Cdu 3

Chapter 11
Petroleum, Complex-Mixture
Fractionation, Gas Processing,
11
Dehydration, Hydrocarbon
Absorption and Stripping
Part 2: Fractionation
The demand for crude oil is often market-driven, as The scale of petroleum fractionation/distillation is
recently observed in the world economic recession generally large, involving atmospheric distillation of
where the price of U.S crude oil fell from U.S. $140 to crude oil, vacuum distillation of bottoms residue, main
$36/barrel by the end of 2008. Crude oil distillation is fractionation of gaseous effluent from catalytic cracking,
a large energy consumer and a large waste generator, so and main distillation of effluent from thermal coking of
refiners need to balance the economics of processing various petroleum fractions. They use heavy equipment
crude oil to viable fractions. The thermodynamics of which consumes large quantities of energy. Figure 11-1
multicomponent distillation, as discussed earlier in the shows a flow diagram of a typical refinery distillation
book, apply to petroleum, synthetic crude oil and other unit.
complex mixtures. Various advanced integrated and
distributed distillation methods can be employed to
minimize waste and thus optimize performance oil 11.1 Characterization of Petroleum
[295]. and Petroleum Fractions
The processing of crude oil requires special consider-
ation for the following reasons: The primary raw material for all petroleum processes is
The feed stocks are generally complex mixtures of crude oil, a naturally occurring liquid produced from
many different types of hydrocarbons along with in- wells drilled into the earth. The well depth and tech-
organic and organic compounds. The number of niques used to produce the oil vary widely around the
carbon atoms in the component molecules ranges world, as does the nature and chemical composition of
from 1 to more than 50, so the compounds have the crude oil itself.
boiling points from –260 F (–162 C) to more than The elementary composition of crude oil will, how-
1000 F (538 C). Further, the number of different ever, fall within the following ranges:
compounds of similar volatility increases very rapidly
with corresponding increase in boiling point, e.g. 16 Element % by weight
of the 18 octane isomers boil within a range of only
Carbon 84–87
22 F (12 C).
Hydrogen 11–14
The fractionation products of these complex mixtures
Sulfur 0.06–8
are often complex mixtures themselves, and their
Nitrogen 0.02–1.7
character and yield varies widely, and often depends
Oxygen 0.68–1.82
upon the source of the feedstock. Crude oils from the
Metals 0–0.14
same locality sometimes exhibit marked variations.
Ludwig’s Applied Process Design for Chemical and Petrochemical Plants; ISBN: 9780750683661
Copyright Ó 2010 Elsevier Inc. All rights of reproduction, in any form, reserved.
CHAPTER 11 Petroleum, Complex-Mixture Fractionation, Gas Processing, Dehydration
Crude oil is classified according to the family of hydro- 11.1.1 Crude Oil Properties
carbons it contains, as paraffin base, naphthene base, ar-
omatic base or asphalt base. The refining process is Crude oil is a mixture of thousands of hydrocarbons; the
dependent on the crude oil base. The relative values of American Petroleum Institute (API) Project 6 has
crude oil depend on the processing costs and the market isolated over 16,000 distinct hydrocarbon compounds
values of the associated products. from a single sample. The tests outlined in the following
LPG and gas Refinery fuel gas

Vapor
Crude unit recovery LPG
Stabilizer Straight-run-gasoline
section
Regular gasoline
Naphtha Reformate
Hydro
Reforming
treating Premium gasoline
Crude unit
Middle distillates Hydro Solvents
atmospheric
Crude treating
tower
oil section Aviation fules
Heavy atmospheric gas oil
Catalytic
cracking Gasoline Diesels
Vacuum gas oil Heating oils
Treating and blending

Fractionator bottoms Greases
Crude unit
Asphalts
vacuum
Solvent Gasoline, naphtha
section
extraction and middle Industrial fuels
Lube
base Hydro distillates
cracking Refinery fuel oil
stocks
Lube oils
Propane Solvent
Waxes
deasphalter dewaxing
Gasoline, naphtha
and middle distillates
Visbreaker
Fuel oil
Asphalt
Delayed
coker
Gasoline, naphtha and middle distillates
Coke
Figure 11-1 A flow diagram of a refinery distillation unit.
270
Petroleum, Complex-Mixture Fractionation, Gas Processing, Dehydration CHAPTER 11
sections are employed to characterize crude oils as po-

Table 11-1 Specific gravities of some crude oils
tential refinery feeds.
Country Crude Sp. gravity: d15
4
11.1.2 Gravity, API Indonesia Minas 0.845
The density of petroleum oil is expressed in the United Iraq Kirkuk 0.845
States as American Petroleum Institute (API) gravity, Iran Cyrus 0.940
where API gravity is related to specific gravity as follows:
Kuwait Kuwait 0.870
Degrees API ¼ ð141:5=Sp: GrÞ 131:5 (11-1)
Nigeria Bonny Light 0.837
Both specific gravity and API gravity are expressed Saudi Arabia Arabian Light 0.858
relative to the density of water at 60 F (15.9 C). The
Venezuela Boscan 1.000
expression for API is non-linear, and so specific gravities
must be used when calculating the gravity for blends.
The API gravity is inversely related to the specific
gravity, so a higher value shows a less dense material Table 11-1 shows typical specific gravities of some
(e.g. the API of water is 10.0). API gravity is preferred crudes.
to specific gravity as a measure of density because the
specific gravities of various crude oils lie very close to
each other and the API gravity alone can differentiate
between them. API gravity is determined with a hy-
11.1.3 Boiling Point Range
drometer and corrected to 60 F (15.9 C).
The boiling point range for a crude oil is determined
In the case of a mixture or blend of various fractions,
with a true boiling point (TBP) distillation test. This
the API gravity is determined as the sum of the in-
test is performed in an apparatus with sufficient frac-
dividual API gravities multiplied by the corresponding
tionation to recover samples of the various products
weight fraction in the mixture.
that can be distilled from crude oil. Results are
reported at one atmosphere. It is impossible to de-
ðAPIÞmix ¼ ðAPIÞ1 W1 þ ðAPIÞ2 W2 þ ðAPIÞ3 W3 þ .. termine the complete boiling point range for most
(11-2) crude oils, as they thermally degrade above tempera-
tures around 650 F (344 C) and a pressure of one at-
where mosphere. Therefore the pressure is gradually reduced
during a TBP distillation test to approximately 40 mm
API1, API2, API3, ¼ the respective API for each
Hg for the heaviest fraction. These conditions are
components
equivalent to a temperature at atmospheric pressure of
W1, W2, W3, ¼ the respective weight fractions
900–950 F (483–510 C). Nelson and others [297]
The specific gravity of crudes can be categorized into four state that the maximum boiling point for crude oil
groups [296]: mixtures is greater than 1400 F (760 C).

Light crude oil, Sp Gr: d15
4
* <0.825 11.1.4 Characterization Factor
15
Medium crude oil, Sp Gr: d4 <0.875
Watson Factor (KW)
Heavy crude oil, Sp Gr: d15
4 <1.000 Watson, Nelson and Murphy [298] proposed the fol-

Extra heavy crude oil, Sp Gr: d15
4 >1.000 lowing characterization factor:
*d15
4 represents density at 15 C with respect to that of
water at 4 C.
K ¼ T0:333
B =Sp: Gr (11-3)

ðMolal average BPt: in RÞ ðVolume average BPt: in RÞ
TB ¼
2
271
where K 10 indicates an aromatic material whereas K ¼ 15

K ¼ characterization factor indicates a paraffinic material. For most crude oils,
TB ¼ the mean average boiling point of the sample, R 10:5 < K < 12:5. Figure 11-2 is a nomograph for es-
Sp. Gr. ¼ specific gravity at 60 F relative to water. timating the K value from the mean average boiling
point, the API gravity and an ASTM D-86 distillation
This characterization factor, which is a rough indicator of the petroleum fraction. The characterization
of hydrocarbon class (or composition) is related to factor, KW is defined in the example given on the
molecular weight, viscosity, critical temperature, etc. figure.
Figure 11-2 Specific Gravity of Petroleum Fractions. Used by permission, Gas Processors Suppliers Association Book Data, 12th Ed., v.1
and 2., (2004).
272
The Universal Oil Product Characterization quantity of lubricating oil fraction that may be recovered
factor, KUOP as product. It can also be used to establish the yields
from coking processes and the coke deposited on the
KUOP is widely used since it expresses quantitatively the
catalyst in catalytic cracking. The higher the carbon
variations in physical properties that vary with the
residue for a crude oil, the lower its value as a refinery
character of the stock. It is defined as the cube root of
feed stock.
the molal average boiling point, T R divided by the
specific gravity at 15.6 C.
11.1.6 Nitrogen Content, wt%
T0:333
KUOP ¼ (11-4)
d Nitrogen compounds in crude oils reduce the activity of
cracking catalysts. High nitrogen content crude oil may
Where T is the average boiling point in R and d is the require special processing to remove the nitrogen. Most
specific gravity at 15.6/15.6 C. of the nitrogen in crude oil is contained in the highest
The volume average boiling point (VABP), obtained boiling fractions.
either from True Boiling Point (TBP) distillation or
American Society for Testing and Materials (ASTM)
distillation, replaces T in the expression for KUOP. 11.1.7 Sulfur Content, wt%
In the case of a crude oil subjected to TBP distillation
(given as volume), Sulfur content and API gravity have the greatest influence
on the value of crude oil. It varies from trace levels to >5
T20 þ T50 þ T80 wt%. High sulfur oils are referred to as ‘‘sour’’ crude oils
Volume average boiling point: T ¼
3 and require more extensive processing than do low sulfur
or in the case of a petroleum cut subjected to ASTM or ‘‘sweet’’ oils. Generally, the lower the API gravity, the
distillation: higher the sulfur content, although there are exceptions
to this rule. Further processing for sour crude oils involves
T10 þ 2T50 þ T90 hydrogen treating processes to convert the sulfur to H2S,
Volume average boiling point: T ¼
4 which can the be removed.
where T20, T50, T80, T90 are the temperatures at
which 20, 50, 80 and 90 per cent by volume of a sample 11.1.8 Salt Content, pounds/1000
distils out. barrels
It is found that highly paraffinic oils have KUOP values
from 12.5–13.0 and cyclic (naphthenic/aromatic) oils All crude oil contains some inorganic salts, generally in
have KUOP values from 10.5–12.5. Thus KUOP values can the form of chlorides. If the salt content exceeds
differentiate between highly paraffinic and highly aro- 10 pounds/1000 barrels, it is necessary to remove them
matic stocks but have problems in classifying naphthenic prior to distillation to reduce corrosion in the distillation
stocks from aromatic ones, since their KUOP values column and its overhead condensers.
overlap.
Note: In the case of a mix, the characterization factor
is determined as: 11.1.9 Metals, parts/million (ppm)
by weight
Kmix ¼ K1 W1 þ K2 W2 þ K3 W3 þ .: (11-5)
Many metals occur in crude oil, the most common being
where nickel and vanadium. Metals tend to concentrate in the
heavier fractions where they act as poisons in the catalytic
K1, K2, K3 ¼ the respective characterization factors for and hydrogen cracking processes. Other metals include
different components copper, zinc, iron and arsenic.
W1, W2, W3 ¼ the respective weight fractions
11.1.10 Pour Point ( F or C)

11.1.5 Carbon Residue, wt%
The pour point (ASTM D97) is used to determine the
This property is determined by reducing a sample to waxiness of petroleum oil. Waxes are paraffins with long
elemental carbon (coke) in the absence of air. It is side chains, or long side chains attached to other
a measure of the asphalt content of the crude oil and the hydrocarbon structures. A waxy crude oil has a high pour
273
hydrocarbon yields, (C1–C5), and a boiling point curve

Table 11-2 Pour points of some crude oils
are the both properties of the oil given by an assay.
Country Crude Pour point, 8C
Algeria Hassid Messaoud 60
11.2.2 Fractional properties
Kuwait Kuwait 42 These are the boiling point or cut-point range and yield
on crude, research octanes for gasoline fractions, smoke
Libya Dahra 1 and freeze points for kerosenes, naphthene and aro-
Venezuela Boscan þ15
matic contents for reforming stock, and cetane number
for diesel fuels.
Crude oil assay data change over time for any oil. This
is caused by two factors; firstly, the crude oil itself
point; i.e. it solidifies at a temperature above room
changes in composition, and secondly, a crude oil sample
temperature. Conversely, a low pour point indicates an
represents a mixture of the crude oil produced in various
absence of paraffinic and/or a presence of aromatic
wells and fields so, as new one are brought on line, the
compounds, as these tend to lower the pour point. A
composition of the mixture changes. It is good practice to
sample of oil is first heated to 115 F (46.4 C) to ensure
request a confirming sample for a crude oil mixture being
complete liquefication. It is then cooled to 90 F
used in any process design/simulation exercise.
(32.6 C). The sample is then placed in a cooling bath
maintained between 15–30 F (8.3–16.7 C) below the
pour point of the oil. The oil is removed from the bath at
5 F (2.8 C) intervals and it tilted to determine whether 11.3 Crude Cutting Analysis
the sample will flow. If no movement is detected within
5 seconds, the solid point has been reached. It is The first seven crude cuts are pure components. They are
defined as the temperature 5 F (2.8 C) above the solid the light ends and consist of H2S, H2, C1, C2, C3, iC4,
point. nC4. The iC5 and nC5 are together and therefore are
The same procedure is used to determine the cloud treated as the eight cut with a TBP end point of 100 F.
point of petroleum oil. This is the temperature at which The sum of light ends through nC5 should be equal to the
a distinct cloudiness or formation of crystals appears in lights vol % distilled at 100 F. However, if the two
the bottom of the test jar. It is typically about 5 F numbers do not agree, then the light ends yields are
(2.8 C) higher than the pour point. Table 11-2 gives pour assumed to be correct and the TBP yields between
points of some typical crude oils. 100–200 F are adjusted to provide a smooth transition
between light ends and the remaining TBP curve.
The ninth cut has a TBP cut point of 100–200 F. Each
11.2 Crude Oil Assay Data successive cut has a TBP end point of 20 F higher than
the preceding component. The 63rd component has end
point of 1200 F. The TBP end point of a crude is difficult
A precise and complete analysis of a crude oil is referred
to estimate with any degree of accuracy in the laboratory
to as crude assay. This is more detailed than a TBP curve.
and is therefore assumed to be 1292 F. The 64th cut is
Crude oil assay data are available for most of the crude
assumed to have a TBP cut range of 1200–1292 F. Tables
oils of the world; some data are company proprietary, but
11-3 and 11-4 show crude assays of Middle East Arab
much is available in open literature. The crude assay
extra light and heavy crudes.
helps to assess the potential sales value of a new oil and
thus plan for its effective utilization.
The following sections cover the information obtained
from an assay. 11.4 Crude Oil Blending
Many refineries process more than one crude oil each
11.2.1 Whole crude oil average being transported to the refinery via pipelines, ships or
properties trucks. Since most refineries do not have adequate crude
oil storage to isolate the various crude oils the storage
Gravity, weight % sulfur, weight % nitrogen, pour point, tanks contain a mixture of the crude oils received by the
viscosities, metal contents, weight percent carbon res- refinery. Generally, crude blending is achieved by mixing
idue, Reid vapor pressure, salt content, flash point, % good quality crudes with heavy low cost crudes to
asphaltene, bottom sediment and water (BS & W), light improve overall profitability for the refinery.
274
Table 11-3 Middle East: Arabian Extra Light 36.7
Crude Data
Sample Date: 01/06/93 Analysis Date:
Petroleum, Complex-Mixture Fractionation, Gas Processing, Dehydration

SG: 0.8386 API Gravity 37.2 Sulfur: 1.38 wt%
Light Hydrocarbon Analysis
wt % lv%
Hydrogen sulfide 0.00 0.00
Methane 0.00 0.00 wt% Iv%
Ethane 0.00 0.00
Propane 0.07 0.12 Total to C3 0.07 0.12
Iso-butane 0.11 0.16
N-butane 0.77 1.11 Total C4s 0.88 1.27 Total to C4 0.95 1.39
Iso-pentane 0.70 0.94
N-pentane 1.40 1.86 Total C5s 2.10 2.80 Total to C5 3.05 4.19
Wide Cut Data

TBP Cut Points C C5–95 95–175 C5–149 149–232 332–342 342–369 369–509 509–550 509D 550D
Yield on Crude wt% 7.91 14.36 15-75 16.35 21.50 5.05 20.68 4.46 18.72 14.26
lv% 9.86 16.03 19.88 17.44 21.47 4.82 19.07 3.97 15.93 11.96
SG 0.6729 0.7512 0.7068 0.7859 0.8396 0.8794 0.9094 0.9418 0.9854 0.9998
API Gravity 78.79 56.87 68.71 48.54 37.04 29.41 24.10 18.74 12.10 10.02
Watson K factor 12.54 12.03 12.21 11.96 11.94 11.85 11.91 12.00 11.86 11.85
Sulfur wt% 0.009 0.028 0.019 0.069 0.921 2.095 2.288 2.480 3.140 3.347
Mercaptan Sulfur wt% 0.018 0.29 0.025 0.025 0.026 0.027 0.028 0.029 0.030 0.031
Iso Paraffins wt% 37.79 31.18 34.19 30.57 - - - - - -
Total Paraffins wt% 86.77 64.88 76.45 58.93 - - - - - -
Naphthenes wt% 11.49 19.60 16.15 18.42 - - - - - -
Aromatics wt% 1.74 15.52 7.40 22.65 29.27 35.95 - - - -
Viscosity at 20 C cSt - 0.87 - 1.39 5.07 18.15 109.80 1510.00 79860.00 411300.
30 C cSt - 0.77 - 1.20 3.94 12.39 62.02 647.10 2382.00 106500.
50 C cSt - 0.62 - 0.93 2.59 6.68 24.87 167.00 3365.00 11850.0
80 C cSt - 0.47 - 0.68 1.61 3.38 9.35 39.73 409.50 1099.0
100 C cSt - 0.40 - 0.56 1.25 2.40 5.78 19.86 146.90 343.50
CHAPTER 11
Cloud Point C - 80.5 - 56.7 12.0 12.0 14.9 21.7 - -
Freeze Point C - 77.3 96.5 53.5 9.0 14.3 - - - -
Pour Point C - 93.1 - 59.5 12.3 11.5 25.7 22.1 7.8 2.0
275
Aniline Point C - 49.3 - 57.1 70.6 76.2 83.1 92.3 102.1 105.1
Smoke Point mm - - - 24.7 19.6 - - - - -
Liminometer - - - 55.3 41.9 - - - - -
(Continued)
Table 11-3 Middle East: Arabian Extra Light 36.7 (Cont’d)
CHAPTER 11
276
Wide Cut Data

Cetane Index - 35.1 - 44.1 53.1 47.4 35.9 9.5 -
Heat Value kcal/kg 10578 10469 10542 10396 10234 9960 9858 9743 9563 9507
RONC 56.27 39.53 46.37 18.56 - - - - - -

RON þ 0.15 g Pb/l 71.69 55.15 61.78 37.02 - - - - - -

MONC 58.21 41.21 47.80 19.07 - - - - - -
MON þ 0.15 g Pb/l 69.95 54.82 61.40 34.83 - - - - - -
Nitrogen wt% 0.0 0.00 0.00 0.00 0.00 0.02 0.05 0.10 0.18 0.21
Vanadium ppm - - - - - - - 0.14 21.27 27.88
Nickel ppm - - - - - 0.00 0.00 0.07 3.81 4.89
RI at 70 C 1.3974 1.3974 - 1.4165 1.4470 1.4696 1.4878 1.5077 1.5368 1.5459
Concarbon wt% - - - - - - 0.18 1.72 14.83 18.93
Asphaltenes wt% - - - - - - - 0.00 1.90 2.49
BMCI 7.55 17.73 14.28 20.89 27.78 37.16 42.90 50.02 56.25 59.20
wt% lv% wt% lv% wt% lv%

Cyclopentane 0.09 0.09 Total CC5 Total to CC5 0.09 0.09
C6 iso paraffins 1.21 1.53
C6 n-paraffins 1.58 1.99
C6 naphthenes 0.49 0.53
Benzene 0.09 0.08 0.07 0.12
Total C6s 3.36 4.13 Total to C6 3.44 4.23

