CONTENTS
1
CONTENTS
2
CONTENTS 1.
2.
SCOPE AND SIGNIFICANCE OF THE RESEARCH
5
1.1
Introduction
5
1.2
Objectives of the Research
7
BACKGROUND
8
2.1
Introduction
8
2.2
Concrete
11
2.3
Composition of Concrete
11
2.3.1
Cement
12
2.3.2
Water
14
2.3.3
Aggregates
15
2.3.4
Additions
16
2.3.5
Admixtures
16
2.4 2.5
2.6
3.
Hydration of the Cement
17
2.4.1
21
Heat of Hydration
Setting
23
2.5.1
Setting Phenomenological point of view
24
2.5.2
Chemical point of view
25
2.5.3
Structural point of view
25
Influence of Laboratory Conditions in the Setting Time
26
2.6.1
Influence of atmosphere temperature
26
2.6.2
Influence of relative humidity
27
TEST PROGRAM
28
3.1
Laboratory and Materials
28
3.1.1
Laboratory
28
3.1.2
Water
29
3.1.3
Sand
29
3.1.4
Cement
30
3.1.5
Vacuum Procedure
30
3.1.6
Composition of the mortar
33
3.1.7
Mixing of mortar
33
3.1.8
Series
34
3.1.9
Nomenclature
35
3.2
3.3
Experimental techniques
36
3.2.1
Penetration Resistance Test (Penetrometer)
36
3.2.2
Langavant-type Semi-Adiabatic Calorimetric Test
38
3.2.3
Ultrasonic test: P-wave Transmission Velocity
48
Previous results
52 CONTENTS
3
3.4 4.
Test schedule
54
RESULTS
57
4.1
Introduction
57
4.2
Experimental Data Treatment
58
4.2.1
The Polynomial Approximation
60
4.2.2
The Cubic Splines Approximation
61
4.3
Penetration Resistance Test: Penetrometer
62
4.3.1
Series 1: Open Storage Cement
64
4.3.2
Series 2: Vacuum Packed Cement
67
4.3.3
Comparison between Open Storage
70
and Vacuum Packed Series
4.4
73
4.3.5
Vacuum Efficiency
74
Ultrasonic test: P-wave Transmission Velocity
76
4.4.1
Series 1: Open Storage Cement
79
4.4.2
Series 2: Vacuum Packed Cement
85 89
4.4.4
Comparison with the Results of Previous Researches
91
4.4.5
Vacuum Efficiency
93
Semiadiabatic Calorimetric Test
95
4.5.1
Series 1: Open Storage Cement
96
4.5.2
Series 2: Vacuum Packed Cement
105
4.5.3
Comparison between the Series
111
4.5.4
Comparison with the Results of Previous Researches
113
4.5.5
Vacuum Efficiency
114
CORRELATION
116
5.1
Procedure
117
5.2
Correlation with regard to IST
118
5.2.1
Penetrometer vs P-wave Transmission test
118
5.2.2
Penetrometer vs Semi-adiabatic Calorimeter test
120
5.3
6.
Comparison with the Results of Previous Researches
4.4.3 Comparison between Open Storage and Vacuum Series
4.5
5.
4.3.4
Correlation with regard to FST
122
5.3.1
Penetrometer vs P-wave Transmission test
123
5.3.2
Penetrometer vs Semi-adiabatic Calorimeter test
124
CONCLUSIONS AND FUTURE LINES OF WORK
128
6.1
Response to the Research Objectives
128
6.2
Future Lines of Work
7.
ANNEX
131
8.
BIBLIOGRAPHY
134
9.
LIST OF FIGURES
137
10.
LIST OF TABLES
143
CONTENTS
4
1.
SCOPE AND THE RESEA RCH
SIGNIFICANCE
OF
1.1 Introduction The present thesis is one of the constituent parts of a broader research that it is being carried out by the Belgium research centers of University of Leuven, University of Gent and The Royal Military Academy of Brussels, in whose construction laboratory and under its supervision this thesis has been carried out. The aim of the cited framework research is to study the rheology of the Self Compacted Concrete (SCC) made of different compositions (superplasticizers, accelerators, retarders and fly ash) and what the characteristics of the fluidity of the mixture are. In practical-on-site meaning: how the SCC mixture is going to flow and how much force will be required for example when it is pumped. Besides, the changes in the mixture fluidity are produced according to the hydration development: when hydration develops, the fluidity comes less and less. Moreover, the objective
of the
wider research is
due to
the
fact
that
superplasticizer’s (SP) sellers do not present the exact SP composition when they commercialize them. Because of this, any hydration study of a SCC mixture can be totally trustworthy due to the lack of knowledge on the exact mixture composition. In response to this, University of Leuven has created its own SP in order to carry out a full research on the SCC hydration behavior together with University of Gent and the Royal Military Academy of Brussels. The two universities aforementioned are in charge of doing the pure rheology tests, where an experimental technique is used to monitor the required force to make a turn in the SCC fresh mixture. Therefore, the necessary energy to do these turns will provide a measure of the rheology of the testing mixture. Unfortunately, when the fresh mixture keeps being turned, the connection between the components’ grains, which are being developed according to the hydration evolution, can be broken and the concrete’s internal structure does not develop as in real on-site-placement. Because the rheology tests are not monitoring a real situation, real pure mixture hydration tests should been also carried out and then putting next to the rheological ones, to achieve real situations results of the SCC mixtures hydration behavior.
1. SCOPE AND SIGNIFICANCE OF THE RESEARCH
5
Therefore, this present thesis, which is titled as “MONITORING OF THE CEMENT HYDRATION BEHAVIOR AND DETERMINATION OF NON-STANDARIZED LABORATORY INDICATORS OF SETTING TIME”, will study the hydration evolution of fresh standard mortar mixtures taking as main key references the Initial and Final setting times. In this way, the results of this thesis will be added to the rheological results obtained by the aforementioned universities. Moreover, Initial and Final setting times have been chosen as main reference points of comparison because both of them are key points in the hydration evolution of a concrete mixture. In addition, there is just ONE standard experimental technique to obtain the Initial and Final setting times of a mortar fresh mixture. This technique is based on the penetration resistance of the mixture and it is carried out with the Penetrometer test, standardized by ASTM C403 (Standard Test Method for Time of Setting of Concrete Mixtures by Penetration
Resistance) (1). VICAT needle test, standardized in EN 196-3 (Methods of testing cement-Part 3:
Determination of setting times and soundness) (2) is another experimental technique that determines the setting time based also on the penetration resistance of the mixture. The difference relies on the fact that VICAT standard requires a mixture made of cement paste instead of cement mortar. It is true that hydration mainly depends on the reactions between cement and water (cement paste), but cement mortar made of water, cement and also sand resembles much better the real conditions of the evolution of the concrete setting. Apart from this, this thesis requires to test on cement mortar instead of cement paste, as this study is set in a framework of the study of the rheology of different SCC mixtures. As described before, there is just one standard test to obtain Initial and Final setting time of a mortar mixture: Penetrometer test.
Nevertheless, this experimental technique
strongly depends on the technician who executes the test and thus the results could bring the possible human error. Moreover, this test does not provide any information about the hydration reactions or the changes on the microstructure of the mixture. In response to this, this study proposes two other experimental techniques which monitor the mixture hydration evolution and therefore could provide the time of setting. The first one is the Ultrasonic test of velocity of the p-wave and the Semi-adiabatic Calorimetric test, which is standardized in EN 196-9 (Methods of testing cement – Part 9:
Heat of hydration –Semi-adiabatic method) (3). None of them counts with standardized indicators to obtain the First and/or Final setting times but as each of them monitor the hydration development according different variables, setting time references can be sought.
1. SCOPE AND SIGNIFICANCE OF THE RESEARCH
6
Finally, the efficiency of the cement vacuum packaging is going to be taken under consideration after some studies which suggest its storage effectiveness in relation to the open storage cement. If after opening a new sack of cement, the remain cement which is not going to be used in the correspondent test, stays in the cement sack until the next test day, this cement will be in contact to the atmosphere during all that time. This means that until the next test day (which could be the following day or the following week), the particles of cement can start to react with the water vapor contained in the atmosphere and it can make the cement less reactive in the process of hydration.
1.2 Objectives of the Research As introduced before, this research aims to: 1. Monitor the hydration evolution of standard mortar mixtures composed of vacuum packaged cement or open storage cement and with different cement ages. 2. Obtain the effectiveness of the cement vacuum packaging in relation to the open storage in terms of setting and hydration key references. 3. Study the correlation between the Ultrasonic test and Semi-adiabatic Calorimetric test to the Penetrometer test in relation to their indicators of Initial and Final setting times. 4. Consider the suitability of the setting indicators from the bibliography for the two nonstandard techniques (Ultrasonic and Semi-adiabatic Calorimetric tests) and relate them to the Penetrometer standard indicators. 5. Suggest new setting indicators in the case that none of the bibliography indicators adjust to the Penetrometer sensibility.
.
1. SCOPE AND SIGNIFICANCE OF THE RESEARCH
7
2.
BACKGROUND 2.1 Introduction Setting of concrete is the gradual transition from the initial liquid-like state to a
load-bearing solid material. Traditionally, two points during setting are considered important for concrete practice. Initial Setting Time (IST) is of importance, as it provides an estimation of when the concrete has reached the point where it has stiffened to such an extent where plasticity and workability are lost and it can no longer be consolidated without damaging the concrete. On the other hand, Final Setting Time (FST) of concrete relates to the point where stresses and stiffness start to develop and the mixture becomes rigid and a solid material. The setting time is extremely important for the construction process. The time of initial set gives the contractors an indication of how long the concrete can be cast, compacted and finished, while the final setting time provides an estimation of the beginning of the concrete strength development and it is used to schedule removal of forms or reshoring, backfilling walls, pretressing and post-tensioning or determining the time for opening the pavements to traffic (4). The setting process is influenced by the rheology effects of the water-cement ratio, aggregates, air voids, bleeding and evaporation but it is primarily influenced by hydration of the cement (5). Cement Hydration is a process of irreversible chemical reactions between cement and water, with the formation of new crystals, and where the cement-water paste sets and hardens. During cement hydration, products of aluminate reactions contribute to early stiffening of concrete; calcium sulfate controls the early aluminate reactions; and products of silicate reactions contribute to concrete’s strength. It has been estimated that on an average 23% of water by weight of cement is required for chemical reaction with Portland cement compounds (6). But what happens if, before adding the water, cement is exposed to the atmosphere for a certain period of time? The consequence is that cement can absorb moisture from the atmospheric air or any other source and react with it. The binding property and strength of cement depend upon its capacity for chemical reaction, which can take place in the presence of water. Then, the strength of such type of cement when used would be adversely affected to the extent such reaction would have taken place (7). For prevention of cement against deterioration and retaining its freshness its storage should be such that no dampness or moisture is allowed to reach cement either from the 2. BACKGROUND
8
ground, walls or from the environment. This becomes particularly important during the humid season and in coastal regions when atmospheric air contains higher amount of moisture in it. Therefore, if binding property and strength of cement depend on the cement capacity for chemical reaction and, on the other hand, open cement has been reacting with the moisture of the atmosphere for a period of time, Initial and Final Setting Times of the fresh mixture should be affected because the cement particles have been already reacting before the addition of water. This
fact
could
cause
changes
in
the
construction
schedule
with
its
corresponding loss of money and resources and also, it can result in important disparities in the comparison of concrete test results in research laboratories. With different experiments, Initial and Final setting times can be predicted for a certain cement, but for test validation, the cement age at the moment of the test must be the same for all cases. As opening a new sac of cement each time an experiment is performed, it would not been reasonable, another solution should be carried out. Vacuum conservation method allows cement to be protected from humidity from the first moment the sac is open. The process consist on high vacuuming certain quantities of cement in sealed airtight plastic bags, so the cement will not contact the atmosphere until the moment the test will carry out. Previous researches at the COBO Department of the Royal Military Academy (RMA) of Brussels have shown differences between the values for the setting between vacuum and non-vacuum cement from the second week of the monitoring.
Unfortunately, any work
tested during the first 14 days, where the difference in setting between the two cements became wider to then start to stabilize. These previous works at the RMA left some guidelines for future works, which this report will attempts to response to. The mentioned guidelines are: 2012 García Cortés, J.M. The efficiency of the vacuum preservation of cement
and its effects on the early hydration of fresh mortar. (8) “Repeat the work: •
With a different kind of cement
•
Change the timing at early ages
•
Create a new “maturity method” for penetrometer.
•
Search a valid indicator for initial setting for SA”.
2. BACKGROUND
9
2012 Ortiz Taboada, N. Study of the effect of admixtures on the early hydration of self-compacting mortars: comparison of experimental methods (9). “…it was noticed that the vacuum preservation of the cement presented some flaws, a study of the efficiency of the vacuum preservation of cement and its effects on the early hydration of fresh mortar”. 2011 William L. “Utility of a vacuum conservation method for cement”. (10) “It is recommendable to confirm these results, omitting the use of silica gel pouches and using the final vacuum conservation method from the start of the tests. It is also necessary to include a larger number of repetitions for each experiment, thus increasing the reliability of the results”. Therefore, the main objective of this research is to clear up the influence of the atmosphere contact on the capacity of chemical reaction according to Initial and Final Setting times through the monitoring of the efficiency of the vacuum preservation of cement in the first two weeks of cement age. For this purpose, half of the purchased cement is vacuumed in small sealed airtight plastic bags and the rest remains in its original already open bags. It is essential that all cement has exactly the same composition, so the cement which was bought was fabricated in the same plant, on the same date and with the same mechanical and chemical characteristics. Then, with a frequency of almost every day during the first two weeks, both vacuum and not-vacuum cement is tested with three different experiment techniques for fresh mortar:
P-wave Transmission Test, Semi-adiabatic Calorimetry and Penetration Resistance
Test. The first one is a non-destructive technique which tests the evolution of the mechanical properties through the transmission of p-waves meanwhile Semi-adiabatic Calorimetry is used to determine the hydration heat generation. On the other hand, Penetration Test is used to determine the change in penetration resistance of a needle in the cement paste. With these three experimental techniques setting times can be deducted. To be able to compare the results of this research to the previous ones of the RMA, there should be more testing days beyond the first two weeks. So, with a lower frequency of testing, data was obtained until two months after the start of the whole test.
