Microstructural Properties of Cement Paste and Mortar Modified by Low Cost Nanoplatelets Sourced from Natural Materials

Nanomaterials have been widely used in cement-based materials. Graphene has excellent properties for improving the durability of cement-based materials. Given its high production budget, it has limited its wide potential for application in the field of engineering. Hence, it is very meaningful to obtain low cost nanoplatelets from natural materials that can replace graphene nanoplatelets (GNPs) The purpose of this paper is to improve the resistance to chloride ion penetration by optimizing the pore structure of cement-based materials, and another point is to reduce investment costs. The results illustrated that low cost CaCO3 nanoplatelets (CCNPs) were successfully obtained under alkali treatment of seashell powder, and the chloride ion permeability of cement-based materials significantly decreased by 15.7% compared to that of the control samples when CCNPs were incorporated. Furthermore, the compressive strength of cement pastes at the age of 28 days increased by 37.9% than that of the plain sample. Improvement of performance of cement-based materials can be partly attributed to the refinement of the pore structure. In addition, AFM was employed to characterize the nanoplatelet thickness of CCNPs and the pore structures of the cement-based composites were analyzed by MIP, respectively. CCNPs composite cement best performance could lay the foundation for further study of the durability of cement-based materials and the application of decontaminated seashells.


Introduction
The progress of nanotechnology has brought new insights to improve the properties of cement-based materials with the aid of nanomaterials in recent years [1][2][3].
The durability exists an important performance of cement-based material. In the field of marine engineering, there are many factors that affect the durability of cement-based materials, including alkali-aggregate reaction, acid erosion, alkali erosion, and chloride ion penetration etc. Yet, the penetration of chloride ions will cause serious and irreversible damage to cement-based materials. Therefore, it is of great significance to improve the impermeability of cement-based materials.
Cement mortar was mixed with cement to sand at mass ratio of 1:3. The water to cement ratio of 0.5. 0.01, 0.02, 0.03, 0.04, 0.07, and 0.1% CCNPs by weight of cement were used in this study.

Test Methods
The chemical composition of mussel seashells powder was determined by X-ray fluorescence spectrometry (XRF, ARL ADVANT'XP). In addition, the particle size distribution of the mussel shell powder was measured by using a laser particle size analyzer (Malvern, Mastersizer 2000). Rigaku D/Max-2500 XRD, Cu-Kα (1.541874 Å) was used to investigate the phase structure of the prepared samples.
The physical properties of CCNPs were tested by the following characterization: particle morphology disclosed through a scanning electron microscope (SEM, JEOL JSM-5900); and the thickness of CCNPs were considered by atomic force microscope (AFM, Veeco Autoprobe CP Research).

Mechanical Test
The compressive strength of cement pastes was examined at various ages of 3, 7, 28 and 56 days using 2 cm cube specimens. A mean value of six paste cubes was used for the compressive strength test. The compressive strength of cement mortar was examined at various ages of 3, 7, 28 days by testing 4 cm cube specimens. A mean value of six mortar cubes was used for the compressive strength test.

Accelerated Chloride Penetration Test
NT build 492 [44] (RCM method) was applied to test the chloride diffusion coefficient of cement mortar samples as shown in Figure 1. The principle is used by the effect of an applied electric field to move the chlorine ions outside the specimen to the inside of the specimen. The RCM method is an effective method to evaluate the resistance against chloride ion permeability of cement-based materials. The test was completed on the basis of establishing migration cell. In this study, specimens were obtained from Φ110 × 220 mm cement mortar cylinders. The migration cell is full of 0.1 M/L NaOH (13.33 g L −1 ), and 10 wt.% NaCl solution. Besides, because of the apparent diffusion coefficient of chloride ion in cement-based materials, the other half of the specimen was spurted with AgNO3 solution to disclose the chloride penetration depth, which can be quickly and accurately measured by silver nitrate colorimetric method.
The calculation of the chloride penetration value from migration test was obtained with the simplified Equation (1) [41]  In this study, specimens were obtained from Φ110 × 220 mm cement mortar cylinders. The migration cell is full of 0.1 M/L NaOH (13.33 g L −1 ), and 10 wt % NaCl solution. Besides, because of the apparent diffusion coefficient of chloride ion in cement-based materials, the other half of the specimen was spurted with AgNO 3 solution to disclose the chloride penetration depth, which can be quickly and accurately measured by silver nitrate colorimetric method.
The calculation of the chloride penetration value from migration test was obtained with the simplified Equation (1) [41] where D RCM is the chloride ion transfer coefficient of unsteady chlorine ion, accurately to 0.1 × 10 −2 m 2 /s. U is the absolute value of the voltage, units by volt (V). L is the thickness of the specimen, accurately to 0.1 mm. T is the mean value of the initial temperature and the end temperature of the anodic solution. X d is the mean value of the penetration depth of the chloride ion, accurately to 0.1 mm. The half of the specimen was measured for 10 values. T is the duration time of the test.

