Next Article in Journal
Optimization Control Strategy for a Central Air Conditioning System Based on AFUCB-DQN
Next Article in Special Issue
A Volume Fracturing Percolation Model for Tight Reservoir Vertical Wells
Previous Article in Journal
Performance Measurement and Optimization for Sustainable Production Process Improvement—Volume II
Previous Article in Special Issue
Biodiesel Production from Waste Cooking Oil: A Perspective on Catalytic Processes
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Processes
Volume 11
Issue 7
10.3390/pr11072067
processes-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Persistent Free Radicals in Petroleum

by
Lina M. Yañez Jaramillo
1,
Joy H. Tannous
2 and
Arno de Klerk
1,*
1
Department of Chemical and Materials Engineering, University of Alberta, 9211-116th Street, Edmonton, AB T6G 1H9, Canada
2
Department of Chemical and Petroleum Engineering, United Arab Emirates University, Al Ain P.O. Box 15551, United Arab Emirates
*
Author to whom correspondence should be addressed.
Processes 2023, 11(7), 2067; https://doi.org/10.3390/pr11072067
Submission received: 31 May 2023 / Revised: 26 June 2023 / Accepted: 6 July 2023 / Published: 11 July 2023
(This article belongs to the Special Issue Petroleum Characterization and Bioprocesses: Numerical and Experimental Investigation)
Browse Figures
Versions Notes

Abstract

:
The persistent free radical content in petroleum is of the order 1018 spins/g (1 μmol/g), with higher and lower values found depending on origin and in different distillation fractions. The field of persistent free radicals in petroleum was reviewed with the aim of addressing and explaining apparent inconsistencies between free radical persistence and reactivity. The macroscopic average free radical concentration in petroleum is persistent over geological time, but individual free radical species in petroleum are short-lived and reactive. The persistent free radical concentration in petroleum can be explained in terms of a dynamic reaction equilibrium of free radical dissociation and association that causes a finite number of species at any given time to be present as free radicals. Evidence to support this description are observed changes in free radical concentration related to change in Gibbs free energy when the bulk liquid properties are changed and responsiveness of free radical concentration to dynamic changes in temperature. Cage effects, solvent effects, steric protection, and radical stabilization affect free radical reaction rate but do not explain the persistent free radical concentration in petroleum. The difference between persistent free radicals in straight-run petroleum and converted petroleum is that straight-run petroleum is an equilibrated mixture, but converted petroleum is not at equilibrium and the free radical concentration can change over time. Based on the limited data available, free radicals in straight-run petroleum appear to be part of the compositional continuum proposed by Altgelt and Boduszynski. Persistent free radical species are partitioned during solvent classification of whole oil, with the asphaltenes (n-alkane insoluble) fraction having a higher concentration of persistent free radicals than maltenes (n-alkane soluble) fraction. Attempts to relate persistent free radical concentration to petroleum composition were inconclusive.

1. Introduction

Persistent free radicals are present in high-molecular-mass natural raw materials like petroleum and coal. The concentration of these free radical species can be quantified using electron spin resonance (ESR) or electron paramagnetic resonance (EPR) spectrometry. This technique exploits the magnetic moment of the unpaired electron, which makes substances with an unpaired electron paramagnetic.
The concentration and electronic environment of unpaired electrons in a substance can be detected by subjecting the material to a magnetic field. When subjected to an external magnetic field, those unpaired electrons that are aligned with the magnetic field will have a lower energy than those unpaired electrons aligned against the magnetic field. When sufficient energy (h.ν) is provided for the unpaired electrons to change their spin state under the influence of a magnetic field (B0), the amount of energy absorbed, as well as the frequency of the electromagnetic radiation, provides information about the concentration and nature of the unpaired electron. Specifically, this relationship is defined as the Landé g-factor (Equation (1)).
g = h . ν µ B . B 0
In this equation, h is the Planck’s constant = 6.626 × 10−34 J/s, ν is the microwave frequency of the ESR in Hz, µB is the Bohr magneton = 9.273 × 10−28 J/Gauss, and B0 is the magnetic field in Gauss. Organic free radicals can easily be distinguished from other paramagnetic species present in petroleum, such as the vanadyl ion (VO2+), based on the difference in g-factor. The unpaired electrons in organic free radicals have g-factors close to that of a free electron, g = 2.0023.
This study is limited to persistent organic free radicals.
Reported concentrations of the organic free radicals in heavy petroleum and coal are typically of the order 1018 spins/g. This is equivalent to about 1 μmol/g. Higher and lower values can be found depending on the origin and distillation range of the material (Table 1) [1,2]. If one then considers the high average molecular mass of heavy petroleum fractions and coal, the concentration of persistent free radical species in molar concentration is around 0.1 mol%.
The persistence of free radicals in petroleum and coal is intriguing because these free radical species appear to have been present over geological time. Free radical persistence over geological time is incongruent with the description of free radicals as reactive species.
Herein lies a conundrum for those interested in petroleum conversion processes that proceed by free radical chemistry, such as visbreaking and coking. The reactions taking place in thermal conversion processes are usually described in terms of free radical initiation, propagation, and termination reactions [3]. The notion is that sufficient temperature is required for homolytic bond dissociation to produce free radicals, i.e., initiation, so that thermal conversion can take place. Yet, as was pointed out, the amount of persistent free radicals present in heavy petroleum fractions is already of the order 0.1 mol%. The implication is that thermal conversion can proceed at a lower temperature than required for initiation by homolytic bond dissociation.
The purpose of this work is to review the field of persistent free radicals in petroleum. It is a specific aim to address and explain the apparent inconsistencies between free radical persistence and reactivity. It is a further aim to explore the relationship between the nature of the free radical species and how it is affected by and related to bulk properties. To do so and to guide the narrative, several questions will be posed and answered to the extent possible. To illustrate specific concepts, the focus will be on petroleum and thermal conversion processes applied to petroleum, although the work can be equally applied to other materials.

2. What Is Meant by Persistent Free Radicals?

The term persistent free radical was defined by Griller and Ingold [4] as “…a radical that has a lifetime significantly greater than methyl under the same conditions…”. They were also at pains to point out that there is a difference between “persistent”, “stable”, and “stabilized” free radicals.
The term stable free radical was reserved for [4]: “…a radical so persistent and so unreactive… under ambient conditions that the pure radical can be handled and stored in the lab with no more precautions than would be used for the majority of commercially available organic chemicals”. A commonly encountered example is molecular oxygen (O2), which is a stable diradical species.
Most persistent free radical species can therefore not be classified as stable free radicals according to the preceding definition. Their persistence depends on their chemical environment and physical state [5]. It is therefore possible that under appropriate circumstances, free radical species that would not normally be persistent appear to be persistent.
The triphenylmethyl radical is a prototypical example of a persistent free radical species, which is stabilized by delocalization of the unpaired electron that is delocalized into the adjacent systems of π-bonds of the phenyl groups [4]. This makes it a stabilized free radical, but it is not a stable free radical that can be isolated and stored for an extended period of time. In fact, the extent of stabilization of the triphenylmethyl radical is about the same as for a benzylic radical because the three phenyl rings are not in the same plane but at an angle to each other, like a propeller. Both electron delocalization and steric hinderance contribute to the persistence, with a half-life of the triphenylmethyl radical in dilute solution that is of the order of milliseconds [6].
The conundrum that arose due to observed free radical persistence in petroleum and the apparent lack of reactivity was an artifact of semantics. In the petroleum community the meaning of the term “persistent free radical” morphed into a description that implied that the radicals were both long-lived and unreactive. This was a false impression. It was demonstrated that the persistent free radical species in petroleum were reactive [7], and evidence of hydrogen transfer at 60–100 °C was presented [8,9]. A change in free radical concentration was also reported after keeping petroleum samples at 60 °C for 2 h [10]. It therefore appears unlikely that one would be able to isolate and store the pure radical species found in petroleum as unreactive materials, which are requirements for classification as stable free radicals according to the terminology of Griller and Ingold [4].
In petroleum, the macroscopically measured concentration of organic free radical species remains constant at constant ambient conditions. The apparent longevity of the free radical species in petroleum meets the criterion of description as persistent free radicals but does not imply anything about the nature of the radicals, radical reactivity, or reason for the radical persistence.

