Cellulose Nanofibrils in Pervious Concrete: Improving Mechanical Properties and Durability

doi:10.21203/rs.3.rs-4782482/v1

Pervious concrete (PC), commonly used in urban pavement is known for its high permeability, which contributes to mitigating the urban heat island. However, the low strength and durability of PC limit its use. The objective of this study is to improve mechanical properties and durability of PC by adding cellulose nanofibrils (CNFs). The results showed that CNFs significantly enhanced compressive strength, flexural strength, and salt frost resistance, with optimal performance at 0.15% CNF. At this concentration, compressive and flexural strengths increased by 26.5% and 25.8%, respectively, despite a slight reduction in permeability. CNFs also improved resistance to salt-induced freeze-thaw cycles, reducing spalling and maintaining a higher dynamic elasticity modulus, particularly at 0.1% and 0.15% dosages. Scanning electron microscope (SEM) analyses revealed that CNFs create a denser, more uniform network of hydrated products, enhancing microstructure and interfacial bonding. This study confirms that CNFs can significantly enhance the mechanical properties and durability of PC.

pervious concrete

cellulose nanofibrils

compressive strength

flexural strength

salt frost resistance

Pervious concrete (PC), due to its unique structural characteristics, has garnered significant attention in contemporary construction and urban planning. Its primary advantages include high permeability, which effectively channels rainwater into the ground, thereby reducing surface runoff and mitigating urban flooding. Additionally, permeable concrete offers environmental benefits such as replenishing groundwater resources, enhancing urban ecological environments, and mitigating urban heat island effects. However, it also presents challenges, including lower strength compared to conventional concrete, higher maintenance costs due to potential pore blockages, and complex construction processes. Given these advantages and disadvantages, research into PC is imperative. Advancements in mix design, optimization of construction techniques, and improvements in material properties can enhance its strength and durability, thereby expanding its application scope. Moreover, investigating its long-term performance and maintenance strategies will help reduce maintenance costs and ensure sustained permeability, thus supporting urban sustainability. To address these challenges, cellulose nanofibrils (CNFs) have been introduced as a modification material for Portland cement. CNFs, a natural polymer, are abundant, biodegradable, and renewable, offering superior mechanical stability. Their elastic modulus and tensile strength can theoretically reach up to 150 GPa and 3 GPa, respectively. The incorporation of CNFs is particularly advantageous for significantly enhancing the physical and mechanical properties of concrete¹.

Barnat-Hunek et al.² revealed that concrete with a 1% CNF admixture exhibited a 34.5% increase in flexural strength and a 23.3% enhancement in compressive strength compared to the control group. Peters et al.³ incorporated CNF concentrations of 1%, 3%, and 5% into reactive powder concrete, observing a modest increase in tensile strength, while the fracture energy improved by 50% relative to the control. Additionally, experiments conducted by Machado et al.⁴ indicated that CNF benefits the early strength development of reactive powder concrete, achieving up to 90% of the 28-day compressive strength within just 7 days. Experiments were conducted using lower dosages of CNF (0.1%, 0.2%, 0.25%), discovering that the batch with 0.2% CNF exhibited a 20.7% reduction in tensile strength. Stephenson et al.⁵ incorporated CNF dosages of 0.1%, 0.5%, and 1.0% into ultra-high-performance concrete, noting that fracture energy peaked at 0.5% CNF, significantly mitigating early shrinkage without negatively impacting compressive strength. Similarly, Dai et al.⁶ embedded CNFs with diameters ranging from 20 to 100 nm into cement paste at dosages between 0.1% and 0.4%. Their results indicated that a 0.15% CNF dosage led to a 20% increase in compressive strength and a 15% improvement in flexural strength. Onuaguluchi et al.⁷ experimented with adding 0.05%, 0.1%, 0.2%, and 0.4% CNF to limestone calcined clay cement, finding that a 0.1% CNF addition resulted in the most significant enhancement in flexural strength. Claramunt et al.⁸ assessed the effects of varying CNF and CNC dosages (0.1% to 8%) on the matrices of Calcium Aluminate Cement (CAC) and Ordinary Portland Cement (OPC). Their study determined that the lower dosage groups of 0.1% and 0.2% CNF demonstrated superior performance in bending strength.

