Search

Quantitative prediction of gas hydrate formation risk is critical for the successful implementation of risk-based approaches to hydrate management in subsea production. Here we use a high pressure, stirred, automated lag-time apparatus and a high pressure acoustic levitator to experimentally obtain smooth probability distributions describing stochastic hydrate formation. Robust, repeatable hydrate nucleation and growth rate probability distributions as a function of induction time and subcooling are measured for various gases, shear rates and inhibitor dosages in systems with interfacial areas ranging from (0.1 to 60) cm2. The results reveal that new engineering models can be used to reliably predict hydrate formation probability. Importantly these new models have solid theoretical foundations which enables them to be generalized with confidence to industrial systems.

The application of bacterial cellulose (BC) as a wrapping material for vacuum-packaged beef was studied and compared against unwrapped beef for up to 3 weeks. The impact of BC wrap on the weight loss, purge accumulation, and drip loss were assessed along with low-field nuclear magnetic resonance, physicochemical, microbiological, and sensorial evaluations. The BC wrap significantly (P < 0.05) reduced purge accumulation in vacuum packages which was confirmed by an increased swelling ratio and scanning electron microscopy images. Colorimetric measurements showed significantly (P < 0.05) increased redness and yellowness values in wrapped samples compared to unwrapped samples. BC wrap did not affect pH, tenderness, and odor of meat, but significantly (P < 0.05) increased lipid oxidation, and numbers of lactic acid bacteria and Brochothrix thermosphacta counts. This study shows that BC wrap has potential as a purge absorbent in vacuum packaged meat. Practical Application: Bacteria cellulose has good water holding capacity that can be utilized to absorb purge exudate from beef. It helps to improve the appearance and consequently consumer acceptance of vacuum packed beef.

Rigorous quantification of hydrate formation probability in the presence of mono-ethylene glycol (MEG) is crucial for understanding hydrate formation risk in oil and gas production systems where the delivery of MEG for complete thermodynamic inhibition is either economically or practically infeasible. Using thermoelectrically cooled, stirred reactors, we have obtained 3,056 hydrate nucleation points in total. With these data we quantify the probability of hydrate formation in under-dosed MEG systems as a function of subcooling from the hydrate phase boundary, at MEG dosages between (5 and 25) wt% with respect to the aqueous phase. Although the addition of the MEG led to reductions in the measured formation temperatures, the corresponding subcooling distributions were similar to those measured in MEG-free systems. Furthermore, isothermal measurements at a fixed subcooling of (3.6 ± 0.1) K across a range of MEG loadings yielded comparable exponential induction time distributions. These results establish that MEG does not significantly affect either the nucleation rate or distribution of hydrate formation probabilities. Initial subcooling-dependent hydrate growth rates were also measured and reduced proportionally with the MEG dosage. Finally, we show how a low dosage kinetic hydrate inhibitor can be used together with MEG to further inhibit hydrate formation, by reducing both nucleation and growth rate. The results detailed here may be useful to the successful implementation of economically-viable, risk-based hydrate management strategies in under-inhibited systems.

Wettability is a critical parameter that controls fluid flow and distribution as well as recovery efficiency in reservoir rocks. The sensitivity of Nuclear Magnetic Resonance (NMR) relaxation measurements to wettability is well-known. Here we develop a new approach to calculate a wettability index based on low field NMR transverse (T2) measurements which is simpler than existing models in the literature and still affords good agreement with Amott-Harvey and USBM wettability test results. The model relies on the emergence of a bulk like relaxation signal from the non-wetting phase in partially oil/water saturated cores. Unlike other models in the literature, which require T2 measurements at up to four different saturation states, our model only requires measurements at two partial saturation states; irreducible brine (Swi) and residual oil (Sor). The performance of the model was tested using experimental results from different core types aged with light and heavy oils containing varied amounts of asphaltene. Different wettability conditions with Amott-Harvey/USBM values ranging from −0.39 to 0.79 were thus tested. When compared with measured reference wettability indices from the Amott-Harvey or USBM method, the model prediction showed good agreement (the calculated mean absolute error was 0.10). Additionally, improved accuracy was achieved relative to existing comparable models in the literature for the range of samples considered.

The effect of salinity on water-in-oil emulsions was systematically studied using a combination of nuclear magnetic resonance (NMR) pulsed field gradient (PFG) measurements of emulsion droplet size...

The tortuosity (τ), defined in the present context as the ratio of the free diffusion coefficient to the restricted diffusion coefficient of a contained fluid, is an important but difficult to measure characteristic of a porous medium, particularly when it is partially saturated with water. We develop and apply methodology, based on nuclear magnetic resonance (NMR) pulsed field gradient techniques, to measure τ for various sandstone rock cores as a function of residual water fraction. The NMR methodology requires the use of bipolar pulsed field gradient stimulated echo pulse sequences to avoid systematic errors due to magnetic susceptibility differences and D2O as a stationary immiscible water phase; this was selected as it provides no 1H NMR signal. Tortuosity of the free pore space was successfully measured using liquid ethane as a probe fluid for three different sandstones over the full accessible range of residual water saturation. Generally, the tortuosity was observed to increase with residual water (D2O) content; however, significant variations were observed between the different sandstones.

Nuclear magnetic resonance (NMR) well logging is a well-established technique for in-situ fluid measurement. It is however not routinely used in the mining industry to quantify water content in high magnetic susceptibility iron ore rock cuttings because the ore’s magnetic properties adversely affect the measurement. In this study the relationship between NMR signal intensity and magnetic susceptibility is studied using a low magnetic field (2 MHz) benchtop NMR instrument. Magnetic susceptibility is commonly measured during well logging protocols and therefore could be used to correct and hence render quantitative NMR measurements of moisture content in iron ore. In this study NMR signal from water-saturated synthetic samples (magnetite/maghemite/hematite mixed with borosilicate glass beads) is measured as a function of iron content and magnetic susceptibility to obtain such a correlation. This correlation is then applied to iron ore rock cuttings. Comparison with gravimetric measurements show that a standard error of estimate of water content across all samples tested of 6.4 wt% using no correction, is reduced to 1.5 wt% using the corrections, corresponding to a relative error reduction from 47% to 11%. Additional comparison with whole rock cores shows that a higher magnetic susceptibility threshold for using the corrections is required for whole rock cores than for the rock cuttings, this difference requires further study. These results encourage the broad use of these types of correlations for NMR measurements of iron rich cuttings and application of NMR well logging to iron ore deposits.

