Mechanical Behavior of Compression-Compacted Dry Concrete Paver Blocks Making Use of Sea Sand and Seawater

: Dry concrete is a kind of concrete whose fresh mixture has almost no ﬂowability and is widely used in the production of small-size unreinforced compression-compacted concrete blocks in plants. Considering the shortage of natural river sand and freshwater for concrete production, this study proposes that sea sand and seawater can be directly used in the manufacture of compression-compacted dry concrete paver blocks. The idea was veriﬁed in the laboratory to ﬁnd suitable mix proportions and forming pressure, which are two key parameters for the production of paver blocks. Furthermore, the effect of sea sand replacement ratio and seawater replacement ratio is investigated, where compression and ﬂexural tensile tests were conducted on lab-made paver blocks at different ages. The experimental results reveal that both the compressive and ﬂexural tensile strengths of paver blocks increased when sea sand and seawater were adopted. It is ﬁnally suggested that sea sand and seawater are suitable for the production of unreinforced paver blocks with enhanced mechanical performance.


Introduction
The construction industry consumes a huge amount of sand every year.During the past decades, many countries and regions have faced a shortage of natural river sand.To solve this problem, some alternatives to natural river sand have been widely investigated, such as sea sand [1], recycled sand [2,3], machine-made sand [4,5], etc.Among them, sea sand is the most competitive one, due to its large and wide storage as well as its geometric similarity to natural river sand.However, the detrimental ions in sea sand prevent its direct use in reinforced concrete (RC) structures.As a result, sea sand generally needs to be desalted to fulfill specific requirements [6] before it can be used in RC structures and prestressed concrete (PC) structures.The desalting of sea sand, however, may introduce other issues like large consumption of freshwater and secondary pollution due to the release of waste salty water.
Some regions also face a shortage of freshwater for concrete production.Even in regions with sufficient rainfall, abnormal weather may lead to a temporary shortage of freshwater.For example, Wenzhou, where the authors live, is a coastal city in East China.The annual rainfall varies between 1100 mm and 2500 mm, which generally means sufficient freshwater supply.However, the total rainfall between October 2020 and January 2021 was only about 44 mm.During this period, freshwater for living can hardly be guaranteed, not to mention the industrial production of concrete.For many islands whose freshwater supply is dependent on the nearby mainland, freshwater supply is always difficult and expensive.In those situations, the direct use of seawater in mixing concrete is very attractive.
There has been explosive growth in the research of seawater sea sand concrete in recent years [1,7,8].Some researchers [1,9,10] tested the compressive strength of sea sand concrete and found that sea sand has a positive or little effect on the short-term compressive strength of concrete.High shell content and organic substances in sea sand may negatively affect the mechanical behavior as well as flowability of concrete [11], which should be considered during the application of sea sand concrete.Some other researchers [12,13] investigated the effect of seawater on the strength and workability of cement paste and found that seawater can increase the compressive strength of cement paste while slightly decreasing its workability.Research on seawater sea sand concrete (SSC) [1,9,13] found that the early-age strength of SSC is generally higher than its normal concrete counterpart.The major reason for the high early strength is that chloride in seawater can affect the hydration process of cement and help form a denser and stronger microstructure [10,14].Three recent reviews [1,7,8] on SSC found that a general consensus has been achieved among researchers that the early-age strength of SSC is higher and the long-term strength is similar to that of normal concrete.Current research on the durability of SSC found that the carbonization performance of SSC is even slightly better than that of normal concrete due to its denser microstructure [15].The adoption of mineral admixtures (silicon powder, fly ash, slag, etc.) can further improve the microstructure and performance of SSC [16,17].Except for research on normal-performance SSC, sea sand and seawater have also been used in the development of high-performance concrete, such as ultra-high-performance concrete [18,19] and self-compacting concrete [20], which further widens the possible applications of seawater and sea sand.
The above studies prove that sea sand and seawater are not detrimental or even beneficial to the performance of concrete.However, the major obstacle to the application of SSC is the corrosion of steel reinforcement in a chloride environment.While the corrosion of steel bars cannot be avoided, some researchers have proposed that alternatives to steel reinforcement should be used in seawater sea sand concrete, such as stainless steel bars [21], fiber-reinforced polymer (FRP) [21][22][23][24][25], steel-FRP composite bars [26], etc. FRP has been proven to possess supreme durability even in seawater environments [27][28][29].The bond between FRP bars and SSC was found to be sufficient [30].The behavior of FRP rebar-reinforced SSC beams, plates, and columns has also been investigated by different researchers [26,31,32].
Based on the above discussion, the recent research on SSC is very hot, and the results are very promising.However, the application of SSC in real projects is very limited so far.One major reason is the lack of a design code or guidelines for SSC, which needs to be based on much more research and will take some time.It is also found that the current research on SSC has been focused on the structural use of SSC.While the current research on SSC generally supports the use of SSC in plain, unreinforced concrete, it will be interesting to try to use SSC in unreinforced non-structural members.As far as the authors' knowledge, the only application of SSC in plain concrete was found in [33], where seawater was directly adopted in the on-site casting of large unreinforced dyke blocks.
Concrete paver blocks (Figure 1) are widely used in cities, such as footways, squares, and parks [34].In the massive production of paver blocks, dry concrete without flowability is preferred compared with normal plastic concrete so that the blocks can be de-molded immediately after consolidation to speed up production efficiency.Dry concrete also uses less cement to help reduce the cost.Due to its poor flowability, the compaction of dry concrete is generally achieved by external mechanical compression, together with mechanical vibration if necessary [35][36][37][38].In this sense, the production of concrete paver blocks is quite similar to roller-compacted concrete (RCC), which is widely used in dam engineering [38].Several previous studies have been dedicated to the adoption of recycled aggregates in the production of environmentally friendly paver blocks [34][35][36].A comprehensive review of dry concrete was recently conducted by [38], where the application and classification, raw materials, manufacturing, static and dynamic strengths, and durability issues of dry concrete were discussed.
massive production of concrete paver blocks.Direct use of sea sand and seawater is both 'green' and 'low-carbon,' as it can reduce the usage of river sand and freshwater and avoid the consumption of freshwater and energy during the desalting of sea sand.To the best of the authors' knowledge, no previous study has been reported on the behavior of such compression-compacted dry concrete blocks made of seawater and sea sand.As the firstever study on seawater sea sand paver blocks, this research first investigated the effect of water-to-cement ratio and forming pressure on the mechanical performance of paver blocks, which verified the appropriateness of the procedure adopted in the laboratory to fabricate the blocks.The effects of the sea sand replacement ratio and the seawater replacement ratio were then investigated.

