IPE-TM-320 Fractionation
Naphtha Splitter Analysis
IPE-TM-320-20
1. Table of Contents
2. Purpose
This procedure describes the analysis for design of a naphtha splitter column using a theoretical stage fractionation model in a process simulator. Aspen Plus was used but any simulator including UniSim can be used in a similar manner. While this procedure looks at a naphtha splitter many other fractionation columns can be analyzed in a similar manner.
3. General
A naphtha splitter is a common fractionation column. The feed to this column is typically C5 to C10 material with very small amounts of lighter and heavier material.
There are many variations on the separation that are desired for this column. Listed below are some of the possible specifications.
- The amount of C6 material in the distillate should be minimized as benzene production is desired. Benzene and other aromatics are produced by reforming the bottom product. The distillate can be used for gasoline blending or feed to a steam cracker.
- The amount of C6 material in the distillate should be maximized to remove benzene precursors as benzene production is not desired. The bottom product normally goes to a reforming unit and most of the C6 material would be converted to benzene. Only a small amount of benzene is normally allowed in gasoline.
- The true vapor pressure or Reid vapor pressure (RVP) of the overhead product may be required to be less than some value so that the distillate may go to a certain type of storage tank. This may result in significant C6 and possibly C7 going out the column in the distillate even though its value is higher as a bottom product. A RVP of 13 psia is typical for atmospheric storage.
- The amount of light C7 material in the distillate should be minimized as this material goes to an isomerization unit where much of the C7 material will crack.
Listed below are three common types of naphtha splitters
Type 1 has the distillate being petrochemical naphtha (PC naphtha). PC naphtha is frequently used for gasoline blending or as steam cracker feed. There is normally a maximum allowed vapor pressure for the distillate. The naphtha splitter bottom is reformed to produce gasoline. C6 material in the bottom is limited to meet a desired benzene level in the reformate. If C6 material in the bottom is held as a specification then more trays and reflux will yield less C7 material in the distillate with the distillate vapor pressure slightly increasing. The value of more than nominal trays and reflux is generally considered to be small. The typical number of trays for this column type is from 30 to 35 and the typical reflux to feed ratio is from 0.5 to 0.75.
Type 2 has the distillate going an isomerization unit. As the isomerization unit will crack a significant percentage of any C7 paraffins it is normally desirable to minimize the C7 material in the distillate. The naphtha splitter bottom is reformed to produce gasoline. C6 material in the bottom is limited to meet a desired benzene level in the reformate. Note that lighter C7 components are more volatile than the heavier C6 components. Due to this, it is best to exclude these lighter C7 components when specifying the distillate during the tray to tray calculations for this column. More discussion on this point is covered in the rest of this procedure. The typical number of trays for this column type is from 50 to 80 and the typical reflux to feed ratio is from 1.0 to 1.5.
Type 3 has the distillate being PC naphtha. Again the PC naphtha is typically used for gasoline blending or steam cracker feed and there is normally a maximum allowed vapor pressure. The bottom of this column goes to a reforming unit that produces aromatics. As aromatics are a desired product, C6 components are desired in the bottom product. However to obtain an acceptable distillate vapor pressure significant C6 and possibly C7 material is allowed in the distillate. For a given distillate vapor pressure, increasing the trays and reflux will shift the distillate composition to have more C6 and less C7 material. The typical number of trays for this column type is from 50 to 65 and the typical reflux to feed ratio is from 0.5 to 0.75. As an alternative, this column may have a lower vapor pressure side product as PC naphtha to atmospheric storage with the higher vapor pressure, lower flow rate distillate to pressurized storage.
The feed to a naphtha splitter is best defined using standard components rather than a series of pseudo-components. This allows for much better understanding and optimization. Occasionally it may still be needed to use pseudo components for the type 1 naphtha splitter. Tray to tray calculations with pseudo components can be very misleading.
The demonstration below can be applied to any of the three types of naphtha splitter designs for optimization.
Below, a type 2 naphtha splitter is used to demonstrate some column analysis methods.