Toluene 0.47 0.45

C8 aromatics 0.71 0.68

Table 11-3 Middle East: Arabian Extra Light 36.7 (Cont’d)

wt% lv% wt% lv% wt% lv%


Crude sulfur % wt 1.320 Crude Sg Sg 0.8410 Crude API 36.7000

Crude viscosity 20 cst 5.990 Crude viscosity 30 cst 4.8100 Crude viscosity 40 cst 3.9600
Crude Pour Point C 39.000 Crude Acidity mg koh/g 0.0500 Crude Concarbon % wt 2.7000
Crude Asphaltenes %wt 0.3500 Crude Vanadium ppm 4.0000 Crude Nickel ppm 2.0000
Crude salt content lb/1000bb 2.1000 Crude water content % vol 0.0500 N-paraffins C5-95 % wt 49.5000
N-paraffins 95-175 % wt. 30.6000 N-paraffins C5-149 % wt 40.3000 Acidity 149-232 mg koh/g 0.0090
Hydrogen 149-232 % wt 14.0500 Naphthalenes 149-232 % vol 0.03200 Color stab 149-232 1.0000
Wax content 232-342 % wt 2.2000 Acidity 232-342 mg koh/g 0.01000 Color stab 232-342 1.0000
Wax content 342-369 % wt 14.9000 Wax content 369-509 % wt 12.300 Basic N2 369-509 ppm 125.0000
Organic O2 369-509 % wt 0.0500 Acidity 369-509 mg koh/g 0.0200 Wax content 509-550 % wt 11.1000
Basic N2 509-550 ppm 225.0000 Acidity 509-550 mg koh/g 0.0600 Wax content 342þ % wt 11.5000
Wax content 369þ % wt 11.0000 Wax Melt Pt. 369þ C 53.000 Acidity 369þ mg koh/g 0.05000
Xylene Equiv. 369þ 12.5000 Resins A 369þ % wt 8.000 Resins B 369þ % wt 1.0000
Pen At 25c 509þ m 0.0500 Softening Point 509þ C 25.0000 Wax content 509þ % wt 9.5000
Wax Melt Pt. 509þ C 62.0000 Pen At 25c 550þ m 0.0500 Softening Point 550þ C 25.0000
Xylene Equiv 550þ 10.0000 Resins A 550þ % wt 12.0000 Resins B 550þ % wt 4.0000
Mcpentene 15-95 % wt 0.2600 Hexane 15-95 % wt 0.19000 C6 – iso-para 15-95 % wt 1.25000
22-dmthlbutane 15-95 % wt 0.0100 23-dmthlbutane 15-95 % wt 0.0800 2-mthlpentane 15-95 % wt 0.68000
3-mthlpentane 15-95 % wt 0.4800 N-heptane 15-95 % wt 0.66000 Mthylcyhexane 15-95 % wt 0.10000
CHAPTER 11
277
Table 11-4 Middle East: Arabian Heavy 27.3
CHAPTER 11
Crude Data:
278
Sample Date: 01/08/93 Analysis Date:

SG: 0.88874 API Gravity: 28.0 Sulfur: 2.90 wt%
Light Hydrocarbon Analysis
wt % lv%
Hydrogen sulfide 0.00 0.00

Methane 0.00 0.00 wt% Iv%
Ethane 0.04 0.10
Propane 0.39 0.68 Total to C3 0.43 0.78
Iso-butane 0.19 0.30
N-butane 0.88 1.34 Total C4s 1.07 1.64 Total to C4 1.50 2.42
Iso-pentane 0.62 0.88
N-pentane 1.13 1.59 Total C5s 1.75 2.47 Total to C5 3.25 4.89
Wide Cut Data

Yield on Crude wt% 5.62 9.12 11.21 10.90 15.90 4.05 19.44 5.11 37.01 31.90
lv% 7.46 10.85 14.21 12.34 16.69 4.04 18.65 4.73 31.65 26.92
SG 0.6683 0.7456 0.6997 0.7839 0.8396 0.8794 0.9094 0.9418 0.9854 0.9998
API Gravity 80.22 58.29 70.73 49.02 37.04 29.41 24.10 18.74 12.10 10.02
Watson K factor 12.55 12.12 12.31 12.00 11.94 11.85 11.91 12.00 11.86 11.85
Sulfur wt% 0.002 0.024 0.006 0.153 1.379 2.625 2.931 3.649 5.357 5.630
Mercaptan Sulfur wt% 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001
Iso Paraffins wt% 39.38 32.51 35.86 31.30 - - - - - -
Total Paraffins wt% 88.63 66.79 79.14 57.98 - - - - - -
Naphthenes wt% 10.58 22.37 16.55 23.06 - - - - - -
Aromatics wt% 0.79 10.84 4.31 18.96 30.07 43.15 - - - -
Viscosity at 20 C cSt - 0.87 - 1.40 5.31 19.47 171.900 3284.00 4.0 E þ 07 4.0E þ 08
30 C cSt - 0.77 - 1.22 4.11 13.18 91.71 1253.00 7066000 6.0E þ 07
50 C cSt - 0.62 - 0.95 2.69 7.01 33.69 271.20 413000 2633000
80 C cSt - 0.48 - 0.70 1.66 3.51 11.63 54.43 18280.00 82910.0
100 C cSt - 0.40 - 0.58 1.29 2.48 6.93 25.27 3904.000 14760.0
Cloud Point C - 82.2 - 58.6 15.20 9.0.0 29.2 38.5 - -

Freeze Point C - 778.2 98.3 54.7 11.3 12.0 - - - -
Pour Point C - 82.2 - 58.6 15.3 8.9 26.9 30.5 6.9 0.3
Aniline Point C - 54.8 - 59.7 68.0 71.0 76.5 80.2 85.0 85.8
Smoke Point mm - - - 26.1 - - - - - -
Liminometer - - - 58.7 - - - - - -
Table 11-4 Middle East: Arabian Heavy 27.3 (Cont’d)
Wide Cut Data

Cetane Index - 38.7 - 45.6 51.5 44.8 32.0 6.3 - -
Heat Value kcal/kg 10591 10477 10552 10401 10214 9890 9761 9602 9259 9204

RONC 57.35 36.81 46.42 14.85 - - - - - -
RON þ 0.15 g Pb/l 72.51 52.33 61.62 33.56 - - - - - -
MONC 59.45 38.26 47.92 15.12 - - - - - -
MON þ 0.15 g Pb/l 70.87 51.91 61.51 31.15 - - - - - -
Nitrogen wt% 0.0 0.00 0.00 0.00 0.00 0.02 0.07 0.13 0.43 0.48
Vanadium ppm - - - - - - 0.10 0.61 162.13 188.01
Nickel ppm - - - - - 0.00 0.11 0.70 51.81 60.00
RI at 70 C - 1.424 - 1.4362 1.4596 1.4755 1.4953 1.5169 1.5717 1.5805
Concarbon wt% - - - - - - 0.19 1.60 22.2 25.50
Asphaltenes wt% - - - - - - - 0.00 12.07 14.00
BMCI 8.23 15.00 11.52 19.74 30.31 41.70 49.910 58.10 66.12 68.78
wt% lv% wt% lv% wt% lv %

Cyclopentane 0.05 0.06 Total CC5 Total to CC5 0.05 0.06
Benzene 0.03 0.03

Toluene 0.15 0.16

CHAPTER 11
279
(Continued)
CHAPTER 11
280
Table 11-4 Middle East: Arabian Heavy 27.3 (Cont’d)
wt% lv% wt% lv% wt% lv %



Crude sulfur % wt 2.900 Crude Sg Sg 0.8905 Crude API 27.3000

Crude viscosity 20 cst 47.100 Crude viscosity 30 cst 30.800 Crude viscosity 40 cst 21.300
Crude Pour Point C 51.000 Crude Acidity mg koh/g 0.2300 Crude Concarbon % wt 8.200
Crude Asphaltenes %wt 4.500 Crude Vanadium ppm 60.000 Crude Nickel ppm 19.000
Crude salt content lb/1000bb 2.6000 Crude water content % vol 0.0500 N-paraffins C5-95 % wt 49.7000
N-paraffins 95-175 % wt. 32.2000 N-paraffins C5-149 % wt 41.6000 Acidity 149-232 mg koh/g 0.0340
Hydrogen 149-232 % wt 14.1400 Naphthalenes 149-232 % vol 0.4500 Naphthalenes 232-342 % vol 6.9700
Color Stab 149-232 1.000 Wax content 232-342 % wt 2.9000 Acidity 232-342 mg koh/g 0.1100
Color Stab 232-342 1.000 Wax content 369-509 % wt 13.3000 Wax content 369-509 ppm 10.6000
Basic N2 369-509 ppm 165.000 Organic O2 369-509 % wt 0.0500 Acidity 369-509 mg koh/g 0.2800
Wax content 509-550 % wt 9.10000 Basic N2 509-550 ppm. 365.000 Acidity 509-550 mg koh/g 0.32000
Wax content 342þ % wt 6.500 Wax content 369þ % wt 6.000 Wax Melt Pt 369þ C 56.000
Acidity 369þ mg koh/g 0.2300 Xylene Equiv. 369þ 17.500 Resins A 369þ % wt 10.000
Resins B 369þ % wt 9.000 Pen At 25c 509þ m 0.0076 Softening Point 509þ C 48.000
Wax content 509þ % wt 3.5000 Pen At 25c 550þ m 0.0022 Softening Point 550þ C 62.5000
Xylene Equiv 550þ 17.500 Resins A 550þ % wt 18.000 Resins B 550þ % wt 14.0000
Mcpentene 15-95 % wt 0.1800 Chexane 15-95 % wt 0.1100 C6 – iso-para 15-95 % wt 0.9300
22-dmthlbutane 15-95 % wt 0.0100 23-dmthlbutane 15-95 % wt 0.0700 2-mthlpentane 15-95 % wt 0.4600
3-mthlpentane 15-95 % wt 0.3900 N-heptane 15-95 % wt 3900 Mthylcyhexane 15-95 % wt 0.0600
11.5 Laboratory Testing

of Crude Oils
11.5.1 True Boiling Point (TBP) Curve
The composition of any crude oil sample is approxi-
mated by a true boiling point (TBP) curve. The
method for determining this is a batch distillation
operation using a large number of stages, usually >60,
and high reflux to distillate ratio (>5). The tempera-
ture at any point on the temperature-volumetric yield
curve represents the true boiling point of the hydro-
carbon material present at the given volume per cent
point distilled. TBP distillation curves are generally
found only for the crude and not for petroleum
products. Figure 11-3 shows a typical TBP curve.
Figure 11-3 Atmospheric TBP distillation curve (Source: Kaes, G.
The volatility of crude oil and petroleum fractions is L., Refinery Process Modeling: A Practical Guide to Steady State
characterized in terms of one or more of the laboratory Modeling of Petroleum Processes, Kaes Enterprises, Inc., 2000).
distillation tests that are summarized in Table 11-5,
namely the American Standard Testing Materials
(ASTM) D86 and D1160 tests. The ASTM D86 dis-
tillation is a rapid batch distillation involving the 11.5.2 ASTM D86 Distillation
equivalent of approximately one equilibrium stage and
no reflux except for that caused by heat losses. The D86 distillation uses a simple Engler flask containing
temperatures are reported at a pressure base of 1 at- a calibrated thermometer to measure the temperature of
mosphere (760 mm Hg). For high altitude laboratories the vapor at the inlet to the condensing tube. An inclined
the barometric pressure corrections to the observed brass condenser in a cooling brass is attached, to remove
temperatures are significant and must not be ignored. all distilled vapors; no liquid reflux is returned to
Table 11-6 shows a sample of these temperature the flask. The condenser tube is cooled in an ice water
corrections. bath to maintain the condensing temperature between
Table 11-5 Laboratory Distillation Tests
Test name Reference Main applicability

ASTM (atmosphere) ASTM D 86 Petroleum fractions or products, including gasoline turbine fuels, napthas,
kerosines, gas oils, distillate fuel oils, and solvents that do not tend to
decompose when vaporized at 760 mm Hg
ASTM [vacuum often 10 torr (1.3 kPa)] ASTM D 1160 Heavy petroleum fractions or products that tend to decompose in the ASTM
D86 but can be partially or completely vaporized at a maximum liquid
temperature of 750 F (400 C) at pressures down to 1 torr (0.13 kPa)
TBP [atmospheric or 10 torr (1.3 kPa)] Nelson*, ASTM D Crude oil and petroleum fractions
2892
Simulated TBP (gas chromatography) ASTM D 2887 Crude oil and petroleum fractions
EFV (atmospheric, superatmospheric, or Nelsony Crude oil and petroleum fractions

subatmospheric)
* Nelson, Petroleum Refining Engineering, 4th ed. Mc. Graw-Hill, New York, 1958, pp 95–99.
y
Ibid, pp 104–105.
Source; Perry’s Chemical Engineer’s Handbook, 7th ed., McGraw-Hill, 1997.
281
Table 11-6 Temperature Corrections for Barometric Pressure
Laboratory pressure, mmHg

Observed
temp, 8F 572 597 622 648 673 698 724
50 12 10 8 7 5 4 2
100 13 11 9 8 6 4 2
150 14 12 10 8 6 4 3
200 15 13 11 9 7 5 3
250 16 14 12 10 7 5 3
300 17 15 13 10 8 6 3
350 18 16 13 11 8 6 4
400 19 17 14 12 9 6 4
450 21 18 15 12 9 7 4
500 22 19 16 13 10 7 4
550 23 20 17 14 11 7 4
600 24 21 18 14 11 8 5
650 25 22 18 15 12 8 5
700 26 23 19 16 12 9 5
For use with ASTM D86 Distillation

Temperature corrections are in 8F.
Source: Gerald L. Kees, Refinery Process Modeling: A Practical Guide to Steady
State Modeling of Petroleum Processes, The Athens Printing Co., Athens, Figure 11-4 The True Boiling Point (TBP) distillation apparatus.
Georgia, 2000.
11.5.3 Boiling Points

32 and 40 F (0–4.5 C). The light components that boil
at temperatures lower than this are lost from the distilled Figure 11-5 illustrates the interconversion between
product. ASTM D-86 distillation 10% to 90% slope, and the dif-
100 ml of sample is distilled and the vapor tempera- ferent boiling points used to characterize fractions of
ture against volume recovered is recorded. The initial crude oil. This is done to determine properties such as
boiling point (IBP) is defined as the temperature at which the volumetric average boiling (VARP) point, weight
the first drop of liquid leaves the condenser tube. The average boiling point (WABP), molal average boiling
final boiling point (FBP) or ‘‘end point’’ is the highest point (MABP), mean average boiling point (MeABP) and
temperature recorded during the test. The total volume cubic average boiling point (CABP). On the basis of
of the distillate is recorded as the recovery. Any liquid ASTM D-86 distillation data, the VARP is [290]:
left in the still after the end point temperature is
recorded is cooled and measured as the residue. The VARP ¼ ðt10 þ t30 þ t50 þ t70 þ t90 Þ=5 (11-6)
difference between 100 ml (initial sample volume) and
the sum of the recovery and the residue is referred to as where the subscripts 10, 30, 50, 70 and 90 refer to
the loss. Repeated tests give 6o F ð3:3o CÞ for the ini- the volume per cent recovered during the distillation.
tial boiling and end points. Intermediate distillation The 10% to 90% slope used as the abscissa in
points are reproducible within 2 ml of distillate which Figure 11-5 is:
corresponds to 6–7 F (3.3–3.9 C). Figure 11-4 shows
a modern TBP apparatus. slope ¼ ðt90 t10 Þ=ð90 10Þ (11-7)
282
Figure 11-5 Characterizing Boiling Points of Petroleum Fractions (From API Technical Data Book). Used by permission, Gas Processors
Suppliers Association Book Data, 12th Ed., v.1 and 2., (2004).
The use of Figure 11-5 is as follows:

From Figure 11-5, Read down from a slope of 5.49
to locate the curve for the distillation VARP in the ap- on the x axis to the interpolated curve closest to 325 F,
propriate set for the type of boiling point desired. in the set drawn with the dashed lines (MeABP). Read
for a known 10–90% slope, to read a correction for across to obtain a correction value of –54 on the
the VABP from the selected VABP curve. ordinate.
MeABP ¼ 325 – 54 ¼ 271 F.
Example 11-1
Determine the mean average boiling point (MeABP) and
the molecular weight for a 56.8 API petroleum fraction 11.5.4 ASTM D1160 Distillation
with the following ASTM distillation data:
This is designed for high boiling point samples in which
% Over Temperature, 8F there is significant cracking at atmospheric pressure.
Typical samples for which D1160 distillations are useful
IBP 100
are heavy heating oils, cracker gas oil feeds, and residual
5 130
oils. The test is similar to the ASTM 86 method, but the
10 153
sample is distilled at a reduced pressure to inhibit
20 191
cracking. The distillation pressure commonly selected for
30 217
ASTM D1160 tests is 10 mm Hg. All observed D1160
40 244
temperatures are pressure corrected to their equivalents
50 280
at 760 mm Hg.
60 319
At 10 mm Hg, petroleum stocks can be distilled
70 384
to a maximum boiling point temperature equivalent to
80 464
950–1000 F (510–538 C) at 760 mm Hg. Lower pres-
90 592
sures are sometimes used to extend this boiling point
EP 640
range for residual stocks. The reduced pressure used for
(Source: Engineering Data Book: Gas D1160 distillations produces a better separation of
Processors Suppliers Association, Tulsa, components than D86 distillation. The D1160 method
Oklahoma, 12th Ed., 2004) closely approximates the TBP method, and the temper-
atures recorded for 50% distilled and higher samples are
Solution assumed to be identical to the corresponding TBP tem-
IBP ¼ initial boiling EP ¼ end point peratures. The D1160 method is sometimes used to
Slope (592 – 153)/80 ¼ 5.49 extend the TBP distillation curve for mixtures containing
VABP ¼ (153 þ 217 þ 280 þ 384 þ 592)/5 ¼ 325 F high boiling materials.
283
11.5.5 Determination of ASTM IBP, 10%,

20–90% Points of Blend
This method is applicable to blends containing distillate
stocks having an ASTM initial boiling point higher than
85 F, and an ASTM end point lower than 700 F. It is based
on the observation that a straight summation line can be
drawn through an ASTM distillation point of a blend.
The slope of the line must be such that the sum of the
proportions of each blend component corresponds to its
intersection with the ASTM distillation curve. For TBP
distillation, the summation lines are parallel to volume
per cent axis on an ASTM distillation plot. ASTM sum-
mation lines slope due to poor fractionation of ASTM
distillation and the slope varies according to distillation
end point. The following are the slopes for varying dis-
tillation points [299]: Figure 11-6 ASTM distillation blending procedure (source:
Parkash, R., Refining Processes Handbook, Elsevier: Gulf
Professional Publishing, 2003).
Distillation Point Slope of Summation line, 8F
IBP –180 F per 100% distilled
10% –180 F per 100% distilled
20% –100 F per 100% distilled
30% –80 F per 100% distilled extrapolated to –1.4%. Therefore, –1.4% becomes the
40% –50 F per 70% distilled new zero of the modified scale, and 10% becomes 11.4%.
50–90% –20 F per 70% distilled The volume distilled at 1.4% is next estimated by the
previous procedure to give the IBP of the ASTM curve.
11.5.6 ASTM 10-90% Points 11.5.8 ASTM End Point of Blend

ASTM distillation curves are drawn for each blend com- The end point of a two-component blend is referred to as
ponent with the temperature on the vertical axis and the the function of the end point, proportion of blend com-
volume per cent distilled on the horizontal axis. Distil- ponent, and slope of the tail of the distillation curve of
lation must be done on a consistent basis for all compo- the higher-boiling component.
nents, i.e. either percent evaporated or percent recovered.
A guess is made of the temperature at which a given
proportion of the blend is distilled, and the corre-
11.5.9 Flash Point
sponding point is marked on the graph. A summation line
The flash point is a measure of the volatile components in
of specified slope is drawn through the point. The vol%
a petroleum middle distillate. It is dependent on the
distilled is read off vertically below the intercept of the
predicted solubility of light hydrocarbons in the middle
summation line and ASTM distillation curve of each
distillate. Various attempts have been made to correlate
component (Figure 11-6). The sums for all blend com-
flash points for petroleum distillates, and all attempts use
ponents should equal the proportion of blend originally
the predicted ASTM 86 10% distilled temperature as
estimated, otherwise a new guess of temperature at
a correlating parameter.
which the specified proportion of blend is distilled is
The flash point test determines the temperature at
made and the procedure repeated.
which the vapors from petroleum oil will ignite or
Where the second estimate does not produce the re-
‘‘flash’’. Three basic procedures are used to determine
quired result, an interpolation is made between the two
this: the Cleveland open cup method (ASTM D92), the
estimates.
Penskey Martens (ASTM D93), and the Tag (ASTM
D53) closed cup methods.
11.5.7 Initial Boiling Point Determination A sample is heated at a specified rate and an open
flame introduced into the emitted vapors at specified
This method is identical to 10–90% points of blend, time intervals. The temperature at which the vapors
except that the distillation curves for the components are ignite is recorded as the flash point. The open cup
284
method is often used because of its simplicity. Below 11.5.13 Reid Vapor Pressure (RVP)
flash points of 510 F (266 C), the open cup flash point is
approximately 10 F (5.6 C) higher than the closed cup The Reid vapor pressure test is used to determine the
flash point for the same sample. The open cup method front-end volatility of products in the gasoline through
heats the oil at a rate of 10 F (5.6 C) per minute with heavy reforming naphtha boiling point range. It correlates
the test flame applied at 30 sec. intervals. Accuracy for with the normal butane content of the sample and the
this method is 5 F (2.8 C). The closed cup tester RVP of gasoline blends is adjusted by adding or removing
heats the oil at a rate of 1.8 F per minute and is the more normal butane. For gasoline, the RVP is 8–14 psi, and the
precise of the two methods. percentage of butanes in the stream is the same as the
RVP. The value is always reported in absolute pressure
units of psi, kPa, bar, etc.
11.5.10 Flash Point, F, As a Function The test procedure for gasolines and naphthas involves
of Average Boiling Point combining one part of chilled liquid sample at 36 F
(2.6 C) with four parts of laboratory air at 70 F
The correlation is applicable up to 1050 F [299]: (21.4 C) in a bomb that has a pressure gauge attached to
Flash ¼ 0.535T – 93, if (T 174 F) the lid. The bomb is placed in a constant temperature
Flash ¼ 0.67T – 116, if (174 T 303 F) bath at 100 F (37.8 C), and when the pressure reaches
Flash ¼ 0.7506T – 140.3, if (303 T 577 F)
Flash ¼ 0.674T – 94.5, if (577 T < 1050 F)
Table 11-7 Relationship between the True and Reid Vapor Pressure
for Miscellaneous Volatile Products
11.5.11 Smoke Point of Kerosenes Vapor

pressure, psi Ratio
The smoke point (S) of kerosene can be estimated by the
Product Reid True True/Reid
following correlation with the aromatic content of the
stream. Natural gasoline 24.0 26.2 1.09