Gas hydrate formation is a stochastic phenomenon of considerable significance for any risk-based approach to flow assurance in the oil and gas industry. In principle, well-established results from nucleation theory offer the prospect of predictive models for hydrate formation probability in industrial production systems. In practice, however, heuristics are relied on when estimating formation risk for a given flowline subcooling or when quantifying kinetic hydrate inhibitor (KHI) performance. Here, we present statistically significant measurements of formation probability distributions for natural gas hydrate systems under shear, which are quantitatively compared with theoretical predictions. Distributions with over 100 points were generated using low-mass, Peltier-cooled pressure cells, cycled in temperature between 40 and −5 °C at up to 2 K·min–1 and analyzed with robust algorithms that automatically identify hydrate formation and initial growth rates from dynamic pressure data. The application of shear...

We present a model for hydrate deposition and sloughing in gas dominated pipelines which allows for rapid estimations of the pressure and temperature profiles along a horizontal pipeline during normal operation in the hydrate forming region in the presence of monoethylene glycol (MEG). Previous models assume that the hydrate deposit growing at the pipe wall is stable, which may lead to an overestimation of the pressure drop over time. Hydrate growth rates were calculated using a classical hydrate kinetic model combined with a simplified two-phase flow model for pipelines in the annular flow regime with droplet entrainment. Hydrate growth at the pipe wall, deposition of hydrate particles from the gas stream and sloughing due to shear fracture of the deposited film contributed to the evolution of the hydrate deposit. The model parameters included a scaling factor to the kinetic rate of hydrate growth and a particle deposition efficiency factor. The fraction of deposited particles forming a stable hydrate film at the pipe wall through sintering and the shear strength of the deposit were introduced as two additional parameters to enable simulation of sloughing events. The tuned model predicted hydrate formation within 40% and pressure drop within 50% of measurements previously obtained in a gas-dominated flow loop over a wide range of subcoolings, MEG concentrations and high and intermediate gas velocities. The observed decrease of the kinetic factor with decreasing gas velocity indicated larger resistances to hydrate growth in the entrained droplets at lower flow rates, while the increase of the deposition parameter with MEG concentration was consistent with a particle adhesion/cohesion mechanism based on the formation of a capillary bridge. The preliminary sloughing model presented in this work, combined with flowloop testing, has allowed the first in-situ determinations of the effective shear strength of the hydrate deposits (in the range of 100–200 Pa) which is a key property to predict hydrate detachment and accumulation in gas-dominated pipelines.

An NMR Q-switch was designed and constructed specifically for use with low-field NMR apparatus. This featured a comparatively simple resistive damping design. It served to reduce the r.f. probe ring-down time, and hence reduced the signal acquisition delay from 25 ms to 9 ms, on an Earth’s magnetic field NMR system. The advantage of this earlier acquisition was demonstrated for both an aqueous suspension of iron oxide particles and using an NMR flow meter.

The current hydrate kinetics model implemented in the multiphase flow simulator OLGA treats hydrate growth in oil-continuous systems by considering the solidification of emulsified water droplets to form a hydrate-in-oil slurry that is assumed to be stable. To date, the validity of this model has not been established for gas-dominant systems, where gas void fractions can exceed 90 vol %. Here, six experimental data sets, collected using a 40-m single-pass gas-dominant flowloop operating in the annular-flow regime, were compared with predictions made using the current hydrate kinetics model. The comparison identified discrepancies in the predicted flow regime and the gas–water interfacial area that significantly affect kinetic hydrate-growth-rate calculations; these discrepancies might be due, in part, to differences in dynamic similarity between flowloop experiments and industrial-scale simulations. By adjusting only the kinetic rate scaling factor, it was not possible to match the pressure drop observed ...

Recovering methane (CH4) via the injection of carbon dioxide (CO2) into a CH4-hydrate-bearing reservoir is a highly attractive mechanism for meeting the world's future energy demand, since it offers the prospect of carbon-neutral energy production. Although the process of CH4–CO2 exchange in hydrate has been demonstrated in numerous laboratory experiments, deployment is hampered by an incomplete understanding of the underlying mechanism of exchange, inconsistencies in reported performance, and insufficient rates of recovery. In this study, the combined effect of thermal stimulation and hydrate-sediment ratio on the rate and degree of CH4 recovery has been quantitatively investigated through a series of large-volume, high-pressure core-flooding experiments on well-characterized synthetic hydrate-sediments using in situ Raman spectroscopy of the effluent vapour. The effect of controlled thermal stimulation has been related to a shell–core model of hydrate exchange, showing that repeated thermal stimulation is effective in temporarily increasing the exchange rate. Furthermore, a numerical analysis of the effective exchange layer thickness for a general particle size distribution is presented: from the results of this work and studies in the literature, we have established a preliminary heuristic for estimating the extent of exchange at constant temperature and pressure. During an initial phase, the effective exchange layer thickness grows at about 9 μm day−1 to a limiting value around 5 μm, where after further exchange occurs at a rate of less than 0.5 μm day−1. The presence of sediment in combination with thermal stimulation was found to significantly increase the rate and overall recovery of CH4, from 28% for a pure hydrate core to 82% for a hydrate-sediment core. This increased recovery is attributed to wetting of the unconsolidated-sediment during exchange. In the context of designing CH4 production from hydrate, these results establish that staged thermal stimulation is an effective means of enhancing CH4 recovery through CO2 exchange and can be improved by applying a series of smaller temperature increments. The enhanced recovery observed in the presence of sediment has important implications for optimized CO2 injection and sequestration, due to the preferential formation of pore-filling hydrate and the associated impact on permeability.

We present a novel nuclear magnetic resonance (NMR) multiphase flow metering system with the ability to interpret the flow regime and quantify both the liquid volumetric flowrate and holdup for gas-liquid flows. The flow measurement apparatus consists of a pre-polarising permanent magnet upstream of an Earth's field radio frequency NMR detection coil. In this work, the system is applied to measure the free induction decay (FID) NMR signal of gas-liquid flows at a range of flow rates in both the stratified and slug flow regimes. Tikhonov regularisation is applied to fit a model equation to the acquired FID signal in order to determine the relevant liquid velocity probability distribution. Signal interpretation applied to the individual NMR scans allows monitoring of both the liquid velocity and holdup with time. The NMR estimate of the liquid holdup is comparable to video analysis of the flowing stream through a transparent pipe section. The accuracy of the NMR metering system is successfully validated against an independent in-line rotameter measurement of the liquid volumetric flowrate during multiphase flow. Finally, analysis of the temporal variation in measured liquid flowrate is shown to clearly distinguish the stratified and slug flow regimes.

