3-1 Research Flow Chart
In this study, a systematic approach to determine the influence how the different steelmaking slags under the different operating conditions, including the rotating speed, liquid-to-solid ratio, particle size interact with the carbonation conversion rate is performed. This way, the best operating condition in engineering aspect is determined and the steelmaking slags before and after the carbonation are analyzed. Then the steelmaking slags are incorporated into the cement clinker at 5-15%. The tests include workability, strength, and durability. Finally, strength prediction models of mortar can be built by means of analyzing chemical composition and strength.
Figure 3-1 depicts the flow chart of this study, including carbonation process, cement replacement and strength prediction.
Objectives
To evaluate the performance for carbonation conversion by different steelmaking slags.
To investigate the properties of cement with partial replacement by different steelmaking slags.
To establish the strength prediction model of mortar replaced by different steelmaking slags.
Literature Review
Carbonation capture and utilization by mineral
Mass transfer in rotating packed bed
Cement chemistry
Figure 3-1 Research Flow Chart in This Study
3-2 Materials
3-2-1 Source of Feedstock
There are many units which can produce alkaline wastes in the process of steel making, such as slags produced by blast furnace, basic oxygen furnace, electric arc furnace and steel refining facility. However, only ground granulated air-cooled blast furnace slag and water-quenched blast furnace slags are allowed to be used as aggregates and pozzolanic materials in concrete according to Taiwan’s regulations.
Other slags are confronted with the restriction of legislation due to their high alkalinity and toxicity, which may cause creaking of concrete and environmental pollution. In this research, basic oxygen furnace slag, refining slag, electric arc furnace slag were chosen to be used as feedstock for carbonation. Through the carbonation process, the disadvantage properties of slags can be improved and the way to utilization in concrete may be clarified.
Materials used in this study were obtained from the sources below:
1. Basic oxygen furnace slag (BOFS) and refining slag (RFS) were obtained from China Steel Corporation (CSC) in Kaohsiung, Taiwan.
2. Electric arc furnace reduction slags (EAFRS) and oxidation slags (EAFOS) were obtained from Tung Ho Steel Enterprise Corporation in Miaoli County, Taiwan.
3. CO2 with high purity (99.5%) was purchased from Ching-Feng Gas Corporation, Taipei, Taiwan.
4. Mass flow controller of CO2 was purchased from Kaoduen Technology, Taiwan.
5. De-ionized water (DIW): Laboratory P.C. Chiang, Lab RB09, Graduate Institute of Environmental Engineering, National Taiwan University, conductivity = 18.24MΩ/cm.
3-2-2 Rotating Packed Bed (RPB)
The carbonation process was performed at the lab by a customized rotating packed bed to enhance the mass transfer efficiency. To promote the micro mixing of gas-stream and slurry and extend retention time, a maze-liked stainless steel packing has to be employed. The inner diameter, outer diameter and height of RPB in this study are 0.06, 0.205 and 0.04 meters respectively, and thus its volume is 1.2*10-3m3. Additionally, the rotating speed ranges from 0 to 1820, corresponding to 118.9 times of gravity.
The size and characteristics of the reactors are shown in Table 3-1
Table 3-1 Parameters of Rotating Packed Bed Used in This Study
3-2-3
Pretreatment of Steelmaking slags
The Slags in this study provided from China Steel and Tung Ho Corp. were first crushed by a crusher into smaller pieces. Secondly, mill the slags into particle sizes below 250 μm with a ball mill. Finally, separate the slags into 5 ranges of particle size with sieves. From coarse to fine are: 150-250, 106-150, 75-106, 53-75, 0-53 μm. The pretreatment procedure is listed in Figure 3-2.