Properties of CCNPs
It can be seen from the Figure 2 that the median particle size of mussel shell powder is 15.9 µm. The particle size of mussel shell powder is between 0.5-200 µm.
where DRCM is the chloride ion transfer coefficient of unsteady chlorine ion, accurately to 0.1 × 10 −2 m 2 /s. U is the absolute value of the voltage, units by volt (V). L is the thickness of the specimen, accurately to 0.1 mm. T is the mean value of the initial temperature and the end temperature of the anodic solution. Xd is the mean value of the penetration depth of the chloride ion, accurately to 0.1 mm. The half of the specimen was measured for 10 values. T is the duration time of the test.

Properties of CCNPs
It can be seen from the Figure 2 that the median particle size of mussel shell powder is 15.9 μm. The particle size of mussel shell powder is between 0.5-200 μm. The chemical composition of mussel seashells was given in Table 1. It can be noted that mussel shells are absolutely made up of inorganic CaCO3 phases (approximately 96.5%) and minor mineral composition. The mineral phase of calcium carbonate is demonstrated to be calcite or aragonite [39,41,43]. In addition, shells also contain small amounts of organic components, which are closely related to the growth of the shells. In mussel shells, the CaO composition is found to be 54.03% with a high loss on ignition (LOI) value of 44.48%. The LOI is owing to the decomposition of CaCO3.  The chemical composition of mussel seashells was given in Table 1. It can be noted that mussel shells are absolutely made up of inorganic CaCO 3 phases (approximately 96.5%) and minor mineral composition. The mineral phase of calcium carbonate is demonstrated to be calcite or aragonite [39,41,43]. In addition, shells also contain small amounts of organic components, which are closely related to the growth of the shells. In mussel shells, the CaO composition is found to be 54.03% with a high loss on ignition (LOI) value of 44.48%. The LOI is owing to the decomposition of CaCO 3 . XRD was used to investigate the phase structure of the prepared samples. Figure 3 shows the XRD patterns of the CCNPs and mussel shells powder, showing the diffraction peaks of the prepared CCNPs and mussel powder, corresponding to the (111), (021), (002), (012), (200), (031), (112), (130), (211), (122), (221), (041), (132), (113), and (231) planes, indicating that the pure CNNPs were prepared. However, it is worth noting that the peak strength of CCNPS is stronger than that of mussel powder, and CCNPS can be stripped from the original mussel powder in all directions. XRD was used to investigate the phase structure of the prepared samples. Figure 3 shows the XRD patterns of the CCNPs and mussel shells powder, showing the diffraction peaks of the prepared CCNPs and mussel powder, corresponding to the (111), (021), (002), (012), (200), (031), (112), (130), (211), (122), (221), (041), (132), (113), and (231) planes, indicating that the pure CNNPs were prepared. However, it is worth noting that the peak strength of CCNPS is stronger than that of mussel powder, and CCNPS can be stripped from the original mussel powder in all directions. Mussel shells have traditional 'brick and mortar' arrangement on the fractured surfaces of the mussel shells [39], which is an organic whole consisting of calcium carbonate aragonite tablets and organic matter. Hence, it serves as a basis for the experimental process in which mussel shells can be processed into platelets. Figure 4 shows the micro morphology of mussel shells powder and CCNPs. Referring to Figure 4a, main size of pristine mussel shells power is about 10 μm. Figure 4b and Figure  4c shows the morphology of CCNPs obtained after treatment of quail shell powder with 5% NaOH solution under water bath treatment conditions. According to Figure 4b, under alkaline conditions, 5% sodium hydroxide solution has a good deproteinization effect on the mussel shells powder [45]. Figure 4c presents microscopic morphology of CCNPs with deciduous shape, and the small unit of CCNPs is well dispersed. As shown in Figure 4b, CCNPs still have large morphological difference but relatively uniform size of 1 μm. It can be observed that the size of CCNPs is obviously smaller than that of mussel shell powder, but their thicknesses cannot be precisely observed from the SEM images.  