De Souza⁹investigated methods for dispersing microcrystalline cellulose and CNF within a cementitious matrix, indicating that ultrasonic disintegration significantly enhanced the dispersion of microcrystalline cellulose in the cement matrix. Mejdoub et al.¹⁰ employed the TEMPO oxidation system to pre-treat CNF, resulting in notable enhancements in the physico-mechanical properties and microstructure of the cement base. Their studies on mechanical and microstructural properties demonstrated that cement with a 0.3% CNF dosage exhibited the lowest porosity and the most substantial increase in compressive strength, a remarkable 43%. Goncalves et al.¹¹ examined the influence of CNF on cement paste subjected to sulfate attack, finding that CNF incorporation could diminish sulfate permeation and mitigate volume expansion. Hoyos et al.¹²utilized CNF to reinforce cement paste, demonstrating significant increases in the yield stress, degree of hydration, and storage modulus, effectively preventing microcracking caused by water expansion and evaporation.

In summary, while most research has focused on the short-term properties of cellulose nanofibril (CNF)-reinforced cement-based materials, there is limited investigation into their long-term durability, particularly their resistance to salt freeze. CNFs can modify rheological properties and nano-reinforce the paste, potentially improving aggregate bonding in pervious concrete (PC). This study prepared high-performance PC samples with CNF dosage ranging from 0.05% to 0.2% by weight of binder to evaluate performance improvements. The mechanical properties, water permeability, and salt freeze resistance were evaluated through laboratory test. Scanning electron microscopy (SEM) was also used to examine CNF contributions to the microstructure and interfacial bonding in the interfacial transition zone (ITZ).

2.1 Materials and Mixture Proportions

Portland cement used in the experiment was P·O 42.5 ordinary Portland cement (OPC), produced from a company in Shandong Province, with specific properties listed in Table 1. Fly ash (FA) was obtained from a power plant in Shandong Province, with main properties outlined in Table 2. The aggregate used was limestone with a particle size of 5-10 mm, and its main physical properties are listed in Table 3. The water reducer used was a third-generation high-performance polycarboxylic acid water reducer produced by a company in Jiangsu province. CNF was produced by a company in Jinan, Shandong Province, with a solid content of 3.15% and parameters listed in Table 4.

Table1. Chemical composition and properties of OPC

Chemical composition (wt.%)					Special surface area (m²/kg)	28d compressive stenghth (MPa)	Flexural stenghth (MPa)
CaO	SiO₂	Al₂O₃	Fe₂O₃	MgO	353	47.8	7.9
63.6	21.2	3.2	3.0	1.0	353	47.8	7.9

Table 2. Properties of fly ash

Mineral additive

Density

(g/cm³⁾

Water content (%)

Loss on ignition (%)

28d activity index (%)

Water demand ratio (%)

FA

2.23

0.5

4.21

73

98

Table 3. Physical properties of the aggregate

Paticle size (mm)	Apparent density (g/cm³⁾	Bulk density(g/cm³⁾	Flat and elongated particle content (%)	Crushing value (%)
5-10	2.691	1.479	9.4	12.1

Table 4. Physical and mechanical properties of CNF

Length (nm)	Diameter (nm)	Tensile strength (MPa)	Young’s modulus(GPa)
500-1000	10-50	8000-10000	100-200

For the PC prepared in this experiment, the water-cement ratio was 0.28, and the water-reducer dosage was 10%, and CNFs were added at proportions of 0.05%, 0.1%, 0.15%, 0.2% by weight of binder, as shown in Table 5.