2. BACKGROUND
10
2.2 Concrete Concrete is a composite material that results from mixing water, cement (binder), aggregate (sand), coarse aggregate (gravel or crushed rocks) and other additions. When aggregate (diameter smaller than 5 mm) is added, the resulting mixture is called mortar, more commonly used for small volume applications. Nowadays, concrete based on Ordinary Portland Cement (OPC) is one of the most consumed materials in the world. Thanks to its versatility, low price and energetic efficiency, it is indispensable in many kinds of construction work around the world. Concrete has a behavior viscoelastic and possesses a compression resistance roughly 10 times bigger than its resistance to traction. For this reason, the concrete is normally reinforced with steel bars which will improve its mechanical characteristics. Mortar and the concrete pre-dates the art of construction. One fresh in Thebes from 1950 before J.C. show how to make concrete and mortar. Egyptians used cement materials obtained by burning gypsum, Romans and Greeks used volcanic ash and turf, mixed with lime and sand, which produced mortar. It is from the Roman words 'caementum' meaning a rough stone or chipping and 'concretus' meaning grown together or compounded, that we have obtained the names for these two now common materials. However after the fall of the Roman Empire, concrete as a construction material disappeared until the beginning of 1900s. In 1756, British engineer, John Smeaton made the first modern concrete (hydraulic cement) by adding pebbles as a coarse aggregate and mixing powered brick into the cement.In 1824, Joseph Aspdin patented a hydraulic lime, which he named Portland cement which was produced using separate clay and lime-bearing materials, so named because of its close resemblance to Portland stone. Portland was improved in 1835 by Isaac-Charles Johnson, who by increasing the ring temperature, produced burnt clinker pieces which improves cement after grinding, heralding the start of real modern Portland Cement. But the really big breakthrough came with the discovery of reinforced concrete. Joseph Louis Lambot in 1854 and Joseph Monier in 1867 obtained patents for using a mixture of iron and cement, but was Thaddeus Hyatt who published in 1877 that the iron is appropriate to absorb the tensile stresses in a concrete structure, in case that they are put in the good place (11) (12).
2. BACKGROUND
11
2.3 Composition of Concrete On a centimeter scale, concrete can be considered as a composite material consisting of mortar and coarse aggregates (gravel or crushed rocks). On a millimeter one, mortar consists of a composition of cement paste and fine aggregate (sand). On a micrometer one, the microstructure of hardened cement paste is composed of unreacted cement, calcium silicate hydrates, calcium hydroxide, capillary pores and other chemical phases. a)
b)
Figure 2.1. a) Standard percentages of concrete ingredients. b) Size range and specific surface area 8 which is the total particle surface area of a unit mass of cement) of concrete ingredients (13)
Specification, performance, production and conformity European criteria is founded in the standard EN 206-1.
2.3.1
Cement
The most commonly used cement nowadays is the Ordinary Portland Cement (OPC). ASTM C 150 defines Portland cement as "hydraulic cement (cement that not only hardens by reacting with water but also forms a water-resistant product) produced by pulverizing clinkers consisting essentially of hydraulic calcium silicates, usually containing one or more of the forms of calcium sulfate as an inter ground addition”. OPC consists of angular particles with dimensions of 1-50 µm, and is produced by a binder formed from a mixture of limestone and calcined clay heated together to a temperature of 1300 to 1500ºC, at which temperature, the material sinters and partially 2. BACKGROUND
12
fuses to form nodular shaped Portland Clinker. Then, the clinker is cooled and ground to fine powder with addition of about 3-5% of gypsum. The product formed by using this procedure is the Ordinary Portland Cement. This mixture is subsequently milled and mixed with water sets and left to harden forming a new product which is stable when submerged in both air and water. Portland cement clinker is a hydraulic material which shall consist of at least twothirds by mass of calcium silicates (3 CaO·SiO2 and 2 CaO·SiO2), the remainder consisting of aluminium- and iron-containing clinker phases and other compounds. Tricalcium silicate and dicalcium silicate are the most important compounds responsible for strength. Name of Compound
Formula
Abbreviated Formula
Mass Content
Tricalcium silicate (alite)
3 CaO.SiO2
C3S
45-75%
Dicalcium silicate (belite)
2 CaO.SiO2
C2S
7-32%
Tricalcium aluminate
3 CaO.Al2O3
C3A
0-13%
4CaO.Al2O3.Fe2O3
C4AF
Tetracalcium aluminoferrite(ferrite)
0-18%
Table 2.1. Typical constituents of Portland clinker
Figure 2.2. Schematic presentation of various compounds in clinker
To manufacture a cement of stipulated compound composition, it becomes absolutely necessary to closely control the oxide composition of the raw materials. The relative proportions of the oxide compositions are responsible for influencing the various properties of cement; in addition to rate of cooling and fineness of grinding. Sulfate is included primarly to control aluminate reactions.
2. BACKGROUND
13
Name of Compound
Oxide
Percent content
Calcium oxide
CaO
60,0-67,0
Silicon dioxide
SiO2
17,0-25,0
Aluminum oxide
Al2O3
3,0-8,0
Ferric oxide
Fe2O3
0,5-6,0
Magnesium oxide
MgO
0,1-4,0
Alkalies
(K2O, Na2O)
0,4-1,3
Sulfate oxide
SO3
1,3-3,0
Table 2.2. Approximate oxide Composition Limits of Ordinary Portland Cement (6)
Types of cement There are different standards for classification of Portland cement. The two major standards are the ASTM C150 used primarily in the U.S. and European EN-197. EN 197 cement types CEM I, II, III, IV, and V do not correspond to the similarly named cement types in ASTM C 150. The European Committee for Standardization (CEN) divides the family of common cement into 27 products, which are grouped into the following five main cement types:
Class
Name
Special characteristics
CEM I
Portland cement
up to 5% of minor additional constituents
CEM II
Portland-composite cement
up to 35% of other single constituents
CEM III
Blastfurnace cement
higher percentages of blastfurnace slag
CEM IV
Pozzolanic cement
up to 55% of pozzolanic constituents(volcanic ashes)
CEM V
Composite cement
blastfurnace slag or fly ash and pozzolana
Table 2.3. European EN-197 classification of Portland cement.
2.3.2
Water
Water and the cement are the active constituents of concrete, unlike the aggregates which are considered inert. This is because when water is added to cement, it is only then that the hydration reaction starts, and it is only then that the cement can begin to play its role of linking. Water plays two roles: Securing the hydration of cement. The amount of water that allows this role is about 25 % in relation of the amount of cement. 2. BACKGROUND
14
Confering certain workability to concrete, which allows his establishment. Normally, lower water to cement ratio (w/c) will yield a stronger, more durable concrete
(14);
while
more
water
will
give
a
freer-flowing
concrete
with
a
higher slump. Impure water used to make concrete can cause problems when setting or in causing premature failure of the structure (15). There is actually water in excess relative to the amount required for hydration. This amount must not be too high, to avoid problems such as bleeding, drying shrinkage and creation of pores in the concrete. This can greatly reduce the strength of concrete. In total, approximately 25 % of water is needed for chemical demands, plus 15 % more for physical demands. We require about 40 % of the mass of cement of water. The quality and temperature of water are also important factors. The criteria for water quality are derived from the standards ASTM C 1602, 1603/C and the EN 1008.
2.3.3
Aggregates
Aggregate is commonly considered inert filler, which accounts for 60 to 80 percent of the volume and 70 to 85 percent of the weight of concrete (see Figure 2.3). Although aggregate is considered inert filler, it is a necessary component that defines the concrete’s thermal and elastic properties and dimensional stability. Aggregate is classified as two different types, coarse and fine. Coarse aggregate is usually greater than 4.75 mm (retained on a No. 4 sieve), while fine aggregate is less than 4.75 mm (passing the No. 4 sieve). The compressive aggregate strength is an important factor in the selection of aggregate (16). Other physical and mineralogical properties of aggregate must be known before mixing concrete to obtain a desirable mixture. These properties include shape and texture, size gradation, moisture content, specific gravity, reactivity, soundness and bulk unit weight. These
properties
along
with
the
water/cementitious
material
ratio
determine
thestrength, workability, and durability of concrete. The shape and texture of aggregate affects the properties of fresh concrete more than hardened concrete. Concrete is more workable when smooth and rounded aggregate is used instead of rough angular or elongated aggregate. Crushed stone produces much more angular and elongated aggregates, which have a higher surface-to-volume ratio, better bond characteristics but require more cement paste to produce a workable mixture. The most common classification of aggregates on the basis of bulk specific gravity is lightweight, normal-weight, and heavyweight aggregates. In normal concrete the aggregate 2. BACKGROUND
15
weighs 1,520 – 1,680 kg/m3, but occasionally designs require either lightweight or heavyweight concrete. Lightweight concrete contains aggregate that is natural or synthetic which weighs less than 1,100 kg/m3and heavyweight concrete contains aggregates that are natural or synthetic which weigh more than 2080 kg/m3 (17).
2.3.4
Additions
Additions are inorganic materials that have pozzolanic or latent hydraulic properties. These very fine-grained materials are added to concrete to improve certain properties. If the addition is pozzolanic or hydraulic, its cementing properties are used in addition to or as a partial replacement of some of the cement, and are classffied as supplementary cementing materials. Fly ash: A by-product of coal-fired electric generating plants, it is used to partially replace Portland cement (by up to 60% by mass) (18). The properties of fly ash depend on the type of coal burnt. In general, siliceous fly ash is pozzolanic, while calcareous fly ash has latent hydraulic properties. Ground granulated blast furnace slag (GGBFS or GGBS): A by-product of steel production is used to partially replace Portland cement (by up to 80% by mass). It has latent hydraulic properties (19). Silica fume: A by-product of the production of silicon and ferrosilicon alloys. Silica fume is similar to fly ash, but has a particle size 100 times smaller. This results in a higher surface to volume ratio and a much faster pozzolanic reaction. Silica fume is used to increase strength and
durability
of concrete,
but
generally
requires
the use of
superplasticizers for workability.[36] High reactivity Metakaolin (HRM): Metakaolin produces concrete with strength and durability similar to concrete made with silica fume. While silica fume is usually dark gray or black in color, high-reactivity metakaolin is usually bright white in color, making it the preferred choice for architectural concrete where appearance is important.
2.3.5
Admixtures
Admixtures are chemical admixtures in the form of powder or fluids that are added to the concrete to give it certain characteristics not obtainable with plain concrete mixes. In normal use, admixture dosages are less than 5% by mass of cement and are added to the concrete at the time of batching/mixing. The common types of admixtures are as follows.
2. BACKGROUND
16
Accelerators speed up the hydration (hardening) of the concrete. Typical materials used areCaCl2, Ca(NO3)2 and NaNO3. However, use of chlorides may cause corrosion in steel reinforcing and is prohibited in some countries, so that nitrates may be favored. Retarders slow the hydration of concrete and are used in large or difficult pours where partial setting before the pour is complete is undesirable. Typical polyol retarders are sugar, sucrose, sodium gluconate, glucose, citric acid, and tartaric acid. Air entrainments add and entrain tiny air bubbles in the concrete, which will reduce damage during freeze-thaw cycles, thereby increasing the concrete's durability. However, entrained air entails a trade off with strength, as each 1% of air may result in 5% decrease in compressive strength. Plasticizers increase the workability of plastic or "fresh" concrete, allowing it be placed more easily, with less consolidating effort. A typical plasticizer is lignosulfonate. Plasticizers can be used to reduce the water content of a concrete while maintaining workability and are sometimes called water-reducers due to this use. Such treatment improves its strength and durability characteristics. Superplasticizers are a class of plasticizers that have fewer deleterious effects and can be used to increase workability more than is practical with traditional plasticizers. Compounds used as superplasticizers include sulfonated naphthalene formaldehyde condensate, sulfonated melamine formaldehyde condensate, acetone formaldehyde condensate and polycarboxylate ethers. Pigments can be used to change the color of concrete, for aesthetics. Corrosion inhibitors are used to minimize the corrosion of steel and steel bars in concrete. Bonding agents are used to create a bond between old and new concrete (typically a type of polymer). Pumping aids improve pump ability, thicken the paste and reduce separation and bleeding.
2.4 Hydration of t he Cement Cement hydration is a series of irreversible chemical reactions between cement and water. During hydration, the cement-water paste sets and hardens, “gluing” the aggregate together in a solid mass. It influences how the plastic concrete behaves when it is being placed and finished and governs how strong and durable the hardened concrete becomes. During the first 72 hours after mixing, concrete can often gain 50 percent of its strength or more.
2. BACKGROUND
17
Chemical reaction between the various compounds in Portland cement and water can be divided in five stages (20):
Stage 1: Mixing (< 15 minutes) ○
Aluminates dissolve and react quickly, with high heat.
○
Sulfate dissolves quickly, too. It reacts with aluminate and
water, forming a gel C-A-S-H, a precursor to ettringite. This gel limits water’s access to aluminate. The reactions slow and the heat drops (Figure 2.4).
Figure 2.4. Controlled alumination
(1)
○
reactions in Stage 1
As it was mentioned in 2.2.1 Cement, sulfate is included in cement primarily to control
aluminate reactions. Moreover: Too little sulfate – immediate hardening of the mix, or flash set, because of the rapid hydration of the C3A: (2)
Too much sulfate – can precipitate out, causing temporary stiffening of the mix, or false set.
Stage 2: Dormancy (2-4 h) ○
While the C-A-S-H gel controls aluminate reactions, the
concrete is cold, plastic and workable. ○
The silicates (alite and belite) slowly dissolve, releasing
calcium ions in solution (Figure 2.5). ○
During dormancy (that it is before initial set) the mix can
be transported, placed, finished and textured.
Figure 2.5. Cement particle dissolving during stage 2
Stage 3: Hardening (2-4 h) ○
The solution eventually becomes supersatured with calcium
ions, triggering the formation of new components: Calcium silicate hydrate (C-S-H, fiber-like particles), which adheres to aggregate and gives concrete its strength. Calcium hydroxide (CH, crystals).
Figure 2.6. New hydration compounds during stage 3
2. BACKGROUND
18
(3) (4) ○
The formation of C-S-H and CH generates heat, causing thermal expansion.
○
Initial Set occurs when enough C-S-H and CH form to mesh together, causing the mix to
stiffen (Figure 2.6). ○
As these products continue to mesh, the concrete begins developing some strength.
