Patent Application
20220002199
The present disclosure relates to the field of rheology-modified concrete and to the field of additive manufacturing.
3D concrete printing (3DCP) requires fine control of the rheological properties of cement composites, requiring a balance between high static yield stress for buildability and shape stability and low/moderate viscosity for extrudability and pumping. To achieve rheological control, is, however, difficult.
Static yield stress describes the material's resistance to flow and it is typically associated with a material's transition from solid to liquid. Viscosity, on the other hand, describes the material's resistance to flow under deformation and is associated with fluidity and pumpability. Cement colloidal forces via van der Waals and electrostatic forces and early hydration, e.g. C—S—H bridging, have been identified as the main interactions controlling such kinetics. Thus, most rheological additives affect both static yield stress and viscosity proportionally, and achieving the high static yield stress desired for 3DCP can result in excessively viscous, unpumpable composites. Accordingly, there is a long-felt need in the art for concrete systems (suitable for 3DCP) that exhibit desirable static yield stress as well as desirable viscosity.
In meeting the described long-felt needs, the present disclosure provides methods, comprising: combining a cementitious material, a cellulosic material, and a nanomaterial so as to give rise to a curable material, (i) the cellulosic material being combined with the nanomaterial before combination with the cementitious material, (ii) the cellulosic material being combined with the cementitious material before combination with the nanomaterial, (iii) the nanomaterial being combined with the cementitious material before combination with the cellulosic material, (iv) the cellulosic material, the nanomaterial, and the cementitious material being combined together, or any combination of (i), (ii), (iii), and (iv).
Also provided is a curable material made according to the present disclosure.
Further provided are methods, comprising: combining a cementitious material, a cellulosic material, and a nanomaterial to form a curable material, the method being performed such that the curable material exhibits at least one of: a pre-selected static yield stress, a pre-selected viscosity, a pre-selected heat of hydration, or pre-selected hydration kinetics, the cementitious material, the cellulosic material, or the nanomaterial being combined with another of the cementitious material, cellulosic material, and nanomaterial before being combined with the third of the cementitious material, cellulosic material, and nanomaterial.
Also disclosed are pre-mixes, comprising: a nanomaterial combined with a cellulosic material.
Additionally provided are methods, comprising combining a pre-mix according to the present disclosure with a cementitious material so as to give rise to a curable material.
Also disclosed are curable compositions, comprising: a cementitious material, a cellulosic material, and a nanomaterial, the cellulosic material and the nanomaterial being present in proportions such that the curable material exhibits at least one of: a pre-selected static yield stress, a pre-selected viscosity, a pre-selected heat of hydration, or pre-selected hydration kinetics.
Further provided are curable materials, comprising a cementitious material; a cellulosic material; and a nanomaterial.
Further provided are methods, comprising the use of a curable material according to the present disclosure in an additive manufacturing process.
The patent or application file contains at least one drawing executed in color. Copies of this patent or patent application publication with color drawing(s) will be provided by the Office upon request and payment of the necessary fee.
In the drawings, which are not necessarily drawn to scale, like numerals can describe similar components in different views. Like numerals having different letter suffixes may represent different instances of similar components. The drawings illustrate generally, by way of example, but not by way of limitation, various aspects discussed in the present document. In the drawings:
Solutions prepared by magnetic stirring lose their effect over time due to reagglomeration and NC falling out of suspension, while solutions prepared by sonication remain stable. Results show that dry dispersion leads to the most enhanced effect (increasing static yield stress by ˜170% compared to neat) and remains stable over 1 week.
The present disclosure may be understood more readily by reference to the following detailed description of desired embodiments and the examples included therein.
Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art. In case of conflict, the present document, including definitions, will control. Exemplary methods and materials are described below, although methods and materials similar or equivalent to those described herein can be used in practice or testing. All publications, patent applications, patents and other references mentioned herein are incorporated by reference in their entirety. The materials, methods, and examples disclosed herein are illustrative only and not intended to be limiting.
The singular forms “a,” “an,” and “the” include plural referents unless the context clearly dictates otherwise.
As used in the specification and in the claims, the term “comprising” can include the embodiments “consisting of” and “consisting essentially of” The terms “comprise(s),” “include(s),” “having,” “has,” “can,” “contain(s),” and variants thereof, as used herein, are intended to be open-ended transitional phrases, terms, or words that require the presence of the named ingredients/steps and permit the presence of other ingredients/steps. However, such description should be construed as also describing compositions or processes as “consisting of” and “consisting essentially of” the enumerated ingredients/steps, which allows the presence of only the named ingredients/steps, along with any impurities that might result therefrom, and excludes other ingredients/steps.
As used herein, the terms “about” and “at or about” mean that the amount or value in question can be the value designated some other value approximately or about the same. It is generally understood, as used herein, that it is the nominal value indicated ±10% variation unless otherwise indicated or inferred. The term is intended to convey that similar values promote equivalent results or effects recited in the claims. That is, it is understood that amounts, sizes, formulations, parameters, and other quantities and characteristics are not and need not be exact, but can be approximate and/or larger or smaller, as desired, reflecting tolerances, conversion factors, rounding off, measurement error and the like, and other factors known to those of skill in the art. In general, an amount, size, formulation, parameter or other quantity or characteristic is “about” or “approximate” whether or not expressly stated to be such. It is understood that where “about” is used before a quantitative value, the parameter also includes the specific quantitative value itself, unless specifically stated otherwise.
Unless indicated to the contrary, the numerical values should be understood to include numerical values which are the same when reduced to the same number of significant figures and numerical values which differ from the stated value by less than the experimental error of conventional measurement technique of the type described in the present application to determine the value.
All ranges disclosed herein are inclusive of the recited endpoint and independently of the endpoints, 2 grams and 10 grams, and all the intermediate values). The endpoints of the ranges and any values disclosed herein are not limited to the precise range or value; they are sufficiently imprecise to include values approximating these ranges and/or values.
As used herein, approximating language can be applied to modify any quantitative representation that can vary without resulting in a change in the basic function to which it is related. Accordingly, a value modified by a term or terms, such as “about” and “substantially,” may not be limited to the precise value specified, in some cases. In at least some instances, the approximating language can correspond to the precision of an instrument for measuring the value. The modifier “about” should also be considered as disclosing the range defined by the absolute values of the two endpoints. For example, the expression “from about 2 to about 4” also discloses the range “from 2 to 4.” The term “about” can refer to plus or minus 10% of the indicated number. For example, “about 10%” can indicate a range of 9% to 11%, and “about 1” can mean from 0.9-1.1. Other meanings of “about” can be apparent from the context, such as rounding off, so, for example “about 1” can also mean from 0.5 to 1.4. Further, the term “comprising” should be understood as having its open-ended meaning of “including,” but the term also includes the closed meaning of the term “consisting.” For example, a composition that comprises components A and B can be a composition that includes A, B, and other components, but can also be a composition made of A and B only. Any documents cited herein are incorporated by reference in their entireties for any and all purposes.
Rheological modifiers are a class of admixtures used to alter the rheological behavior of cement-based materials in the fresh state, namely yield stress, viscosity, shear thinning, or shear thickening. Clays are an example of a mineral rheological modifier that is commonly incorporated into cement-based systems to enhance shear thinning. And a special class of clays, nanoclays (NCs), have been found to be particularly effective in introducing high shear thinning and high yield stress development.
In 3D concrete printing, shear thinning and high yield stress development are highly desirable. However, studies have found that an increase in rate and magnitude of static yield stress and improved control of viscosity is needed. One may hypothesize that this may be addressed through enhanced dispersion. NCs exhibit very high specific surface area, which makes them prone to aggregation. Further, their aggregation behavior is highly sensitive to pH, and cement systems are highly alkaline. Therefore, to achieve sufficient rheological performance of 3D concrete printing with NCs, one must have improved dispersion and stability to enhance their effect. Here, our approach to this is to produce a hybrid admixture solution composed of NCs and methyl cellulose through different processing techniques, including a dry dispersion technique to coat cement particles (and dry methyl cellulose) with NC.
Although NC and MC are each utilized as concrete admixtures, there is no existing technique where they are processing together to form a hybrid admixture system prior to addition to the concrete mix (i.e., to the knowledge of the inventors, they are used as separate admixtures only). Common rheological modifiers induce coupled effects on the rheological properties, namely an increase in yield stress is coupled with an increase in viscosity. This coupled effect is not always desirable, especially for 3D concrete printing where a high yield stress is needed for shape stability, but viscosity must be controlled to prevent blocking during the pumping/printing process. The work described herein presents a hybrid mix of MC and NC, and our results are demonstrating that this hybridization allows for:
Decoupling of the rheological properties (i.e. decrease in viscosity and increase in yield stress, and vice versa) and ability to achieve both shear thinning and significant increase in yield stress (up to 5 times control)
Improvement of the efficiency, dispersion and suspension stability of the NCs to prolong the shelf life of the suspended admixture.
As a result, the rheological properties of 3D printing concrete (3DCP) mixes can be engineered to produce specific yield stress or viscosity to maximize print quality. Other techniques are also described to produce ready dry mixes with significantly longer shelf life (greater than several months) via dry dispersion. The following are example mixing methods:
One can extend the dry dispersion technique (1.) to other nanomaterials (i.e. graphene nanoplatelets, silica nanoparticles, alumina nanoparticles, and calcium carbonate nanoparticles) onto cement for other functionalities.
A comparison was performed to evaluate the effects of dry dispersion with sonication on the compressive and tensile strength of cement mortars at 1, 3, 7 and 28 days to test the strength development rates. Additionally, we looked at the application of NC and MC to improve the rheology and printing performance of magnesium oxide (MgO) binders. The hydration reaction of MgO produces brucite which is physically weak. Brucite can transform with CO2 into different magnesium carbonates that are physically strong.
Conventional casting techniques creates dense MgO elements that require high pressure and exposure to CO2 in order to enable carbonation and consequently gain physical strength. This, however, can be avoided by creating internal carbon delivery channels through 3D printing. In this investigation, we studied the rheology of MgO paste at 0.9 and 1.1 w/b ratio at different NC and MC dosages. We then selected a mixture of NC and MC at each w/b to produce similar static yield stress and tested the mechanical performance of printed 1 inch cylinders to compare the effects of 3D printing and infill patterns.
Type I/II ordinary Portland cement and distilled water are used in all cement pastes specimen. The water to cement ratio (w/c) is kept at 0.34 and the additives are added as weight replacement of cement. An exemplary, non-limiting methyl cellulose was purchased from Alfa Aesar and Sigma Aldrich in dry powder form. This type of MC has a molecular weight of 14,000 and viscosity of 15 cPs at 2.0% and 20° C. According to the manufacturer data sheet, MC exhibits a linear relationship between content and viscosity, as shown in
Both NC and MC are typically supplied in powder form, but solutions can be made or supplied. For NC, they can be either dry mixed (dm) with the cement as powder, or utilize dry dispersion (dd) which is a novel dispersion method that coats cement with NC particles. Dry mixing however is not an effective method to disperse NC. In solution, the most common method used for NC is magnetic stirring, but generally sonication utilizes significantly higher energy and it is the most common method to disperse nanomaterials. For MC, dry mixing is an effective tool since MC powder is micro sized and does not have the same agglomeration behavior as NC. MC solutions can be made with common dissolution processes. Because of the number of ways each additive can be produced, there are a number of ways a hybrid can be formed.