Table 5. Mixture proportion design (kg/m³)

Group ID	OPC	CA	FA	Water	Water-reducer	CNF
CNF-0	8316	325.60	924	2577.4	9.24	0
CNF-0.05	8316	325.60	924	2405.1	9.24	4.62
CNF-0.1	8316	325.60	924	2232.7	9.24	9.24
CNF-0.15	8316	325.60	924	2060.8	9.24	13.86
CNF-0.2	8316	325.60	924	1888.4	9.24	18.48

2.2 Experimental Methods

2.2.1 Pre-dispersed method of CNF

The suspension of CNF is shown in Fig.1 (a). The pre-dispersion method of CNF is as follows: CNF and water were mixed in a ratio of 1:2 and subjected to ultrasonic treatment in a CNF-specific ultrasonic cell disruptor, with a treatment time of 15s in fast mode and 30s in slow mode. After the disruption, the CNF was placed in a magnetic stirrer and stirred at room temperature for 15 min at a stirring frequency of 1200-1500 rpm. Finally, it was dispersed using an ultrasonic disperser at room temperature for 45 min to obtain the CNF solution, as shown in Fig.1 (b).

2.2.2 Mechanical test

Compressive strength and flexural strength of PC were measured according to Chinese National Standard GB/T 50081-2019¹³. The 28 d compressive strength was measured on 150 mm×150 mm×150 mm cubes, and specimens were loaded at a rate of 0.5 MPa/s. Four-point flexural strength was measured using 100 × 100 × 400 mm prisms at 28 d at a loading rate of 0.5 MPa/s.

2.2.3 Porosity and water permeability test

The porosity test was conducted according to the standard CJJ/T 135-2009¹⁴, and static water balance method was used for estimation. The permeability coefficient was determined according to DB11/T775-2020¹⁵, using a permeability coefficient tester.

2.2.4 Salt frost resistance test

The salt frost resistance of PC was tested using the single-surface method specified in the Chinese National Standard GBT 50082-2009¹⁶, with 150 mm × 150 mm × 150 mm cubes. In the test, the central temperature of the specimen was alternatively start from 20°C, and decrease to -20°C at a rate of 10°C/h, maintaining this temperature for 3 hours. Then, it start from -20°C and increase to 20°C at a rate of 10°C/h, maintaining this temperature for 1h. Mass loss and dynamic elasticity modulus were calculated at intervals of 20 cycles.

2.2.5 SEM test

Microscopic morphology was depicted using the Gemini SEM 300 field emission scanning electron microscope produced by Carl Zeiss. The cured specimens were crushed into blocks, dried in a drying oven at 45°C for 24 hours, and then photographed.

3.1 Compressive Strength

The compressive strength of specimens in all groups are shown in Fig.2. As Fig.2 depicted, the 7 d compressive strength generally first increases and then decreases with increasing CNF content, peaking at 0.1% CNF with a maximum strength of 18 MPa, which represents a 20% increase over the control (PC without CNF). Subsequently, a slight decline is observed. However, the 0.2% CNF mix still shows a 16.7% increase in strength, reaching 17.5 MPa. Conversely, the 0.05% CNF mix exhibits a 2% decrease in 7 d compressive strength compared to the control, with a strength of 14.7 MPa.

Furthermore, Fig.2 indicates that within the 0-0.15% CNF admixture range, the 28 d compressive strength consistently shows a steady increase, achieving a peak value of 23.4 MPa at 0.15% CNF, which is 26.5% higher than that of the control. At a 0.2% CNF admixture, the 28 d compressive strength slightly declines, yet it remains 15.7% above the control.

The advantageous effects of CNF on concrete strength primarily arise from its enhancement of the cement hydration process. With its substantial specific surface area and hydroxyl groups (OH-), CNF efficiently forms hydrogen bonds with the hydrogen ions in water, boosting its hydrophilic properties. This enhanced water retention capability minimizes bleeding and inhibits sedimentation within the mix. Typically, sedimentation leads to the accumulation of cement paste at the base of the concrete, diminishing the thickness of the cement binder layer at the top and thereby reducing load-bearing capacity. A uniformly distributed cement paste facilitates effective stress transfer, enhances the encapsulation of the upper cement, and improves overall strength. The improved hydration efficiency and enhanced interface performance between paste and aggregate are attributed to the shortcut diffusion effect¹⁷. This effect posits that the inclusion of CNF creates internal pathways for moisture transport within cement particles, thus enhancing hydration efficiency and potentially improving the interface performance between paste and aggregate.

