer.
The quadratic regression models for initial and 1‑hour cement paste flow were statistically significant, with good fit (R² ≈0.72–0.79) and adequate precision (signal‑to‑noise >7.9).
Significant interactions were identified, especially acid‑ether ratio × reducer, temperature × reducer, and chain transfer agent × phosphorus monomer.
The optimized EPEG‑PCE demonstrated better initial spread (570 mm) and 2‑hour retention (495 mm) than a commercial reference, with comparable strength development.
FT‑IR confirmed the presence of hydroxyl, carboxyl, and ester groups; GPC showed a narrow molecular weight distribution (PDI 1.71) and high monomer conversion (84.3%).