Dry Mixing and Dry Dispersion
Dry mixing is the simplest method that combines cement with NC or MC. This can be achieved using a hand mixer or a cement stationary mixer. In essence, this method utilizes typical mixing equipment used to make cement paste, mortar, or concrete. While this method can be used with both NC and MC, in some cases the methods does not always resolve agglomerates. However, this method is useful in dispersing MC with cement in contents up to 2.0 wt. % by mass of cement.
First, NCs (at a weight replacement of cement) are added to ethanol with 200 proof or higher. Any ethyl alcohol can be used in which cement can not produce any hydrate products and can maintain its liquid state at 40° C. Sonication is a high energy process and subsequent cooling of the solution and heat transfer of the alcohol are parameters to consider. The overall liquid to solid ratio of the ethanol: (cement+NC) should be maintained at 2 for contents up to 2.0 wt. %. For every additional 1 wt. % content, this ratio must be increased by an increment of 0.5. For example, at 3.0 and 4.0 wt. % NC content, the corresponding liquid to solid ratios are 2.5 and 3, respectively. This will ensure adequate heat transfer and energy transfer as NC produce significant colloidal forces that increase the static yield stress. The solution is then placed in an ice bath and cooled to 5° C. or lower. Sonication is then initiated using a probe sonicator operating at 500 W generating around 100 W in solution. The sonication is continuous until 10 kJ/g of NC is achieved. At that state, cement is introduced into the ethanol-NC solution while sonication is undergoing.
Once all cement is added, sonication is continued in 2 second pulses (2 sec on/2 sec off) until 30 kJ/g of NC is achieved. As soon as sonication is complete, the solution is placed in distillation which can be performed using a Bunsen burner, electric mantle or using a heated oil bath as shown in
Using the heated oil bath shown in
Once distillation is complete, the NC coated cement cake is transferred to the oven at 105° C. for 24 hours to ensure complete removal of ethanol and to prevent hydration with water vapor from air humidity. Photos of the cement before and after oven drying are shown in
Upon removal from the oven, the NC coated cement cake is ground using mortar and pestle or any other method that does not significantly impact particle gradation. The resulting NC coated cement is then ready for use and the dispersive state is maintained for months.
One way to create NC solutions is using magnetic stirring such as the one shown in
The most common dispersion method of nanomaterials is sonication, specifically using a sonication probe. In this method, the specific content of NC is added to water inside a stainless steel beaker and the sonication probe is inserted. Because sonication generates high energy, there is an associated increase in solution temperature. As a result, the solution may be cooled, e.g. using an ice bath. Sonication is performed continuously at an effective 100 W until 5,000 J/g of NC is achieved. One can use a larger solution to minimize any evaporation that can occur due to overheating.
Dissolution
MC is water soluble only at temperatures below 55-60° C. If MC powder is added to water directly, the outer shell of powder agglomerates would dissolve, preventing water to penetrate to the inside of the agglomerate, which prevents complete dissolution. Therefore, the first step is to heat water to 70-80° C. (above the MC dissolution temperature and below water evaporation temperature) and then add the specific content of MC powder gradually while magnetic stirring is underway at 200-800 rpm. Heating and stirring are continued for 30-60 minutes. The beaker can be covered to minimize any evaporation or spillage.
The solution is then immediately transferred to an ice bath to begin dissolution, where stirring is continued to prevent agglomeration until the solution temperature reaches 5° C. The solution during this process shifts from a white color to colorless, indicating complete dissolution. The solution is then sealed and left at room temperature to dissipate the air bubbles accumulated during the last stage, a process that can take up to 2 days without external aid.
Admixture Systems
Because there are a number of ways MC and NC can be dispersed and added to cement paste, there are a significant number of combinations to create hybrids. The order in which NC and MC are added can impact their effect on rheology:
NC powder dispersion (dd) with MC powder dry mixed with cement:
Dd is performed on cement first to disperse NC then MC powder is dry mixed with the cement. Water is added as free water.
NC powder dispersion (dd) with MC solution.
Dd is performed on cement first then MC is added as solution. Water can be either added as 100% of the mixing water or at lower percentage maintaining the total MC content in combination with some additional mixing water to meet the target water-to-cement ratio of the concrete mix. Without being bound to any particular theory, the addition of free water is preferable to maximize dissolution of cement; however, attention must be given to the viscosity of the MC solution as critically saturated MC solutions will form gels.
NC Solution with MC Powder Dry Mixed with Cement
MC powder is mixed with cement first then NC solution is added prepared via stirring or sonication. The NC solution can contain 100% of the water or free water in addition to high concentration of NC solution can be used. Solutions with high NC concentrations are more susceptible to deagglomeration.
NC Solution with MC Solution
Hybrid Solution
Dilute MC Solution in NC Solution
Dilution of one solution into the other can be done using magnetic stirring or sonication, as described elsewhere herein. Because sonication utilizes high energy that causes heat generation, which in turn can break the dissolution state of MC. This process is better using magnetic stirring or with short pulses of sonication such as 2 on/5 off or greater.
Dilute NC Solution in MC Solution
Disperse NC Powder in MC Solution
This process replaces distilled water used in magnetic stirring or sonication with MC solution. However, because sonication utilizes high energy that causes heat generation, which in turn can break the dissolution state of MC. This process is better using magnetic stirring or with short pulses of sonication such as 2 on/5 off or greater.
Disperse NC Prior to MC Dissolution
This process takes essentially a NC solution that is prepared using sonication or magnetic stirring as the base solution for MC dissolution described elsewhere herein. In this case, it may be useful to have NC well dispersed when MC powder is introduced into the solution.
Separate Solutions
Adding two solutions at two different stages to the cement. The content of MC and NC in water should be corrected to maintain the overall correct dosages. For example, for every 100 g of cement, 1 g of NC, 1 g of MC and 34 g of water are required. In terms of separate solutions and using 1/1 ratio of solutions, that translates to 1 g of NC in 17 g water and 1 g of MC in 17 g of water.
Dry Dispersion of NC onto MC
The process of dd described herein coats cement particles with NC particles. The same process can be used to coat MC powder with NC particles where the cement is substituted by MC powder. The resulting hybrid powder can then be dry mixed with the cement.
Example Results
To show the efficacy of the hybrid system, we show the results of each of NC and MC systems separately before showing the effect of hybridization. The results herein are from using w/c ratio of 0.34 unless noted otherwise.
The effect of method of dispersion on the static yield stress of 1 wt. % NC cement paste are shown in
We also compared the efficiency of NC based on the method of dispersion; magnetic stirring (mag), dry dispersion (dd) and dry mixing (dm). Since magnetic stirring and sonication show no differences immediately after solution synthesis, we utilized magnetic stirring to represent solution dispersion via sonication as well. Results of static yield stress,
The steady state viscosity was measured for the same cement systems and the results are shown in
For reference and comparative purposes, we present the effect of MC on cement rheological parameters, i.e. static yield stress, viscosity and storage modulus in
Rheology—Hybrid Solutions
To study the compatibility between NC and MC, we measured the static yield stress and viscosity. The hybrid mixes utilized NC and MC with contents between 0.1-1.5 wt. % NC and 0.1-2.0 wt. % MC. The hybrids were prepared by using magnetic stirring for NC and dry mixing for MC. This method utilizes the lowest dispersion energy and have been proven to be effective for MC and NC. We also compared this method to other methods and measured the stability of the dispersion by measuring the static yield stress immediately after dispersion, 1-hour after and 1-week (168 hours) after dispersion/synthesis.
Static Yield Stress
As described herein, the addition of NC increases the static yield stress of cement paste proportionally to NC content up to 4.0 wt. %. We also showed that at 0.1 and 0.5 wt. % MC there's a reduction in static yield stress, insignificant change at 1 and 1.5 wt. % and an increase at 2.0 wt. %. We examine the effect of NC addition at each of the previous MC contents and the results are shown in
Plastic Viscosity
Because the addition of MC increases viscosity, one may wish to characterize the changes in viscosity due to hybridization. This was captured by measuring the steady state viscosity and they are expressed in values normalized with respect to the Neat/reference cement paste in
Suspension
To examine the effect of method of synthesis on the rheological response of cement pastes, we tested for static yield stress immediately after synthesis, one hour after and one week after (168 hours) and the results are shown in
Examining the results immediately after dispersion in
The developed admixtures are primarily to control rheological properties. However, we also tested their effect on key concrete properties—hydration kinetics and strength.
The developed admixtures are primarily to control rheological properties. However, we also tested their effect on other properties—hydration kinetics and strength.
Calorimetry
In order to measure the effects of the admixtures on cement hydration, isothermal calorimetry was used to record heat of hydration of cement pastes. Specimen were prepared similarly to how they are prepared for rheology tests of hybrids where NC were added in solution via magnetic stirring, MC were added as powder and hybrids were prepared using a combination of the two. Because NC and MC are added as replacement of cement, some reduction in heat of hydration and delays are expected due to dilution effects that scale with the content. In these calorimetry results, the total heat of hydration (measured by the area under the curve up to 36 hours) is statistically indifferent. And we discuss the effect of NC, MC and combination on accelerating or decelerating hydration, which translates to faster or slower setting, respectively.
To identify the effects of NC only and MC only on hydration kinetics as a baseline, each of the two systems were examined at 0.5 wt. % content increments up to 1.5 wt. % NC and 2.0 wt. % MC. The results are shown in
Analysis of the previous results show that the effect of NC and MC on the characteristic hydration kinetics are linear with respect to the content of NC. Thus, in order to understand the effects of hybridization on the hydration kinetics two sets of hybrid systems were tested: i) maintaining the highest content of MC at 2.0 wt. % and varying NC content and ii) maintaining 1.5 wt. % NC content while varying MC content. The heat of hydration of each of these groups are presented in
Mechanical Properties
To test the effect of the rheological admixtures on the mechanical behavior of cement systems, the compressive and flexural strength of cement mortars prepared with NC only, MC only and hybrids of the two were measured at 7 and 28 days. MC were added as dry mixed powders and NC were added as solutions via magnetic stirring. The hybrids were prepared similarly by first dry mixing MC as powder with cement then adding NC as solution via magnetic stirring. In all mixes, the cement paste was prepared first then sand was added to create cement mortar with a sand to cement ratio of 2. For each mix, 6 cubes and 6 prisms were cast using standard 2×2×2 in and 1.5×1.5×6 in molds, respectively. The specimens were air-cured to simulate harsh curing conditions since large applications of 3D printing can't be water-cured. 3 specimens were tested at each age (7 days and 28 days) and the average was taken to be the representative value. Cubes were loaded in compression at 2,000 N/sec and prisms were loaded using three-point bending with a span of 5.5 inches and load rate of 200N/sec. All specimens were prepared using w/c ratio of 0.37 instead of 0.34.
