GRAPHICAL METHOD SIMPLIFIES DIESEL CLOUD POINT DETERMINATIONS

Sept. 24, 1990
H.U. Khan, M.M. Mungali, K.M. Agrawal, G.C. Joshi Indian Institute of Petroleum Dehradun, India A graphical method for the determination of the cloud point of diesel oil blends has been developed that uses only the composition of the blends. The method has advantages over other methods because information on the n-paraffin content and weighing factors for the volume fractions of the components are not required. Cloud points are determined experimentally for only a few of the potential volume
H.U. Khan, M.M. Mungali, K.M. Agrawal, G.C. Joshi
Indian Institute of Petroleum
Dehradun, India

A graphical method for the determination of the cloud point of diesel oil blends has been developed that uses only the composition of the blends.

The method has advantages over other methods because information on the n-paraffin content and weighing factors for the volume fractions of the components are not required.

Cloud points are determined experimentally for only a few of the potential volume percentages of the fractions used in a diesel blend.

From these data, a plot is constructed from which cloud points of all possible compositions of the fractions used in the blend can be estimated.

The cloud points obtained by the graphical method compare well and are within repeatable limits of the ASTM test method.

DIESEL BLENDS

Diesel fuel is produced by blending various distillate fractions, generally having boiling ranges of 150-380 C. When diesel blends are developed for cold climates, cloud point is the primary factor for consideration.

Determination of cloud point of one blend formulation normally takes about 2 hr by the ASTM D-2500 test method. Therefore, a considerable amount of laboratory time would be required to determine the cloud points of several potentially useful diesel blends.

With diesel fuels that are blended from three or four distillate fractions, there are many possible formulations that will yield the desired cloud point target. For these blends, rigorous optimization is impractical.

Where there is no quantitative relationship between the blend formulation and cloud point, optimization will not help in long-term planning of distillate utilization. This has created the need for the development of a simple and quick method for cloud point determination of diesel blends.

Various attempts have been made to correlate cloud point with the composition and n-paraffin concentration of distillate fuels.1-9 However, these studies have limited applications because most of them are based on one or two paraffins in the base fuel.

But diesel fuels contain a large variety of n-paraffins with chain lengths ranging from C9 to C27.2-4 Recently, mathematical correlations between the blend composition and cloud point have been proposed.10

These correlations suffer from the drawback that an initial value for the weighing factor must be known. If not, the correlation can be very time consuming and tedious.

An extension to the correlation was developed by Seglin that improves the algorithm for determining the equation.11

The graphical, cloud point determination method was developed from the blend compositions with only a small number of data required.

EXPERIMENTAL BASIS

Four distillate fractions, boiling in the ranges of 140-180, 180-230, 230-300, and 300-360 C. were obtained by distilling Bombay High crude oil. The physiochemical characteristics of these distillates are shown in Table 1.

These distillates correspond to heavy naphtha (Z1), kerosine (Z2), light gas oil (Z3), and heavy gas oil (Z4). Using different volume percents of Z2, Z3, and Z4, and keeping Z1 at 5 vol %, 21 different diesel oil blends were prepared.

The volume percents of the components in the blend were varied in steps of 10 vol %. The cloud points of these blends were then determined following the ASTM D-2500 procedure (Table 2).

NOMOGRAPH DEVELOPMENT

Using the data in Table 2, the volume percent of Z2 and Z4 in each blend was plotted against their respective cloud points on a three-dimensional plot, with Z2 on the X axis, cloud point on the Y axis, and Z4 on the Z axis. The plots result in the formation of the three curves (AB, AC, and DE) in Fig. 1.

From each point on the curve AB, curves similar to AC and DE were extrapolated, corresponding to the constant values of Z2 and Z3.

Each point on curve AC was then joined with the intersecting points to get constant-value curves (straight lines) for Z4. The extrapolation resulted in the formation of a triangular graph, which for 5 vol % heavy naphtha (Z1 = 5 vol %) as one component of the blend, the cloud point of any desired blend containing varying volume percentages of kerosine (Z2), intermediate distillate (Z3), and heavy gas oil (Z4), can be read.

Where more or less heavy naphtha could be in the blend, additional graphs could be constructed for each heavy naphtha volume percentage. Each graph, then would give the entire range of cloud points of any combination of the other fractions making up the final diesel blend.