○
Final Set occurs when the concrete achieves a defined stiffness, about when the concrete
is hard enough to walk on. ○
The gel-like C-A-S-H transforms into a needle-like solid (ettringite) that contributes
somewhat to early strength. (5) ○
It is critical to apply curing compound thoroughly (or conduct other curing practices) as
soon as possible after finishing, before the concrete begins hardening. It slows water evaporation from the surface, retaining mix water for hydration and reducing the risk of plastic shrinkage cracking.
Stage 4: Cooling (several hours) ○
Soon after final set, the buildup of C-S-H and CH begins to
limit access of water to undissolved cement. Silicate reactions slow (Figure 2.7). Heat peaks and begins to drop. ○
The sulfate will be depleted. Any remaining aluminate then
reacts with the C-A-S-H. This has little effect on concrete properties except for a brief increase in heat. ○
Figure 2.7. Hydrations compounds during stage 4
As it cools, concrete contracts. It can cause tensile stress and concrete can crack.
Stage 5: Densification (can continue for years) ○
Belite dissolves and reacts more slowly than alite.
○
Belite reactions also produce C-S-H and CH, forming a solid
mass (Figure 2.8). ○
The longer the cement in concrete hydrates (that is, belites
and any remaining alites react with water):
The greater the concrete’s strength.
The lower its permeability.
Figure 2.8. Hydration compounds during stage 5
2. BACKGROUND
19
The greater its potential durability. To sum up, cement and hydration of Portland cement can be schematically
represented in Figure 2.9:
Component elements in raw materials
O2
Si
Ca
Al
Fe
Oxide composition in raw materials
CaO SiO2 Al2O3 Fe2O3
On burning, clinker formed
Compound composition
C 3S
C 2S
C 3A
C4AF
On grinding clinker
Portland cements
Various types On hydration
Products of hydration
C-S-H
CH
C-A-S-H
"CFH" "CAH"
Calcium Silicate Hydrate Portlandite Ettringite Aluminate Hydrates Figure 2.9. Summary of the process from fabrication of cement to hydration
Figure 2.10.Formation of new materials in the hydration process
2. BACKGROUND
20
2.4.1
H e a t o f H yd r at i o n
As it has been said, during the process of hydration, heat is given off to the environment; this is because the reaction of cement with water is exothermic. This liberation of heat is called heat of hydration. The five stages are now illustrated by a curve that represents changes in heat during the first hours and days of hydration:
C3A hydration C3S hydration
Figure 2.11. The five stages of hydration mapped on a heat of hydration vs. time curve
It can be observed that just after mixing there is a high but very narrow peak corresponding to the reaction of solution of aluminates and sulphates and the formation of the protective gel C-A-S-H, the precursor of ettringite. This initial heat evolution ceases quickly when the solubility of aluminium is depressed by gypsum and starts the dormant period, where the mixture is relatively inactive. After the dormant period, at a time corresponding roughly to the initial set (20), an acceleration occurs as the C-S-H and CH are formed, the heat rate starts to increase rapidly and reaches a maximum peak, and also may be due to the reaction of C3S (6) As the reactions gradually slow down, the mixture starts cooling and a second smaller peak appears due to the formation of monosulphate. During the final stage it is possible that a further narrow peak occurs after one or two days (21). The total quantity of heat generated in the complete hydration will depend upon the relative quantities of the major compounds present in cement, but they also factors which make the specimen hydrates at different rates.
2. BACKGROUND
21
Different compounds hydrate at different rates and liberate different quantities of heat. Figure 2.12 shows the rate of hydration of pure compounds. Since retarders are added to control the flash setting properties of C 3A, actually the early heat of hydration is mainly contributed from the hydration of C3S.
Figure 2.12. Rate of Hydration of Pure Compounds. Adaptation from (22)
C3S readily reacts with water and produces more heat of hydration. It is responsible of the early strength of concrete. A cement with more C 3S content is better for cold weather concreting. The quality and density of calcium silicate hydrate formed out of C 3s is slightly inferior to that formed by C2S. The early strength of concrete is due to C3S. C2S hydrates rather slowly. It is responsible for the later strength of concrete. It produces less heat of hydration. The calcium silicate hydrate formed is rather dense and its specific surface is higher. In general, the quality of the product of hydration of C 2S is better than that produced in the hydration of C3S. On the other hand, the reaction of pure C 3A with water is very fast and this may lead to flash set. To prevent it, gypsum is added at the time of grinding the cement clinker. The hydrated aluminates do not contribute anything to the strength of concrete but their presence is harmful to the durability of concrete particularly where the concrete is likely to be attacked by sulphates. Anyway, as it hydrates very fast, it may contribute a little to the early strength.
2. BACKGROUND
22
Figure 2.13. Development of Strength of Pure Compounds (6)
Other factors influencing heat development in concrete include the cement fineness and content, water-cement ratio, placing and curing temperature, the presence of mineral and chemical admixtures, and the dimensions of the structural element. Fineness of cement also influences the rate of development of heat but not the total heat. It is because higher fineness provides a greater surface area to be wetted, resulting in an acceleration of the reaction between cement and water. This causes an increase in the rate of heat liberation at early ages, but may not influence the total amount of heat developed over several weeks. In general, higher cement contents result in more heat development. With equal cement content, higher water-cement ratios cements have more water and microstructural space available for hydration, resulting in an increased rate of heat development. Nevertheless, the water-cement ratio effect is minor compared to the effect of cement content. On the other hand, higher ambient temperatures greatly accelerate the rate of hydration and the rate of heat liberation at early ages. Chemical admixtures that accelerate hydration also accelerate heat liberation and admixtures that retard hydration delay heat development. Mineral admixtures, such as fly ash, can significantly reduce the rate and amount of heat development (9) (23).
2.5 Setting The concept of setting was already introduced in the explanation of the process of hydration, but as it is the main variable under study, it deserves an independent chapter.
2. BACKGROUND
23
The setting mainly concerns the cement paste, as the aggregate itself is inert and does not directly affect the setting. One possible definition for setting is the gradual transition from the initial liquid state to a load solid material. Traditionally, two points during setting are considered important for concrete practice. Initial Setting Time (IST) is of importance, as it provides an estimate of when the concrete has reached the point where it has stiffened to such an extent where plasticity and workability are lost and it can no longer be consolidated without damaging the concrete. On the other hand, Final Setting Time (FST) of concrete relates to the point where stresses and stiffness start to develop and the mixture becomes rigid and a solid material. The setting time is extremely important for the construction process. The time of initial set gives the contractors an indication of how long the concrete can be cast, compacted and finished, while the final setting time provides an estimate of the start of the concrete strength development and it is used to schedule removal of forms or reshoring, backfilling walls, pretressing and post-tensioning or determining the time for opening the pavements to traffic (4) Soroka (24), gives an interesting description of setting from a phenomenological, chemical and structural point of view:
2.5.1
S e t t i n g Ph e n o m e n o l o g i c a l p o i n t o f vi e w
Mixing cement and water produces a plastic and workable cement paste. During the dormant period, these early properties remain unchanged, but a certain stage, called the initial set, the cement paste stiffens to such a degree that the plasticity and workability is lost. Thereafter, the paste continues to stiffen until it becomes a rigid and solid material, this is the final set. Both initial and final set are of practical importance. The initial setting time determines the period that the concrete is workable on a construction site and can be cast and vibrated without damaging its structure. After casting the concrete, the construction work cannot continue until final set has occurred. At that age, the concrete is rigid enough to remain stable unsupported. The formwork can then be removed when a certain strength level is reached, depending on the loading conditions.
2. BACKGROUND
24
2.5.2
C h e m i c al p o i n t o f vi e w
Setting depends on the hydration reactions, which influence the two main setting mechanism, coagulation and bridging: 1) After mixing, water becomes an electrolyte due to the partial solution of the cement minerals. Since the H2SiO42- ions remain at the surface of the cement grains, they are negatively charged and coagulation is hindered by electrostatic repulsion. However, chemical reactions at the surface with the H2SiO42- reduce this electrostatic repulsion so the Van der Waals forces lead to the coagulation of the cement grains (25). 2) The precipitation of hydration products in the contact zone between the coagulated cement grains, binds them together forming a rigid structure. Less than 5% hydrated material is already sufficient to make the cement paste perfectly rigid (25). Accordingly, final setting occurs when a specific degree of hydration is reached, depending on the w/c ratio. Normal setting thus occurs during the acceleratory hydration period characterized by the rapid formation of C-S-H and CH.
2.5.3
S t r u c t u r al p o i n t o f v i e w
As hydration reactions take place, the cement grains are encapsulated by a dense layer of deposited hydration products. As the hydration develops, the thickness of the layer increases, friction increases between the grains and workability is lost. Initial set is reached. As the cement hydration proceeds, the bonds between the cement grains start to create continuity. According to the percolation theory, the first long-range connectivity in a sample is achieved when a critical fraction of particles is bonded to each other, called the percolation threshold (26). From then, the percolation probability increases rapidly and reaches a value of nearly 1 (4). When it happens, the mixture changes from plastic to solid state. The elastic modulus, compressive resistance, Poison's coefficient develop (27).Then, the paste can be regarded as a porous solid material and final set has occurred. Afterwards, the hydration products continue to expand into the pore space during the hardening. The strength of the sample starts to increase significantly, while the largest increase in stiffness has already occurred.
2. BACKGROUND
25
Figure 2.14. Cement grain hydration and percolation threshold. Adapted from (28)
2.6 Influence of Laboratory Conditions in the Setting Time 2.6.1
I n f l u e n c e o f a t mo sp h e r e t e m p er a t u r e
As it was abovementioned in 2.3.1 Heat of Hydration, the hydration of Portland cement is a temperature dependent reaction; therefore the mechanical properties that are measured as indicators of concrete set are highly temperature dependent as well. Some research (29) (5) (30)
have noted this in the development of the penetration setting test. It
was noted that, with the decrease in temperature from 37º to 10ºC, the time of initial set increased by as much as 400 percent (31). The final test results for all the cases were that setting occurs setting occurs earlier at higher temperatures and later at lower temperatures.
Figure 2.15. Effect of temperature on penetration resistance (31).
Burg (32) also presented the results from the influence of temperature in concrete strength. As expected, early age compressive strength of concrete cast and cured at high temperature was greater than concrete cast and cured at 23ºC. However, after seven days, 2. BACKGROUND
26
compressive strength of concrete cast and cured at high temperature was lower than concrete cast and cured at 23ºC. On the other hand, concrete cast and cured at low temperature had initial strength lower than concrete cast and cured at 23ºC. However, later age strength either equaled or exceeded that of concrete cast at 23ºC.
2.6.2
I n f l u e n c e o f r e l a t i ve h u mi d i t y
Established maturity formulas which are a function of time interval and temperature do not incorporate the effect of humidity on the strength development of Portland cement mortars. However, it is apparent that the effect of humidity may considerably change the hydration process even under same temperature. This result indicates that well-known maturity formula which is function of time interval and temperature is not valid for the climates with relative humidity values lower than 75% (33).
2. BACKGROUND
27
3.
TEST PROGRAM Three types of experiments were carried for monitoring the process of hydration of
the fresh mortar: Ultrasonic P-wave Transmission Test, Semi-adiabatic Calorimetric and Penetration Resistance Test. Each one of the three mentioned tests has their own way of analysis to obtain the cement setting times.
3.1 Laboratory and Materials 3.1.1
Laboratory
All tests have been carried out in the Concrete Laboratory of the Royal Military Academy of Brussels. According to ASTM C403 (1) and the European Standard 196-9 (3), Penetration Resistance Test and Semi-adiabatic should be performed under certain conditions of ambient temperature and relative humidity. The requirements are: The laboratory where the mortar is mixed shall be maintained at a temperature of (20 ± 2) °C. The room where the test is carried out shall be maintained at a temperature of (20,0 ± 1,0) °C. Unfortunately, the ambient conditions of the RMA laboratory cannot be set as fixed. Because of this, some alternative measures have been adopted to try to maintain the ambient conditions as the Standard stipulates. These measures have been: ○
To maintain the temperature: As the tests have been carried during the winter, the
daily laboratory temperature used to be under 20ºC. For this reason, two powerful heaters have been placed in different points of the lab and they have been continuously working all day-all night.
○
To maintain the relative humidity: a second use has been given to the water tank
available in the lab for submerging concrete specimens for another kind of tests. A resistance is placed in the tank which increases the temperature of the water until a value around 60ºC. Also, a water pump is placed in the tank, in order to make the water moves around the resistance and increase the water surface which is exposed to the air.
3. TEST PROGRAM
28
Figure 3.1. Water tank with heating resistance and water pump to increase the relative humidity of the laboratory
3.1.2
Water
For mixing, tap water at 20ºC of temperature was used. This is a requirement standardized for the fresh mortar tests.
3.1.3
Sand
Sand should also follow the requirements stated by EN 196-1, “CEN Standard sand” (34). It should be natural, siliceous sand consisting of rounded particles and with a silica content of at least 98%. Moisture also hast to be less than 0.2% in mass. Sand has also to comply with the following particle size distribution determined by sieve analysis:
Figure 3.2. Particle size distribution of the CEN Reference sand. Adaption from EN 196-1 (9)
Standard sand used in the tests was provided in individual plastic bags with a content of (1350 ± 5) g (exactly quantity required for the EN-196 mortar mixing).
3. TEST PROGRAM
29
Figure 3.3. CEN Normalized Sand used in this research
3.1.4
Cement
All cement used in this research was produced on the same date on the same industry and with the same characteristics. For all the tests, three 25 kg sacks were bought in the sailing point of Holcim Belgium cements, Brussels, on the 22th November 2012. The type of cement chosen is CEM I 52.5 R HES, which, according the European Standard EN197:2000 (35), means: ○
CEM I: Portland cement with up to 5% of minor additional constituents.
○
52.5 R: Strength class of 52.5 R. It means Rapid Cement which has an early strength
≥ 30 MPa on the second day, a standard strength at least of 52.5 MPa after 28 days and an initial setting time ≥ 45 min. ○
HES: High Early Strength Cement.