3.2 Flexural strength

Fig.3 depicts the variations in 28d flexural strength of PC when augmented with five distinct CNF dosages. Similar to its influence on compressive strength, the flexural strength generally increases with the addition of CNF, peaking before a slight regression. At a CNF dosage of 0.15%, the flexural strength reached 3.9 MPa, marking a 25.8% increase compared to the control. At dosages of 0.1% and 0.2%, the strengths recorded were 3.6 MPa and 3.5 MPa, respectively, representing improvements of 16% and 13% over the control.

Research has shown that the high aspect ratio of CNF can bridge micro-cracks in concrete, modifying their morphology and preventing their propagation. This reduces inherent defects and micro-cracks. However, When CNF content exceeds 0.15% leads to aggregation at the cement paste surface, impairing the bond between the cement paste and aggregates, thereby diminishing mechanical performance.

3.3 Permeability properties

Fig.4 illustrates the relationship between the 7d permeability coefficient and CNF dosage for various test specimens. Without CNF, the permeability coefficient is 13.4 mm/s. At a CNF dosage of 0.1%, the permeability coefficient increases to 14.33 mm/s, which is 6.9% higher than the control, indicating the most significant improvement in the permeable performance of the concrete. In contrast, a 0.05% CNF dosage results in a 13.6% reduction in the permeability coefficient compared to the control. When CNF dosages increase to 0.15% and 0.2%, the permeability coefficients decrease by 10.2% and 12.7%, respectively.

Porosity is a critical factor affecting the permeability coefficient of PC. The trend in 7d porosity closely mirrors changes in the 28d permeability coefficient. At a CNF dosage of 0.1%, the porosity of PC is 19.2%, a 2.1% increase compared to the control. However, this increase is minimal at a 0.2% dosage. For CNF dosages of 0.05% and 0.15%, porosity decreases by 17.6% and 8%, respectively.

The relationship between permeability coefficient and porosity is interconnected. CNFs enhance the viscosity of the cement paste, increasing the number and size of interconnected pores. This results in a larger effective cross-sectional area for water flow, reducing resistance and increasing flow rate and volume, thus enhancing the permeability coefficient¹⁸. At a 0.1% CNF dosage, optimal improvement is observed, with uniform encapsulation of aggregate surfaces maximizing the permeability coefficient.However, at a 0.2% dosage, an increase in porosity with a decrease in the permeability coefficient occurs. This is because porosity measurements include both interconnected and semi-interconnected pores, but only interconnected pores contribute to permeability. Higher porosity may indicate more interconnected pores, but uneven vertical distribution of cement paste can reduce the permeability coefficient. This variation may also result from challenges in consistency controlling during the molding process.

3.4 Salt frost resistance

Figs.5 demonstrate the apparent changes of various specimens before and after freeze-thaw cycles under salt frost conditions.

Before exposure to salt frost, the specimens with various CNF dosages had uniformly smooth surfaces and sharp, well-defined edges, free of visible pores or depressions. After freeze-thaw cycles, the specimens exhibited a coating of white and pale yellow substances, along with general structural loosening. The edges showed different levels of aggregate detachment and deformation, including warping and rounding.The CNF-0 group experienced the most severe edge chipping and aggregate exposure, with significant corner rounding and visible through-cracks, leading to layer-by-layer spalling damage. Each dosage group displayed distinct degrees of surface deterioration. The CNF-0.1 group showed cement paste spalling and exposed aggregates, while the CNF-0.15 group had edge chipping. The CNF-0.2 group revealed increased pore diameters at contact surfaces. The CNF-0.05 and CNF-0.2 dosage groups retained the best overall structural integrity, whereas the CNF-0.1 and CNF-0.15 groups exhibited superior aggregate uniformity.

These observations relate to the behavior of water in concrete during freezing. Soluble salts, such as sodium (Na⁺) and potassium (K⁺), in the pore solution increase the solute concentration, exacerbating structural damage due to increased expansive pressure from ice formation. This process, driven by a concentration gradient, promotes the permeation of external water into the concrete, leading to higher saturation levels.