3. What Makes Free Radicals in Petroleum Persistent?

3.1. Cage Effect

When radicals are formed from non-radical precursors, two radicals are formed in close proximity to each other and the reaction yield in the liquid phase is usually less than in the gas phase for the exact same reaction and conditions. To explain this difference, reference is made to the concept of a cage effect [11,12].
The cage effect can be explained in terms of the effect of the solvent molecules surrounding free radicals, which effectively forms a solvent cage around the newly formed radicals (Figure 1). Multiple collisions may have to take place before the radicals can diffuse past those solvent molecules surrounding them to escape the solvent cage. This could increase the probability of recombination, since the newly formed radical species are neighbors in solution. Considering that all species in solution are surrounded by other molecules, Lorand [11] elected to reserve the term cage only for the solvent molecules surrounding a pair of newly generated radicals that were generated by the same event, so-called geminate radicals. The cage effect is therefore the consequence of the lower diffusion rate in the liquid phase compared to the gas phase. The extent to which the cage effect will affect a specific reaction is dependent on the properties of the liquid phase that affects the diffusion rate.
For solvents that are mixtures, bulk liquid viscosity is not necessarily a good predictor of the cage effect. It was shown that cage recombination rate was poorly related to bulk viscosity (η, Pa.s) but was instead related to the inverse diffusion coefficient (1/D, s/m2) they called “microviscosity” [13,14,15], which has the same units as inverse kinematic viscosity (1/ν, s/m2).
The cage effect is a purely physical phenomenon. The solvent can of course also affect the rate of radical recombination reactions by affecting the stability of the free radicals. These chemical interactions are solvent-effects [16] and are discussed in Section 3.3. Changes in recombination rate due to chemical interaction with the solvent molecules within the cage are not considered a cage effect.
One of the explanations that was put forward in the petroleum literature for the presence of persistent free radicals in petroleum was that of the cage effect [17]. Caging by aggregation was offered as the explanation for persistence over time and to explain how some radicals could survive under strong reducing conditions [18]. As envisioned, the interaction was more than just a physical caging effect because the interaction was not just decreasing the diffusion rate to slow mass transport; it was an interaction that effectively prevented reaction with the free radicals species. Along similar lines, it was postulated that free radicals played a role in the aggregation of asphaltenes [19], although in that work, aggregation was not used as an explanation for free radical persistence; it was instead a consequence of persistent free radicals in petroleum.
Aggregation is caused by chemical interaction or reaction between species that leads to an increase in the molecular weight of the aggregate when compared to the species involved in aggregation. Aggregation may then appear to physically keep the free radical persistent because aggregates can also de-aggregate, which is the reverse reaction. Thus, despite evidence for the relationship between aggregation and the free radical nature of the species involved in aggregation, this is not a cage effect but a reaction that appears to be causing a physical effect.
At the same time, aggregates when present in petroleum can cause a cage effect, as noted by Gray et al. [20]: “Free radical species that form within nanoaggregates due to thermal reactions would be restricted in their ability to react with the bulk liquid phase”. The newly formed radicals are still within the aggregate and the aggregate presents a more formidable diffusion barrier than monomeric species in solution. The physical restriction imposed by the aggregate structure is a cage effect because it is a physical obstacle that reduces the diffusion rate of the newly formed radical species. No claims were made in that work [20] that the radicals formed within aggregates would persist indefinitely, only that the radical species would have more restricted access to the bulk liquid phase.
For the sake of argument, the concept of a cage effect can be applied to persistent free radical species in petroleum. The cage effect can explain how solvent molecules surrounding a persistent free radical species can shield it from radical–radical termination reactions, but it cannot explain how radical–radical termination reactions can be prevented. Even if the petroleum viscosity is very high, the diffusion coefficient in the liquid phase is not zero, which means that the rate of radical–radical termination reactions can only be slowed down. Analogously, if the free radical species are trapped within an aggregate structure, the radical termination is prevented only if the aggregate persists indefinitely. Since it was shown that persistent free radicals are reactive, this is not the case. In conclusion, the persistence of free radicals in petroleum cannot be explained by a cage effect.

3.2. Steric Protection

One of the causes of free radical persistence is steric protection [4], which affects the reaction rate of the free radical. Pendant groups surrounding the free radical center may physically hinder the approach of a reagent to the radical center, thereby reducing the probability of reaction during an interaction. The longevity of the free radical is increased, and meeting the criterion for being described as a persistent free radical (see Section 2) is easily met.
The size and nature of the pendant groups surrounding the free radical center may also affect the stability of the free radical. For example, Rüchardt [21] showed the relative impact of different pendant groups on the bond dissociation energy (BDE), which was related to the hybridization of the atom where the free radical is located, and more specifically to the bond angles of the pendant groups. Bulkier groups can also cause higher strain in the molecule, resulting in a greater relief when going from a tetrahedral (sp3) to trigonal planar (sp2) hybridization [6]. Thus, in addition to the steric protection offered by the pendant groups, the pendant groups also affect the thermodynamic stability of the free radical, which should be considered when assessing the impact of steric protection on reaction rate. This will be discussed in Section 3.3.
In the case of straight-run petroleum, its free radical concentration remains invariant. If this is due to steric protection, the steric protection must be absolute. Acevedo et al. [18] demonstrated that it was possible to reduce the free radical concentration of a petroleum asphaltenes fraction by boiling it in tetrahydrofuran (boiling point 66 °C). Since treatment at such mild conditions is unlikely to cause skeletal rearrangement to change the extent of steric protection of the persistent free radicals, steric protection cannot be used as explanation for free radical persistence over geological time.
Alili et al. [7] purposefully investigated the impact of steric effects on free radical reactivity of petroleum asphaltenes. Compounds with different pendant groups surrounding C=C were used as probe reagents to evaluate the relative reactivity of persistent free radicals in the asphaltenes fraction. The probe reagents themselves imposed steric restrictions on reaction. As expected, the reaction rate of reagents with sterically crowded C=C was slower, but with the exception of tetraphenyl ethylene, all other probe reagents resulted in measurable conversion during reaction with persistent free radicals in asphaltenes within 1 h at 250 °C (Figure 2) [7].
In conclusion, although the impact of steric protection of free radical species in petroleum cannot be disregarded, it does not offer an explanation for free radical persistence over geological time.