Materials
The materials used in this study include PO 42.5 cement, tap water, sea sand, and crushed stone aggregates.The cement was provided by a local company and fulfilled the requirements of the corresponding product code.Normal tap water was used as the mixing water directly [39] and was used to produce man-made seawater in Section 3.5.Un-desalted sea sand was obtained from a local desalting plant.The coarse aggregates adopted were single-grade crushed stone with a nominal aggregate size of 5-10 mm.Sieve analysis was conducted on both sea sand and coarse aggregates (Figure 2).The grading curve of sea sand is shown in Figure 2a, together with the lower and upper limits of Zone III sand defined by Chinese code [40].Based on the grading curve, the sea sand is classified as fine sand.For coarse aggregates (Figure 2b), about 80% of the aggregates fall in the range of 4.75 mm and 9.5 mm, while the other 20% fall in the range of 2.36 mm and 4.75 mm.As a non-structural member, paver blocks generally do not contain any form of reinforcement and, of course, do not have the problem of steel corrosion.As a result, the authors believe that sea sand and seawater have great potential to be directly used in the massive production of concrete paver blocks.Direct use of sea sand and seawater is both 'green' and 'low-carbon,' as it can reduce the usage of river sand and freshwater and avoid the consumption of freshwater and energy during the desalting of sea sand.To the best of the authors' knowledge, no previous study has been reported on the behavior of such compression-compacted dry concrete blocks made of seawater and sea sand.As the first-ever study on seawater sea sand paver blocks, this research first investigated the effect of water-to-cement ratio and forming pressure on the mechanical performance of paver blocks, which verified the appropriateness of the procedure adopted in the laboratory to fabricate the blocks.The effects of the sea sand replacement ratio and the seawater replacement ratio were then investigated.

Materials
The materials used in this study include PO 42.5 cement, tap water, sea sand, and crushed stone aggregates.The cement was provided by a local company and fulfilled the requirements of the corresponding product code.Normal tap water was used as the mixing water directly [39] and was used to produce man-made seawater in Section 3.5.Un-desalted sea sand was obtained from a local desalting plant.The coarse aggregates adopted were single-grade crushed stone with a nominal aggregate size of 5-10 mm.Sieve analysis was conducted on both sea sand and coarse aggregates (Figure 2).The grading curve of sea sand is shown in Figure 2a, together with the lower and upper limits of Zone III sand defined by Chinese code [40].Based on the grading curve, the sea sand is classified as fine sand.For coarse aggregates (Figure 2b), about 80% of the aggregates fall in the range of 4.75 mm and 9.5 mm, while the other 20% fall in the range of 2.36 mm and 4.75 mm.