S ¼ 53:7=arom0:5 þ 0:03401 API1:5 þ 1:0806 Natural gasoline 20.0 21.8 1.09
(11-8) Natural gasoline 18.4 20.1 1.092

where Natural gasoline 17.8 20.3 1.14
S ¼ Smoke point, mm
Natural gasoline 16.0 17.5 1.09
arom ¼ aromatic content, LV %
API ¼ API gravity of the cut. Light gasoline 14.0 15.4 1.10
Light gasoline 12.0 12.9 1.07

11.5.12 Luminometer Number Natural gasoline 12.0 12.9 1.07
The luminometer number is qualitatively related to the Gasoline 10.0 10.4 1.04
potential radiant heat transfer from the combustion
products of aviation turbine fuels. It provides an in- Straight-run gasoline 8.3 9.7 1.17
dication of the relative amounts of radiation emitted by Gasoline 8.0 8.3 1.04
combustion products of gas turbines fuels.
The luminometer number (Ln) is related to the smoke Gasoline 6.0 6.3 1.04
point by the following correlation (ASTM specification
Refinery gasoline 5.9 7.2 1.16
D-1740):
Gasoline 5.0 5.2 1.04
Ln ¼ 12:03 þ 3:009 SP 0:0104 SP2 (11-9)
Gasoline 3.0 3.1 1.03
SP ¼ 4:16 þ 0:331Ln þ 0:000648 L2n (11-10)
Gasoline 2.0 2.1 1.05
where
Source: W.L. Nelson, Petroleum Refinery Engineering, 4th. Ed., McGraw-Hill
Ln ¼ the luminometer number Book Co., New York, p 133, [1958].
SP ¼ the smoke point
285
Table 11-8 RVP and Flash point of some crude oils where
B ¼ average boiling point, C
Country Crude RVP, bar Flash point, 8C
T ¼ temperature, C
Iraq Kirkuk 0.29 <20 VPmmHg ¼ vapor pressure, mmHg
VPpsi ¼ vapor pressure, psi
Kuwait Kuwait 0.51 <20
Nigeria Nigerian 0.26 <20

11.6 Viscosity
Qatar Qatar 0.50 <20
Venezuela Bachaquero 0.06 46 The viscosity of oil is its internal resistance to flow and
a measure of its usefulness as a lubricating substance.
Crude oil viscosities are usually used for calculating the
a constant value, this is recorded as the Reid vapor
pressure drop in pipelines, and for deriving the specifi-
pressure. Table 11-7 shows the Reid vapor pressure for
cations of equipment such as exchangers and pumps. The
gasoline streams with the largest deviations occurring for
Saybolt Universal viscosity test (ASTM D88) is the most
the samples at the highest vapor pressures. The Reid
common method used by refiners. This measures the
vapor pressure measurements have been used to assess time taken (in seconds) for a sample of 60 cc of oil to flow
the amount of light hydrocarbons that may be lost during from a container through a calibrated orifice at a pre-
filling and draining operations of tanks [296]. scribed temperature. The usual test temperatures are
RVP 1 100 and 210 F (37.8 and 99.2 C).
Losses ðvolume%Þ ¼ (11-11)
6 Saybolt viscosities are referred to as seconds. The
The Reid vapor pressure is expressed in psi (1 psi ¼ 6.9 Saybolt Furol viscosity test (ASM D88) is a similar test
kPa). Table 11-8 shows the RVP of some crude oils. for heavy fuel oils and uses a larger orifice. Saybolt Furol
Measurement of the vapor pressure and flash point of viscosities are reported at 122 and 210 F (50.3 and
crude oils indicates the amount of light hydrocarbons 99.2 C).
present. Crude oils with a vapor pressure of >0.2 bar at
37.8 C have a flash point <20 C.
11.6.1 Conversion to Saybolt Universal
Viscosity
11.5.14 Vapor Pressure of Narrow
Hydrocarbon Cuts Kinematic viscosity in centistokes (mm2/s) can be
converted to Saybolt universal viscosity in Saybolt uni-
Vapor pressure of narrow hydrocarbon cuts can be versal seconds (SUS) units at the same temperature by
estimated from their boiling points by the correlation of API databook [300] procedure or using ASTM conver-
Van Kranen and Van Ness [299]: sion tables [301].
½1:0 þ 0:03264 cSt

SUS ¼ 4:6324 cSt þ (11-15)
½3930:2 þ 262:7 cSt þ 23:97 cSt2 þ 1:646 cSt3 105
where

ð232:0 þ BÞ ð1120 TÞ SUS ¼ Saybolt universal viscosity in Saybolt universal
x ¼ 6:07918 ð3:19837Þ
ð232 þ TÞ ð1120 BÞ seconds units at 37.8 C
(11-12) cSt ¼ kinematic viscosity, centistokes (mm2/s)
where where
SUS ¼ SUS viscosity at 37.8 C
VPmmHg ¼ 10x (11-13)
SUSt ¼ SUS viscosity at the required temperature
VPpsi ¼ 10x $0:01932 (11-14) t ¼ temperature at which SUS viscosity is required, C
286
11.6.2 Conversion to Saybolt Furol compared to that of the standard member to determine
Viscosity its octane member.
Kinematic viscosity in centistokes (mm2/s) at 50 and

98.9 C can be converted to Saybolt Furol viscosity in 11.8 Cetanes
Saybolt Furol seconds (SFS) units at the same temper-
ature by API databooks [302]: The cetane number is used to measure the performance
of a fuel in a diesel engine. Although determined in
a laboratory engine (ASTM engine test D613), it is cal-
SFS50 ¼ ð0:4717ÞðVIS50 Þ culated with a formula that relates cetane number, API
13:924 gravity and ASTM D86, 50% distilled.
þ 2
ðVIS50 Þ 72:59VIS50 þ 6:816 All the laboratory tests are used to determine the
qualities of feed stocks and products from petroleum
(11-16)
processes. They are essential in design and simulation
5:610 since they can be used to define petroleum pseudo-
SFS98:9 ¼ ð0:4792Þ ðVIS98:9 Þ þ
ðVIS98:9 Þ2 þ 2:130 components to represent the streams for the unit oper-
(11-17) ations calculations.
where 11.8.1 Cetane Index

SFS50 ¼ Saybolt Furol viscosity at 50 C, Saybolt Furol
seconds The cetane index (CI) is estimated from the API gravity
VIS50 ¼ kinematic viscosity at 50 C, centistokes (mm2/s) and 50% ASTM distillation temperature in F of the
SFS98.9 ¼ Saybolt Furol viscosity at 98.9 C, Saybolt diesel. It is defined by [303]:
Furol seconds
VIS98.9 ¼ kinematic viscosity at 98.9 C, centistokes CI ¼ 420:34 þ 0:016ðAPIÞ2 þ 0:192ðAPIÞ log ðMÞ
(mm2/s) þ 65:01ðlog MÞ2 0:0001809 M2
(11-18)
11.6.3 Equivalents of Kinematic (cSt), or

Saybolt Universal (SUS), and Dynamic
Viscosity ðmÞ CI ¼ 454:74 1641:416D þ 774:74 D2 0:554B
þ 97:803 logðBÞ2
Saybolt Universal Seconds ¼ Centistokes 4.6347 for
y ðcStÞ > 50 (11-19)
Saybolt Furol Seconds ¼ Centistokes 0.4717
where
Kinematic viscosity ðcStÞ API ¼ API gravity

M ¼ mid-boiling temperature, F
Dynamic viscosity; m ðcPÞ
¼ 103 D ¼ density at 15 C
kg
Density; r B ¼ mid-boiling temperature, C
m3
11.9 Diesel Index
11.7 Octanes The diesel index (DI) is correlated with the aniline point
of diesel as:
The octane test is used to measure the knocking char- DI ¼ aniline point API=100 (11-20)
acteristics of a fuel in a laboratory gasoline engine. The
engine is calibrated by measuring the knocking value of The relationship between diesel index and cetane index
a standard material; 2-2-4 trimethyl pentane (224 TMP). of diesel is:
By definition, an octane number of 100 is assigned to
224 TMP. The knocking value of the test sample is DI ¼ ðCI 21:3Þ=0:566 (11-21)
287
Nelson [304] presents four fundamental principles for

11.10 Determination of the Lower the operation of crude columns and other multi-draw
Heating Value of Petroleum columns as:
Fractions 1. The yield of a given product is primarily a function of
the composition of the feed mixture, not the degree
If the API gravity of a distillate is above 42 , the net heat of separation.
(H) of combustion in Btu/lb can be determined as
2. The number of trays only slightly alters the boiling
follows:
range of the products as defined by ASTM initial
boiling points (IBP) and end points.
H ¼ 8505:4 þ 846:81K þ 114:92 API
3. The initial boiling point of side draw products is
þ 0:12186 API2 ð9:951ÞðKÞðAPIÞ (11-22) always low, and it must be corrected by either steam
stripping or reprocessing.
If the API gravity of distillate or residue is below 42 , the 4. The end point of a side draw product is primarily
net heat (H) of combustion in Btu/lb can be determined controlled by opening or closing the draw valve to
as follows: change the yield.
H ¼ 17213 þ 5:22 API 0:5058 API2 (11-23)

11.12 Chromatographically
Correct the preceding heating values for sulfur content of Simulated Distillations
the distillate or residue.
Gas chromatography methods have been developed to
Heat ¼ H ð1 0:01 SÞ þ 43:7 S (11-24) predict the boiling point distribution corresponding to
that determined by the detailed TBP method. Four
where standard tests are currently employed by refinery test
laboratories, namely; D3710, D2887, D2887 extended,
Heat ¼ lower heating value, Btu/lb
and HTSD. Table 11-9 summarizes simulated distilla-
K ¼ Watson K
tion procedures. The methods of operation for each are
API ¼ API gravity similar, and the most commonly used method is illus-
S ¼ Sulfur content (%) trated in ASTM Standard D 2887-73. This method is
limited to petroleum fractions with a final boiling point
of 1000 F (538 C). The sample must also have a boil-
11.11 Aniline Point Blending ing point range 100 F (55 C). Carbon disulfide (CS2)
is used for the carrier gas. The chromatographic
The aniline point of a gas oil is indicative of its aromatic column is first calibrated with a standard mixture of
content. Aromatic hydrocarbons exhibit the lowest and
paraffins the highest values. Aniline point (AP) blending
is not linear, and therefore blending indices are used.
Table 11-9 Simulated Distillation Procedures [15].
The conversion of aniline point to aniline index is
given by: Report
Method Application Standard Range basis
ANLIND ¼ 1:25 AP þ 0:0025 ðAPÞ2 (11-25)
D3710 gasolines, C5–C15 <500 F volume %
or, alternatively the aniline point is: naphthas
D2887 cycle oils, C5–C44 <1015 F weight %

ANLIND 0:5 gas oils
AP ¼ 200 1:5625 þ 1:25
100 D2887 – heavy gas oil, C5–C60 <1140 F weight %
(11-26) ext. crude
HTSD heavy gas oil, C5–C120 <1382 F weight %

where crude
ANLIND ¼ the aniline index Note: The standard is a mixture of normal paraffins with carbon number ranges
AP ¼ the aniline point as shown.
288
normal paraffin hydrocarbons. A calibration curve of Crude oils and heavy residues have such high boiling
retention time in the column versus normal boiling points that the column temperature cannot be increased
point is developed for normal paraffins by increasing sufficiently to measure the highest boiling fractions.
the column temperature and recording the areas of Hence, the total area of the sample cannot be known and
each recorded peak of the individual components at the boiling point curve for the sample cannot be de-
timed intervals. The sample is next introduced into the termined accurately. HTSD is a modification of the
column; the retention times and the areas of the D2887 procedure and extends the boiling range of hy-
sample components are recorded. The components in drocarbon to a final boiling point of z1382 F (750 C). A
the sample are then assigned boiling points by using the careful choice of chromatographic conditions is used to
calibrated curve developed for the normal paraffin cause the high boiling components to be eluted from the
standard. column some 500–600 F (260–316 C) below their
Cooling water
Gas/light naphtha
To recovery columns
Pump back
reflux
Sour water
Steam
Heavy naphtha
Crude distillation unit
Kerosene
Light diesel
Heavy diesel
Flash zone
Atmospheric residue
to VDU unit
Superheated
steam
Crude oil Desalter unit
Figure 11-7 Atmospheric crude column with pumparound and pumparound reflux.
289
normal boiling points, and as such provides great poten-

Table 11-10 Washwater requirements of Desalters
tial for analysis of heavy petroleum residues.
Crude oil API Washwater, Vol. % Temperature, 8F
11.13 Process Description
API > 40 3–4 240–260
Figure 11-7 shows the crude fractionation unit in more 30 < API < 40 4–7 260–280
detail with the pumparound. Cold crude oil from storage
is pumped through a series of exchangers, where it is API < 30 7–10 280–300
preheated by exchange with intermediate products from
the crude and vacuum distillation columns.
Crude desalting is the first processing step in a re- applied at potentials of 16,000–35,000 V. In the desalter,
finery. This is done to remove the salts, suspended solids the salts (e.g. magnesium chlorides) selectively migrate
and water from the unrefined crude before introducing it to the aqueous phase which forms a brine solution at the
to the crude distillation unit (CDU). Salt is present in bottom of the desalter, while the crude oil floats above
crude oil in the form of dissolved or suspended crystals in and forms a separate stream. Efficiencies of up to
water that is emulsified with the crude oil. The desalting 90–95% water removal are achieved in a single stage, and
process washes the salt from crude oil with water. It can up to 99% in a two-stage desalting process. Figure 11-8
often be difficult to produce efficient and economical shows an electric desalting method and Table 11-11
water/oil mixing, water wetting of suspended solids, and shows typical operating conditions in the desalting unit.
separation of oil from wash water. The separation of oil After desalting, the crude oil contains from 0.25–0.5
and wash water is affected by the specific gravity, vis- liquid volume percent water from solubility and emul-
cosity and pH of the crude oil, and the ratio of water/ sion effects.
crude used for washing. The pressure of the desalted crude oil is then raised in
The crude oil is injected with five to six liquid volume a booster pump, and it continues through a preheat train
percent water at a temperature 200–300 F (93–149 C). where its temperature is raised by exchange with hot
Both the ratio of water to oil and the operating conditions intermediate streams from the crude and vacuum col-
are determined by the specific gravity of the crude oil, umns. Upon exit from the last preheat exchanger the oil
and typical operating conditions are shown in Table 11-10. is heated to the desired column inlet temperature in the
The salts dissolve in the wash water, and then the oil and crude furnace. Furnace outlet temperatures typically
water phases are separated in a settling vessel, either by range from 630–725 F (333–385 C), depending on the
adding chemicals to break up the emulsion, or by the crude oil composition.
application of an electrostatic field to coalesce the The hot oil mixture from the furnace then flows
droplets of salt water. Either an AC or DC field may be through a transfer line into the flash zone of the main
First stage desalter Second stage desalter

V-100 V-101
Electrical power Electrical power
Desalted crude
Heater H-100
Unrefined
Crude oil Emulsifier Emulsifier
P-100 steam MX-100 MX-101
Washwater
P-101
Effluent water
Acid injection
P-102
NH3
Figure 11-8 Two-stage desalter.
290
the column. The stripping steam also lowers the partial

Table 11-11 Operating conditions
pressure of these hydrocarbons in the column flash zone.
Parameter Units Value This in turn lowers the temperature required to distill
the overhead products from the column feed.
Crude to desalter* bpsd 98000 The hot liquid from the flash zone drops through the
column stripping section, where it is stripped by steam
Water to desalter gpm 145
injected into the bottom of the column. The hot vapors
Water to crude ratio % 5 from the flash zone rise, contact the colder reflux flowing
down the column which condenses out the side liquid
Demulsifier injection ppmw 10–15 products. Reflux flow is provided by the overhead con-
Pressure
denser and by pump-around circuits in which liquid
drawn from the side of the condenser is pumped through
Crude to desalter psig 125 exchangers and cooled by exchange with crude oil in the
preheat train. The cooled pump-around liquid is typically
DP mixing valve psi 20 returned to the column a few trays above the draw tray.
Temperature Pump-around flow rates are high, because heat is being
exchanged with crude oil that is at a high flow rate.

Crude to desalter F 270 The pump-around cooling circuits accomplish three
objectives: firstly, they remove latent heat from the hot
Water to desalter F 265
flash-zone vapors and help condense the side products;
Crude from desalter
F 260 secondly, they improve the efficiency of the crude pre-
heat train by allowing heat recovery at higher tempera-
Analysis results ture levels than an overhead condenser alone could
Crude inlet salt lb/1000 bbl 3.94 achieve. This decreases the work required from the
crude furnace. Finally, they reduce the vapor flow rate
Crude inlet salt ppmw 12.87 through the column so reducing required size. Since
crude column sizes are primarily determined by the
Crude outlet salt ppmw 1.2
vapor flow rates, pump-around cooling is an essential
Outlet BS & W % Mass 0.05 factor in their design. However, the pump-around cir-
cuits degrade the fractionation between the side prod-
Water ucts by reducing the internal liquid flows between side
products.
Inlet salt content ppm 100
The liquid side draw products are usually stripped in
Outlet salt content ppm 310 steam strippers or recoiled strippers to remove light
materials and control their flash points, so the composi-
Inlet oil content ppm 7 tion of the lightest portion of a liquid side product is
Outlet oil content ppm 10 controlled by a side stripper, not the main column. In the
case of the gas oil side product, stripping is used to
pH inlet 6.5 remove diesel boiling range materials from the cracking
stocks for feeding to fluid catalytic crackers (FCCs).
Outlet pH 6.5
The maximum boiling point of the liquid side prod-
Outlet pH after NH3 ucts are controlled on the main column by their draw
rates. For example, to increase the maximum boiling
Injection 7 point of the kerosene product, it is necessary to decrease
the flow of the diesel product (which has a higher boiling
* 30.4 API Crude. range) and increase the flow of the kerosene product.
Note: Basis 98000 bpsd crude.
Source: Parkash, S., Refining Processes Handbook, Elsevier: Gulf Professional
This adjustment allows heavier components to travel up
Publishing, Co. 2003. the column to the kerosene draw tray, increasing the
maximum boiling point of the kerosene product. There
may also be a decrease in the lightest portion of the diesel
product because of this adjustment, but this effect is
fractionation column. Here it mixes with steam and light small when compared to the effect of the diesel steam
hydrocarbons from the column stripping zone. Here, stripper. Table 11-12 shows the products of crude oil
sufficient steam is used to strip the lightest hydrocarbons distillation, and Table 11-13 shows the approximate
from the gaseous portion of the oil that rises to the top of boiling ranges for crude oil fractions.
291
Table 11-12 Products of crude oil distillation
Fraction Boiling range Remarks

1. Gases (mainly methane to the extent of 65–90% <20 C Resembles natural gas and is useful for fuel and
together with some ethane, propane and butane) chemicals. In refineries most of these gases are
burned in furnace as fuel due to the costs involved
in the recovery
2. Naphtha
a. Light Naphtha (C5–C6 hydrocarbons) 70–140 C Useful for both fuel and chemicals. But light
naphtha is not preferred in gasoline because on
catalytic reforming yields the carcinogenic
compound benzene
b. Heavy Naphtha (C7–C9 hydrocarbons 140–200 C Base for gasoline
3. Atmospheric gas oil

a. Kerosene (contains C9–C16 compounds). 175–275 C Used as heating fuel
b. Diesel fuel (contains C15–C25 compounds, 200–370 C Fuel for diesel engines
mostly linear)
4. Heavy fractions
a. Lubricating oil (contains C25–C40 compounds). >370 C As lubricating oil
b. Residual or heavy fuel oil. ————— Useful as boiler fuel
Source: Koch Baby Majoorana, Modern Petroleum Chemistry: An overview, Kannatheri Publications Kochi, 2003.
temperature to which the crude can be heated in the