The salt frost resistance of CNF-modified PC is evaluated based on the observed mass loss after the freeze-thaw cycles, as detailed in Fig.6.

Fig.6 shows that as the number of salt freezing cycles increases, flaking per cycle escalates for all specimens. Increasing CNF content significantly enhances resistance to freeze-thaw degradation, particularly at 0.1% and 0.15% CNF, where there is a notable reduction in mass loss compared to the control. After 120 cycles, the performance of CNF-0.2 is similar to that of CNF-0.05, indicating that beyond a certain concentration, further increasing CNF content may not improve salt frost resistance.

Initially, mass loss primarily occurs from the cement paste, while in later stages, significant degradation is observed due to aggregate detachment. The proliferation of concrete cracks further exacerbates freeze-thaw damage, accelerating aggregate separation from the cement matrix. CNF inclusion considerably reduces mass loss and enhances resistance to salt frost damage by inhibiting crack propagation and preventing aggregate separation. Additionally, CNFs may alleviate internal concrete pressures by facilitating gas permeation through their crisscrossed distribution within the matrix.

Fig.7 presents the changes in the dynamic elasticity modulus for the specimens of all groups.

The relative dynamic elasticity modulus measures elastic deformation resistance, reflecting the interactions between atoms, ions, or molecules, with higher values indicating greater stiffness and stress resistance. Increasing CNF content slows the decline in relative dynamic elasticity modulus. The CNF-0 group shows the steepest drop, reaching the lowest density after 120 cycles, while the CNF-0.1 group retains a higher density. Both CNF-0.2 and CNF-0.15 demonstrate enhanced durability with a more gradual decrease values.The decline in relative dynamic elasticity modulus is due to structural damage from freeze-thaw cycles, primarily caused by micro-cracks from freezing moisture. CNFs reinforce the internal matrix, improving connections and distribution, which helps mitigate the spread of micro-cracks and preserve material integrity and performance across cycles.

3.5 SEM test

SEM tests was used to analyze the microstructure of CNF-modified PC. The micrographs shown in Figs.8 and 9 revealed the distribution of CNF within the cementitious matrix and its effects on the pore structure of PC.

The SEM images showed that CNFs were uniformly distributed within the cementitious matrix, forming a dense network structure, which led to a more compact microstructure with reduced pore connectivity. This enhancement improved the mechanical properties and salt frost resistance of PC, as CNFs acted as reinforcing agents, bridging cracks and increasing toughness. The improved salt-freeze resistance is attributed to several factors: CNFs enhance the interfacial bonding between the cementitious matrix and aggregates, reducing the ingress of water and chloride ions. They also act as nucleating agents, promoting the formation of denser hydration products and minimizing harmful expansive products during freeze-thaw cycles. Additionally, CNFs refine the pore structure, reducing capillary pore connectivity and thus boosting resistance to salt frost damage.

The study examined the impact of CNF dosage on the mechanical properties and durability of PC. The findings revealed that incorporating CNFs significantly enhanced the compressive strength, flexural strength, permeability, and salt frost resistance of the concrete. These improvements are attributed to the refined microstructure and enhanced interfacial bonding due to CNF addition. Overall, CNFs demonstrate significant potential as a sustainable and effective additive for improving PC performance. Key conclusions from the research include:

CNF can enhance the strength of PC while maintaining good permeability. Higher CNF dosages significantly increase compressive and flexural strengths, with a moderate impact on permeability. With a 0.15% dosage, the 28 d compressive and flexural strengths are exceptional, resulting in a 10.2% reduction in permeability. The 0.1% dosage group showed a 6.9% higher permeability coefficient than the control.
CNF can significantly improve salt freeze resistance of PC. With more freeze-thaw cycles, non-CNF-incorporated concrete showed severe surface damage, while CNF reduced unit area spalling. The relative dynamic elasticity modulus was higher in the 0.1% and 0.15% CNF dosage groups compared to others.
Appropriate CNF dosages narrow micro-cracks at the matrix-aggregate interface and enhance cement-to-aggregate bonding by forming a uniform, dense network of hydrated products. This improvement in the interface region significantly boosts the mechanical properties of pervious concrete.