3.1.5
Vacuum Procedure
With the vacuum procedure, the new cement, already purchased, is introduced in plastic bags to maintain its original conditions of humidity along the time through isolation from mass transfer between the atmosphere and the cement. It means that it is not isolated from the ambient temperature but there may not be any transfer of relative humidity or other agents from the atmosphere to the cement. As it was presented in 3.1.4 Cement, three 25 kg sacks of CEM I 52,5 R HES were purchased on 22nd November 2012 and transported immediately to the RMA Concrete Lab. At 11:15 am the first 25 kg sack was open and its content was introduced by +900 g parts 3. TEST PROGRAM
30
in plastic bags. Then, each of the plastic bags with the cement was vacuumed with an offthe-shelf food vacuum conservation device (DO316L, produced by DOMO) connected to a powerful vacuum pump. When the internal pressure gets a value of around 6 mbar the bag is subsequently sealed. The disadvantage of this stronger pump is that a certain quantity of the cement can be sucked out of the bag, reducing also the quality of the heat seal and polluting the vacuum pump oil. Because of these reasons, the bag is filled with an extra quantity of cement (about +15 g) and after vacuuming it, it is placed alone or with another one or two in a second bag and this one is also vacuum in order to reduce the impact of a possible leak in the inner vacuum bag.
Figure 3.4. Process of vacuum. At the end of the process it can be notice how the internal air of the bag is removed
At 11:45 am, the two other sacks of cement were opened. Half of one of them was also divided into +900g parts and vacuumed and sealed. It is important to mention that the plastic bags used to package the cement were already dried during the previous 24 hours in a vacuum oven to reduce their possible humidity.
CEM I 52,5 R HES
Date of Purchased: 22nd November 2012
Number of Sack
Hour of Opening
Use
Patch Name
1
11:15
Vacuum Package
Cement 1st
2 3
11:45
Vacuum and Open Open Storage
Cement 2nd
Table 3.1. Resume of the purchased cement division
3. TEST PROGRAM
31
CEM I 52,5 R HES Ciment rapide haute performance Le produit et ses applications Le ciment CEM I 52,5 R HES est un ciment portland dont l’unique “constituant principal” est le clinker portland (K). La teneur en clinker est supérieure à 95 %.
SAC
Domaines d'application préférentiels Bétons en milieu non agressif (classes d'environnement E0, EI et EE selon la norme NBN B15-001), qui demandent un décoffrage, une manutention ou une mise en service très rapide Bétons de classes de résistance très élevée Bétons précontraints Préfabrication de produits en béton
VRAC
Avantages du CEM I 52,5 R HES Durcissement très rapide Résistance élevée à très courte échéance et résistance très élevée à moyenne échéance
Spécifications techniques Caractéristiques mécaniques et physiques **
Prise Besoin en eau Début Fin
Recommandation particulière • Recommandé en période hivernale pour mettre le béton hors gel
% hh:mm hh:mm
32 2:40 3:30
≥ 0:45 ≤ 12:00
≤1
≤ 10
N/mm2 N/mm2 N/mm2
31 42 63
≥ 20 ≥ 30 ≥ 52,5
Surface spécifique Blaine
m2/kg
500
-
Masse volumique absolue
kg/m3
3160
-
Masse volumique apparente
kg/m3
1138
-
%
≤ 0,1
≤ 3,0
-
1,2
Refus au tamis de 200 µm
Composition chimique ** Valeur C
-
RÉSULTATS (%)
SPÉCIFICATIONS (%) EN 197-1
CaO
62,7
-
SiO2
17,7
-
Al2 O3
5,5
-
40
Fe2O3
4,1
-
30
MgO
0,8
-
20
Na2 O
0,39
-
K2O
0,70
-
SO3 Cl-
3,7
≤ 4,0
0,05
≤ 0,10
Perte au feu
1,2
≤ 5,0
Résidu insoluble
0,25
≤ 5,0
Evolution de la résistance à la compression d'un béton standard * 60
Résistance à la compression (N/mm2) 50
10
Temps ( jours) 0 1
Le ciment CEM I 52,5 R HES est marqué CE (en tant que CEM I 52,5 R), ce qui garantit la conformité à la norme EN 197-1. En outre, il répond à plusieurs normes nationales et porte différentes marques de qualité nationales comme indiqué ci-contre :
SPÉCIFICATIONS EN 197-1 ET PTV 603
mm
Résistances béton Usine d'Obourg certifiée
RÉSULTATS
Résistance à la compression 1 jour 2 jours 28 jours
Stabilité
Contre-indications • Bétons en milieu agressif (classes d'environnement EA2 et EA3 selon la norme NBN B15-001) • Bétons pour constructions massives • Utilisation de granulats sensibles à la réaction alcalisgranulats pour les bétons en milieu humide
UNITÉS
7
14
21
28
La figure donne l'évolution de la résistance à la compression sur cubes de 150 mm d'arête, obtenue dans notre laboratoire sur un béton à base du CEM I 52,5 R HES. Les caractéristiques principales du béton sont : • granulométrie continue : concassé calcaire 4/20 + sable de rivière gros • dosage en ciment : 350 kg/m3 • fluidité : affaissement (slump) de 120 mm • facteur E/C : environ 0,54 PAYS
NORME
DÉNOMINATION
MARQUE
Belgique
NBN EN 197-1 NBN B12-110
CEM I 52,5 R HES
Benor
France
NF EN 197-1 NF P15-318
CEM I 52,5 R CP2
NF
Pays-Bas
NEN EN 197-1
CEM I 52,5 R
KOMO
* Remarque : La résistance d’un béton dépendant de beaucoup de facteurs, la courbe de la figure n'est pas nécessairement représentative pour l'évolution des résistances d'un béton quelconque à base de CEM I 52,5 R HES. ** Remarque : Les résultats repris dans les tableaux sont basés sur des valeurs moyennes et sont donnés à titre purement indicatif et n’ont en aucun cas un caractère contractuel. En conséquence, ils ne sauraient engager la responsabilité de Holcim (Belgique) s.a.
Holcim (Belgiq La fiche MSDS de ce produit est disponible sur www.holcim.be
Technical helpdesk :
[email protected] 3. TEST PROGRAM
32
3.1.6
Composition of the mortar
EN 196- 1 (34) establishes the proportions of the mortar compounds by mass rate: Quantity according one
Proportion by mass
Standard Mixture
One part of cement
450 ± 2 g
Three parts of CEN Standard sand
1350 ± 5 g
One half part of water (water/cement ratio = 0.5)
225 ± 1 g
Table 3.2. Proportion of the mortar compounds by mass rate in a standard EN 196-1 mixture
3.1.7
M i x i n g o f mo r t a r
For Semi-adiabatic Calorimeter Test and for the P-wave Transmission one, the different constituents are mixed with the aid of the automatic mortar mixer automix 65 L0006/AM, with a 5l bowl capacity and 4 programmable mixing cycles conforming to EN 196-1, EN 196-3, DIN 1164-5 and DIN 1164-7 and 1 mixing cycle programmable by the operator. For Penetration Resistance Test; a non-auto mixer with larger capacity has been used because of biggest quantity of mortar needed for the test. For this research, programmable mixing cycle is according to EN 196-1, due to the fact that any addition or admixture is being utilized. The procedure of the EN 196-1 is the following: Before starting automix program of the mixer, cement has to be already in the bowl. Then: Start
Final
time
time
00:00
Action
Mixing Speed
00:30
Slow addition of water
Slow
00:30
01:00
Addition of sand
Slow
01:00
02:00*
02:00
02:30
02:30
03:30
Resting time
03:30
04:30
Fast
Fast Stir mortar adhering from the wall and bottom parts
Resting time
Table 3.3. EN 196-1 Mixing Program for basic mortar mixtures
3. TEST PROGRAM
33
a)
b)
Figure 3.5. a) 65 L0006/AM Automix mixter. b) Non-auto mixer for PE mortar mixtures
3.1.8
Series
The aim of this research is monitoring the changings in setting times for vacuum packaged cement and open storage one to give conclusions about the hydration process of the fresh mortar. Therefore, there are going to be just two series of mortar mixture, which are going to be used for all the techniques during the whole research: ○
Series 1: Mortar using Open Storage Cement Series 1 Compound
Percentage in weight (%)
Quantity (g)
CEM I 52.5 R HES (open)
0.222
450
CEN Normalized sand
0.667
1350
Water (w/c = 0.5)
0.111
225
Table 3.4. Mortar composition of Series 1
○
Series 2: Mortar using Vacuum Packaged Cement Series 2 Compound
Percentage in weight (%)
Quantity (g)
CEM I 52.5 R HES (vacuum)
0.222
450
CEN Normalized sand
0.667
1350
Water (w/c = 0.5)
0.111
225
Table 3.5. Mortar composition of Series 2
3. TEST PROGRAM
34
3.1.9
Nomenclature
Every specimen needs to be classified according to its series, date and experimental technique used. The final name will be a correlation between these factors:
Researcher
Technique
Date
Series
Number of Specimen
Classification Name LAUIP2511O-1
P-wave Transmission (IP)
Open storage (O)
1, 2
LAUIP251O-2 LAUSA2511O-1 LAUSA2511O-2
Laura (LAU)
Semi-adiabatic (SA)
25rd November (2211)
Penetrometer (PE)
LAUPE2511O-1 LAUPE2511O-2 Vacuum package (P)
LAUIP2511P-3 3, 4
LAUIP2511P-4 LAUSA2511P-3 LAUPE2511P-3 LAUPE2511P-4
Figure 3.6. Diagram of the nomenclature of each specimen
To give an example of the description of one specimen name: LAUIP2511O-1: Researcher: LAU- Laura Technique: IP- P-wave Transmission Date: 2511 – 25 November 2012 Series: O- Open storage Number of Specimen: 1
3. TEST PROGRAM
35
3.2 Experimental techniques 3.2.1
P e n e t r a t i o n R e s i st an c e T e s t ( P e n et r o m e t e r )
Penetration resistance methods (Vicat, Penetrometer, etc.) are based on the determination of the depth of penetration of probes (steel rods or pins) into concrete. This provides a measure of the hardness or penetration resistance of the material that can be related to its strength. When the needle penetrates into the cement paste, it exerts two reaction forces on the needle: skin resistance exerted on the “penetrated" surface of the needle which is dictated by shear resistance and a compressive resistance exerted on the tip of the needle (36).
Figure 3.7. Schematic illustration of the forces involved in the penetrometer test.
Penetrometer follows two Standards: 1. EN 196-1 (37) “Methods of Testing cement – Part 3: Determination of setting times and soundness” states that setting time is determined by observing the penetration of a needle into cement paste of standard consistence until it reaches a specified value. 2. ASTM C 403 (1), “Standard Test Method for Time of Setting of Concrete Mixtures by Penetration Resistance” This method can also be used to determine the effects of some variables, such as water content, type and amount of cementitious material or admixtures, upon the time of setting of concrete.
T e s t p r o c ed u r e Mortar mixture equivalent in mass to 4 times the 196-1 standard mixture is prepared in the larger non-auto mixer of Figure 3.5 b). Then, the mixture is placed in non-absorptive wood containers of 140 mm (depth) x 150 mm (width) x 150 mm (length) that allows at 3. TEST PROGRAM
36
least ten penetrations throughout one test in accordance with clear distance requirements specified in the procedure. Each penetration should have a distance of 2.5 cm of either wall of the container or a clear distance of at least 2.5 cm or two diameters between penetrations. The mortar is penetrated to a depth of 25 mm in 10 s. To avoid the disturbance of bleed water, it is removed before any penetration. Different size needles are used depending on the mixture and its state (1/2”, 1/10”, 1/20”, 1/40”). The penetration resistance exerted by the mortar is calculated by dividing the force by the cross-sectional area of the needle. The equipment used is shown in Figure 3.8.
Figure 3.8. Penetrometer at the RMA Concrete Lab.
Each test day, 4 batches are tested: 2 for Open Storage series mixture and the other 2 for Vacuum Packaged Series one. Each couple of batches is made from the same mortar mixture, and the mixtures are made one immediately after the other, to ensure the most similar ambient conditions.
D e t e r m i n a t i o n o f se t t i n g t i m e Initial setting time is determined as the value in time correspondent to the penetration resistance of 3.4 MPa (500 psi). For final setting time, it corresponds to a Penetration resistance of 27.6 MPa (4000 psi). For plot results and determine the times of setting, 9.5.3 ASTM C 403 encourages to use the following power regression: 3. TEST PROGRAM
37
(6)
Where: = penetration resistance = elapsed time and
= regression constants In the following chart, the results from the specimen LAUPE2211O-1 are plot. A
power regression curve is plot according to equation (6) and correlation coefficient is calculated. In addition, setting times are identified according to the values aforementioned.
Penetration Resistance vs Time LAUPE2211O-1
Power (LAUPE2211O-1)
6000
Penetrometer Resist. (psi)
5500 5000 4500 4000 3500 3000 2500
y = 2,32E-14x6,93 R² = 9,92E-01
2000 1500 1000 500 0 210
IST
230
250
270
290
FST
310
330
Mortar Age (min) Figure 3.9. Original data from the Penetrometer Resistance Test (
) and power trendline (
) with its
respective equation (y) and correlation coefficient (R2). Initial and Final Setting times are also displayed
In the case that the correlation coefficient for the regression analysis (after removal of outliers) is less than 0.98, procedure 9.5.1 from ASTM C 403 should be used.
3.2.2
L a n g a v a n t - t yp e S em i - A d i a b a t i c C a l o r i m e t r i c T e s t
As it was described in 2.3.1 Heat of hydration, heat of hydration is the heat generated when cement and water react, because it is an exothermic reaction. The quantity of heat generated mainly depends on the chemical composition of the cement, with C 3A and C3S being the phases primarily responsible for high heat evolution. The water/cement ratio, fineness of the cement and temperature of curing also are relevant factors in the increase of the heat of hydration.
3. TEST PROGRAM
38
Studying temperature variations is interesting not only because it determines the degree of hydration, but also temperature gradients may indicate if they will be early-age cracking and consequently influences the serviceability of the material. During normal concrete construction, the heat is dissipated into the soil or the air, and resulting temperature changes within the structures are not significant. However, in some situations, particularly in massive structures, the heat cannot be readily released. The temperature rise cause expansion while the concrete is hardening. If this is significantly high and the concrete undergoes nonuniform or rapid cooling, stresses due to thermal contraction in conjunction with structural restraint can result in the early-age cracking (23). Different techniques can be used to measure the heat of hydration. These include (9): Heat of solution Isothermal calorimetric Semi-adiabatic calorimetric Heat of solution. This method involves dissolving cement in an acidic mixture within a calorimeter
and
measuring
the
temperature
rise,
its
main
advantage
being
the
measurements can be taken during long periods. It has fallen into disuse due to its limited relevance to concrete performance and high degree of operator skill required to obtain good results. It is important to note this method has not been proven reliable where supplementary cementing materials are sed. This method is standardized by EN 196-8 Methods of testing cement- Part 8: Heat of hydration- Solution Method. Isothermal calorimetric. Samples are placed in a calorimeter which is maintained at constant temperature. As heat is released it is measured. It is important to note the sample will therefore be at a constant temperature. This method is not yet standardized. (Semi)-adiabatic calorimetric. This method consists in the introduction of a sample inside a calorimeter and measuring the temperature variation. In this method heat losses are allowed, and accounted for by comparison to a reference calorimeter containing a fully hydrated mortar of equivalent thermal mass. It is classified in two categories, depending on the type of insulation: ○
Langavant semi-adiabatic calorimetric: uses a dewar flask to insulate. This method is normalized in the standard EN 196-9 (Methods of testing cement- Part 9: Heat of hydration-Semi adiabatic method) (38).