4. Initial Estimates and Specifications for a Column
It is normally best to analyze the column using a model with 100% efficient theoretical stages. This procedure includes tables and graphs that are used for the optimization work where the number of stages in each column section may be varied. Only at the end of the process work should the actual number of trays be assigned based an overall column efficiency and simulation stages that were used. The use of actual trays for tables and graphs has the disadvantage that there is always some ambiguity in the efficiency and there is some “noise” from the rounding required when converting between actual trays and theoretical stages. Only whole numbers can be used for either and hence rounding is required as one is converted to the other. This rounding causes “noise” in the graphs (or data) that is undesirable during the optimization.
Some simulators allow the use of a stage efficiency, usually a Murphree efficiency. Normally the use of a simulator stage efficiency is undesirable as it slows the calculation, it is different than an overall column efficiency and it has little added value.
To start a process simulator column model initial values are needed for the following variables.
- Total number of stages
- Feed and side product stage location
- Pressure for each stage
- Estimated values for temperature and vapor rate of each stage
- Estimated product rates
From experience a naphtha splitter is expected to have from 30 to 60 stages with the feed near the middle of the column. The procedure IPE-TM-320-15 can be used as a source for typical column design values.
For this naphtha splitter there are only two products, light and heavy naphtha. The light naphtha will contain mainly C5 and C6 material. The pressure needs to be high enough that its bubble point temperature is obtainable with the desired cooling medium. Some conservatism (higher receiver pressure) is reasonable to help the column operate if the feed is slightly lighter than was specified in the design basis. For this column a receiver pressure from 10 to 40 psig is expected. The other stage pressures can be set based on an assigned condenser circuit pressure drop and an assumed stage pressure drop. For an air condenser with a hot vapor by-pass control, 9 psi is typical (4 psi for the condenser, 3 psi for the control valve and 2 psi for line loss). The stage pressure drop is set based on an expected actual tray pressure drop divided by the design tray efficiency. For most columns an average tray pressure drop is less than 0.12 psi. For low pressure columns, such as this one, the average tray pressure drop may be a little higher, perhaps 0.14. The selected tray pressure drop needs to be divided by the design tray efficiency. For this service an efficiency of 75 to 80% has typically been used. Using a pressure drop of 0.18 psi/theoretical stage is normally reasonable.
The receiver temperature needs to be reasonably above the cooling medium temperature. For naphtha splitters a receiver temperature from 120 to 160 ºF is typical. The column bottoms temperature is expected to be in the range of 300 to 350 ºF. The top tray temperature is expected to be 30 to 50 ºF hotter than the receiver temperature. Estimated temperature values for the receiver, top stage and bottom stage are normally provided. The simulator will use linear interpolation to supply initial temperature estimates for the remaining stages.
The initial vapor rate from each stage is typically set by the simulator using the provided estimated product and reflux rates plus the feed vapor rate. UniSim allows for specification of a reflux to feed ration. Aspen Plus does not allow the specification of reflux to feed ratios a hard value for the reflux rate is frequently used. For this column an expected reflux to feed ratio of between 0.5 and 1.0 can be used to set the initial reflux rate specification.
The initial distillate rate is set based on the assumption that the distillate will be all the C6 minus material that is in the feed.
All of the above values are based on experience or shortcut methods. It is expected that these values can be improved as experience is gained with each column.
5. Understanding the Column Behavior
5.1 Stage Profile
After the execution of the stage to stage model, a review of the stage profile is appropriate. Hopefully the model has converged but review of non converged column stage profile data frequently still provides some value. Table 1 shows the stage profile report of a converged model for a naphtha splitter. Reviewing this report allows confirmation that the stage pressures were correctly set and that the top and bottom temperatures are acceptable for the desired condensing and reboiling mediums. Review of the L/V ratios can help provide some understanding of the column. Fractionation is controlled by the fraction of feed that goes out each product, the alpha values of the key components and the L/V ratios. The closer the L/V ratio is to 1.0, the better the fractionation. Most columns that are providing reasonable component separation will have L/V ratios that are from 0.25 to 4.0. L/V ratios outside of this range are normally only appropriate when a very sloppy separation is desired. L/V ratios that are between 0.9 and 1.1 provide an extremely good component separation for that column section. Some column sections have L/V ratios in this range not out of a need for very good component separation but as a result of very low net material flow, │L-V│. This net material flow is constant for each column section. As such it can be used to identify the feed and product locations and amounts.