3.3. Radical Stabilization Effects

The stabilization of free radicals can be viewed in terms of the rate of their formation and disappearance. Radical stabilization has been described in terms of (i) thermodynamic stabilization, which is related to the bond dissociation energy (BDE) of the bond leading to radical formation, with contributing factors such as hyperconjugation, resonance, and captodative effects; and (ii) kinetic stabilization, which is controlled by the steric hindrance at a radical center and solvent stabilization effects.
While hyperconjugation makes reference to a “donation” of electrons from a σ-bond to a p-orbital which is electron deficient, resonance finds similarity with hyperconjugation with the difference that the electron is delocalized in order to stabilize the radical and lower BDE [21]. Another factor contributing to the thermodynamic stability is the captodative effect. The captodative effect refers to the presence and combined action of electron donating and electron withdrawing groups on a radical center. The stability results from the combination of having an electron donor and an electron acceptor substituent on the radical center that enables additional resonance structures that involve a separation of charge [22,23].
It should be pointed out that thermodynamic stabilization generally does not lead to the formation of persistent free radicals and that the lifetime of free radicals are determined mainly by kinetic stabilization effects [4].
Kinetic stabilization due to steric protection was discussed in Section 3.2, and it was already concluded that this type of stabilization is insufficient to explain free radical persistence over a geological time period as is observed in petroleum. A second type of kinetic stabilization is that provided by the properties of the bulk liquid or solvent.
One type of solvent effect that plays a role relates to the formation of solvent–radical complexes, causing a redistribution of the π-electron density [24,25]. For example, Isenberg and Baird [26] reported on the efficiency of polar solvents in which neutral molecules dissociate due to the solvent’s polarity lowering the energy by a specific interaction between the solvent and the radical.
Another type of solvent effect is related to the Gibbs free energy (ΔG) of the reagents and products from free radical decomposition in solution [5,16]. The ΔG of the free radical species in solution is not necessarily the same as that of the material from which the free radical species is formed, or the products from free radical termination. If the free radical species have lower ΔG in solution than that of the termination products, there is an added energy barrier to overcome. This will affect the equilibrium composition but may also lead to a kinetic effect.
The kinetic stabilization effects can lead to free radical persistence in the sense that the free radical lifetime is greater than that of a methyl radical under the same conditions but do not explain how free radicals can remain persistent over geological time.

3.4. Dynamic Reaction Equilibrium

Persistent free radicals in petroleum may be stabilized by several factors, none of which could explain the longevity of the free radicals over such a long time period. Furthermore, it was shown that the free radicals in petroleum are reactive. When the straight-run petroleum is perturbed through a change in temperature or composition, there is an impact. It is necessary to explain both the long-term free radical persistence and reactivity of the persistent free radicals.
The existence, persistence, and reactivity of persistent free radicals in petroleum can be described by a dynamic equilibrium between associated radical pairs and their dissociated forms [5,7]. This is illustrated by Figure 3. Such a description does not require the longevity of any individual radical species and it describes the persistence in terms of the relative rates of dissociation and association reactions.
The description of persistent free radicals in petroleum can be viewed in an analogous way to the description of persistent ions in water. There is a dynamic equilibrium in water between associated ion pairs (H2O) and their dissociated forms (H+ + OH−). No individual ionic species remains persistent, but due to the dynamic nature of the equilibrium, at constant conditions, the concentration of dissociated ion pairs in the water remains invariant over time.
What makes the description of the dynamic equilibrium associated with persistent free radicals in petroleum more difficult is that it is not a simple binary combination. The free radical species A• and B• in Figure 3 are not only engaged in a simple equilibrium with A–B but also all of the other free radical species—C•, D•, E•, and so forth—and the different potential binary combinations thereof. The macroscopically observed persistent free radical concentration is then the result of the multicomponent equilibrium of all radical species and combinations of radical species.
What evidence exists for this description of persistent free radicals in petroleum, and can this description provide a viable explanation for the observations reported in literature?

3.4.1. Impact of Bulk Liquid Composition

One aspect of the impact of the solvent on free radicals that was noted in Section 3.3 was the influence of the solvent on the ΔG of the free radical species in solution. If the persistent free radical species in petroleum is engaged in dynamic reaction equilibrium, then a change in the bulk liquid properties should have an impact on the free radical concentration. This was indeed what was reported.
It was found that the concentration of solvent affected the measured concentration of free radicals in petroleum (Table 2) [1]. To be clear, the concentration of the free radical species is expressed on the basis of the amount of analyte, the petroleum, which takes the dilution into account (i.e., number of spins/g of analyte). The data were taken from a single study because the quantification of free radical content has several pitfalls [27] which may introduce discrepancies in the absolute numbers reported between studies.
The change in the free radical concentration of the petroleum shown in Table 2 is a consequence of the change in bulk liquid composition. In straight-run heavy gas oil, the free radical concentration at ambient conditions was of the order 1 × 1017 spins/g but increased to 8 × 1017 spins/g when diluted to 5 wt% in toluene.
In the example shown in Table 2, the impact of changing the bulk liquid composition on the equilibrium free radical concentration was considerable. When this is interpreted in terms of ΔG, toluene stabilized dissociated free radicals in solution better than the petroleum fraction. The observation is consistent with the increased stabilization of chlorine radicals by aromatic solvents reported before [28], and presumably for a similar reason.
The bulk liquid composition can also be changed in different ways by making use of different solvents at the same concentration.
The effect of different solvents on the persistent free radical concentration in bitumen has been previously evaluated. Khulbe et al. [29] evaluated different solvents in the quantification of free radicals and noted that the number of free radicals in the bitumen–solvent or asphaltenes–solvent solution was dependent on the nature of the solvent. In their work, it was observed that the persistent free radical concentration of the analyte decreased exponentially with an increase in the dipole moment of the solvent.
In a study by Tannous et al. [30], it was also found that at the same level of petroleum dilution the type of solvent affected the free radical concentration. However, this study did not find a relationship between the free radical concentration in the solution with the dipole moment of the solvent. Additionally, bulk properties of the solvent such as molecular weight, refractive index, density, and viscosity were not correlated to the free radical concentration of the analyte. Table 3 [30] illustrates the effect of the type of solvent on the persistent free radical concentration. The measurements were all performed at the same concentration of bitumen.
Here, it is important to point out that some solvents are “lossy”, and this can affect the quality factor (Q-factor) of an ESR, which will affect the signal intensity in relation to the amount of free radicals present [27,31]. In practical terms, the Q-factor is a relationship between the amount of energy stored in the ESR per unit time and the amount of energy lost per unit time. As long as this ratio remains constant, the ESR spectrometer can be calibrated to quantify the free radical concentration of unknown samples against the concentration of free radicals in a known standard. When all samples are measured using the same type of ESR hardware, ESR probes, and bulk solvent, the Q-factor remains the same. However, when the bulk solvent is changed, there is a risk that the Q-factor may change.
The study by Tannous et al. [30] did not explicitly determine whether there was a change in Q-factor, but this was evaluated in the study by Elofson et al. [32]. They compared measurements of the free radical concentration of petroleum asphaltenes samples sealed in a capillary tube placed in different solvents and the same petroleum asphaltenes dissolved in those same solvents. It was reported that there was no change in Q-factor for benzene, toluene, pyridine, quinoline, and tetralin, with a constant free radical concentration measured for the asphaltenes sealed in the capillary tube. They noted that the free radical concentration changed when the asphaltenes were dissolved in the solvents, which means that the change was not due to a change in Q-factor, but due to the interaction of the solvent with the asphaltenes. The only solvent that was tested for which the Q-factor changed was nitrobenzene.
It can be concluded that variation in the nature and concentration of solvent species at constant temperature caused a change in the measured persistent free radical content of petroleum. This is consistent with a description of persistent free radicals in terms of dynamic reaction equilibrium that will be affected by changes in ΔG caused by the bulk liquid properties.

3.4.2. Impact of Temperature

Reaction equilibrium is usually a function of temperature. If the persistent free radical concentration is a consequence of dynamic reaction equilibrium, it is expected that the measured concentration of free radicals should change with temperature. It is further expected that the free radical concentration should increase with temperature, because the dissociation is endothermic. This is unlikely to be a monotonic trend over a wide temperature range because the free radicals are reactive, and over time, at elevated temperature, it is expected that the measured free radical concentration will also be affected by non-equilibrium reactions.
The temperature dependence of free radical concentration is somewhat masked by the effect of Curie’s law. The Curie law states that the mass magnetic susceptibility (χ) of a paramagnetic substance is proportional to the inverse temperature (1/T, K−1). Whether the Curie law applies to all petroleum fractions is unclear. The work of Elofson et al. [32] showed little or no evidence that the Curie law was obeyed over the temperature range −196 to +23 °C for five different asphaltenes samples, although some temperature dependent changes were observed. Conversely, Hernández et al. [33] claimed that the Curie law was obeyed over the temperature range −183 to +237 °C for three different asphaltenes samples, although the data indicated that the fit was only approximate. Malhotra and Graham [34] also reported that bitumen and n-heptane solubility fractions obeyed the Curie law over the temperature range −263 to +27 °C.
A further challenge to such temperature dependent measurements is the reactivity of the material. As mentioned before, there were measurable changes in composition observed at 60 and 100 °C within 1–2 h [8,9,10]. Differently put, given sufficient time at elevated temperature, the free radicals will participate in some reactions that are not equilibrium reactions.
There is nevertheless evidence that the persistent free radical content in petroleum samples dynamically increase, as the temperature is dynamically increased from ambient conditions to about 150 °C [35,36].
Since the species in petroleum that are responsible for its persistent free radical nature are the products of transformations that took place over geological time, the near instantaneous responsiveness to changes in temperature and bulk properties (Section 3.4.1) points to a dynamic reaction equilibrium.