Mussel shells have traditional 'brick and mortar' arrangement on the fractured surfaces of the mussel shells [39], which is an organic whole consisting of calcium carbonate aragonite tablets and organic matter. Hence, it serves as a basis for the experimental process in which mussel shells can be processed into platelets. Figure 4 shows the micro morphology of mussel shells powder and CCNPs. Referring to Figure 4a, main size of pristine mussel shells power is about 10 µm. Figure 4b,c shows the morphology of CCNPs obtained after treatment of quail shell powder with 5% NaOH solution under water bath treatment conditions. According to Figure 4b, under alkaline conditions, 5% sodium hydroxide solution has a good deproteinization effect on the mussel shells powder [45]. Figure 4c presents microscopic morphology of CCNPs with deciduous shape, and the small unit of CCNPs is well dispersed. As shown in Figure 4b, CCNPs still have large morphological difference but relatively uniform size of 1 µm. It can be observed that the size of CCNPs is obviously smaller than that of mussel shell powder, but their thicknesses cannot be precisely observed from the SEM images. XRD was used to investigate the phase structure of the prepared samples. Figure 3 shows the XRD patterns of the CCNPs and mussel shells powder, showing the diffraction peaks of the prepared CCNPs and mussel powder, corresponding to the (111) (231) planes, indicating that the pure CNNPs were prepared. However, it is worth noting that the peak strength of CCNPS is stronger than that of mussel powder, and CCNPS can be stripped from the original mussel powder in all directions. Mussel shells have traditional 'brick and mortar' arrangement on the fractured surfaces of the mussel shells [39], which is an organic whole consisting of calcium carbonate aragonite tablets and organic matter. Hence, it serves as a basis for the experimental process in which mussel shells can be processed into platelets. Figure 4 shows the micro morphology of mussel shells powder and CCNPs. Referring to Figure 4a, main size of pristine mussel shells power is about 10 μm. Figure 4b and Figure  4c shows the morphology of CCNPs obtained after treatment of quail shell powder with 5% NaOH solution under water bath treatment conditions. According to Figure 4b, under alkaline conditions, 5% sodium hydroxide solution has a good deproteinization effect on the mussel shells powder [45]. Figure 4c presents microscopic morphology of CCNPs with deciduous shape, and the small unit of CCNPs is well dispersed. As shown in Figure 4b, CCNPs still have large morphological difference but relatively uniform size of 1 μm. It can be observed that the size of CCNPs is obviously smaller than that of mussel shell powder, but their thicknesses cannot be precisely observed from the SEM images.  High magnification AFM was utilized to analyses the thickness and the finer structure of CCNPs, which was presented in Figure 5. Compared with the mussel shell powder, the size of CCNPs is smaller and the thickness is obviously reduced. As can be seen from Figure 5a, show the flake shape morphology of CCNPs, which is consistent with the results in Figure 4c. At the same time, Figure 5b illustrates the height of CCNPs (namely its thickness) ranges from 1 nm to 4 nm. Compared with the particle size of mussel shell powder in Figure 2, the particle size of CCNPs decreased from about 16 to 1 µm. Combined with SEM and AFM analysis, the result obviously reveals that CCNPs could be successfully prepared by using alkali treatment assisted with ultrasonic.
Materials 2018, 11, x FOR PEER REVIEW 6 of 11 High magnification AFM was utilized to analyses the thickness and the finer structure of CCNPs, which was presented in Figure 5. Compared with the mussel shell powder, the size of CCNPs is smaller and the thickness is obviously reduced. As can be seen from Figure 5a, show the flake shape morphology of CCNPs, which is consistent with the results in Figure 4c. At the same time, Figure 5b illustrates the height of CCNPs (namely its thickness) ranges from 1 nm to 4 nm. Compared with the particle size of mussel shell powder in Figure 2, the particle size of CCNPs decreased from about 16 to 1 μm. Combined with SEM and AFM analysis, the result obviously reveals that CCNPs could be successfully prepared by using alkali treatment assisted with ultrasonic.

