Sample age effect on parameters of dynamic nuclear polarization in certain difluorobenzen isomers/MC800 asphaltene suspensions

1 Introduction

Asphaltene is the heaviest and most viscous aromatic component of crude oil. It is also the most polar component, being soluble in aromatic solvents such as toluene and xylene but insoluble in normal aliphatic hydrocarbons such as n-heptane and n-hexane [1,2,3]. Asphaltene precipitates only when unstable. Its stability is affected by different parameters, including the pressure, temperature, and characteristics of crude oil mixtures [4,5].

Asphaltenes, are a part of petroleum, are considered to be polar species, have greater aromatic complexes, and consist of heteroatoms (e.g., O, S, and N), alkyl chains, and certain metals [6]. Asphaltenes have unpaired electrons, which are observed in crude petroleum by using the electron spin resonance (ESR) experiments [7].

Dynamic nuclear polarization (DNP) uses liquid state interactions between nucleus and unpaired electron spins to produce enhanced nuclear polarization after the saturation of ESR radical transitions. The maximum DNP enhancement rate of the nuclear magnetic resonance (NMR) signal is calculated by the electron-to-nuclear gyromagnetic ratio, γe/γn. For ¹⁹F spin, this ratio is 700.

It becomes apparent that studies of the coupling between I, nuclear spin reservoir, and S, electronic spin reservoir, may put forward an insight into the dynamics and structure of these sophisticated materials. The magnetic interaction, which is the source of DNP, between the nucleus and unpaired electron, is much stronger than the inter-nuclear interaction experienced in brood solvent region. The intensities and spectral densities of the intermolecular scalar could arise from direct contact between the solvent nuclei and unpaired electrons of the radical or from a temporary overlap of the delocalized molecular orbitals during the diffusion process [8,9,10]. The overhauser dynamic nuclear polarization type mechanism (ODNP) for the hyperpolarization should be the most efficient as crude oil is historically treated as a viscous liquid [11]. As the comparatively high mobility of oil molecules, including radicals, requires an appropriate regulation of electron–nuclear interaction, this condition leads to an efficient propagation of polarization between nuclei and electrons in oil. The main aspect of ODNP is that when the microwave pumping frequency is in precisely the same resonance as the EPR transition, the full effect of electron polarization conversion on the nuclear system is obtained. In fact, as early as the 1950s, Poindexter [12,13] recorded Overhauser form of DNP in the external magnetic field B ₀ = 18 G on protons of “light” or “medium” crude oils that have a viscosity of 25–40 mPa s.

The nuclear spins are the ¹⁹F spins in this sample, and they are the ¹⁹F center of the solvent medium, whereas the electron spins are the free electrons on the asphaltene micelles. Unpaired electrons have been dislocated to the unfilled carbon bonds of asphaltene particles compact aromatic structure [14]. We revealed a dipolar interaction between the solvent fluorine atoms and unpaired colloidal asphaltene electrons through the DNP process. In a low magnetic field, electron spin saturation, as well as DNP improvements, were achieved, and also DNP parameters have been determined.

This study aimed to investigate the DNP parameters (A _∞, U _∞, ρ, f, s, and K) of difluorobenzene isomer solvents in MC800 asphaltene free radical against time at room temperature at 1.53 mT low magnetic field.

2 Theory

The whole DNP phenomenon theory through the Overhauser effect in free radicals solutions has already been published in the technical literature [15,16,17,18,19,20]; thus, in this article, we present just a short overview of the theory that applies to our study. In Figure 1, a schematic display of a 1/2-spin and a two-spin electron system energy level is displayed. In this figure, m _S and m _I are, respectively, electron and nucleus magnetic quantum numbers.

Figure 1

Diagram showing the energy level of a nucleus that has a 1/2-spin and two-spin electron system. m _I and m _S denote the magnetic quantum numbers of nucleus and electron, respectively. ω _i and w _i (i = 0, 1, 2) represent the transition frequencies and probabilities. Here ω _S/ω _I equals to 700 for ¹⁹F nucleus.

The possibilities of transition through relaxation may be determined in this system as follows (Table 1).

Here, w 1 = w 1 ′ + w 1 ″ ⋅ 2 w 1 ′ corresponds to the transition probability owing to spin coupling of nuclear electron, and 2 w 1 ″ denotes the probability of transition because of various processes of relaxation.

An expression for the motion equation of nuclear polarization (NP) is given as [21].

Π z and P _z indicate the unpaired electron spins and the NP of the nuclear spins. P ₀ and Π 0 are the thermal equilibrium polarizations.