Specimen Fabrication
The massive production of paver blocks in the plant is finished by a highly automatic production line, which covers the mixing and delivery of raw materials, the forming and de-molding process, curing and packing, etc.In this experimental study, the paver blocks

Specimen Fabrication
The massive production of paver blocks in the plant is finished by a highly automatic production line, which covers the mixing and delivery of raw materials, the forming and de-molding process, curing and packing, etc.In this experimental study, the paver blocks are fabricated in the laboratory using specially designed steel molds with the help of a hydraulic loading machine.The internal dimensions of steel molds in this study are length × width × thickness = 200 mm × 100 mm × 100 mm.Several steel molds were prepared.Each mold consists of one bottom plate, four side plates, and four connection bolts, as shown in Figure 3a.The bottom plate and side plates are connected together by four bolts and some shear slots, as shown in Figure 3b,c.The thickness of all side plates is 20 mm, and that of the bottom plate is 10 mm.It should be mentioned that in the paver block plant, the thickness of the steel mold is much larger (i.e., 40 mm), and the side steel plates were welded together instead of using bolt connections.The stiffness of the steel molds used in this study is much smaller than that used in the plant, which may lead to an excessive rebound of the steel mold under high forming pressure, as will be explained in Section 3.3.A solid steel block (Figure 3d) with plan dimensions slightly smaller than 200 mm × 100 mm was used to receive the compressive load to consolidate the concrete.The mold can be used to fabricate blocks with thicknesses smaller than 100 mm.In this study, the thickness of all blocks is 60 mm.The paver blocks were fabricated according to following procedure (Figure 4): (1) Concrete mixing: the sand, aggregates were mixed for one minute, then the cement was added and mixed for another minute, finally the water was added and mixed for three minutes.The concrete mixture is very dry as shown in Figure 4a.(2) Casting: about 2.7 kg concrete mixture was filled in a mould and flattened manually (Figure 4b), followed by the installation of top steel block; (3) the mold together with top steel block was placed at the center of a hydraulic loading machine (Figure 4c).The load was increased from 0 to 400 kN (20 MPa) in one minute.The load was then removed after a time duration of 1 minute.(4) de-moulding and curing: the side plates of the mould was removed The paver blocks were fabricated according to following procedure (Figure 4): (1) Concrete mixing: the sand, aggregates were mixed for one minute, then the cement was added and mixed for another minute, finally the water was added and mixed for three minutes.The concrete mixture is very dry as shown in Figure 4a.( 2) Casting: about 2.7 kg concrete mixture was filled in a mould and flattened manually (Figure 4b), followed by the installation of top steel block; (3) the mold together with top steel block was placed at the center of a hydraulic loading machine (Figure 4c).The load was increased from 0 to 400 kN (20 MPa) in one minute.The load was then removed after a time duration of 1 min.(4) de-moulding and curing: the side plates of the mould was removed immediately and the blocks were kept static on the bottom plate for 24 h and then transferred to a standard curing room (20 ± 2 • C, ≥95% relative humidity) until testing (Figure 4d).

Testing Methods
In this study, the compressive strength and flexural tensile strength of the were tested following Chinese code [41].For each mix, the strengths were tested ages of 3-day, 7-day, 14-day, and 28-day.At each age, three blocks were test compression, and another three were tested for flexure.Before testing, the block immersed in 20℃ water for 24 hours and then taken out and wiped with a cloth s the blocks were in a saturated surface dry (SSD) condition as required by the code.
For the compression test, a block was placed at the center of a hydraulic lo machine (Figure 5a).The load was increased with a constant loading rate of 0.4 MP 8 kN/s) [41] until the block was crushed.The compressive strength was calculated ratio between the maximum load and the nominal area of the block.
The flexure tensile tests were conducted by a 300-kN material testing machine.testing, the block was mounted on a jig with a span length of 150 mm, as shown in 5b.A constant loading rate of 0.04 MPa/s (or 64 N/s) [41] was then applied until the ruptured.The flexural tensile strength was then calculated by the following equatio

Testing Methods
In this study, the compressive strength and flexural tensile strength of the blocks were tested following Chinese code [41].For each mix, the strengths were tested at the ages of 3-day, 7-day, 14-day, and 28-day.At each age, three blocks were tested for compression, and another three were tested for flexure.Before testing, the blocks were immersed in 20 • C water for 24 h and then taken out and wiped with a cloth so that the blocks were in a saturated surface dry (SSD) condition as required by the code.
For the compression test, a block was placed at the center of a hydraulic loading machine (Figure 5a).The load was increased with a constant loading rate of 0.4 MPa/s (or 8 kN/s) [41] until the block was crushed.The compressive strength was calculated as the ratio between the maximum load and the nominal area of the block.

