Scrivener Publishing
100 Cummings Center, Suite 541J
Beverly, MA 01915-6106
Publishers at Scrivener
Martin Scrivener (martin@scrivenerpublishing.com)
Phillip Carmical (pcarmical@scrivenerpublishing.com)
Edited by
Ying (Alice) Wu, John J. Carroll, Mingqiang Hao and Weiyao Zhu
This edition first published 2020 by John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, USA and Scrivener Publishing LLC, 100 Cummings Center, Suite 541J, Beverly, MA 01915, USA
© 2020 Scrivener Publishing LLC
For more information about Scrivener publications please visit www.scrivenerpublishing.com.
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, except as permitted by law. Advice on how to obtain permission to reuse material from this title is available at http://www.wiley.com/go/permissions.
Wiley Global Headquarters
111 River Street, Hoboken, NJ 07030, USA
For details of our global editorial offices, customer services, and more information about Wiley products visit us at www.wiley.com.
Limit of Liability/Disclaimer of Warranty
While the publisher and authors have used their best efforts in preparing this work, they make no representations or warranties with respect to the accuracy or completeness of the contents of this work and specifically disclaim all warranties, including without limitation any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives, written sales materials, or promotional statements for this work. The fact that an organization, website, or product is referred to in this work as a citation and/or potential source of further information does not mean that the publisher and authors endorse the information or services the organization, website, or product may provide or recommendations it may make. This work is sold with the understanding that the publisher is not engaged in rendering professional services. The advice and strategies contained herein may not be suitable for your situation. You should consult with a specialist where appropriate. Neither the publisher nor authors shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages. Further, readers should be aware that websites listed in this work may have changed or disappeared between when this work was written and when it is read.
Library of Congress Cataloging-in-Publication Data
ISBN 9781119592068
Cover image: Jim J. Caroll
Cover design by Kris Hackerott
The eighth edition of the International Acid Gas Injection Symposium (AGIS VIII) was held at the end of September 2019 in Calgary, Canada. Presentations and posters covered various topics related to (1) acid gas injection [AGI], (2) carbon capture and storage [CCS], and (3) the use of CO2 for enhanced oil recovery [EOR]. This volume contains select papers presented at the Symposium. These included primary research such as the solubility of CO2 and the ionic speciation of CO2 in amine solutions. It also includes safety considerations such as pipeline leak detection and the dealing with a blowout from an acid gas well. And through to applications, including field projects. It is worth noting that Chap 1 is the paper that was presented by Mitch Stashick and was awarded the Best Student Paper at the Symposium.
Mitchell J. Stashick, Gabriel O. Sofekun and Robert A. Marriott*
Department of Chemistry, University of Calgary, Calgary, Alberta, Canada
Abstract
Transportation and handling of molten sulfur is inevitably challenging due to the anomalous rheological behaviour of sulfur with changing temperature. After melting at 115 °C, sulfur’s viscosity remains close to 10 cP. The onset of the λ-transition region is observed at T ≈ 160 °C. The viscosity drastically rises in this region, increasing to a maximum of about 93000 cP at 187 °C. This anomaly is due to the cleavage of sulfur rings and production of reactive diradical sulfur species that concatenate to create sulfur polymers. The entanglement of these sulfur chains causes the dramatic rise in viscosity. Within this work, new data is reported that examines the modifying effects of hydrogen sulfide (H2S) within liquid sulfur. This may be used when considering the potential injection of liquid sulfur into depleted reservoirs for safe long-term storage. H2S, when present in liquid sulfur, can either physically dissolve or chemically react to generate polysulfane. The chemical reaction causes significant changes in the sulfur chain distribution and consequently changes the viscosity curve of liquid sulfur as a function of temperature. This study reports viscosity measurements performed from 120 < T < 280 °C with concentrations of H2S in sulfur ranging from 0 ppmw to 284 ppmw.