Pore Size Distribution
The porosity and pore distribution are two main factors affecting the durability and strength of cement-based materials. Cementitious materials with lower porosity and finer pores show better durability and higher strength. The effect of the incorporation of CCNPs on the porosity and pore size distribution was determined by MIP is discussed below. Figure 6 illustrates the pore size distribution of CCNPs composite cement pastes. Based on the MIP test, after 28 days curing, Table 2 shows the total porosity of 0, CCNPs-0.01, CCNPs-0.02, and CCNPs-0.04 when CCNPs account for 0, 0.01, 0.02, and 0.04% of the weight of the cement. According to Table 2, the results illustrate the total porosities are 15.59, 15.42, 15.51, and 15.58%, correspondingly. Apparently, there are no significant changes in the total porosity, but pore size distribution is affected a lot by the incorporation of CCNPs. For CCNPs-0.01, the harmless pore (<20 nm) increased from 18.54% to 22.97%, and the less harm pore (20-50 nm) increased from 37.91% to 47.52%. In addition, the performance displays that it could optimize the pore structure of cementbased materials with a little amount of CCNPs in the cement mix, which could be attributed to the flake structure of CCNPs (refer to Figure 4c). CCNPs are nanometer scale, with the thickness of several nanometers, which easily generate sheet wrinkle and interlock with some more sheets. Moreover, CCNPs are hydrophilic material, it could be believed that nanoscale CCNPs can lead to the ordinary influence by optimization of pore structure in cement-based materials [2][3][4].

Pore Size Distribution
The porosity and pore distribution are two main factors affecting the durability and strength of cement-based materials. Cementitious materials with lower porosity and finer pores show better durability and higher strength. The effect of the incorporation of CCNPs on the porosity and pore size distribution was determined by MIP is discussed below. Figure 6 illustrates the pore size distribution of CCNPs composite cement pastes. Based on the MIP test, after 28 days curing, Table 2 shows the total porosity of 0, CCNPs-0.01, CCNPs-0.02, and CCNPs-0.04 when CCNPs account for 0, 0.01, 0.02, and 0.04% of the weight of the cement. According to Table 2, the results illustrate the total porosities are 15.59, 15.42, 15.51, and 15.58%, correspondingly. Apparently, there are no significant changes in the total porosity, but pore size distribution is affected a lot by the incorporation of CCNPs. For CCNPs-0.01, the harmless pore (<20 nm) increased from 18.54% to 22.97%, and the less harm pore (20-50 nm) increased from 37.91% to 47.52%. In addition, the performance displays that it could optimize the pore structure of cement-based materials with a little amount of CCNPs in the cement mix, which could be attributed to the flake structure of CCNPs (refer to Figure 4c). CCNPs are nanometer scale, with the thickness of several nanometers, which easily generate sheet wrinkle and interlock with some more sheets. Moreover, CCNPs are hydrophilic material, it could be believed that nanoscale CCNPs can lead to the ordinary influence by optimization of pore structure in cement-based materials [2][3][4]. It can also be observed that with the 0.04% of CCNPs incorporating to cement mortar, the result shows that both the less harm pore and large pore increase while the harmless pore decreases compared with the control sample. That is because when CCNPs are introduced into the cement, they will be surrounded by the hydration product of the cement and remain in it. When chloride ions intrude into cement, the permeation path of chloride ion can be changed and extended, thus improving its impermeability [5].
However, with the increasing content of CCNPs, the properties of cement-based materials decrease even lower than those of control samples. It could be concluded that, due to the effect of CCNPs stacking effect and its irregular shapes, coupled with the combined effect of weakening the nanoscale effect, when the amount of CCNPs increases, more water was needed for the dispersion of CCNPs, which easily results in an aggregation effect in the mix. As reported [40][41][42][43][44][45], the pore structure of cement-based materials influences its physical properties directly, including compressive strength and durability. It can also be observed that with the 0.04% of CCNPs incorporating to cement mortar, the result shows that both the less harm pore and large pore increase while the harmless pore decreases compared with the control sample. That is because when CCNPs are introduced into the cement, they will be surrounded by the hydration product of the cement and remain in it. When chloride ions intrude into cement, the permeation path of chloride ion can be changed and extended, thus improving its impermeability [5].
However, with the increasing content of CCNPs, the properties of cement-based materials decrease even lower than those of control samples. It could be concluded that, due to the effect of CCNPs stacking effect and its irregular shapes, coupled with the combined effect of weakening the nanoscale effect, when the amount of CCNPs increases, more water was needed for the dispersion of CCNPs, which easily results in an aggregation effect in the mix. As reported [40][41][42][43][44][45], the pore structure of cement-based materials influences its physical properties directly, including compressive strength and durability.