Figure 3-2 Material Preparation of Steelmaking Slags
Items Units Value
Rotational speed rpm 0 ~ 1820
Gravity value G 0 ~ 118.9
Inner radius (D1) m 0.06
Outer radius (D2) m 0.205
Mean radius(Dmean) m 0.132
Axial height (ZB) m 0.04
Volume (VB) m3 1.2*10-3
3-3 Equipment
3-3-1 Thermal Gravimetric Analysis (TGA)
Thermal gravimetric analysis (TGA, STA 6000, PerkinElmer) is a method of thermal analysis where the mass of a sample is measured over time during the temperature changes. This measurement provides information about physical and chemical phenomena. The former concludes phase transitions, absorption and desorption. The latter contains chemisorption, thermal decomposition, and solid-gas reactions. Each component in the solid sample will be decomposed in different temperature scales, which are regarded as the unique fingerprint for the component.
Thus, the various substances can be determined by recognizing different decomposition temperature ranges.
Figure 3-3 TGA (STA 6000) used in this study
In this study, steelmaking slags after carbonation would be analyzed by calculating the weight loss from the thermal decomposition. Four zones of weight loss could be found in the curve of temperature to differential thermal gravity (DTG) in Figure,
correspond to: (1) 50℃ to 105℃; (2) 105℃ to 380℃; (3) 380℃ to 500℃; (4) 500℃
to 900℃.
Zone (1) is relative to moisture from the calcium sulphoalumiates and hexagonal tetracalcium aluminatehydrat (Marsh and Day, 1988).
Zone (2) is associated with the decomposition of organic components (Jo, Park et al., 2014).
Zone (3) is relative to pyrogenic decomposition of calcium hydroxide (Jo, Park et al., 2014).
Zone (4) is associated with the decomposition of calcium carbonation (Chen et al.,2011).
To be more precise, weight loss from 650℃ to 850 ℃ are used in calculation.
3-3-2 Scanning Electron Microscope (SEM)
A scanning electron microscope (SEM) is a type of electron microscope that produces images of a sample by scanning the surface with a focused beam of electrons.
The electrons interact with atoms in the sample, producing various signals that contain information about the sample's surface topography and composition. The morphology investigations of steelmaking slags and mortars were performed with a scanning electron microscope (SEM) (JIB-5410, JEOL) in this study. The surface topography,
distributions of the elements in samples were also detected by energy dispersive x-ray spectroscopy (EDX), which can identify CaCO3 on the surface of slag and cement.
Therefore, the performance of CaCO3 precipitation can be evaluated by comparing the SEM images of fresh and carbonated slags and mortars.
Figure 3-4 Scanning Electron Microscope (SEM)
Figure 3-5 Energy Dispersive X-ray Spectroscopy (EDX)
3-3-3 X-Ray Fluorescence (XRF)
X-ray fluorescence (XRF) is the emission of characteristic “secondary” (or fluorescent) X-rays from a material that has been excited by bombarding with high-energy X-rays or gamma rays. The phenomenon is widely used for elemental analysis and chemical analysis, particularly in the investigation of metals, glass, ceramics and building materials. The electrons in the inner layer can be excited by extra energy from the X-ray. After that, the outer layer electrons may transit to the inner layer to fill up vacancy and release a secondary X-rays (fluorescent) in the meanwhile. Different elements can be determined by the wavelength of fluorescent due to their specific energy bandgap. The obtained wavelength is calculated by Moseley's law:
𝜆 = 𝐾(𝑍 − 𝑠)-2 (3-1)
Additionally, abundance of each element can be obtained from energy of fluorescent by quantum theory of light:
𝐸 = ℎ𝜈 = ℎ𝑐/𝜆 (3-2)
3-3-4 X-ray Diffractometer (XRD)
X-ray diffractometer (XRD) is a method that analyzes the diffraction pattern to obtain the material composition, the structure or morphology of the atoms or molecules inside the material. Crystals can be used as gratings for X-rays. The coherent scattering produced by these large numbers of atoms (atoms, ions, or molecules) will interfere with the light and thus increase or decrease the intensity of scattered X-rays.