Keywords: Rheology, viscosity, λ-transition, sulfur, polymers, hydrogen sulfide
Sulfur requires transportation and handling on an industrial scale. After acknowledging that there were approximately eighty million metric tons of elemental sulfur produced globally in 2018, this becomes abundantly clear [1]. Sulfur recovery from oil refineries and sour gas treatment plants (sour gas is natural gas with hydrogen sulfide (H2S) > 5.7 mg∙m-3) has continued to increase, bringing about more deliberation on the utilization and/ or storage of sulfur product. When the value of sulfur is not high enough to economically justify transportation and sale of the material, it is often common to block pour for long-term storage. Block pouring sulfur is a relatively inexpensive method where sulfur is solidified within retaining walls on large plots of land. However, there are some disadvantages to this technique. The land on which the blocks are poured must be leased or owned and this can cause some financial strain. Also, bacteria classified as Thiobacilli are known to metabolize sulfur by oxidizing the material to sulfate 
and in doing so produce sulfuric acid (H2SO4). This along with rain and further weathering results in the need for water treatment of the acid runoff, incurring additional cost. Lastly, large amounts of weathering sustained by sulfur blocks can yield high content of materials such as dirt, sand and moisture. This can create issues of purity if the price of sulfur rises and it makes economic sense to sell the product again [2]. Alternatively, producers have chosen to inject H2S (acid gas injection) to mitigate stranded sulfur. With acid gas injection, one needs to consider the energy required for compression of gas, storage under pressure and the loss of energy due to eliminating the Claus plant.
In future operations, an alternative to block pouring or acid gas injection could include the potential injection of liquid elemental sulfur into depleted reservoirs for safe long-term storage. This method would diminish some of the issues experienced with block pouring, but would likely sustain other costs needed for infrastructure and operation. Also, other fundamental understanding is currently lacking in order to pursue this approach. The conditions under which an injection such as this would take place must be known and understood in how they will impact the rheological behavior of sulfur.
One of several conditions to consider is that all sulfur produced in Sulfur Recovery Units at refineries and gas processing facilities will contain some residual H2S. This is because sulfur is recovered before complete conversion from H2S. Generally, recovered sulfur from a plant process can have up to 380 ppmw of dissolved H2S depending on the facility set up [3]. Another condition to consider is the temperature of the depleted underground reservoir. For this, the range of interest for potential injection was found to be around 120 < T < 280 °C. It should be recognized that this temperature range includes the λ-transition region of liquid sulfur. Also, high shear must be considered as this will be encountered in sulfur pumps and the near-wellbore region. Studying sulfur’s shear flow with and without H2S over the specified temperature range could therefore greatly improve the fundamental understanding needed for injection of molten sulfur.
The viscosity temperature dependence of sulfur within the λ-transition region can be explained by a scission-recombination equilibrium for polymers and the term reptation. In this occurrence, reptation is defined as the thermal motion of entangled macromolecular sulfur chains. At T > 160 °C, cleavage of sulfur S8 rings results in the generation of reactive diradical sulfur species that combine to create sulfur polymers [4]. The average polymer chain lengths increases as a function of temperature leading to more entanglement and higher viscosity values. At 187°C, the maximum polymer chain length is reached, along with the maximum viscosity. Beyond this temperature, the viscosity decreases due to the thermal scission of polymer chains; however, if H2S is present, the viscosity profile of sulfur changes depending on concentration. H2S is known to chemically react with diradical sulfur species during the λ-transition to produce polysulfane, as shown in Eq. (1-1). This reaction terminates the polymerization, which reduces the average sulfur polymer chain length. Therefore, a reduction in the viscosity of the liquid is observed [5-7].
Oxygen-free, nitrogen gas (N2; 99.998% as per certificate of analysis) was purchased from Praxair Technology, Inc. Hydrogen sulfide gas (H2S; 99.6% containing N2, Ar and trace CO2 as per certificate of analysis and in-house gas chromatography analysis) was purchased from Praxair Technology, Inc.