Enhanced natural gas recovery (EGR) with supercritical (sc)CO2 sequestration offers the prospect of increased natural gas recovery. High-fidelity reservoir simulations offer a method to quantify the risk of contamination of produced gas by the injected scCO2. Simulations of scCO2 mixing with the reservoir gas have been reported; however the effects of connate water on EGR have not been effectively explored. We extend a prior EGR simulation tool (Patel, May and Johns, 2016; Ref. [1]) to incorporate connate water accounting for its effect on dispersivity and permeability; chemical equilibrium is modelled using a novel, computationally efficient Lagrange multiplier-based approach. The code is applied to a ‘quarter five-spot’ benchmark scenario. The inclusion of connate water generally resulted in a reduction in breakthrough time and a decrease in methane recovery. The connate water's largest effect was to change the scCO2 flow field, which sank towards the reservoir floor, flooded the lowermost accessible layers and entered the production well via a high throughput channel (‘coning’). The magnitude of these effects were, however, sensitive to well perforation depth, the influence of which was subsequently studied systematically. Well perforation depth was found to determine the duration of these sinking and coning events in a non-linear manner.

The development of kinetic hydrate inhibitor (KHI) performance rankings is essential for these chemicals’ implementation within risk-based hydrate management strategies. Here, we obtain high resolution induction time distributions and growth rate measurements with a high pressure, stirred, automated lag time apparatus (HPS-ALTA) to make rigorous comparisons between two industrial KHIs (Inhibex 501 and 713). These comparisons reveal performance rankings can be subcooling dependent, and that growth rate alone is insufficient for accurate screening. Application of a Classical Nucleation Theory (CNT) based model to these results suggests that nucleation in systems dosed with Inhibex 501 and 713 respectively occurs on 7 and 700 times as many sites as compared to KHI-free systems, while the nucleation work required at these sites is increased by a factor of 9 and 28. When combined with gamma distributions and fluid residence times, these measurements can help inform the operational risk assessment of a hydrate blockage.

Gas hydrates are ice-like solids that may form in crude oil flowlines under high pressure and at low temperature, resulting in the formation of viscous hydrate-laden oil slurries. The magnitude of the slurry viscosity has been suggested as a primary means of determining the risk and severity of flow blockage, but there is a dearth of data available to calibrate predictive models of hydrate-in-oil slurry viscosity. This work deploys a controlled-stress, high-pressure rheometer to characterize the rheological properties of methane hydrate-in-crude oil slurries, which were generated in situ from water-in-oil emulsions. A vane blade rotor was found to maintain sufficient methane saturation in the oil phase during the hydrate growth period to allow near full conversion of the water. Dynamic measurements with the rheometer were able to separate the contributions to the viscosity change as the emulsion converted to a hydrate slurry due to (i) formation of the solid particles and (ii) reduction of the methane content in the oil continuous phase. For 5–30 vol% watercut systems, the slurry viscosity increased between 20 and 60 times during hydrate growth, whereas, under equivalent conditions for a 20% watercut emulsion, the viscosity increase due to the desaturation of methane from the oil phase was less than a factor of two. The steady-state hydrate-in-oil slurry demonstrated shear thinning behavior at both 1 and 5° C. The measured slurry relative viscosity deviated between 12 and 212% from the current industry-standard hydrate-in-oil slurry viscosity model, indicating the need for model improvement. After an eight-hour annealing period to simulate subsea shut-in, the yield stress of the hydrate-in-oil slurries varied between 3 and 25 Pa over 5 to 25 vol% hydrate.

Surface tension predictions of hydrocarbon mixtures at vapour-liquid equilibrium are crucially important and used in the design of many unit operations in the production of liquefied natural gas (LNG). Predictive models for surface tension are not well tested for hydrocarbon mixtures at high pressures due to the limited data available at relevant conditions. A differential capillary rise apparatus consisting of a high-pressure sapphire equilibrium cell was constructed and used to measure the surface tension of methane + propane mixtures along three isotherms T = (272.23, 285.51 and 303.34) K at pressures up to 9 MPa. The capillary diameters were calibrated using pure saturated propane at 271 K, and the technique was validated by measurements of pure ethane. The measured surface tensions were compared with the available data and the predictions of various models, including the Parachor method and Linear Gradient Theory: both were able to describe the data within their uncertainty. However, the default surface tension model for hydrocarbon mixtures implemented in the widely-used software package REFPROP 9.1 gave poor predictions; this was rectified through the implementation of the Parachor method in a beta-version of REFPROP 9.2. Critical points were also measured from observations of critical opalescence and the disappearance of the bulk interface. The measured mixture critical points were consistent with, and extended, literature values for this binary system. The measured critical points differed by between 0.5 and +27 K from predictions made with the GERG-2008 equation of state (EOS) and between −5 and +10 K for the Peng-Robinson EOS.

Enhanced Gas Recovery (EGR) involves CO2 injection into natural gas reservoirs to both increase gas recovery and trap CO2. EGR viability can be determined by reservoir simulations; however these require a description of fluid dispersion (mixing) between the supercritical CO2 and natural gas. Here we quantify this dispersivity (α) in sandstone rock plugs as a function of residual water fraction. To ensure the accuracy of such data, we designed a novel core flooding experimental protocol that ensured an even spatial distribution of water, minimised erroneous entry/exit contributions to mixing, and minimised dissolution of the CO2 into the water phase. Dispersivity was found to increase significantly with water content, although the differences in α between sandstones were eliminated upon the inclusion of residual water. This enabled development of a correlation between α and water content and, hence, between the dispersion coefficient and Peclet number that is readily incorporable into reservoir simulations. This article is protected by copyright. All rights reserved.