In energy dispersive analysis, the X-rays diffracted by the material sample are directed into a detector, which produces a "continuous" distribution of pulses, and the voltages of which are proportional to the incoming photon energies. This signal is processed by a multichannel analyzer (MCA), which produces an accumulating digital spectrum processed to obtain analytical data. In wavelength dispersive analysis, the X-rays diffracted by the material sample are directed into a diffraction grating monochromator. The adopted diffraction grating is usually a single crystal. By varying the angle of incidence and take-off on the crystal, a single X-ray wavelength can be selected. The wavelength obtained is given by Bragg's law:
𝑛 ∙ 𝜆 = 2𝑑 ∙ 𝑠𝑖𝑛(𝜃) (3-3)
Figure 3-6 X-ray Diffraction (XRD)
3-4 Methods
3-4-1 Carbonation Conversion Process
In the beginning of this study, all of the steelmaking slags had to be analyzed by X-ray fluorescence (XRF) and X-ray diffractometer (XRD) to understand their chemical
composition. After that, the experiment was designed and performed as the schematic set-up diagram in Figure 3-7.
All of the design of experimental factors in this study is shown in Figure 3-8. The pure CO2 steam flowed with a rate of 0.5 L/min and the slurry was pumped by peristaltic pump at the rate of 1L/min. All of the experiments were operated for 60 minutes. The operating parameters, including liquid to solid ratio (10 to 50 mL/g), rotating speed (700 to 1300 rpm), and particle size (32 to 160 μm), were evaluated on the performance of the carbonation reaction.
Figure 3-7 Schematic diagram of experimental set-up for carbonation in a RPB:
(1) CO2 Cylinder; (2) Gas flow controller; (3) High-gravity RPB; (4) Slurry; (5) Stirring and Heating Machine; (6) Peristaltic Pump.
Firstly, high purity (99.5%) carbon dioxide gas flowed into the RPB reactor and the gas flow rate was adjusted by the gas flow controller. Secondly, the steelmaking slags and tap water was completely mixed with different L/S ratio in a beaker to create slurry. Thirdly, the slurry was pumped into the RPB reactor by a peristaltic pump. After entering the RPB reactor, the slurry was sprayed on the inner layer of the packing and moved outward by the centrifugal force. In the meanwhile, CO2 flowed as a countercurrent dissolved and reacted in the slurry liquid drops. The reacted slurry was discharged from the bottom of the RPB back into the beaker to keep recycling, and the
carbonation, the slurry was filtered by a 0.45 μm filter paper and the filtered solid sample was dried in the oven (105℃, 24 hours) for the thermal gravimetric analysis (TGA) measurement to evaluate the carbonation conversion at different reaction time.
Figure 3-8 Designs of Experimental Factors in Carbonation Process
During the analyzing procedure of TGA, the samples were heated directly from 50 to 900 °C at the heating rate of 10°C /min under pure nitrogen gas at 19.8 mL/min.
After the samples were heated, a TG curve and DTG curve diagram could be obtained to provide the information on temperature at the maximum peak and other important peak parameter. By subtracting the weight loss from 650℃ to 850℃, where calcium carbonate was decomposed into carbon dioxide and calcium oxide, the amount of CO2
captured in the carbonation process could be obtained by eq. (3-4):
𝐶𝑂2(𝑤𝑡. %) = ∆𝑚𝐶𝑎𝐶𝑂3
𝑚105℃ × 100 (3-4)
Where CO2 (wt.%) means the amount of CO2 actually captured in the dry mass of each sample, ∆mCaCO3 is the weight loss fraction due to the decomposition of CaCO3, and m105℃ is the weight without moisture. Thus, the CO2 captured by slags could be expressed as the mass of CO2 captured per mass of slags, which is shown in Eq. (3-5).：
CO2uptake(%) = CO2(wt%)
100−CO2(wt%) (3-5)
Additionally, the theoretical mass fraction of CO2 capture achievable based on the composition of the fresh slags and stoichiometry was expressed as ThCO2, which is shown in Eq. (3-6) with the assumption that all of the CaO could be transformed to CaCO3.