Results and Discussion
In this section, the effect of different parameters on the compression and flexural strength of paver blocks is presented and discussed.It should be mentioned that the experiments on different sub-sections were finished in different time periods using different batches of cement.As a result, the strength of paver blocks using nominally the same mix may have different strengths in different subsections, and the results of different subsections should not be compared directly.

Failure Modes
During the compression test, the blocks generally failed by crushing and spalling the concrete, especially near the corners.A typical specimen after a compression test is shown in Figure 6a.In the flexural test, the block is under three-point bending, and the failure is generally characterized by a single major crack near the loading point.A typical block after a flexural tensile test is shown in Figure 6b.The failure modes of blocks with different mix proportions are generally similar and will not be repeated or discussed in the following subsections.

Effect of Water-to-Cement Ratio
For normal plastic concrete, the water-to-cement (W/C) ratio has the largest effect on the strength of the concrete.This subsection investigates the effect of W/C ratio on the strength of compression-compacted paver blocks.Four mix proportions with different W/C ratios were investigated, as shown in Table 1.The mix proportions were modified from a commercial production line of C40 paver blocks.The dosage of cement, sand, and The flexure tensile tests were conducted by a 300-kN material testing machine.Before testing, the block was mounted on a jig with a span length of 150 mm, as shown in Figure 5b.A constant loading rate of 0.04 MPa/s (or 64 N/s) [41] was then applied until the block ruptured.The flexural tensile strength was then calculated by the following equation [41]: where P is the maximum load, L is the span length (150 mm), b is the width of the block (100 mm), and t is the thickness of the block (60 mm).

Results and Discussion
In this section, the effect of different parameters on the compression and flexural strength of paver blocks is presented and discussed.It should be mentioned that the experiments on different sub-sections were finished in different time periods using different batches of cement.As a result, the strength of paver blocks using nominally the same mix may have different strengths in different subsections, and the results of different subsections should not be compared directly.

Failure Modes
During the compression test, the blocks generally failed by crushing and spalling the concrete, especially near the corners.A typical specimen after a compression test is shown in Figure 6a.In the flexural test, the block is under three-point bending, and the failure is generally characterized by a single major crack near the loading point.A typical block after a flexural tensile test is shown in Figure 6b.The failure modes of blocks with different mix proportions are generally similar and will not be repeated or discussed in the following subsections.

Effect of Water-to-Cement Ratio
For normal plastic concrete, the water-to-cement (W/C) ratio has the largest effect on the strength of the concrete.This subsection investigates the effect of W/C ratio on the strength of compression-compacted paver blocks.Four mix proportions with different W/C ratios were investigated, as shown in Table 1.The mix proportions were modified from a commercial production line of C40 paver blocks.The dosage of cement, sand, and coarse aggregates was kept unchanged while the water amount was changed so that the W/C ratio varied between 0.29 and 0.35.The sand ratio is 0.53, which is larger than normal plastic concrete due to the large voids among the single-grade coarse aggregates.A large sand ratio also helps improve the surface smoothness of the final product.In this subsection, the mechanical pressure used to compact the concrete was taken as 20 MPa for all blocks.Un-desalted sea sand and tap water were used for all mixes.

Failure Modes
During the compression test, the blocks generally failed by crushing and spalling the concrete, especially near the corners.A typical specimen after a compression test is shown in Figure 6a.In the flexural test, the block is under three-point bending, and the failure is generally characterized by a single major crack near the loading point.A typical block after a flexural tensile test is shown in Figure 6b.The failure modes of blocks with different mix proportions are generally similar and will not be repeated or discussed in the following subsections.

