Depolymerization of polysaccharides through the generation of free radicals at a platinum surface: a novel procedure for the controlled production of free-radical oxidations

Xanthan Stability at Elevated Temperatures

Summary

Xanthan stability is examined to define more clearly the polymers temperature limitations as a mobility-control agent. Experiments were performed to probe the relative importance of hydrolysis, oxidation, and helix-coil transition in xanthan degradation. In the absence of oxidizing agents (i.e., dissolved oxygen), results indicate that free-radical, oxidation/reduction reactions are not the dominant mechanism for xanthan degradation. Depending on the pH, acid-catalyzed hydrolysis and base-catalyzed fragmentation reactions may play important roles. With Arrhenius calculations, it was estimated that under ideal conditions (no dissolved oxygen, pH 7 to 8, and moderate to high salinities), a xanthan solution could maintain at least half of its original viscosity for a period of 5 years if the temperature does not exceed 75 to 80°C [167 to 176°F]. New polymers will be needed for chemical floods where xanthan does not have sufficient stability.

Introduction

About 50% of the oil in the U.S. that could be recovered by chemical flooding exists in reservoirs that have temperatures above 60°C [140°F].1 Stable, water-soluble polymers will generally be required during these floods to provide mobility control. Much progress has been made toward understanding the temperature limitations of acrylamide-based polymers - particularly in brines that contain divalent cations.2–6 However, there is still considerable uncertainty about the stability of xanthan solutions at elevated temperatures.

The goal of this work is to define more clearly the limits of stability for xanthan at elevated temperatures in the absence of dissolved oxygen. This paper first summarizes the known composition and structure of xanthan. Second, the literature is examined to review those types of mechanisms that could contribute to chemical cleavage of the xanthan-polymer backbone during long-term exposure to reservoir conditions. Next, techniques are briefly described for preparing, monitoring, and maintaining solutions with undetectable concentrations of dissolved oxygen (<2 ppb). Results are reported for experiments that probe the relative importance of hydrolysis, oxidation, and helix/coil transitions in xanthan degradation. Finally, temperature limitations associated with the use of xanthan as a mobility-control agent are discussed.

Xanthan Composition and Structure

Fig. 1 illustrates the chemical composition7 of xanthan. The backbone of the molecule is composed of glucose monomers connected by ß(1–4) gycosidic linkages. A side chain that contains the trisaccharide sequence mannose/glucuronic-acid/mannose is attached to every other glucose residue in the backbone. In each side chain, an O-acetyl group is usually bound to the mannose residue closest to the main chain of the polymer. Some of the terminal mannose units in the side chains may contain a ketal-linked pyruvate group. Depending on the bacterial strain producing the polymer, the fraction of side chains containing pyruvate may be 0%, 100%, or some intermediate value.7–11

Weight-average molecular weights reported for native xanthan samples have ranged from 2 million to 50 million Daltons.12–16 Most of the molecular weight values are in the low end of this range. Polydispersity indexes (Mw/Mn) between 1.4 and 2.8 have been published.13–15 The radius of gyration of native xanthan has been estimated to be between 0.1 and 0.4 µm in saline solutions.16,17 Xanthan has been modeled as a rigid-rod molecule whose length is between 0.6 and 1.5 µm14,16 and whose diameter is about 2nm [20 Å].16,18,19 Because the contour length of xanthan is thought to be in the range of 2 to 10 µm,15,18 the molecule is presumed to have some flexibility rather than being strictly a rigid rod.14,20,21 Consequently, xanthan has also been modeled as a worm-like chain whose persistence length is between 50 and 120 nm [500 and 1,200 Å].20,22,23 The rigidity of native xanthan is attributed to the helical structure of the molecule. Some researchers have argued that the xanthan helix is composed of a single polysaccharide strand,15,19,22 while others regard xanthan as a double-stranded helix.13,17 Still other researchers suggest that xanthan can assume different ordered configurations (including single- and double-stranded helixes) depending on salinity, temperature, and sample history.11,24

Xanthan Reactions and Degradation Mechanisms

Helix/Coil Transition.

Xanthan is capable of undergoing a helix/coil transition (or perhaps more correctly, an "order/disorder" transition) as temperature is increased.25,26 In aqueous solutions of very low salinity, this transition can occur near room temperature. The transition or "melting" temperature, Tm, increases in direct proportion to the logarithm of the salt concentration. The following formula, based on the data of Holzwarth,25 relates Tm to the molar sodium concentration, [Na+]:Tm=122+ 30 log[Na+]. (1)

For comparison, the data of Milas and Rinaudo26,27 can be described by use ofTm=125+ 43 log[Na+]. (2)

The melting temperature is much more sensitive to divalent cations than to monovalent cations. Eq. 3, which is based on Holzwarth's data, quantifies the effect of molar calcium concentration, [Ca++], on Tm:Tm=310+ 70 log[Ca++]. (3)

The melting temperature can be influenced by the acetate and pyruvate contents of the xanthan.28 Even so, Tm is expected to be quite high for the salinity and hardness levels that are characteristic of most formation brines. For example, in a 3 wt% NaCl solution, Tm is estimated to be 113°C [235°F] with either Eq. 1 or 2. The presence of calcium raises the melting temperature dramatically. In a 0.3 wt% CaCl2 brine, Eq. 3 predicts a value of 200°C [392°F] for Tm.