Sulfur used for this study was air prilled and supplied by Keyera Energy from their Strachan, Alberta, gas plant in 2017. Analyses of the air prills were performed and reported in Table 1.1. Moisture (volatiles) content, total nonpolymer sulfur content/overall purity, and ash content were determined by methods used at ASRL, reported by Sofekun et al. [4]. Total carbon content and H2S content were also performed on the sulfur prills. Additionally, it was necessary to perform analyses of H2S content before and after rheometric measurements. Total carbon content was measured using an analysis method established by Dowling et al. [8]. H2S content was measured using an analysis method detailed by Sofekun et al., Marriott et al., and Tuoro and Wiewioroski [3, 4, 9, 10]. Large sulfur samples that had been equilibrated with H2S were partially used to record infrared spectra (six total measurements) prior to each set of rheometric measurements. All samples were analyzed using a Nicolet 380 FT-IR (Thermo Electron Corp.) spectrometer.
Table 1.1 Analytical results of the Keyera air prilled sulfur.
| Moisture Content (wt. %) | Purity (wt. %) | Ash Content (ppmw) | Total Carbon Content (ppmw) | Total H2S Content (ppmw) |
| 0.14 | 99.995 | 8 | 34 | < 0.2 |
The experimental system was composed of an Anton-Paar MCR 302 rheometer with a custom built gravity-fed charging rig, a high-pressure N cylinder and a thermostated Fourier Transform Infrared (FT-IR) cell used2 fromprevious studies [4, 9, 10]. A head pressure tee for inducing H2S partial pressures along with a purging setup connected to KOH traps for removal of H2S before venting was added. This permitted the safe introduction of differing concentrations of H2S in liquid sulfur samples. All valves and tubing used for both the pressure tee and gravity-fed charging rig were made of 316L stainless steel. The system was housed inside a negative-pressure bay equipped with automatic toxic gas detectors.
Geometry and measuring limitations for the Anton-Paar MCR 302 were detailed by Sofekun et al. [4]. Air checks and motor adjustments were performed at a rotational speed of 0.3 min-1 at set time intervals as recommended by the instrument manufacturer. This confirmed on a regular basis that the rheometer was in good working condition.
The procurement of viscosity data sets corresponding to each differing H2S concentration in sulfur were carried out using the following experimental procedure: Two hundred grams of high purity sulfur prills were placed in an oven overnight at 135 °C. The experimental system was prepared for rheometric measurements of liquid elemental sulfur by inputting set points for the temperature controllers on the gravity-fed charging rig and rheometer cylinder jacket of 135 °C. After being held at the set temperature for 2 hrs, the pressure tee was removed from the top of the reservoir. The reservoir was then filled with 80 mL of liquid elemental sulfur. Following this, the pressure tee was reconnected to the top of the reservoir, purged and pressurized with pure H2S or a H2S/N2 mixture. The gases in the headspace of the reservoir were then allowed to equilibrate with the sulfur for at least 24 hrs. Following equilibration, the H2S dissolved in sulfur was extremely stable. This allowed the pressure tee to be purged with N2 and subsequently, the safe loading of the rheometer cell.
To load the rheometer cell, a plug on the sample cell was removed and the pressure tee was disconnected. This allowed for liquid sulfur to be transported through the charging rig by gravity after opening a valve on the sulfur reservoir. This valve was metered until sulfur filled the rheometer cell. Analyses by FT-IR to determine the concentration of dissolved H2S in sulfur were performed by sampling twice through a drain valve below the rheometer before filling the rheometer cell once more with a final sample used for viscosity measurements. The plug on the sample cell was then retightened, all valves were confirmed closed, and the sulfur reservoir was capped.
For measurement and recording of data, the desired T and shear rate, γ, were programmed using the Rheoplus software interface. Heating of the sample was done at 5 °C intervals from 120 to 280 °C. The measurements at each temperature were made after specified equilibration times. Average Newtonian viscosities were obtained at each temperature using a data processing code written in Microsoft Visual Basic for Applications in Excel. The code yields uncertainty in each viscosity value at a 95% confidence interval.