| Table 1 Stage Profile | Table 1 Stage Profile | Table 1 Stage Profile | Table 1 Stage Profile | Table 1 Stage Profile | Table 1 Stage Profile |
|---|---|---|---|---|---|
| Stage | Temperature | Pressure | Heat duty | Liquid flow | Vapor flow |
| F | psi | MMBtu/hr | lbmol/hr | lbmol/hr | |
| 1 | 137 | 28.7 | -37.94 | 2087 | 0 |
| 2 | 191 | 33.7 | 0 | 2238 | 2866 |
| 3 | 206 | 33.9 | 0 | 2242 | 3017 |
| 4 | 213 | 34.0 | 0 | 2233 | 3021 |
| 5 | 217 | 34.2 | 0 | 2220 | 3012 |
| 6 | 220 | 34.3 | 0 | 2206 | 2999 |
| 7 | 223 | 34.5 | 0 | 2191 | 2985 |
| 8 | 226 | 34.6 | 0 | 2172 | 2970 |
| 9 | 229 | 34.8 | 0 | 2141 | 2950 |
| 10 | 233 | 34.9 | 0 | 2084 | 2920 |
| Feed 11 | 241 | 35.1 | 0 | 3979 | 2817 |
| 12 | 249 | 35.2 | 0 | 4034 | 2970 |
| 13 | 252 | 35.4 | 0 | 4057 | 3025 |
| 14 | 255 | 35.5 | 0 | 4070 | 3049 |
| 15 | 257 | 35.7 | 0 | 4078 | 3061 |
| 16 | 258 | 35.8 | 0 | 4084 | 3069 |
| 17 | 259 | 36.0 | 0 | 4089 | 3075 |
| 18 | 260 | 36.1 | 0 | 4094 | 3081 |
| 19 | 261 | 36.3 | 0 | 4098 | 3085 |
| 20 | 262 | 36.4 | 0 | 4101 | 3089 |
| 21 | 263 | 36.6 | 0 | 4105 | 3093 |
| 22 | 264 | 36.7 | 0 | 4108 | 3096 |
| 23 | 265 | 36.9 | 0 | 4111 | 3099 |
| 24 | 265 | 37.0 | 0 | 4114 | 3102 |
| 25 | 266 | 37.2 | 0 | 4117 | 3105 |
| 26 | 267 | 37.3 | 0 | 4120 | 3108 |
| 27 | 267 | 37.5 | 0 | 4123 | 3111 |
| 28 | 268 | 37.6 | 0 | 4125 | 3114 |
| 29 | 269 | 37.8 | 0 | 4128 | 3117 |
| 30 | 269 | 37.9 | 0 | 4131 | 3119 |
| 31 | 270 | 38.1 | 0 | 4134 | 3122 |
| 32 | 270 | 38.2 | 0 | 4136 | 3125 |
| 33 | 271 | 38.4 | 0 | 4139 | 3128 |
| 34 | 272 | 38.5 | 0 | 4142 | 3130 |
| 35 | 272 | 38.7 | 0 | 4145 | 3133 |
| 36 | 273 | 38.8 | 0 | 4147 | 3136 |
| 37 | 273 | 39.0 | 0 | 4150 | 3139 |
| 38 | 274 | 39.1 | 0 | 4153 | 3141 |
| 39 | 275 | 39.3 | 0 | 4155 | 3144 |
| 40 | 275 | 39.4 | 0 | 4157 | 3146 |
| 41 | 276 | 39.6 | 0 | 4160 | 3149 |
| 42 | 277 | 39.7 | 0 | 4161 | 3151 |
| 43 | 277 | 39.9 | 0 | 4163 | 3153 |
| 44 | 278 | 40.0 | 0 | 4162 | 3154 |
| 45 | 279 | 40.2 | 0 | 4160 | 3154 |
| 46 | 280 | 40.3 | 0 | 4154 | 3151 |
| 47 | 282 | 40.5 | 0 | 4138 | 3145 |
| 48 | 285 | 40.6 | 0 | 4104 | 3129 |
| 49 | 290 | 40.8 | 0 | 4033 | 3095 |
| 50 | 300 | 40.9 | 40.58 | 1009 | 3024 |
Table 2 shows a column stage profile that includes the feeds and products. Review of this information will insure that the feeds and products are going to and coming from the desired stages. For this column the vapor part of the feed, 45.81 lb-mol/hr was fed to stage 10 and the liquid part of the feed, 1741.88 lb-mol/hr was fed to stage 11. This information allows a check on feed thermal condition. The vaporization of the feed can also be seen in the change in the liquid and vapor rates listed in the profile of Table 1.