Xanthan stability at elevated temperatures is much greater in saline solutions than in deionized water.27 The helical conformation of xanthan in saline solutions is thought to protect the molecule from chain scission. In low-salinity solutions, however, xanthan will be in the disordered, coil conformation and, therefore, may be much more susceptible to chemical attack. A key question is whether helix/coil transitions are important during xanthan degradation in saline solutions.

Helix/Coil Transition.

Xanthan is capable of undergoing a helix/coil transition (or perhaps more correctly, an "order/disorder" transition) as temperature is increased.25,26 In aqueous solutions of very low salinity, this transition can occur near room temperature. The transition or "melting" temperature, Tm, increases in direct proportion to the logarithm of the salt concentration. The following formula, based on the data of Holzwarth,25 relates Tm to the molar sodium concentration, [Na+]:Tm=122+ 30 log[Na+]. (1)

For comparison, the data of Milas and Rinaudo26,27 can be described by use ofTm=125+ 43 log[Na+]. (2)

The melting temperature is much more sensitive to divalent cations than to monovalent cations. Eq. 3, which is based on Holzwarth's data, quantifies the effect of molar calcium concentration, [Ca++], on Tm:Tm=310+ 70 log[Ca++]. (3)

The melting temperature can be influenced by the acetate and pyruvate contents of the xanthan.28 Even so, Tm is expected to be quite high for the salinity and hardness levels that are characteristic of most formation brines. For example, in a 3 wt% NaCl solution, Tm is estimated to be 113°C [235°F] with either Eq. 1 or 2. The presence of calcium raises the melting temperature dramatically. In a 0.3 wt% CaCl2 brine, Eq. 3 predicts a value of 200°C [392°F] for Tm.

Xanthan stability at elevated temperatures is much greater in saline solutions than in deionized water.27 The helical conformation of xanthan in saline solutions is thought to protect the molecule from chain scission. In low-salinity solutions, however, xanthan will be in the disordered, coil conformation and, therefore, may be much more susceptible to chemical attack. A key question is whether helix/coil transitions are important during xanthan degradation in saline solutions.

Role of singlet oxygen in the degradation of hyaluronic acid

To investigate the effect of singlet oxygen on the molecular properties of hyaluronic acid, the polymer was irradiated in the presence of a dye sensitizer for singlet oxygen. Viscosity and circular dichroism techniques were used to monitor these changes. The relative viscosity of the polymer solution decreased steadily with increasing duration of irradiation, indicating an apparent decrease in molecular weight of hyaluronic acid. Circular dichroism measurements of the irradiated sample, however, did not show any appreciable change in the secondary structure, but do suggest that the generated singlet oxygen changes the tertiary structure and that this change is followed by a minor depolymerization.

Oxidant injury of lung parenchymal cells

Hyperoxia and paraquat ingestion are two clinical examples of lung injury thought to be mediated by oxidant mechanisms. An in vitro cytotoxicity assay using freshly explanted 51Cr-labeled lung tissue as the target was used to quantify the ability of hyperoxia and paraquat to directly injure lung parenchymal cells in an environment where indirect mechanisms such as recruitment of inflammatory cells were not possible. There are clear species differences in the susceptibility of lung parenchyma to direct injury by hyperoxia (95% O2) and paraquat (10 microM--10 mM) for 18 h at 37 degrees C, with human and rat lung being more sensitive than rabbit lung. Oxygen radical inhibitors, particularly catalase (1,100 U/ml) and alpha-tocopherol (10 micrograms/ml), reduced hyperoxia and paraquat-induced lung injury, although their ability to do so depended on the oxidant and the species. The simultaneous use of hyperoxia and paraquat accelerated the in vitro lung parenchymal cell injury in each species tested. These studies demonstrate that both oxygen and paraquat can directly injure the cells of the lower respiratory tract without enlisting the aid of additional blood-derived inflammatory cells. In addition, the 51Cr-labeled lung explant assay used for these studies allows for the quantitative assessment of direct lung cell injury and thus may prove useful as an in vitro model by which to investigate lung injury of other etiologies.