The effect of NC and MC on the compressive strength are shown in
The addition of NC and MC reduces the compressive strength of cement mortar proportionally to their content by 3 MPa and 5.5 MPa per 1 wt. % content of NC and MC at 7 days strength, respectively. The addition of both NC and MC, however, shows an increase in strength from 7 to 28 days unlike that of Neat. Without being bound to any particular theory, this could be attributed to the improved water retention due to the additives water adsorption properties. Although the addition of both NC and MC results in a decrease in compressive strength at 28 days, the maximum decrease is at 19% from Neat. Furthermore, the decrease in compressive strength at 28 days stays proportional to NC content but is not affected by changing MC content. This could be attributed to the higher static yield stress and stiffness resulting in dry mixtures that are difficult to compact manually. Similar to compressive strength, a reduction in flexural strength is observed with the addition of only either MC or NC. Although some increase is observed between 7 and 28 days strength, only 2MC shows insignificant decrease from that of Neat. Nevertheless, the decrease at 28 days remains under 11% from that of Neat.
To test the effect of hybridization, we selected four hybrid mixes: 2MC0.5NC, 2MC1.5NC, 0.5MC1.5NC and 1MC1NC, shown in
Comparing 2MC1.5NC with 0.5MC1.5NC, reducing MC content from 2.0 to 1.5 wt. % shows no statistically significant difference in compressive strength. In fact, both 2MC1.5NC and 0.5MC1.5NC show no statistical difference with respect to 1NC and 2NC. This can be attributed to the effect of NC in mitigating the loss of heat of hydration, as discussed in the previous calorimetry section. The decrease in MC content from 2MC1.5NC to 0.5MC1.5NC on the other hand reflects positive increase in flexural strength at both 7 and 28 days. At the lower MC dosage of 0.5MC combined with 1.5NC, the flexural strength at both 7 and 28 days are statistically indifferent to that of Neat. On the other hand, the addition of 1NC to 1MC seems insufficient to reverse the adverse effects of 1MC on compressive strength as 1MC1NC shows no statistical difference from that of 1MC but reverses the decrease in flexural strength resulting in strength values that are similar to that of Neat. Examining the results at 28 days of all hybrids, one may discern that the effects of NC on the compressive strength development are proportional to NC content. Mixes containing 0.5 wt. % NC show the lowest increase followed by 1.0 wt. % NC then 1.5 wt. % NC. At 28 days, both 2MC1.5NC and 0.5MC1.5NC show statistically insignificant difference compared with Neat in regards to compressive strength. Furthermore, all hybrids show good flexural strengths comparable to that of Neat. The decrease observed at 2MC1.5NC represents only 15% reduction.
These mixes offer substantially improved rheological properties for 3D printing as well, especially 0.5MC1.5NC as it produces the maximum increase in static yield stress of 515% and only 80% increase in viscosity when compared with Neat. Compared to ordinary 1.5 NC mix, the static yield stress of the hybrid is 80% higher.
Dry Dispersion
As discussed, dd can produce the highest efficiency and longest stability of NC dispersions. Here, we examine the state of dry dispersed NC on cement particles through SEM and expand this method to other nanomaterials: graphene nanoplatelets, alumina nanoparticles, silica nanoparticles and calcium carbonate nanoparticles.
The effects of NC interactions on cement paste, especially its rheological behavior, are unique to NC. The degree of influence of the nanomaterial, targeted property, or added functionality are dependent on their physical and chemical properties. However, it is expected that for any nanomaterial type and application, enhanced dispersion leads to enhanced performance. Because of the geometry of nanomaterials, the surface area of few grams of nanomaterials when well-dispersed can exceed the surface area of cement. This can, in some instances, yield some level of clustering or agglomeration where multiple nanomaterials are present at the same location.
In order to ensure that the dispersion energy utilized in the description earlier is suitable for other nanomaterials, we tested the heat of hydration of cement paste at 0.46 w/b with 4 wt. % SNP at 15, 30 and 45 kJ/g for the step when the sonication is applied when the cement is added while maintaining 10 kJ/g for the first step. The results of heat of hydration are for all mixes are shown in
Nanoclays
SEM scans of nano-coated cements were collected for all the different nanomaterials.
As discussed, dd achieves the maximum increase in static yield stress per NC content for any dispersion method studied in this report. However, dd can also be argued to be of limited scalability if desired at high quantities, as processing can be lengthy and energy intensive. Therefore, we explored the potential of utilizing concentrated systems by mixing high content NC dd cement with plain cement (versus using NC dd cement only with a lower NC concentration) to reach a target NC content. That is, while maintaining the overall NC dosage by weight of cement, we examined the rheological properties of cement paste when NC are used to coat all cement particles versus when NC are used to coat only part of the cement content and the rest of the cement is left uncoated/untreated. To best test this, we examined producing 1, 2, 3 and 4 wt. % contents of NC by mixing 10, 20, 30 and 40% of cement with 10 wt. % NC coatings and the results are shown in
The results show that the effective dosage of NC, regardless of the percentage of NC dd cement amount used, maintains the same rheological properties or show further enhancement in static yield stress or elastic modulus when some of the cement is uncoated. Without being bound to any theory, these results suggest that dd can be utilized as a partial replacement of cement to minimize cost while maintaining the benefits of increased efficiency. It also allows utilizing different mixes of dd cements with other nanomaterials, such as the ones discussed in the following section.
Dry Dispersion Other than Cement
To negate the effects of dd on cement, the same dd process on cement can be used but replacing cement with MC to create hybrid nanomodified polymer powder. This results in nanocoated MC polymers such as the one shown in
Other Nanomaterials
Dd was used to coat cement with other nanomaterials: alumina nanoparticles (ANP), silica nanoparticles (SNP) and calcium carbonate nanoparticles (CCNP), which are 15-20 nm spherical nanoparticles, and graphene nanoplatelets (GNP) with 3-14 nm thickness and length and width <2 μm. These particles are often hard to uniformly disperse at high contents in cement paste due to their extreme small size and high rate of reagglomeration.
Furthermore, one of the most challenging nanomaterials to disperse with cement are GNP. To create a stable GNP solution suspension, the GNP must be functionalized, a dispersion agent must be used or, often, a combination of the two. Dd is unique it which non-functionalized GNP coat the surface of the cement. This offers the potential to create tailorable conductive cement products that can be used as smart self-sensing materials, in which the mechanical stress and damage can be evaluated by recording the changes in conductivity. It should be understood, however, that other conductive materials besides GNP can be used, e.g., carbon nanotubes, carbon black, other carbonaceous nanoparticles, and the like.
Electrical Conductivity
One of the most challenging nanomaterials to disperse with cement are GNP. To create a stable GNP solution suspension, the GNP must be functionalized, a dispersion agent must be used or, often, a combination of the two. Dd is unique it which non-functionalized GNP coat the surface of the cement. This offers the potential to create tailorable conductive cement products that can be used as smart self-sensing materials, in which the mechanical stress and damage can be evaluated by recording the changes in conductivity or capacitance. The increase in conductivity with GNP through dd has been measured with cement paste in its fresh state at as little as 0.1 wt. % dosage as shown in
Heat of Hydration
Both SNP and CCNP are nanoparticles that are used in cement applications to promote hydration and improve the mechanical performance. In order to check this, we looked at the heat of hydration via calorimetry of cement paste at 0.46 w/b ratio containing 4 and 10 wt. % replacement of cement and the results are shown in
Mechanical Performance
In order to measure whether the dispersion of nanomaterials on the surface of cement at high content result in different mechanical performance than when nanomaterials are dispersed in solution, we tested the compressive and tensile strength of mortars at 4 wt. % of SNP, CCNP and NC prepared via sonication and dry dispersion. The compressive strength was measured by the crushing of standard 2 in cubes at 2000 N/sec. The tensile strength was determined through split tension test (also known as Brazilian split tension) for 2 in diameter and 4 in height cylinders at 1.4 MPa/min. Sonicated solutions were used immediately after sonication and all nanomaterials were dispersed without any surfactants or stabilizers. Due to the high static yield stress of mixes with 4 wt. % NC, the w/b ratio examined was raised to 0.46 to produce workable slurries without the use of superplasticizers.
The compressive and tensile strength of mortars prepared at 4 wt. % SNP are shown in
Similarly to specimens prepared with SNP, specimens containing CCNP show some improvement in tensile strength at 28 days with insignificant differences in compressive strength. The tensile strength development rate with sonicated CCNP shows similar increase where the 28 days strength was reached at 7 days. All specimens prepared with sonication showed similar or improved mechanical performance regardless of testing age to that of Neat. On the other hand, some adverse effects on the mechanical performance are observed when CCNP are dispersed using dry dispersion. The compressive strength at 28 days however is similar to that of Neat and the decrease in tensile strength is lower than 10%.
Significant decrease in compressive strength is observed due to the addition of NC at 4 wt. % regardless of dispersion method. As indicated in the prior section, addition of NC at contents greater than 1 wt. % result in a reduction in compressive and tensile strengths. The decrease in compressive strength however is much more significant in sonication compared to dry dispersion. The decrease is attributed to the high water adsorption causing the mixture to be excessively stiff resulting in larger air voids and higher porosity. When preparing all mortars with NC, additional compaction was applied compared to the specimen with SNP and CCNP to reduce all voids. Because of the increased porosity due to the mix's dryness, the reduction in tensile strength could be considered critical at 50% compared to that of Neat at 28 days. On the other hand, the compressive strength using dd is only 13% lower than that of Neat. Thus, while the addition of NC is unfavorable for mechanical performance, addition of NC using dd shows better performance than with sonication which increase this method's desirability for concrete 3D printing. Furthermore, as explored in the previous section, the addition of MC to NC can mitigate the excessive dryness and can result in an increase in tensile strength.
Printing Performance
A 60 mL syringe gantry 3D printer is used to produce prints with cement paste incorporating 1.0 wt. % NC with 1 wt. % MC where NC are added as magnetically stirred solution and MC as dry mixed powder. The syringe has a 14-gauge dispensing straight stainless steel needle with an inner diameter of 1.6 mm. Printing layer height varies based on geometry and is in the range of 0.8 mm to 1.5 mm. The gantry speed is set to 800 mm/min during printing and a plunger is used to apply force for extrusion. The gantry system has a printing area of 250 mm×250 mm and vertical height of 100 mm. Printing codes are either manually generated or automatically using commercially available software; Simplify3D.
The printing performance of our proposed hybrid system is demonstrated here through a buildability test and producing multiple complex items, as shown in
Application of MC and NC to MgO Binders
In order to characterize the effects of the NC and MC as a rheological additive for 3D printing, we tested the mixture of NC and MC at different dosages for pastes prepared with MgO at 0.9 and 1.1 w/b ratios. Because of the finer particle size of the MgO powder, the w/b ratio needed for hydration is significantly higher than that of portland cement. As a result, we examined higher NC and MC dosages. To characterize the rheological properties, we examined the static yield stress, plastic viscosity and elastic modulus of fresh paste. We also examined the application of NC and MC on the print quality maintaining similar values of static yield stress.