Qing Y, Zhiyong U, Yiqiang W. Study Progress on Cellulose Nanofibril. Scientia Silvae Sinicae. 2012;48(7):145-152.
Barnat-Hunek D, Smarzewski P. Influence of hydrophobisation on surface free energy of hybrid fiber reinforced ultra-high performance concrete. Constr Build Mater. 2016;102(Jan):367-377. https://doi.org/10.1016/j.conbuildmat.2015.11.008
Peters SJ, Rushing TS, Landis EN, et al. Nanocellulose and microcellulose fibers for concrete. Transp Res Rec. 2010;2142:25-28. https://doi.org/10.3141/2142-04
Machado F, Pedroti LG, Lemes J, et al. Addition of cellulose nanofibers in reactive powder concrete. Springer International Publishing; 2017. https://doi.org/10.1007/978-3-319-51382-9_57
Stephenson KM. Characterizing the behavior and properties of nano cellulose reinforced ultra high performance concrete. 2011. https://digitalcommons.library.umaine.edu/etd/1579
Dai H, Jiao L, Zhu Y, et al. Nanometer cellulose fiber reinforced cement-based material. CN201510549498.1; 2015.
Onuaguluchi O, Panesar DK, Sain M. Properties of nanofibre reinforced cement composites. Constr Build Mater. 2014;63(Jul):119-124. https://doi.org/10.1016/j.conbuildmat.2014.04.072
Claramunt J, Ventura H, Filho R, et al. Effect of nanocelluloses on the microstructure and mechanical performance of CAC cementitious matrices. Cem Concr Res. 2019;119:64-76. https://doi.org/10.1016/j.cemconres.2019.02.006
Souza L, Cordazzo M, Souza L, et al. Investigation of dispersion methodologies of microcrystalline and nano-fibrillated cellulose on cement pastes. Cem Concr Compos. 2022;126:104351. https://doi.org/10.1016/j.cemconcomp.2021.104351
Mejdoub R, Hammi H, Suñol JJ, et al. Nanofibrillated cellulose as nanoreinforcement in Portland cement: Thermal, mechanical and microstructural properties. J Compos Mater. 2016;50(18):2491-2503. https://doi.org/10.1177/002199831667
Goncalves J, El-Bakkari M, Boluk Y, et al. Cellulose nanofibres (CNF) for sulphate resistance in cement based systems. Cem Concr Compos. 2019;100:100-111. https://doi.org/10.1016/j.cemconcomp.2019.03.005
Hoyos CG, Zuluaga R, Ganan P, et al. Cellulose nanofibrils extracted from fique fibers as bio-based cement additive. J Clean Prod. 2019;235:1540-1548. https://doi.org/10.1016/j.jclepro.2019.06.292
GB/T 50081-2002. Standard for Test Method of Mechanical Properties on Ordinary Concrete. 2002.
CJJ/T 135-2009. Technical Specification for Pervious Cement Concrete Pavement. 2009.
DB11/T 775-2010. Technical Specification for Pervious Concrete Pavement. 2010.
GB/T 50082-2009. Standard for Test Methods of Long-term Performance and Durability of Ordinary Concrete. 2009.
Cao Y, Zavaterri P, Youngblood J, et al. The influence of cellulose nanocrystal additions on the performance of cement paste. Cem Concr Compos. 2015;56:73-83. https://doi.org/10.1016/j.cemconcomp.2014.11.008
Yang Y, Cheng J, Guo X. Discussion on the relationship between porosity and permeability coefficient of pervious concrete. Concr Cem Prod. 2007;000(004):1-3.

Cellulose Nanofibrils in Pervious Concrete: Improving Mechanical Properties and Durability

Status:

Version 1

Abstract

Figures

1 Introduction

2 Experimental Program

3 Results and analysis

4 Conclusion

References

Status:

Version 1