Table 11-13 Approximate ASTM boiling point ranges for crude oil
furnace without incipient cracking.
fractionation
2. Flash zone operating conditions: flash zone temper-
Boiling range ature is limited by advent of cracking; flash zone
Fractions 8F 8C pressure is set by fixing the reflux drum pressure
and adding the value to the line and tower pressure
Light naphtha 85–210 30–99 drop.
Heavy naphtha 190–400 88–204 3. Over-flash: this is the vaporization of crude over
and above the crude overhead and sidestream
Kerosene 340–520 171–271 products. Over flash is generally kept in the range
of 3–6 LV% (LV ¼ liquid volume). Over-flash
Atmospheric gas oil 540–820 288–438
prevents coking of wash section plates and carry-
Vacuum gas oil 750–1050 399–566 over of coke to the bottom sidestream, and also
ensures a better fractionation between the bottom
Vacuum residue 1000þ 538þ sidestream and the tower bottom by providing
reflux to plates between the lowest sidestream and
the flash zone. A larger over-flash also consumes
more energy, thus over-flash is kept to the mini-
11.14 Process Variables in the mum needed for the the quality requirement of the
Design of Crude Distillation Column bottom side stream.
4. Steam stripping: the bottom stripping steam is used
The following are important variables in the design of to recover light components from the bottom liquid.
a crude oil distillation unit: In the flash zone of an atmospheric distillation
1. The nature of the crude: water content, metal con- column, approximately 50–60% of the crude oil
tent and heat stability. The latter limits the input is vaporized. The unvaporized crude travels
292
down to the stripping section of the columns, where 11.14.2 A Case Study: Process
four to six plates strip out any low boiling point Description of Light Arabian Crude Using
distillates still contained in the remaining crude. The
steam rate used is approximately 5–10 lbs/bbl of
UniSimÒ Simulation Software
stripped product. The flash point of the stripped
Honeywell UniSimÒ simulation software was used to
stream can be adjusted by varying the steam flow
simulate the flow of crude oil from the pre-flash unit to
rate.
the crude distillation fractionators. The following illus-
5. Fractionation: this is the difference between the 5% trates the process:
point on the ASTM curve of a heavy cut and the After characterization (Figure 11-9), the crude oil
95% point on the ASTM curve of a lighter cut of passes through a preheat train at 10,000 barrel/day at
two adjacent side products. A positive difference is 29 API and is fed into a pre-flash separator operating at
referred to as a gap and a negative difference is re- 450 F (232 C) and 75 psia. The vapor from this sepa-
ferred to as an overlap. rator bypasses the crude furnace and is re-mixed with the
hot crude at 650 F (343 C) pre-flash liquids leaving the
furnace. The combined stream is then fed to atmospheric
11.14.1 Process Design of a Crude Oil fractionation column as shown in Figure 11-10. The
column has a total condenser, three coupled side
Distillation Column strippers and three pump-around circuits. Light gases
The following steps in the design of a crude distillation and a naphtha product are produced overhead, a kero-
column are: sene product from the first recoiled side stripper,
a diesel product from the second steam-stripped side
Prepare TBP distillation and equilibrium flash vapor- stripper, and an atmospheric gas oil (AGO) from the
ization curves for the crude to be processed. Several third side stripper. This last is further processed in
methods are available for converting TBP data to EFV the vacuum distillation unit. Figure 11-11 shows a snap
curves. shot of the process flow diagram of the crude distilla-
Using crude assay data, construct TBP curves for all tion unit with the splitters, and Table 11-14 shows
products except gas and reduced crude. These are typical operating conditions of the atmospheric distil-
then converted to ASTM and EFV curves by lation column.
Edmister [305], Maxwell [306] or computer The two basic steps in the process simulation are:
methods.
1. Setup involves listing the components between
Prepare material balances of the crude distillation
C1–C4 light ends, as well as the pseudo components
column, on volume and weight basis, showing crude
that will be used to represent the C5þ and above
input and product output. In addition, plot the
portion of the crude oil. The oil characterization
physical properties, such as cut range on TBP and
procedure as described earlier is used to convert
LV% and vol% vs. SG, molecular weight, mean aver-
the laboratory data into petroleum pseudo-
age boiling point, and enthalpy curves for the crude
components.
and its various products.
Fractionating requirements: Ideal fractionation is the 2. Steady state simulation modeling involves simulat-
difference between the 5% and 95% points on ing the pre-fractionation train, which consists of
ASTM distillation curves obtained from ideal TBP a separator and heater. This determines the feed to
curves of adjacent heavier and lighter cuts. Having the atmospheric fractionators, and includes the
fixed the gaps as a design parameter, the difference pre-flash separator, crude furnace and mixer which
between ideal gap and actual gap required is referred recombine the preflash vapor and furnace outlet
to as the deviation. streams.
The deviation or gap can be correlated with an F factor, Simulation of the atmospheric fractionator involves
which is the product of the number of plates between adding the column streams inlets to the flowsheet and
two adjacent side draws and the internal reflux ratio. carrying out the appropriate calculation.
Internal reflux is defined as the volume of hot reflux at Before doing this, the steam feeds and the energy
60 F below the draw-off plate of the lighter product, requirements of the column must be defined. Three
divided by the volume of liquid products at 60 F, except streams are fed to various locations in the column. The
gas lighter than the adjacent heavier products. This im- atmospheric tower is represented by 29 ideal stages (not
plies that the reflux ratio and the number of plates are including the condenser). The pressures of the overhead
interchangeable for a given fractionation. condenser, the top stage and the bottom of the column are
293
Figure 11-9 True Boiling Point Distillation – Arab Extra Light 36.7
19.7 psia; 28.7 psia and 32.7 psia respectively. The con- tray. All calculation procedures keep the column
denser therefore experiences a pressure drop of DP ¼ 9 psi. pressures constant. The results from the assumed pro-
The temperature estimates for the condenser, top stage, files are used for the next series of calculations. Com-
and bottom stage are 100 F (37.8 C); 250 F (121 C) and putations are further complicated by the fact that the
600 F (316 C) respectively. Water is removed via a side equilibrium K values for the components on each tray
water draw from the three phase condenser. are affected by the tray composition as well as the
Solution of fractionation calculations involves a trial temperature and pressure. Further, all of the tray-to-
and error procedure. Temperature and flow profiles are tray calculations are coupled into a large set of equa-
assumed for all trays in the column and vapor–liquid tions, which describes the component mass balances
equilibrium and heat balances are calculated for each and the heat balances for every tray, the vapor-liquid
Figure 11-10 Process flow diagram of the crude distillation unit.
294
Figure 11-11 A snap shot of crude distillation unit with pump-arounds and side-strippers.
equilibrium relationships between the trays and the tray heat balances, the tray equilibrium relationships
overall heat and mass balances for the entire operation. (e.g. expressed as the tray liquid bubble point condi-
Parameters are added to define process conditions tion), the performance specifications and the tray
or specifications that must be met by the final solution. equilibrium K values. The latter tolerance is required
A converged solution is defined as a set of independent because the K values are dependent on composition.
variables, i.e. flows, heat inputs, tray temperatures, etc., Therefore, it is essential that the K-values converge
which produce a solution for all of the column equa- together with all the column equations. Several
tions within certain specified tolerances. Tolerances are methods have been described in the literature to solve
prescribed for the component material balances, the distillation problems rigourously. One of the most
295
useful method for petroleum distillation columns is the

Table 11-14 Atmospheric Tower Operating Conditions
inside-out method, which is described elsewhere [255].
Operating parameter Units Value

Temperature F 11.14.3 Column Conventions
Transfer line 660
Column conventions are similar for all process simula-
Flash zone 657 tors. All distillation calculation methods are derived
for theoretical trays, i.e. trays in which perfect vapor-
Tower top 359 liquid equilibrium is achieved. The condenser and over-
Kerosene draw-off 469 head accumulator drum are considered to be the first
tray. A net liquid product from this tray is returned to the
Pumparaound draw off 548 column as reflux. Liquid and vapor products are with-
drawn from the reflux drum. Where only a liquid prod-
Pumparaound return 345
uct is withdrawn these conditions are not met. For many
Light diesel draw off 603 columns, the liquid product is sub-cooled below its
bubble point. The same convention is followed and the
Tower bottom 648 overhead system is still referred to as the first tray. The
top tray of the column is considered to be the second tray
Pressure psig
for a column with a condenser. The tray at the bottom is
Reflux drum 2.0 referred to as the Nth tray.
The reboiler is normally considered to be the highest
Tower top 10.3 tray number, labelled N þ 1. Columns may have side
Flash zone 14.7 liquid and vapor draws and side water draws. Side strip-
ping columns (e.g. for kerosene, light and heavy gas oils,
Reflux ratio, reflux/Liq.dist. 0.6 diesel, etc.) may also be attached to the column. Condi-
tions for side columns are solved along with the main
Stripping steam
column, by the inside-out method. Pump-around side
To atmospheric tower lbs/bbl Resid. 5.5 draws are used to remove heat from the column, which
may have multiple feeds. UniSim designates the con-
To kerosene stripper lbs/bbl Resid 5.9 denser-reflux drum trays as the condenser and the second
tray as tray number one. It also designates the reboiler
To diesel stripper lbs/bbl Resid 2.1
(when present) as the reboiler and tray N as the highest
Atmospheric heater tray number in the column model. For columns with no
condenser-reflux drum, e.g. a recoiled stripper, the top
Process fluid conditions tray of the column is designated as tray one by all simu-
Temperature in
F 453 lators. For columns with no reboiler, e.g. an absorber, the
bottom tray of the column is designated as tray N by all

Temperature out F 660 simulators.
Pressure drop psi 138

11.14.4 Performance Specifications
Tube skin temperature (avg.) F 735
Definition

Stack gas temperature F 725
Poorly chosen specifications are the cause of nearly all
Fractionation efficiency convergence failures when solving distillation problems
using the inside-out method. Additionally, specifications
5%–95% ASTM distribution gap
for the column model should be similar to those used
Atmospheric naphtha-kerosene Gap þ 10 in actual operation so actual operating experience is
useful for the modeler. For every fixed value imposed
Kerosene-light diesel Gap – 36 on a column model, there must be a corresponding
Note: Basis 154000 bpsd Kuwait Crude Run. degree of freedom or unspecified parameter. Changing
these parameters enables the user to find sets of con-
ditions for the column that meet the specified
296
performance criteria. For distillation models, the 11.15 Gas Processing

choices of parameter to vary are the column heat sinks,
the side product draw rates, and the column heat Gas processing or treating may be defined as removal of
sources. For example, a conventional column with two hydrogen sulfide (H2S) and carbon dioxide (CO2) from
products, a reboiler and a condenser has two possible a feedstock. H2S must be removed for health reasons and
degrees of freedom and so may have two performance to prevent corrosion. Natural gas pipeline specifications
criteria specified. require no more than ¼ grain/100 SCF. This is equiva-
lent to 4 ppmv or 7 ppmw (for a 0.65 specific gravity
gas). By contrast, the human nose can detect 0.13 ppmv
11.14.5 Troubleshooting Actual Columns and the threshold limit value for prolonged exposure is
10 ppmv.
Troubleshooting with computer simulation must be de- CO2 must be removed to prevent corrosion and be-
pendent on mathematically converged column solutions, cause it lowers the heating value of natural gas. Some
as it is impossible to attain valid conclusions using sim- processes allow most of CO2 to ‘‘slip’’ while selectively
ulation models for which the column mathematical removing H2S. Where CO2 can be tolerated down-
constraints are unsatisfied. Thus the first stage in trou- stream, it is more economical to remove only H2S. H2S
bleshooting is to ensure converged results with the can be selectively removed in order to enrich the H2S
column model. The column models must be ascertained content of the acid gas stream feeding a sulfur recovery
as being in predictive mode for troubleshooting. The unit. However, in such cases the concentration of CO2
models based on local tray efficiency models will yield should be retained below 80% in the acid gas feed. Some
poor results because they cannot predict an alternate of the reasons for selective H2S removal are:
operating point. Computer simulation has successfully H2S enrichment for sulfur recovery
being applied to petroleum distillation problems; how-
Economical design (capital and operating costs)
ever the necessity of representing petroleum mixtures
Special cases such as energy recovery.
with pseudo-components has been a serious impediment
to the development of accurate models. Therefore,
where a computer model matches the actual operation,
there are definitive reasons. Differences between the 11.15.1 Natural Gas
model and the plant are commonly reflected in the
product properties, product compositions, product rates, Natural gas is mixture of hydrocarbon and non-hydrocarbon
column temperatures, top reflux rates, internal liquid gases found in porous rock formations below the
flow rates and heat duties. Kaes [307] has described earth’s surface, usually with crude oil. The main com-
various step wise approach when matching models to ponent of natural gas is methane, together with small
plant data. amounts of heavier hydrocarbons and some non-
Figures 11-12 and 11-13 show temperature and hydrocarbon gases, such as carbon dioxide (CO2),
pressure versus tray position from the top. Zones of helium, hydrogen sulfide (H2S) and nitrogen (N2). It
relatively constant temperature in the column indicate can further be classified as:
tray damage or missing trays. Associated gas: the gas occurs in the free state in
Simulation program of the crude oil using UniSimÒ contact with a large crude oil reservoir. Associated gas
software (crude-oil-distillation.usc) can be accessed usually contains less methane than non-associated gas,
from the Elsevier website. Results of the simulation of but is richer in higher molecular weight hydrocarbons.
crude oil distillation are shown in Appendix L, and Table Non-associated gas: the gas remains in the free state,
11-15 shows rules for the design and operation of pe- associated with minimal amounts of crude oil. It is richer
troleum fractionators. in methane and leaner in other higher molecular weight
Oil refineries are often designed to handle various hydrocarbons than associated gas.
crude oils with varying boiling assays. The refinery re- Dissolved gas: the pressure in the oil reservoir is
sponds to market conditions by making different product high, so the gas remains in solution. It is released as
ranges at different times of the year. The crude oil a result of drilling.
fractionation column and associated equipment (e.g. heat Natural gas can be in other ways classified as: dry or
exchanger networks) must be flexible enough to handle lean gas which contains mostly methane (CH4); wet gas
these changes, whilst maintaining the boiling point curves which contains considerable quantities of higher hydro-
for every product. This waves designing a fractionating carbons; sour gas which contains much H2S; sweet gas
column an even more complex operation. which contains little H2S; residue gas from which higher
297
paraffins have been extracted, and finally casing head gas The diluents, being non-combustible, reduce the value of
which is derived from an oil well by extraction at the the gas as a fuel and thus can be used as fillers to reduce
surface. its heat content. The contaminants can cause harmful
The non-hydrocarbon components of natural gas can effects and can also affect transportation and production
be classified as: equipment plants.
Diluents such as N2, CO2 and water vapor Natural Gas Liquids (NGL): Liquid hydrocarbon
Contaminants such as H2S and other sulfur
mixtures extracted from natural gas. They fractionate
compounds. into ethane (C2H6), propane (C3H8) and butane
Figure 11-12 Temperature vs. Tray Position from Top.
Figure 11-13 Pressure vs. Tray Position from Top.
298
Table 11-15 Some Rules for Design and Operation of Petroleum Fractionators
299
(C4H10). Liquefied petroleum gas (LPG) containing (see: www.HydrocarbonProcessing.com). The main
mainly C3H8 and C4H10 is the main commercial product process types are:
made from natural gas and it can be stored and handled as Reaction type: Monoethanolamine (MEA), dietha-
a liquid at ambient temperatures and moderate nolamine (DEA), methyldiethanolamine (MDEA),
pressures. diglycoamine (DGA)
Natural gasoline is a mixture of hydrocarbons Physical solvent type: Flour solvents (propylene car-
extracted from natural gas mostly containing pentanes bonate), Selenol
and some heavier hydrocarbons. Initially, liquefied pe-
Physical/chemical type: Sulfinol.
troleum gas and natural gasoline were used solely as fuels,
but recently liquefied petroleum gas has been used as Carbonate type: Potassium carbonate
a source of propane and butane for the petrochemical Solution batch type: Lo-Cat, Chemsweet
industry. Natural gasoline is also used as a source of bu- Bed batch type: Iron Sponge, Molecular sieve.
tanes (C4H10), isobutanes (iC4H10), pentanes (C5H12) Figure 11-14 provides a guide line for selecting gas
and isopentanes (iC5H12) for reforming, alkylation, syn- processes.
thetic rubber manufacture, etc. Table 11-16 shows the
general composition of natural gas.
11.15.3 Reaction Gas Processes
11.15.2 Gas Processing Methods The amines (MEA, DEA, MDEA and DGA) are the
most popular treating solutions under the following
There are many gas processing technologies as described conditions:
in Hydrocarbon Processing Magazine and Handbook
Low acid and gas partial pressures (product of system
pressure and concentration of acid gases H2S and
CO2 in the feed) of roughly 50 psia and below.
Low acid gas concentrations in the product of roughly
Table 11-16 Composition of natural gas* 4–8 ppmv (0.25–0.5 grains/100 scf).
Heavier hydrocarbons present (provide better filtra-
Dry gas Sour gas Gas condensate tion for DGA).
Component (mol %) (mol %) (mol %) For CO2 removal with no H2S, a CO2 partial pressure
CO2 – 6.7 0.7 of 10–15 psi [308].
Among the amines, MEA is preferred for low contactor
H 2S – 3.3 –
pressures and stringent acid gas specifications such as
N2 and air 0.8 – – H2S well below 0.25 grains/100 scf, and CO2 as low as
100 ppmv. MEA is degraded by carbonyl sulfide (COS)
CH4 95.8 84.0 74.5 and carbon disulfide (CS2), with the reactions being only
partially reversible with a reclaimer. (The reclaimer helps
C2H6 2.9 3.6 8.3
remove degradation products from the solution and also
C3H8 0.4 1.0 4.7 aids in the removal of suspended solids, acids and iron
compounds. Reclaimer operation is a semi-continuous
iC4H10 0.1 0.3 0.9 batch operation, where it is filled with hot amine solution
nC4H10 Trace 0.4 1.9 and if necessary, soda ash is added.)
DEA is preferred when the total system pressure is
iC5H12 – – 0.8 >500 psi or where pipeline quality gas is not required.
The 0.25 grains/100 scf constraint is more difficult to
nC5H12 – – 0.6
reach with DEA. COS and CS2 have few detrimental
C6H14 – 0.7 1.3 effects on DEA.
The process flow scheme for DEA plants resembles
C7H16 – – 6.3 that for the MEA process, however the advantages and
*
disadvantages of DEA compared to MEA are:
Sometimes natural gas contains some helium content 0–0.5 mol %.
The mole/mole loading typically used with DEA
Source: Majoorana, S., K. B., Modern Petroleum Chemistry, Kannatheri
Publications, 2003. (0.35–1.3 mole/mole) is much higher than that used
(0.3–0.4) for MEA.
300
Figure 11-14 Guide for selecting gas sweetening processes (Source: Branan, Carl, R., ‘‘Rules of Thumb for Chemical Engineers’’ Gulf
Publishing Co., 1994).
Because DEA does not form significant amounts of problems. Some of the advantages of selective removal
non-regenerable degradation products, a reclaimer is of H2S are:
not required. Smaller amine regeneration plants are needed.
DEA is a secondary amine, and so is chemically Reduced solution flow rates resulting from a reduction
weaker than MEA, hence less heat is required to in the amount of acid gas removed.
strip the amine solution.
Enriched H2S concentrations in the acid gas resulting
DEA is typically used for refinery and manufactured in reduced problems in sulfur recovery.
gas streams that have COS and CS2 present.
MDEA is preferred for selective H2S removal and lack of
degradation by COS and CS2. 11.15.4 Physical Solvent Gas Processes
DGA is preferred for cold climates and high (50–70
wt%) solution strength for economy. The freezing point The physical solvent types of gas processing are considered
for 50% DGA solution is –30 F. By comparison, solution when acid gases in the feed are above 50–60 psi, heavy
strength for MEA is 15–25 wt%, and for DEA, 25–35 wt%. hydrocarbon concentration in the feed gas is low, bulk re-
Because of the high amine degradation rate DGA moval of the acid gas is required and high selective removal
systems require reclaiming to remove the degradation of H2S is desired. Usually, physical solvents are economical
product. because regeneration occurs by flashing or stripping which
\MDEA is a tertiary amine which is used to selec- requires little energy. The solvents are regenerated by
tively remove H2S to levels required by pipeline speci- multi-stage flashing at low pressures; regeneration at low
fications at moderate to high pressure, if increased temperatures with an inert stripping gas; heating and
concentration of CO2 in the residue gas does not cause stripping of solution with steam/solvent vapors.
301
SelenolÒ: (licensed by Norton Company) uses dime- Low corrosion/carbon steel.