○
Adiabatic test which is totally isolated from the environment.
3. TEST PROGRAM
39
At the RMA Concrete Lab, the hydration heat is measured through Langavant semiadiabatic calorimeter. This device is based on the (Semi)-adiabatic calorimetric method, which is normalized in the standard Test data from a semi-adiabatic calorimeter provides a means to quantify the heat of hydration development as the hydration of the mixture progresses. The test consists of placing a specimen of fresh mortar in a heat-insulated box (calorimeter) in order to determine the quantity of heat released in accordance with the development of the temperature. At a given point, the heat released by hydration equals the cumulative heat input into both the calorimeter and specimen plus the heat that has dissipated to the outside since the initial time. Temperature rise of the sample will depend upon the total heat generated, the heat flux generated and the thermal efficiency of the system. The heat of hydration is expressed in joules per gram of cement. The temperature of the mortar is compared with the temperature of fully hydrated sample of equivalent thermal mass in a reference calorimeter and it shall be considered to be the ambient temperature and shall be maintained during the test within ± 0.5 ºC.
A P P A R A T US
Figure 3.10. Langavant semi-adiabatic calorimeters in the RMA Concrete Lab
a) Calorimeter It is an isolated flask sealed with an insulated stopper and encased in a rigid casing which acts as its support. Its heat losses are less than 100 J/(hK) and is defined by RILEM TC 119 TCE 1 (1997). During the test, the distance between each of the calorimeters shall be approximately 12 cm. The velocity of the ventilation air around the calorimeters shall be less than 0,5 m/s-1. 3. TEST PROGRAM
40
b) Reference Calorimeter It has the same characteristics that another calorimeter but it contains a mortar box in which there is a sample of mortar mixed at least 12 months previously (and is considered to be inert).
c) Platinum resistance thermometers The hydratation heat inside the sample in the calorimeter device is measured by a PT100 Platinum Resistance Thermometer (PRTs) from Pico® Technology. The principle of operation is to measure the resistance of a platinum element. It has a resistance of 100 ohms at 0 °C and 138.4 ohms at 100 °C. The relationship between temperature and resistance is approximately linear over a small temperature range: for example, the error at 50ºC is 0.4ºC if it is linear in the range of 0ºC to 100ºC. The most recent definition of the relationship between resistance and temperature is International Temperature Standard 90 (ITS-90).
Figure 3.11. PT100 Platinium Resistance Thermometer
This linearization is done automatically by commercial software, when using Pico signal conditioners. The linearization equation is: ) )
(7)
Where:
Rt is the resistance at temperature t, R0 is the resistance at 0 °C, and
A= 3.9083 E-3
B = -5.775 E-7
C = -4.183 E -12 (below 0 °C), or
C = 0 (above 0 °C)
3. TEST PROGRAM
41
The thermometer data is sent to the computer through the PT-104 data logger by Pico® Technology, which is a high-resolution temperature converter. It can be used to measure temperature, resistance and voltage.
Figure 3.12. PT-104 Data logger
d) Mortar box Cylindrical container fitted with a cover, having a volume of approximately 800 cm3, designed to contain the sample of mortar under test. The lid of the mortar box is fitted with a cylindrical pipe, placed to put up the platinum thermometer.
Each mortar box is
discarded after each test and should be impermeable to water vapor.
Figure 3.13. Storage of used mortar boxes
T e s t p r o c ed u r e One standard mixture is made for each Series (Open Storage and Vacuum Package). Then, following the EN procedure, the mortar box is filled with the mixture and introduced in the calorimeter. The weight of the empty and full box should be recorder.
3. TEST PROGRAM
42
A f t e r t h e t e s t : C a l cu l a t i o n o f t h e h ea t o f h y d ra t i o n The heat of hydration Q at elapsed time , can be calculated using the equation (38): (8)
∫
Where: = Heat of hydration (J/g) = Mass of cement contained in the test sample (g) = Hydration time (h) = Total thermal capacity of the calorimeter (J/K) = Coefficient of heat loss of the calorimeter (J/hK) = Temperature rise of the test sample at time t (K)
The first term in equation (9) represents the heat accumulated in the calorimeter A, and the second term represents the heat lost into the ambient atmosphere B: (10)
Where: = Heat accumulated in the calorimeter (J/g) = Heat lost into the environment (J/g) Heat accumulated in the calorimeter
As it was seen before in equation (11), the
heat accumulated is calculated using: (12)
The total thermal capacity of the calorimeter
depends upon the composition of
the test sample and the properties of the calorimeter: (13)
)
Where: = Thermal capacity per unit of mass of cement plus sand (J/Kg) = Average thermal capacity per unit of mass of water (J/Kg) = Thermal capacity per unit of mass of the mortar box (J/Kg) 3. TEST PROGRAM
43
= Thermal capacity of the empty calorimeter (J/K) = Mass of cement (g) = Mass of water (g) = Mass of empty mortar box plus lid (g) Heat lost into the environment
It can be calculated for known periods of
hydration represented by the time elapsed between successive measurements of the temperature of the test sample. As it was presented in equation (14), heat lost into the environment is: (15)
∫
This can be simplified in: ̅
∑̅
(16)
Where: = Elapsed time between the measurement of temperature at point in time, and the next measurement at point in time,
) ,(h)
̅ = Average of the temperature rise of the test sample, between times ),
),
)
and
(K): )
̅
(17)
̅ = Average coefficient of total heat loss of the calorimeter in the period of time (J/hK): ̅
̅
(18)
Where: and
= calorimeter calibration constants
= temperature (K) Finally, the hydration heat flux is defined as follows: (19)
3. TEST PROGRAM
44
Where: = hydration heat flux (J/gh)
M a t u ri t y M e t h o d Maturity method is a technique to account for the combined effects of time and temperature on the strength development of concrete during the curing period. For each concrete mixture, the relationship between strength and the maturity index is established beforehand. Saul presented the following principle that has become known as the maturity rule:
“Concrete of the same mix at the same maturity (reckoned in temperature-time) has approximately the same strength whatever combination of temperature and time goes to make up that maturity.” The function of the equivalent age of concrete is:
∑
(
)
(20)
Where: = Equivalent age at the reference temperature = Apparent activation energy (J/mol) = Average temperature of concrete during time interval
(K)
= Specified reference temperature (K) = Time interval (hours) = Universal gas constant. Reference temperature
is traditionally taken in Europe as 20ºC. The activation
energy determines the overall effect of temperature within the maturity function and values from 40 000 -45 000 J/mol for concrete containing Type I cement without admixtures is recommended by ASTM C 1074 (2004). It was found that, given the apparent activation energy of a mixture, the maturity method was able to greatly reduce the variability in setting times caused by temperature differences. An example of the maturity approach as applied to penetration data can be seen in the following Figure (5):
3. TEST PROGRAM
45
Figure 3.14. Maturity function applied to ASTM C 403 penetration resistance readings (adapted from (39)
H e a t o f h y d r a t i o n a n d S et t i n g t i m e Initial set has been traditionally linked to somewhere between stage two and three of the heat of hydration evolution. In the same way, final set has been linked to a point between stage three and stage four.
In 2007, Sandberg and Liberman (40), basing on
previous publications and their own testing, proposed two methods to obtain the set time of a fresh mixture according to the temperature data collected with the use of an insulated thermal testing device, the Fractions and the Derivatives Methods. Schindler, 2004 (41), also proposed a new method based on the well-known Maturity Method: 1) Fractions Method 2) Derivatives Method 3) Matury-based Model
3. TEST PROGRAM
46
Fractions Method Initial and final setting times are defined as percentages of the total semi-adiabatic temperature rise of a specimen. Under standard laboratory curing conditions, default values of 21% and 42% are considered as initial and final set respectively.
Figure 3.15. Initial and final set as defined by the Fractions Method.
Derivatives Method Initial set is defined as the maximum curvature (second derivative) of the main alite (C3S) hydration peak and final set is defined as the maximum slope (first derivative) of the main alite (C3S) hydration peak. A plot of the first and second derivative of hydration data can be seen in the following figure along with points defined as initial and final set.
Figure 3.16. Initial and final set as defined by the Derivatives Method.
3. TEST PROGRAM
47
3.2.3
U l t r a s o n i c t e s t : P- w a v e T r a n s mi s si o n V e l o c i t y
During the last decades, non-destructive techniques, which already had been used for damage assessment of hardened concrete, also started to serve to continuously monitor the changing material properties of early-age concrete. Measurement of the ultrasonic wave velocity through mortar or concrete allows the observation of the setting behavior. These velocity curves actually show the development of the elastic properties such as Young’s modulus and the Poisson’s ratio. It has been also demonstrated for Portland cement concrete without additions that the changes in the ultrasonic wave velocity are related to the hydration degree during the first 24 h and are affected by the water-to-cement ratio (42) (4). Three types of propagation mechanical waves (also called stress waves) are created when the surface of a large solid elastic medium is disturbed by a dynamic or vibratory load: compressional waves (also called longitudinal or P-waves) shear waves (also called transverse or S-waves) and surface waves (also called Rayleigh waves) For a given solid, compressional waves have the highest velocity and surface waves the lowest. In concrete, the velocities of the shear and surface waves are typically 60 and 55%, respectively, of the compressional wave velocity The particular velocity of a wave depends on the elastic properties and density of the medium. For elastic, homogeneous solid media the compressional wave velocity is given by the following (43): (21)
√
Where = compressional wave velocity (it typically ranges for concrete from 3000 to 5000 m/s) =
)
)
))
= dynamic modulus of elasticity = density = dynamic Poisson’s ratio When a propagating wave pulse impinges on an interface with a medium having distinct material properties, a portion of the wave energy is scattered away from the original wave path. For example, voids, cracks, and aggregate particles in concrete act to scatter 3. TEST PROGRAM
48
some of the initial energy of the compressional wave pulse away from the original wave path. The magnitude of the scattering is especially intense if the wavelength of the propagating wave is the same size or smaller than the size of the scatterer, resulting in rapid wave attenuation. For concrete, the upper limit of usable frequency is about 500 kHz as the associated wavelength is approximately 10 mm, which is in the size range of the coarse aggregate particles. As a result, the path length that can be effectively traversed at this upper limit of frequency before the wave pulse becomes completely scattered is only several centimeters. Greater path lengths can be traversed using lower frequencies (thus larger wavelengths): a frequency of 20 kHz can usually traverse up to 10 m of concrete. In the ultrasonic pulse velocity test method, an ultrasonic wave pulse through concrete is created at a point on the surface of the test object, and the time of its travel from that point to another is measured. Knowing the distance between the two points, the velocity of the wave pulse can be determined. It has been reported that the type of cement did not have a significant effect on the pulse velocity (44). The rate of hydration, however, is different for different cements and it will influence the pulse velocity. As the degree of hydration increases, the modulus of elasticity will increase and the pulse velocity will also increase. The use of rapid-hardening cements results in higher strength for a given pulse velocity level. About the effect of water− cement (w/c) ratio on the pulse velocity it has shown that as the w/c increases, the compressive and flexural strengths and the corresponding pulse velocity decrease assuming no other changes in the composition of the concrete (45). The effect of age of concrete on the pulse velocity is similar to the effect on the strength development of concrete. Jones18 reported the relationship between the pulse velocity and age. He showed that velocity increases very rapidly initially but soon flattens. This trend is similar to the strength vs. age curve for a particular type of concrete, but pulse velocity reaches a limiting value sooner than strength. He further concluded that once the pulse velocity curve flattens, experimental errors make it impossible to estimate the strength with accuracy. Temperature variations between 5 and 30°C have been found to have an insignificant effect on the pulse velocity.
W a v e s a n d p r o c e s s o f h y d ra t i o n The point where the ultrasonic velocity starts to increase corresponds to the end of the dormant period (see Figure 2.11) and the formation of a percolation cluster of solid 3. TEST PROGRAM
49
particles. The subsequent rapid increase in velocity follows the rapid change in the connectivity of the solid phase. On the other hand, it has been study that the evolution of the ultrasonic velocity is significantly influenced by acceleration admixtures and additions such as fly ash (46).
Waves and setting time The fraction of connected solid particles (solid percolation fraction) is the most important microstructure parameter determining the chance of the wave velocity. During setting, the cement hydrates star to percolate and form complete pathways of connected particles for the ultrasonic pulse wave. After percolation, the formation of additional hydration products still increases the elastic moduli and thus the wave velocity. 1) During the dormant stage, the velocity is characterized by a constant low velocity value and is mainly determined by the air content of the mixing water and the amount of air bubbles entrapped during mixing. Theoretically, a small amount of entrapped air (1%) can decrease the initial velocity measured on concrete (w/c = 0.5) from 2100 to 250 m/s. 2) After the dormant stage, velocity increases rapidly due to two phenomena: a.
Migration of air bubbles to the surface due to bleeding an percolation of the solid phase by which more
b. More solid particles are bound together by the formation of hydration products 3) During the deceleration stage, the ultrasonic velocity increases merely gradually when a completely connected solid framework has been formed. The further change of the velocity then follows the total fraction of solid phase. Finally, the velocity reaches an asymptotic value. In contrast to Penetrometer Test, acquisition of setting time IS NOT STANDARIZED. In contrast, several researchers have proposed different points that could correspond to IST and FST. Setting times will be the mortar age correspondent to the following points of pwave velocity or p-wave velocity gradient:
Initial set P-wave velocity curve
1500 m/s threshold
Final set
P-wave gradient velocity curve First inflection point
Maximum inflection point
P-wave velocity curve
2975 m/s threshold
2/3 maximum velocity
P-wave velocity gradient 20% maximum gradient Second inflection point 3. TEST PROGRAM
50
Mean of Inflection points Table 3.6. Setting time indicator for P-wave Transmission Test.