4. Are Persistent Free Radicals in Straight-Run and Converted Petroleum Different?

It would be convenient if the description of persistent free radicals for straight-run and converted petroleum is the same. The evidence that was presented to show that the apparent persistence of free radicals in petroleum is due to dynamic reaction equilibrium (Section 3.4) relied mostly on measurements made using straight-run petroleum and petroleum fractions.
It stands to reason that when persistent free radicals are found in converted petroleum products, the same arguments could be applied to explain the longevity of free radicals in terms of dynamic reaction equilibrium. However, there are some important differences in equilibration and the reaction pathways available to persistent free radicals in straight-run and converted petroleum:
(i) The species that give rise to the observed persistent free radicals in straight-run petroleum are species that over geological time, exhausted their reactive pathways to more stable products within the mixture. The straight-run material is an equilibrated mixture. When the petroleum is subjected to conversion, the material is no longer equilibrated.
(ii) Both hydrogen transfer and methyl transfer reactions can take place at meaningful rates at temperatures below the onset of thermal cracking [8,9,37,38,39]. There is evidence that petroleum processing at milder conditions than normally associated with thermal conversion leads to changes in the composition and properties of the petroleum. For example, changes in physical properties and spectroscopy of bitumen were clearly visible over a period of 8 h exposure to temperature of 150 °C under inert gas atmosphere [40]. After heating bitumen to 70 °C over a period of 5.5 h, gas evolution was measurably increased over the baseline emissions at 25 °C [41]. These are just two examples to illustrate that the temperature threshold for reactive change is in petroleum with a high persistent free radical content is quite low.
(iii) Conversion of petroleum changes the bulk liquid composition. It has already been shown that a change in bulk liquid composition will affect the concentration of free radicals. Depending on the severity of the change, the converted petroleum may also contain additional hydrogen acceptor species, such as olefins. Under inert ambient storage, the converted petroleum will continue to change, with changes in the persistent free radical content and properties of the material being apparent over a period of weeks (Table 4) [42].
To avoid the complication of dealing with non-equilibrated materials, subsequent questions about persistent free radicals in petroleum will be limited to straight-run materials. At the same time, some of the observations may equally apply to converted petroleum products, although no specific support for extending the conclusions to converted products is presented in this study.
Finally, it should be noted that exposure of straight-run petroleum to air and light can, over time, affect the free radical concentration. For example, the bond dissociation energy for free radical formation to initiate new free radical reactions can be provided by light of a sufficiently short wavelength (typically ultraviolet light) [43].

5. Is Persistent Free Radical Concentration Part of the Compositional Continuum?

One of the most powerful concepts in the characterization of petroleum is the concept of the compositional continuum. To quote from Altgelt and Boduszynski [44], “…compositional trends in fractions of increasing boiling point are continuous and that this continuity extends even to nondistillable residues”. This continuity was demonstrated for many properties, but the work did not include quantification of the free radical content. Nevertheless, there was an expectation that the compositional features giving rise to the persistent free radical nature of petroleum would also follow the compositional continuum.
There is a common trend that can be observed in the characterization of straight-run petroleum in terms of its distillation fractions. As the atmospheric equivalent boiling point of the distillation fraction increases, the persistent free radical content also increases. Atmospheric distillate does not have a detectable amount of persistent free radicals. As the boiling point of the petroleum fractions further increases, it is found that for light vacuum gas oil (LVGO), heavy vacuum gas oil (HVGO), and vacuum residue (VR), the increase in boiling point range corresponds to an increase in the persistent free radical concentration of those fractions (Table 5) [1].
The atmospheric equivalent boiling point temperature of petroleum is correlated with the average molecular weight of the material [44]. The compositional continuum can be rationalized by viewing the composition in terms of the probability of finding a specific structural element for species with a specific molecular weight. For example, there is a threshold molecular weight below which it is not possible to find trinuclear aromatic species, but beyond this threshold molecular weight, it is possible to find species that contain a trinuclear aromatic substructure.
This reasoning can be applied to species that form persistent free radicals. It is postulated that there are specific structural requirements that must be met by species to participate in the dynamic reaction equilibrium involved in maintaining a persistent free radical concentration. For example, we can see how specific structural groupings with condensed hydrocarbon rings that have an odd number of carbons (Figure 4) would naturally give rise to species that have a free radical nature. The likelihood of finding such structural elements increases with increasing molecular weight.
The simplest condensed aromatic structure with an odd number of carbons that is prone to free radical formation is indene, which forms the indenyl radical (•C9H7). Although indene is in the atmospheric distillate, with an atmospheric equivalent boiling point temperature of 182 °C, upon heating, it self-reacts to form a product with measurable persistent free radical content [45].
The aforementioned observations about the potential link between persistent free radical nature and structural elements in species explains why there is the expectation that persistent free radical content is a property that can be described in terms of the compositional continuum. Unfortunately, the only study that the authors are aware of that reported on the measured relationship between molecular weight and persistent free radical content is that by Rudnick and Tueting (Figure 5) [46].
In the study by Rudnick and Tueting [46], two straight-run oils were separated using preparative size exclusion chromatography and characterized using ESR to determine the persistent free radical content. The free radical concentration was determined based on a calibration with 1,l-diphenyl-2-picryl-hydrazyl (DPPH), and measurement reproducibility was reported to be within ±15% relative [46]. Based on current understanding, the molecular weights are likely over-estimated [47], but this does not detract from the value of this study, which employed a single consistent measure of molecular weight.
According to the data in Figure 5, the persistent free radical content has a regular relationship with molecular weight, which, in turn, is related to the atmospheric equivalent boiling point of the petroleum. Thus, both Table 5 and Figure 5 presented data showing an increase in persistent free radical content with an increase in atmospheric equivalent boiling point temperature. Based on the limited evidence presented, it appears that persistent free radical content in straight-run petroleum is a property that is part of the compositional continuum.

6. How Is Persistent Free Radical Concentration Related to Petroleum Composition?

Continuing from the tentative conclusion that the persistent free radical concentration is part of the compositional continuum of petroleum (see Section 5), it would be useful to know how the persistent free radical concentration is related to petroleum composition. Properties of petroleum that are part of the compositional continuum may also be correlated to the free radical concentration since both are correlated to atmospheric equivalent boiling point temperature. The challenge is therefore to differentiate between properties that are just correlated because they are both part of the compositional continuum and those that are not only correlated but also causally related to the origin of the persistent free radicals.
It is unlikely that a clear answer to the question can be provided with the current state of knowledge about persistent free radicals in petroleum. Some possible relationships can nevertheless be explored.

6.1. Aromatic Carbon Content in Relation to Free Radical Concentration

The example presented in Figure 4 suggested that there might be a causal relationship between the abundance of specific aromatic species and the persistent free radical concentration. Although that specific example was meant to be illustrative, it nevertheless presented a plausible relationship between aromatic carbon content and free radical concentration.
The persistent free radical concentration of different materials was reported along with their aromatic fraction [2,48], which indicated a relationship between these properties. The relationship spanned many different materials, which included converted materials from petroleum and coal.
However, the observation should be interpreted with caution. The apparent relationship between the aromatic fraction and the free radical concentration is not restricted to the aromatic carbon content, nor does it imply that the reported relationship involved carbon-centered radicals. In fact, further evidence was provided that extended the relationship to the heteroatom content of the materials, which is discussed next (Section 6.2).

