**3. The establish of group-type analysis method by CLC**

#### **3.1 Optimum chromatographic condition**

As a base line, some pure reagents were chosen as model components prepared for CLC. These model compounds were tetracosane for saturates, dibenz[ah]anthracen for aromatics and acetanilide for resins. There is no appropriate pure reagent used for asphaltene fraction, so the insoluble fraction of tetrahydrofuran in one asphalt sample was used for asphaltene fraction.

Through a series of investigations,the optimum chromatographic operation was performed. The final optimum conditions were obtained as follows: Chromatographic column was glass column being 90 mm length, 6 mm i.d. The amount of silica gel used was from 1 to 1.5 gram.

process and this lack made it not suitable for other analysis with preparation fraction. It should be pointed that the conventional method such as ASTM method use amount of solvent is large and some solvents has high toxicity [4, 5]. Moreover, there are too troublesome for some operation in traditional method. Hence, the separation of products

Refereeing the literatures [4-10], the authors of this paper establish an optimum CLC method to analyze group-type of heavy oils through a series of studies. This paper detail introduces this method and its many applications which include preparation of high-level road asphalt, the characterization of molecular weight distributions (MWDs) and analysis of

The dimension of glass chromatographic column is 90 mm length and 6 mm I. D. Silica gel with particle size range from 100 to 200 meshes was provided by marine chemical plant of Qingdao China. Silica gel was active under temperature of 180oC for 4 hours before use. Oxide of alumna 0.047-0.147 mm used was purchased from chemical and medical reagent company in Shanghai China. Muffle furnace (50°C-1000°C) and oven was used for sample

N-heptane, dichloromethane, trichloromethane as eluent solvents all were analytical grade reagents produced by Tianjin Chemical Reagent Factory (China). Pure reagents as model compounds were supplied by Aldrich Chemical Company (USA), including tetracosane

Fourier transforms FT-IR spectra were measured by a Bio-Rad Excalibur Series FTS 3000 spectrometer in the range of 4000-400 cm−1 using KBr pellets. 1H NMR measurements were

As a base line, some pure reagents were chosen as model components prepared for CLC. These model compounds were tetracosane for saturates, dibenz[ah]anthracen for aromatics and acetanilide for resins. There is no appropriate pure reagent used for asphaltene fraction, so the insoluble fraction of tetrahydrofuran in one asphalt sample was used for asphaltene

Through a series of investigations,the optimum chromatographic operation was performed. The final optimum conditions were obtained as follows: Chromatographic column was glass column being 90 mm length, 6 mm i.d. The amount of silica gel used was from 1 to 1.5 gram.

(99.5% pure), dibenz[ah]anthracen ( 98%, pure), and acetanilide ( 99% pure), etc.

made with a Bruker Avance 500 spectrometer operating at 500.1 MHz.

**3. The establish of group-type analysis method by CLC** 

**3.1 Optimum chromatographic condition** 

containing heavy components remains a difficult task up to now.

heterocyclic aromatic components of heavy oils.

**2.1 Column, support and heating apparatus** 

**2. The establish of CLC method** 

preparation and heating.

**2.3 Analytical instruments** 

**2.2 Reagents** 

fraction.

The amount of alumina was from 1.5 to 1.8 gram. Total sample used was about 0.1 gram. The solvent of heptanes, mixture of heptanes/ dichloromethane (1/2.5, V/V) and mixture of dichloromethane/ trichloromethane (1/3, V/V) were as elutes corresponding to saturated hydrocarbon, aromatic hydrocarbon and resin respectively. The amount of heptanes, heptanes/ dichloromethane, and dichloromethane/ trichloromethane was 20ml, 35ml and 30ml respectively. Each fraction collected was dried in vacuum under 60oC until the weight keep constant.

Through above group analysis, the experimental deviation and recovery of CLC method are summarized in Table 1. From Table, it can be seen that the average of deviation and recover are -1.546% and 100.681% respectively; the results are good.


Table 1. Experimental deviation and recovery of model compound.

#### **3.2 Check of chromatographic resolution rate by FT-IR**

The result of CLC method was checked by Fourier transform infrared (FT-IR) method .The spectra IR were acquired in the transmission mode as 64 scan in the IR range from 4000 to 500cm-1 at a resolution of 4cm-1. KBr standard pellets were used, and the samples were dried and then mixed with KBr, ground, and palletized.

IR spectrums of pure reagents including tetracosane, dibenz(ah)anthracen and acetanilide were obtained and used for standards. The IR spectrums of different fractions collected from flow out separated of the mixture reagents, and spectrums were compared with above standard spectrums. The results were shown in Figure 1.

Column Liquid Chromatography 5

0 2 4 6 810

ppm

The recover rate and experiment deviations for model compounds were summarized in

Compared with routine ASTM method, these optimum chromatographic conditions show many advantages. First, the reagent and sample consumed was fewer than total solvent of 300 ml of classic ASTM method. Second, the dichloromethane and trichloromethane used in

Coal is used as the main source of energy in China. The crude oil produced in China is paraffinic; therefore, it is not suitable for road asphalt. China is trying to produce high grade

present study, compared with toluene and benzotrichloride used, has lower toxicity.

Fig. 2. 1H NMR results of pure reagents and different fraction.

**3.4 Evaluation of analysis of group composition by CLC** 

Table 1. It can be seen that the experiment result are fine.

**4. Applications of group type analysis by CLC 4.1 The application in making high grade road asphalt** 

road asphalt from the mixture of coal and petroleum [11, 12].

Fraction 3

Acetanilide

Fraction 2

Fraction 1

Tetracosane

Dibenz(ah)anthracen

Fig. 1. Infrared spectrum for pure reagents and different fraction.

It is important to indicate that the IR spectra of fraction 1 collected (from 1202# sample) show similarity with pure tetracosane reagent. IR spectra for fraction 2 and fraction 3 show accordant results with dibenz(ah)anthracen and acetanilide respectively.

#### **3.3 Check of chromatographic resolution rate by 1 H NMR**

The CLC method was checked also by 1H NMR. It measured different fractions collected from flow out separated of the mixture reagents and spectrums were compared with above standard spectrums. The high resolution 1H NMR spectra of pure model compounds and fraction 1-3 are shown in Figure 2.

It is difficult to separate complex and heavy sample, however the IR and 1H NMR analysis of the prepared fractions from CLC were all good agreement with pure reagents. This observation indicate the optimum CLC parameter in this work guarantee a good qualitative results.

Tetracosane

Fraction 1

Fraction 2

Acetanilide

Fraction 3

Absorbance

Dibenz[ah]anthracen

4000 3500 3000 2500 2000 1500 1000 500

Wavenumber (cm-1

It is important to indicate that the IR spectra of fraction 1 collected (from 1202# sample) show similarity with pure tetracosane reagent. IR spectra for fraction 2 and fraction 3 show

The CLC method was checked also by 1H NMR. It measured different fractions collected from flow out separated of the mixture reagents and spectrums were compared with above standard spectrums. The high resolution 1H NMR spectra of pure model compounds and

It is difficult to separate complex and heavy sample, however the IR and 1H NMR analysis of the prepared fractions from CLC were all good agreement with pure reagents. This observation indicate the optimum CLC parameter in this work guarantee a good qualitative

Fig. 1. Infrared spectrum for pure reagents and different fraction.

**3.3 Check of chromatographic resolution rate by 1**

fraction 1-3 are shown in Figure 2.

results.

accordant results with dibenz(ah)anthracen and acetanilide respectively.

)

**H NMR** 

Fig. 2. 1H NMR results of pure reagents and different fraction.

#### **3.4 Evaluation of analysis of group composition by CLC**

The recover rate and experiment deviations for model compounds were summarized in Table 1. It can be seen that the experiment result are fine.

Compared with routine ASTM method, these optimum chromatographic conditions show many advantages. First, the reagent and sample consumed was fewer than total solvent of 300 ml of classic ASTM method. Second, the dichloromethane and trichloromethane used in present study, compared with toluene and benzotrichloride used, has lower toxicity.