Rheological Properties
MgO paste at 1.1 and 0.9 w/b ratios has a significantly low static yield stress of 2.4 and 10.4 Pa, respectively. In fact, at such low values, the MgO paste can be considered zero yield stress suspensions reflecting significantly weak colloidal network and forces. Addition of NC increases the static yield stress of MgO paste at 1.1 and 0.9 w/b ratios proportionally to NC content by 120 and 184 Pa per 1 wt. % NC, respectively as shown in
The increase in static yield stress due to the addition of NC causes significant stiffening of the colloidal structure leading to an increase in the elastic modulus proportional to NC content as shown in
The plastic viscosity of neat MgO pastes at 1.1 and 0.9 w/b ratios have 0.83 and 0.26 Pa·s values, respectively as shown in
Printing Performance
To test the corresponding NC and MC effects on print quality, a rotating spirograph structure shown in
After rheological characterization, a static yield stress of 360 Pa was chosen to meet the printing requirements to produce 1″×1″ cylinders. The corresponding NC and MC dosages to achieve the static yield stress for 1.1 and 0.9 w/b were 3 wt. % NC+1.5 wt. % MC and 1.75 wt. % NC+1.0 wt. % MC, respectively. Cast specimens were prepared with and without additives in 1″×1″ cylindrical molds and demolded after 24 hours. 3D printed specimens were printed using a syringe gantry system with layer height and width of 1.55 mm. A minimum of four specimens were prepared for each test and all specimens, cast and 3D printed, were covered in plastic for the first 24 hours to mitigate water evaporation. After 24 hours, carbon cured specimens were placed in a CO2 incubator at 20% CO2, 25° C. and 80±5% relative humidity (RH). Control specimens were cured at similar ambient conditions at 25° C., 80±5% (RH) and ambient CO2 (˜0.041%). Printing was performed via a syringe gantry printer with nozzle inner diameter of 1.55 mm and movement speed of 40 mm/sec. One open and one closed infill pattern were chosen to investigate the impact of printed CO2 delivery channels. The open infill provided four delivery channels whereas the closed infill was made of concentric circles and intended to mimic cast specimens. Additionally, capped open infill specimens where the top and bottom layers are replaced by closed infill layers were printed to investigate whether CO2 penetration of one layer thickness is sufficient to mimic fully exposed internal structure. A second open infill specimen set with six channels was introduced to examine the effect of exposed infill pattern on the compressive strength with similar capped and uncapped profiles. To ensure that any strength differences between open and closed infills are due to increased carbonation of the interior elements, the shells (without any infill) were printed and tested as well. The exposed infill shell has an average 2.85 mm thickness whereas the closed infill shell has an average 3.1 mm thickness. The differences in thickness are caused by different line overlap parameters to achieve the required geometries. Up to twelve specimens were prepared for each specimen type and the results reported are the average of minimum of four specimens. For sample sizes greater than four, sampling was held at one standard deviation from the population.
Cast-in and Role of Admixtures and Carbon Curing
This study was aimed towards examining whether enabling 3D printing of MgO cement through the utilization of NC and MC as rheological additives can increase CO2 penetration leading to an increase in compressive strength. As a result, the additives selected for printing at 1.1 and 0.9 w/b were chosen to maximize printing performance by reaching similar static yield stress. Thus, the dosage of NC and MC is not similar between both w/b ratios and they may impact compressive strength differently. To analyze the effect of admixtures on the compressive strength, cast specimens were prepared with and without admixture at their respective w/b ratios and tested for 3- and 28-days strengths as shown in
3D Printing Effects
3D printed specimens show significantly higher compressive strength than cast ones with similar mixture by up to 360% and 455% at 3 days and 380% and 390% at 28 days for 1.1 and 0.9 w/b ratios, respectively as shown in
Comparing 0.9 and 1.1 w/b ratios, 0.9 w/b specimens show overall higher compressive strength at 3 days whereas 1.1 w/b specimens show overall higher strength at 28 days (capped open infill are statistically indifferent). We suspect this to be due to difference in porosity between the two. Since all specimens are cured at 80% RH, water evaporation due to drying continues to occur for the duration of curing. Because specimens made with 1.1 w/b have higher water content, higher evaporation could be expected creating higher microstructural porosity increasing CO2 penetration and creating larger space for magnesium carbonates to form.
To further investigate the differences between 3D printed specimen, an additional open infill was introduced at 1.1 w/b and the shells of both open and closed infills were tested and the results are shown in
The mass of carbon-cured and ambient condition specimens was recorded at 1, 3 and 28 days. Since both carbonation and ambient conditions are controlled for similar temperature and relative humidity, water evaporation rate in both conditions should be similar before carbonation and will result in a decrease in mass over time while carbon curing can result in an increase in mass due to the formation of new products of magnesium carbonates. The change in mass of carbonated specimen measured then is the overall increase in mass due to carbonation and the reduction in mass due to water evaporation. The loss of water can be estimated by the change in mass of ambient condition. However, since carbonation can decrease surface porosity due to the formation of new magnesium carbonate products on the surface, carbonated specimens are likely to experience less water evaporation. Therefore, the analysis of change in mass shown in
The higher surface area of 3D printed specimens makes them susceptible to higher water evaporation as well as increased carbon penetration. However, results show that all 3D printed specimens show higher percentage increase in mass due to carbonation than cast specimen regardless of age and no significant differences between infill patterns are observed. At 3 days age, there are no significant differences in mass change between both w/b ratios. However, all specimens (printed and cast) show higher mass change at 1.1 w/b compared to 0.9 w/b at 28 days. These results support that higher w/b ratio can result in higher water evaporation increasing porosity and enhancing carbon penetration yielding higher compressive strength as discussed previously.
Additional Disclosure—Hybrid System
As described elsewhere herein, here, we combine NC with a water soluble viscosity modifying admixture (VMA) to increase cohesion, increase static yield stress and improve overall printing performance of cement composites. We further examine different ways of synthesizing our hybrid system and test their efficiency after 1 week of producing the admixture.
The NC used in this study are palygorskite or attapulgite clays supplied in highly purified powder form. They are 30 nm in diameter, 1.5-2.0 μm in length and carry a uniform negative charge along their length with positive charges at the ends. The VMA used is a soluble low molecular weight cellulose ether supplied commercially in powder form. Cement is type 1/II Portland cement and its chemical composition is shown in Table 2 below. The water to cement ratio (w/c) is kept at 0.34. Additions of NC and VMA are 0-2% by mass of cement are tested.
Both VMA and NC used in this study are supplied in powder form and can be added to the cement paste in either powder or solution form, where the latter is suspending or dissolving the material in water. To synthesize NC solution, magnetic stirring is one of the most commonly used methods and is suggested by the manufacturer. However, sonication remains the most common method to disperse nanomaterials, as it offers significantly higher energy than shear mixing. In this study, we compare all three methods; mixing NC as powder with cement and producing NC solution by either magnetic stirring or sonication. When magnetic stirring is used, the solution is mixed at 500 rpm for 1 hour. Sonication is performed using a sonicator probe at 300 Watts achieving 6,500 J/g of NC. On the other hand, a VMA solution can only be prepared via magnetically stirring the powder in high temperature water, then continuing stirring as the temperature drops to solubility levels.
To produce hybrids, we select magnetic stirring to prepare the NC solution and combine it with VMA via three methods. The first method, M1, is adding VMA as powder mixed with cement and adding NC in solution form. The other two methods M2 and M3 create hybrid solutions of both VMA and NC. In M2, two separate solutions are prepared and then the solutions are combined together. The resulting solution has half the original concentration of each of its constituents.
For example, to prepare 1 wt. % NC with 1 wt. % VMA, two separate solutions of 2 wt. % NC and 2 wt. % VMA are prepared, then the NC solution is added and mixed in the VMA solution. In the third method, M3, both NC and VMA are added as powder to one solution creating one hybrid solution.
The vane and cup setup is used to measure the effect of the additives on the rheological properties of cement paste. The protocol for producing cement paste is kept consistent between all specimen and fresh paste is prepared for every test. A pre-shear at strain rate of 260 sec−1 is applied for 1800 seconds to ensure all samples are at deflocculated state. At the end of the pre-shear, the steady-state viscosity is collected. A zero-stress condition is then applied for 300 seconds to allow structural build-up. A strain rate of 0.1 sec−1 is applied afterwards to measure the static yield stress where the material transitions from solid to fluid. A minimum of 3 tests are performed and averaged to quantify the rheological properties. All tests are performed at 25° C.
In order to examine the shelf-life effect of our hybrid system we test the static yield stress immediately after synthesis of the hybrid solution or dispersion of NC in solution, as well as exactly 1 week after (168 hours). Solutions are kept covered in controlled lab conditions (24° C.) with minimal handling. Similar mixing and rheological protocols are used for both specimens.
A 60 mL syringe gantry 3D printer, shown in
To study the effect of combining VMA with NC on cement paste rheology, NC contents of 0.1, 0.5, 1.0 and 1.5 wt. % are tested with VMA contents of 0, 1.0 and 2.0 wt. %. For this investigation, the simplest hybridization method of adding NC as solution and VMA as powder (referred to as M1 in
The addition of VMA to cement paste results in an increase in viscosity proportional to VMA content, as shown in
The increase in static yield stress of all hybrid mixes can be attributed to an increase in effectiveness of NC interactions, VMA interactions or new interactions between NC and VMA. Since changes in NC content does not affect viscosity while increasing VMA content increases viscosity, the change in viscosity can be associated with proportional change in VMA interactions. Comparing the change in static yield stress in
The static yield stress of cement pastes prepared with 1 wt. % NC alone and a hybrid of 1 wt. % NC+1 wt. % VMA using three dispersion/hybridization methods are compared. The results of static yield stress measured immediately after solution preparation and exactly 1 week after are summarized in
Since magnetic stirring shows a decay in performance and a common method used with NC, we utilize it in examining the three hybridization methods, as discussed. In the fresh solution state, M2 offers the highest increase in static yield stress. However, this method goes on to exhibit the largest loss of efficiency at 62% after 1 week. Because M2 combines two separate solutions to synthesize a new hybrid one, each constituent solution has higher content of additive in its original state. That is, to produce 1 wt. % NC and 1 wt. % VMA hybrid solution in M2, the constituent solutions each has 2 wt. % NC and 2 wt. % VMA. Maintaining dispersion at higher nanomaterials content is harder and the decay in dispersion of NC is worsened. Hybrid solutions via hybrid synthesis (M3) on the other hand show significantly higher stability, indicating that solubilizing VMA when NC is in the well dispersed state not only significantly drives static yield stress but further improves the stability of NC dispersion. This is evident when comparing the loss of 24% of M3 compared to 48% with NC dispersed via magnetic stirring. Finally, combining NC solution with VMA powder (M1) seems to offer a median performance both as a fresh solution and after 1 week. This method is the most suitable for scaling for industrial use as it utilizes the lowest energy, ease of processing and offer rheological properties suitable for 3DCP processes. If longer shelf-life is required, NC solutions prepared via sonication show no loss of performance after 1 week. Without being bound to any particular theory, the M1 method of synthesis utilizing sonication can provide a high stability and a high performance after 1 week.
The printing performance of our proposed hybrid system is demonstrated here through a buildability test and producing multiple complex items, as shown in
While not shown here, printing with NC system alone often results in clogging of the syringe due to segregation resulting in water bleeding. We acknowledge that such effects are due to the nature of the syringe extrusion system, which creates an uneven pressure profile within the syringe. However, such problems are uncommon in this study when utilizing the hybrid system. These results suggest that NC alone, despite having high static yield, may not be sufficient for syringe-based 3DCP.