Figure 3.17. Indicator points, depending the author, for Initial set (in red) and Final set (in blue)
Apparatus For this series of experiments the multi-channel \Ultrasonic Multiplex IP-8 tester manufactured by UltraTest GmbH is used.
Figure 3.18. Ultrasonic Test at the RMA Construction laboratory
Procedure Each specimen is placed in a silicon mold (70mm diameter and 60mm high) with an ultrasonic sender and a receiver positioned diametrically opposite to each other. At 1 minute intervals, the data acquisition (DAQ) card of the computer generates an electric pulse which is sent through the amplifier to the ultrasound transmitter. This transducer incorporates a piezoelectric element, which converts electrical signals into mechanical 3. TEST PROGRAM
51
vibrations initiating the ultrasonic wave. After travelling through the cement-based sample, the ultrasonic wave is detected by the ultrasound receiver and reconverted to an electrical signal which is sent back to the DAQ card through a preamplifier. P-wave velocity is then calculated from the transmission time which is measured and the distance between the two transducers.
3.3 Previous results William (10) started to study the effect of vacuum conservation method with two different techniques: using silica-gel pouches in the vacuum bags (CUB series) and without using them (CUA series). Three different experiment techniques were used: ○
Semi-Adiabatic Calorimetric
○
P-wave Transmission Test
○
Isothermal Calorimetric The results were better for the bags without using silica-gel pouches (CUB) and
retardation in setting times for unpackaged cement (CUC) rises to about 80 minutes after 13 weeks, meanwhile for the CUB series this retardation is not considerably, showing the efficiency of the vacuum packaging.
Figure 3.19. Mortar age at qmax (SA method) in function of the time since packaging (cement age).
García (8) has studied the effect of vacuum in terms of Initial and Final setting, from 0 to 10 weeks of cement age. For having validated results, three different tests were performed: ○
Penetrometer test
○
P-wave Transmission test
○
Semi-Adiabatic Calorimetric Method 3. TEST PROGRAM
52
An overview of his final results is shown below:
b) Final Setting
a) Initial Setting Open
Vacuum
Open Mortar Age (min)
Mortar Age (min)
Vacuum
375
270 245 220 195 170
350
325
300
275 0
2
4
6
Cement Age (weeks)
8
10
0
2
4
6
8
Cement Age (weeks)
10
Figure 3.20. Test results for Setting Times obtained with Penetrometer standard ASTM C403 by García Cortés, 2012
b) Final Setting
a) Initial Setting
Open
Vacuum 560
250
540
Mortar Age (min)
Mortar Age (min)
Open 270
230 210 190 170
Vacuum
520 500 480 460
0
2
4
6
8
10
0
Cement Age (weeks)
2
4
6
Cement Age (weeks)
8
10
Figure 3.21. Test results for Setting Times obtained with Penetrometer standard ASTM C403 by García Cortés, 2012
a ) Fin a l Se t t in g Open
Vacuum
Mortar Age (min)
600 580 560 540 520 500 0
2
4
6
8
10
Cement Age (weeks) Figure 3.22. Semi-adiabatic Test results.
3. TEST PROGRAM
53
As it is illustrated, Initial and Final setting times appear later for open cement than for vacuum one. It can be noticed that it is from the second week of cement age when the difference of one or another setting times remains constant (in the 10 th week other difference is also observed just for the initial setting time). Therefore, there should be another study during the first two weeks of cement age.
3.4 Test schedule According to the research objective, intensive testing should be carried during the first days of cement age for the monitoring of the early hydration process of the two series to compare: Open storage and Vacuum Package one. As a result, tests will be done with a daily frequency during the first week of cement age, every two days during the second week, weekly for the third and fourth week and then every fourteen days until finishing all the stock of vacuum cement which lasted till two months and 9 days after the beginning of the monitoring.
SCHEDULE 22 NOV 2012 - 31 ENE 2013 Total: 3SA. 4UT. 4PE Start Date
22/11/2012
23/11/2012
24/11/2012
25/11/2012
26/11/2012
Series 2
Start hh.mm
Aging (h)
11.18
0.03
12.37
1.22
14.46
3.31
1(1)
14.59
3.44
1(2)
9.45
23,92
10.28
23,97
11.40
24,02
12.04
24,03
13.50
48,09
14.21
48,13
15.02
48,16
1(1)
15.19
48,17
1(2)
17.09
48,25
14.08
72,10
14.38
72,14
8.52
95,88
9.10
95,91
10.28
95,95
10.43
95,98
10.53
95,98
11.18
95,98
IP8
SA
Series 1 PE
IP8
SA
PE 2
2(1,2)
2 2 2(3,4) 1(3) 2 2
2(1,2) 2 2 2 2 2(1,2) 2(3,4) 1(3) 1(1) 3. TEST PROGRAM
54
28/11/2012
30/11/2012
03/12/2012
05/12/2012
12/12/2012
18/12/2012
19/12/2012
09/01/2013
31/01/2013
10.13
143,94
10.34
143,97
11.08
143,97
11.17
144,00
11.36
144,01
14.18
144,11
1(1)
14.41
144,12
1(3)
10.41
191,96
11.09
191,98
13.24
192,07
13.51
192,11
14.39
192,14
15.30
192,16
10.58
263,97
11.32
263,99
12.22
264,03
14.17
264,11
14.36
264,12
14.54
264,13
9.23
311,90
9.42
311,91
10.18
311,94
10.30
311,95
10.39
311,95
10.49
311,96
11.38
480,00
12.03
480,03
13.46
480,08
14.05
480,12
14.25
480,13
14.41
480,12
11.36
623,99
11.56
624,03
13.43
624,10
14.03
624,10
11.27
647,99
11.47
648,00
13.16
1152,06
13.48
1152,09
14.34
1152,12
14.50
1152,13
15.25
1152,15
15.47
1152,17
10.31
1679,95
11.05
1679,99
14.47
1680,13
15.11
1680,16
15.51
1680,19
16.17
1680,19
2 2 2(1,2) 2(3,4) 1(2)
2 2 2(1,2) 2(3,4) 1(3) 1(1) 2 2 1(2) 2(1,2) 2(3,4) 1(1) 2 2 2(1,2) 2(3,4) 1(2) 1(1) 2 2 2(1,2) 2(3,4) 1(2) 1(1) 2(1,2) 2(3,4) 1(1) 1(2) 2 2 2 2 2(1,2) 2(1,2) 1(2) 1(1) 2 2 2(1,2) 2(1,2) 1(1) 1(2) 3. TEST PROGRAM
55
TOTAL SAMPLES
22
12
12
20
12
26
TOTAL TESTS
11
12
12
10
12
13
Figure 3.23. Schedule the complete monitoring at the RMA Concrete Lab
Where: ○
Samples in black means that the cement used in their mixtures came from bag nº1, which was open on 22/11/12 at 11:15 am.
○
Samples in red means that the cement used in their mixtures came from bag nº2, which was open on 22/11/12 at 11:45 am.
○
The numbers in brackets () show the channels used in the test for those specimens.
3. TEST PROGRAM
56
4. RESULTS 4.1 Introduction The results of the hydration behavior monitoring of the fresh mortar mixtures of the two series are presented in this chapter. The monitoring took two months and nine days of testing,
where
P-wave
Transmission,
Semi-adiabatic
Calorimeter
and
Penetrometer
techniques were carried on with different time frequencies: daily, every two days, weekly and every two weeks. The two series to compare are: Series 1: Mortar made with Open Storage Cement Series 2: Mortar made with Vacuum Packed Cement Others parameters of study are the cement age, mortar age, setting times, ambient temperature and relative humidity and the specific variables of each test. The total number of carried tests of each experimental technique is: Number of
Number of
tests
specimens tested
Open
10
20
Vacuum
11
22
Semi-Adiabatic
Open
12
12
Calorimeter
Vacuum
12
12
Open
13
26
Vacuum
12
24
Experimental Technique
P-wave Transmission
Penetrometer
Series
Table 4.1. Resume of numbers of tests carried out and the number of specimen tested
Each specimen is named according to the chapter 3.1.8 Series, where the procedure for nomenclature is explained in Figure 3.6. Two examples are showed: LAUIP25110-1 Code
Significance
Code
Researcher
LAU
Technique
IP
P-wave Transmission
PE
2511
25th November 2012
3101
Date
Laura
LAUPE3101P-34
LAU
Significance Laura Penetrometer 31st January 2013
4. RESULTS
57
Series
O
Open storage
P
Number of Specimen
1
1
34
Vacuum Packed Average
between
specimens 3 & 4
Table 4.2. Nomenclature codes for specimens
4.2 Experimental Data Treatment The experimental data acquisition has been made manually for the Penetrometer test and automatically for P-wave Transmission and Semi-adiabatic Calorimeter tests. Due to the characteristics of the Penetrometer test, just 6 – 12 data are obtained for each batch in each test but, on the other hand, around 3000 data are acquired for each specimen for Pwave Transmission and Semi-adiabatic Calorimeter tests. Because of this large number of data, the data treatment for these two last tests will be done by software MATLAB programmed codes. On the other hand, Penetrometer data will be treated with Microsoft Excel. Firstly, experimental data for the three tests will be plotted and first analyzed. No analytic treatment is needed. The graphics produced will be: Penetrometer test Penetration Resistance (MPa) vs Mortar Age (min)
P-wave Transmission test
Semi-adiabatic Calorimeter test
Wave velocity (m/s) vs Mortar Age (min)
Temperature ºC vs Mortar Age (min)
Then, the experimental data will be treated according each technique procedure in order to obtain the necessary curves for obtaining Initial and Final setting times. The curves obtained for acquiring the setting times will be: Penetrometer test Penetration Resistance (MPa) vs Mortar Age (min) (based on the average of each pair of batches)
P-wave Transmission test
Semi-adiabatic Calorimeter test
Wave velocity (m/s) vs Mortar Age (min) Wave velocity gradient (m/s2) vs Mortar Age (min)
Hydration Heat Q (J/g) Hydration heat Flux Q (J/gh)
For obtaining the new curves listed above, data treatment with mathematic analytical software is needed. In the case of Penetrometer data, setting times values will be obtained from the trendlines of the experimental data calculated with Microsoft Excel. To know how goodness the approximation is, the coefficient of determination (R2) is calculated.
This coefficient gives the proportion of the variance of one variable that is
predictable from the other variable and represents the percent of the data that is the closest to the curve of best fit. Because of this, R square assistances when choosing the best regression line for each case. Coefficient of determination is obtained from the coefficient of 4. RESULTS
58
correlation (Pearson product moment correlation coefficient) when it is multiplied by itself. The formula of the coefficient of determination is:
[∑ ∑ ̅
̅ ∑
̅ ] ̅
(1)
Where: = Experimental data x-axis coordinate ̅ = Regression curve x-axis coordinate = Experimental data y-axis coordinate ̅ = Regression curve x-axis coordinate
For the case of P-wave Transmission and Semi-adiabatic, once the experimental data is transformed to the new magnitudes (p-wave velocity gradient, accumulated Q, flux Q), the results are plotted after an interpolation between the data obtained. Logan William’s research, 2010 (L., 2011), entitled "Optimization of the methodology
for analyzing continuous data ultrasonic testing of concrete and mortar", was to study the approximation of such data by a continuous curve. He concluded that the two most reliable experimental curves for numerical approximation methods were the polynomial method and the spline. According to the previous paragraph, the approximation of the continuous data for P-wave Transmission and Semi-adiabatic tests will be calculated with the following approximation methods: Polynomial (polyfit) approximation method Splines approximation method
Figure 4.1. Interpolation methods
4. RESULTS
59
4.2.1
T h e P o l yn o mi a l A p p r o x i m a t i o n
The polynomial method consists in the approximation of the curve by a polynomial of degree , with the coefficient
for each degree
: (2)
The degree
of the polynomial has to be wisely chosen. In effect, the highest it is,
the closest the polynomial curve follows the actual data, but the peaks due to the measures noise will have more influence on the curve definition. This is shown in figure 5.4. In addition, a degree
too high could lead to Runge's phenomenon, resulting in a separation
of the curve modeled with respect to the actual curve.
Figure 4.2. Approximation of experimental data by the polynomial method
Therefore, the degree
should be chosen so as to follow the experimental data but
trying to avoid that noise has too much influence on the results. In practice, the value of is chosen between 10 and 60 optimized using the Matlab fmincon function, in order to minimize the RMSE ("root mean square error"). The RMSE measures the difference between the experimental values and the calculated by the model. It is given by the following formula: ∑ √
(3)
Where: = Number of measurements = values of p-wave velocity obtained by the approximation curve = values of p-wave velocity from the experimental data
Once the degree n is chosen, the coefficients of the polynomial are calculated with the function polyfit of Matlab.
4. RESULTS
60
4.2.2
T h e C u b i c Sp l i n e s A p p r o xi m a t i o n
A spline is a function piecewise defined by different polynomials. This method is usually preferred rather than polynomial’s one because it is possible to achieve similar results, while using lower degree polynomials. This avoids Runge’s phenomenon.
Figure 4.3. Approximation of experimental data by the splines method
4. RESULTS
61
4.3 Penetrati on Resistance Test: Penetrometer A total of 50 specimens have been tested during the 2 months and 9 days. The plot for all the results can be found below. What it is represented, it is the penetration resistance in MPa versus the mortar age (time from water addition) in minutes. Open storage series (Series 1) is plotted in cold colors (in chronologically order: yellow-green-blue-purple) and Vacuum Packed series (Series 2) is shown in warm colors (in chronologically order: orangered-pink).
35
30
Penetrometer Resist. (Mpa)
Penetration Resistance vs Mortar Age
LAUPE2211O LAUPE2311O-12 LAUPE2311P-34 LAUPE2411O-12 LAUPE2411P-34 LAUPE2511O-12 LAUPE2511P-34 LAUPE2611O-12 LAUPE2611P-34 LAUPE2811O-12 LAUPE2811P-34 LAUPE3011O-12 LAUPE3011P-34 LAUPE0312O-12 LAUPE0312P-34 LAUPE0512O-12 LAUPE0512P-34 LAUPE1212O-12 LAUPE1212P-34 LAUPE1912O-12 LAUPE1912P-34 LAUPE0901O-12
25
20
15
10
5
0 200
220
240
260
280
300
320
340
360
380
400
420
Mortar age (min) Figure 4.4. Representation of the penetration resistance obtained each test day for Series 1 (O) and Series 2 (P) cement. Each line represents the average between the points obtained by the two samples for each mixture (12 or 34)
Regarding the Figure 4.4, it can be observed that the curves are displaced to the right in time as later the specimen’s date is. This displacement is more remarkable for the series of Open storage, as we see that the cold colors curves are which predominate at the right side of the graphic. Accordingly, the efficiency of vacuum is early detected with just observing the graphic during few seconds. Deeper study is followed showing the results of Initial and Final setting times for all the series during the entire monitoring test and then the efficiency of the vacuum is found.