Effect of Water-to-Cement Ratio
For normal plastic concrete, the water-to-cement (W/C) ratio has the largest effect on the strength of the concrete.This subsection investigates the effect of W/C ratio on the strength of compression-compacted paver blocks.Four mix proportions with different W/C ratios were investigated, as shown in Table 1.The mix proportions were modified from a commercial production line of C40 paver blocks.The dosage of cement, sand, and  The compressive strength (f c ) and flexural tensile strength (f rup ) of each single paver block at different ages are listed in Table 2, and the average strengths at different ages are plotted in Figure 7.It can be seen that both compressive strength and flexural tensile strength generally increase with an increase in the W/C ratio at all ages (Figure 7).When the W/C ratio was increased from 0.29 to 0.35, the average 28-day compressive strength of the blocks increased by 70.6% (35.7 MPa vs. 60.9MPa), while the average 28-day flexural tensile strength increased by 28.2% (3.40 MPa vs. 4.36 MPa).However, for plastic concrete, it is well accepted that concrete strength generally decreases with an increase in the W/C ratio.The reason for the disagreement is believed to be that with an increase in the W/C ratio, the concrete is easier to compact, and denser concrete generally leads to an increase in strength.Based on the experience in the massive production of paver blocks in the first author's plant, if the W/C ratio is too high, the block may not keep its shape when the mold is immediately removed after forming, as is conducted in the massive production of paver blocks.This will affect the accuracy of the final products.Also, a very large W/C ratio may also lead to a larger shrinkage of blocks during curing.As a result, a W/C ratio of 0.35 was adopted in the following parts of this study, although the strengths may further increase beyond this W/C ratio.
Figure 8 shows the development of strength with concrete age.In the figure, the vertical axis represents relative strength normalized by the corresponding 28-day strength.The strength development curve predicted by the CEB-FIP model code [42] is also shown in the figure as a dashed line.It can be seen from the figure that early-age strength generally develops quicker than that estimated by the CEB-FIP model code.At the age of 3 days, the relative compressive strength and relative flexural tensile strength reached 78.9% and 75.9%, respectively, compared with 59.8% predicted by the CEB-FIP code.The reason is believed to be due to the way that concrete was compacted, where substantial strength can be established by the interlocking of aggregates soon after the forming process.Figure 8 shows the development of strength with concrete age.In the figure, the vertical axis represents relative strength normalized by the corresponding 28-day strength.The strength development curve predicted by the CEB-FIP model code [42] is also shown in the figure as a dashed line.It can be seen from the figure that early-age strength generally develops quicker than that estimated by the CEB-FIP model code.At the age of 3 days, the relative compressive strength and relative flexural tensile strength reached 78.9% and 75.9%, respectively, compared with 59.8% predicted by the CEB-FIP code.The reason is believed to be due to the way that concrete was compacted, where substantial strength can be established by the interlocking of aggregates soon after the forming process.Figure 8 shows the development of strength with concrete age.In the figure, the vertical axis represents relative strength normalized by the corresponding 28-day strength.The strength development curve predicted by the CEB-FIP model code [42] is also shown in the figure as a dashed line.It can be seen from the figure that early-age strength generally develops quicker than that estimated by the CEB-FIP model code.At the age of 3 days, the relative compressive strength and relative flexural tensile strength reached 78.9% and 75.9%, respectively, compared with 59.8% predicted by the CEB-FIP code.The reason is believed to be due to the way that concrete was compacted, where substantial strength can be established by the interlocking of aggregates soon after the forming process.

Effect of Forming Pressure
For paver blocks, the compaction of concrete is achieved by external mechanical pressure.The magnitude of pressure (i.e., forming pressure) affects the degree of consolidation as well as the strength of paver blocks.In this subsection, three different forming pressures were investigated, namely 10 MPa, 20 MPa, and 30 MPa, respectively.The concrete mix proportion with a W/C ratio of 0.35 (Table 1) was adopted.Un-desalted sea sand and tap water were used for all forming pressures.
The compressive strength (fc) and flexural tensile strength (frup) of paver blocks with different forming pressures are shown in Table 3.To better understand the effect of forming pressure, the average strengths at different ages are plotted in Figure 9.

Effect of Forming Pressure
For paver blocks, the compaction of concrete is achieved by external mechanical pressure.The magnitude of pressure (i.e., forming pressure) affects the degree of consolidation as well as the strength of paver blocks.In this subsection, three different forming pressures were investigated, namely 10 MPa, 20 MPa, and 30 MPa, respectively.The concrete mix proportion with a W/C ratio of 0.35 (Table 1) was adopted.Un-desalted sea sand and tap water were used for all forming pressures.
The compressive strength (f c ) and flexural tensile strength (f rup ) of paver blocks with different forming pressures are shown in Table 3.To better understand the effect of forming pressure, the average strengths at different ages are plotted in Figure 9.When the forming pressure is increased from 10 MPa to 20 MPa, the compressive strength increases substantially.Take 28-day strength as an example; the compressive strength increased by 41.4% (47.3 MPa vs. 66.9MPa).The flexural tensile strength, however, only slightly increased by 1.8% (4.45 MPa vs. 4.53 MPa).The increase in compressive strength is believed to be due to better compaction under increased forming pressure.When forming pressure is further increased from 20 MPa to 30 MPa.However, both the compressive strength and the flexural tensile strength of paver blocks decrease.The 28-day compressive strength decreases by 22.6% (51.8 MPa vs. 66.9MPa), and the flexural tensile strength decreases by 16.1% (3.80 MPa vs. 4.53 MPa).The reason for the strength degradation is believed to be due to the elastic rebound of the steel mold, which causes internal damage to the blocks because the rebound increases with the increase in forming pressure.Some visible deformation and minor cracks were even observed during the preparation of the blocks.The irregular damage leads to a larger scatter in the mechanical strength of those blocks.The larger scatter perhaps explains why the 3-day strengths of blocks with a forming pressure of 30 MPa were higher than those of 20 MPa but were lower at other ages.It may also explain why their 28-day strengths were only slightly higher than those of the 14-day.Based on the above results, a forming pressure of 20 MPa is deemed suitable for the steel mold of the present study.The forming pressure is taken as 20 MPa in the following subsections.