Thus, the present disclosure provides, inter alia, a new hybrid additive system using nanoclays and VMA. We show that the hybrid system significantly outperforms NC or VMA alone in increasing the static yield stress, reaching almost an order of magnitude increase from the reference cement paste—918% increase with 1 wt. % NC with 2 wt. % VMA. We also show that while utilizing the VMA there is an increase in viscosity, although the increase is not proportional to the increase in static yield stress. Further, during 3D printing the use of VMA increases printing ink cohesion, reducing clogging and gaps in the print, as opposed to using NC alone. We also look at the stability of dispersed/synthesized solutions, comparing static yield stress using freshly prepared solutions versus week old solutions. Results show a decay in performance except when mixing NC as dry powder or when using sonication. While a decay in performance is observed in all hybrid mixes, it is essential to highlight that regardless of the method of synthesis, all hybrid mixes significantly outperform 1 wt. % NC alone, even after decay. The proposed hybrid system is used to produce a number of complex prints showing high level of control over the rheological properties, translating to high buildability, shape stability, extrudability and detail. The method of hybridization we suggest is to prepare NC solution via magnetic stirring or sonication and combine it with VMA, where the VMA is premixed in powder form with cement. Thus, the disclosed technology allows for control over rheology, static yield, and curing kinetics, allowing users unprecedented control and choice over these parameters and the ability to satisfy a broad range of use cases.
Additional Disclosure
Nanoclays (NC) can serve as thixotropy modifiers for fresh concretes, and show potential to meet the rheological demands of 3D concrete printing. Here, we propose a dry dispersion technique that produces NC-coated cement and compare to conventional methods of dispersion. Pastes incorporating NC were tested via shear rheology, scanning electron microscopy, and isothermal calorimetry. Dry dispersion was found to be the most effective method, where incorporating 4.0 wt. % NC increased the static yield stress and storage modulus of cement paste by 1500% and 550%, respectively, with a minimal increase in viscosity of 90%. Results of small amplitude oscillatory shear and isothermal calorimetry indicated NC can enhance fresh-state stiffening through seeding, although shear rheology results of kerosene-based cement systems indicated the increase in static yield stress by NC is mainly due to ionic forces. Finally, we discuss how these properties translate to high buildability and stable layer deposition.
Fresh concrete is a non-Newtonian fluid material with rheological properties that vary vastly with respect to chemical composition, particle gradation, environmental conditions, method of preparation and shear history. The interest in 3D concrete printing (3DCP) has generated significant demand for increased understanding of cement rheology and the role of admixtures. Roussel recently identified several rheological properties to be critical for 3DCP, namely static yield stress, dynamic yield stress, critical strain, viscosity, elastic modulus and structuration rate [1]. These rheological properties can be described by colloidal forces driven by Van der Waals attraction and electrostatic forces from adsorbed ions, and progression of cement hydration (e.g. calcium silicate hydrate (C—S—H) bridging) that originate in colloidal flocculation with characteristic time of few seconds [2-4]. The main intrinsic rheological properties resisting failure in 3DCP are static yield stress and structuration rate, as shown in Eq. (1) [1], for systems that do not rely solely on accelerated hardening:
τc(t)=Tc0+Athixt>ρgH/√{square root over (3)} Eq. (1)
where τc(t) is static yield stress at time t after deposition, τc0 is the static yield stress just after deposition, Athix is the structuration rate, p is the density, g is gravitational acceleration, and H is the total object height. Some more recent works indicated that this equation gives a rather conservative estimation of the desirable static yield stress for the total object height [5, 6]. Of course, other properties such as critical strain, elastic shear modulus and Young's modulus control other factors limiting layer width, height and velocity [1]. Static yield stress, structuration rate, critical strain, and elastic modulus can be further used to describe shape stability—the ability to maintain the deposited layer's shape within tolerable deformation.
In order to meet the high rheological demand of 3DCP, four approaches can be adopted. The first approach is to utilize a number of supplementary cementitious materials (SCMs), such as silica fume (SF) [7-14], fly ash or volcanic ash [9, 11-13, 15], or ground granulated blast furnace slag (GGBS) [11, 12, 14, 15]. The second approach is to rely on retarder/accelerator systems in which wet concrete supports only a few layers before it completely hardens [9, 16-18]. The third approach is to rely on chemical admixtures such as polycarboxylates (PCE) and high-range water reducing admixtures (HWRWA) [8, 9, 14], or nanomaterials such as purified alumino magnesium silicates (nanoclays (NC) or attapulgite clay) [7-9, 11, 14, 15, 19], nanosilica (NS) [10, 14] or nanobentonite [10, 13]. Of course, these approaches are not mutually exclusive and can be used simultaneously. The variation in cement additives and replacements combined with the scarcity of concrete 3D printers and lack of standards introduce large variations, creating challenges in fully characterizing their effect on printing performance. Furthermore, because cement rheology is shear history dependent, the vast configurations of extrusion and pumping systems in printers can cause additional variations in printing performance.
One of the main challenges faced in 3DCP additives is the coupling effect between rheological properties. For example, an increase in static yield stress and structuration rate is often coupled with an increase in viscosity, potentially compromising pumpability [20]. NC offers great potential for increasing static yield stress and rate of structuration [21-23] with minimal effects on viscosity [8, 24, 25], also described as exhibiting enhanced thixotropy, and have been shown extensively to improve the printing performance of cement-based composites in terms of buildability [9, 11, 14, 15, 19], shape stability [7, 8, 15], robustness (low variability in static yield stress) [7] and stiffness [19]. Highly purified NC are negatively charged uniformly along their length with high positive charges at their ends that produce a house of card effect in idle solutions [14, 24]. It is hypothesized that a similar mechanism is active within the pore solution in NC-cement composites that causes the increase in static yield stress [14, 24]. While such claims have not been verified, they will induce strong ionic interactions nonetheless.
As a nanomaterial, the effectiveness of NC is highly dependent on dosage and method of dispersion. Typical contents used in literature are in the range of 0.01-0.5 wt. % [7-9, 15, 19, 26, 27] while some authors used up to 1.0 [28], 1.2 [11], 2 [14, 25], 2.5 [24] and 3 [29] wt. %. [11, 14, 15] showed that NC has no significant effect on viscosity while [8, 24, 25] showed an increase in viscosity. Such discrepancies can be partially explained through differences in preshear conditions [23]. However, they are also a result of the variation in NC dispersion. For example, blending NC in water is the most common way of preparing NC-cement composites [8, 9, 14, 15, 19, 24, 27] while others blend NC dry with cement [7, 28, 29]. The blending time ranges from 1 to 5 minutes [15, 24, 27] or is unreported [8, 9, 14, 19]. The speed of blending also varies between 140 rpm (1 min) [24], 400 rpm (5 min) [9], and 12,000 rpm (2 min) [27]. Thus, the level of dispersion of NC can be a key reason behind discrepancies in literature. It should be noted that NC also comes in colloidal/liquid pre-dispersed form [30].
Quanji et al. studied the effect of NC dosage and showed that higher contents of NC result in higher static yield stress and degree of thixotropy up to 3.0 wt. % [29], but that increasing the dosage of NC beyond 1.3 wt. % resulted in a decreased rate of thixotropy. However, dispersion was not the focus of this work and their method of dry mixing NC with cement does not guarantee optimum dispersion. Parveen et al. reported that typical mixing processes employed for mixing in cementitious mortars are insufficient in producing uniform dispersion of nanomaterials such as CNTs [31]. On the other hand, sonication horns/probes are typically classified as the most effective method in deagglomerating and producing uniform dispersion of nanomaterials [32]. However, even when NC are well dispersed in water, achieving a state of uniform dispersion in solution does not guarantee uniform dispersion in the composite [31]. In fact, Yazdanbakhsh and Grasley showed through simulations that achieving uniform dispersion in cement composites requires homogeneous and deagglomerated cement particles [10], which cannot be achieved by adding NC solutions to cement gradually. In order to meet the high rheological demands of 3DCP, high dosage and high efficiency utilization of NC is desired.
In this disclosure, we provide a dry dispersion method to coat cement grains with NC, producing nanomodified cement. Although this method has been implemented for carbon-based nanomaterials in ceramics [33] and cements [34-37], to the knowledge of the authors it has not been implemented for nanoclays or other inorganic particles in cements. The influence of NC prepared via dry dispersion are compared with that prepared via other methods, i.e. dry mixing and magnetic stirring in solution. NC are incorporated into cement pastes and the pastes are tested for rheological properties, i.e. static yield stress, storage modulus, storage modulus evolution and viscosity, heat of hydration via isothermal calorimetry and scanning electron microscopy (SEM) imaging. We also examine kerosene-based NC cement systems to distinguish Van der Waals from ionic and electrostatic forces. Through the obtained results, we aim to expand upon Roussel's work on the origin of thixotropy to account for NC interactions, and break down the working mechanisms of NC in cement pastes. Finally, we discuss how the attained rheological properties with NC translate to high buildability potential and stable layer deposition with slenderness ratios of up to 10 for 3D concrete printing.
Motivation and Background
To fully characterize the kinetics of NC influence on cement rheology, we have used the characteristic soft and rigid interaction mechanisms described by Roussel et al. [2] and expanded it to include the effects of NC by introducing NC-cement and NC-NC interactions. In this work, we refer to all early hydration products including early ettringite formations as C—S—H bridge forces for simplicity. The origin and classification of rigid interactions warrants significant more in-depth analysis of the internal structure that are outside the scope of this work but are subject of future research. Our analysis of the origin of thixotropy of NC-cement paste is depicted in
Addition of NC into cement leads to three types of interactions: cement-cement, cement-NC and NC-NC, represented in 2D simplified schematics in
To better understand the relationship between cement-NC and cement-cement interactions, we can estimate the percentage of cement particle surface area that can be theoretically covered by well dispersed NC, % Acover, measured according to Eq.(2):
The specific surface area (SSA) is defined as the surface area per unit volume. Cement has an SSA around 1 m2/g [40] while NC similar to the one used in this paper have been measured as 107 m2/g using BET method [41]. Given that the solid density of NC is 2.287 g/cm3 [26] and the solid density of cement is 3.15 g/cm3, % Acover for the different NC dosages used is shown in Table 2. The first limit of % Acover represents the case where NC is sandwiched between two cement particles, while the second limit represents the case where NC covers the surface of cement on one side and is free on the other, effectively reducing the contact SSA by half. While the second limit seems more likely, it also assumes that all NC needles will be densely packed on the surface of the cement. However, because NC needles are negatively charged along their length, such dense packing is energetically unfavorable. Thus, the values of % Acover presented aim to provide limits to when all the surface of cement will be covered. In actuality, % Acover decreases due to NC and cement agglomeration and increases due to cement dissolution. Therefore, we consider our theoretical limit only at the beginning of dissolution processes and with effective dispersion.
The extreme dimensionality of NC is clearly reflected by the exceptionally high values of % Acover exceeding 100% within the tested NC dosage. In fact, within 1.27-2.54 wt. %, well dispersed NC have an effective surface area to cover the surface area of cement particles completely. Hence, if all cement particles are covered by NC particles, the interaction between two cement particles is insignificant to cement rheology, as they exceed 30 nm in spacing. This means that increasing the content above 1.27-2.54 wt. % in well dispersed states allows only for additional NC-NC interactions where cement-NC are maximized and soft cement-cement interactions are eliminated. However, due to nucleation and seeding effects of NC, some additional rigid cement-cement interactions can still occur. Because cement dissolution and hydration are such complex phenomena and the challenges in measuring the state of NC dispersion within cement paste at a few 10s of seconds after deflocculation, this theoretical model cannot be verified experimentally. However, this discussion highlights some key aspects on the effect of NC addition to soft colloidal cement interactions.