thylether or polyethylene glycol as a solvent. A SelenolÒ Low foaming.
plant can be designed to provide some selectivity for Low vapor losses.
H2S, e.g. the plant can be designed to provide pipeline
quality gas (0.25 grains H2S/100 scf) while slipping 85%
Disadvantages
of the CO2 present in the original sour gas.
The Fluor Solvent: propylene carbonate is used pri- Sulfinol absorbs heavy hydrocarbons, which may pro-
marily for removal of CO2 from high pressure gas vide problems if the acid gas is then fed to a sulfur
streams. The temperature of the lean solution in the recovery unit.
absorber is usually well below ambient, and some method The process uses expensive chemicals and the user
of refrigerating the solvent is required. Propylene car- must pay a process royalty, but losses are low and
bonate is not economical below about 12% acid in the the licensee receives many engineering services.
feed [284].
Rectisol: This method, developed by Lurgi Co. and
Linde A.G, uses methanol as a solvent. Because of the 11.15.7 Carbonate Process
vapor pressure of methanol, the process is normally ap-
plied at temperature of –30 F to –100 F. The process has The basic process was developed by the U.S Bureau of
been found suitable for the purification of natural gas for Mines and uses an aqueous solution of potassium car-
LNG production. It is best suited to situations where bonate (K2CO3). The contactor and still both operate at
there is little ethane and heavier components in the feed. temperatures in the range of 230–240 F. The process is
Purisol: The process has been developed by Lurgi, not suitable for gas streams containing only H2S. Natural
and licensed by the Ralph Parsons Company. The solvent gas streams have been economically treated with potas-
used in this process is N-methyl-2-pyrolidone (NMP), sium carbonate. Medium to high acid gas concentrations,
a high boiling point liquid. combined with the high pressures of natural gas trans-
mission lines yield the high partial pressures of acid gases
required for the process.
11.15.5 Physical and Chemical The overall reaction for CO2 and H2S with potassium
Processes carbonate can be represented by:
The Sulfinol process developed by Shell Development K2 CO3 þ CO2 þ H2 O42KHCO3 (11-27)
Company uses both physical and chemical processes for K2 CO3 þ H2 S4KHS þ KHCO3 (11-28)
gas treatment. It blends a physical solvent and an amine
to obtain the advantages of both. The physical solvent is For stringent specifications of outlet H2S and CO2,
SulfolaneÒ (tetrahydrothiophene dioxide) and the amine special designs or a two-stage process may be required.
is usually DIPA (diisopropanol amine). The flow scheme Maddox [309] states that a two-stage process would
is identical to an amine plant. typically be used for acid gas feed concentrations of
20–40%. Maddox gives the following rules of thumb for
process flow scheme selection:
11.15.6 Advantages and disadvantages
of the SulfinolÒ process Single-stage process will remove CO2 down to 1.5%
in the treated gas.
Advantages A split stream cooling modification will produce
0.8% CO2.
The solvent SulfolaneÒ, like other solvents, has
Two-stage process or two-stage process with cooler
higher capacity for acid gas at higher acid gas partial
will go below 0.8% CO2.
pressures. At these higher partial pressures, the
SulfinolÒ process has lower circulation rates (higher He further shows how the major process concerns of
solution loading) and better economy than corrosion, erosion and column instability must be met in
MEA. SulfinolÒ is preferred over MEA when the the design and operation of a hot carbonate process, since
H2S/CO2 ratio is >1:1, while at high acid gas partial these items will affect the capital and operating/main-
pressures. tenance costs.
In solution, the amine DIPA is able to achieve Various processes such as the Benfield, Caracara and
pipeline quality gas (0.25 grains H2/100 scf). Giammarco processes attempt to improve on the basic
COS, CS2, and mercaptans (RSH) are removed. CO2 potassium carbonate process by using activators to in-
slightly DIPA degrades, but reclaimation is easy. crease the rate of CO2 absorption.
302
11.15.8 Solution Batch Process selectively treating the gas in order to raise the H2S
concentration to the sulfur recovery plant.
The Lo-CatÒ process can be used to sweeten or convert High water concentration and hydrocarbons in the
H2S to sulfur. It catalytically converts H2S in the gas feed gas can cause design and operational problems for
stream to solid sulfur and will not remove CO2, COS, the sulfur recovery plant, and therefore the effect of
CS2, or mercaptans. Iron is held in dilute solution (high these components must be considered when selecting
circulation rates) by the common chelating agent EDTA the gas treatment process. Further decisions are simpli-
(ethylene diamine tetra acetic acid). The iron oxidizes fied by considering gas composition and operating con-
the H2S to sulfur. The solution is circulated batch-wise ditions. For example, high partial pressures (50 psi) of
to an oxidizer for regeneration. acid gases increase the likelihood of using the physical
Chemsweet, from C.E. Natco, is another H2S only solvent method. Alternatively, the presence of significant
process. It uses an aqueous dispersion of zinc oxide and quantities of heavy hydrocarbons in the feed prevents the
zinc acetate to oxidize H2S and form zinc sulfide. The use of physical solvents. Generally, low partial pressures
process can handle H2S up to 100 grains/100 scf at of acid gases and low outlet specifications would require
pressures from 75–1400 psig. Mercaptan concentrations the use of amines for adequate treatment. Table 11-17
above 10% of the H2S concentration can cause problems provides a summary for a number of processes.
by forming zinc mercaptides [Zn(SH)RH]. The resulting
zinc mercaptide will form sludge and possibly cause Chemical Reactions
foaming problems. The overall chemical equilibrium reactions applicable for
H2S and CO2 reacting with primary and secondary
Process Selection amines are:
The treatment of gas can affect the design of the entire For hydrogen sulfide H2S removal:
gas processing facility, including the methods chosen for

sulfur recovery and acid gas removal, liquids recovery, RNH2 þ H2 S4RNHþ
3 þ HS Fast (11-29)
liquids fractionation and liquid product treating. De-
RNH2 þ HS 4RNHþ
3 þS

Fast (11-30)
signers should consider the following factors when
selecting a gas treatment process: H2 Sg 4H2 SðSolutionÞ (11-31)
Specification of the acid gas. H2 SðSolutionÞ 4Hþ þ HS (11-32)
Hydrocarbon composition of the gas.
Type and concentration of impurities in the H2 O4Hþ þ OH (11-33)
sour gas. ppðH2 SÞ ¼ HðH2 SÞ ½H2 S (11-34)
Temperature and pressure at which the sour gas is
available, and at which the sweet gas must be For carbon dioxide (CO2) removal:
delivered.

Volume of gas to be processed. 2RNH2 þ CO2 4RNHþ
3 þ RNHCOO Fast
Specification of the residue gas. (11-35)
Selectivity required for acid gas removal.
RNH2 þ CO2 þ H2 O4RNHþ
3 þ HCO3 Slow
Air pollution regulations regarding sulfur removal
and/or tail gas clean-up (TGCU). (11-36)
Capital and operating costs. RNH2 þ HCO þ
3 4RNH3 þ CO
3 Slow (11-37)
Liquid product specifications.
H2 O þ CO2 42Hþ þ CO
3 (11-38)
Royalty costs for the process.
þ
Process selection and economics depend on knowing all H2 O4H þ OH (11-39)
components present in the gas. Impurities such as COS, ppðCO2 Þ ¼ HðCO2 Þ ½CO2 (11-40)
CS2 and mercaptans can present significant impact on
the design of both the gas treating and downstream The above reactions proceed to the right at high pres-
processing facilities. Further, if the gas processing sures and/or low temperatures and to the left at high
equipment is to be used together with the recovery of temperatures and/or low pressures. Secondary amines
liquids, then the requirements for H2S, CO2, and mer- undergo similar reactions to primary amines. Tertiary
captans removal may be affected. amines can only react with CO2 via the acid/base re-
With sulfur recovery, the composition of the acid action in Equation 11-29. The CO2 reaction is slowed by
gas stream feeding the sulfur plant must be considered. the time required to dissolve CO2 and its conversion to
CO2 concentration >80% in the acid gas may require bicarbonate. This may be the reason that tertiary amines
303
Table 11-17 Process Capabilities for Gas Treating
Normally capable Removes mercaptans Selective H2S Solution

of meeting ¼ grainsyy H2S and COS sulfur removal degraded (by)
Monoethanolamine Yes Partial No Yes (COS, CO2, CS2)
Diethanolamine Yes Partial No Some (COS, CO2, CS2)
DiglycolamineÒ Yes Partial No Yes (COS, CO2, CS2)
Methyldiethanolamine Yes Slight Yesz No
SulfinolÒ Yes Yes Yesz Some (CO2, CS2)
Selenol Yes Slight Yesz No
Hot Pot-Benfield Yes* Noþ No No

**
Fluor Solvent No No No No
Iron Sponge Yes Partial Yes –
Mol Sieve Yes Yes Yesz –
Stretford Yes No Yes CO2 at high concentrations
Lo-Cat Yes No Yes CO2 at high concentrations
Chemsweet Yes Partial for COS Yes No
*
Hi–pure version.
D
Hydrolyzes COS only.
z
Some selectivity is exhibited by these processes.
**
Can be met with special design features.
zz
¼ grain H2S /100 scf z ppm H2S
Source: Gas Processor Suppliers Association, Engineering Data Book, Vols. 1 and 2, 12th Ed., (2004).
exhibit a greater selectivity for H2S in the presence of the top of the contactor being used as the absorption
CO2. section, with make-up water to the unit being used as the
wash liquid. The sweet gas exiting the contactor is sat-
urated with water, so dehydration is required prior to the
Process Description sale of gas. The amine-rich solution leaving the contactor
The process flow diagram for an alkanolamine treating flows through a flash tank to remove absorbed hydro-
plant is shown in Figure 11-15. carbons. From there, the solution passes through the
Here, the sour gas enters through an inlet separator for rich/lean exchanger where heat is absorbed from the lean
the removal of liquids and/or solids. From the separator, solution. The heated, rich amine then passes to the top
the gas stream enters the bottom of the contactor where section of the stripper. As the solution flows down the
it contacts the amine solution flowing down from the top column to the reboiler, it is stripped of H2S and CO2.
of the column. The acid gas components in the gas react The amine solution then passes through the rich/lean
with the amine to form a regenerable salt. As the gas exchanger and a solution cooler, to reduce the lean so-
continues to pass up the contactor, more acid gases lution temperature to approximately 10 F above that the
chemically react with the amine. The sweetened gas inlet gas in order to stay above the hydrocarbon dew
leaves the top of the contactor and passes through an point. The process is repeated as the lean solution is
outlet separator to trap any solution which may be car- returned to the contactor.
ried over. Where MEA is the sweetening agent, or the Finally, acid gas stripped from the amine exits from
contactor is operating at unusually high temperatures, the top of the stripper, where it passes through a con-
a water wash may be used to attempt to recover some of denser to cool the stream and to recover water. The re-
the vaporized amine from the gas leaving the contactor. If covered water is usually returned to the stripper as
a water wash is used, it consists of three to four trays at reflux. The acid gas passes through an outlet separator
304
Sweet gas
Acid gas
Outlet separator
Condenser
Cooler
Rich/lean
exchanger
Outlet separator
Contactor
Stripper
Off gas
Reclaimer
Flash tank
Reboiler
Sour gas
Lean solvent booster

Inlet separator
Figure 11-15 Typical gas sweetening by chemical reaction.
and, depending on composition and quantity, is vented, a depropanizer column, yielding a liquid product with
sent to sulfur recovery plants, or compressed for sale. specified propane content.
UniSimÒ simulator (gas-processing USC) is used to
model a natural gas processing facility that uses propane
11.15.9 Process Description of Gas refrigeration to condense liquids from the feed, and
Processing Using UniSimÒ Simulation a distillation tower to process the liquids. Figure 11-16
Software show a snap-shot of the flowsheet, and Figures 11-17
shows the phase envelope of the feed having the bubble
A Case Study and the dew points. Figure 11-18 shows the hydrate line
A natural gas stream containing N2, CO2 and C1–nC4 is intersecting the phase envelope. The critical temperature
processed in a refrigeration system to remove the heavier and pressure of the combined feed are 12.28 C and 9976
hydrocarbons. The lean, dry gas produced will meet kPa respectively. The combined feed stream enters an
a pipeline hydrocarbon dew point specification. The liq- inlet separator, which removes the free liquids. Over-
uids removed from the rich gas are processed in head gas from the SEPARATOR is fed to the gas/gas
Figure 11-16 Process flow diagram of a gas processing unit.
305
Figure 11-17 Phase envelope of natural gas feed.
exchanger, where it is pre-cooled by already refrigerated hydrocarbons condense such that the eventual sales gas
gas. The cooled gas is then fed to the chiller, where fur- meets a pipeline dew point specification. The cold stream
ther cooling is accomplished through exchange with is then separated in a low temperature separator (LTS).
evaporating propane (e.g. C3 Duty stream). In the The dry cold gas is fed to the gas/gas exchanger and then
chiller, which is modeled as a COOLER, enough heavier to sales, while the condensed liquids are mixed with free
Figure 11-18 Phase envelope of natural gas feed.
306
liquids from the inlet separator. These liquids are For DEA (conventional):
processed in a depropanizer column to produce a low
propane content bottoms product. Figures 11-19 and 11- GPM ¼ 45 ðQy=xÞ (11-42)
20 show the phase envelope of the sales gas and with the
In S.I. units:
hydrate line outside the boundary of the phase envelope
having critical temperature and pressure as –33.93 C and
Flow ðm3=hÞ ¼ 360 ðQy=xÞ (11-42a)
8070 kPa. Results of simulation of natural gas processing
(gas-processing. HSC) are shown in Appendix L. (0.5 mol acid gas pick-up per mole DEA assumed)
For DEA (high loading):
Simplified Design Calculations for MEA, DEA
and DGAÒ GPM ¼ 32 ðQy=xÞ (11-43)
The following equations are used for making rough es- In S.I. units:
timates of the principal parameters for conventional
MEA, DEA and DGAÒ amine treating facilities, when Flow ðm3=hÞ ¼ 256 ðQy=xÞ (11-43a)
both H2S and CO2 are both present in the gas. The
(0.7 mol acid gas pick-up per mole DEA assumed)
procedure involves calculating the amine circulation rate
For DGAÒ:
and using it as the principal variable in estimating other
parameters [290]. GPM ¼ 55:8 ðQy=xÞ (11-44)
For MEA:
In S.I. units:
GPM ¼ 41 ðQy=xÞ (11-41)
Flow ðm3=hÞ ¼ 446 ðQy=xÞ (11-44a)
In S.I. units:
Flow ðm3=hÞ ¼ 328 ðQy=xÞ (11-41a) (0.39 mol acid gas pick-up per mole DGAÒ assumed)
(DGAÒ concentrations are normally 50-60% by
(0.33 mol acid gas pick-up per mole MEA assumed) weight)
Figure 11-19 Phase envelope of Sales gas.
307
Figure 11-20 Phase envelope of Sales gas.
where should not be used if the combined H2S and CO2 con-
centration in the gas is above 5 mole %. They also are
Q ¼ Sour gas to be processed, Mescd, MSm3/day limited to a maximum amine concentration of about 30%
x ¼ Amine concentration in liquid solution, wt%, by weight.
mass % The diameter of an amine plant contactor is:
y ¼ Acid gas concentration in sour gas, mole %
0:5
After the amine circulation has been estimated, heat and Q
Dc ¼ 44 pffiffiffi (11-45)
heat exchange requirements can be calculated from the P
information in Table 11-18; the pump power re-
quirements can be estimated from Table 11-19. In S.I. units:
0:5
Equations 11-41 to 11-44a provide conservative Q
Dc ¼ 10750 pffiffiffi (11-45a)
(high) estimates of required circulation rates. They P
Table 11-18 Estimated Heat Exchange Requirements
Duty Area
Btu/hr kW Sq. ft. m2
Reboiler (Direc fired) 72,000 GPM 93 m3/h 11.30 GPM 4.63 m3/h
Rich-Lean Amine HEX 45,000 GPM 85 m3/h 11.25 GPM 4.60 m3/h
Amine cooler (air cooled) 15,000 GPM 19.3 m3/h 10.20 GPM 4.18 m3/h
Reflux condenser 30,000 GPM 38.6 m3/h 5.20 GPM 2.13 m3/h
Source: Gas Processor Suppliers Association, Engineering Data Book, Vols. 1 and 2, 12th Ed., (2004).
308
Reboiler
Table 11-19 Estimated Power Requirements
H ¼ 72,000 216 ¼ 15.6 106 Btu/hr
Main Amine GPM PSIG m3/h kPa (ga) A ¼ 11.30 216 ¼ 2441 ft2
Solution Pumps 0.00065 ¼ hp 0.00031 ¼ kW Rich- Lean amine exchanger
H ¼ 45,000 216 ¼ 9.72 106 Btu/hr
Amine Booster GPM 0.06 ¼ hp m3/h 0.20 ¼ kW
Pumps A ¼ 11.25 216 ¼ 2430 ft2
Amine cooler
Reflux Pumps GPM 0.06 ¼ hp m3/h 0.20 ¼ kW H ¼ 15,000 216 ¼ 3.24 106 Btu/hr
A ¼ 10.2 216 ¼ 2203 ft2
Aerial Cooler GPM 0.36 ¼ hp m3/h 1.20 ¼ kW
Reflux condenser
Source: Gas Processor Suppliers Association, Engineering Data Book, Vols. 1 and
H ¼ 30,000 216 ¼ 6.48 106 Btu/hr
2, 12th Ed., (2004). A ¼ 5.20 216 ¼ 1123 ft2
Power requirements from Table 11-19
Main amine pumps, hp ¼ 216 900 0.00065 ¼ 126.4
Amine booster pumps, hp ¼ 216 0.06 ¼ 13
where
Reflux pumps, hp ¼ 216 0.06 ¼ 13
Dc ¼ Contactor diameter in inches (before rounding up Aerial cooler, hp ¼ 216 0.36 ¼ 78
to the nearest 6 inches), or contacting diameter in Contactor diameter from Equation 11-45
mm (before rounding up to nearest 100 mm). 0:5
Q ¼ MMscfd gas to contactor, MSm3/day gas to Dc ¼ 44 pffiffiffi
Q
contactor. P
P ¼ Contactor pressure, psia, kPa(abs). 0:5
40
¼ 44 pffiffiffiffiffiffiffiffiffiffiffiffi
The diameter of the regenerator is: 914:7
pffiffiffiffiffiffiffiffiffiffiffiffi ¼ 50:6 inches or 48 inches rounded down:
Dr ¼ 3:0 GPM (11-46)
Regenerator diameter below feed point:
In S.I. units:
pffiffiffiffiffiffiffiffiffiffiffiffi
qffiffiffiffiffiffiffiffiffiffiffiffi Dr ¼ 3:0 GPM
Dr ¼ 160 m3 =h (11-46a) pffiffiffiffiffiffiffiffi
¼ 3:0 216 ¼ 44:1 inches or 48 inches ðbottomÞ
where rounded up to nearest 6 inches:
Dr ¼ Regenerator bottom diameter in inches, mm.
Regenerator diameter above feed point:
The diameter of the section of the still above the feed
Dra ¼ 0.67 48 ¼ 32.2 inches or 36 inches (top)
point can be estimated as 0.67 times the bottom
rounded up to nearest 6 inches.
diameter.
Example 11-2 11.16 Gas Dryer (Dehydration)