4. RESULTS
62
Each of the curves above plotted are calculated from the average between the power trend equations of the two batches from the same mixture. Following the criteria of 3.1.8 Series, each curve is named in the legend like “LAUPE2311O-12”, which means that this curve comes from the average between the trendlines of the experimental results of “LAUPE2311O-1” and “LAUPE2311O-2” batches. Below, it could be seen an example of how this average is obtained: b) Pen etr ation Resistan ce vs Tim e
a) Pen etr atio n Resistan ce vs Tim e LAUPE0512P-3 Power (LAUPE0512P-3)
LAUPE0512P-4 Power (LAUPE0512P-4)
LAUPE0512P-34 35 Penetrometer Resist. (MPa)
6000 Penetrometer Resist. (psi)
Power (LAUPE0512P-34)
5000
y = 1,218E-16x7,756E+00 R² = 0,988
4000 3000 2000 1000
y = 1,868E-15x7,255E+00 R² = 0,993
30 25 20 15 10
y = 2,996E-18x7,522E+00
5 0
0 240
260
280
300 320 340 Mortar Age (min)
360
240
260
280 300 320 Mortar Age (min)
340
360
Figure 4.5. a)Experimental Penetration resistance results and trendlines and b) Average of the two trendlines of the experimental results
(It is important to notice that these power equations are just valid in the range of time -mortar age- where the experimental results have been obtained, because we cannot predict the penetration resistance behavior in other ranges of mortar age). With the two trendlines equations of the Figure 4.5 a), penetration resistance values are obtained for mortar ages between 260-340 minutes. Then, the average for each pair of penetration resistance values is calculated and an average power equation is obtained, which is represented in the Figure 4.5 b). With this equation Initial and Final setting time for this mixture are acquired. The average equation (Figure 4.5 b)) is equaled to the specified value for Initial or Final setting time in MPa “y”, according to the standard ASTM C403. When “x” is isolated, time of mortar age in minutes is obtained for the setting time.
Procedure 1)
4. RESULTS
63
4.3.1
S e r i e s 1 : O p en S t o r a g e C e m e n t
Open storage series corresponds to the cement which has not been vacuum Packed after opening the sacks from the cement plant. For each Penetrometer test, 1.8 kg of cement is taken from the commercial 25 kg cement sacks and 8 kilograms mortar mixture is made with it (this is four times an EN 196-1 standard mixture). Then, the mortar is poured and compacted in the two waterproof wooden containers (batches 1 and 2), as it is better explained in 3.2 Experimental Techniques. The following chart represents the curves calculated from the experimental data average for each Series 1 mixture, following the procedure explained in the Figure 4.5 a) and b).
Series 1 35
LAUPE2211O LAUPE2311O-12 LAUPE2411O-12 LAUPE2511O-12 LAUPE2611O-12 LAUPE2811O-12 LAUPE3011O-12 LAUPE0312O-12 LAUPE0512O-12 LAUPE1212O-12 LAUPE1912O-12 LAUPE0901O-12 LAUPE3101O-12
Penetrometer Resist. (Mpa)
30
25
20
15
10
FST T
5
IST 0 200
220
240
260
280
300
320
340
360
380
400
420
Mortar age (min) Figure 4.6. Penetration Resistance versus mortar age for mortar mixtures prepared with Open Storage Cement. Marks corresponding to Initial Setting Time (3,45 MPa) and Final Setting Time (27,58 MPa) are also showed
As it was aforementioned, curves corresponding to older cement ages displace to the right side of the chart, so setting time for these specimens occurred later than for the specimens of the first days of testing. It can also be observed that the difference in Initial setting time between the first test and the last test is about 90 minutes, meanwhile for Final setting time this difference reaches the 100 minutes. Initial and Final setting time values are calculated using the Procedure 1) with the equations of the experimental data average curves and displayed in the table below. To see how the setting times change according to the date of test (cement age), the following two variables are calculated:
4. RESULTS
64
Where:
ST = Setting time, it can be IST or FST (min) = Setting time of a mixture tested on the date “i” (min) = Setting time of a mixture tested on the test date before the date “i” (min) = Setting time of the mixture tested on the first date of the monitoring (min)
According to 2.6 Influence of Laboratory Conditions in Setting Time, temperature and relative humidity can affect the concrete setting time, so the evolution of temperature and relative humidity in the RMA Concrete lab during the test days has been recorder in order to contrast possible influences of the lab conditions in the results of setting time. To help the study of the evolution of the setting time and its comparison with the values of laboratory conditions, a color code is used to remark when the values increase or decrease, to show up any possible correlation between them.
Low increase
Significantly increase
Moderate increase
Big increase
Date
22nov 23nov 24nov 25nov 26nov
Moderate decrease
28nov
30nov
03dic
191,96 263,97 311,90
Cement Age(h)
0,0
23,92
48,09
72,1
95,88
143,94
Cement Age(d)
0
1
2
3
4
6
8
11
(min)
231
227
226
231
234
236
233
-4
-1
5
3
1
-4
-5
0
3
45,6
54,1
51,6
-0,5
8,5
19,7
(min) (min) RH%
46,1 (%)
TºC
20,5 (ºC) (min) (min) (min)
312
Low decrease
05dic
12dic 19dic
09ene
31ene
480
648
1152,1 1679,9
13
20
27
48
70
257
277
269
278
299
320
-2
24
20
-8
8
22
20
4
2
26
46
38
47
68
89
46,5
47,2
47,2
39,5
41,6
30,6
47,7
43,9
45,3
-2,5
-5,1
0,7
0
-7,7
2,1
-11
17,1
-3,8
1,4
19,7
19,8
20,4
20,3
20,3
19,4
19,1
18,8
19,2
19,3
19,6
-0,8
0
0,1
0,6
-0,1
0
-0,9
-0,3
-0,3
0,4
0,1
0,3
313
307
309
307
307
310
341
361
349
353
390
413
0
-6
2
-2
0
3
31
20
-12
4
37
23
0
-6
-3
-5
-5
-3
29
48
37
41
77
100
Table 4.3. Results of the monitoring of the setting time for the Series 1
4. RESULTS
65
b) Setting Times vs Relative Humidity
a) Settin g Tim es vs Tem p er atu r e FST
TºC
IST
FST
RH% 60
21 400
400
55
20 20
300
19 250
Mortar Age (min)
350
T ºC
Mortar Age (min)
21
50 350
45 40
300
35 30
250
19
200
18 0
1
2
3
4
6
25 200
20 0
8 11 13 20 27 48 70
1
2
3
4
6
8 11 13 20 27 48 70
Cement Age (days)
Cement Age (days)
Figure 4.7. Evolution of Initial (IST) and Final (FST) Setting times versus Temperature (a) and Relative Humidity (b)
After an examination of the previous results, a first early conclusion can be drawn: from the 11th day of cement age, a significantly increase in minutes is noticed in both Initial and Final setting time of mortar. Until that date of cement age, the results for setting time varied in a range of just ± 6 minutes. It was on the 11th day of testing when the IST experimented a moderate increase of 26 minutes from the first test and FST experimented also a moderate increase of 29 minutes. From that point, the setting time increased for all the tests but one, on the 12 th December. Then, the increments in setting time were significantly and at the end of the monitoring, the IST for the mixture was 320 minutes, which means a difference of 89 minutes from the first test. On the other hand, the FST was 413 minutes, which corresponds to a difference of 100 minutes from the first test. It also implies that the difference in time from the IST of a mixture and its FST also increased in 11 minutes. Influence of the laboratory atmospheric conditions on the setting time results According to the correlation of the setting time results to the evolution of temperature and relative humidity of the laboratory, conclusions are difficult to draw during the first 8 days of testing because of the irregular setting results. What is almost clear it is that temperature has not considerable influence on the setting results, mostly because during all the monitoring it just varied ±1.7ºC. Moreover, it was a different situation for relative humidity, which varied in the range of [30.6% – 50.1%]. Therefore, it may explain the moderate decrease which broke the increasing tendency on the test of the 12th of December. On that date, IST decreased in 8 minutes, FST in 12 minutes and relative humidity had the lowest level during the whole test 4. RESULTS
66
RH %
IST
time: 30.6%. This very low humidity condition in the air could produce acceleration in the setting of the mortar due to the possible transfer of moisture from the mixture to the air. It also makes sense that the decrease in FST was higher than for IST, because for the FST, the mixture has been exposed to the atmosphere for a longer time. In addition, it can be also noticed that it is not just the actual relative humidity which has an effect on the setting, also the accumulated humidity may do. It means that for cement which had been storage without any vacuum system, the different ambient conditions it had been exposed could have an effect in its later hydration process. It can be observed, for example, on the date of 19th December, where despite the actual RH was almost 50%, the increase in setting was lower as it was expected, likely due to the really low value of %RH during the previous days.
4.3.2
S e r i es 2 : V a c u u m Pa c k e d C e m e n t
Vacuum Packed Series corresponds to the cement which was vacuumed from the first moment it was exposed to the atmosphere. As the vacuum cement was packed in 900 g plastic sealed bags, two bags are necessary to make the 4 kg mortar necessary for a mixture equivalent of 4 times the EN-196 standard mortar mixture. The following chart represents the curves calculated from the experimental data average for each Series 2 mixture, following the procedure explained in the Figure 4.5 a) and b).
Series 2 35
LAUPE2211O LAUPE2311P-34 LAUPE2411P-34 LAUPE2511P-34 LAUPE2611P-34 LAUPE2811P-34 LAUPE3011P-34 LAUPE0312P-34 LAUPE0512P-34 LAUPE1212P-34 LAUPE1912P-34 LAUPE0901P-34 LAUPE3101P-34
Penetrometer Resist. (Mpa)
30 25 20 15 10
FST
5
IST 0 200
220
240
260
280
300
320
340
360
380
400
420
Mortar Age (min) Figure 4.8. Penetration Resistance versus mortar age for mortar mixtures made by Vacuum Packed Cement. Marks corresponding to Initial Setting Time (3.45 MPa) and Final Setting Time (27.58 MPa) are also showed
4. RESULTS
67
What comes in sight in first place it is that the curves correspondent to the later tests do not have a displacement as significant as it was for the open storage series. It means that despite the age of the older cement, setting time does not vary as much as it does for the Series 1. It can be observed that IST has a variation of about 25 minutes and FST of almost 40 minutes, which is much lower than for the Series 1. The calculations and procedure followed to study the Series 1 are also applied for Series 2 and represented in the table below. It is important to notice that the monitoring of this series starts with the same experimental results than in the open series, because the cement used for both series was the same on the first day (new purchased cement).
Date Cement Age (h) Cement Age (d) (min)
22nov
23nov
24nov
25nov
26nov
28nov
30nov
03dic
05dic
12dic
19dic
09ene
31ene
0,0
23,97
48,13
72,14
95,91
143,97
191,98
263,99
311,91
480,03
648
1152,1
1680,0
0
1
2
3
4
6
8
11
13
20
27
48
70
231
236
233
231
233
238
239
245
252
252
261
242
259
5
-3
-2
1
5
1
6
6
0
10
-20
17
5
2
0
1
7
8
14
21
20
30
10
28
45,6
54,1
51,6
46,5
47,2
47,2
39,5
41,6
30,6
47,7
43,9
45,3
-0,5
8,5
-2,5
-5,1
0,7
0
-7,7
2,1
-11
17,1
-3,8
1,4
19,7
19,7
19,8
20,4
20,3
20,3
19,4
19,1
18,8
19,2
19,3
19,6
-0,8
0
0,1
0,6
-0,1
0
-0,9
-0,3
-0,3
0,4
0,1
0,3
319
310
304
304
312
309
329
332
335
337
320
342
7
-8
-6
0
8
-3
20
3
3
2
-17
23
7
-2
-8
-8
0
-3
17
19
22
25
7
30
(min) (min)
46,1
RH% (%)
20,5
TºC (ºC) (min) (min) (min)
312
Table 4.4. Results of the monitoring of the setting time by Penetrometer test for Series 2.
4. RESULTS
68
a) Settin g tim es vs Tem p er atu r e FST
TºC
IST
21 400
FST
RH%
58
400
53
350
20
300
19 250
19
200
18 0
1
2
3
4
6
8 11 13 20 27 48 70
Cement Age (days)
T ºC
20
Mortar Age (min)
Mortar Age (min)
21
48 350
43 38
300
33 28
250
23 200
18 0
1
2
3
4
6
8 11 13 20 27 48 70
Cement Age (days)
Figure 4.9. Evolution of Initial (IST) and Final (FST) setting times by Penetrometer test versus Temperature (a) and Relative Humidity (b) for Series 2
In contrast to the Series 1, Vacuum Packed Series results do not show such a clear tendency of setting time displacement. It is difficult to select a date when setting behavior starts to change according to cement age, but there are certain points that can be discussed. For Initial setting time value, it varies during the first 8 days of testing, date when it starts to slightly increase until the day 27th of testing, when it increase in 10 minutes more, to then have a big decline of 20 minutes and Finally to increase in another 17 minutes, which means a total growth of 28 minutes from the first day of testing. In disparity, Final setting time value follows a negative tendency until the 8 th day, when it suddenly experiences a high increase of 20 minutes. After it, the tendency turned positive but with slight increases until the 9th January when, as well as IST, FST also had a big decline of 17 minutes to then finishes with another big increase of 23 minutes. Finally, in the last test, FST was 342 minutes of mortar age, which corresponds to an increase of 30 minutes from the first day. Influence of the laboratory atmospheric conditions on the setting time results Regarding the laboratory ambient conditions, it can be clearly observed that the actual or accumulated relative humidity does not have any noticed effect on the setting value, because there is not any correspondence between the tendencies of setting and RH. In relation to temperature, as well as in the Series 1 case, it does not have any impact on the results. According to the large drop for both IST and FST on the 9th January, no clear explanation can be given. If these values are compared with the results on the same date 4. RESULTS
69
RH %
IST
b) Setting times vs Relative Humidity
for open storage series, it can be seen that there is not any correlation between them. In relation to the actual ambient conditions, it seems that they neither affect the setting, because there is not any significant change. Indeed, the relative humidity of the previous test day experienced a very high increase, so if it would have influenced, it should have increase the setting time.