There is a theoretical limit at which nanomaterials in general can interact with cement. Beyond such limit, the main additional interactions within the system are nanomaterial interactions like NC-NC.
There is an exchange between soft cement-cement and NC-cement interactions due to the addition of NC given their geometry.
The state of NC dispersion plays a key role in controlling the mechanisms of NC interactions, as such has a direct influence on their effective SSA and in turn the dosage at which % Acover=100% is achieved.
Materials and Methods:
2.1 Materials:
Ordinary Type I cement and distilled water were used to prepare pastes. The chemical composition of the cement is provided in Table 3. Different levels of NC substitution by weight of cement were used in this study ranging from 0.5 to 4.0 wt. %. Table 4 summarizes all mixes with NC content based on their dispersion method. The prefix in NC mixes represents the content as weight replacement of cement and the suffixes are; “mag” referring to NC dispersed in water via magnetic stirring, “dd” referring to NC dispersed on cement via dry dispersion, and “dm” referring to NC that is added in the dry, as-received state during mixing. Neat refers to the reference paste that is unprocessed and Neatdd is reference cement that undergoes the dry dispersion process without NC presence. Neat, mag, dd and dm mixes were prepared using water-to-binder (w/b) ratio of 0.34 by mass. Additionally, in some mixes the water content was completely replaced with kerosene, a nonpolar solvent (dielectric constant ε=1.8) and absolute viscosity of 0.00164 Pa. s, as the liquid phase to diminish the influence of electrostatic force between particles and hydration. They are denoted with “_kero”. Kerosene mixes were prepared using similar mass ratio of 0.34. The cement-kerosene pastes stayed homogenous and no bleeding was observed during the tests. The negative and positive charges along NC lengths and ends introduce dipole-dipole forces in polar solvents such as water allowing dispersion without introducing additional chemical forces from surfactants. Because kerosene is non-polar, all kerosene systems used dd cements, as a stably dispersed NC suspension cannot be achieved in kerosene at the studied addition rate without introducing additional chemical forces. Because NC are added as replacement of cement, a dilution effect is expected decreasing cement-cement interactions and reducing the maximum heat of hydration similar to increasing the w/b from 0.34 to 0.354 at 4 wt. %.
2.2 Mixing:
Magnetic stirring (mag) was used to prepare a solution dispersion of NC in distilled water at 400 rpm for a minimum of 1 hour. A large solution of 350-400 mL was produced, which was always stirred immediately before use to prepare cement pastes. Dry mixing (dm) cement paste was prepared by using a hand mixer for 15 seconds to combine cement with NC powder, before adding distilled water. Dry dispersion (dd) is detailed in the following section.
Special care was taken to ensure all pastes had similar shear history leading up to rheological testing. To prepare mag cement paste, the NC solution was added to the cement, whereas in the case of dm or dd paste distilled water was added to the cement and NC composite. This process was done within 30 seconds and NC, cement and water were then mixed for 120 seconds at similar speed using a hand mixer. The highest mixable content of NC in solution was 3.0 wt. %. The paste was then loaded into the cup and vibrated for 5 seconds to remove air bubbles before loading into the rheometer. The vane was inserted 600 seconds from the time of contact between cement and water. A similar mixing approach was used for calorimetry but the paste was added to a glass vial and inserted in the calorimeter 300-330 seconds from the time of contact between cement and water.
2.3 Dry Dispersion:
The basics of the approach reported by [33-37] were further developed to ensure ease of repeatability and applicability to a diverse selection of nanomaterials. Isopropyl alcohol was replaced with ethyl alcohol, magnetic stirring was used in conjunction with probe sonication, distillation was used in place of desiccation or evaporation, and magnetic stirring was used through distillation. NC were first sonicated continuously in ethanol solution at 10 kJ/g NC while being simultaneously stirred using a magnetic stirrer. The energy used in this step is twice the one reported by [10] since dispersion in ethanol is more challenging than in water due to its lower polarity at relative polarity of 0.654. Ethanol was used instead of other alcohols due to its low boiling temperature and polarity. The solution was kept inside an ice bath to prevent excessive evaporation of the ethanol. Cement was then introduced to the solution maintaining continuous sonication and stirring. Sonication was applied to achieve 30 kJ/g NC after addition of cement at 2 second pulses. The NC-cement ethanol solution was then moved to a distillation apparatus to recover the ethanol. During distillation, the solution was also stirred to minimize sedimentation. 80% of the original ethanol was recovered within 30 minutes. The NC coated cement cake was then broken down and placed in a drying oven set at 260° F. (126° C.) for 24-48 hours to ensure complete removal of ethanol and to prevent hydration with air moisture. Upon removal of cement, the cement grains were further processed using mortar and pestle and kept in airtight bags until mixing with water to produce NC-cement pastes.
2.4 Rheological Protocol:
A four-blade vane and cup geometry were used in a stress-controlled HAAKE MARS III rheometer set at 25° C. The cup had an inner diameter of 26.6 mm, the vane blades were 21.9 mm wide and the gap between the vane and the cup was 8 mm. At least three different measurements were collected per mix, where additional tests of up to 7 tests per mix were carried out when variance was greater than 10%. These parameters ensure that results with means greater than 1±0.074 times the mean are different with 95% confidence interval which have been sufficient based on the authors experience when shear history was similar. Five mixes required additional tests namely 1NCmag, 1.5NCmag, 2NCmag, 2NCdm and Neatddkero. Shear history of 3DCP materials depend on the type of reservoir, pump, transportation and extrusion system of the printer. Thus, it is critical for rheological studies to examine the rheological properties of cement-based materials from a reference deflocculated state. High contents of additive that target high structural build up (as demanded by 3D printing processes) require significantly longer preshear to reach a deflocculated steady state [23]. Such is especially critical for NC, which are charged particles that significantly increase the rate of flocculation [27, 42, 43]. Thus, a relatively long preshear of 20 minutes was used in this study to ensure all tests were at a well-defined reference state. It should be noted that for NC prepared in solution via magnetic stirring, at high NC contents the cement paste was too stiff and no flow could be initiated by the rheometer, as it exceeded the equipment's maximum torque. Therefore, mixes containing 2.5 and 3.0 wt. % (2.5NCmag and 3.0NC mag) required manual initial torquing to initiate structural breakdown prior to applying pre-shear to protect the testing apparatus.
The rheological protocol is described in 59. It starts with a preshear, where the plastic viscosity was determined once steady-state was reached. The prolonged preshear was needed to ensure all mixes reached a steady state response before measuring the static yield response. Due to the high shear strain rate applied during preshear, the formation of hydration products was retarded. This is reflected with the downward sloping apparent viscosity evolution as shown in
Our protocol measured plastic viscosity, static yield stress, and storage modulus in series, in order to reduce the overall number of tests and to simulate the 3DCP process. In a general extrusion-based 3D concrete printing scheme, the fresh cement-based material is subjected to prolonged shearing during pumping and extrusion, then deposited, where the deposited layer is able to hold its shape if it exhibits sufficient green strength to sustain self-weight induced stresses and in some cases the weight of subsequent layers. The initial pre-shear simulates shearing from pumping and extrusion, from which we obtain corresponding viscosity. From there, static yield stress is measured to determine the capacity for shape stability shortly after deposition. And finally, the fresh material's stiffness and evolution over time will indicate the extent of elastic deformation of the layer, as well as overall structural stability, as printing continues. So although the static yield stress and storage modulus of a given paste sample will be measured at different shear histories, these parameters can be directly compared against different mix designs, which was the aim of this study.
2.5 Calorimetry
The hydration kinetics were investigated for the hydrating mixes (excluding kerosene system) at 25° C. using isothermal calorimetry (TAM Air III). A new paste was prepared for each test. 5 g of paste was loaded inside each standardized glass ampule, where a total of three tests were performed per mix. Data was collected for 48 hours and all samples generated a total of 185 kJ±3 kJ by 36 hours indicating the generation of a similar total amount of hydration products.
2.6 SEM Scans
Scanning electron microscope images were collected for NC powder and unhydrated 2.0NCdd powder, as well as 7-day air-cured 2.0NCmag and 2.0NCdd cement pastes. 2.0 wt. % content was selected as the median content examined in this paper. All samples were coated with 1 nm gold palladium (Au—Pd) via 108 Manual Sputter Coater. Scans were collected using Zeiss Sigma VP SEM with a resolution of 12 Å at 2-5 kV. Hydrated samples were obtained from fractured pieces of hardened cement samples that were produced using the same mixing approach as described prior.
2.7 Particle Size Analysis
The particle size distribution was measured using Beckman Coulter Laser Diffraction Particle Sizing Analyzer; LS 13 320. The device has a working range of 17 nm to 2000 μm and uses a sonicated aqueous submersion technique. This test was utilized to examine the effect of dry dispersion processing on the cement particle distribution. Plain cement was processed at similar energy to that used for 1NCdd and the difference in particle distribution before and after was recorded.
3. Results:
3.1 SEM:
The extreme dimensions of nanomaterials generate high surface energy as the number of atoms on the surface are higher than those inside, reaching upwards of 50% at 3 nm. As a result, nanomaterials such as NC tend to agglomerate in order to reduce their free surface energy and stabilize [31].
Typical dispersion methods are applied for nanomaterials in water so the dispersive state of NC can only be investigated in solution prior to combining with cement, or post hydration. Imaging post hydration however requires high resolution environmental SEMs and the ability to arrest hydration abruptly and rapidly. For example, Makar and Chan were able to utilize FEG-SEM and flash freezing using liquid nitrogen to examine growth of hydration products and the dispersive state of single walled CNTs [34]. Because dd coats cement with NC in the absence of water (and hydration products, as a result) scans such as the one shown in
Cement powder was tested before and after dd processes to examine the effect of sonication and distillation on the cement particle gradation and rheology. As a result of dd, particles sized in the 100-200 μm range completely disappeared in Neatdd, as shown in
The rheological responses of NC-cement pastes at different NC contents were measured for all three dispersion methods.
The results of plastic viscosity, which was measured as the steady-state viscosity, is shown in
Storage modulus measurements from the SAOS amplitude sweep are presented in
Storage modulus results show higher degree of sensitivity to dispersion energy compared to static yield stress, where the NC efficiencies are 3.68 and 6.9 times higher for mag and dd, respectively, compared to dm. Such findings suggest that higher dispersion energy could correspond to more uniformly dispersed, single NC needles enhancing degree of percolation and subsequent nucleation and rigidification, or serving as links themselves. It should be noted that this is subject to the limit of energy to achieve uniform dispersion without damaging the particles. The origin of the increase in storage modulus is further examined by looking at the evolution of the storage modulus and examining the hydration kinetics through calorimetry.
