Electrical effects generated by mechanical loading of hardened Portland cement paste

This work presents results on the effect of applied load on the electrical properties (resistivity and capacitance) of hardened cement paste. Impedance spectroscopy measurements in the high frequency region (100 kHz–15 MHz) show two time constants, one associated to the solid phase, and the other to the liquid filling the pores. The time constant associated with the liquid phase is sensitive to external loads acting on the specimen. The observed variations are explained in terms of electrolyte drag from interlaminar spaces and structure pores and vice versa. The importance of those electrolyte movements concerning structure fatigue and rebars corrosion is also discussed.     

Water in hardened cement paste

Summary Some properties of adsorbed water in hardened cement paste and in the three completely hydrated main constituents of portland cement have been examined by means of differential thermoanalysis, thermo-gravimetry and nuclear magnetic resonance. When a dried specimen is re-wetted, the adsorbed water first contributes to the build-up of a mono-layer. Only in the higher humidity range is the sorbed water partly bound as inter-layer hydrate water and as hydrate water. According to the first results of the nuclear magnetic resonance measurements the water adsorbed in the monolayer behaves like a two-dimensiional Van-der-Waals-gas.

---

Résumé Quelques propriétés de l'eau adsorbée dans la pate de ciment durcie et dans les trois constituants principaux complètement hydratés du ciment Portland ont été étudiées avec l'analyse thermique différentielle, la thermo-gravimétrie et la résonance magnétique nucléaire. Lorsqu'une éprouvette séchée est réhumidifiée, l'eau adsorbée contribe d'abord à la formation d'une mono-couche. L'eau adsorbée, en tant qu'eau d'hydratation, n'est partiellement liée dans la structure du sel que pour les humidités les plus élevées. Selon les tout premiers résultats fournis par les mesures de la résonance magnétique nucléaire, l'eau adsorbée dans la mono-couche se comporte comme un gaz bi-dimensionnel de Van-der-Waals.

     

Numerical homogenization of hardened cement paste

Based upon a three-dimensional computer-tomography of hardened cement paste, a finite-element mesh at micrometer length scale is introduced. Effective material properties are obtained through numerical homogenization techniques using representative volume elements. Statistical tests, two- and three-dimensional computations and a comparison with experimental data are shown. For the hydration products of hardened cement paste a visco-plastic constitutive equation of Perzyna type including isotropic damage is introduced. The inelastic material parameters are identified solving an optimization problem through a combination of a stochastic genetic algorithm and the deterministic Levenberg-Marquardt method. The time-consuming evaluations of the corresponding objective function are distributed within a network environment automatically.     

Hygral stress in hardened cement paste

A specially constructed stress cell, permitting variation of degree of restraint, was used to measure the hygral stress produced in thin-walled hardened cement paste cylinders due to water absorption. The effects of porosity, relaxation and relative humidity on the hygral stress were investigated using a Portland and a Portland composite cement. It was found that capillary suction transports water to the gel pores causing an initial rapid stress development. This is followed by a gradual increase over ca. 3 days governed by the redistribution kinetics of water molecules in the cement gel. The hygral stress developed is proportional to the volume fraction of cement gel. The cement gel itself produces a uniaxial stress of ca. 8 MPa for degrees of restraint above ca. 80%. About 70% of the stress is caused by changes in surface energy at the gel particle/pore water interface; the remainder is due to the disjoining pressure. A change in surface energy of 0.17 J/m² was estimated based on measurements of specific surface and porosity. The development of hygral stress is also controlled by stress relaxation. This appears to be enhanced by the effect of the disjoining pressure which weakens bonds between gel particles to create a more mobile structure under stress.     

Internal friction and dynamic modulus transitions in hardened cement paste at low temperatures

Summary Low-temperature dynamic mechanical measurements in the system hardened cement paste and water indicated the existence of an internal friction peak and a modulus transition. The transition occurred in the temperature range −160° to −60°C. The transition was found to be enhanced by increasing the moisture content and the extent of the internal surface. The transition is believed to be associated with a phase change of the adsorbed water. Hence the name “adsorbate transition” seems appropriate. Though the true nature of the low-temperature phase is not known, possible means of determining the characteristics of this phase are referred to and suggested. Stress induced phase transition is regarded as a possible cause for the internal friction peak. The method is regarded as having a great potential value in studying water adsorbing systems. In particular, calibration of the internal friction peak versus some other measure of the internal surface area, yields a new method for determining the extent of internal surface areas on a relative scale.