| Table 2 Feeds and Products | Table 2 Feeds and Products | Table 2 Feeds and Products | Table 2 Feeds and Products | Table 2 Feeds and Products | Table 2 Feeds and Products | Table 2 Feeds and Products |
|---|---|---|---|---|---|---|
| Stage | Temperature | Liquid feed | Vapor feed | Mixed feed | Liquid product | Vapor product |
| F | lbmol/hr | lbmol/hr | lbmol/hr | lbmol/hr | lbmol/hr | |
| 1 | 137.3 | 0.00 | 0.00 | 0.00 | 778.78 | 0.00 |
| 2 | 191.2 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 3 | 205.8 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 4 | 212.7 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 5 | 217.0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 6 | 220.3 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 7 | 223.2 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 8 | 226.0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 9 | 229.1 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 10 | 233.4 | 0.00 | 45.81 | 0.00 | 0.00 | 0.00 |
| 11 | 241.0 | 1741.88 | 0.00 | 0.00 | 0.00 | 0.00 |
| 12 | 248.5 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 13 | 252.4 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 14 | 254.9 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 15 | 256.7 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 16 | 258.1 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 17 | 259.3 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 18 | 260.4 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 19 | 261.3 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 20 | 262.2 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 21 | 263.0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 22 | 263.7 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 23 | 264.5 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 24 | 265.2 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 25 | 265.9 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 26 | 266.6 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 27 | 267.2 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 28 | 267.9 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 29 | 268.5 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 30 | 269.2 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 31 | 269.8 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 32 | 270.4 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 33 | 271.1 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 34 | 271.7 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 35 | 272.3 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 36 | 272.9 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 37 | 273.4 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 38 | 274.0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 39 | 274.6 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 40 | 275.2 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 41 | 275.8 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 42 | 276.5 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 43 | 277.2 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 44 | 278.0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 45 | 278.9 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 46 | 280.2 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 47 | 282.0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 48 | 285.1 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 49 | 290.4 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
| 50 | 300.5 | 0.00 | 0.00 | 0.00 | 1008.91 | 0.00 |
5.2 Component Recoveries
A review of the component recoveries into each of the products provides fundamental understanding about the column behavior. Component recoveries can be displayed as the fraction recovery in each product stream and the absolute component rates in the each feed and each product. Both of these have significant value and should be reviewed. Table 2 shows these values for a naphtha splitter.
For this sample a review of the absolute component rates in each product stream allows the following to be noted.
- The light key component is clearly CH with 6.41 mol/h going into the bottoms stream. A significant amount of Bz is in the feed but only 0.02 mol/h go into the bottoms.
- The choice of the proper heavy key component is not clear from the absolute component rates. Significant amount of most of the undersireable C7 paraffins are present in the overhead.
For this sample the component recoveries allow the following items to be noted.
- Two of the C7 paraffin’s are actually lighter than the light key, CH. Recoveries into the distillate of 22DMP and 24DMP are 0.9845 and 0.9919 respectively whereas CH recovery into the distillate is 0.9240. In addition the component 224MB has a recovery into the distillate of 0.8884 which is very close to that of As such 22DMP, 24DMP and 224MB should not be included in a group heavy key specification. Including these light components in a heavy group can easily result in convergence difficulty as they can not be forced out the bottom of the column while the light key component goes out the overhead.
- The choice of a single C7 paraffin heavy key component is still not absolutely clear. However, as 0.1587 fraction of 2MH is recovered into the distillate and this is the largest contributor of group heavy key material into the distillate, 2MH should be a reasonable choice for the single component heavy key.
| Table 3 Component Recoveries | Table 3 Component Recoveries | Table 3 Component Recoveries | Table 3 Component Recoveries | Table 3 Component Recoveries | Table 3 Component Recoveries | Table 3 Component Recoveries |
|---|---|---|---|---|---|---|
| Fraction | Fraction | Molar Flow Rate | Molar Flow Rate | Molar Flow Rate | ||
| Component | Distillate | Bottoms | Feed | Distillate | Bottoms | |
| C2 | 1.0000 | 0.0000 | 1.69 | 1.69 | 0.00 | |
| C3 | 1.0000 | 0.0000 | 15.33 | 15.33 | 0.00 | |
| NC4 | 1.0000 | 0.0000 | 50.31 | 50.31 | 0.00 | |
| IC5 | 1.0000 | 0.0000 | 99.57 | 99.57 | 0.00 | |
| NC5 | 1.0000 | 0.0000 | 125.81 | 125.81 | 0.00 | |
| CP | 1.0000 | 0.0000 | 15.67 | 15.67 | 0.00 | |
| 22DMB | 1.0000 | 0.0000 | 5.69 | 5.69 | 0.00 | |
| 23DMB | 1.0000 | 0.0000 | 10.98 | 10.98 | 0.00 | |
| 2MP | 1.0000 | 0.0000 | 75.13 | 75.13 | 0.00 | |
| 3MP | 1.0000 | 0.0000 | 43.55 | 43.55 | 0.00 | |
| NC6 | 1.0000 | 0.0000 | 102.59 | 102.59 | 0.00 | |
| MCP | 1.0000 | 0.0000 | 61.06 | 61.06 | 0.00 | |
| CH | 0.9240 | 0.0760 | 84.36 | 77.95 | 6.41 | |
| BZ | 0.9996 | 0.0004 | 55.83 | 55.81 | 0.02 | |
| 3EP | 0.0459 | 0.9541 | 1.86 | 0.09 | 1.77 | |
| 33DMP | 0.3064 | 0.6936 | 2.19 | 0.67 | 1.52 | |
| 24DMP | 0.9845 | 0.0155 | 5.74 | 5.65 | 0.09 | |
| 23DMP | 0.1225 | 0.8775 | 9.95 | 1.22 | 8.73 | |
| 22DMP | 0.9919 | 0.0081 | 3.37 | 3.34 | 0.03 | |
| 3MH | 0.0839 | 0.9161 | 39.81 | 3.34 | 36.47 | |
| 2MH | 0.1587 | 0.8413 | 36.77 | 5.84 | 30.93 | |
| 223MB | 0.8884 | 0.1116 | 1.00 | 0.89 | 0.11 | |
| NC7 | 0.0124 | 0.9876 | 93.29 | 1.15 | 92.14 | |
| ECP | 0.0025 | 0.9975 | 7.74 | 0.02 | 7.72 | |
| 11DMCP | 0.1640 | 0.8360 | 21.87 | 3.59 | 18.28 | |
| 1T2DMCP | 0.0563 | 0.9437 | 98.12 | 5.52 | 92.60 | |
| 1T3DMCP | 0.0576 | 0.9424 | 98.12 | 5.65 | 92.47 | |
| TOL | 0.0005 | 0.9995 | 105.67 | 0.05 | 105.62 | |
| 3MHEP | 0.0000 | 1.0000 | 26.48 | 0.00 | 26.48 | |
| 2MHEP | 0.0139 | 0.9861 | 52.83 | 0.74 | 52.09 | |
| 33MHEP | 0.0003 | 0.9997 | 52.83 | 0.01 | 52.82 | |
| NC8 | 0.0000 | 1.0000 | 69.70 | 0.00 | 69.70 | |
| 11MCP | 0.0000 | 1.0000 | 21.08 | 0.00 | 21.08 | |
| ECH | 0.0000 | 1.0000 | 59.95 | 0.00 | 59.95 | |
| EBZ | 0.0000 | 1.0000 | 12.58 | 0.00 | 12.58 | |
| P-X | 0.0000 | 1.0000 | 33.91 | 0.00 | 33.91 | |
| M-X | 0.0000 | 1.0000 | 36.78 | 0.00 | 36.78 | |
| O-X | 0.0000 | 1.0000 | 18.47 | 0.00 | 18.47 | |
| 2244MP | 0.0000 | 1.0000 | 50.87 | 0.00 | 50.87 | |
| NC9 | 0.0000 | 1.0000 | 29.13 | 0.00 | 29.13 | |
| NBCP | 0.0000 | 1.0000 | 8.84 | 0.00 | 8.84 | |
| NPCH | 0.0000 | 1.0000 | 22.49 | 0.00 | 22.49 | |
| CUMENE | 0.0000 | 1.0000 | 1.54 | 0.00 | 1.54 | |
| NPBZ | 0.0000 | 1.0000 | 1.69 | 0.00 | 1.69 | |
| 1M2EBZ | 0.0000 | 1.0000 | 0.14 | 0.00 | 0.14 | |
| 1M3EBZ | 0.0000 | 1.0000 | 3.37 | 0.00 | 3.37 | |
| 1M4EBZ | 0.0000 | 1.0000 | 0.56 | 0.00 | 0.56 | |
| 135MBZ | 0.0000 | 1.0000 | 1.83 | 0.00 | 1.83 | |
| 124MBZ | 0.0000 | 1.0000 | 1.69 | 0.00 | 1.69 | |
| 123MBZ | 0.0000 | 1.0000 | 0.14 | 0.00 | 0.14 | |
| 4MC9 | 0.0000 | 1.0000 | 7.48 | 0.00 | 7.48 | |
| NBCH | 0.0000 | 1.0000 | 0.24 | 0.00 | 0.24 |
5.3 Relative Volatility Data
Relative volatility (alpha) data are fundamental fractionation properties. The relative volatility is the ratio of the vapor liquid equilibrium K value (y/x) of one component to that of a base component. The recoveries of C7 paraffins in the overhead can be analyzed by their relative volatility to In most cases a heavy component is chosen as the base component. However, for this problem CH, a light component is used as the base component as it is clearly the light key component and there are multiple possible heavy key components of interest. As the light key is the base component, the smaller the alpha value is the easier the fractionation. It is convenient to review this data via a plot as in Figure 1. Review of this plot allows for the following issues.