6.2. Heteroatom Content in Relation to Free Radical Concentration

Generally speaking, carbon-centered radicals have g-factors closer to that of a “free electron” compared to heteroatom-centered radicals that have slightly higher values [27].
For pure compounds, the g-factor and hyperfine splitting of the ESR signal can provide detailed information about the nature of the radical center. Hyperfine splitting, or hyperfine coupling, is when the unpaired electron is coupled with a nearby nucleus with non-integer spin number, such as 1H, 13C, and 14N. The coupling affects the energy needed for the change in spin orientation. Hyperfine splitting in ESR is analogous to the effect of spin-spin coupling of nuclei in nuclear magnetic resonance (NMR) spectrometry.
However, the ESR spectrum of the organic radicals in petroleum consists only of a single absorption. In complex mixtures, the overlapping signals of many different radical species obscure the hyperfine splitting, and the observed g-factor (Equation (1)) becomes the weighted average of the mixture.
Yen and Sprang [49,50] pointed out that the heteroatom content in free radical species caused a systematic change in the g-factor of the organic radical peak. Specifically, it was claimed that there was a linear relationship between the g-factor and the sum of the N, O, and S elemental content in petroleum and coal. Retcofsky et al. [51] showed similar trends between the g-factor and O and S content in coals, and Elofson et al. [32] also presented data that suggested that there was a relationship between the g-value and heteroatom content. On the other hand, Malhotra and Buckmaster [52] found that the relationship, although observed, was not always statistically significant at 99% level of confidence.
The relationship between g-factor and free radical concentration is of interest because both are macroscopic averages that contain information about the nature of the free radicals. An increase in g-factor implies that, on average, there is a larger contribution of heteroatom-centered free radicals or an increase in carbon-centered radicals in proximity to heteroatoms and other radical centers that may cause a shift in the g-factor.

6.3. Watson K-Factor in Relation to Free Radical Concentration

The persistent free radical concentration in petroleum was expected to be the result of both the composition of the petroleum and the change in composition with boiling point, as is the case with other properties that follow the compositional continuum. It was therefore of interest to see whether the Watson K-factor (KW) was a property relationship that allowed the normalization of the impact of the compositional properties and distillation profile in relation to free radical concentration.
The KW is an empirical measure in petroleum refining that combines the mean atmospheric equivalent boiling point temperature (T50, °R) and density expressed as specific gravity (SG) to give an indication of composition (Equation (2)) [53].
K W = T 50 3 S G
The values for KW practically range from around 10.5 for very aromatic and naphthenic oils to around 13.0 for very paraffinic oils.
It was found that for oil sands bitumen from different locations that had similar values of KW, the persistent free radical concentrations varied more than could be explained by experimental uncertainty (Table 6) [1,54,55]. The mean boiling points for these materials were close, and it is known that the distillation profiles for Athabasca and Cold Lake bitumens are quite similar [56]. Although the values in Table 6 were not from the same study, the studies were from the same laboratory using the same analytical protocol. It is therefore tentatively concluded that the KW does not adequately capture the properties that determine the persistent free radical concentration.

7. Are Persistent Free Radicals in Petroleum Found Only in the Asphaltenes Fraction?

A solubility class of petroleum that is often encountered as a topic of study is asphaltenes [57]. Asphaltenes are defined as a material that is insoluble in an n-alkane solvent but soluble in an aromatic solvent. There are several standard test methods for the separation of asphaltenes from petroleum—for example ASTM D2007 [58], ASTM D4124 [59], and ASTM D6560 [60]—and asphaltenes are also encountered as an industrial product obtained from solvent deasphalting as process [53,56,57].
Persistent free radicals in the petroleum partition between the n-alkane soluble (maltenes) fraction and n-alkane insoluble (asphaltenes) fraction. The free radical concentration is not necessarily an additive property. Asphaltenes separated from petroleum have a higher persistent free radical content than the whole petroleum, and the maltenes separated from petroleum have a lower persistent free radical content (Table 7) [55]. The persistent free radicals in petroleum are therefore not limited to only the asphaltenes. In fact, the persistent free radical concentrations of several crude oils that were measured and reported, had little or no measurable n-pentane insoluble asphaltenes fraction [32].
As one would expect, the persistent free radicals in asphaltenes have features in common with persistent free radicals in whole petroleum, namely, the responsiveness of the free radical concentration to changes in solvent environment and dynamic changes in temperature (see Section 3.4) [29,61].
In the study by Adams et al. [62], the n-pentane insoluble asphaltenes fractions from different crude oils were further separated by size exclusion chromatography into different molecular weight fractions, and the free radical concentration of each fraction was determined by ESR spectrometry. The relationship between free radical concentration and molecular weight found is shown in Figure 6 [62].
The increase in persistent free radical concentration with the increase in molecular weight observed for whole petroleum (Figure 5) is also, to some extent, seen in the asphaltenes (Figure 6). The trend, broadly speaking, is one where the free radical concentration increases with molecular weight, but it is not a monotonic increase, and for several of the asphaltenes, the free radical concentration passed through a maximum with an increase in molecular weight.
The molecular weights were determined in a consistent way [62], but (as mentioned before) the values are too high based on current understanding [47]. The higher-molecular-weight fractions from size exclusion chromatography likely represent a higher extent of molecular-level aggregation [63], with a correspondingly larger difference between the monomeric molecular weight and reported value.
Other studies that report both free radical concentration and molecular weight [18,64] also found that the free radical concentration increased somewhat with an increase in molecular weight, but when sufficient data was available to see a trend, the trend was not monotonous.

8. Conclusions

The stated purpose of this work was to review the field of persistent free radicals in petroleum and to explain the apparent inconsistencies between free radical persistence and reactivity. The main conclusions were as follows:
(a) There is a difference between the longevity of an individual free radical species and the macroscopic average of free radical longevity in petroleum. Individual species in petroleum may be termed persistent free radicals because their longevity exceeds that of the methyl radical, but the individual free radical species in petroleum are short-lived and reactive. It is the macroscopic average free radical concentration in petroleum that persisted over geological time, not that of individual free radical species.
(b) The persistent free radical concentration in petroleum can be explained in terms of a dynamic reaction equilibrium of free radical dissociation and association that causes a finite number of species at any given time to be present as free radicals. Evidence to support this description are the change in free radical concentration related to changes in the Gibbs free energy of the system when the bulk liquid properties are changed, as well as the responsiveness of free radical concentration to dynamic change in temperature.
(c) Cage effects, solvent effects, steric protection, and radical stabilization affect the reaction rate of free radicals in petroleum, but these phenomena are not responsible for the macroscopically observed persistent free radical concentration in petroleum.
(d) The main difference between persistent free radicals in straight-run petroleum and those in converted petroleum is that the former system has been equilibrated over geologic time, whereas the latter is not equilibrated. The free radical concentration in straight-run petroleum at constant conditions is time invariant, whereas the free radical concentration in converted petroleum can change over time, even at ambient conditions.
(e) The persistent free radical concentration in petroleum distillation fractions increases with an increase in atmospheric equivalent boiling point of the distillation fraction. Based on the limited evidence presented, it appears that persistent free radical content in straight-run petroleum is a property that is part of the compositional continuum postulated by Altgelt and Boduszynski.
(f) Separation of petroleum into solubility fractions causes a partitioning of the free radical species. The asphaltenes (n-alkane insoluble material) fraction of whole petroleum has a higher concentration of persistent free radicals, and the maltenes (n-alkane soluble material) fraction has a lower concentration of persistent free radicals than whole petroleum. As expected, the free radicals in the solubility fractions have the properties of the free radicals in the whole petroleum, which includes responsiveness to changes in solvent environment and dynamic changes in temperature.
(g) Attempts to relate persistent free radical concentration to petroleum composition were inconclusive.