---

Résumé Les measures dynamiques à basse température faites sur le système pate de ciment durcie-eau montrent l'existence d'une valeur maximale du frottement interne et d'un changement de module. Ce changement de module se produit entre −160° et −60°C et on établit qu'il est accentué par l'accroissement de la teneur en eau et de la surface interne. L'auteur pense que ce changement est associé à un changement de phase de l'eau adsorbée. On pourrait donc la désigner comme ?transition d'adsorbat?. Quoique la nature véritable de la phase à basse température soit inconnue, l'auteur suggère des moyens de détermination des caractéristiques de cette phase. Le changement de phase induit par la contrainte est considéré comme une cause possible de la valeur maximale du frottement interne. L'auteur pense que la méthode peut avoir une grande valeur pour l'étude des systèmes à adsorption d'eau. En particulier, de l'étalonnage de la valeur maximale du frottement interne en fonction de quelque autre mesure de la surface interne découle une nouvelle méthode pour déterminer l'extension relative des surfaces internes.

     

Comparative investigation on nanomechanical properties of hardened cement paste

Three types of nanomechanical methods including static nanoindentation, modulus mapping and peak-force quantitative nanomechanical mapping (QNM) were applied to investigate the quantitative nanomechanical properties of the same indent location in hardened cement paste. Compared to the nanoindentation, modulus mapping and peak-force QNM allow for evaluating local mechanical properties of a smaller area with higher resolution. Beside, the ranges of elastic modulus distribution measured by modulus mapping and peak-force QNM are relatively greater than that obtained from nanoindentation, which may be due to a result of the shaper probe and local confinement effect between multiple phases. Moreover, the average value of elastic modulus obtained using peak-force QNM were consistent with those obtained by modulus mapping, while the different in modulus probability distribution could be related to the different nanomechancial theories and contact forces. The probability distributions of elastic modulus measured using nanomechanical methods to provide a basis for the different types of phases existing in cement paste. Based on the observation with high spatial resolution, cement paste can be likely found as nanocalse granular material, in which different submicron scale or basic nanoscale grain units pack together. It indicates that the peak-force QNM can effectively provide an effective insight into the nanostructure characteristic and corresponding nanomechanical properties of cement paste.     

Activation energy of creep of hardened cement paste

Summary By taking into account the increase of internal friction of hardened cement paste under permanent stress a relation for the macroscopic activation energy is derived, from which activation energy can be evaluated for the complete creep process. In the same way another equation for the change of activation energy is found as function of the duration of load. By means of extrapolation the activation energy at the beginning of the creep process can be determined approximately. Then the activation energy can be calculated throughout the duration of the creep process. Microscopic activation energy of hardened cement paste is calculated from relaxation experiments. The values of all tested specimens are more than two orders of magnitude greater than those from other solids and from water. Thus it can be concluded, as far as the size of the particles and the amount of displacement is concerned, that the contribution of the shift of single molecules is negligible. On the contrary, it can be deduced that gel particles must be considered as being the creep unity.

---

Résumé On établit une relation pour l'énergie d'activation macroscopique en tenant compte de l'accroissement du frottement interne de la pate de ciment durcie sous une charge permanente; elle permet d'évaluer l'énergie d'activation pour l'ensemble du processus de fluage. De la même manière, une autre équation pour la modification de l'énergie d'activation appara?t comme une fonction de la durée de chargement. Par extrapolation, l'énergie d'activation au début du processus de fluage peut être déterminée approximativement. L'énergie d'activation peut être ensuite calculée autant que dure le processus de fluage. D'après les expériences de relaxation, on calcule l'énergie d'activation microscopique de la pate de ciment durcie. Les valeurs pour toutes les éprouvettes essayées sont de plus de deux ordres de grandeur supérieures à celles trouvées pour d'autres solides et pour l'eau. On peut donc conclure, autant qu'il s'agit des dimensions des particules et de l'importance des déplacements, que la composante formée par le déplacement des simples molécules appara?t comme négligeable. Au contraire, on peut en déduire que les particules de gel doivent être considérées comme des unités de fluage.

     

Pore structure of hardened cement paste and mortar

The paper describes an experimental investigation into the pore structure of hardened cement paste and mortar. The pore structure was studied using mercury porosimeter. Ordinary portland cement and natural river sand were used. Pore structure determination was carried out for both the cement paste and mortar mixes over four hydration periods and five water-cement ratios. The threshold radius which was found to be prominent in the hardened cement paste, flattens out as the fine aggregate content increased.     

Effect of gamma irradiation on properties of hardened cement paste

Carbonation of hydrated cement paste (HCP) under γ-irradiation is researched in this paper. It is supposed that radiation-induced carbonation takes part simultaneously with “natural” carbonation. “Natural” carbonation is governed by diffusion and therefore takes part in the material close to surface only, whereas radiation-induced carbonation manifests in all irradiated material. The hypothesis has been tested by means of three sets of experiments. Within these experiments the following changes were investigated: (a) changes of composition of hydrated cement paste (transformation of portlandite to calcite), (b) changes of porous space [changes of average pore diameter (APD)], (c) changes of mechanical properties (hardness). HCP was analysed in the surface layer as well as in the centre of samples. The results show clearly that radiation at least accelerates carbonation driven by diffusion and support a hypothesis that besides natural carbonation there is also independent carbonation caused by γ-irradiation.     