The inability of superoxide dismutase to inhibit the depolymerization of hyaluronic acid by ferrous ions and ascorbate

Superoxide radicals were investigated as to their capability of depolymerizing the hyaluronic acid of the bovine vitreous body. Using viscometry it was found that O2 radicals, generated by the hypoxanthine/xanthine oxidase method or the combination of NADH and phenazine methosulphate, degraded hyaluronic acid. This reaction was suppressed by superoxide dismutase, catalase, and peroxidase. In contrast, the depolymerization of hyaluronic acid by oxidation-reduction systems like ascorbic acid or ferrous ions was abolished by catalase and peroxidase while superoxide dismutase showed no effect.

Effect of oxygen-derived free radicals on hyaluronic acid

To investigate possible mechanisms of hyaluronic acid depolymerization, superoxide anion and other secondary oxygen-derived free radicals were generated in vitro and allowed to act upon a hyaluronate substrate. Superoxide, generated either enzymatically with xanthine oxidase or by stimulation of polymorphonuclear leukocytes, reduced the viscosity of hyaluronate solutions dramatically while the chromatographic profiles of the glycosaminoglycan shifted toward lower molecular weights. Superoxide-treated hyaluronate also became susceptible to further degradation by beta-N-acetylglucosaminidase A. Experiments with scavengers of various toxic oxygen-derived free radicals clearly implicated these reactants as mediators of hyaluronate depolymerization. Generation of superoxide by leukocytes in vivo may account for the loss of synovial fluid viscosity that accompanies inflammatory joint disease.

[The change of hyaluronic acid of the vitreous humour by oxidation-reduction-systems (author's transl)]

Purified hyaluronic acid of ox vitreous humour was isolated treating the acetone precipitate of a vitreous humour homogenate with 1 M NaCl solution and thereafter with cetylpyridiniumchloride. Both disc-electrophoresis and hydroxyproline content proved the absence of collagen in the purified hyaluronic acid. FeSO4, ascorbate, and cysteine changed the hyaluronic acid molecule and lowered the viscosity of the hyaluronic acid solution, EDTA alone did not affect the viscosity but enhanced the effectiveness of iron ions or ascorbate on the viscosity of the solution. Catalase prevented the reduction of the viscosity by the above mentioned substances. Therefore, it is suggested that H2O2 and free radicals are generated during the reaction. The free radicals produced are responsible for the change of the hyaluronic acid molecule.

[Electron microscopic investigations of vitreous collagen after treating the vitreous with liquefying substances (author's transl)]

By electron microscopic studies collagenase, hyaluronidase, HCl, ascorbic acid, and iron ions have been found to attack the collagen fibers of bovine vitreous. Because of the possible role of ascorbic acid in collagen synthesis and the ability of ascorbic acid to degrade hyaluronic acid and collagen we suggest that the ascorbic acid of the vitreous essentially participates in construction and metabolism of the vitreous body.

[The influence of riboflavin on vitreous homogenate (author's transl)]

Sunlight causes a decrease of viscosity of a mixture of riboflavin and ox vitreous homogenate while without riboflavin no reaction can be observed. The mechanism of this reaction is not yet clarified. It is possible that the reaction is closely related with the degradation of the viscosity of a hyaluronic acid solution by ascorbic acid cause of the production of H2O2 in both reactions. As an other mechanism the transfer of light energy on hyaluronic acid by riboflavin can be assumed.

Interaction of carcinogenic metals with tissue and body fluids

In addition to cobalt, metallic nickel, cadmium and other metals (e.g. zinc and copper) dissolve when incubated with horse serum at 37°. The dissolving property of copper in serum resembles that of cobalt, its solubility being increased greatly in the presence of oxygen, whereas the solubilities of cadmium, zinc and most preparations of nickel are the same aerobically and anaerobically. In all of these "metal-sera" the cations are bound, although in different proportions, both by proteins and by small diffusible molecules.Although Co(2+) ions and cobalt-serum cause limited catalytic oxidation of fresh serum, most of the oxygen uptake by suspensions of metallic cobalt in serum, or by more simple model systems, is due to absorption of oxygen by the metal powder; the consequences of this are discussed.Metallic cobalt, cadmium and nickel dissolve readily when incubated with sterile homogenates of rat muscle (and other tissues), the dissolved cations being bound predominantly by small, diffusible molecules, rather than by the protein components. Binding by small molecules, in preference to proteins, also occurs when the metals dissolve in vivo. In both the in vivo and in vitro systems, the metallic ions are not bound by a specific cation carrier, but are distributed amongst a number of components. These components have greater affinities for the dissolved metals than serum proteins and seem likely to be the normal cation carriers in vivo. As in serum, solubility in muscle homogenates is not a specific property of the carcinogenic metals as other, non-carcinogenic metals also dissolve. The specificity of the former metals, therefore, is attributed to the subsequent effects of the dissolved cations after intracellular incorporation.