Author Contributions

The authors contributed equally to this work. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

This research made use only of previously published data and the origin and experimental details can be found in the cited literature noted in this study.

Acknowledgments

We would like to acknowledge people that unwittingly served as inspiration for the research presented here. Gareth and Sandra Eaton, for their work on quantitative electron spin resonance spectrometry, and Dicho Stratiev, for sharing a wealth of industrial data and insights on thermal conversion processes with the research community.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Tannous, J.H.; De Klerk, A. Quantification of the free radical content of oilsands bitumen fractions. Energy Fuels 2019, 33, 7083–7093. [Google Scholar] [CrossRef]
  2. Chang, H.-L.; Wong, G.K.; Lin, J.-R.; Yen, T.F. Electron spin resonance study of bituminous substances and asphaltenes. In Asphaltenes and Asphalts; Yen, T.F., Chilingarian, G.V., Eds.; Developments in Petroleum Science Volume 40 B; Elsevier: Amsterdam, The Netherlands, 2000; Volume 2, pp. 229–280. [Google Scholar]
  3. Raseev, S. Thermal and Catalytic Processes in Petroleum Refining; Marcel Dekker: New York, NY, USA, 2003. [Google Scholar]
  4. Griller, D.; Ingold, K.U. Persistent carbon-centered radicals. Acc. Chem. Res. 1976, 9, 13–19. [Google Scholar] [CrossRef]
  5. Forrester, A.R.; Hay, J.M.; Thomson, R.H. Organic Chemistry of Stable Free Radicals; Academic Press: London, UK, 1968. [Google Scholar]
  6. Parsons, A.F. An Introduction to Free Radical Chemistry; Blackwell Science: Oxford, UK, 2000. [Google Scholar]
  7. Alili, A.S.; Siddiquee, M.N.; De Klerk, A. Origin of free radical persistence in asphaltenes: Cage effect and steric protection. Energy Fuels 2020, 34, 348–359. [Google Scholar] [CrossRef]
  8. Styles, Y.; De Klerk, A. Sodium conversion of oilsands bitumen-derived asphaltenes. Energy Fuels 2016, 30, 5214–5222. [Google Scholar] [CrossRef]
  9. Naghizada, N.; Prado, G.H.C.; De Klerk, A. Uncatalyzed hydrogen transfer during 100–250 °C conversion of asphaltenes. Energy Fuels 2017, 31, 6800–6811. [Google Scholar] [CrossRef] [Green Version]
  10. Petrakis, L.; Grandy, D.W. An electron spin resonance spectrometric investigation of natural, extracted and thermally altered kerogenous materials. Geochim. Cosmochim. Acta 1980, 44, 763–768. [Google Scholar] [CrossRef]
  11. Lorand, J.P. The cage effect. In Inorganic Reaction Mechanisms, Part II; Progress in Inorganic Chemistry; Edwards, J.O., Ed.; Wiley and Sons, Inc.: New York, NY, USA, 1972; Volume 17, pp. 207–325. [Google Scholar]
  12. Koenig, T.; Fischer, H. “Cage” effects. In Free Radicals; Kochi, J.K., Ed.; Wiley and Sons, Inc.: New York, NY, USA, 1973; Volume 1, pp. 157–189. [Google Scholar]
  13. Barry, J.T.; Berg, D.J.; Tyler, D.R. Radical cage effects: Comparison of solvent bulk viscosity and microviscosity in predicting the recombination efficiencies of radical cage pairs. J. Am. Chem. Soc. 2016, 138, 9389–9392. [Google Scholar] [CrossRef]
  14. Barry, J.T.; Berg, D.J.; Tyler, D.R. Radical cage effects: The prediction of radical cage pair recombination efficiencies using microviscosity across a range of solvent types. J. Am. Chem. Soc. 2017, 139, 14399–14405. [Google Scholar] [CrossRef]
  15. Li, X.; Ogihara, T.; Abe, M.; Nakamura, Y.; Yamago, S. The effect of viscosity on the diffusion and termination reaction of organic radical pairs. Chem.-Eur. J. 2019, 25, 9846–9850. [Google Scholar] [CrossRef]
  16. Reichardt, C. Solvents and Solvent Effects in Organic Chemistry, 3rd ed.; Wiley-VCH: Weinheim, Germany, 2003. [Google Scholar]
  17. Mujica, V.; Nieto, P.; Puerta, L.; Acevedo, S. Caging of molecules by asphaltenes. A model for free radical preservation in crude oils. Energy Fuels 2000, 14, 632–639. [Google Scholar] [CrossRef]
  18. Acevedo, S.; Escobar, G.; Ranaudo, M.A.; Piñate, J.; Amorín, A. Observations about the structure and dispersion of petroleum asphaltenes aggregates obtained from dialysis fractionation and characterization. Energy Fuels 1997, 11, 774–778. [Google Scholar] [CrossRef]
  19. Zhang, Y.; Siskin, M.; Gray, M.R.; Walters, C.C.; Rodgers, R.P. Mechanisms of asphaltene aggregation: Puzzles and a new hypothesis. Energy Fuels 2020, 34, 9094–9107. [Google Scholar] [CrossRef]
  20. Gray, M.R.; Yarranton, H.W.; Chacón-Patiño, M.L.; Rodgers, R.P.; Bouyssiere, B.; Giusti, P. Distributed properties of asphaltene nanoaggregates in crude oils: A review. Energy Fuels 2021, 35, 18078–18103. [Google Scholar] [CrossRef]
  21. Rüchardt, C. Relations between structure and reactivity in free-radical chemistry. Angew. Chem. Int. Ed. Engl. 1970, 9, 830–843. [Google Scholar] [CrossRef]
  22. Stella, L.; Janousek, Z.; Merényi, R.; Viehe, H.G. Stabilization of radicals by “capto-dative” substitution –C–C addition to radicophilic olefins. Angew. Chem. Int. Ed. Engl. 1978, 17, 691–692. [Google Scholar] [CrossRef]
  23. Viehe, H.G.; Janousek, Z.; Merényi, R.; Stella, L. The captodative effect. Acc. Chem. Res. 1985, 18, 148–154. [Google Scholar] [CrossRef]
  24. Huyser, E.S. Solvent effects in free-radical reactions. In Advances in Free Radical Chemistry; Williams, G.H., Ed.; Academic Press: London, UK, 1965; Volume 1, pp. 77–153. [Google Scholar]
  25. Gendell, J.; Freed, J.H.; Fraenkel, G.K. Solvent effects in electron spin resonance. Chem. Phys. 1962, 37, 2832–2841. [Google Scholar] [CrossRef]
  26. Isenberg, I.; Baird, S.L., Jr. Solvent effects in radical ion formation. J. Am. Chem. Soc. 1962, 84, 3803–3805. [Google Scholar] [CrossRef]
  27. Eaton, G.R.; Eaton, S.S.; Barr, D.P.; Weber, R.T. Quantitative EPR; Springer: New York, NY, USA, 2010. [Google Scholar]
  28. Russell, G.A. Solvent effects in the reactions of free radicals and atoms. J. Am. Chem. Soc. 1957, 79, 2977–2978. [Google Scholar] [CrossRef]
  29. Khulbe, K.C.; Mann, R.S.; Lu, B.C.-Y.; Lamarche, G.; Lamarche, A.-M. Effects of solvents on free radicals of bitumen and asphaltenes. Fuel Process. Technol. 1992, 32, 133–141. [Google Scholar] [CrossRef]
  30. Tannous, J.H.; Tulegenova, D.; De Klerk, A. Effect of solvents on persistent free radical content in the absence of reactions. Energy Fuels 2022, 36, 5253–5266. [Google Scholar] [CrossRef]
  31. Dalal, D.P.; Eaton, S.S.; Eaton, G.R. The effects of lossy solvents on quantitative EPR studies. J. Magn. Reson. 1981, 44, 415–428. [Google Scholar] [CrossRef]