氧化石墨烯对水泥净浆流动度及水泥石结构和性能的影响

采用Hummers法及超声破碎分散法制备了氧化石墨烯(GO)纳米相分散液,研究了GO对水泥净浆流动度和水泥石微观结构的影响,用FT-IR、AFM、XRD及SEM对GO及水泥石结构进行了表征。结果表明GO的掺入降低了净浆流动性,每增加0.01%的GO需要增加0.02%的聚羧酸系减水剂(PCs)以保持水泥净浆流动度在3 h内在200 mm以上,GO的掺入能够使水泥石的微观结构发生明显的变化,当GO/PCs掺量为0.01%/0.24%~0.03%/0.28%时,水化龄期7 d的水泥石出现了大量分散均匀的花状微晶体,当GO/PCs掺量为0.05%/0.32%~0.07%/0.36%时,同龄期水泥石中出现大量的片状晶体,在水化龄期延长到28 d时有转化为密实结构的趋势,结果说明GO具有调控水泥水化产物形貌的能力及增强增韧的作用,此研究结果对于提高水泥基材料的力学性能具有重要意义。     

Surface tension skrinkage and strength of hardened cement paste

Summary The surface tension and the surface energy of a solid is reduced by the presence of an adsorbed water film. When the surface tension of a porous material with a large interior surface is reduced, the length increases whereas the strength decreases. With the help of Griffith's theory of crack propagation it is possible to calculate the surface energy. For hardened cement paste with a water cement ratio of 0.45 and 0.6 the surface energy is found to be 1370 erg/cm² and 657 erg/cm² respectively. The cement paste was allowed to hydrate for 28 days without loss of moisture at a temperature of 20 °C. These results are in agreement with the surface energy of porous glass with a similar interior surface.

---

Résumé La tension superficielle et l'énergie superficielle d'un solide se trouvent diminuées par la présence d'une pellicule d'eau adsorbée. Quand la tension superficielle d'un matériau poreux à grande surface interne est diminuée, la longueur augmente tandis que la résistance diminue. En s'appuyant sur la théorie de la propagation des fissures de Griffith, il est possible de calculer l'énergie superficielle. Pour une pate de ciment durcie ayant un rapport eau/ciment de 0,45 et de 0,6, l'énergie superficielle se révèle être de 1370 erg/cm² et de 657 erg/cm² respectivement. On a laissé la pate de ciment s'hydrater durant 28 jours sans perdre d'humidité à une température de 20 °C. Ces résultats concordent avec l'énergie superficielle du verre poreux possédant une surface similaire.

     

Experimental study of thermal expansion of hardened cement paste

Thermal expansion of hardened cement paste has been determined within the temperature range of ?30°C and +80°C. Mositure content varied from virtually zero to saturaton. Immediate thermal response is followed by time dependnt ration. Immediate thermal response is followed by time dependent partial recovery. A tentative explanation for this behaviour is given.     

Size and boundary effects on desiccation cracking in hardened cement paste

The density of cracks or size of fragments formed in hardened cement paste upon first drying is affected by specimen size as measured with a crack-impregnation technique in free shrinking specimens with a thickness of 4 cm. Fragment size on the drying surface increased with distance away from the specimen corner, resulting in smaller average surface crack densities in larger specimens. Size effect on three- dimensional crack density, that was measured from sections through the impregnated specimens, was weaker. The size effect is explained by higher residual thermal stresses in larger specimens due to the cement hydration process. For comparison a desiccation crack pattern in a 5-mm-thick cement paste layer on a marble substrate was studied. Residual thermal stresses in this specimen were probably low and a uniform crack-pattern with a Gaussian-like fragment size distribution formed.     

Physical structure of hardened cement paste. A classical approach

It is explained how the classical model of the structure of hardened Portland cement paste was deduced by T.C. Powers and coworkers from data on water vapor adsorption isotherms. Moreover, it is shown how the fractional volumes of all major constituents in the physical structure of room temperature cured Portland cement paste can be estimated from information on water-cement ratio and degree of hydration of the cement. A large element of Powers’ model remains sound after almost forty years, and it remains the only model upon which quantitative calculations of the volumetric composition of hardened Portland cement pastes can be based. The paper is written for civil engineers in order to promote a better understanding of some fundamental ideas upon which many later developments in concrete technology are based.