USING THE PLOTS

Some examples, shown in Fig. 1, indicate how the cloud point is read from the plot for various blend compositions.

For a blend of 5 vol % heavy naphtha (Z1), 40 vol % kerosine (Z2), 25 vol % intermediate distillate (Z3), and 30 vol % heavy gas oil (Z4), a line parallel to the X axis is drawn from the 30 vol % point on the Z axis (Point G). Another line, parallel to the Z axis is drawn through the 40 vol % point on the Z2 axis.

Where these lines intersect (Point F), a line parallel to the Y axis is drawn up to the Z3 composition of 25 vol % (Point H). Another line from Point F is drawn to intersect with point K.

The cloud point value is located by drawing a line from Point H, parallel to line KF. This line crosses the Y axis at Point J, the value of the cloud point for the blend.

For this example, the cloud point of a blend of 5 vol % heavy naphtha, 40 vol % kerosine, 25 vol % intermediate distillate, and 30 vol % heavy gas oil, is 278 K. or 5 C.

If the blend is 5 vol % heavy naphtha, 5 vol % kerosine, 50 vol % intermediate distillate, and 40 vol % heavy gas oil, the cloud point is 283 K. or 10 C. The lines used to determine the cloud point of this second blend are shown in Fig. 1. Similar lines can be constructed for all of the various possible blend combinations of the four blend components.

To check the validity of the plot of Fig. 1, and to determine the cloud point of the unknown blends, 21 different blends were prepared, with compositions different from the one used to prepare Fig. 1. The cloud points corresponding to the composition of these blends were then obtained from Fig. 1 using the above procedure.

Table 3 compares the cloud points determined from Fig. 1 to cloud points of the same compositions determined from the ASTM D-2500 procedure. The results obtained from the plot are quite close to the values obtained by the laboratory procedure.

OTHER BLENDS

Three-component diesel blends can also be plotted the same way as Fig. 1. One such plot was made from data generated by the equation developed by NOVA/Husky and represents three refinery distillate streams that could be blended into diesel fuel at various compositions (Fig. 2).10

The plot shown in Fig. 2 covers light, intermediate, and heavy distillate. Note that in the case of the three-component blend, all three compositions are shown on the plot, and therefore, the cloud points of all possible blends of the three components are on a single plot.

The cloud point determinations of Fig. 2 were also checked against ASTM D-2500 procedures. Results showed that the graphical data agreed closely with ASTM cloud points.

Unfortunately, the three-dimensional graphical presentation of cloud point data can be used for blends of no more than four components; however, diesel blends of more than four components are unusual.

Because cloud point, pour point, and cold filter plugging point (CFPP) all correlate well, similar graphical techniques could be used to determine these properties in the same manner.

REFERENCES

  1. Brownawell, D.W., and Hollyday, Petroleum, Vol. 48, No. 463, 1962, p. 209.

  2. Dimitroff, E., and Dietzmann, H.E., preprint of the American Chemical Society, Petroleum Division, Vol. 14, 1969, p. B 132.

  3. Holder, G.A., and Winkler, J., Journal of the Institute of Petroleum, Vol. 51, No. 499, 1965, p. 228.

  4. Petrovic, K., and Vitorovic, D., Journal of the Institute of Petroleum, Vol. 59, No. 565, 1971 p. 20.

  5. Walsh, R.P., and Mortime, J.V., Hydrocarbon Processing, Vol. 50, 1971, p. 153.

  6. Affens, W.A., Hall, J.M., Hold, S.. and Hazle, R.N., Fuel, Vol. 63, 1984, p. 543.

  7. Reddy, D.R., Fuel, Vol. 65, 1986, p. 1647.

  8. Krishna, R. Bhattacharjee, S., Joshi, G.C., Erdol und Kohle, Bol. 42, 1989, pp. 72-75.

  9. Krishna, R., Joshi, G.C., Agrawal, K.M, Energy and Fuel, Vol. 3, 1989, pp. 15-20.

  10. Tsang, C.Y., Ker, V.S.F., Miranda, R.D., and Wesch, C.J."Equation predicts diesel cloud points," OGJ, Mar. 28, 1988, pp. 33-38.

  11. Seglin, L., "Extensions developed for NOVA/Husky cloud point equation," OGJ, Oct. 24, 1988, pp. 68-70.

Copyright 1990 Oil & Gas Journal. All Rights Reserved.