Effect of Sea Sand
This subsection investigates the effect of sea sand on the mechanical strengths of paver blocks.Four nominally the same concrete mixes were adopted, which is Mix #4, When the forming pressure is increased from 10 MPa to 20 MPa, the compressive strength increases substantially.Take 28-day strength as an example; the compressive strength increased by 41.4% (47.3 MPa vs. 66.9MPa).The flexural tensile strength, however, only slightly increased by 1.8% (4.45 MPa vs. 4.53 MPa).The increase in compressive strength is believed to be due to better compaction under increased forming pressure.When forming pressure is further increased from 20 MPa to 30 MPa.However, both the compressive strength and the flexural tensile strength of paver blocks decrease.The 28-day compressive strength decreases by 22.6% (51.8 MPa vs. 66.9MPa), and the flexural tensile strength decreases by 16.1% (3.80 MPa vs. 4.53 MPa).The reason for the strength degradation is believed to be due to the elastic rebound of the steel mold, which causes internal damage to the blocks because the rebound increases with the increase in forming pressure.Some visible deformation and minor cracks were even observed during the preparation of the blocks.The irregular damage leads to a larger scatter in the mechanical strength of those blocks.The larger scatter perhaps explains why the 3-day strengths of blocks with a forming pressure of 30 MPa were higher than those of 20 MPa but were lower at other ages.It may also explain why their 28-day strengths were only slightly higher than those of the 14-day.Based on the above results, a forming pressure of 20 MPa is deemed suitable for the steel mold of the present study.The forming pressure is taken as 20 MPa in the following subsections.

Effect of Sea Sand
This subsection investigates the effect of sea sand on the mechanical strengths of paver blocks.Four nominally the same concrete mixes were adopted, which is Mix #4, shown in Table 1.The sea sand replacement ratio is defined as the mass ratio between the masses of un-desalted sea sand and total sand.The desalted sand in the present study was obtained by immersing the sea sand in sufficient freshwater for 72 h, where the freshwater was refreshed every 24 h.After 72 h, the sand was taken out of the water and oven-dried to obtain desalted sand.Three different replacement ratios, namely 0%, 50%, and 100%, were investigated.Since sea sand derived from different sea areas may contain different chloride ions due to variations in seawater salinity, it is decided to further investigate the chloride content of sea sand.To achieve this, some sea sand was immersed in man-made seawater with two times the globe's average salinity (7.0%) to produce some 'saltier' sea sand.The 'saltier' sand was produced using a similar procedure to the desalted sand but using man-made seawater instead of freshwater.The chloride and sulfate content of desalted sea sand, natural sea sand, and 'saltier' sea sand were tested following [43], and the results are listed in Table 4.The soluble chloride ion content was tested following a standard titration procedure, and silver nitrate was used as an indicator.The sulfate content was tested using the gravimetric method, where barium chloride was used to fix sulfate in barium sulfate precipitation.It can be seen that the chloride content of sea sand is 14.2 times that of desalted sea sand, while the chloride content of 'saltier' sea sand is 1.9 times that of natural sea sand.The compressive strength (f c ) and flexural tensile strength (f rup ) of paver blocks with different sea sand replacement ratios are shown in Table 5, as well as Figure 10.It can be seen that the compressive and flexural tensile strength of paver blocks generally increases with the increase in sea sand replacement ratio, and the blocks made with 'saltier' sea sand possess the highest compressive and flexural tensile strength.Take 28-day strength as an example; the compressive and flexural tensile strengths of blocks with a replacement ratio of 100% are 10.7% (57.1 MPa vs. 51.6MPa) and 8.7% (4.49MPa vs. 4.13 MPa) higher than blocks made from completely desalted sea sand (i.e., a replacement ratio of 0%).The compressive strength and flexural tensile strength of blocks made with 100% 'saltier' sea sand are 6.3% (60.7 MPa vs. 57.1 MPa) and 1.8% (4.57MPa vs. 4.49 MPa) higher than those of blocks made with 100% natural sea sand.Based on Table 5, it can be seen that the strength of paver blocks increases with the replacement ratio or the amount of chloride introduced by the sea sand.Previous hydration heat evolution analysis [13,14] found that chloride can accelerate the formation of Friedel's salt, which in turn promotes the hydration of aluminate and ferrite minerals.Scanning electron microscopy (SEM) analysis [13] also indicates that more C-S-H nuclei are found around C 3 S during the early age of hydration, which helps form a denser microstructure and higher mechanical strength.The same mechanism is believed to account for the strength enhancement of sea sand paver blocks in this study.