40.0 MMscfd of gas available at 900 psig, and containing
Design
0.5% H2S and 2.5% CO2 is to be sweetened using 25% by
weight DEA solution. If a conventional DEA system is to
Natural gas and associated condensate are often pro-
be used, what amine circulation rate is required, and
duced from a reservoir saturated with water. Addition-
what will be the principal parameters for the DEA
ally, the gas and condensate often contain CO2 and H2S
treatment system?
which require removal. Also, many sweetening agents
employ aqueous solutions for treating the gas. De-
Solution
hydration is the process used to remove water from
Using Equation 11-42, the required solution circulation is: natural gas and natural gas liquids (NGLs) and it is
necessary to:
GPM ¼ 45$ðQy=xÞ ¼ 45 ð40:0 3:0=25Þ
Prevent the formation of hydrates and condensation
¼ 216 gallons of 25% DEA solution per minute: of free water
Prevent corrosion
Heat exchange requirements are shown in Table Meet a water content specification.
11-18
309
Dehydrating the natural gas after the sweetening process streams, dehydration of gases and in gas purification to
involves: remove sulfur dioxide from a stack gas. In addition, ad-
Dehydrating by refrigeration sorption is also employed to fractionate fluids that are
Dehydration with CaCl2 difficult to separate by other methods. Porous desiccants
(adsorbents) with a large internal surface area are used
Dehydration by distillation
for industrial applications, as the amount of adsorbate
Dehydration by membrane permeation
collected per unit area is very small.
Dehydration by gas stripping The most common commercially use desiccants are
Absorption by liquid desiccants alumina, silica gel, and molecular sieves. The last of
Adsorption by solid desiccants. which posseses the highest water capacity, will give the
lowest water dew points, and can be applied to sweeten
Adsorption is a process that involves the transfer of a ma-
dry gases and liquids.
terial from one phase to a surface where it is bound by
Figure 11-21 shows how desiccants can be used in de-
intermolecular forces, and involves the transfer from a gas
hydrators containing two or more towers and Table 11-20
or liquid to a solid surface. It could also involve transfer
presents the most important properties of commercial
from a gas to a liquid surface. The adsorbate is the material
solid desiccants. Figure 11-21a shows a typical geom-
being concentrated on the surface and the material that it
etry of solid desiccant dryer. One of the towers is in
accumulates is known as the adsorbent.
operation, using steam to adsorb water from gas, while
The oil and chemical industries use adsorption pro-
the other tower is regenerated and cooled. The regen-
cesses in the cleanup and purification of wastewater
erated gas is heated to a range of 450 F to 600 F,
Regeneration gas FRC
Regeneration
gas compressor Water knockout
Regeneration gas 450 to 600°F
Regeneration
gas cooler
Inlet separator
Water
Wet
feed gas
Regenerating
Adsorbing and cooling
Regeneration
gas
600°F Dust filter
Regeneration Dry gas

gas heater
Valve open
Valve closed
Figure 11-21 Solid desiccant dehydrator twin tower system.
310
Table 11-20 Typical Desiccant Properties
Bulk density, Heat capacity, Approx. minimum moisture

Desiccant Shape lb/ft3 Particle size Btu/(lb. 8F) content of effluent gas
Alumina Alcoa F200 Beads 48 7 14 Tyler mesh 0.24 90 F dew point
1/8’’ / 3/16’’ / ¼’’
Activated Alumina UOP A-201 Beads 46 3-6 mesh or 5-8 mesh 0.22 5–10 ppmv
Mol Sieve Grace – Davison 4A Beads 42–45 4-8 mesh or 8-12 mesh 0.23 0.1 ppmv (150 F)
Molecular Sieve UOP 4A-DG Extrudate 40–44 1/8’’ or 1/16’’ 0.24 0.1 ppmv
Mole Sieve Zeochem 4A Beads 45–46 4-8 mesh or 8-12 mesh 0.24 0.1 ppmv
Silica Gel SorbeadÒ-R Beads 49 5 8 mesh 0.25 60 F dew point
Silica Gel SorbeadÒ -H Beads 45 5 8 mesh 0.25 60 F dew point
Silica Gel Sorbead($)-WS Beads 45 5 8 mesh 0.25 60 F dew point
Source: Engineering Data Book: Gas Processor Suppliers Association, 12th Ed., Tulsa, Oklahoma, 2004
depending on the type of solid desiccant and the nature The basic design procedures for dehydrating a satu-
of service [292]. rated natural gas at a specified dew point are:
The use of solid desiccant is limited to applications 1. Determine the process conditions and the dryer
with very low water dewpoint requirements, where process flow diagrams.
simultaneous control of water and hydrocarbon
dewpoints are required, and in very sour gases. In 2. Select the drying cycle and calculate the water load.
cryogenic plants, solid-desiccant dehydration is much 3. Select the desiccant type and compute the capacity
preferred over methanol injection, as it prevents and volume required.
hydrate and ice formation. 4. Size the dryer and check for pressure drop.
5. Calculate the desiccant reactivation heating and
a cooling requirements.
Regeneration cycle inlet
Here, design procedures are presented for sizing a solid
2.0 ft.
desiccant dryer for removing moisture from gas streams.
1.3 ft.
The process involves the following calculations.
Floating screen The water pickup
Desiccant volume
7.0 ft. Desiccant bed
Dump port Gas velocity through the bed
Vessel weight
2.0 ft. Dryer regeneration requirements.
Support grating
The design is based upon the following assumptions
Baffle plate [292]:
1. The temperature difference between the heater
Adsorption cycle Inlet
outlet temperature and the peak vessel outlet tem-
1.0 ft. perature is 50 F.
2. The average bed temperature is based on 75% of the
4.0 ft.
bed at heater outlet temperature and 25% at the
Figure 11-21a Geometry of a typical solid desiccant dryer. peak vessel outlet temperature Equation 11-56.
311
3. Specific heats: steel, 0.12 Btu/lb F, and desiccant Superficial gas velocities for adequate contact time
0.25 Btu/lb F. should range between 30 and 60 ft/min.
4. Heat of water desorption 1400 Btu/lb H2O Dryer shell thickness, inch:
adsorbed. PR
t¼ (11-54)
5. Flat heads used on vessel ends; steel density is SE 0:6P
480 lb/ft3.
Vessel weight:
6. Total vessel weight increased by 10 per cent for
supports. t
WVES ¼ 480 p D ðL þ h þ RÞ ð1:10Þ (11-55)
7. Heat losses to the dryer heating period calculated at 12
5%.
Add an extra 2 ft (h ¼ 2) of straight side (i.e. for 6 in. of
3/8 in. inert balls above and below the desiccant bed and
11.16.1 The Equations 1 ft for an inlet-gas distributor).
Temperature of desiccant bed, F:
The equations used for calculating these requirements
are: TBED ¼ 0:75 ðHTÞ þ 0:25 ðHT 50Þ (11-56)
The total water adsorbed, lb:
The total heat required to regenerate the dryer (ideal),
ðFlowÞðCycleÞðH2 O ContÞ Btu:
H2 O load ¼
24
HTOT ¼ ðWVES Þð0:12ÞðTBED TÞ
MM std ft3 lb
ðhÞ
day MM std ft3 þ ðWD Þð0:25ÞðTBED TÞ
¼
ðh=dayÞ
(11-47) þ ðH2 O loadÞð1400Þ ð1:05Þ (11-57)
Desiccant volume:
where:
ðH2 O loadÞð100Þ
VDES ¼ (11-48)
ðpickupÞ ðrD Þ WD ¼ ðrD Þ ðVDES Þ (11-58)
The calculated area based on desiccant volume: Loss factor for non-steady state heating, F:

VDES HT T
ADES ¼ (11-49) F ¼ ln (11-59)
L 50
The bed diameter: Heat added to regeneration gas to regenerate the desic-
cant bed, Btu:

4ADES 0:5
DBED ¼ (11-50) HREG ¼ HTOT ,F (11-60)
p
The total regeneration gas requirement, lb:
Actual desiccant bed area:
HREG
p D2BED HGAS ¼ (11-61)
A¼ (11-51) H1 H2
4
The actual gas flow rate at flowing conditions, ft3 =min: Standard cubic feet of regeneration gas:
HGAS ð380Þ
14:7 T 1 SCF ¼ (11-62)
Q ¼ ðFlowÞ ðZÞ MW
P þ 14:7 520 1440

T
¼ ðFlowÞ ðZÞ ð19:6314Þ (11-52) 11.16.2 Pressure Drop ðDPÞ
P þ 14:7
Superficial gas velocity, ft/min: The pressure drop is estimated from the desiccant
manufacturer’s data and relevant correlations. In addition,
Q DP can be calculated by Wunder’s [310] graphical corre-
VGAS ¼ (11-53)
A lation of gas superficial velocity against DP in ft water per ft
312
Figure 11-22 Pressure drop for an 8 mesh silica gel desiccant (source: Wunder, J. W., ‘‘How to Design a Natural Gas Drier’’, Oil & Gas
Journ., Aug. 6, pp. 137-148, (1962)).
bed. Wunder used a factor of 1.6 for a fouled bed. DP Desiccant attrition due to high gas velocities is controlled
through a new desiccant bed is initially low. After a short by proper bed design. The superficial gas velocity is used
time in service, DP rises because of bed settling and then to estimate Alcoa’s momentum down flow velocities
slowly increases over the active life of the desiccant, be- [310]. However, desiccant attrition should not be
cause of more settling and some attrition. Figure 11-22 a problem when the momentum number is equal to or
shows DP for an 8-inch silica gel desiccant. These data are less than 30,000. Although this number is based on
based on using air. For other gases, the given values should granular alumina, it can be used for silica gel, which has
be multiplied by ðMW gas=MW airÞ0:9. The pressure crushing characteristics similar to alumina. The mo-
drop for a clean bed is given by: mentum is expressed as:

ft:H2 O 0:4335 psi MW gas 0:9 Momentum ¼ ðVGAS Þ ðMWÞ
P þ 14:7
DP ¼ 14:7
ft:bed ft:H2 O MW air (11-65)
(11-63) 30; 000
where
11.16.3 Fouled Bed
VGAS ¼ superficial gas velocity, ft./min
MW ¼ molecular weight of gas
DP ¼ ðclean bed DPÞ ð1:6Þ (11-64) P ¼ system pressure in psig.
313
Desiccant Reactivation The clean bed DP using Equation 11-63 is:

The desiccant bed is reactivated at as high a temperature
0:9
as permissible to regain optimum capacity and minimum ft:H2 O 0:4335 psi 18
dew points. In natural gas drying service, the normal DP ¼ 0:87 ð15:0 ft:bedÞ
ft:bed ft:H2 O 29
reactivation temperature for silica gel is between 350 F
¼ 3:68 psi
and 400 F. Because the feed gas contains high levels of
pentane-plus materials, Wunder [310] suggested that
reactivation can be carried out at 450 F to desorb the Part 2 Hydrocarbon Absorption and Stripping
heavy hydrocarbons.
At a temperature above 450 F, the attrition rate for (With Contributions by Dr. P. A. Bryant)
silica gel is high. However, occasionally at 600 F, reac- Many operations in petrochemical plants require the
tivation will restore some desiccant’s initial activity thus absorption of components from gas streams into ‘‘lean’’
extending the life of the silica gel. Therefore, the reac-
tivation procedure is designed to periodically heat the Table 11-22 Input Data and Computer Output for Gas Dryer Design
desiccant to 600 F [310].
Data name: Data113. Dat
65.0 700.0 80.0
Example 11-3 0.9228 12.0 47.0
Size a dryer that uses silica gel using the data in 15500.0 1.0 8.0
50.0 15.0 500.0
Table 11-21.
18.0 490.0 225.0
Gas Dryer Design

Solution 6 3
Gas flow rate, 10 std. ft /day: 65.000
Computer program PROG 113 has been developed to Dryer operating pressure, psig.: 700.000
calculate the total regeneration flow, the heat re- Gas flowing temperature, F: 80.0
quirement for dryer regeneration, the momentum and Gas compressibility factor: 0.9228
superficial gas velocity. Table 11-22 illustrates the Drying cycle, hr.: 12.0
Water content of gas, lb/106 std.ft3: 47.
input data and the computer output. The total re- Allowable stress, psi: 15500.0
generation flow is 44649.0 lb and the momentum is Joint efficiency: 1.0
30,584. The heat requirement for dryer regeneration is Useful design capacity of desiccant, (%): 8.0
5.6 106 Btu. The calculated superficial gas velocity is Desiccant density, 1b/ft3: 50.0
34.9 ft/min. Heater outlet temperature, F: 500.0
Molecular weight of regeneration gas: 18.0
From Figure 11-22: Enthalpy of gas before regeneration heater, 490.0
The superficial gas velocity ¼ 35 ft./min. Btu/1b:
Desiccant bed length ¼ 15.0 ft. Enthalpy of gas after regeneration heater, 225.0
System pressure ¼ 700 psig. Btu/1b:
Total water adsorbed, 1b/h.: 1527.500
Desiccant volume, ft3.: 381.8750
Calculated cross-sectional area, ft2.: 25.4583
Table 11-21 Dryer Parameters Bed diameter, ft.: 5.6934
Flow ¼ 65 MM std ft3/day Desiccant H2O cap ¼ 8% Bed radius, ft.: 34.1603
Bed length, ft.: 15.00
P ¼ 700 psig Desiccant density ¼ 50 lb/ft3 Actual gas flow rate, ft3/min.: 889.697
Superficial gas velocity, ft./min.: 34.947
T ¼ 80 F Desiccant bed length ¼ 15 ft Dryer shell thickness, in.: 1.5857
Vessel weight, 1b.: 24767.
Z ¼ 0.9228 Heater outlet ¼ 500 F Desiccant weight, 1b.: 19094.
Temperature of desiccant bed, F: 487.500
H2O Cont ¼ 47 lb/MM std ft3 MW ¼ 18 Total heat required to regenerate dryer, Btu: 0.555954E þ 07
Loss factor: 2.1282
S ¼ 15,500 lb/in2 H1 ¼ 490 But/lb Heat added to regeneration gas, Btu: 0.118320E þ 08
Total regeneration gas requirement, lb: 0.446490E þ 05
E ¼ 1.0 H2 ¼ 225 Btu/lb Standard cubic feet of regeneration gas, 0.942590E þ 06
std. ft3:
Cycle ¼ 12h h ¼ 2.0 ft. Momentum: 0.305838E þ 05
314
oils or solvents. The resultant ‘‘rich’’ oil is then stripped b. Fraction absorbed
or denuded of the absorbed materials. The greatest use of
this operation utilizes hydrocarbon materials, but the YNþ1 Yi ANþ1
i Ai
basic principles are applicable to other systems provided Eai ¼
¼ Nþ1
(11-67)
YNþ1 Yo Ai 1
adequate equilibrium data are available.
Several methods [17, 18, 29, 40, 62, 67, 223] for where Y*o is often considered zero or very small.
handling this design have been offered, and each has in- Solve using Ai values.
troduced a concept to improve some feature. An ap-
proximation method combining Kremser-Brown [40, 67] 7. Mols each component absorbed/hr.
and a more complete method of Edmister [18] will be
summarized. Figure 11-23 summarizes the system and ¼ ðVo YðnþlÞi Þ ðEai Þ
terminology.
8. Mols each component absorbed/(mol inlet lean oil)
ðhrÞ ¼ XiR
11.17 Kremser-Brown-Sherwood
Method - No Heat of Absorption 9. Mols of each component in gas out of top of
absorber:
This method [18] gives reasonably good results for sys- ¼ (mols component in inlet gas) (mols com-
tems involving relatively lean gas and small quantities ponent absorbed)
absorbant. For rich gases the error can be considerable 10. Mols of component in gas out of top of absorber/
(more than 50% for some components). It has given mol of inlet rich gas:
generally good results on natural gas and related systems.
YNþ1 Y1
Eai ¼
YNþ1 Y
o
11.17.1 Absorption: Determine
Component Absorption in Fixed Tray Solve for Y1.
Tower (Adapted in part from Ref. 18) 11. Correct values from first calculation, Steps 1
through 10, using the SXiR values of Step 8, as
1. Calculate the total mols. of gas inlet to the absorber, follows:
identifying the quantities of individual components.
12. Calculate Ai:
2. Assuming the tower pressure as set and that an
average of top and bottom temperatures can be se- X
Lo
lected (these may become variables for study), read Ai ¼ 1þ XiR (11-68)
VNþ1 Ki
equilibrium Ki values from charts for each compo-
nent in gas.
3. Assume or fix a lean oil rate. 13. Calculate absorption efficiency, Eai. Using new Ai
value:
4. Calculate
ANþ1
i Ai
Lo Mols=hr lean oil in Eai ¼ ; read Figure 11-24
¼ (11-66) ANþ1
i 1
VNþ1 Mols=hr rich gas in
Assume this is a constant value for tower design. 14. Calculate mols absorbed/hr:
5. Calculate absorption factor, Ai ¼ Lo/(VNþ1) (Ki), ¼ (Eai) (mols component in inlet rich gas)
using values of (2) and (4) above for each 15. Mols of each component in gas out of top of ab-
component. sorber/hr ¼ (mols component in) (mols compo-
6. Calculate the fraction absorbed for each component absorbed)
nent, assuming a fixed overall tray efficiency for an 16. Mols of component in outlet gas/mol inlet rich gas.
assumed number of actual trays (or an existing Solve for Y1i
column with trays).
a. Theoretical number of trays, N ¼ (tray efficiency,
YNþ1 Y1i
Eai ¼
Eo,) (number of actual trays) YNþ1 Y
o
315
2. Calculate:
If the X1 in equilibrium with Y1 is desired: a. Mols component in inlet gas/hr
P b. Mols in outlet gas
Y1 ð1 þ Xir Þ
X1 ¼ P (11-69)
K Y1 100 ð%recoveryÞðtotal mols inÞ
¼
100
17. Improved values can be obtained by recalculation
from Step 11 if there is too great a difference be- c. Mols component absorbed ¼ inlet outlet
tween the ‘‘S mols absorbed’’ from trial no. 1 and
3. Calculate: Eai for specified component (specified
trial no. 2.
in 1.)
First trial Mols component in Mols component out

Eai ¼
Mols component in
L
Component Inlet Y(ND1)i Mols/hr In Ki A ¼ VKi ¼ specified fraction recovery
4. Minimum (L/V) for specified component:

P
Y
ðLo =VNþ1 Þmin ¼ KEa P
1.00 S 1þ X
Fraction
Absorbed Mols Mol/hr Assuming equilibrium at bottom, SY ¼ 1. Ignoring
Component Eai Absorbed XiR Off Gas Y1i(out) SX gives slightly conservative value,

Lo
¼ Ki Eai
VNþ1 min

5. Operating (Lo/VNþ1)o
S S S ¼ (specified unit) (Lo/VNþ1)min
6. Operating
For second trial, see Step 11.
18. A graphical stepwise is procedure offered by Lo 1
Aio ¼ (11-71)
Sherwood [62] and also summarized by Reference VNþ1 o K i
18. Y and X are plotted and handled stepwise as in
7. Theoretical number of plates at operating (Lo/VNþ1):
distillation. The equilibrium line equation is, for any
solve for N.
single component:
P ANþ1 Aio
Yi Yi Eai ¼ io
(11-72)
¼ Ki P (11-70) ANþ1 1
Xi 1 þ Xi io

ðAio Eai Þ
For a complete denuded inlet solvent at the top ðN þ 1ÞlogðAio Þ ¼ log
ð1 Eai Þ
SX ¼ 0, using K at top column conditions. The slope
of the operating line ¼ Lo/VNþ1 ¼ mols lean oil
entering/mols wet gas entering. 8. Actual number of trays at operating (Lo/VNþ1):
No ¼ N/Eo
Eo values may be calculated from Figure 10-54 or
11.17.2 Absorption: Determine the assumed at 20 to 50% as an estimating value for
Number of Trays for Specified Product hydrocarbon oil and vapors, pressures atmospheric
Absorption to 800 psig, and temperatures of 40 F to 130 F (see
Table 11-23).
1. For fixed tower temperature, pressure, gas feed
rate, specified or assumed operating (Lo/VNþ1) 9. Lean oil rate:
times minimum value, specified component recov-
ery from inlet gas. Lo ¼ ðAi ÞðKi ÞðVNþ1 Þo ; mols=hr (11-73)
316
transfer. The condensing pressure (and column op-

Table 11-23 Absorption-Stripping Approximation Tray Efficiencies**
erating pressure) may be dictated by the available
Pressure Temp., Range steam pressure used in stripping, or the pressure on
Type range, psig 8F efficiency % the inert stripping gas.
2. From the K charts, determine Ki values for each
Absorption
component at the column temperature and
Hydrocarbon 0–800 30–130 35–50 pressure.
Oils & Vapors 3. From a fixed percentage of recovery for key com-
Propane-key 100–2,100 – 30–37*–38 ponent (¼ Esi for key component), mols component
stripped/hr ¼ Gmi ¼ (Lmþ1) (Xmþ1) (Esi)
Butane-key 100–2,100 – 28–33*–36
4. Estimate stripping efficiency for components other
Stripping than the key by:

Hydrocarbon Oils 0–130 300–550 50–80 Kkey
with Steam Es2 ¼ Eskey (11-75)
K2
Unsaturates in Oil 0–50 – 25–35
with closed reboiler
* Average value Assume known SXi [
** Hull, R. J. and K. Raymond, Oil and Gas Journal, Nov. 9, 1953 – March 15,
1954 [32], used by permission. V o Lo Vo/Lo (LXi)mD1/Lo 1DSXi Es, key (1DSXi)
Same
for
all
10. For other components: Eai is estimated by trials
SYi [
K1
Eai ¼ ðEai Þ ; (11-74)
K2 Es; key ð1 þ SXi Þ
SGmi Kkey
with a limiting value of unity. (Step 3)/Vo 1 D SYi (1 D SYi) Ekey ð1 þ SYi Þ

11.17.3 Stripping: Determine the
Number of Theoretical Trays and Stripping
Steam or Gas Rate for a Component
Recovery [18] Note that no recovery can be greater than 1.0, there-
fore any value so calculated is recorded as 1.0, indi-
The rich gas from the absorption operation is usually cating that the component is completely stripped
stripped of the desirable components and recycled from the rich oil. Calculate mols stripped per hour
back to the absorber (Figure 11-23). The stripping for each component as in Step 2.
medium may be steam, or a dry or inert gas (methane,
5. The minimum stripping medium (steam or gas) to
nitrogen, carbon oxides, hydrogen, etc.). This depends
lean oil ratio is estimated by a trial and error pro-
upon the process application of the various
cedure based on key component: By assuming sev-
components.
eral values of Vo, plot Vo/Lo versus Es, Key (1 þ SXi)/
1. The rich oil flow rate and absorbed component KKey (1 þ SYi). The point where they are equal
compositions (this is the only composition of con- gives the minimum value for Vo/Lo. This calculation
cern, not the oil composition, unless reaction or can be thought of as assuming equilibrium at the gas
change takes place under the system conditions) are outlet end and being slightly conservative by in-
known. From the temperature levels of the available cluding the (1 þ SXi) term. Operation at this point
condensing fluids (water, refrigerant, etc.), deter- requires infinite number of plates; therefore, values
mine a column operating pressure which will allow larger than the minimum should be used. For eco-
proper condensation of the desirable components at nomical as well as reasonable operation, several
the selected temperature, allowing for proper values of (Vo/Lo)oper should be tried and corre-
temperature difference ðDTÞ for efficient heat sponding plates evaluated.
317
V0 Ym
Off gas
Vn+1 Y1
L0 X0 Lm+1 Xm+1
1 m Condenser
2 m–1
3 m–2
Pre-heater
Raw gasoline
Cooler
To finishing
Lean oil
n–2 3
n– 1 2
n 1
Wet gas Stripping stream

Vn+1 Yn+1 V0 Y0
L0 Xn Lm+1 X1
Rich oil
Exchanger
Absorber Stripper
Figure 11-23 Flow diagram of absorption-stripping for hydrocarbon recovery from gaseous mixture. Used by permission, Edmister, W.
C., Petroleum Engr., Sept. (1947) to January (1948).
Vo (operating) ¼ (assumed (Vo/Lo)oper) (Lo inlet), 8. Actual number of trays at operating reflux:
mols/hr
Mo
6. Calculate Si for the key component, using the value Mact ¼ (11-79)
Eo
of (1 þ SXi) calculated in Step 5. Calculate
9. Calculate for each component corrected amount
X X Ymi ðStep 3Þ stripped:
ð1 þ Xi Þ ¼ 1 þ For each component:
Vo ðoperÞ
P P
Vo 1þ Y Vo 1þ Y
then; Sio ¼ Ki P i (11-76) Si ¼ K i P i (11-80)
Lo oper 1 þ Xi Lo oper 1 þ Xi
ðVo =Lo Þoper ¼ fixed in step 5:
Sometimes the last term on the right can be X X
neglected. 1þ Yi and 1 þ Xi come from step 6:
7. Calculate number of theoretical trays, M.
10. From Figure 11-24, read (SM þ 1 S)/(SM þ 1 1) ¼
SMþ1 Sio Esi for each component at the fixed theoretical
io
Esi ¼ (11-77) required trays and at each individual Si value.
SMþ1
io 1
11. For final detail, recalculate mols stripped per hour
from new Esi values and the total quantities of
Sio Esi
ðMo þ 1ÞlogSio ¼ log (11-78) each component in the incoming rich oil. If values
1 Esi do not check exactly, adjustments can be made in
318
Figure 11-24 Absorption and stripping factors, Ea or Es vs. effective values Ae or Se (efficiency functions). Used by permission, Gas
Processing Suppliers Association, Engineering Data book, vol. 12, 12th ed., Tulsa, Oklahoma (2004).
steam rate and SYi to give exact values. In many Operating line:
cases this accuracy is not justified since the
method is subject to some deviation from theo- Slope ¼ Lo =Vo
retically correct values. Mols lean oil leaving stripper
¼
12. A graphical solution is presented by Edmister [18] Mols stripping steam ðor gasÞ entering
and handled like step-wise distillation.
At lean end, Yi ¼ 0 (or nearly so in most cases); if not,
Equilibrium line: starts at origin of X–Y plot. plot accordingly.
For assumed X values, calculate Y corresponding for key
component from 11.17.4 Stripping: Determine Stripping-
P Medium Rate for a Fixed Recovery [18]
Xi 1 1þ X
¼ P i
Yi Ki 1 þ Yi
1. The composition and quantity of rich oil, and per-
At lean oil end of tower: SXi ¼ 0 and SYi ¼ 0. cent recovery of a specified key component are
Slope of equilibrium line is Y/X ¼ Ki known, also column pressure and temperature.
At rich oil end of tower: 2. Using Figure 11-24, assume a value for theoretical
P plates, read Se corresponding to specified value of
1þ Y
Yi ¼ Ki Xi P i (11-81) recovery Esi for key component, since:
1þ Xi R
where Xi, SXi and SYi are known. R ¼ rich end. Es ¼ ðSMþ1 SÞ=ðSMþ1 1Þ
319
Note that with this procedure, the effect of the 11.18 Absorption: Edmister Method
number of theoretical plates available can be deter-
mined. In an existing column where the number of
This method [18] is well suited to handling the details of
trays is fixed, the theoretical trays can be obtained by
a complicated problem, yet utilizing the concept of av-
evaluating an efficiency for the system.
erage absorption and stripping factors. It also allows for
3. The value of Se ¼ Si for key component obtained in the presence of solute components in the solvent and the
Step 2 is equal to: loss of lean oil into the off gas. Reference 18 presents
P more details than are included here. It is Edmister’s
Vo 1þ Y original publication of the basic method for absorbers and
Si ¼ K i ¼ P i
Lo 1þ Xi strippers, and it also generates the treatment to include
distillation towers and presents the same graphical re-
Using key component: lationships in a slightly modified form.
P
Vo 1 þ Y Si
P i ; known Absorption: Lean Oil Requirement for Fixed
Lo 1 þ Xi Ki
Component Recovery in a Fixed Tower [18]
1. The rich gas is known, the number of theoretical trays is
Set up table: use Ki for each component to calculate fixed (or assumed and corresponding result obtained),
Column 4. the operating pressure and temperature can be fixed.
2. For key component and its fixed recovery, Ea, read
Mols/
Ae from Figure 11-24 at the fixed theoretical
hr in Vo
K
ð1þSYi Þ number of trays, N:
Compo- Rich Ki at Col. i L ð1þSXi Þ
Mols/hr
o
nent Oil Condition Esi Stripped
ANþ1
e Ae
Ea ¼
ANþ1
e 1

3. Assume:
S S a. Total mols absorbed
b. Temperature rise of lean oil (normally 20–40 F)
From values of Si calculated (¼Se), read Esi values
from Figure 11-24 at the number of theoretical c. Lean oil rate, mols/hr, Lo
trays assumed in Step 2. Note that the Se corresponds 4. Using Horton and Franklin’s [29] distribution rela-
to the number of trays selected, hence will give a value tion for amount absorbed (or vapor shrinkage), per
for performance of the system under these particular tray:
conditions.
1=N 1=N
4. Calculate the mols of each component stripped/hr V1 V1 Vi
¼ or: ¼ ¼
VNþ1 VNþ1 Viþ1
¼ ðLMþ1 XMþ1 ÞR ðEsi Þ (11-82)
5. Calculate Vo, mols/hr of stripping medium required Mols off gas leaving top tray.
(steam or gas): ¼ V1 ¼ VNþ1 – Mols absorbed (assumed)
Mols gas leaving bottom tray No. N
P
Vo 1 þ Y ¼ VN ¼ VNþ1 (V1/VNþ1)1/N
From Step 3; P is known ðequals Si=KiÞ
Lo 1 þ X
Mols gas leaving Tray No: 2 ðfrom topÞ
for key component. V1
Multiply by Lo. ¼ V2 ¼
V1 1=N
VNþ1
Then; multiply result by
P
ð1 þ mols=hr in rich oil=Lo Þ Liquid leaving top tray No. 1 ¼ L1 ¼ Lo þ V2 – V1
P
ð1 þ mols=hr stripped ðStep 3Þ=Vo Þ where
V2 ¼ vapor leaving tray No. 2 from top, mols/hr
This is equal to Vo. Note that Vo also is in the right hand Lo ¼ lean oil entering (assumed completely free of
side of the denominator, so fractions must be cleared. rich gas components), mols/hr
320
LN ¼ liquid leaving bottom tray, mols/hr 12. If the result does not yield the desired amount of
VN ¼ vapor leaving bottom tray, mols/hr the key component absorbed, then reassume the
Liquid leaving bottom tray lean oil quantity, Lo and recalculate. Adjustments
¼ L4 ¼ Lo þ Mols absorbed (assumed) may have to be made separately or simultaneously
5. Calculate: At top, L1/V1 in the assumed absorption quantity until an ac-
At bottom, LN/VN ceptable result is obtained. After two or three trials
6. Use Horton-Franklin method to estimate tempera- a plot of the key variables will assist in making the
tures at tower trays: proper assumptions.
TN Ti VNþ1 Viþ1
¼ (11-83) Absorption: Number of Trays for Fixed
TN To VNþ1 V1
Recovery of Key Component
where
Here we also consider the more general case when the
To ¼ lean oil temperature, F
lean oil contains some of the components to be absorbed
TN ¼ bottom tray temperature, F
from the entering gas. The relationships are most con-
Ti ¼ tray, i temperature, F
veniently written as follows [18], for a given component:
TNþ1 ¼ inlet rich gas temperature, F
These relations assume constant percent absorption v1 ¼ f s lo þ ð1 f a Þvnþ1 (11-84)
per tray, and temperature change proportioned to the
vapor contraction per tray. For estimating use only. Snþ1
e Se
fs ¼ (11-85)
Temperature bottom tray ¼ TN ¼ TNþ1 þ Snþ1 1
e
(assumed rise)
Anþ1
e Ae
Temperature top tray fa ¼ (11-86)
Anþ1
e 1
VNþ1 VTray 2
¼ TN ðassumed riseÞ
VNþ1 V1 Re-arranging Equation 11-86 gives:
7. Read the K values from equilibrium charts for Ae 1
1 fa ¼ (11-87)
components in feed at temperatures of (a) top tray Anþ1
e 1
and (b) bottom tray.
Substituting Equations 11-85 and 11-86 into Equation
8. Calculate ATi and ABi for each component.
11-84 gives
ATi (for top conditions) ¼ Li/(KiVi)
ABi (for bottom conditions) ¼ LN/(KiVN) nþ1 nþ1
Se Se Ae Ae
where: v1 ¼ l o þ 1 vnþ1
Snþ1
e 1 Anþ1
e 1
ATi ¼ absorption factor for each component at
conditions of top tray (11-88)
ABi ¼ absorption factor for each component at which can be written as
conditions of bottom tray
Se 1 Ae 1
9. Read Ae values corresponding to ATi and ABi values v1 ¼ 1 nþ1 lo þ nþ1 vnþ1
from Figure 11-25. Se 1 Ae 1
(11-89)
10. Read Eai values for fraction absorbed from
Figure 11-24 using the Ae values of Step 9 and the where
fixed or assumed number of theoretical trays. v1i ¼ molar gas flow rate of component ‘‘i’’ leaving plate
11. Calculate the mols of each component absorbed by: 1 in the absorber
(Mol component in inlet rich gas) (Eai) loi ¼ molar flow of component ‘‘i’’ in entering liquid to
Suggested tabulation: the absorber
Mols Rich K Absorption Factors Eai, Frac. Mols

Component Gas In Top Bottom AT AB Ae Absorbed Absorbed

321
Figure 11-25 Effective absorption and stripping factors used in absorption, stripping and fractionation as functions of effective factors.
Used by permission, Edmister, W. C, Petroleum Engr., Sept. (1947) to Jan. (1948).
vnþ1i ¼ molar gas flow rate of component ‘‘i’’ in entering Lean Oil
gas to the absorber
The selection of lean oil for an absorber is an economic
fsi ¼ the fraction of Vnþli that is absorbed by the liquid study. A light lean oil sustains relatively high lean oil
phase loss, but has the advantage of high mols/gal compared
fai ¼ the fraction of Vnþ1 that is absorbed by the liquid to a heavier lean oil. The availability of a suitable ma-
n ¼ absorber theoretical number of trays, also ¼ N terial has a large influence on the choice. For example,
m ¼ stripper theoretical number of trays, also ¼ M a lean oil three carbon numbers heavier than the
A material balance on the key component fixes v1, lo and lightest component absorbed is close to optimum for
vnþ1 for the key. Ae is estimated after AT and AB are some applications. In natural gas plant operations,
calculated using approximate conditions at the top and a lean oil heavier by about 10-14 carbon numbers has
bottom, with a multiple of the minimum solvent rate, been reported [284].
which is estimated by assuming equilibrium at the
bottom of the tower. Se is estimated from ST and SB.
Note that ST ¼ 1/AT and SB ¼ 1/AB. A trial and error Presaturators
solution for the number of theoretical stages is effected A presaturation to provide lean oil/gas contact prior
by using Equation 11-89 (or 11-88 and Figure 11-24). to feeding the lean oil into the tower is a good way of
Values of v1 for the non-keys can be calculated by using getting more out of an older tower. Absorber tray
these relationships directly the calculated value of n; if efficiencies are low, and therefore a presaturation that
necessary, the entire procedure can be repeated, using achieves equilibrium can provide the equivalent of
the better estimates of the component flow rates in the a theoretical tray. This can easily equal 3-4 actual
leaving streams that were estimated in the first iteration. trays [284].
322
Example 11-4: Absorption of Hydrocarbons The other estimates in column 5 of Table 11-24 are cal-
with Lean Oil culated in a similar manner. Note that the C5s are as-
The gas stream shown in Table 11-24 is fed to an iso- sumed to be completely absorbed for this first iteration.
thermal absorber operating at 90 F and 75 psia. 90% of 2. Inlet rate of rich oil.
the n-butane is to be removed by contact with a lean oil
stream consisting of 98.7 mol% non-volatile oil and the The maximum mol fraction n-C4 in the leaving liquid
light components shown in column 2 of Table 11-24. is taken as that in equilibrium with the incoming
Estimate the composition of the product streams and the gases. Thus, for n-C4,
required number of theoretical stages if an inlet rate of
ln 0:017
1.8 times the minimum is used. ¼ ¼ 0:02615
ðLn Þmin 0:65
Solution A material balance of n-C4 yields:
1. Initial estimates of extent of absorption of non-keys.
ln ¼ 0:002 Lo þ ð0:9Þ 33:6
As a rough approximation, assume the fraction
absorbed of a given component is inversely pro- With the absorption efficiencies assumed above,
portional to its K value (Equation 11-74). For example:
Ln ¼ Lo þ 126:34
n-C4 ; in off-gas ¼ 33:6 ð0:9Þð33:6Þ ¼ 3:36
Combining the above equations yields the estimate of
0:65 the minimum lean oil rate:
C1 ; in off-gas ¼ 1; 639 ð0:9Þ ð1; 639:2Þ
42:5
¼ 1; 616:6 0:002ðLo Þmin þ ð0:9Þ33:6
¼ 0:02615
0:65 ðLo Þmin þ 126:38
C2 ; in off-gas ¼ 165:8 ð0:9Þ 165:8
7:3
or (Lo)min ¼ 1,115.3 mols/hr
¼ 152:5
Thus, Lo ¼ 1.8 (1,115.3) ¼ 2,007.5 mols/hr
Table 11-24 Calculations Summaries for Example 11-4
Off-Gas
Initial
estimate
Lean of net After After After
Feed oil in amount Initial first second third Lean Rich oil
908F K gas in (mol. absorbed estimate iteration iteration iteration oil in out
Component 75 psia (mols/hr) Fraction) (mols/hr) (mols/hr) (mols/hr) (mols/hr) (mols/hr) (mols/hr) (mols/hr)
Methane 42.5 1,639.2 – 22.6 1,616.6 1,597.5 1,598.4 1,598.4 – 40.8
Ethane 7.3 165.8 – 13.3 152.5 141.2 141.8 141.8 – 24.0
Propane 2.25 94.9 – 24.67 70.23 49.84 50.76 50.76 – 44.14
i-Butane 0.88 17.8 0.001 11.83 5.97 3.10 3.13 3.13 1.91 16.58
n-Butane 0.65 33.6 0.002 30.24 3.36 3.36 3.36 3.36 3.83 34.07
i-Pentane 0.28 7.9 0.004 7.9 0 2.08 2.03 2.03 7.66 13.53
n-Pentane 0.225 15.8 0.006 15.8 0 2.51 2.44 2.44 11.49 24.85
Heavy Oil 0 – 0.987 0 0 0 0 0 1,889.83 1,889.83
1,975.0 1.000 126.34 1,848.66 1,799.57 1,801.92 1,801.92 1,914.72 2,087.78
323
3. Effective Absorption Factor for n-C4. 6. Second Iteration.

Using the previously calculated values, the net amount
The total rich oil out is estimated as
absorbed is 1,975 1,799.59 ¼ 175.41 mols/hr. The
Ln ¼ 1,975 þ 2,007.5 1,848.66 ¼ 2,133.84
minimum rate of lean oil is calculated from
The absorption factors are calculated by
AT ¼ 2,007.5/(.65) (1,848.66) ¼ 1.671
ð0:002ÞðLo Þmin þ 0:9ð33:6Þ
AB ¼ 2,133.84/(.65) (1,975) ¼ 1.662 ¼ 0:02617
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ðLo Þmin þ 175:41
Ae ¼ AB ðAT þ 1Þ þ 0:25 0:5
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Ae ¼ 1:662ð2:671Þ þ :25 0:5 ¼ 1:665 from which (Lo)min ¼ 1,061.2 mols/hr and
The stripping factor, Se, is taken as 1/Ae ¼ 0.60006 Lo ¼ 1.8 (1,061.2) ¼ 1,910.2 mols/hr
An overall material balance gives Ln ¼ 2,085.6.
4. Calculation of the required number of theoretical The effective absorption factor for n-C4 is Ae ¼ 1.627,
stages. and Se ¼ 0.6145, n is calculated from
Using Equation 11-88 for n-butane,
0:6145 1
3:36 ¼ 1 3:82
0:6006nþ1 0:6006 0:6145nþ1 1
3:36 ¼ 4:0
0:6006nþ1 1 1:627 1
þ 33:6
1:627nþ1 1
1:665nþ1 1:665
þ 1 33:6
1:665nþ1 1
from which n ¼ 5.20 theoretical number of stages.
which is equivalent to The non-key components are computed and tabulated
in Column 7 of Table 11-24.
0:6006 1 7. Third Iteration.
3:36 ¼ 1 4:0
0:6006nþ1 1
A third iteration gave (Lo)min ¼ 1,063.73, Lo ¼
1:665 1
þ 33:6 1,914.72, Ln ¼ 2,087.8, and V1 ¼ 1,801.92, with no
1:665nþ1 1 change in the calculated off-gas component flows.
Solving for n by trial and error yields, n ¼ 5.12 The stripping calculations are handled in a manner
similar to the steps above, and using the figures
5. Calculation of absorption of non-keys. indicated.
Equation 11-89 is used with n ¼ 5.12 to calculate v1,
as for example, for i-butane,
AT ¼ 1.233 Intercooling for Absorbers
AB ¼ 1.227 Most absorbers require some intercooling between stages
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi or trays to remove heat of absorption, and to provide
Ae ¼ 1:227ð2:333Þ þ 0:25 0:5 ¼ 1:229 internal conditions compatible with proper or required
absorption. Some temperature rise (10–30 F) is usually
1
Se ¼ ¼ 0:8137 designed into the initial conditions. The rise above this
1:229 must be handled with intercoolers.
vo ¼ 17.8 The total intercooler duty is the difference between
lo ¼ (0.001) (2,005.7) ¼ 2.01 the total heat in the rich gas and lean oil, and the total
heat out of the off-gas and rich oil all at the terminal
0:8137 1
v1 ¼ 1 2:01 calculated or design conditions. The total duty is often
0:81376:12 1 divided between several coolers placed to re-cool the oil

1:229 1 as it passes down the column. If intercoolers are not used,
þ 17:8
1:2296:12 1 then the absorption cannot meet the design terminal
¼ 3:097 outlet conditions, and the quantity of material absorbed
will be reduced. If the intercooling is so great as to sub-
ln ¼ 17.8 þ 2.01 – 3.097 ¼ 16.71 cool, then greater absorption may be achieved, but this
The remaining non-keys in the off-gas are calculated in can be controlled by the intercooler operation.
a similar manner and are tabulated in Column 6 of Table A second approach to the same result involves the
11-24. Note that the calculated values are somewhat same requirements as for a balanced ‘‘heat’’ design; the
different from the assumed values in Column 5. heat of absorption of the actual components absorbed
324
must equal the sum of the heat added to the lean oil and
to the lean gas. For hydrocarbon materials these factors
can be developed by using total heats.
The relation of Hull and Raymond [32] considers heat
loss through the column wall, and indicates that either
the total heat of absorption, or the rich oil outlet tem-
perature for system balance can be calculated. Thus, if
a reasonable temperature balance is not obtained, the
heat load for the intercoolers can be set.
WLo CpLo ðTRo TLo Þ þ WDG CpDG ðTDG TIG Þ