In addition to obtaining G′ from the amplitude sweep, we also monitored G′ evolution to obtain a measure of rate of stiffening. Applying Eqn (3), we focused our analysis on Grigid, i.e. rate of linear increase of G′ over time, which is considered to be a measure of rigidification due to the growth of early hydrates. The results are shown in
According to Roussel et al., there exist rigid and soft colloidal critical strains associated with the breakage of C—S—H links and colloidal network which are at the order of few hundredths % and few % strains, respectively [2]. At the order of few % strain and prior to yield, the stiffness of the percolated network is governed by soft colloidal interactions as rigid C—S—H links are ruptured past the rigid critical strain [2]. Nevertheless, the yield stress is still a function of the stress required for the breakage of both networks. Consequently, the macroscopic stress τmacro (the stress applied to the percolated network up to and within the yield stress) can be expressed as a product of the macroscopic strain γmacro in the order of few % and macroscopic elastic modulus G′macro; that is, τmacro=G′macro γmacro [2]. Hence, we measure G′macro within the linear regions of the stress-strain response where the strain is in the order of few % representing the stiffness of the soft colloidal network and the results are shown in
Similar to previous results, high G′macro sensitivity was observed with dd followed by mag then dm. With respect to each dispersion method on static yield stress, storage modulus and G′macro, NC efficiency showed higher sensitivity to properties that include soft interactions, i.e. static yield and G′macro, compared to only rigid interactions, i.e. SAOS. For example, the upper limit of increase in static yield stress for dd is 1500% compared to 550% in storage modulus. There is also a significant shift in the increase magnitude between the corresponding macroscopic stress at strain values in the 10−3-10−2 range compared to 10−5 to 10−4 (See the small box in
3.4 Rheology—Kerosene System
To further study colloidal interactions, we use dd NC-cement in kerosene solutions. This method of dispersion was used because NC are not dispersible in kerosene by using magnetic stirring and because dd is significantly more efficient than dm. Further, kerosene prevents hydration, eliminating the influence of any early hydrates and leaving only colloidal interactions. It is known that there are two main types of non-contact colloidal interaction existing in a cementitious system if we do not consider the steric hindrance induced by polymer additives: Van der Waals forces and electrostatic forces from adsorbed ions. By using kerosene as the suspending medium, we can diminish the influence of electrostatic force in cement paste. Kerosene is a nonpolar solvent, as opposed to water or isopropyl/ethyl alcohol, and hence we expect that electrostatic interactions are weak compared to aqueous phases and may be considered as negligible on the basis of solvation energy arguments [44]. We also looked at different NC contents up to 4.0 wt. %, measuring both static yield and storage modulus, as shown in
3.5 Calorimetry—Cement Paste Systems
In section 3.3, we discussed that the results of storage modulus suggest stronger rigid structure, either through linking of cements by early hydrates or the NC themselves. And the results of Grigid indicated NC dispersed by all methods led to an increase in rate of storage modulus evolution, although threshold levels were observed. To further characterize this effect and expand its relationship with hydration kinetics, we can analyze the results of isothermal calorimetry of dm, mag and dd shown in
NC have been suggested to have nucleation or seeding potential [14, 47, 48], which would result in a forward time shift in point #1 with an increase in NC content. NC-cement pastes prepared with dm and dd showed an insignificant difference at 1 wt. % and a forward shift (earlier) with increasing NC content up to 3.0 wt. % for point #1. Cement pastes prepared via mag showed a forward shift in peak #1 irrespective of NC content, statistically speaking. These results only agree marginally with the relationship between nucleation potential and expected shift in point #1. Thus, either the resolution of the isothermal calorimetry test was not enough to capture this phenomenon, or nucleation potential cannot fully explain the origin of how NC is increasing modulus. To clarify the contribution of hydration versus nanoclay on structure will require further investigation on build-up kinetics and is out of the scope of the present paper. Nevertheless, all NC cement pastes, regardless of their method of dispersion, show higher peaks of heat of hydration at the acceleration period in agreement with other researchers [7, 14, 29]. This also agrees with the results of storage modulus evolution (
Discussion
Our results showed an increase in static yield stress at significantly higher levels than storage modulus, 1500% compared to 550%, respectively, which suggests that the contribution of soft ionic forces due to addition of NC are greater than rigid forces. Our prior discussion in the background and motivation further shows that the main driver of the soft colloidal forces are NC-NC interactions. Since our results show higher NC efficiency for dd, mag and then dm in static yield stress, plastic viscosity, storage modulus and macroscopic elastic modulus, the state of NC dispersion must be critical to the number of NC-NC and NC-cement interactions. Dispersion of nanomaterials encompasses two main features: disentangling or breaking apart agglomerates and distributing nanomaterials uniformly within the medium. The prior can be characterized through imaging while the latter is examined through consistency and magnitude of measured behavior. Yazdanbakhsh and Z. Garsley concluded that uniform dispersion in cement composites requires deagglomerated cement particles [16]. Because cement most often is in an agglomerated state [3], uniform solution dispersion of nanomaterials cannot produce uniform dispersion in cement composites. Of the three dispersion methods investigated, dd is the only method where both NC and cement are deagglomerated at the same time due to sonication of both in ethanol. Therefore, we suggest that of the three investigated dispersion methods, dd has the most uniform dispersion with the highest number of deagglomerated or individual NC needles. Since cement pastes prepared via mag and dm utilize similar mixing energy to distribute NC within the cement matrix, the investment of energy to disentangle and deagglomerate NC is critical to maximize NC efficiency on the rheological response of cement paste.
3DCP remains in its initial stages of development, which includes significant developments in rheological models and additives, printers, extruders, pumps, printing properties and mechanical/structural performance. Similar to current construction methods, this technology should be applicable in pre-cast facilities as well as on site. The additive system used to achieve the rheological demand should be efficient, scalable, low maintenance and versatile. Currently, significant delays can be expected due to printing errors, weather or machinery related delays. For example, Diggs-McGee et al. showed that for the construction time of 5 days, the actual printing time is 14 hours and the elapsed printing-active time is 31 hours [49]. Nanomodified cement through dd offers a simple method of using NC cement on site and offers an extended shelf life, compared to solution dispersions that may segregate over time. This method also shows higher efficiency in increasing the static yield stress by up to 1500% with a maximum increase in viscosity of 90%. The stiffness of the resulting layers characterized by G′macro is also significantly higher than other methods, a key factor in buildability and shape stability [1].
The static yield stress can be directly applied to the maximum layer height possible according to Roussel: τy≥μgh0, where τy is the static yield stress, ρ is the density, g is the gravity and h0 is the individual layer height [1].
We examined various dispersion methods for NC (dry mixing, magnetic stirring in solution, and dry dispersion) and tested their impact on rheological and hydration behavior of cement pastes. Dry dispersion is a new dispersion method that successfully coats cement grains with NC needles. NC cement paste tests included rheological measurements (i.e. static yield stress, viscosity, and small amplitude oscillation shear), SEM imaging, and isothermal calorimetry. Our results show that there is a significantly higher increase in static yield stress, up to 1500%, than in storage modulus, limited to 550%, with the addition of ddNC up to 4%, and no increase in either of these parameters in the kerosene NC-cement systems. As a result, we suggest that soft colloidal interactions due to adsorbed ions play a more significant role in increasing the static yield stress than rigid interactions or Van der Waals forces. Within these soft adsorbed ionic forces, we discussed how NC-cement, NC-NC and cement-cement interactions are affected by the addition of NC. We also examined the effect of NC dispersion on rheology, and we discussed NC potential for 3DCP. Our results show:
Static yield stress and storage modulus increased from reference cement paste by up to 1500% and 550%, respectively. The increase was proportional to NC content up to 4% wt with minimal increase in viscosity by up to 90%.
The efficiency of NC in altering the rheological response of cement paste was higher for methods with higher dispersion energy: dry dispersion, magnetic stirring followed by dry mixing.
Heat of hydration during the acceleration period increased with NC content and dispersion energy.
It is possible to significantly reduce the required NC dosage for 3DCP by utilizing dry dispersion, which is a method that has no dispersion decay as it disperses on solid rather than in solution.
The higher NC efficiency in static yield stress compared to storage modulus observed in our results indicate that NC-NC soft interactions (specifically from adsorbed ions per our kerosene system investigation) are the main driver of static yield stress structuration.
NC increases C—S—H growth and potentially surface-based C—S—H nucleation, which corresponds to increased rigidification and stiffness.
The effect of NC on static yield stress, storage modulus and its storage modulus evolution can lead to high buildability and shape stability for 3DCP.
Aspects
The following Aspects are illustrative only and do not serve to limit the scope of the present disclosure or the appended claims.
Aspect 1. A method, comprising: combining a cementitious material, a cellulosic material, and a nanomaterial so as to give rise to a curable material, (i) the cellulosic material being combined with the nanomaterial before combination with the cementitious material, (ii) the cellulosic material being combined with the cementitious material before combination with the nanomaterial, (iii) the nanomaterial being combined with the cementitious material before combination with the cellulosic material, (iv) the cellulosic material, the nanomaterial, and the cementitious material being combined together, or any combination of (i), (ii), (iii), and (iv).
As described elsewhere herein, the method may be performed so as to give rise to a dry material. The method can also be performed in solution, e.g., in water or other solution. Without being bound to any particular theory or embodiment, combination can be effected by sonication, stirring/agitation, or any combination thereof. The components can be combined in any order, e.g., the cellulosic material can be combined with the nanomaterial, and the resultant combination can then be combined with the cementitious material. Any or all of the foregoing combining can be accomplished in the presence of water or other solvent, but this is not a requirement, as an or all of the foregoing can be accomplished in the absence of a solvent or with a solvent essentially absent (with water being an example such solvent), i.e., in a dry or nearly dry manner.
Aspect 2. The method of Aspect 1, wherein the cellulosic material is combined with the nanomaterial such that the nanomaterial is dispersed on the cellulosic material. The combination can be effected by, e.g., sonication, vibration, stirring/agitation, or any combination thereof. A solvent may be present, but this is not a requirement, as the combination can be effected in a dry or even nearly dry manner.
Aspect 3. The method of Aspect 1, wherein the nanomaterial is combined with the cementitious material such that the nanomaterial is dispersed on the cementitious material. The combination can be effected by, e.g., sonication, vibration, stirring/agitation, or any combination thereof. A solvent may be present, but this is not a requirement, as the combination can be effected in a dry or even nearly dry manner.
Aspect 4. The method of Aspect 1, wherein the cellulosic material is present as a solution before combination with the nanomaterial or the cementitious material.
Aspect 5. The method of Aspect 1, wherein the nanomaterial material is present as a solution before combination with the cellulosic material or the cementitious material.
Aspect 6. The method of Aspect 1, wherein the cellulosic material is combined with the nanomaterial before combination with the cementitious material.
Aspect 7. The method of Aspect 1, wherein the cellulosic material is combined with the cementitious material before combination with the nanomaterial.
Aspect 8. The method of Aspect 1, wherein the nanomaterial is combined with the cementitious material before combination with the cellulosic material.
Aspect 9. The method of Aspect 1, wherein the cementitious material comprises a hydraulic, calcium-based cement or MgO. Portland cement is one suitable material. The cementitious material can include one or more supplementary cementitious materials (e.g. fly ash, slag, silica fume) and alternative binders (i.e. reactive magnesium-based cements). It should be understood, however, that MgO can be utilized in addition to or even as a cementitious material. As discussed elsewhere herein, MgO can react with CO2 to give rise to carbonates that provide strength and also act as carbon sinks. MgO can be used, e.g., as a substitute for Portland cement. One or more geopolymers can also be present; a geopolymer can be, e.g., a compound or mixture of compounds consisting of repeating units such as silico-oxide (—Si—O—Si—O—), silico-aluminate (—Si—O—Al—O—), ferro-silico-aluminate (—Fe—O—Si—O—Al—O—) or alumino-phosphate (—Al—O—P—O—), created through a process of geopolymerization.