  32. Elofson, R.M.; Schulz, K.F.; Hitchon, B. Geochemical significance of chemical composition and ESR properties of asphaltenes in crude oils from Alberta, Canada. Geochim. Cosmochim. Acta 1977, 41, 567–580. [Google Scholar] [CrossRef]
  33. Hernández, M.S.; Coll, D.S.; Silva, P.J. Temperature dependence of the electron paramagnetic resonance spectrum of asphaltenes from Venezuelan crude oils and their vacuum residues. Energy Fuels 2019, 33, 990–997. [Google Scholar] [CrossRef]
  34. Malhotra, V.M.; Graham, W.R.M. Characterization of P.R. Spring (Utah) tar sand bitumen by the EPR technique: Free radicals. Fuel 1983, 62, 1255–1264. [Google Scholar] [CrossRef]
  35. Yokono, T.; Obara, T.; Sanada, Y.; Shimomura, S.; Imamura, T. Characterization of Carbonization Reaction of Petroleum Residues by Means of High-Temperature ESR and Transferable Hydrogen. Carbon 1986, 24, 29–32. [Google Scholar] [CrossRef]
  36. Dappe, V.; Ben Tayeb, K.; Vezin, H.; Mariette, S.; Serve, O.; Livadaris, V. Effect of Thermal Treatment of Different Petroleum Fractions: Characterization by in situ EPR Spectroscopy. Energy Fuels 2020, 34, 12026–12032. [Google Scholar] [CrossRef]
  37. Rüchardt, C.; Gerst, M.; Ebenhoch, J. Uncatalyzed transfer hydrogenation and transfer hydrogenolysis: Two novel types of hydrogen-transfer reactions. Angew. Chem. Int. Ed. 1997, 36, 1406–1430. [Google Scholar] [CrossRef]
  38. Tannous, J.H.; De Klerk, A. Methyl and hydrogen transfer in free radical reactions. Energy Fuels 2020, 34, 1698–1709. [Google Scholar] [CrossRef]
  39. Payan, F.; De Klerk, A. Hydrogen transfer in asphaltenes and bitumen at 250 °C. Energy Fuels 2018, 32, 9340–9348. [Google Scholar] [CrossRef]
  40. Yañez Jaramillo, L.M.; de Klerk, A. Partial upgrading of bitumen by thermal conversion at 150–300 °C. Energy Fuels 2018, 32, 3299–3311. [Google Scholar] [CrossRef]
  41. Jha, K.N.; Montgomery, D.S.; Strausz, O.P. Chemical composition of gases in Alberta bitumens and in low-temperature thermolysis of oil sand asphaltenes and maltenes. In Oil Sand and Oil Shale Chemistry; Strausz, O.P., Lown, E.M., Eds.; Verlag Chemie: New York, NY, USA, 1978; pp. 33–54. [Google Scholar]
  42. Yan, Y.; Prado, G.H.C.; De Klerk, A. Storage stability of products from visbreaking of oilsands bitumen. Energy Fuels 2020, 34, 9585–9598. [Google Scholar] [CrossRef]
  43. Ellis, C.; Wells, A.A.; Heyroth, F.F. The Chemical Action of Ultraviolet Rays; Reinhold: New York, NY, USA, 1941. [Google Scholar]
  44. Altgelt, K.H.; Boduszynski, M.M. Composition and Analysis of Heavy Petroleum Fractions; Marcel Dekker: New York, NY, USA, 1994. [Google Scholar]
  45. Tannous, J.H.; De Klerk, A. Asphaltenes formation during thermal conversion of deasphalted oil. Fuel 2019, 255, 115786. [Google Scholar] [CrossRef]
  46. Rudnick, L.R.; Tueting, D.R. Variation of free radical concentration with molecular weight in petroleum feedstocks. Fuel Sci. Technol. Int. 1989, 7, 57–68. [Google Scholar] [CrossRef]
  47. Podgorski, D.C.; Corilo, Y.E.; Nyadong, L.; Lobodin, V.V.; Bythell, B.J.; Robbins, W.K.; McKenna, A.M.; Marshall, A.G.; Rodgers, R.P. Heavy petroleum composition. 5. Compositional and structural continuum of petroleum revealed. Energy Fuels 2013, 27, 1268–1276. [Google Scholar] [CrossRef]
  48. Yen, T.F.; Erdman, J.G.; Saraceno, A.J. Investigation of the nature of free radicals in petroleum asphaltenes and related substances by Electron Spin Resonance. Anal. Chem. 1962, 34, 694–700. [Google Scholar] [CrossRef]
  49. Yen, T.F.; Sprang, S.R. ESR g-values of bituminous materials. Prepr. Pap.-Am. Chem. Soc. Div. Petrol. Chem. 1970, 15, A65–A76. [Google Scholar]
  50. Yen, T.F.; Sprang, S.R. Contribution of E.S.R. analysis toward diagenic mechanisms in bituminous deposits. Geochim. Cosmochim. Acta 1977, 41, 1007–1018. [Google Scholar] [CrossRef]
  51. Retcofsky, H.L.; Hough, M.R.; Maguire, M.M.; Clarkson, R.B. Nature of the free radicals in coals, pyrolyzed coals, solvent-refined coal, and coal liquefaction products. In Coal Structure; Gorbaty, M.L., Ouchi, K., Eds.; Advances in Chemistry; American Chemical Society: Washington, DC, USA, 1981; Volume 192, pp. 37–58. [Google Scholar]
  52. Malhotra, V.M.; Buckmaster, H.A. 9 and 34 GHz EPR Study of the free radicals in various asphaltenes: Statistical correlation of the g-values with heteroatom content. Org. Geochem. 1985, 8, 235–239. [Google Scholar] [CrossRef]
  53. Kaiser, M.J.; De Klerk, A.; Gary, J.H.; Handwerk, G.E. Petroleum Refining. Technology, Economics, and Markets, 6th ed.; CRC Press: Boca Raton, FL, USA, 2020. [Google Scholar]
  54. Wang, L.; Zachariah, A.; Yang, S.; Prasad, V.; De Klerk, A. Visbreaking oilsands-derived bitumen in the temperature range of 340–400 °C. Energy Fuels 2014, 28, 5014–5022. [Google Scholar] [CrossRef]
  55. Yañez Jaramillo, L.M.; De Klerk, A. Is solubility classification a meaningful measure in thermal conversion? Energy Fuels 2022, 36, 8649–8662. [Google Scholar] [CrossRef]
  56. Gray, M.R. Upgrading Oilsands Bitumen and Heavy Oil; University of Alberta Press: Edmonton, AB, Canada, 2015. [Google Scholar]
  57. Ramirez-Corredores, M.M. The Science and Technology of Unconventional Oils; Academic Press: London, UK, 2017; pp. 41–222. [Google Scholar]
  58. ASTM D2007; Standard Test Method for Characteristic Groups in Rubber Extender and Processing Oils and Other Petroleum-derived Oils by the Clay-Gel Absorption Chromatographic Method. ASTM International: West Conshohocken, PA, USA, 2020.
  59. ASTM D4124; Standard Test Method for Separation of Asphalt into Four Fractions. ASTM International: West Conshohocken, PA, USA, 2018.
  60. ASTM D6560; Standard Test Method for Determination of Asphaltenes (Heptane Insolubles) in Crude Petroleum and Petroleum Products. ASTM International: West Conshohocken, PA, USA, 2022.
  61. Niizuma, S.; Steele, C.T.; Gunning, H.E.; Strausz, O.P. Electron spin resonance study of free radicals in Athabasca asphaltene. Fuel 1977, 56, 249–256. [Google Scholar] [CrossRef]
  62. Adams, J.Q.; Altgelt, K.H.; LeTourneau, R.L.; Lindeman, L.P. Free radical concentrations in gel permeation fractions of asphaltenes from different crude oils. Prepr. Pap.-Am. Chem. Soc. Div. Petrol. Chem. 1966, 12, B140–B144. [Google Scholar]
  63. Andersen, S.I. Effect of precipitation temperature on the composition of n-heptane asphaltenes. Fuel Sci. Technol. Int. 1994, 12, 51–74. [Google Scholar] [CrossRef]
  64. Schultz, K.F.; Selucky, M.L. ESR measurements on asphaltene and resin fractions from various separation methods. Fuel 1981, 60, 951–956. [Google Scholar] [CrossRef]
Processes 11 02067 g001 550