Effect of Seawater
In this subsection, the effect of the seawater replacement ratio is investigated.Concrete Mix #4, shown in Table 1, is adopted again.The difference lies in the water adopted, where a certain percentage of seawater is used to replace freshwater in the mixing of concrete.The seawater is man-made seawater with a salinity of 3.5%, which was prepared following ASTM D1141-98 [44].In this study, only two major ions, namely chloride (Cl −1 ) and sulfate anion (SO4 2− ), are introduced into man-made seawater by adding sodium chloride (NaCl) and sodium sulfate (Na2SO4) to freshwater.The proportions of man-made seawater are shown in Table 6.Three different replacement ratios, namely 0%, 50%, and 100%, were adopted.Un-desalted sea sand is adopted for all mixes in this subsection.The compressive strength (fc) and flexural tensile strength (frup) of paver blocks with different seawater replacement ratios are shown in Table 7.The results are also plotted in Figure 11.It can be seen that both compressive strength and flexural tensile strength generally increase with an increase in the seawater replacement ratio.At the age of 3 days, the compressive strength and flexural tensile strength of paver blocks with 100% seawater are 64.8%(40.7 MPa vs. 24.7 MPa) and 106.2% (3.36 MPa vs. 1.63 MPa) higher than those with only freshwater (i.e., a replacement ratio of 0%).At the age of 28 days, the corresponding strength gains are 17.1% (55. 5

Effect of Seawater
In this subsection, the effect of the seawater replacement ratio is investigated.Concrete Mix #4, shown in Table 1, is adopted again.The difference lies in the water adopted, where a certain percentage of seawater is used to replace freshwater in the mixing of concrete.The seawater is man-made seawater with a salinity of 3.5%, which was prepared following ASTM D1141-98 [44].In this study, only two major ions, namely chloride (Cl −1 ) and sulfate anion (SO 4 2− ), are introduced into man-made seawater by adding sodium chloride (NaCl) and sodium sulfate (Na 2 SO 4 ) to freshwater.The proportions of man-made seawater are shown in Table 6.Three different replacement ratios, namely 0%, 50%, and 100%, were adopted.Un-desalted sea sand is adopted for all mixes in this subsection.The compressive strength (f c ) and flexural tensile strength (f rup ) of paver blocks with different seawater replacement ratios are shown in Table 7.The results are also plotted in Figure 11.It can be seen that both compressive strength and flexural tensile strength generally increase with an increase in the seawater replacement ratio.At the age of 3 days, the compressive strength and flexural tensile strength of paver blocks with 100% seawater are 64.8%(40.7 MPa vs. 24.7 MPa) and 106.2% (3.36 MPa vs. 1.63 MPa) higher than those with only freshwater (i.e., a replacement ratio of 0%).At the age of 28 days, the corresponding strength gains are 17.1% (55.5 MPa vs. 47.4MPa) and 30.5% (4.32 MPa vs. 3.31 MPa), respectively.Based on the test results, similar to normal plastic concrete [1,9], seawater can also enhance the early-age strength of paver blocks.

Conclusions
Dry concrete with almost no flowability is very popular in the production of unreinforced precast concrete blocks due to its low cost and high production efficiency.In this study, sea sand and seawater were proposed to be directly used in the production of compression-compacted dry concrete paver blocks.Compression tests and flexural tensile tests were conducted on paver blocks with different mix proportions.In total, 168 paver blocks (200 mm × 100 mm × 60 mm) were prepared in the lab and tested.Fourteen different concrete mixes were adopted in the present study, and the parameters investigated included four different water-to-cement ratios (0.29, 0.31, 0.33, and 0.35), three different forming pressures (10 MPa, 20 MPa, and 30 MPa), three different sea sand replacement ratios (0%, 50%, and 100%), and three different seawater replacement ratios (0%, 50%, and 100%).The following conclusions can be drawn based on the present study: 1.The compressive strength and flexural tensile strength of paver blocks increase with an increase in the water-to-cement ratio in the range of this study (i.e., 0.29 to 0.35).The reason is that dry concrete with a larger water-to-cement ratio is easier to compact under mechanical pressure.2. When the forming pressure is increased from 10 MPa to 20 MPa, the 28-day compressive strength and flexural tensile strength of blocks increase by 41.4% and 1.8%, respectively.However, when the forming pressure further increased from 20 MPa to 30 MPa, the 28-day compressive strength and flexural tensile strength of blocks decreased by 22.6% and 16.1%, respectively.The strength decrease is believed to be due to the elastic rebound of the steel mold, and a thicker steel mold should be used if a higher forming pressure is adopted.