The high operating pressures and distances of deep water tiebacks increase the likelihood of hydrate blockage during transient operations such as shut-in and restart. In many cases, complete hydrate avoidance through chemical management may become cost prohibitive, particularly later in the field’s life. However, a subclass of crude oils has been observed in which hydrate blockages do not form during restart, rendering active hydrate prevention unnecessary. Over the past 20 years, limited information has been reported about the chemical or physical mechanisms that enable this plug-resistive behavior. This study presents an extensive and systematic method of characterizing whether an oil may naturally resist hydrate plug formation, including (i) chemical and physical property analysis, (ii) water-in-oil emulsion behavior, and (iii) the effect of the oil on hydrate blockage formation mechanics. This last set of experiments utilizes both a high-pressure rheometer and a sapphire autoclave to allow direct obse...

A high-pressure stirred autoclave cell equipped with a focused beam reflectance measurement (FBRM) probe and a particle video microscope (PVM) was used to study hydrate formation and plugging in gas–water systems as a function of shear rate. These probes allowed estimates of the mean hydrate particle size and number of hydrate particles to be correlated with the hydrate volume fraction and the hydrate slurry’s resistance-to-flow. Before reaching the hydrate volume fraction φtransition at which the hydrate slurry first exhibits a measurable increase in resistance-to-flow at ≈(16 ± 2) vol %, clear changes in the measured number and size of the hydrate particles were observed. Initially, hydrate particles within the FBRM probe’s field of view decreased in size and increased in number until a maximum was reached at concentrations of 2–9 vol % (increasing with shear rate). However, with continued hydrate growth, the number of particles within the FBRM probe’s field of view unexpectedly decreased and eventually...

By-line Nuclear Magnetic Resonance (NMR) measurements of emulsion droplet size distributions are presented based on pulsed field gradient (PFG) measurements. These are performed on temporarily immobilised samples extracted from a main process stream with corrections applied for any temporal variations in sample composition. The overall methodology is initially applied to pure fluids and then a range of water-in-oil emulsions. It is then demonstrated on an emulsification flow loop in which three commercial demulsifiers are separately applied; significant variation in their performance with respect to increasing emulsion droplet size (and thus emulsion destabilisation) is observed. Finally, a more rapid PFG method, Difftrain, is successfully demonstrated with the measured mean emulsion droplet size being used as the input into standard PID control of applied shear and hence the extent of emulsification.

Hydrate formation was studied using water droplets acoustically levitated in high-pressure natural gas. Despite the absence of solid interfaces, the droplets' area-normalised nucleation rate was about four times faster than in steel autoclave measurements with interfacial areas roughly 200 times larger. Multiple stages of stochastic, template-free hydrate growth were observed.

Complex organic molecules are of great importance to academic and industrial chemistry and typically synthesised from smaller building blocks by multistep reactions. The ability to perform multiple (distinct) transformations in a single reactor would greatly reduce the number of manipulations required for chemical manufacturing, and hence the development of multifunctional catalysts for such one-pot reactions is highly desirable. Here we report the synthesis of a novel hierarchically porous framework, in which the macropores are selectively functionalised with a SO4/ZrO2 solid acid coating, while the mesopores are selectively functionalised with MgO solid base nanoparticles. The spatially orthogonal distribution of acid and base catalytic sites permits the efficient two-step cascade deacetylation and Knoevenagel condensation of dimethyl acetals to cyanoates, and the simultaneous esterification and transesterification of a high free fatty acid content triglyceride.

Current methods of estimating boil-off gas (BOG) rates for large-scale liquefied natural gas (LNG) storage tanks are largely empirical and based on limited available experimental data. More accurate models would be extremely valuable for estimating the potential for excessive BOG generation during LNG storage and transportation scenarios as well as providing critical inputs into the design of BOG re-liquefaction systems. This study reports a series of experiments that have been conducted for LNG-like binary mixtures of methane and ethane to measure the BOG production and resultant pressure change under various industrially relevant conditions. Experimental data and observations made in this work are compared with both the available literature and with the predictions of a new non-equilibrium model that uses the GERG-2008 equation of state to calculate relevant LNG and BOG properties. The data reveal three distinct stages of BOG evolution, here labelled as self-pressurisation, transient, and homogenous. It is observed that, in the self-pressurisation stage, the thickness of a thermally stratified layer adjacent to the liquid–vapor interface increases with time. The transient stage is defined to commence when the system reaches the specified relief pressure and the homogeneous stage is reached upon the effective elimination of thermal stratification in the LNG. Good agreement exists between this new model and the experimental and literature data acquired during the self-pressurisation and homogeneous stages. In the transient stage, the model does not accurately quantify the BOG rate indicating a need to incorporate the effects liquid thermal stratification in future model development.

The formation of gas hydrates is crucial to many technological applications including energy production, energy storage, desalination, and CO2 capture. Kinetic hydrate inhibitor (KHI) chemicals have been used industrially for over 25 years to suppress hydrate formation in key industrial applications. However the mechanisms by which they operate, and specifically whether they delay nucleation or just retard growth, remain open questions. Here induction time probability distributions for methane hydrates were measured at several constant subcoolings as a function of KHI concentration. This allowed the hydrate nucleation and growth rates at each condition to be separately quantified through the observed induction times and initial gas consumption rates, respectively. Adding a KHI produces a Gamma-distribution of induction times, characterized by two parameters: nucleation rate and the average number of events associated with detection. This is qualitatively different to the exponential distributions of induction time probabilities observed in systems without any KHI. These data reveal how KHIs both increase the nucleation work required to form a critical nuclei and increase the effective number of sites where nucleation could occur. By showing unambiguously how KHIs delay hydrate nucleation at low subcoolings, this work opens systematic pathways for developing improved chemicals for either hydrate inhibition or promotion. The approach to inhibitor testing demonstrated here may also help natural gas producers assess the costs and benefits of competing designs for hydrate management.

The boil-off gas (BOG) produced from liquefied natural gas (LNG) mixtures in cryogenic storage tanks must be predicted reliably as a function of tank shape, heat ingress, thermal stratification, pressure, and liquid volume fraction. However, current methods of estimating BOG rates for large-scale tanks are entirely empirical and based on limited available data, with no models available for reliable predictions. This affects the ability of LNG carriers to optimise BOG compressor sizing. A new apparatus was developed to explore the effects of heat flux, liquid stratification, volume, and mixture composition on the measured boil-off rate. The apparatus is demonstrated using liquid nitrogen with BOG rates quantified as a function of various heat fluxes, pressures, and initial liquid volume fractions. Three distinct periods of boil-off were observed: the pressurisation, transient, and steady-state stages. The data are compared with the available literature and the predictions of a new dynamic model accounting for heat transfer from the super-heated vapour. Excellent agreement is observed between model predictions and the data measured during the pressurisation and steady-state stages. However, the model does not capture the BOG rate observed in the transient stage, suggesting liquid thermal stratification should be considered in future models for LNG boil-off.