Figure 1. Schematic representation of solvent molecules surrounding a newly formed radical pair that is referred to as the solvent cage.
Figure 1. Schematic representation of solvent molecules surrounding a newly formed radical pair that is referred to as the solvent cage.
Processes 11 02067 g001
Processes 11 02067 g002 550
Figure 2. Conversion of different probe reagents at 250 °C for 1 h with industrially produced n-pentane insoluble asphaltenes from Athabasca bitumen with a persistent free radical content of 2.1 × 1018 spins/g [7].
Figure 2. Conversion of different probe reagents at 250 °C for 1 h with industrially produced n-pentane insoluble asphaltenes from Athabasca bitumen with a persistent free radical content of 2.1 × 1018 spins/g [7].
Processes 11 02067 g002
Processes 11 02067 g003 550
Figure 3. Dynamic reaction equilibrium of free radical pairs as explanation of both longevity and reactivity of persistent free radicals in petroleum.
Figure 3. Dynamic reaction equilibrium of free radical pairs as explanation of both longevity and reactivity of persistent free radicals in petroleum.
Processes 11 02067 g003
Processes 11 02067 g004 550
Figure 4. Examples of condensed aromatic structures with an odd number of carbons that could potentially exist as a free radical (as shown) in dynamic reaction equilibrium with another free radical to give rise to a finite persistent free radical concentration in petroleum.
Figure 4. Examples of condensed aromatic structures with an odd number of carbons that could potentially exist as a free radical (as shown) in dynamic reaction equilibrium with another free radical to give rise to a finite persistent free radical concentration in petroleum.
Processes 11 02067 g004
Processes 11 02067 g005 550
Figure 5. Relationship between persistent free radical concentration and average molecular weight of two straight-run oils, redrawn using the data from Rudnick and Tueting [46].
Figure 5. Relationship between persistent free radical concentration and average molecular weight of two straight-run oils, redrawn using the data from Rudnick and Tueting [46].
Processes 11 02067 g005
Processes 11 02067 g006 550
Figure 6. Relationship between persistent free radical concentration and average molecular weight of different asphaltenes, redrawn using the data from Adams et al. [62].
Figure 6. Relationship between persistent free radical concentration and average molecular weight of different asphaltenes, redrawn using the data from Adams et al. [62].
Processes 11 02067 g006
Table
Table 1. Persistent free radical content in petroleum and coal [1,2].
Table 1. Persistent free radical content in petroleum and coal [1,2].
MaterialFree Radical ContentReference
(spins/g)(µmol/g)
Athabasca bitumen0.9–1.1 × 10181.5–1.8[1]
Cold Lake bitumen0.9–1.1 × 10181.5–1.8[1]
Coal (low rank) a2.5–5.3 × 10184.2–8.8[2]
Coal (high rank) b7.2–25 × 101812–42[2]
a Carbon content range 77–85%; b Carbon content range 89–97%.
Table
Table 2. Change in free radical concentration in heavy gas oil (HGO) when ESR measurements were performed at different concentrations of heavy gas oil in toluene [1].
Table 2. Change in free radical concentration in heavy gas oil (HGO) when ESR measurements were performed at different concentrations of heavy gas oil in toluene [1].
HGO Concentration (wt%)Free Radical Concentration
(spins/g HGO)(µmol/g HGO)
4.68.7 × 10171.4
114.0 × 10170.67
202.3 × 10170.38
222.0 × 10170.34
371.4 × 10170.23
491.1 × 10170.18
Table
Table 3. Change in free radical concentration in bitumen when ESR measurements were performed at the same concentration of bitumen, but different solvents [30].
Table 3. Change in free radical concentration in bitumen when ESR measurements were performed at the same concentration of bitumen, but different solvents [30].
SolventFree Radical Concentration
(spins/g Bitumen)(µmol/g Bitumen)
Carbon disulfide1.44 × 10182.4
2,3-Benzofuran0.99 × 10181.6
Cumene0.92 × 10181.5
Toluene0.76 × 10181.3
Tetrahydrothiophene0.76 × 10181.3
Diphenyl sulfide0.63 × 10181.0
Table
Table 4. Changes in converted petroleum during storage at ambient conditions under inert atmosphere [42].
Table 4. Changes in converted petroleum during storage at ambient conditions under inert atmosphere [42].
Storage Time (Weeks)Free Radical ConcentrationRefractive Index at 20 °CDensity at 20 °C (kg/m3)
(spins/g)(µmol/g)
01.9 × 10183.11.5749986.6
81.6 × 10182.71.5774997.1
201.3 × 10182.11.58091001.1
Table
Table 5. Persistent free radical concentration in straight-run distillation fractions of Athabasca bitumen [1].
Table 5. Persistent free radical concentration in straight-run distillation fractions of Athabasca bitumen [1].
Distillation FractionFree Radical Concentration
(spins/g)(µmol/g)
Light vacuum gas oil7.8 × 10171.3
Heavy vacuum gas oil8.7 × 10171.4
Vacuum residue1.4 × 10182.3
Table
Table 6. Comparison of different straight-run oil sands bitumen samples in terms of Watson K-factor (KW), mean boiling temperature (T50), and persistent free radical concentration [1,54,55].
Table 6. Comparison of different straight-run oil sands bitumen samples in terms of Watson K-factor (KW), mean boiling temperature (T50), and persistent free radical concentration [1,54,55].
SampleKWT50Free Radical ConcentrationReference
(°R)(°C)(spins/g)(µmol/g)
Cold Lake bitumen11.215305779.5 × 10171.6[1,54]
Athabasca bitumen11.214895541.6 × 10182.7[55]
Table
Table 7. Partitioning of persistent free radicals in Athabasca bitumen from Nexen Long Lake between the maltenes (n-heptane soluble) and asphaltenes (n-heptane insoluble) fractions [55].
Table 7. Partitioning of persistent free radicals in Athabasca bitumen from Nexen Long Lake between the maltenes (n-heptane soluble) and asphaltenes (n-heptane insoluble) fractions [55].
SampleFree Radical Concentration
(spins/g)(µmol/g)
Whole petroleum1.6 × 10182.7
Maltenes fraction1.1 × 10181.8
Asphaltenes fraction3.4 × 10185.6
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Yañez Jaramillo, L.M.; Tannous, J.H.; de Klerk, A. Persistent Free Radicals in Petroleum. Processes 2023, 11, 2067. https://doi.org/10.3390/pr11072067

AMA Style

Yañez Jaramillo LM, Tannous JH, de Klerk A. Persistent Free Radicals in Petroleum. Processes. 2023; 11(7):2067. https://doi.org/10.3390/pr11072067

Chicago/Turabian Style

Yañez Jaramillo, Lina M., Joy H. Tannous, and Arno de Klerk. 2023. "Persistent Free Radicals in Petroleum" Processes 11, no. 7: 2067. https://doi.org/10.3390/pr11072067

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop