Abstract

The FT-IR spectra of the stable C76 and C84 isomers of D2 symmetry, isolated by the new, advanced extraction and chromatographic methods and processes, were recorded by the KBr technique, over the relevant region from 400 to 2000 cm−1, at room temperature. All the observed infrared bands are in excellent agreement with the semiempirical QCFF/PI, DFT, and TB potential calculations for these fullerenes, which is presented in this article, as the evidence of their validity. The molar absorptivity ε and the integrated molar absorptivity ψ of their IR absorption bands were determined and reported together with the relative intensities. Excellent agreement is found between the relative intensities of the main and characteristic absorption maxima calculated from and from the values in adequate integration ranges. These results are significant for the identification and quantitative determination of the C76-D2 and C84-D2:22 fullerenes, either in natural resources on Earth and in space or in artificially synthesized and biomaterials, electronic, optical, and biomedical devices, sensors, polymers, optical limiters, solar cells, organic field effect transistors, special lenses, diagnostic and therapeutic agents, pharmaceutical substances in biomedical engineering, and so forth.

1. Introduction

Fullerenes C60 and C70 were detected in a series of astrophysical objects and space environments [16], such as certain planetary [7, 8] and protoplanetary [9] nebulae, postasymptotic giant branch stars, young stellar objects [10], reflection nebulae [11], certain R-Coronae Borealis stars, and carbon rich stars [1216], as well as in some resources on Earth [17, 18]. The identification and quantitative assessment of these molecules, both in natural and in artificially synthesized materials, were made possible by the measurement of their IR spectra, the dependence of these spectra on temperature, the molar absorptivity, and integrated molar absorptivity of their absorption bands [226].

It is expected that also higher fullerenes can be found in space, besides C60 and C70. Calculations [27] suggest that, on a per carbon atom basis [1], higher fullerenes are thermodynamically even more stable than C60, C70 [28], and from the hydrogenated derivatives fulleranes [17, 18, 2931]. Their formation through coalescence of smaller fullerenes [32] and by laser ablation of carbon [1719, 33, 34] also leads to the conclusion about their possible presence in nature.

For the qualitative detection of C76 and C84 fullerenes, the knowledge of the infrared band position and band widths, as well as the evolution of these parameters with temperature, is necessary. This need was fulfilled, for instance, by the previous works [1, 3542] in the infrared spectroscopy of C76 and C84, whereas quantitative assessment of these fullerenes requires knowledge about intensities of their IR absorption bands, which is provided in the current work.

In the first phase of this research, the only stable C76-D2 isomer [4345] and the most abundant, stable isomer of the higher fullerene C84 with D2 symmetry, C84-D2:22 [4654], were isolated from carbon soot, by new and advanced chromatographic methods and processes [3542], in comparison to previous methods for the separation of higher fullerenes under pressure [5563]. Their IR (KBr) spectra were recorded over the entire relevant region, from 400 to 2000 cm−1 in transparence mode [3542], and in the absorption mode in this article.

A comparison of the experimentally observed vibrational frequencies in the IR absorption spectra of the isolated C76-D2 and C84-D2:22 samples [35, 38] with the semiempirical QCFF/PI, DFT, as well as TB potential theoretical calculations for these fullerenes [44, 45, 4850], is presented in this article, indicating their validity.

In this work also, the molar extinction coefficients and the integrated molar extinction coefficients of their main and characteristic IR absorption bands were determined.

These data are important for the qualitative and quantitative determination of the C76-D2 and C84-D2:22 isomers, either in natural resources on Earth and in space or in artificially synthesized materials, electronic and optical devices, diagnostic and therapeutic agents for the applications in biomedical engineering, and so forth.

2. Experimental Methods

In the first phase of this research, C60, C70 [2426], and the higher fullerenes, mainly C76 and C84 [3542], were Soxhlet-extracted with a series of different and previously unapplied solvents or combinations of solvents from the samples of carbon soot, produced by electric arc (MER Corporation, Tucson, USA). The extraction procedures were performed until the complete disappearance of color in a Soxhlet extraction thimble. Solvents used were n-heptane, toluene, chlorobenzene, p-xylene, a mixture of o/m/p-xylene, and pyridine, as well as the successive use of toluene and chlorobenzene and p-xylene and pyridine. The yields, as well as the compositions of all the extracts, were determined by spectroscopic and chromatographic methods. The procedures for increases of fullerenes yields, as well as for additional selective extraction of higher order fullerenes, were found [2426, 3542].

In the second phase, C60, C70, and the higher fullerenes C76 and C84 (the only stable C60-Ih, C70-, and C76-D2 isomers of the first three mentioned fullerenes and the most abundant, stable C84 isomer of D2 symmetry) were chromatographically separated from the obtained soot extracts on the activated Al2O3 columns, by new and advanced methods [3542].

The main difference and advancement of these methods [3542], in comparison to previous methods under pressure [5563], is the isolation of the purified stable isomers of the higher fullerenes C76 and C84 (the C76-D2 and C84-D2:22 isomers), successively after the basic fullerenes, in one phase of each of the processes, under atmospheric pressure and smaller flow of 1.5 mL/min, in increased milligrams yields. The other advantages of the developed methods [35, 42] are the use of significantly smaller amounts of the initial materials, as well as less expensive laboratory equipment. In these methods [35, 42], the entire materials and energy expense, the time spent on the purification processes, and environmental pollution were decreased, using smaller amounts of less toxic solvents. The yields and the purities of the isolated fullerenes were increased or maximized [35, 36, 39].

Purification of the higher fullerenes under pressure, on a preparative scale, either by flash chromatography or by HPLC, generally required larger amounts of the initial materials and repeated chromatographies, and the fullerenes were obtained in smaller yields [5563].

In our new methods [3542], the elution was performed continuously with several different original, defined gradients of solvents: from pure hexane or 5% toluene in hexane to pure toluene. The amounts of the initial materials used were as follows: fullerenes extracts, 10 mg, and finely granulated Al2O3, 50 g, activated for 2 h at 105°C, and eluent (1.5 to 1.75 L) per chromatographic separation [3542]. Starting from 10 mg of the soluble soot extract, in average ca. 1 mg of C76 and ca. 1 mg of C84 were isolated in purified form per one chromatographic process, or up to few milligrams in some cases. The time spent on the purification processes was from 16.7 to 19.4 h [36, 39].

For comparison, using flash chromatography to separate fullerenes [55], on alumina, with hexane or 5% toluene in hexane as eluent, required about 50 times larger quantities of the initial materials, such as 500 mg of crude fullerenes extract, 2500 mg of alumina, and about 12.5 L of solvent for one chromatographic fraction, C60, or 75 L for six chromatographic fullerene fractions, per one chromatography and the large size of columns. The entire time of this purification process, including repeated chromatographies, was 66 hours and purified higher fullerenes were obtained in lower yields. From the total amount of 2500 mg of toluene soluble soot extract, 12 mg of C76 and 2 mg of C84 were isolated.

From these data, it follows [36, 39] that 21 times larger amounts of the initial materials (extract, stationary phase, and solvent) and 2 times longer time are needed for obtaining 1 g of purified C76, and 125 times larger amounts of the initial materials and 10 times longer time are required for obtaining 1 g of purified C84 by the mentioned flash chromatography process [54], in comparison to our protocols [3542].

In the previous method under pressure [57, 58], the purified basic and higher fullerenes were eluted according to their molecular weights on the monomeric ODS column, using large volumes of solvents, in comparison to our new methods [3542]. Several tens of liters of a mixture of toluene and methanol (55 : 45, v/v) per chromatography were used, at a flow rate of 40 mL/min [57, 58]. In the new methods [3542], under atmospheric pressure and smaller flow rate of 1.5 mL/min, significantly smaller volumes of solvents were used for the elution of the purified basic and higher fullerenes in one phase, 1.5 to 1.75 L per chromatography.

The IR spectra of all the chromatographically purified fractions of the basic and the higher fullerenes from this research, as well as of the obtained soot extracts, were previously recorded on a Perkin Elmer FT-IR 1725 X spectrometer by the KBr pellet technique, from 400 to 4000 cm−1, at a resolution of 1 cm−1, in the transparence mode [2426, 36, 37, 3942].

The IR spectra of the C76-D2 and C84-D2:22 samples, isolated by the new and advanced chromatographic methods [3542], were also recorded on a Thermo Scientific FT-IR spectrometer Nicolet IR-6700, by the KB disk technique, in the range of 400–2000 cm−1, at a resolution of 1 cm−1, in the transparence mode [35, 38], as well as in the absorption mode in this article.

2.1. Measurement of the Molar Absorptivity and Integrated Molar Absorptivity of C76-D2 and C84-D2:22

Chromatographically isolated C76-D2 (0.249 mg) and C84-D2:22 (0.270 mg) were mixed with 70.8 mg and with 77.8 mg of KBr, respectively. The obtained powder was compressed at the 4 tons/cm2 with the Perkin Elmer press.

The resulting pellets were placed in the FT-IR spectrometer. Measurements of the intensities (heights) of the absorption bands, as well as of the integrated band intensities of C76-D2 and C84-D2:22, with automatic subtraction of the baseline, were made possible through the OMNIC software from Thermo Scientific, dedicated to the FT-IR spectrometer. This software has also been recently used for the measurement of relative intensities of IR absorption bands of C60 and C70 [4].

The masses of the resulting pellets were 71.0 mg and 78.1 mg, and the percentages of carbon determined by the elemental analysis were 0.351 and 0.346. Their measured thicknesses () were 0.67 mm~0.07 cm and 0.74 mm~0.07 cm, the diameters () were 0.7 cm, and the half diameters () were 0.35 cm.

The volumes of the pellets () were determined from the abovementioned and parameters, by the equation . The obtained values of the volumes, as well as the thicknesses of pellets, were also confirmed using KBr density (2.753 g/cm3) [4] and the masses of pellets.

Concentrations () of fullerenes C76 and C84 in the pellets, as the number of moles per unit of volume, were calculated using the masses of C76 and C84 in the pellets, their molar masses of 912.76 g/mol and 1008.84 g/mol, and the volumes of pellets.

The ()−1 values were determined for the C76-D2 and the C84-D2:22 samples in KBr pellets from the abovementioned experimental parameters. The ()−1 value found for C76-D2 was 1409.7 L·cm−1·mol−1 and the ()−1 value found for C84-D2:22 was 1436.0 L·cm−1·mol−1.

3. Results and Discussion

In the recent works [1, 3542], the IR spectra of the higher fullerenes C76 and C84 and their stable isomers of D2 symmetry have been studied. The dependence on temperature of the position and width of their infrared absorption bands has been determined [1, 35]. The molar extinction coefficients and integrated molar absorptivity of the infrared absorption spectra of C60 and C70, as well as of related hydrogenated derivatives, fulleranes, have also been recently determined [25]. However, neither the molar absorptivity nor the integrated band intensity of C76-D2 and C84-D2:22 has been reported.

Determination of molar absorptivity of the isolated higher fullerenes, in L·cm−1·mol−1, at a given wavenumber, , was achieved through (1), previously applied for C60 and C70, as well as for hydrogenated fullerenes [26, 64], according to Lambert and Beer law, using the absorbance read at a given wavenumber: The determined values of ()−1 for both the C76-D2 and the C84-D2:22 samples are reported in the Experimental Methods.

It was found that the peak height measurements that correspond to the absorbance are sensitive to changes in the resolution of the spectrometers used [26, 64]. The measurement of the integrated intensity that corresponds to the total area below a given absorption band is much less sensitive to instrumental resolution than the peak height measurement [26, 64].

Thus, the absorbance and the integrated band intensities in the obtained original IR spectra of the isolated C76-D2 and C84-D2:22 samples were determined using the OMNIC software of our spectrometer, in both cases subtracting automatically the baseline.

The integrated molar absorptivity of the C76-D2 and C84-D2:22 fullerenes, expressed in cm mol−1 or 10−5 km mol−1, was determined by (2), previously applied for the basic fullerenes, as well as for fulleranes [26, 64]:

In this equation, λ is the wavelength and is the molar absorptivity measured with a spectrometer with unlimited resolution, integrated over the whole band. In practice, by substituting (1) into (2), we get [26, 64]

The original, characteristic, representative IR spectrum of the isolated sample of the C76-D2 isomer is obtained in this article in the absorption mode, Figure 1, for determination of the molar absorptivity and integrated molar absorptivity of its absorption bands, which is important for the quantitative assessment of this fullerene and represents the main work of this article. It was previously provided in transparence mode, in supplemental material of our article [35], for the qualitative determination.

The main three, most intense, dominant C76 maxima, registered in this research [3542], appear at 967, 1082, and 1187 cm−1, with some weak, distinct shoulders. Characteristic, sharp absorption bands unique to C76 occur in the first relevant part at 893 and 823 cm−1, with a neighboring shoulder at 792 cm−1. Several other bands are present at 703 cm−1 with a shoulder at 742 cm−1, at 605 cm−1 with the shoulders at 647 and 665 cm−1, and at 484 cm−1 with the shoulders at 538, 462, 456, and 426 cm−1. Pronounced and intense bands are present in the higher frequency region at 1386 cm−1 with the shoulders at 1397 and 1364 cm−1, at 1493 cm−1 with a neighboring shoulder band at 1462 cm−1, as a doublet, and at 1735 cm−1. Maximum at 1312 cm−1 appears with the neighboring shoulders at 1273 and 1248 cm−1, as a triplet. Complete absorption in this spectrum [35] is in agreement with the theoretical calculations for C76-D2, as well as for its dianion [44, 45].

In the previous articles [35, 37], a comparison of the experimentally observed absorption frequencies in the IR spectra of the chromatographically isolated C76-D2 samples, recorded on Perkin Elmer [37, 4042] and on Thermo Scientific FT-IR spectrometer Nicolet IR-6700 at room temperature [35], with the semiempirical QCFF/PI theoretical calculations for this fullerene [35, 37, 4042, 44], as well as with the IR spectra of C76, recorded on three different temperatures between −180°C and +250°C [1, 35], was presented. On the basis of the obtained excellent agreement [35, 37, 4042, 44], the validity of both the experimental results [35, 37, 4042] and the mentioned theoretical calculations for C76-D2 [44] was indicated [35, 37, 44]. In the more recent article [35], a larger number of experimentally registered vibrational frequencies of C76 were presented and theoretically confirmed [35, 44].

There is also a good agreement between the absorption bands in our infrared spectra at room temperature [3542] and the recent spectra of C76-D2 at three different temperatures [1]. Only some smaller shifts, as well as some changes of their relative intensities with the temperature, were observed [1, 35].

In this article, a comparison of the experimentally obtained vibrational frequencies (cm−1) in the IR absorption spectra of the chromatographically isolated C76-D2 samples (IR1-IR3), recorded from 400 to 2000 cm−1, on a Thermo Scientific FT-IR spectrometer Nicolet IR-6700 [35], with the different theoretical calculations, by the QCFF/PI method (Calc. 1, from 286 to 1668 cm−1) [44] and DFT method for C76 (Calc. 2, from 206.7 to 1602.7 cm−1) [45], as well as for (Calc. 3, from 195.7 to 1556.0 cm−1) [45], is presented in Table 1. Excellent agreement is obtained between the experimental results [35] and all the aforementioned theoretical calculations for this fullerene [44, 45], as the evidence of their validity.