A stirred, high-pressure, visual microscopy-based apparatus was used to investigate solid-formation kinetics in LNG-relevant binary mixtures. The apparatus can detect solid crystals as small as 20 µm in a 3.5 mm field of view and is capable of measurements at temperatures as low as 90 K and pressures up to 20 MPa. The apparatus is described in detail and is presented alongside experimental results for two benzene + methane mixtures. Measurements of the solid–fluid equilibrium temperatures for the benzene + methane mixtures at 10 MPa were determined by raising the temperature of the mixture in (0.5 to 1) K steps and observing whether the crystals were still present after 2 h. The equilibrium temperatures were consistent with those predicted using the software package ThermoFAST within the combined experimental and model uncertainties. Formation measurements were carried out using a 100 ppm benzene-in-methane sample. Isobaric constant-cooling experiments at pressures of 8 and 10 MPa were used to construct subcooling formation-probability distributions for this mixture by identifying the temperature at which solid crystals were first observed to form on a copper substrate. The subcooling values at formation ranged from (4.4 to 11.0) K and were used to generate a cumulative probability of formation. The measured formation-probability distribution was fit to a model based on Classical Nucleation Theory to yield an estimate of 5 mJ/m2 for the effective surface free energy of solid benzene in liquid methane on the copper substrate. These results pave the way to probabilistic estimates of risk for solids formation in cryogenic heat exchangers.

Hydrocarbon production from unconventional shale reservoirs has gained increasing worldwide focus, spurred largely by improvements in production techniques. Nuclear magnetic resonance (NMR) is increasingly being explored as a means of characterizing such shale rock. Compared to conventional sandstone and carbonate rocks, shales present a much lower porosity and permeability with NMR signal arising from both producible hydrocarbon fluid content as well as semi-solid organic solids and bitumen. Interpretation of the resultant NMR signal from shales is consequently comparatively more complex. To this end we apply combined NMR free induction decay (FID) and Carr-Purcell-Meilboom-Gill (CPMG) measurements and interpret the spliced data using a combined Gaussian and exponential inversion method. This was applied to a range of shale samples with variable moisture content in an attempt to assess the accuracy of such an approach to delineate the inherent predominantly bitumen signal from that of the added water. The analysis was performed at a range of frequencies (20, 40 and 60 MHz), which are applicable to more widely accessible bench-top NMR spectrometers. Consistently, the Gaussian component of the resultant T2*|T2 distribution was found to be independent of moisture content, scaling rather with the total organic content of the respective shale cores. In contrast, the exponential component of the resultant T2*|T2 distribution was found to scale linearly with the moisture content of the cores.

EGR (Enhanced natural gas recovery) with CO2 sequestration offers the prospect of significant environmental and economic benefits by increasing gas recovery while simultaneously sequestering the greenhouse gas. Field-scale deployment is currently limited as the risks of contamination of the produced gas by injected CO2 are poorly understood. Reservoir simulations offer a method to quantify the risk but only if sufficiently accurate. For the first time, finite element simulations are presented for several EGR scenarios that incorporate the most accurate models available for fluid mixture and rock properties. Specifically, the GERG-2008 EOS (equation of state) is utilised to describe the supercritical fluid mixture's density, as are reference correlations linked to the most accurate experimental data available for diffusivity and viscosity. Realistic values for rock dispersivity and tortuosity determined from high-accuracy core-flooding and NMR (nuclear magnetic resonance) experiments were also integrated. The relative impacts of these properties were investigated for a benchmark layered reservoir with a quarter 5-spot well pattern. Recovery efficiency at different CO2 injection rates was also investigated and was determined to be the dominant consideration: a 100-fold rate increase improved recovery from 53% to 69% while CO2 breakthrough time decreased by less than expected. Collectively, these results emphasise the importance of accurate reservoir simulations for EGR.

Gas hydrate formation is a critical flow assurance risk in oil and gas production, as remediation of blockages may require weeks of operating downtime and represent a significant safety hazard. While many studies over the past two decades have focused on quantifying hydrate blockage risk in crude oil systems, there is a dearth of information available with which to assess hydrate growth rate or blockage severity in natural gas systems, which typically operate between stratified and annular flow regimes. In this investigation, a single-pass gas-dominant flowloop was used to measure hydrate growth and particle deposition rates with variable liquid holdup (1–10 vol%) and subcooling (1–20 °C). A particular focus of this study was the impact of reducing the gas phase velocity to achieve lower liquid entrainment and, therefore, decrease hydrate formation rate. Reducing the gas velocity from 8.7 to 4.6 m/s at a constant subcooling around 6 °C reduced the total formation rate by a factor of six. At these conditions, the sensitivity of hydrate formation rate to velocity was about 40 times greater than the sensitivity to subcooling. This reduction in gas velocity also halved the estimated rate of hydrate deposition on the pipeline wall. Finally, new observations of hydrate wash-out are reported, whereby significant localized hydrate deposits were effectively removed by modulating the subcooling of the flowloop wall from 6 °C to 3.5 °C. The results provide new insight to inform the next generation of predictive hydrate growth and deposition models for gas-dominant flowlines.

Gas hydrates are crystalline solids comprised of a three-dimensional network of water cages that can trap small hydrocarbon molecules at high pressure and low temperature. Formation of gas hydrates can lead to blockages in subsea oil and gas flowlines, where these conditions are readily achieved. As gas reserves mature over time, new CO2-rich reservoirs can be prolific hydrocarbon producers for markets willing to bear the additional production costs. However, there is a lack of understanding about the formation behavior and growth morphology of CO2-rich hydrates. In this work, hydrate growth rate and resistance-to-flow (torque) data for a ternary CO2-rich gas mixture were measured in a high-pressure sapphire autoclave apparatus and were compared with baseline data for a pure methane system. Images of hydrate growth and eventual plug formation were captured for methane hydrate and ternary gas systems, where early hydrate growth in the latter was dominated by an opaque film nucleating at the gas-water-wall interface and growing into both gas and water phases with time. Only 7 vol% methane hydrate was required to increase the motor torque above its baseline value, while at least 20 vol% hydrate was required when formed from the CO2-rich ternary gas. These new observations provide insight into the effect of guest species solubility on hydrate growth and resistance-to-flow in the resultant slurries, which are key parameters when assessing the risk of hydrate blockage in oil and gas pipelines.