ThCO2(mol/kg) = CaOtotal(kg/kg)
MWCaO(kg/mol) (3-6)
Where the MWCaO is the molecular weights of CaO (kg/mol), CaOtotal is the weight fraction of CaO of fresh slags analyzed by XRF. Consequently, the carbonation conversion yield could be estimated by Eq. (3-7):
𝜁Ca(%) =
CO2(wt%)
100−CO2(wt%)x 1 MWCO2(kg/mol)
THCO2(mol/kg) (3-7)
The carbonation conversion yield (ζCa (%)), was defined as the actually captured amounts of CO2 in dry mass of carbonated feedstock with the calculated theoretical extent of carbonation based on the reactive-oxide content of feedstock.
3-4-2 Properties of Cement Replacement
In this study, the steelmaking slags ware utilized as cementitious material with a substitution ratio of 5 to 15%. However, utilization of fresh slags as cementitious material may cause fatal structural damage due to the existence of free-CaO and alkali metal oxides which will expand as the formation of calcium carbonate (CaCO3) and alkaline-silica gel (Na2SiO32H2O). Thus, carbonation process was expected to be able to eliminate free-CaO and neutralized alkalinity to improve mechanical properties. In the experiment, the workability of cement pastes and mortars were first determined by the standard consistency, setting time, and flow test. Later, the best ratio of water to binding materials obtained from the workability test would be used for the strength test and the durability test. Figure 3-9 shows the flow chart of cement replacement experiment in this research.
Figure 3-9 Flow Chart of Cement Replacement Experiment in this Research
3-4-2-1 Normal Consistency
According to Chinese National Standards (CNS) 3590, normal consistency is intended to be used to determine the amount of water required to prepare hydraulic cement pastes with normal consistency. The experiment was carried out in a Vicat apparatus which is depicted in Figure 3-10. The testing procedure is listed below:
a. Mix 650 g of cement with a specific amount of water by the standard of CNS 3655 to produce a cement paste.
b. Form the cement paste into an approximate ball and toss it six times through a free Type of Steelmaking Slags
c. Fill the conical specimen mold with cement paste sample and remove the excess.
d. Place the specimen mold on a glass plate on its larger end.
e. Lower the plunger to the cement surface. Tighten the set-screw. Then set the movable indicator to the upper zero mark of the scale.
f. Release the set-screw to let the plunger sink into the paste naturally.
g. Note the reading after 30 seconds.
h. Repeat the procedure until the reading sets within 10 ± 1 mm.
i. Calculate the ratio of water needed to cement added in weight.
Figure 3-10 Vicat Apparatus for Normal Consistency Test
3-4-2-2 Setting Time
Setting time is essential to cement because it determines the available time for construction before the cement solidified. The experiment was also carried out in a Vicat apparatus but introduced the Vicat needle instead of the plunger. The procedure of
setting time test in this study was followed with the standard of CNS 786, which was listed as followed:
a. Prepare a cement paste and fill the specimen mold with it in the same way as the normal consistency (CNS 3590).
b. Rest the specimen in constant humidity environment for 30 min.
c. Perform the penetration test by lowering the needle of the rod until it rests on the surface of the cement paste.
d. Tighten the set-screw and set the indicator at the upper end of the scale.
e. Release the set-screw quickly to allow the Vicat Needle penetrate into the paste for 30 seconds.
f. Note the reading of penetration for every 15 minutes. Remember to make each penetration test at least 6 mm away from any previous penetration and at least 9 mm away from the inner side of the mold.
g. Record the results of all penetration tests and determine the time when a penetration of 25 mm is obtained by interpolation.
h. The time between the first contact of cement and water and the penetration of 25 mm is termed as initial setting time.
i. Determine the Vicat final time of setting end point to be the first penetration measurement that does not mark the specimen surface with a complete circular impression.
Figure 3-11 Vicat Apparatus for Setting Test.