In conclusion, a possible answer to this considerable drop in
setting time could be a result of a deficient vacuum.
4.3.3
C o m p a r i so n b e t w e en O p e n S t o r a g e an d V a c u u m Packed Series
An abstract of the previous results for Series 1 and Series 2 is shown in the following figure, where Initial and Final setting time are plotted. It is also accompanied by the representation of the evolution of relative humidity (%) because it is the laboratory condition supposed to influence the results of penetration resistance.
Setting Times vs Cement age IST SERIES1
FST SERIES1
IST SERIES2
FST SERIES2
RH%
450
60
400
50 45
350
40 35
300
RH%
Mortar Age (min)
55
30 25
250
20 15
200
10 0
1
2
3
4
6
8
11
13
20
27
48
70
Cement age (days) Figure 4.10. Setting times for Series 1 and 2 by Penetrometer test and Relative Humidity versus Cement age. Notice that the intervals for the variable “cement age” do not have the same length in the x-axis
Regarding this figure, the different tendencies can be observed and also the moment when the setting significantly starts to delay because of the cement age. It can be easily noticed that during the first week, there is no correlation between the age of the cement and the setting times of the mortar mixture. However, it can be perceived how from the day 8th there is a positive correlation between the two variables of the study (when Cement age and Mortar Age): Series 1 trend ( inclination than Series 2 (
,
,
) experiments higher degree of
). Moreover, it can be seen how the changes in relative
humidity may also have big influence on the setting values, and how this influence is much bigger in the Open Storage Series than in Vacuum Packed one (see the decrease on the day 20th). 4. RESULTS
70
Splitting the previous figure from the date when the tendencies for the two Series start to separate (day 8th), we can calculate the regression curves for the evolution of the Initial and Final setting times according to the cement age. Therefore, if we introduce one value of cement age in the regression curve, we can approximately predict the setting time of a mixture, as long as it has the same characteristics described in this study and a cement age included in the range 8 to 70 days old.
Setting Times vs Cement Age IST SERIES1
FST SERIES1
IST SERIES2
FST SERIES2
Log. (IST SERIES1)
Log. (FST SERIES1)
Log. (IST SERIES2)
Log. (FST SERIES2)
Mortar Age (min)
450
400
y = 38,832ln(x) + 240,56 R² = 0,8543
350
y = 7,2777ln(x) + 306,83 R² = 0,2699
300
y = 33,113ln(x) + 174,93 R² = 0,8881
250
y = 5,0075ln(x) + 234,57 R² = 0,228
200 0
10
20
30
40
50
60
70
Cement age (days) Figure 4.11. Evolution of Setting times by Penetrometer test in function of Cement age from the day 8th of testing
It has been shown that logarithmic correlation shows the better approximation between the trendline and the setting times obtained from the experimental data from the day 8th of testing. The coefficient of determination R 2 has also verified it, mostly for Series 1, because R2>8,5 for both cases (IST and FST). On the other hand, although logarithmic correlation offers also the better regression type for Series 2 over other types of regression curves like exponential, linear or polynomial, the R 2 is still very low (R2≈2,5). This is due to the fact that there is not solid direct correlation between the variables of cement age and mortar age for Series 2. The fact that the better regression curve for Series 1 is the logarithmic expresses that just after the day 8th of cement age, the rate by which setting times increase is higher than the rate of a mixture using older cement. Setting Time
Equation
R2
Series 1: Open Storage Cement IST
y = 33,113ln(x) + 174,93
0,888
FST
y = 38,832ln(x) + 240,56
0,854 4. RESULTS
71
Series 2: Vacuum Packed Cement IST
y = 5,0075ln(x) +234,57
0,228
FST
y = 7,2777ln(x) + 306,83
0,267
Table 4.5. Logarithmic regression curves’ equations for Initial and Final setting times of the two Series, where “x” corresponds to the cement age and “y” means the respectively setting time. Functions valid just for the range x = [8, 70] days
In the table below, a summary of the main results of setting time for both Series are presented in order to evaluate the importance of the changes in the setting times according to the cement age: IST
Series
FST
t = 0days
t = 11days
∆
t = 70days
∆
t = 0days
t = 11days
∆
t = 70days
∆
1
231
257
26
320
89
312
341
29
413
101
2
231
245
14
259
28
312
329
17
342
30
12
61
12
71
Table 4.6. Initial and Final setting times for Series 1 and Series 2 on the first and last day of the monitoring
At the end of the all hydration monitoring, we observe that the Initial setting time for the Open Storage Series increases in 89 minutes (around 1 h 30 min) and Final setting time increases in 101 minutes (around 1 h 40 min) what supposes important differences to take into account in research or construction procedures. As it has been shown, setting times for Vacuum Packed Series do not change so much: 28 minutes for Initial setting time and 30 minutes for Final setting time. Although it is not usual that neither in laboratories or construction sites use cement of 70 days old, it has to be highlighted that cement with more than 8 days old has already experimented changes in setting time enough important regarding concrete researches or other applications, because it means that the process of cement hydration changes (highly important for researches in fresh concrete) with also possible consequences in the final strength and properties of the hardener concrete (important for researches in hardener concrete). This change, for Series 1 it is an increase of 26 minutes for Initial setting time and an increase of 29 minutes for Final Setting time. For the Series 2, the increase is lower, but still notorious for the applications aforementioned.
4. RESULTS
72
4.3.4 C o m p a r i so n w i t h t h e R e s u l t s o f Pr e v i o u s R e s e ar c h e s The results of the hydrations monitoring of the standardized mortar mixture with the evaluation of the setting times obtained by the Penetrometer standard ASTM C403 can be compared to the results obtained in (García Cortés, 2012). In this study, the monitoring of the hydration behavior also lasted until 70 days of cement age, but the setting time values for the last test are not very trustworthy as they experiment a considerable drop regarding the tendency. This could have happened because a drop in the relative humidity of the laboratory, but there are not records to verify it. Therefore, we took the previous test date, the day 56th, as well as the results for the tests on the day 0 and the immediately following, the 14th. As there are not experimental results for the days 14 th and 56th of monitoring in this study to compare with the results of García Cortés, 2009, the regression curves’ equations of Table 4.5 are going to be used to obtain the approximation of setting times for those dates,
as it can be seen in Table 4.7.
b) Final Setting
a) Initial Setting Open
Open
Vacuum 375 Mortar Age (min)
270 Mortar Age (min)
Vacuum
245
220
350
325
300
195
170
275 0
2
4 6 Cement Age (weeks)
8
10
0
2
4 6 Cement Age (weeks)
8
10
Figure 4.12. Test results for Setting Times obtained with Penetrometer standard ASTM C403 by García Cortés, 2012
Series
1 2
IST t = 0days LAU
JM
231
228
FST
t = 14days
t = 56days
t = 0days
t = 14days
t = 56days
LAU
JM
LAU
JM
LAU
JM
LAU
JM
LAU
JM
262
262
308
256
374
397
361
239
255
234
331
343
248
312
326
340
336
331
Table 4.7. Comparison of this study’s results (LAU) with García Cortés, 2012 (JM) of the setting times for Series 1 and Series 2 for t = 0 days, t = 14 days (2 weeks) and t = 56 days (8 weeks) of cement age
4. RESULTS
73
Despite the results of this study for open series start to diverge and increase in setting time from the day 8th-11th, results for García diverge on the 2nd week and then they stabilize around the same value, both for Initial and Final setting time. It means that for both cases, García monitoring setting time for open series increased just around 30-40 minutes from week 0 to week 8th, meanwhile for the present research setting time increased 90-100 minutes. As there is not information about ambient conditions during Garcia’s tests, not conclusions can be drawn for this difference in the behavior of open storage series. About Series 2, penetration resistance results for García showed a difference in setting between week 0 and week 8 very low, about just 10 minutes, and sometimes even less than the setting on the week 0. Maybe it is due to the fact that García’s cement bags were better vacuumed.
4.3.5 V a c u u m E f f i c i en c y It has been demonstrated that vacuum packaging does not totally preserve the cement from the aging but convincing efficiency over the open storage method has been noticed. The vacuum efficiency is determinate following the equation: (4)
Where: = Setting time corresponding to the Vacuum Packed Series on cement day number i. = Setting time corresponding to the Vacuum Packed Series on the first cement day. = Setting time corresponding to the Open Storage Series on cement day number i. = Setting time corresponding to the Vacuum Packed Series on the first cement day.
To appreciate how effective the vacuum procedure has been, Table 4.8 shows the minutes saved by the vacuum in relation to the results in setting of the open storage series. Date
22nov
23nov
0
1
2
3
4
6
8
11
13
20
27
48
70
Series 1 (min)
231
227
226
231
234
236
233
257
277
269
278
299
320
Series 2 (min)
231
236
233
231
233
238
239
245
252
252
261
242
259
0
-9
-7
0
2
-2
-6
12
26
18
16
58
61
0
0
0
0
53,8
0
0
46,2
55,5
46,5
35,2
84,7
68,9
Cement Age (days)
IST
Time
saved
(min)
Vacuum Efficiency (%)
24nov 25nov 26nov 28nov 30nov 03 dic 05dic 12dic 19dic 09ene 31ene
4. RESULTS
74
Series 1 (min)
312
313
307
309
307
307
310
341
361
349
353
390
413
Series 2 (min)
312
319
310
304
304
312
309
329
332
335
337
320
342
0
-6
-4
5
3
-5
0
12
29
15
16
70
70
0
0
67,1
0
0
0
0
42,1
59,9
39,5
39,6
90,5
70,1
FST Time
saved
(min)
Vacuum Efficiency (%)
Table 4.8. Vacuum Efficiency in terms of difference of Final Setting Time between Series 1 and 2 with the results obtained by the Penetrometer test
These results confirm once more that it is during the 8th and 11th day of testing when the cement age starts to affect the setting results and it is also when the vacuum packaging starts to be efficient. It seems that this efficiency grows along the time: the results have showed that during the 11th to 27th of cement age, the efficiency for both Initial and Final setting time is around 40% and from the 48 th day it rises until values of 70% and even 90% of efficiency.
Vacuum Efficiency (%)
Vacuum Efficiency IST
100 90 80 70 60 50 40 30 20 10 0
0
4
8
FST
12 16 20 24 28 32 36 40 44 48 52 56 60 64 68 72 Cement age (days)
Figure 4.13. Efficiency of the vacuum packaging over the open storage system from the 8th day of testing. Representing results from Table 4.8
Figure 4.13 verifies that the efficiency of the vacuum system increases along the cement age and then it seems to stabilize around the 80% of efficiency and it does it in the same way for IST and FST. This tendency is logical as the evolution of the setting times versus the cement age has a logarithmic correlation, as it was esteemed in Figure 4.11. Drops in the vacuum efficiency as it can be seen for the days 20th and 27th could be caused by incomplete vacuum in the day 0.
4. RESULTS
75
4.4 Ultrasonic te st: P-wave Transmission Velocity A number of 42 specimens have been tested during the A whole research: 20 corresponding to Series 1 (Open Storage Cement) and the other 22 to Series 2 (Vacuum Packed Cement). All the specimens were tested in pairs; it means that there were two batches for each standardized mixture. Most of the times, both Series 1 and 2 were tested at the same time because the Ultrasonic Multiplex IP-8 tester allows testing four channels at the same time. The measurements of p-wave velocity are taken each minute. As it was explained in 3.2.3 Ultrasonic Test: P-wave Transmission Velocity a pulse of p-wave is emitted each minute through the fresh mortar specimen to the receiver. Furthermore, there will be as many experimental data as minutes the test lasts. The results from the whole monitoring of the hydration behavior of the mixtures by the transmission of p-waves are presented in the Figure 4.14. To aid the distinction of the curves, a color scale has been used. Curves from the same day of test are plot in the same color, even for the two different Series. Curves from the first days of testing are plot in warmer colors (yellow-magenta-red) and curves from the last days of testing are represented in colder colors (blue-cyan-green-black).
P-wave Velocity vs Mortar Age
Figure 4.14. IP8 Experimental data for Series 1 (O) and Series 2 (P) series where the velocity of the pwaves is plot versus the mortar age.
Each curve of the Figure 4.14 represents the p-wave velocity behavior for each specimen during the whole testing period. It can be seen that there are hardly any variations in the wave speed during the first 30 min after the adding of the mixing water but then how the wave propagation velocity changes rapidly during the first 12 hours of the 4. RESULTS
76
hydration to continue with a second domain where the velocity increases much slower, towards an asymptotic value of around 4750 m/s. At first sight, it can be perceived that curves in colder colors are displaced to the right, which means that the specimens of older cement experiment a certain delay in setting in relation to the specimens of new cement. Moreover, it can also be seen that curves for older cement specimens experiment a lower maximum p-wave velocity. Then, the numerical derivative of the p-wave velocity evolution is calculated as a function of time, using a first order centered difference algorithm. In what follows, the time history of this derivative will be referred to as the gradient of the p-wave velocity evolution. Figures 4.15 a) and b) show the p-wave velocity gradient plotted with the two
interpolation methods explained in 4.2 Experimental Data Treatment for the specimens LAUIP1812P-37. Despite not all the curves have exactly the same shape for all the specimens, the curves of this specimen are a representative sample of the major part of the results.
Figure 4.15. a) P-wave velocity gradient versus time using Polynomial (polyfit) approximation method. b) P-wave velocity gradient versus time using Splines approximation method.
After observation and comparison of the graphics made by the two interpolation methods, it can be noticed that both approximations are fairly similar until maximum gradient but, after that point, splines approximations present some unusual fluctuations which may give misleading results for the second maximum inflection point and 20% maximum gradient, key points to obtain setting time values. Due to this fact, polynomial approximation method has been chosen, also based on its smaller root means square error and standard deviation for the points in the proximity of the inflexion points. The p-wave velocity gradient curves, approximated with Polynomial interpolation for all tests are presented in the following figure. The order of the polynomial is optimized to obtain a root mean squared error (RMSE) 0. On the other hand, qmax slope “m” is