Assessment of the viability of enhanced gas recovery (EGR), in which CO2 is injected into natural gas reservoirs, requires accurate and appropriate reservoir simulations. These necessitate provision of parameters describing dispersion between the fluids. Here we systematically measure fluid dispersion in various rock cores (sandstones and carbonates), both dry and at irreducible water saturation, at reservoir conditions. In this manner we evaluate the impact of the irreducible water on the miscible displacement processes. As such this represents the first measurement of dispersion as a function of water saturation for supercritical gases in consolidated media. Complementary measurements of water spatial distribution along the rock axis, as well as the pore size distribution occupied by the water were performed using magnetic resonance techniques. Irreducible water was found to increase dispersivity by a factor of up to 7.3. The dispersion coefficient (K) was measured as a function of velocity and the data for both dry and water-containing samples were successfully combined on a K–Peclet number (Pe) plot, enabling ready future inclusion into EGR reservoir models. The power-law dependence of K upon Pe produced an exponent of 1.2 for dry and water-saturated sandstones and 1.4 for dry and water-saturated carbonates, consistent with literature results (Bijeljic et al., 2011; Honari et al., 2015).

Formation of water-in-crude oil emulsions is a pervasive problem for crude oil production and transportation. Here we investigate the effectiveness of a comparatively low pressure CO2 treatment in terms of breaking these water-in-crude oil emulsions. To this end, we used unique benchtop nuclear magnetic resonance (NMR) technology to measure the droplet size distribution (DSD) of the emulsions. Treatment with 50 bar CO2 for 2 h resulted in significant emulsion destabilization; this was replicated when CO2 was replaced by N2O, which has a solubility in both the aqueous and oil phases similar to that of CO2. Low solubility gases, N2 and CH4, by contrast had no effect on emulsion stability. Treatment with CO2 was also found to have no effect on a model water-in-paraffin oil emulsion stabilized by a synthetic surfactant (Span 80). Collectively, this supported the hypothesis that emulsion destabilization results from CO2 precipitation of asphaltenes as opposed to emulsion droplet film disruption during depressu...

Membrane systems are commonly used in the water industry to produce potable water and for advanced wastewater treatment. One of the major drawbacks of membrane systems is biofilm formation (biofouling), which results in an unacceptable decline in membrane performance. Three novel in situ biofouling characterization techniques were assessed: (i) optical coherence tomography (OCT), (ii) planar optodes, and (iii) nuclear magnetic resonance (NMR). The first two techniques were assessed using a biofilm grown on the surface of nanofiltration (NF) membranes using a transparent membrane fouling simulator that accurately simulates spiral wound modules, modified for in situ biofilm imaging. For the NMR study, a spiral wound reverse osmosis membrane module was used. Results show that these techniques can provide information to reconstruct the biofilm accurately, either with 2-D (OCT, planar optodes and NMR), or 3-D (OCT and NMR) scans. These non-destructive tools can elucidate the interaction of hydrodynamic...

As industry moves toward the production of oil and gas resources in deep offshore environments, the prospective formation of natural gas hydrates under low temperature, high-pressure conditions poses an increasing risk of pipeline blockage. Successful management of this risk requires a precise forecast of the hydrate growth rate. Such predictions should incorporate quantitative descriptions of both deterministic and probabilistic hydrate phenomena. In this work, we present a first step towards such a quantitative risk assessment for systems that form hydrate slurries, based on a consideration of the stochastic nature of hydrate formation. Our framework introduces a new approach to risk assessment, by coupling a laboratory-derived probabilistic nucleation model with existing deterministic calculations for hydrate slurry viscosification. This new approach is used to extend a previously-described hydrate risk algorithm, the Hydrate Flow Assurance Simulation Tool (HyFAST), which enables the rapid assessment of hydrate slurry viscosification using the most advanced models available. Importantly, while the previous version of HyFAST was constrained to calculations in flowloop and autoclave geometries, for which it is sufficient to consider a single volume element, the new version of the HyFAST algorithm allows calculations for flowlines wherein a series of multiple volume elements are considered. The advanced hydrate formation models within this algorithm allow identification of critically important hydrate formation scenarios, and may serve as a screening tool for cases that then require study within rigorous hydrodynamic packages such as OLGA® or LedaFlow® to examine behaviour during key transient events. By coupling the outputs of our predictive algorithm and quantitative risk framework, we demonstrate a new capability to assess what constitutes an acceptable level of hydrate formation. We use this algorithm in a series of case studies that compare the effectiveness of common hydrate mitigation strategies over the life of a reservoir, including the optimization of a wellhead choke opening for both production rate and hydrate blockage risk.

The formation of gas hydrate plugs in deep water oil and gas flowlines poses severe operational and safety hazards. Previous work has established a mechanism able to describe plug formation in oil-continuous systems, which relies on the assumption that all the water remains emulsified in the oil phase. However, light hydrocarbon fluids, including condensates, may not stabilize water-in-oil emulsions, and the current mechanistic model cannot reliably assess the risk of plug formation in this scenario. This study presents a comprehensive set of experiments conducted in a high-pressure sapphire autoclave apparatus using 10 to 70 vol% water in partially-dispersing mineral oil at three fixed rotational speeds: 300, 500 and 900 RPM. Pressure and temperature were monitored continuously in the autoclave, providing direct estimates of hydrate growth rate, alongside measurements of the motor torque required to maintain constant mixing speed. A new conceptual mechanism for plug formation has been developed based on the visual observations made during these experiments, where a small hydrate fraction (2–6 vol%) in the oil phase was observed to disrupt the stratified water–oil interface and help disperse the water into the oil. This disruption was followed by an increase in the hydrate growth rate and particle agglomeration in the oil phase. In the final stages of hydrate growth for systems with low turbulence and high watercut, hydrate particles in the visual autoclave were observed to form a moving bed followed by full dispersion of water and oil, rapid hydrate growth and deposition on the wall. These rapid hydrate growth and deposition mechanisms significantly increased the maximum resistance-to-flow for partially-dispersing systems in comparison with mixtures that are fully dispersed under similar conditions.