3-4-2-3 Flow Test
Flow test is designed to determine the flow of hydraulic cement mortars, and of mortars containing cementitious materials other than hydraulic cements. The test was performed on the flow table as Figure 3-12, following the guide of CNS 1010 and CNS 15992.
a. The standard mortar shall be mixed by 1 time of binder, including Portland cement and steelmaking slags, 2.75 times of graded standard sand, and 0.485 times of water by weight. The quantities of materials to be mixed at one time in the batch of mortar for making six test specimens included 500g of cement with steelmaking slags, 1375g of sand, and 242 ml of water.
b. The mixture of mortars were added into the stirring machine and operated by following the standard of CNS 3655.
c. Place a layer of mortar about 25 mm in thickness in the mold and tamp 20 times with the tamper and repeat it again to fill up the mold.
d. Lift the mold away from the mortar 1 min after completing the mixing operation.
Immediately drop the flow table for 25 times in 15 seconds.
e. Measure the mean diameter of the mortar along the four lines scribed on the flow table.
g. Repeat the procedure until fitting the result within 110 ± 5 %
i. Calculate the ratio of water needed to cement added in weight.
Figure 3-12 Flow Table for Flow Test in this study
3-4-2-4 Compressive Strength
Compressive strength indicates the maximum pressure that the mortar is capable of bearing. The test was carried out via compression machine and the procedure was in conformance to CNS 1010.
a. Apply a thin coating of release agent to the interior faces of the mold (cubic of 50 mm) and non-absorptive base plates and assemble the mold.
b. Prepare the mortar as the same way in the flow test (CNS 15992).
c. Fill the mold with mortar of 25 mm in thickness. Tamp the mortar in each cube compartment 32 times in about 10 s in 4 rounds, each round to be at right angles to the other and consisting of eight adjoining strokes over the surface of the specimen.
d. Repeat procedure c. again to fill up the mold.
e. Immediately after molding, keep all test specimens in the molds in the moist closet or moist room for 20 to 24 hours with their upper surfaces exposed to the moist air.
f. Transform the specimens into saturated limewater immediately after removal from the moist closet. Prepare for compressive test by curing as age of 3, 7, 28 days.
g. Wipe out the water on the mortar specimen and smooth the uneven parts.
h. Adding compressive stress on the specimen until it collapsed. The process should be done in 20 to 80 seconds.
i. Record the data and calculate the average compressive strength of mortar specimen.
3-4-2-5 Autoclave expansion
The autoclave expansion test provides an index of potential delayed expansion caused by the hydration of CaO or MgO, or both, when present in hydraulic cement.
The test was performed in autoclave following the guide of CNS 1258:
a. Prepare the cement paste in the same way to standard consistency.
b. Mold the test specimen in two approximately equal layers, each layer being compacted with the thumbs or forefingers by pressing the paste into the corners, around the gauge studs, and along the surface of the mold until a homogeneous
c. Compact the top layer, cut off the paste flush with the top of the mold with a thin-edged trowel, and smooth the surface with a few strokes of the flat trowel.
d. At 24 h ±30 min after molding, remove the specimens from the moist atmosphere.
e. Immediately obtain a length comparator reading for each specimen, and place in the autoclave at room temperature in a rack. To maintain an atmosphere of saturated steam vapor, the autoclave shall contain enough water, at an initial temperature of 20 to 28℃.
f. Close the valve and raise the temperature of the autoclave at a rate that will bring the gauge pressure of the steam to 2 MPa in 45 to 75 minutes from the time the heat is turned on.
g. Maintain the 2 ± 0.07 MPa pressure for 3 hours.
h. At the end of the 3-hour period, shut off the heat supply and cool the autoclave at such a rate that the pressure will be less than 0.07 MPa at the end of 90 minutes.
i. At the end of the 90 minutes period, slowly release any remaining pressure by partially opening the vent valve until atmospheric pressure is attained.
j. Then open the autoclave and place the test specimen in water at a temperature above 90°C.
k. Maintain the water surrounding the specimens at 23°C for an additional 15 minutes.
l. Remove one specimen at a time from the water blot the pins, but not the specimen,
l. Remove one specimen at a time from the water blot the pins, but not the specimen,