Retarder effect on hydrating oil well cements investigated using in situ neutron/X-ray pair distribution function analysis

Abstract

Understanding the role of retarder on the chemical nature and molecular architecture of hydrating cement paste is essential for engineering oil well cements with additives. Here, synchrotron X-ray and total neutron scattering with pair distribution function (PDF) analysis were performed in combination with calorimetry and nuclear magnetic resonance (NMR) to examine the retarder effect in hydrating tri-calcium silicate (C₃S) and Class G oil well cement paste. Primarily, the retarder, Diethylenetriamine pentamethylene phosphonic acid (DTPMP) influenced the hydration by affecting the Ca-O and Ca-Si pair correlation providing evidence of calcium playing a predominant role in the retardation process. Secondary effects related to Calcium-Silicate-Hydrate (C-S-H) nuclei poisoning influencing the suppression of calcium hydroxide precipitation were observed. These findings provide insights into the retardation mechanism of hydrating cement paste influenced by calcium depletion when subjected to phosphonate retarders.

Introduction

Chemical retarders play a pivotal role in controlling and suppressing calcium-silicate hydrate (C-S-H) hydration in oil well cements. It is essential to decipher the retarding mechanism between the additives and cementitious surfaces to predict their compatibilities for use in down-hole oil well conditions [1]. The investigation of these interactions of retarding additives with hydrating oil well cements helps us understand the detailed microstructure that is formed during the initial stage of hydration. With the inclusion of additives, the changes in cement paste microstructure occur at various length scales along with the conversion of free and bound water during the hydration process [2,3]. There are four possible retardation mechanisms that are generally proposed to govern the retarding action of retarders among Portland cements [4]. These are (a) complexation of calcium ions in pore solution [5], (b) adsorption of retarder on the surface of the cement via reducing the rate of ion release into the pore solution, (c) poisoning of nuclei leading to inhibition of nucleation and growth of hydrates, primarily the C-S-H, and (d) precipitation of a semipermeable layer via perturbation of the silicate-aluminate-sulfate balance [6]. Depending on the retarding additives used for cementing one or more combinations of the above mechanisms are responsible for delaying the setting time by limiting the rate of cement hydration as well as preventing initial strength formation.

Phosphonic acids and their salts have been used extensively as set retarding additives for well cementing at elevated temperatures and pressures [7,8]. Phosphonate retarders have hydrolytic stability at an elevated temperature while lowering the viscosity of high-density cement slurries and extending the induction period during hydration [1,9]. The retardation mechanism suggests growth of the hydrates is inhibited by the adsorption of phosphonate groups onto the nuclei of cement hydrates. For example, when Nitrilotris methylene phosphonic acid (NTMP) binds to the calcium pore solution through precipitating the surface of cement grains, growth inhibition of C-S-H nuclei occurs [10]. Another study used impedance spectroscopy to trace the retardation effect in cement pastes using aminotri(methylenephosphonic acid) (ATMP), 1-hydroxyethylidene-l,l-diphosphonic acid (HEDP) and Diethylenetriamine pentamethylene phosphonic acid (DTPMP) [11]. These results showed that phosphonate compounds had strong chelating effects potentially poisoning C-S-H and calcium hydroxide. To date, most studies report on the use of DTPMP retarder for oil well-cementing application using traditional instrumental techniques to investigate the working mechanism without providing necessary details on the microstructural evolution during the hydration process [7,[10], [11], [12]]. The retardation mechanism in hydrating cement paste with additives is a complex process; requiring careful selection of experimental techniques to provide accurate detection of the mechanisms responsible for inhibiting the growth of the hydration products.

A wide variety of methods has been used in oil well cement characterization. Calorimetric methods and nuclear magnetic resonance spectroscopy (NMR) along with laboratory XRD have been found to be of particular value for understanding the hydration process in oil well cements. NMR techniques include time domain NMR analysing proton spin-lattice (T₁) and spin-spin (T₂) relaxation times for cement hydration, structure of hydrated phases in C-S-H, and frequency domain MAS (Magic Angle Spinning) NMR for determination of the various silicate condensation productions, Qⁿ (Q⁰, Q¹ and Q²) [[13], [14], [15], [16], [17], [18]]. Furthermore, MAS (Dynamic Nuclear Polarization) DNP 2D NMR experiments have been employed to detect the atomic-level structure of the calcium silicate hydrate gel [19].

Scattering techniques available at synchrotron sources using high energy X-rays and spallation neutron sources are of primary interest to the cement and colloids community [20]. One such experimental technique that is useful in deciphering molecular mechanisms in complex colloidal systems is X-ray or neutron total scattering with the Pair Distribution Function (PDF) analysis [21]. The pair distribution function, as obtained from the Sine Fourier transform of the experimental total scattering (i.e., Bragg and diffuse scattering) data, gives the probability of finding an atom at a distance “r” from a given atom in a structure [22].

Here, the reduced Pair Distribution Function, G (r) is used, which is defined as:

$$G\left(r\right)=\frac{2}{\pi }{\int }_{\mathit{Qmin}}^{\mathit{Qmax}}F\left(Q\right)sin\left(\mathit{Qr}\right)\mathit{dQ}$$

In Eq. (1), F(Q) is the reduced structure function, i.e., F(Q) = Q [S(Q) − 1], where Q is the magnitude of moment transfer with Q_min and Q_max being the minimum and maximum values of Q, and S (Q) is the experimental total scattering structure function [21]. PDF studies using neutrons and X-rays have been well documented for cementitious materials including calcium silicate hydrates (C-S-H) in Portland cement-based systems and alkali-activated binders [[23], [24], [25], [26], [27], [28], [29], [30], [31]]. With the availability of advanced instrumental capabilities at these sources, gaining critical insight into the molecular structure of the multi-scale hydration product formation is made possible, and one can accurately predict the structural and dynamic changes during hydration. In this study, Neutron and X-ray PDF, MAS NMR and isothermal calorimetry are used to examine the retardation effect on hydrating Class G and C₃S pastes using DTPMP retarders. To the best of our knowledge, the current study is the first PDF analysis of retarder with commercially available oil well cement and tricalcium silicate (C₃S). Here, critical insight into the working mechanism of phosphonate based retarders in hydrating cement paste is examined via X-ray/Neutron PDF techniques.