Hypothesis Shear-induced droplet diffusion of flowing hard spheres is relatively well understood and has been extensively studied both experimentally and via simulations. The same however is not true of soft spheres, specifically emulsions, despite their broad and extensive industrial relevance. Here we seek to demonstrate that appropriate NMR techniques can be used to quantitatively measure shear-induced droplet diffusion. Limited literature indicates that dilute dispersions of soft spheres experience significantly larger shear-induced droplet diffusion relative to otherwise equivalent hard sphere suspensions. Here we explore whether this effect persists to high concentrations. Experiments Nuclear Magnetic Resonance (NMR) pulsed field gradient (PFG) techniques were used to measure shear-induced droplet diffusion for capillary flow of various water-in-oil (w/o) emulsions in a direction transverse to flow. Two adaptations were necessary – the acquired signal was analyzed so as to quantitatively distinguish restricted molecular diffusion within the emulsion droplets from shear-induced diffusion of the droplets, whilst flow-compensated PFG pulse sequences were shown to be necessary to account for any erroneous effects due to flow. A range of w/o emulsions were considered to enable measurement of shear-induced droplet diffusion as a function of both water content and mean shear rate. The surfactant content of these emulsions was adjusted such that they presented similar (stationary) emulsion droplet size distributions (DSD) which were also measured using NMR PFG techniques. Findings The droplet shear-induced diffusion data for the emulsion systems were compared against relevant results from the literature. Consistent with predictions for dilute systems, significantly greater droplet diffusion was measured relative to hard sphere suspensions at all concentrations, and a quadratic dependence was found between droplet diffusion and mean droplet size. For more concentrated emulsions, a peak in the droplet diffusion–concentration relationship was observed for the first time in emulsions, prior to the onset of emulsion inversion.

In the present work, Raman spectroscopy was used to study the structure of water molecules in the vicinity of glass particles with different hydrophobicity, immersed in water and in tetrahydrofuran and cyclopentane hydrates. The glass particle surfaces were clean (hydrophilic), coated with N,N-dimethyl-N-octadecyl-3-aminopropyl trimethoxysilyl chloride (partially hydrophobic), or coated with octadecyltrichlorosilane (hydrophobic). The Raman spectra indicate that, prior to nucleation, water molecules in the vicinity of hydrophobic surfaces are more ice-like ordered than those in the bulk liquid or near either hydrophilic or partially hydrophobic surfaces. Furthermore, the degree of hydrogen-bond ordering of water observed prior to hydrate nucleation, as measured by the ratio of the inter- and intramolecular Raman OH bands, was found to have an inverse relationship with the mean induction time for hydrate formation. Following hydration formation, no significant difference in the water molecule structure was observed in the hydrate phase based on their Raman OH bands, irrespective of surface hydrophobicity. These observations made with Raman spectroscopy provide the foundations for a quantitative link between hydrate nucleation promotion and water-ordering near solid surfaces, which could enable direct comparisons with results from corresponding molecular dynamics simulations.

Liquid irrigation is one of the key process control parameters following the construction of an ore leaching heap. This study uses 3D magnetic resonance imaging (MRI) to examine non-invasively the effect of liquid flowrate changes on heap hydrology when drip irrigation is used. Experimental results from a vertical column show that the increase in flowrate causes an increase in the number of rivulets in the ore bed. The new rivulets were found to be thicker, and their development caused an increase in liquid–solid contacting area which is considered advantageous for metal ion recovery. Experiments performed on larger samples showed that the effects of flowrate changes were limited to the region directly below the drip emitter because the increase in flowrate caused an increase in macro-pore flow and not capillary retention of liquid. Therefore the increase in flowrate was not found to perturb liquid distribution patterns in a way that would be substantially advantageous to heap leaching recoveries.

The stability of hydrate-in-oil dispersions is a critical parameter in assessing the risk of flowline blockage due to particle aggregation or wall deposition. Many studies of hydrate particle transportability have used deionized water to form the dispersion; however, the resulting lack of ions means that the crude oil’s natural surfactants will be less active, which does not represent production conditions. This study presents a new investigation of both hydrate-in-oil dispersion stability and water-in-oil emulsion stability, measured with a differential scanning calorimeter (DSC) and low-field nuclear magnetic resonance (NMR) apparatus, respectively. The results show that hydrate-in-oil dispersion stability increases directly with sodium chloride (NaCl) mass fraction in the aqueous phase; above 5 wt % NaCl, the dispersion was observed to be stable over ten hydrate formation–dissociation trials. This was comparable with the dispersion stability observed previously when an ionic surfactant was dosed at 2 w...

Magnetic Resonance Imaging (MRI) velocimetry was applied to study non-invasively the water flow field inside a spiral-wound desalination membrane module (diameter: 2.5 in.; length: 18.5 in.), located in a pressure vessel, at typical practice operational conditions as a function of alginate fouling, simulating extracellular polymeric substances (EPS). Cross-sectional velocity images were acquired at an in-plane spatial resolution of 0.137 mm at multiple locations along the length of the reverse osmosis module and were acquired as a function of alginate concentration. At a total system alginate concentration of 3.25 mg/l, significant changes in the cross-sectional velocity map were observed near the module inlet due to alginate fouling, with limited changes observed in the middle and outlet regions of the module. When the total system alginate concentration was increased to 75 mg/l, it caused the module brine seal to fail resulting in significant local water flow by-passing the membrane module. This was clearly discernible in this opaque membrane system using MRI and resulting in dramatic changes in fluid velocity distribution through the membrane module. These observations of significant flow field heterogeneity as fouling develops are consistent with ‘irreversible’ fouling effects noted frequently in practice by the water treatment industry.

Kinetic hydrate inhibitors (KHIs) are polymeric based chemicals that delay the nucleation and/or suppress the growth rate of gas hydrate crystals. While KHIs have been used successfully to mitigate hydrate blockage risk during oil and gas production, the mechanisms by which they function remain unclear. In this work, multiple high-pressure stirred automated lag time apparatus (HPS-ALTA) were used to investigate the impact of a KHI on the subcooling formation probability distributions of methane hydrates and the subsequent initial growth rates. Over 3000 hydrate formation events were measured around 12 MPa using seven independent HPS-ALTA cells with KHI concentrations of up to 3 wt% in water. The addition of KHI made hydrate formation much less stochastic: significant reductions occurred in both the width of the formation probability distribution for a given cell, and in the offsets between distributions measured with different cells. Average initial hydrate growth rates were reduced by approximately a factor of 5 as KHI concentration increased, even though the average driving force (subcooling) increased by a factor of up to 3. However, above a KHI concentration of 0.3 wt%, a diminishing return was observed in both the nucleation delay and growth rate suppression. A Classical Nucleation Theory (CNT) framework was applied to investigate whether polymer adsorption onto active nucleation sites could explain the observed delay in formation onset. However, the CNT kinetic parameter extracted from the measured formation probability data increased with concentration, which is opposite to the dependence predicted by the polymer-adsorption model of nucleation suppression by KHIs.