Four charges of liquid sulfur without H2S were measured to compare the maximum viscosity value, ηmax, with the previous study [4] and values reported by Bacon and Fanelli, Doi, and Ruiz-Garcia et al. [11-13]. At 187 °C, the values were in agreement with past studies, ηmax ≈ 93000 cP. When comparing the ηmax of Sofekun et al. to the average viscosity value of the four charges in this study, there was a relative change of 2.10%. This indicated that no hydrocarbon impurities were present in the rheometer cell during this study [4]. Average Newtonian viscosities were then measured at every 5 °C across the λ-transition of sulfur containing differing concentrations of H2S. The average Newtonian viscosity data for each specified H2S concentration in this study, was plotted as a function of temperature (Figure 1.1). The data plotted for sulfur with approximately no H2S present was reported by Sofekun et al. [4].
Some general trends were gathered from Figure 1.1. One clear deduction was that as H2S concentration increases, the viscosity of liquid sulfur decreases. Another definitive trend was that the viscosity curve or peak maximum shifted to higher temperatures with larger concentrations of dissolved H2S. Although measured viscosity values by Fanelli and Rubero were much lower due to high concentrations of H2S, the trends reported were in accordance with trends found in these studies [5, 7]. This study reports reliable data with more industrially-relevant concentrations of H2S in sulfur. The concentrations from this study would be more commonly found in recovered sulfur from natural gas processing plants or oil refineries and therefore would be more relevant in consideration of storage by subsurface injection.
To speculate on the trends or changes in viscosity observed upon the addition of H2S in sulfur, Eq. (1-1) must once again be contemplated. The H2S present in sulfur acts as another reaction pathway where polymerization is terminated to form a shorter polysulfane chain. Normally, without H2S, the sulfur chain length distribution at each temperature would be determined by the chemical equilibrium established by scission and recombination of polymers. However, due to the new reaction pathway, the kinetic rate of termination interferes with the rate of recombination and/or polymerization resulting in a shift of the viscosity-temperature profile. Besides the profile shape change, there was a reduction in viscosity. As mentioned before, this can be attributed to shorter polymer chain lengths. Shorter polymers results in less entanglement which influences the reptative behavior. With fewer entangled macromolecular chains, there are less intermolecular interactions to break allowing for an ease of thermal motion. Chain relaxation is faster and fluid deformation requires less force.
Figure 1.1 Average Newtonian viscosity versus temperature for liquid sulfur containing different concentrations of H2S. ●, heating run of 0 ppmw H2S in sulfur; +, heating run of 15.9 ppmw H2S in sulfur; ▲, heating run of 35 ppmw H2S in sulfur; ×, heating run of 79 ppmw H2S in sulfur; ■, heating run of 159 ppmw H2S in sulfur; ♦, heating run of 284 ppmw H2S in sulfur [4].
The rheometric properties of liquid sulfur with varying concentrations of dissolved H2S were measured. This was accomplished by using an AntonPaar MCR 302 rheometer with a custom built gravity-fed charging rig and a head pressure tee for inducing H2S partial pressures. Sulfur with H2S concentrations ranging from 0 ppmw to 284 ppmw were measured at 5 °C intervals over temperatures of 120 °C to 280 °C. This permitted investigation of the impact of H2S on the λ-transition region of sulfur. Introduction of H2S led to overall reductions in viscosity, as well as a peak shift to higher temperatures for the entire viscosity profile of sulfur. This was presumed to be due to H2S terminating the polymerization of sulfur, which gave shorter polymer chain lengths and altered each temperature’s chemical equilibrium.
The rheometric data presented within this study for liquid sulfur with industrially relevant concentrations of H2S should help in understanding the feasibility of operations contemplating this materials subsurface injection into depleted reservoirs. Currently, more efforts are being made to produce a semi-empirical correlation model based on the reptation model of Cates to estimate the properties discussed in this study [4, 14]. A semi-empirical model could aid in the simulation of subsurface injection or assist operators in the use of sulfur pumps. Future research could also include investigating the impact of other impurities on liquid sulfur’s rheometric properties. If experimental data were available, these properties could be estimated using a similar model.