T 1 − 1 = w 0 + 2 w 1 + w 2 correspond to the overall rate of relaxation of the nuclear spin-lattice, and f = ( w 0 + 2 w ′ 1 + w 2 ) / [ w 0 + 2 ( w ′ 1 + w ″ 1 ) + w 2 ] stands for the leakage factor, which reveals the degree of efficiency that electron spins relax nuclear spins. The above-mentioned leakage factor is ranging from 0 (no nuclear–electron coupling relaxation) and 1 (there is not any other relaxation mechanisms). ρ = ( w 2 – w 0 ) / ( w 0 + 2 w ′ 1 + w 2 ) stands for the coupling factor of nuclear–electron that relies on the type of nuclear–electron interactivity. Here, ρ may range between +0.5 that corresponds to purely dipolar interactivities and −1 that corresponds to purely scalar interactivities. Finally, s = (Π₀ – Π_z)/Π₀ represents the saturation factor equivalent to 1 (0 ≤ s ≤ 1) in full EPR saturation.

At the steady state, namely, dP _z/dt = 0, the formulation of Overhauser enhancement factor is as follows:

Here, γ I and γ S correspond to the gyromagnetic ratios of nucleus and electron in turn ( γ I > 0 ; γ S < 0 ).

In the event that the saturation state is met, the reciprocal of the enhancement factor is demonstrated below which is extrapolated for infinite ESR power,

There, f, leakage factor can be formulated as follows:

Here, T 1 ' − 1 = w 0 + 2 w 1 ′ + w 2 denotes the rate of nuclear relaxation as a result of coupling of nuclear–electron, and T 10 − 1 = T 1 ' − 1 = 2 w ″ 1 is the rate of nuclear relaxation caused by other processes. The leakage factor reciprocal is denoted as follows:

T 1 ' − 1 is proportionate to c, the concentration of electron spin (or free radical), alternatively. In this way, equation (3) puts forward

Here, the ultimate electron spin concentration is as follows:

Since, ∣ γ S / γ I ∣ = + 700 for ¹⁹F, equation (7) presents

In the white spectral region, namely in extreme narrowing case ( ω S t t ≪ 1 ). In addition, an earlier study [22] denotes the K parameter that describes the relative significance of both scalar and translational dipolar interactivities,

Here, J 1 D ( 0 ) and J Sc ( 0 ) equal to the translational and scalar dipolar spectral density functions. Finally, K is computed by using the experimental ρ value at high temperatures and in weak fields [19,23,24] as follows:

3 Methods

3.1 Solvents and asphaltene

MC-800 liquid asphalt is used to obtain the asphaltene. The asphalt is dissolved in a tenfold excess volume of benzene and precipitated in a further tenfold excess volume of petroleum ether at a boiling temperature interval of 40 and 60°C (namely 1-volume asphalt, 10-volume benzene, and 100-volume petroleum ether) [25,26]. Filtration was used to collect the precipitate, which was then dried. The resultant was a black semisolid asphaltene. Asphaltene made up 14% of the weight of MC-800 liquid asphalt.

The solvents were obtained from Fluka (Switzerland) and Aldrich Chem. Co. (USA) and utilized exactly as they were supplied. The purities, viscosities, molecular weights, boiling temperatures, freezing temperatures, densities, dipole moments, and their computed NMR sensitivities of the chosen solvents are reported in Table 2. Figure 2 depicts the molecular structures of these solvents.

Table 2

NMR sensitivities and basic constants of the solvents chosen

Solvent	Molecular weight (g/mol)	Melting point (°C)	Boiling point (°C)	Density (g/cm3)	Viscosity (cP)	Dipole moment (D)	NMR sensitivity (×1022 spin/cm3)
o-Difluorobenzene (C6H4F2)	114.09	−34	92	1.158	0.548	2.40	1.22
m-Difluorobenzene (C6H4F2)	114.09	−59	83	1.163	0.603	1.50	1.23
p-Difluorobenzene (C6H4F2)	114.09	−13	88–89	1.514	0.608	0	1.24

Figure 2

Structure of 1,2 difluorobenzene (ortho), 1,3 difluorobenzene (meta), and 1,4 difluorobenzene (para) molecules (green color “fluorine,” blue color “hydrogen,” and gray color “carbon”).

3.2 Preparation of the samples

The samples had been produced at three distinct degrees of radical concentration. The samples were sealed Pyrex tubes (18 mm diameter and resistant to liquid nitrogen temperature) after being degassed by means of minimum five freeze–pump–thaw cycles that uses liquid nitrogen at about 10⁻⁵ Pa in a Leybold–Heraeus vacuum system. Figure 3 demonstrates how the samples were degassed and sealed in Pyrex tubes. The degassing process is required because the presence of oxygen in the sample prevents the EPR transition from being easily stimulated. To the naked eye, the solutions look substantially identical, as there is no apparent indication of fast flocculation.

Figure 3

The sample is degassed and sealed in Pyrex tubes.