---

Résumé On expose comment Powers et ses collaborateurs ont déduit le modèle classique de la structure de la pate de ciment Portland durci à partir de données sur les isothermes d’absorption de la vapeur d’eau. On montre, en outre, comment on peut obtenir une évaluation numérique du pourcentage en volume de tous les composants essentiels formant la structure physique de n’importe quelle pate de ciment Portland conservée à température ambiante à partir d’informations sur le rapport eau/ciment et le degré d’hydratation du ciment. Une grande partie du modèle de Powers reste valable après 40 ans ou presque, et c’est le seul modèle sur lequel on puisse établir des calculs quantitatifs de la composition volumétrique de pates de ciment Portland durci. Cet article, écrit à l’intention des ingénieurs en génie civil, a pour but d’améliorer la compréhension de certaines idées fondamentales sur lesquelles s’appuient de nombreux développements ultérieurs de la technologie du béton.

     

Thermal dilation and internal relative humidity of hardened cement paste

The coefficient of thermal dilation (CTD) of hardened cement paste and concrete is a function of the state of internal moisture in the pore system. It has been theorized that changes in the pore fluid pressure induced by temperature change causes additional dilation when the material is partially saturated. Drying shrinkage stresses in early-age concrete also evolve from changes in the pore fluid pressure. The Kelvin-Laplace equation relates changes in the pore fluid pressure to the measured internal relative humidity (RH). This research investigated the role of pore pressure changes on the CTD through internal RH measurements. A maximum change in humidity (ΔRH) due to temperature change (ΔT) was measured when the initial humidity was at an intermediate value. Likewise, the maximum CTD was also measured at an intermediate initial RH. Based on these findings, the additional thermal strain caused by changes in pore fluid pressure was modelled using internal RH measurements as a primary parameter.     

Crack patterns in steel fiber reinforced cement paste

Fracture patterns, produced when a crack advancing from a notch in cement paste intersected a steel fiber placed perpendicular to it, were studied by SEM methods. The specimens were small notched compact tension specimens that could be wedge loaded within the scanning electron microscope chamber using a wet cell. Three distinct cracking patterns were identified. In most cases the fiber caused a shift in the crack path, accompanied by microcracking and by separation of the main crack into several branches. Observations of the surface of the groove under the steel fiber indicated that the offsetting segment of the crack path in the matrix under the fiber was tortuous, and in many cases it was characterized by branching, discontinuities, and microcracks. The microstructure of the interface region consisted of CH-rich duplex film at the actual interface, backed by a highly porous zone consisting mainly of C-S-H. The offsetting crack segments tended to damage the preexisting interface zone microstructure under the steel fibers.

---

Résumé On a étudié par les méthodes de microscopie électronique SEM des configurations de fissures qui se forment quand une fissure se propageant à partir d'une entaille dans la pate de ciment recontre une fibre d'acier placée perpendiculairement. On a utilisé de petides éprouvettes compactes de traction entaillées qui pouvaient être chargées en coin à l'intérieur de l'enceinte de microscopie électronique avec une cellule humide. On a identifié trois configurations différentes de fissures. En général, la rencontre de la fibre a provoqué un déplacement de la trajectoire de la fissure, accompagné d'une microfissuration et de la division de la fissure principale en plusieurs branches. En observant la surface de la rainure, sous la fibre d'acier, on a remarqué que le segment déporté de la fissure dans la matrice avait un parcours sinuex et était souvent caractérisé par des ramifications, des discontinuité et des microfissures. La microstructure de la région de l'interface se présentait comme un double film riche en CH à l'interface même, reposant sur une zone très poreuse de C-S-H principalement. Les segments de fissure déportés avaient tendance à endommager la microstructure de la zone d'interface préexistante sous les fibres d'acier.

     

Density mapping of hardened cement paste using ptychographic X-ray computed tomography

Ptychographic X-ray computed tomography (PXCT) allows for a non-destructive, three-dimensional mapping of the electron density. Its quantitativeness combined with a resolution in the 100 nm range makes it a suitable tool for the assessment of densities of the individual phases in complex materials, such as hardened cement pastes. Here we present results of an experiment performed on a cylindrical sample of epoxy-impregnated hardened cement paste of about 30 μm in diameter. Two-dimensional cross sections of the three-dimensional electron density map show a microstructure that bears distinct similarity to that observed by back-scattered scanning electron microscopy. Domains of various residues of cement grains, calcium hydroxide, calcium carbonate, epoxy-resin impregnated calcium silicate hydrates, epoxy-resin impregnated porosity, and unimpregnated porosity are revealed and are manifested as distinguishable peaks in the histogram of the three-dimensional electron density map. On assumptions of (i) a priori knowledge of the chemical composition and (ii) the purity of the analysed regions, the mass densities of the above mentioned individual material phases are estimated. The potential of PXCT for the science of cement and concrete is discussed.