Conclusions
Dry concrete with almost no flowability is very popular in the production of unreinforced precast concrete blocks due to its low cost and high production efficiency.In this study, sea sand and seawater were proposed to be directly used in the production of compression-compacted dry concrete paver blocks.Compression tests and flexural tensile tests were conducted on paver blocks with different mix proportions.In total, 168 paver blocks (200 mm × 100 mm × 60 mm) were prepared in the lab and tested.Fourteen different concrete mixes were adopted in the present study, and the parameters investigated included four different water-to-cement ratios (0.29, 0.31, 0.33, and 0.35), three different forming pressures (10 MPa, 20 MPa, and 30 MPa), three different sea sand replacement ratios (0%, 50%, and 100%), and three different seawater replacement ratios (0%, 50%, and 100%).The following conclusions can be drawn based on the present study: 1.
The compressive strength and flexural tensile strength of paver blocks increase with an increase in the water-to-cement ratio in the range of this study (i.e., 0.29 to 0.35).The reason is that dry concrete with a larger water-to-cement ratio is easier to compact under mechanical pressure.

2.
When the forming pressure is increased from 10 MPa to 20 MPa, the 28-day compressive strength and flexural tensile strength of blocks increase by 41.4% and 1.8%, respectively.However, when the forming pressure further increased from 20 MPa to 30 MPa, the 28-day compressive strength and flexural tensile strength of blocks decreased by 22.6% and 16.1%, respectively.The strength decrease is believed to be due to the elastic rebound of the steel mold, and a thicker steel mold should be used if a higher forming pressure is adopted.

3.
Both compressive strength and flexural tensile strength increase with the increase in sea sand replacement ratio.The 28-day compressive strength and flexural tensile strength of blocks made with 100% un-desalted sea sand are, respectively, 10.7% and 8.7% higher than blocks made with 100% desalted sea sand.4.
Both compressive strength and flexural tensile strength increase with the increase in seawater replacement ratio, especially at early ages.The 3-day compressive strength and flexural tensile strength of blocks made with 100% seawater are, respectively, 64.8% and 106.2% higher than blocks made with 100% freshwater.The corresponding increments at 28 days are 17.1% and 30.5%, respectively.5.
The current study suggests that sea sand and seawater can be directly used in the production of dry concrete paver blocks without steel reinforcement.The mechanical performance can be increased while the cost can be reduced at the same time.

Figure 1 .
Figure 1.Typical concrete paver blocks in the first author's plant.

Figure 1 .
Figure 1.Typical concrete paver blocks in the first author's plant.

Figure 3 .
Figure 3. Steel mold: (a) steel plates and bolts before assembly; (b) after assembly (side view); (c) after assembly (top view); (d) installation of the top steel block.

Figure 3 .
Figure 3. Steel mold: (a) steel plates and bolts before assembly; (b) after assembly (side view); (c) after assembly (top view); (d) installation of the top steel block.
P is the maximum load, L is the span length (150 mm), b is the width of the (100 mm), and t is the thickness of the block (60 mm).

Table 2 .
Compressive and flexural tensile strengths of paver blocks with different W/C ratios.

Table 3 .
Compressive and flexural tensile strengths of paver blocks with different forming pressures.

Table 4 .
Chloride and sulfate content of different sands.

Table 5 .
Compressive and flexural tensile strengths of blocks with different sands.

Table 6 .
Proportions of man-made seawater.
* The dosage of NaCl was increased to consider the chloride from MgCl2, CaCl2, SrCl2.

Table 6 .
Proportions of man-made seawater.The dosage of NaCl was increased to consider the chloride from MgCl 2 , CaCl 2 , SrCl 2 . *

Table 7 .
Compressive and flexural tensile strengths of blocks with different seawater replacement ratios.

Table 7 .
Compressive and flexural tensile strengths of blocks with different seawater replacement ratios.