Oil shale samples were provided from four major oil shale locations in central and southern Jordan (Figure 1). Full sample preparation methodology and characterization experiments were presented in Alnawafleh and Fraige [22] [23] . Analyses performed covered OS petrography, mineralogy, organic carbon content, and Fisher assay. Size distribution test was carried out at Al-Hussein Bin Talal University. A laboratory scale jaw crusher was used to reduce the size of oil shale samples. The crushed OS then fed to a ball mill for 10 mins. After sieving the ground OS, retained OS masses were recorded and the size distribution curves for each sample then constructed.

The average elemental composition % of the total solid samples was analyzed using EuroVector elemental analyzer model: EA 3000 A available at AL al-Bayt University, Jordan.

Shale oil solvent extraction was carried out in three solvation protocols: Soxhlet extraction, extraction via mixing and stirring, and lastly super-critical fluid extraction via extraction reactor designed locally for this purpose (see Alnawafleh and Fraige, 2015a). The purpose is to compare the performance of extraction between them. Eleven different solvents and two solvent mixtures were used in the extraction experiments. Solvent mixture one (Mix. 1) is THF (50%); Acetone (50%), whereas, solvent mixture two (Mix. 2) is Chloroform (50%); Acetone (30%); Methanol (20%).

In the Soxhlet extraction, four grams of finely comminuted oil shale were transferred to a thimble which was then placed in the Soxhlet chamber (100 ml). The Soxhlet chamber was fitted to a distillation flask containing 200 ml of the extraction solvent. Each extraction run was carried out for 24 hrs. Extraction via mixing and stirring involves stirring the OS sample on magnet hot plate at 1000 rpm using 200 ml of solvent. The stirred sample then filtered. Supercritical extraction involves shale oil extraction via the designed reactor and relatively high pressure and temperature conditions. Within the reactor, OS sample and solvent used are charged and the pressure and temperature are raised to the desired values. In all protocols, the extraction solvent was separated form liquor via rotary evaporator and the extract was then dried under a stream of N2 gas. The organic matter (OM) yield was then determined.

Chromatographic column of silica gel and alumina (1.3 cm i.d., 70 cm height) was used to fractionate the obtained extract from Soxhlet extraction using solvent Mix. 2. The 30 g of silica gel layer used in the column is placed in the lower part of the chromatographic column while the 10 g of alumina layer, in the upper part of the column. Measured quantity of 0.25 g of oil extract was added to the top of the column and eluted sequentially by 100 ml of hexane, 75 ml chloroform-hexane mixture (2:1), and 70 ml ethanol to separate aliphatic, aromatic and NSO fractions respectively.

High performance of digital NMR spectrometer in CDCl3, Model: Bruker 300 MHz/ Switzerland was used to study qualitatively the fractional composition of oil shale extract.

The characterization of the studied oil shales; their mineralogy, composition, texture and quality are reported in Alnawafleh and Fraige [22] [23] . A brief summary is presented in Table 1. The studied oil shales differ in their organic matter content. Their

mineralogy and internal texture are quite similar. Slight differences are present from one location to another. Their oil yielding capacities are quite different.

Grain size distribution of OS samples from four OS deposits is shown in Figure 2. Results show that 80% of all studied samples sizes are less than 350 μm. All deposits reveal quite similar trend but different size distributions. This is normal due to their chemical and physical variability.

The average elemental composition % of the total solid samples calculated from duplicate analyses results is presented in Figure 3. All OS deposits show quite similar distribution for the determined four elements. Highest percentage is obtained from El-Lajjun deposit, whereas the lowest is for Juf Ed-Darawish OS deposit. The elemental composition values are higher than that reported for Egyptian OS by Al-Alla and Nassef [24] from Al-Quseir area. Quite high sulfur content in El-Lajjun OS may limit its future exploitation.

The effect of solvent type on the shale oil extractability is shown in Figure 4 and Figure 5, respectively. The OM yields obtained from different solvents used are clearly unequal for the studied oil shales. The most effective extraction solvent is THF. Highest yield is obtained from polar solvents. This is consistent with the results reported by Shawaqfeh and Al-Harahsheh [14] . Solvation yield can be enhanced using solvent mixtures rather than using pure solvents [18] . The type of solvent used is reported to play a significant effect on the yield and the composition of the obtained oil [25] . The quantity of solvent used in the extraction is also very important factor [24] . A study by Tamimi and Uysal

[9] on Jordanian OS showed that the best extraction is obtained when the ratio of solvent to oil shale is in the order of 1:2. Since four OS deposits are studied, it is important to mention that the type of OS for each deposit as well as its original OM and type has clear effect on the extraction yield.

The effect of grain size on shale oil extractability is shown in Figure 6 and Figure 7, respectively. Results show that grain size has minor effect on OM yield. Results are consistent with that reported by Anabtawi and Uysal [15] . Grain size effect on shale oil extractability is indicated from the OM yield fluctuation on different grain sizes. The OM yield via soxhlet extraction is found to be high when compared with that of mixing and stirring extraction. The suitable size for extraction is reported to be depends on the extraction kinetics, dissolution thermodynamics, and the mode of mass transfer from the particles [9] .

The effect of stirring time on shale oil extractability is shown in Figure 8. The OM yield

shows slight increase with the increase in mixing time. This is due to the dissolving of shale oil during the early stages of contact between solvent and OS material. After enough extraction time, the extraction completed and the ultimate yield will be achieved. Mixing improves the extraction yield due to increasing agitation and turbulence. The effectiveness of mixing on extraction depends on several factors include the rate of mixing, the design of the mixer, the mode of operation, and the solvent to oil shale ratio [9] .

The effect of extraction temperature on shale oil extractability is shown in Figure 9 and Figure 10, respectively. Generally, the OM yield increases with temperature increase. Significant OM yield is obtained using solvent mixtures under super-critical conditions for both El-Lajjun and Sultani OS deposits compared with the OM yield obtained from traditional methods of shale oil extraction (Figure 10). Super-critical conditions are preferred to be used in the field of shale oil extraction [9] [14] [17] [18] .

Shale oil composition obtained from column chromatography fractionation is shown in Figure 11. Results show that shale oil composition differs from one deposit to another. The aliphatic hydrocarbons are obtained from El-Lajjun and Ash-Shoubak shale oils. With exception of that of Sultani, the aromatic fraction content is generally greater than aliphatic fraction content. All deposits are quite rich in the NSO in their shale oils. High residue fraction is obtained from Juf Ed-Darawish shale oil. The NMR analysis results are presented in Figure 12 and Table 2 respectively. Variability in shale oil composition is clearly indicated. The 1H NMR spectra of the extracted fraction from all OS deposits show saturated aliphatic and aromatic hydrocarbons. Unsaturated hydrocarbons (alkenes and alkynes) are found in shale oil obtained from Ash-Shoubak and Juf Ed-Darawish oil shales.

Shale oil compositional variability results obtained from this study supported by the physical and chemical OS variability reported in literature, e.g. Alnawafleh et al. [4] , indicate that different extraction techniques suit different OS deposits, and such variability should be considered in any future utilization of OS resources in Jordan. The extracted shale oil refinery, therefore, will require special facilities. Accordingly, the economic and environmental aspects of shale oil extraction in Jordan should be highly considered.