The IR spectra of all the chromatographically isolated samples of the C76-D2 isomer from this research, recorded on the two mentioned spectrometers, have similar properties. All the observed vibrational frequencies and the general pattern of these spectra [3542] are in agreement with the semiempirical QCFF/PI [44] and DFT theoretical calculations for C76-D2 [45], as well as for its dianion C76- [45].

The achieved agreement between our experimental results [3542] and all the aforementioned theoretical predictions of the IR absorption frequencies of C76-D2 [44, 45], which is presented in this article in Table 1 and Figure 1 [35, 44, 45], is better in comparison to previous, partial experimental results for the obtained C76 samples, from other separation processes, by other IR techniques [5962].

It is important to mention that the obtained generally good correlation between the overall configuration of absorption and all the observed vibrational frequencies in our recent experimental IR spectra for the neutral C76-D2 [3542] and the next obtained infrared multiphoton electron detachment (IR-MPED) spectrum of the unsolved gas phase dianion C76- [45], as well as with the adequate most recent B3LYP/TZVP DFT calculations, presented in this article in Table 1, Figure 1 [35, 45], provides significant experimental evidence [3542] that the dianionic molecule retains its overall symmetry (i.e., D2 point group) with 1A1 ground state with respect to the neutral cage [45].

From the IR spectrum of C76-D2 in a mode, presented in Figure 1, the absorbance values , as well as the integrated absorbance values of the absorption bands, were determined using the OMNIC software.

The molar absorptivity , calculated according to (1), the integrated molar absorptivity , calculated according to (3), and the integration ranges of absorption bands of this fullerene are reported in Table 2.

It can also be seen from Table 2 that excellent agreement is found between the relative intensities of the main and characteristic absorption maxima of C76-D2 computed from and from the Ψ values, in adequate integration ranges, taking as 100 the most intense vibration mode of C76-D2 at the frequency of 967 cm−1.

The original, characteristic, representative IR absorption spectrum of the isolated sample of the isomer C84-D2:22 is obtained in this article in the absorption mode, Figure 2, for determination of the molar absorptivity and integrated molar absorptivity of its absorption bands, which is important for its quantitative determination, as the main work of this article. It was previously provided in transparence mode [35], for qualitative determination.

A group of sharp, characteristic absorption bands is present between ca. 700 and 840 cm−1 [3542], at 711, 746, 779, and 843 cm−1, followed by the bands at 635 and 473 cm−1 in the first relevant part. Dominant and pronounced C84-D2:22 maxima appear in the higher frequency region, between ca. 1390 and 1120 cm−1, as well as a group around 1600 cm−1. The main, most intense band is present at 1385 cm−1, followed by the bands at 1263 cm−1 and 1122 cm−1. Intense bands also appear at 1456–1465 cm−1, 1599–1616 cm−1, and 1731 cm−1. The entire absorption in this spectrum [35] corresponds to the theoretical predictions for C84-D2:22 [4850].

In the previous article [35], a comparison of the experimentally observed absorption frequencies in the IR spectra of the chromatographically isolated C84-D2:22 samples, recorded on a Thermo Scientific FT-IR spectrometer Nicolet IR-6700 at room temperature [35, 38], with the semiempirical QCFF/PI theoretical calculations for this fullerene [48], as well as with the IR spectra of C84 (mixture of isomers), recorded on three different temperatures between −180°C and +250°C [1, 35], was presented. On the basis of the obtained excellent agreement [35, 38, 48], the validity of both the experimental results [35, 38] and the mentioned theoretical calculations [48] was indicated [35].

Most of the absorption maxima in our IR spectra of C84-D2:22 at room temperature [3542] are also in good agreement with the recent IR spectra of C84 (mixture of isomers) at different temperatures between −180°C and +250°C [1], as presented in the previous article [1, 35, 38]. However, significant changes of relative intensities of the main bands, as well as some shifts, were observed [1, 35].