3.4 Setting the DNP Parameters

To calculate the A _∞ values, the reciprocal of the enhancement factor (namely [(P _z – P ₀)/P ₀]⁻¹) should be derived as a function of the ESR power’s reciprocal value. In this context, ESR power is proportionate to either H 1 e − 2 or V eff − 2 , where V _eff value represents the high frequency voltage on the ESR coil. For complete ESR saturation, Π _z value will be zero, and the saturation factor (s) will come close to 1. The ESR power needs to reach ∞, or V eff − 2 to 0, in order to achieve complete saturation. Therefore, the A ∞ − 1 value in the linear graph is equal to the junction of the extrapolated best-fit line and V eff − 2 = 0 line. Figure 4 depicts variation of –[(P _z – P ₀)/P ₀]⁻¹ vs V eff − 2 for three concentrations of asphaltene/fluorobenzene samples. The A ∞ − 1 values are the intersections of the V eff − 2 = 0 line and the best suited extrapolated line. Figure 4 demonstrates the R ² values to assess the consistency between the experimental points and the linear fit function. A maximum inaccuracy of approximately ±10% is found for a single point in the graphic.

Figure 4

Variation of ( P z − P 0 ) P 0 − 1 versus V eff − 2 for asphaltene/fluorobenzene derivative samples at three distinct concentration levels. Here, A ∞ − 1 values equivalent to the junction points of the extrapolated best-fit and the V eff − 2 = 0 lines. (The red lines represent 1,4-difluorobenzene, the blue lines represent 1,3-difluorobenzene, and the green lines represent 1,2-difluorobenzene.) (a) Ref. [30] Data and (b) 2021 data.

In Figure 5, as A ∞ − 1 values proportionate to the reciprocal concentration of asphaltene c ⁻¹ , the ultimate enhancement factor U _∞ may be calculated simply. The U ∞ − 1 values are the junction points on the best-fit extrapolated lines and c ⁻¹ = 0 line (f = 1, ∞ concentration).

Figure 5

Instances of the evaluations for ultimate enhancement factor at room temperature. Here, the U ∞ − 1 values are equivalent to the junction points of extrapolated best-fit lines and the c ⁻¹ = 0 line (ref. [30] 2014 and 2021 data).

After U _∞ is determined, the nuclear–electron coupling parameter value ρ may be derived by using equation (8). The A ∞ / U ∞ ratio that is calculated from equations (3) and (7) is used to achieve f value for each of the concentrations. Finally, by using the equations (2) and (3), the s value was attained from the enhancement factor A _end for the maximum existing ESR power divided by A ∞ .

4 Results and discussion

5 Conclusion

The following results are obtained as a result of the experimental study that is conducted in line with the aim of this study.

To study colloidal suspensions solvent mediums of MC800 asphaltene consisting of 1,2 difluorobenzene, 1,3 difluorobenzene, and 1,4 difluorobenzene. Overhauser affect type DNP method is utilized at room temperature in a 1.53 mT low-level magnetic field. For testing environment, at each solvent medium sample, three different concentrations of asphaltene was used. Consequently, there were nine samples in total. In the EPR spectrum of the asphaltenes, the spectrum of MC800 asphaltene in the 1,3 difluorobenzene solvent medium was discovered to be a single Gaussian peaking about 45.5 MHz. The pattern of the EPR spectrum agrees well with findings of Kirimli’s study (2014) [30].

In the study, ρ value, namely the nuclear–electron coupling parameter, was observed to be lower after samples that had produced a colloidal 1,2 difluorobenzene, 1,3 difluorobenzene, and 1,4 difluorobenzene solvent media of MC 800 asphaltene for 7 years. The nuclear–electron coupling parameter ρ ranges between pure dipolar value of +0.5 and pure scalar value of −1.0 in low magnetic fields, and the new ρ values discovered to be within these limits. In our experimentation, new ρ values are found to be positive, which suggests that the dipolar portion in spin–spin interactivities is prevalent. K value depicts the significance of almost all the samples according to the scalar coupling parameter. In other words, in awaited samples, the scalar section of the intermolecular spin–spin interactivities rises, and the dipolar section of the interactivities diminishes. For the 1,4 difluorobenzene solution medium, the lowest absolute U∞ value (thus, ρ value as well) has been achieved. This suggests that the asphaltene has been much more efficiently dispersed and suspended by 1,2-difluorobenzene and 1,3-difluorobenzene. In other respects, in terms of electrical dipole moment value, the 1,4 difluorobenzene has the lowest value.

We think that in biological samples, comprehensive information about the chemical environment may be acquired using free radicals and aromatic fluorinated solvents that are predisposed to the scalar interactivity in EPR oximetry, which is promoted through the Overhauser effect [31,35].

This research makes a positive contribution to the study of the asphaltene surface and aids related research with DNP findings. Furthermore, DNP investigations are helpful applications for the empirical characterization of asphaltene solutions.

Figure 6

The EPR spectrum acquired in the study (▯) and Kirimli’s EPR spectrum (2014) (Δ) [30].