Aspect 10. The method of any one of Aspects 1-9, wherein the nanomaterial comprises a nanoclay. A nanoclay can be, e.g., a purified magnesium alumino-silicate, commonly known as palygorskite or attapulgite. Such materials can be needle-like in shape. Such materials can have a diameter of, e.g., 15-40 nm and length of, e.g., 1-3 μm. These needles can be charged, e.g., negatively charged along their length and positively charged at the ends. A nanoclay can be entirely one charge (e.g., positive charge, negative charge), but can also include regions of different charge.
Other nanomaterials (e.g., graphene nanoplatelets, silica nanoparticles, alumina nanoparticles, calcium carbonate nanoparticles, single- and multi-wall carbon nanotubes, and other nanoparticles) can also be used; such materials can be used with nanoclays, but can also be used in place of nanoclays. Without being bound to any particular theory or embodiment, a conductive nanomaterial (e.g., graphene, carbon nanotube) can be used. Such conductive materials allow for the formation of conductive structures whose conductivities (and attendant structural condition) can be monitored. A structure can be formed that includes one or more electrodes, which electrodes can be used to monitor electrical conductance within one or more regions of the structure.
Aspect 11. The method of any one of Aspects 1-10, wherein the cellulosic material comprises a cellulosic polymer. Such a polymer can have a methoxy substitution between 27.5-31.5 wt. % at a degree of substitution of 1.5-1.9.
Aspect 12. The method of any one of Aspects 1-11, wherein the cellulosic material defines a molecular weight in the range of from about 4,000 to about 140,000 (e.g., from about 4,000 to about 140,000, from about 10,000 to about 120,000, from about 20,000 to about 100,000, from about 35,000 to about 70,000. A molecular weight in the range of about 10,000 to about 18,000 can be especially suitable, e.g., for a range of tailorable behavior. Example cellulosic material and manufacturers include, e.g., Sigma-Aldrich, Spectrum Chemicals, BeanTown Chemical, and Acros Organics. Methyl cellulose is one example cellulose; other celluloses (e.g., cellulose ethers) can be used, e.g., methylhydroxyethyl cellulose (MHEC), ethylhydroxyethyl cellulose (EHEC), methylhydroxypropyl cellulose (MHPC), hydroxyethyl cellulose (HEC), hydrophobically modified hydroxyethyl cellulose (HMHEC), and the like. It should be understood that cellulose can be substituted by a variety of substituents, as methylcellulose is only illustrative form of cellulose that can be used. Likewise, the molecular weight of the methylcellulose used in the examples herein is illustrative only, as cellulose of varying molecular weights can be used. Likewise, the degree of substitution of cellulose can vary; the degree of substitution can be up to 3, although more typical values are 1.3-2.6.
Aspect 13. The method of any one of Aspects 1-12, wherein the cellulosic material is present at from about 0.1 to about 6 wt % (and all intermediate values and ranges) in the curable material. For example, the cellulosic material can be present at from 0.1 to about 6 wt % in the curable material, or from about 0.2 to about 5.5 wt %, or from about 0.5 to about 5 wt %, or from about 0.8 to about 5 wt %, or from about 1 to about 4.5 wt %, or from about 1.5 to about 4 wt %, or from about 2 to about 3.5 wt %, or even from about 2.5 to about 3 wt %.
Aspect 14. The method of any one of Aspects 1-13, wherein the nanomaterial is present at from about 0.1 to about 10 wt % (and all intermediate values and ranges) in the curable material. For example, the cellulosic material can be present at from 0.1 to about 10 wt % in the curable material, or from about 0.2 to about 9 wt %, or from about 0.5 to about 8 wt %, or from about 0.8 to about 7 wt %, or from about 1 to about 6 wt %, or from about 1.5 to about 5 wt %, or from about 2 to about 4 wt %, or even from about 2.5 to about 3 wt %.
Aspect 15. The method of any one of Aspects 1-14, wherein the curable material (e.g., as a cement paste phase) has a static yield stress of, e.g., from about 8 to about 10,000 Pa. The static yield stress can be measured by, e.g., measuring the cement paste phase using a cup and vane geometry at a given strain rate. The static yield stress can be, e.g., from about to about 10,000 Pa, or from about 10 to about 9,000 Pa, or from about 50 to about 7500 Pa, or from about 100 to about 5000 Pa, or from about 250 to about 4000 Pa, or from about 300 to about 3500 Pa, or from about 500 to about 2500 Pa.
Aspect 16. The method of any one of Aspects 1-15, wherein the curable material (e.g., as a cement paste phase) has a plastic viscosity of, e.g., from about 0.3 Pa·s to about 18 Pa·s, e.g., from about 0.3 to about 18 Pa·s, from about 0.5 to about 15 Pa·s, from about 1 to about 12 Pa·s, from about 2 to about 10 Pa·s, from about 3 to about 8 Pa·s, from about 5 to about 7 Pa·s.
Aspect 17. The method of any one of Aspects 1-16, further comprising dispensing an amount of the curable material in an additive manufacturing process.
Aspect 18. A curable material made according to any one of Aspects 1-16.
Aspect 19. A method, comprising: combining a cementitious material, a cellulosic material, and a nanomaterial to form a curable material, the method being performed such that the curable material optionally exhibits at least one of: a pre-selected static yield stress, a pre-selected viscosity, a pre-selected heat of hydration, or pre-selected hydration kinetics, the cementitious material, the cellulosic material, or the nanomaterial being combined with another of the cementitious material, cellulosic material, and nanomaterial before being combined with the third of the cementitious material, cellulosic material, and nanomaterial. As but some examples, one can arrive at a curable material having a static yield stress of, e.g., from about 8 to about 10,000 Pa. One can also arrive at a curable material having a viscosity of from about 0.3 to about 18 Pa·s. One can also arrive at a curable material having a desired set of hydration kinetics and/or a desired heat of hydration.
Aspect 20. The method of Aspect 19, wherein the cementitious material, the cellulosic material, or the nanomaterial is combined with another of the cementitious material, cellulosic material, and nanomaterial such that the cementitious material, the cellulosic material, or the nanomaterial is dispersed on the another of the cementitious material, cellulosic material, and nanomaterial.
Aspect 21. A pre-mix, comprising: a nanomaterial combined with a cellulosic material.
Aspect 22. The pre-mix of Aspect 21, wherein the nanomaterial is dispersed on the cellulosic material. A pre-mix can optionally comprise a cementitious material and/or MgO, as described elsewhere herein.
Aspect 23. A method, comprising combining the pre-mix of Aspect 21 or Aspect 22 with a cementitious material so as to give rise to a curable material.
Aspect 24. The method of Aspect 23, further comprising dispensing an amount of the curable material in an additive manufacturing process.
Aspect 25. The method of Aspect 24, further comprising forming at least part of a structure with the additive manufacturing process.
Aspect 26. A curable composition, comprising: a cementitious material, a cellulosic material, and a nanomaterial, the cellulosic material and the nanomaterial being present in proportions such that the curable material optionally exhibits at least one of: a pre-selected static yield stress, a pre-selected viscosity, a pre-selected heat of hydration, or pre-selected hydration kinetics.
Aspect 27. The curable composition of Aspect 26, wherein the nanomaterial comprises a nanoclay.
Aspect 28. The curable composition of Aspect 28, wherein the curable composition is essentially free of water. As described elsewhere herein, curable compositions can be formed by dry dispersion, which dry dispersion can in some instances include one or more of stirring, agitation, and sonication.
Aspect 29. A curable material, comprising: a cementitious material; a cellulosic material; and a nanomaterial. Suitable cementitious materials, cellulosic materials, and nanomaterials are described elsewhere herein.
Aspect 30. The curable material of Aspect 29, wherein the cementitious material comprises a hydraulic, calcium-based cement or MgO.
Aspect 31. The curable material of any one of Aspects 29-30, wherein the nanomaterial comprises a nanoclay.
Aspect 32. The curable material of any one of Aspects 29-31, wherein the cellulosic material comprises a cellulosic polymer.
Aspect 33. The curable material of any one of Aspects 29-32, wherein the cellulosic material defines a molecular weight in the range of from about 4,000 to about 140,000 (e.g., from about 4,000 to about 140,000, from about 10,000 to about 120,000, from about 20,000 to about 100,000, from about 35,000 to about 70,000. A molecular weight in the range of about 10,000 to about 18,000 can be especially suitable, e.g., for a range of tailorable behavior.
Aspect 34. The curable material of any one of Aspects 29-33, wherein the cellulosic material is present at from about 0.1 to about 6 wt % in the curable material (and all intermediate values and ranges) in the curable material. For example, the cellulosic material can be present at from 0.1 to about 6 wt % in the curable material, or from about 0.2 to about 5.5 wt %, or from about 0.5 to about 5 wt %, or from about 0.8 to about 5 wt %, or from about 1 to about 4.5 wt %, or from about 1.5 to about 4 wt %, or from about 2 to about 3.5 wt %, or even from about 2.5 to about 3 wt %.
Aspect 35. The curable material of any one of Aspects 29-34, wherein the nanomaterial is present at from about 0.1 to about 10 wt % (and all intermediate values and ranges) in the curable material. For example, the cellulosic material can be present at from 0.1 to about 10 wt % in the curable material, or from about 0.2 to about 9 wt %, or from about 0.5 to about 8 wt %, or from about 0.8 to about 7 wt %, or from about 1 to about 6 wt %, or from about 1.5 to about 5 wt %, or from about 2 to about 4 wt %, or even from about 2.5 to about 3 wt %) in the curable material.
Aspect 36. The curable material of any one of Aspects 29-35, curable material has a static yield stress (e.g., as a cement paste phase) of from about 8 to about 3,000 Pa (e.g., from about 8 to about 3000 Pa, from about 10 to about 2500 Pa, from about 50 to about 2000 Pa, from about 75 to about 1750 Pa, from about 100 to about 1500 Pa, from about 150 to about 1250 Pa, from about 200 to about 1200 Pa, from about 300 to about 1000 Pa, or even from about 500 to about 900 Pa) when measuring the cement paste phase.
Aspect 37. The curable material of any one of Aspects 29-37, wherein the curable material has a plastic viscosity (e.g., as a cement paste phase) of from about 0.3 Pa·s to about 18 Pa·s when measuring the cement paste phase.
The following references are listed only for the convenience of the reader; the inclusion of a reference is not any acknowledgment that the reference is material to the patentability of the disclosed technology.
The present application claims priority to and the benefit of U.S. patent application No. 63/047,430, “Nanomaterial And Cellulosic Rheology Modifiers For 3D Concrete Printing” (filed Jul. 2, 2020), the entirety of which application is incorporated herein by reference for any and all purposes.
This invention was made with government support under Contract No. 1653419 awarded by the National Science Foundation. The government has certain rights in the invention.
| Number | Date | Country | |
|---|---|---|---|
| 63047430 | Jul 2020 | US |