Chloride Penetration Test
The result of the chloride penetration impermeability of cement mortar specimens is shown in Figure 7. It could be concluded that the incorporation of CCNPs can effectively affect the permeability of chloride in cement mortar. When the amount of CCNPs increased from 0.01% to 0.04%, the chloride penetration impermeability of the cement mortar is getting lower, that is, permeability resistance of cement mortar improves. Furthermore, cement mortar with 0.04% of CCNPs incorporated exhibits the lowest chloride depth penetration, reduced by 15.7% than that of the control samples. However, when the content of CCNPs is higher than 0.04%, the permeability coefficient of the cement mortar will increase, but it is still lower than that of the control samples. Until the content of CCNPs reaches to 0.2%, the permeability coefficient of the mortar is higher than that of the control samples. In short, a small amount of CCNPs plays an important role in optimizing the permeability of cement mortar and enhancing its durability, when the amount of CCNPs exceeds a certain value of 0.04%, the effect of the addition of CCNPs on the impermeability of the mortar will be weakened until the opposite effect occurs.

Chloride Penetration Test
The result of the chloride penetration impermeability of cement mortar specimens is shown in Figure 7. It could be concluded that the incorporation of CCNPs can effectively affect the permeability of chloride in cement mortar. When the amount of CCNPs increased from 0.01% to 0.04%, the chloride penetration impermeability of the cement mortar is getting lower, that is, permeability resistance of cement mortar improves. Furthermore, cement mortar with 0.04% of CCNPs incorporated exhibits the lowest chloride depth penetration, reduced by 15.7% than that of the control samples. However, when the content of CCNPs is higher than 0.04%, the permeability coefficient of the cement mortar will increase, but it is still lower than that of the control samples. Until the content of CCNPs reaches to 0.2%, the permeability coefficient of the mortar is higher than that of the control samples. In short, a small amount of CCNPs plays an important role in optimizing the permeability of cement mortar and enhancing its durability, when the amount of CCNPs exceeds a certain value of 0.04%, the effect of the addition of CCNPs on the impermeability of the mortar will be weakened until the opposite effect occurs.

Compressive Strength of Cement Paste and Cement Mortar
As the curing age extended, the compressive strength increased correspondingly, because the process of the cement hydration had gradually accomplished. Figure 8 shows the change of compressive strength of cement paste with respect to four different CCNPs dosage ratios of 0, 0.01, 0.02, and 0.04%. It is noted that the incorporation of CCNPs shows a significant effect on cement paste compressive strength, with value ranging from 55 to 92 MPa, accordingly. The evolution of cement hydration shows that the trend of strength changes gradually milder. Moreover, referring to Figure  8, the compressive strength achieves the highest value when incorporating 0.01% of CCNPs regardless of curing ages, and the compressive strength increased by 37.9% than that of the control samples, respectively. However, when the amount of CCNPs increases to 0.04%, increment of compressive strength is lower than that of 0.01%. In general, the compressive strength increases with the CCNP dosage.  Figure 9 shows the change in compressive strength of cement mortar with 0, 0.01, 0.02, and 0.04% for four different CCNPs dose ratios. With the extension of curing time, the compressive strength of cement mortar is gradually increased. It can be observed that when the CCNPs are incorporated to cement mortar, there was no significant change in compressive strength at 28 days compared to the control sample. The reason for this phenomenon is that a large amount of sand is added to the cement mortar, which greatly dilutes the content of CCNPs in the cement, thereby weakening the effect of CCNPs on the mechanical properties of the cement paste. However, at the curing time of 3 days, the