- Two of the components, 22DMP and 24DMP, are lighter than
- One component, 223MB, boils close to CH (there is an azeotrope at stage 9).
- All of the C7 paraffins have alpha value curves that are parallel to each other and are fairly constant over the column.
Figure 1 Alpha Values versus Stage Number
The Fenske equation allows the calculation of the minimum number of stages. Identification of a single component for the light key and a single component for the heavy key is required. For these components a single alpha value is needed as are their concentrations or flows in the distillate and bottoms. For the sample problem the light key is CH and the heavy key and base component is 2MH. Rather than a more complex average it is sufficient to use the alpha value at the middle stage, stage 26. Based on the data in Figure 1 the alpha value is 1/0.845 or 1.183. An inverse of the alpha value from Figure 1 is needed as the heavy key needs to be the base.
Fenske Equation
For this simulation the number of stages, N, was 50 so the ratio of N/Nmin is 2.02. For most optimized columns this value is between 2.0 and 3.0. Typically a value of 2.5 is justified. Based on this result, additional stages are likely justified. Optimization of the number of stages should be based on the results of the simulator study, not the results of this shortcut method. Procedure PCS-13 discusses methods for this optimization.
It is possible to use shortcut methods to predict the behavior of this column with alternate number of stages. The Gilliland correlation can be used to predict the minimum reflux using values from the simulation and the Fenske equation. The Eduljee curve fit to the Gilliland is effective for this effort. With the minimum trays and minimum reflux values available the Gilliland correlation can now be used to prepare a reflux versus stage curve. In addition condenser and reboiler duties can be estimated for any stage value by using the condenser and reboiler duties from the simulation. While this may add understanding about the column behavior all final optimization work should be based on simulator results rather than shortcut method results.
Gilliland Correlation
Eduljee Equation
© 2026 Inflection Point Engineering, LLC. All rights reserved. The content of this page — including calculation methods, reference data, written analysis, interactive tools, and source code — is the intellectual property of Inflection Point Engineering, LLC and is protected under applicable copyright, trademark, and trade secret laws. Unauthorized reproduction, redistribution, modification, or derivative use in whole or in part is prohibited without prior written consent.
Disclaimer. This material is provided for informational and educational purposes only and does not constitute professional engineering advice. Calculations, reference data, and methodologies are based on published standards and accepted engineering practice but are not a substitute for engineering judgment, site-specific analysis, or review by a licensed Professional Engineer. Inflection Point Engineering, LLC makes no warranties, express or implied, regarding the accuracy, completeness, or fitness for a particular purpose of any content presented here, and shall not be liable for any direct, indirect, incidental, or consequential damages arising from its use. Users assume all risk associated with applying this content to real-world design, operations, or decisions.
© 2026 Inflection Point Engineering, LLC. All rights reserved.