In this article, a comparison of the experimentally obtained vibrational frequencies (cm−1) in the IR absorption spectra, of the chromatographically isolated C84-D2:22 samples (IR1-IR3), recorded from 400 to 2000 cm−1, on a Thermo Scientific FT- IR spectrometer Nicolet IR-6700 [35, 38], with the different theoretical calculations for this fullerene, by the QCFF/PI method (Calc. 1, from 179 to 1711 cm−1) [48], DFT (Calc. 2, from 211 to 1674 cm−1) [49], and TB potential method (Calc. 3, from 190 to 1726 cm−1) [50], is presented in Table 3. Excellent agreement between the experimental results [35, 38] and the aforementioned theoretical calculations for this fullerene [4850] provides the evidence of their validity.

The IR spectra of all the chromatographically isolated samples of the isomer C84-D2:22 from this research, recorded on the mentioned spectrometers, have similar properties. All the observed vibrational frequencies and the overall appearance of these spectra [3542] are in excellent agreement with the semiempirical QCFF/PI, DFT, and TB potential calculations for this fullerene [4850].

The achieved agreement between our experimental results [3542] and the aforementioned theoretical predictions for this molecule [4850], which is presented in this article in Table 3 and Figure 2 [35, 38, 4850], is better in comparison to previous experimental results for the obtained C84 samples (partially separated isomers) from other separation processes, by other IR techniques [6063]. This was also mentioned in the previous article [38].

From the IR spectrum of C84-D2:22 in a mode, presented in Figure 2, the absorbance values , as well as the integrated absorbance values of the absorption bands, were determined using the OMNIC software.

The molar absorptivity , as well as the integrated molar absorptivity , calculated according to (1) and (3), and the integration ranges of the absorption bands of this fullerene are presented in Table 4.

Also in this case, as can be seen from Table 4, excellent agreement is found between the relative intensities of the main and characteristic absorption maxima of C84-D2:22 calculated from and from the values, in adequate integration ranges, taking as 100 the most intense vibration mode of C84-D2:22 at the frequency of 1385 cm−1.

4. Conclusion

In this research, the stable C76 and C84 isomers of D2 symmetry were isolated from carbon soot, by new and advanced chromatographic methods and processes [3542]. The IR-KBr spectra of the isolated fullerenes were obtained over the entire fullerenes fingerprint region, 400–2000 cm−1, on a Thermo Scientific FT-IR spectrometer, in transparence mode [35, 38], as well as in the absorption mode in this article.

Based on comparison of the experimentally observed infrared absorption frequencies of the isolated C76-D2 and C84-D2:22 samples [35, 38] with the semiempirical QCFF/PI, DFT, and TB potential calculations for these fullerenes [44, 45, 4850] and the obtained excellent agreement [35, 38, 44, 45, 4850], presented in this article, the validity of both the experimental results [35, 38] and all the mentioned theoretical calculations [44, 45, 4850] is confirmed. These research results can be used for their qualitative determination.

The molar extinction coefficients and the integrated molar extinction coefficients of the IR absorption bands of the C76-D2 and C84-D2:22 isomers were determined at room temperature in KBr matrix. Excellent agreement is found between the relative intensities of the main and characteristic absorption maxima of these fullerenes calculated from the values and from the values in adequate integration ranges. These results can be used for their quantitative determination.

All the obtained data are important for the identification and quantitative assessment of the C76-D2 and C84-D2:22 isomers, either in natural resources on Earth and in space or in artificially synthesized materials, electronic and optical devices, such as polymers, composites, nanophotonic and biocompatible materials, optical limiters, sensors, special lenses with optical absorption properties closer to human eye light sensitivity, diagnostic and therapeutic agents, pharmaceutical substances, and biomaterials.

Conflicts of Interest

The authors declare that there are no conflicts of interest regarding the publication of this paper.

Acknowledgments

The authors are grateful to the Ministry of Education, Science and Technological Development of the Republic of Serbia and to the University of Belgrade for financial support of this research (Project III 45009).