Forward osmosis (FO) and reverse osmosis (RO) membrane processes differ in their driving forces: osmotic pressure versus hydraulic pressure. Concentration polarization (CP) can adversely affect both performance and lifetime in such membrane systems. In order to mitigate against CP, the extent and severity of it need to be predicted more accurately through advanced online monitoring methodologies. Whilst a variety of monitoring techniques have been used to study the CP mechanism, there is still a pressing need to develop and apply non-invasive, in situ techniques able to produce quantitative, spatially resolved measurements of heterogeneous solute concentration in, and adjacent to, the membrane assembly as caused by the CP mechanism. To this end, 23Na magnetic resonance imaging (MRI) is used to image the sodium ion concentration within, and near to, both FO and RO composite membranes for the first time; this is also coupled with 1H MRI mapping of the corresponding water distribution. As such, it is possible to directly image salt accumulation due to CP processes during desalination. This was consistent with literature expectations and serves to confirm the suitability of 23Na MRI as a novel non-invasive technique for CP studies.

The risk of natural gas hydrate formation in oil and gas flowlines increases when production moves into deep-water regions under high pressure conditions. Unlike hydrate dissociation, which is deterministic, the formation of gas hydrates is a well-known stochastic phenomenon which has not been fully characterized to date. Currently, it is often assumed that gas hydrate will form when the subcooling exceeds 3.6 °C (6.5 °F); this value is a heuristic based on limited field-tests. To more accurately predict when and where hydrates are likely to form, large experimental datasets are required to quantify probabilistic hydrate formation. Two generations of High Pressure Stirring Automated Lag Time Apparatus (HPS-ALTAs) were developed to repeatedly measure the formation driving force in term of subcooling and induction time. These HPS-ALTAs are capable of generating statistically significant datasets – typically on the order of 100 formation events – that enable the calculation of formation probability as a function of subcooling from hydrate equilibrium. The current results demonstrate that hydrate formation is strongly dependent on shear, with mean subcooling being reduced by 45% when the mixing speed was increased by a factor of seven. Additionally, initial induction time distributions have been measured at two subcooling values, and are described well by an exponential distribution as expected from theory.

Gas hydrates are ice-like solids that can form in crude oil flowlines at high pressure and low temperature; in severe cases, hydrate particles may aggregate in the oil phase and ultimately occlude flow in the line. In order to quantify the risk of plug formation, an accurate universal predictive model for hydrate-in-oil slurries is required. Currently, only one model has been proposed to predict the viscosity of hydrate-in-oil slurries, with an average deviation of 70% from new experimental data. In this work, a temperature-controlled, high-pressure rheometer with a vane blade rotor was deployed to study the rheological properties of hydrate-in-crude oil slurries, which were reacted from water-in-oil emulsions with and without anti-agglomerants (AAs). The results indicate that the addition of AA greatly reduced the viscosity of the hydrate-in-oil slurry, while the slurry maintained shear thinning behavior. The infinite shear viscosity data showed the basis of the current hydrate slurry model is does not appropriately capture the rheological properties of hydrate suspensions. By considering the shape effect of hydrate particles, a new model for hydrate-in-oil slurry has been proposed.

Transport in subsea gas gathering networks of hydrocarbons liberated from natural hydrate deposits poses many of the same risks as faced in more traditional hydrocarbon production. There is an inherent need to ensure that these fluids do not recombine into hydrates within the flow network, a familiar problem in flow assurance. This paper describes a risk based approach to assessing hydrate formation in the transport flowline of a produced hydrate deposit. The bathymetry is typical of those found off the Australian coast, with fluid properties as expected for a methane rich natural hydrate deposit. To fully assess the ‘risk’ of hydrate formation, probabilistic analysis based on Monte-Carlo simulation of the blowdown pressure and nucleation temperature are performed. This is coupled with targeted transient analysis of the severity of hydrate formation using a new plugin for OLGA®, which focuses on the mechanics of hydrate formation in gas dominant systems.

The viscosity of {xCH4 + (1 – x)C4H10} with x = 0.9484 has been measured at temperatures and pressures in the range (200 to 423) K and (2 to 30) MPa, respectively, corresponding to densities between (20 and 371) kg·m–3. The measurements were made using a vibrating-wire-viscometer with the wire clamped at both ends and operated in steady-state mode with a combined relative uncertainty of 1 %. The viscometer was also used to investigate the ability of a vibrating-wire instrument to determine the upper and lower dew pressures of the mixture in the retrograde region at (263 and 273) K. The dew pressures were determined by identifying the point along an isothermal pathway at which the slope of the wire’s resonance half-width with pressure exhibited a discontinuity. At the upper dew pressures near 10 MPa the results were consistent to within 0.2 MPa of predictions made using the GERG-2008 equation of state (EOS), while at the lower dew pressures near 3 MPa the agreement was within 0.3 MPa. To facilitate future ...

Asphaltenes are the heaviest and most polar class of compounds in crude oil, which may precipitate out of solution due to changes in the pressure, composition, or temperature. During production, aggregation between asphaltene solids may lead to viscosification of the oil phase and/or deposition of the solids on the flowline wall. This study presents the first measurement of asphaltene interparticle cohesive forces using a micromechanical force (MMF) apparatus, which is similar to that used previously to investigate gas hydrate interparticle cohesion. Asphaltene solids were precipitated from two crude oils, and cohesive force measurements were performed for particle pairs with diameters ranging from 100 to 200 μm. In air, the measured cohesive forces between the asphaltene particles were approximately one-half of those measured between hydrate particles in cyclopentane-saturated nitrogen vapor. Asphaltene cohesive force was measured in liquid cyclopentane, to provide a comparison against cyclopentane hydra...