Compressive Strength of Cement Paste and Cement Mortar
As the curing age extended, the compressive strength increased correspondingly, because the process of the cement hydration had gradually accomplished. Figure 8 shows the change of compressive strength of cement paste with respect to four different CCNPs dosage ratios of 0, 0.01, 0.02, and 0.04%. It is noted that the incorporation of CCNPs shows a significant effect on cement paste compressive strength, with value ranging from 55 to 92 MPa, accordingly. The evolution of cement hydration shows that the trend of strength changes gradually milder. Moreover, referring to Figure 8, the compressive strength achieves the highest value when incorporating 0.01% of CCNPs regardless of curing ages, and the compressive strength increased by 37.9% than that of the control samples, respectively. However, when the amount of CCNPs increases to 0.04%, increment of compressive strength is lower than that of 0.01%. In general, the compressive strength increases with the CCNP dosage.

Compressive Strength of Cement Paste and Cement Mortar
As the curing age extended, the compressive strength increased correspondingly, because the process of the cement hydration had gradually accomplished. Figure 8 shows the change of compressive strength of cement paste with respect to four different CCNPs dosage ratios of 0, 0.01, 0.02, and 0.04%. It is noted that the incorporation of CCNPs shows a significant effect on cement paste compressive strength, with value ranging from 55 to 92 MPa, accordingly. The evolution of cement hydration shows that the trend of strength changes gradually milder. Moreover, referring to Figure  8, the compressive strength achieves the highest value when incorporating 0.01% of CCNPs regardless of curing ages, and the compressive strength increased by 37.9% than that of the control samples, respectively. However, when the amount of CCNPs increases to 0.04%, increment of compressive strength is lower than that of 0.01%. In general, the compressive strength increases with the CCNP dosage.  Figure 9 shows the change in compressive strength of cement mortar with 0, 0.01, 0.02, and 0.04% for four different CCNPs dose ratios. With the extension of curing time, the compressive strength of cement mortar is gradually increased. It can be observed that when the CCNPs are incorporated to cement mortar, there was no significant change in compressive strength at 28 days compared to the control sample. The reason for this phenomenon is that a large amount of sand is added to the cement mortar, which greatly dilutes the content of CCNPs in the cement, thereby weakening the effect of CCNPs on the mechanical properties of the cement paste. However, at the curing time of 3 days, the  Figure 9 shows the change in compressive strength of cement mortar with 0, 0.01, 0.02, and 0.04% for four different CCNPs dose ratios. With the extension of curing time, the compressive strength of cement mortar is gradually increased. It can be observed that when the CCNPs are incorporated to cement mortar, there was no significant change in compressive strength at 28 days compared to the control sample. The reason for this phenomenon is that a large amount of sand is added to the cement mortar, which greatly dilutes the content of CCNPs in the cement, thereby weakening the effect of CCNPs on the mechanical properties of the cement paste. However, at the curing time of 3 days, the compressive strength of the cement mortar increased with the increase of the amount of CCNPs added, indicating that the CCNPs play a nucleating role in the cement mortar and promote the hydration of the cement.

Conclusions
Seashell landfill brings about a great deal of pollution and waste into the environment. Recycling waste seashells can effectively solve the problems. In this study, Nanoscale CCNPs were obtained from alkali-treated mussel shells powder. With incorporating CCNPs to the cement-based materials at lower dosage, the results indicate the positive effect on the resistance of chloride ion penetration and the compressive strength. Besides, it can bring considerable economic and environmental efficiency. Therefore, the following conclusions can be made: Nanoscale CCNPs can be successfully obtained from mussel shell powder by alkali treatment. Specific test process: the mussel shell powder was placed in the mass fraction of 5% sodium hydroxide solution, 80 °C water bath heating, and supplemented by 1.5 h of ultrasound correspondingly.
CCNPs have an efficacious influence on chloride ion impermeability and compressive strength by refinement of the pore structure of cement-based materials. The cement mortar incorporating 0.04% CCNPs resulted in a 16.1% reduction in chloride penetration as compared with the control mortar. Also, the cement paste including 0.01% CCNPs provided 37.9% higher compressive strength than the control paste. Yet, CCNPs should remain in the low dosage because of the agglomeration effect.
Author Contributions: P.H. conducted the experiments and data analysis and wrote part of this paper. C.L. provided the original ideas and conducted data analysis. L.L. wrote part of this paper. W.L. participated in the revision work. Z.X. provided the original ideas.