þ Ws Cps ðTRo TIG Þ
¼ DHs 0:024 UA00 ðTAV AAMB Þ
(11-90)
Figure 11-27 Hydrocarbon systems; overhead gas minus lean oil
Figure 11-26 presents actual total heats of absorption temperature for components absorbed in top ‘‘theoretical’’ tray (or
based on experimental studies [32]. As long as the hy- top actual three trays). Used by permission, Hall, R. J., and
drocarbon absorption is in the range of 80–120 F, the Raymond, K., Oil & Gas Journ. Nov. 9 (1953) thru Mar. 15 (1954).
values read from the graph should apply.
Estimation of discharge gas temperature may be made
from Figure 11-27 based on test data. faster than the lighter components (lower A values).
The design of absorbers has not received the empirical More heavier materials are also absorbed as the tem-
design guidelines study so prevalent in distillation prob- perature of absorption is lowered. Thus, a cold lean oil
lems. The graph of Hutchinson [34], Figure 11-28, is has more capacity per gallon than a warm oil. This re-
convenient to examine a problem to determine pre- quires a study of the entire process and not just the one
liminarily the effects of design. This curve is compatible unit. Heat economy, oil flows, and tower costs all enter
with the conventional absorption factor graphs. The into a full evaluation of the absorber as it fits into the
percent extraction gives a quick evaluation of the possi- plant system.
bilities of accomplishing the desired absorption. As the Many designs are set up by assuming the number
number of trays in the absorber is increased, the amount of theoretical trays, using the best available in-
of heavier material (larger A values) absorbed increases formation for tray efficiencies and then calculating
the expected performance. A series of such studies
might be made.
11.18.1 Absorption and Stripping

Efficiency
Unfortunately, the efficiencies for tray and overall
column operation are incomplete and nullify to a cer-
tain extent some very high quality theoretical perfor-
mance design. Tray efficiencies may be estimated by
Figure 10-54 or Table 11-23. Table 11-25 gives a sum-
mary of Sherwood’s absorption and stripping calcula-
tion method.
Example 11-5: Determine the Number of Trays

for Specified Product Absorption
A gas stream is to have the ethylene removed by ab-
sorption in a lean oil of molecular weight 160, sp. gr.
0.825. The inlet gas is at 70 psig and 100 F and the oil is
Figure 11-26 Component heats of absorption. Used by at 80 F. The gas rate is 16,000,000 scf/day (60 F). Ex-
permission, Hall, R. J., and Raymond, K., Oil & Gas Journ. Nov. 9 amine the tower performance for 98% ethylene recovery
(1953) thru Mar., 15 (1954). at 1.25 times the minimum Lo/VN.
325
Figure 11-28 Absorption equilibrium curve. Used by permission, Hutchison, A. J. L., Petroleum Refiner, V. 29 (1950), p. 100, Gulf Pub.
Co., All rights reserved.
Feed Gas
Mols ethylene in ¼ (20.5/100) (1,756) ¼ 360 mols/hr
Component Mol or Volume % Mols ethylene in outlet gas ¼ (100 – 98/100)(360) ¼
7.20 mols/hr
H2 18.5
Mols ethylene absorbed in oil ¼ 360 7.2 ¼ 352.8
CH4 22.3
mols/hr
C2H4 20.3
C2H6 0.5 2. Specified ethylene separation ¼ 0:98 ¼ Eai
C3H6 22.0 Mols in Mols out
C3H8 0.7 ¼
Mols in
n- C4H10 2.5
n- C5H12 13.0 3. Minimum L/V for ethylene:

100.0 Lo
¼ Ki Eai
VNþ1 min
Determine the oil rate and the number of theoretical Average tower conditions for K:
and actual trays required. (Note: this example illustrates Temperature ¼ (100 þ 80)/2 ¼ 90 F
that unreasonable results must be examined, not just Pressure: allow 20 psi pressure drop, then top pressure
accepted.) would be 70 20 ¼ 50 psig
Average: (50 þ 70)/2 ¼ 60 psig
Solution K (ethylene) at 60 psig and 90 F ¼ 11.5 (from equi-
16; 000; 000 librium charts).
1. Inlet rich gas rate ¼
14:7 460 þ 60
359 ð24Þ
14:7 460 þ 32 Lo
¼ ð11:5Þð0:98Þ ¼ 11:28
¼ 1; 756 mols=hr VNþ1 min
326
Table 11-25 Summary of Sherwood Absorption and Stripping Calculation Method

This method is a combination of graphical and analytical using ‘‘equi- Subscripts L and R refer to compositions at lean and rich ends, re-
librium line’’ and ‘‘operating line’’ concepts thus illustrating diffu- spectively. Lo, Vo and Vnþ1 refer to lean oil stripping steam and
sional driving force wet gas quantities, respectively
All compositions (X and Y) are referred to constant and known streams, Method is applied in both absorption and stripping calculations by
such as: inlet wet gas to absorber, lean oil to absorber or from making preliminary calculations to permit drawing XY diagram for
stripper, and inlet stream to stripper key component and using graphical results to complete
calculations
Absorption Stripping
Equations: Equations:
X
P
YR YL Anþ1 A L+ XR XL Smþ1 S KV+ 1 þ Y
Ea ¼ ¼ nþ1 ; A¼ ð1 þ XÞ Es ¼ ¼ mþ1 ; S¼ P
YR Y+ A 1 KVnþ1 XR X+ S 1 L+ 1 þ X
The stripping factor S has

P Vo =Lo as P the ratio of stripping medium to
The absorption factor PA has Lo =Vn þ 1 as the ratio of the lean oil to the lean oil and ð1 þ YÞ=ð1 þ XÞ term corrects for the ratio
inlet gas and ð1þ XÞ term changes the lean oil to the rich oil. K is for the transfer components from the liquid to the gas. K is
determined
P at an average oil temperature evaluated
Estimate X and calculate the amounts of each component absorbed P at anP average oil temperature
Estimate X and Y at the rich end and calculate the amounts of
by the above equation. Then make XY diagram for the key each component stripped by the above equation. Then make XY
component diagram for the key component
XY Diagram XY Diagram
Equilibrium line is drawn with the slope at origin and values of XR and Equilibrium line is drawn with the slope at origin and values of XR at
YR at rich end, both being obtained from the equilibrium equation rich end, both being obtained from the equilibrium equation
P P
Y 1 þ X Y 1 þ X
K¼ P K¼ P
X Y X 1þ Y
P P P
At the lean end,P X ¼ 0, and slope of the equilibrium line ¼ At the lean end X ¼ 0 and Y ¼ 0, and slope of the equilib-
Y=X ¼ KL ð YL Þ. P P rium line ¼ Y=X ¼ KL P P
At the rich end, values of YR, XR and YR are known At the rich end, values of YR, XR and YR are known
P P
YR 1 þ XR 1 þ Y
XR ¼ P YR ¼ KR XR P R
KR YR 1 þ XR
(Continued)
327
Table 11-25 Summary of Sherwood Absorption and Stripping Calculation Method (Cont’d)
Operating line is drawn with the slope and intercept on the Y axis, or Operating line is drawn with the slope and intercept on the X axis, or
the slope and trial to get proper number of steps. the slope and trial to get proper number of steps.

Lo mols of lean oil entering absorber Lo mols of lean oil leaving stripper
slope ¼ ¼ slope ¼ ¼
Vnþ1 mols of wet gas entering absorber Vo mols of stripping steam entering stripper
Final calculations: Final calculations:

Repeat the analytical or graphical calculations with necessary adjust- Repeat the analytical or graphical calculations with necessary ad-
ments until results of both are the same. justments until results of both are the same.
* These equations which are expressions of the L/V ratio at rich end and the average K, are consistent with the graphical calculation procedure for the key component.
(Source: Edmister, W. C., Hydrocarbon absorption and fractionation process design methods, part 6., The Petroleum Engineer, October, 1947).
4. Operating (Lo/VNþ1)o ¼ (1.25) (11.28) ¼ 14.1 Lean oil rate ¼ Lo ¼ Ai (Ki) (VNþ1)o ¼ (1.227)
5. Operating (11.5)(1,756)
Lo ¼ 24,788 mols/hr

Lo 1 14:1
Aio ¼ ¼ ¼ 1:227 25; 000ð160Þ
VNþ1 o Ki 11:5 GPM oil ¼ ¼ 9; 700
ð8:33Þð0:825Þð60Þ
6. Theoretical number of trays at operating (Lo/VNþ1):
This is unreasonable, and is due to the effect of:
ANþ1 Ai ð1:227ÞNþ1 1:227 a. Operating pressure being too low, thus giving
Eai ¼ ¼ ¼ 0:98
ANþ1 1 ð1:227ÞNþ1 1 a high K value
b. Ethylene being a light component and difficult to
1:227 0:98 absorb.
ðN þ 1Þ log 1:227 ¼ log
1 0:98
c. Temperature being too high.
8. Recalculation of steps 3 through 8.
N ¼ 11.29 Assume operating pressure is 700 psig, K ¼ 1.35.
7. Actual number of plates at operating (Lo/VNþ1): Note that this same K value could have been
Efficiency of oil at 90 F, and sp. gr. of 0.825 corre- achieved by lowering the operating temperature to
sponding to API of 40. –90 F. This is also not practical from the oil stand-
Viscosity ¼ 0.81 centipoise point or even from the economics of operating the
For O’Connell’s efficiency correlation, entire system and refrigeration system at this level,
Figure 10-54. unless (1) the refrigeration is available and (2)
a suitable oil is available.
0:825ð62:3 lb=ft3 Þ Min (L/V) ¼ (1.35) (0.98) ¼ 1.32
¼ 0:0345
ð11:5Þð160Þð0:81Þ Operating (L/VNþ1)o ¼ (1.25) (1.32) ¼ 1.65
Operation Aio ¼ 1.65/1.35 ¼ 1.222
Theoretical number of trays at operating (Lo/VNþ1):
Reading curve (3), Eff. ¼ 14%
This value is low, but agrees generally with the spe-
1:222 0:98
cific data in O’Connell’s [49] report. Although ðN þ 1Þ log 1:222 ¼ log
1 0:98
Drickamer’s data are not so specific for absorption,
the graph of this correlation gives Eff. ¼ 20% for the
0.81 cP. Because no better information is available, N ¼ 11.42 trays
use Eff. ¼ 15%.
Actual number of trays, No ¼ N/Eo ¼ 11.29/0.15 ¼
75.26 ð0:825Þð62:3Þ
Efficiency : ¼ 0:294
Use No ¼ 75 trays ð1:35Þð160Þð0:81Þ
328
Sp:Gr: solvent ð1:0Þð62:3Þ

Reading curve 3, Eff. ¼ 29% HP ¼ ¼
Actual number of trays ¼ 11.42/0.29 ¼ 39.4. Say 40 ðKC3 H4 ÞðM:W: solventÞ ð0:07Þð180Þ
trays. ¼ 4:94
Lean oil rate ¼ (1.222) (1.35) (1,756) ¼ 2,900
HP 4:94
mols/hr ¼ ¼ 2:15
m 2:3
ð2; 900Þð160Þ
GPM ¼ ¼ 1;125
ð8:33Þð0:825Þð60Þ Efficiency ¼ 46.5% (From Figure 10-54)
Use: 45%
Actual number trays in column ¼ 40
This is still a large quantity of oil to absorb the Theoretical number of trays based on 45% efficiency ¼
ethylene. Under some circumstances it might be less (40) (.45) ¼ 18
expensive to separate the ethylene by low-tempera- 4. Using Eai ¼ 96.5% for C3H4. read Ai from
ture fractionation. Figure 11-24.
At n ¼ 18
Example 11-6: Determine Component Ae ¼ Ai ¼ 1.04
Absorption in a Fixed-Tray Tower Ai ¼ Lo/KVNþ1
An existing 40-tray tower is to be examined to determine VNþ1 ¼ 890 mols/hr
the absorption of a rich gas of the following analysis: Lo ¼ (1.04) (0.07) (890) ¼ 64.8 mols/hr
Lo/VNþ1 ¼ 64.8/890.0 ¼ 0.0728
5.
Component Mols/hr Mol wt 1048F Ki 176 psia
Inlet
H2 500.0 2 59.0
CH4 20.9 16 56.0 Lo
Compo- Ai ¼
C2 H 4 131.5 26 8.1 VNþ1 Ki
nent Y(ND1)i Mols/hr Eai
CO 230.0 28 12.3
C3 H 4 3.5 40 0.07 H2 0.562 500.0 0.001235 0.001235
C4 H 2 4.1 50 0.009 CH4 0.0235 20.9 0.00130 0.00130
C2H2 0.1478 131.5 0.009 0.009
890.0 CO 0.258 230.0 0.00592 0.00592
C3H4 0.00393 3.5 1.04 0.965
C4H2 0.00462 4.1 8.1 1.000
1. The key component is methyl acetylene, C3H4. Re-
covery will be based on 96.5% of this material. 0.99985 890.0
The tower average temperature will be assumed ¼
104 F. Mols/hr Mols/hr
The operating average pressure will be ¼ 161 psig. Compo-
nent Absorbed XiR Off Gas Yli (out)
2. The Ki values are tabulated for the conditions
of (1), and were determined from laboratory H2 0.617 0.00953 *499.383 0.561
test data for the special solvent oil being con- CH4 0.0272 0.00042 20.8728 0.02348
sidered. C2H2 1.185 0.0183 130.315 0.14647
Ki ¼ 14.7 (Hi)/176, where Hi is Henry’s CO 1.36 0.021 228.64 0.2565
constant expressed as atm/mol fraction for each C3H4 3.37 0.052 0.13 0.00014
component. C4H2 4.1 0.0633 0 0
Note that conventional K charts are only ap-
plicable to hydrocarbon oil systems, and do not 10.659 0.1645
apply for any special solvents. * These are subtraction differences and do not infer
that the results are this accurate.
3. Calculate tray efficiency for C3H4.
Using O’Connell’s correlation:
Typical calculations: for hydrogen
(solvent) at 104 F ¼ 2.3 cP
Ai ¼ 0.0728/59.0 ¼ 0.001235
Sr. gr. ¼ 1.0
From Figure 11-24 at n ¼ 18 theoretical number of
Mol. wt. solvent ¼ 180
trays, and Ai ¼ 0.001235 read Eai, except that in this low
K ¼ 0.07
329
region some values cannot be read accurately. When Ai is Mols of component in off-gas out top of absorber:
considerably less than 1.0, use Eai ¼ Ai (very little light ¼ 500.0 0.72 ¼ 499.28
material recovered), and when Ai is quite a bit larger than These results do not justify recalculation for greater
1.0, use Eai ¼ 1.0 (heavy material mostly recovered). accuracy. Note that 98% of the C3H4 is absorbed instead
Mols component absorbed/hr ¼ (VNþ1) (Y (Nþ1)i) Eai ¼ of 96.5% as initially specified. This could be revised by
(890) (0.562) (0.001235) ¼ 0.617 reassuming a (slightly) lower oil rate, but this is not
Mols component absorbed/mol lean oil ¼ XiR ¼ 0.0617/ considered necessary.
64.8 ¼ 0.00953 The off-gas analysis Yi represents mols gas out per mol
Mols of component in off-gas out top of absorber: entering rich gas.
¼ 500.0 0.617 ¼ 499.383 mols/hr For a new design, a study should be made of the number
Mols component in, out top of absorber/mol inlet rich of trays against the required lean oil for a given
gas absorption.
YNþ1 Yli
Eai ¼ Let Y
o ¼ 0
YNþ1 Y
o 11.19 Gas Treating Troubleshooting
0:562 Y1i
0:001235 ¼
0:562 Troubleshooting of the glycol unit can be simplified, if the
Yli ¼ 0:5614 unit is appropriately instrumented, operating records,
glycol analyses, process flow sheets and drawings of
6. Correcting values and recalculating vessel internals are readily available [311, 312]. Inter-
preting the records requires knowledge of the design
Inlet
P condition for the plant. Additionally, major changes in
Mols/ A ¼ Lo ð1 þ XiR Þ pertinent variables such as inlet gas flow rate, tempera-
i
Component Y(ND1)i hr VN þ1 Ki Eai ture or pressure require determination of a new set of
optimum operating conditions. Branan [284] illustrated
H2 0.562 500.0 0.00144 0.00144
the corrective measures for the possible causes of
CH4 0.0235 20.9 0.001513 0.001513
common problems in glycol plant as follows.
C2H2 0.1478 131.5 0.01048 0.01048
CO 0.258 230.0 0.00689 0.00689
C3H4 0.00393 3.5 1.21 0.98 11.19.1 High Exit Gas Dew Point
C4H2 0.00462 4.1 9.43 1.00
Change in gas flow rate, temperature or pressure.
Mols/hr Off Gas Insufficient glycol recirculation (should be 1.5 to 3 gal
TEG/lb water removed).
Component Absorbed XiR Mols/hr Yli
Poor glycol reconcentration (exit gas dew point is
H2 0.72 0.0111 499.28 0.561 5–15 F higher than dew point in equilibrium with
CH4 0.0317 0.000488 20.869 0.0232 lean glycol concentration).
C2H2 1.38 0.0213 130.12 0.1463
CO 1.585 0.0244 228.415 0.257
C3H4 3.43 0.0529 0.07 0.00008 11.19.2 High Glycol Losses
C4H2 4.1 0.0633 0 0
First determine where loss is occurring: from contactor (i.e.,
11.246 0.1734 878.754 0.9775 down pipeline); out of still column; or leaking from pump.
Loss from contactor
Typical calculations, using hydrogen: Inlet separator passing liquids
Carryover due to excessive foaming
Lo X
Ai ¼ 1þ XiR Mist extractor plugged or missing
VNþ1 ðKi Þ Lean glycol entering contactor too hot
0:0728 Plugged trays in contactor
¼ ð1 þ 0:1645Þ ¼ 0:00144
59:0 Excessive gas velocity in contactor
Tray spacing less than 24 in.
Mols component absorbed/hr ¼ (0.00144) (500) ¼ 0.72 Mist pad too close to top tray
Mols component absorbed/mol lean oil ¼ 0.72/64.8 ¼ Loss from separator
0.0111 Glycol dumped with hydrocarbons
330
Loss from still contactor and timing fill rate of gauge column on
Excessive stripping gas chimney tray section.
Flash separator passing condensate Check pump valves to see if worn or broken.
Packing in still column is broken, dirty, plugged Plugged strainer, lines, filters.
Insufficient reflux cooling at top of still Vapor lock in lines or pump.
Too much cooling at top of still Low level in accumulator.
Temperature at top of still is too high Excessive packing gland leakage.
Leaks, spills, etc. Contactor pressure too high.
Check piping, fittings, valves, gaskets
Check pumps, especially packings 11.19.6 High Pressure Drop Across
Pilferage Contactor
Many manufacturers will estimate maximum glycol Excessive gas flow rate.
losses to be 1 lb or 1 gas per MMscf. Improper operation Operating at pressures far below design.
can increase the actual glycol losses to 1, 10 or even 100 Plugged trays.
gallons per MMscf. One leaking pump can waste 35
Plugged demister pads.
gallons per day.
Glycol foaming.
11.19.3 Glycol Contamination 11.19.7 High Stripping Still Temperature

Carryover of oils (e.g. compressor lube oils), brine, Inadequate reflux.
corrosion inhibitors, well treating chemicals, sand,
Still column flooded.
corrosion scales, etc. from inlet separator.
Glycol foaming.
Oxygen leaks into glycol storage tanks, etc.
Carryover of light HC in rich glycol.
Inadequate pH control (low pH) increases corrosion.
Leaking reflux coil.
Overheating of glycol in reboiler due to excessive
temperature or hot spots on firetube.
Improper filtration; plugged filters; bypassing of 11.19.8 High Reboiler Pressure
filters.
Improper cleaning of glycol unit; use of soaps or acids. Packing in still column is broken and/or plugged with
tar, dirt, etc.
Restricted vent line, not sloped.
11.19.4 Poor Glycol Reconcentration Still column is flooded by excessive boil up rates and/
or excessive reflux cooling.
Low reboiler temperature. Slug of liquid HC enters stripper, vaporizes on
Insufficient stripping gas. reaching reboiler, and blows liquid out of still.
Rich glycol is leaking into lean glycol in the glycol heat
exchanger or in the glycol pump. 11.19.9 Firetube Fouling/Hot
Overloading capacity of reboiler.
Spots/Burn Out
Low fuel rate or low Btu content of fuel.
Glycol foaming in the still column. 1. Buildup of salt, dust, scales, etc. on firetube – check
Dirty and/or broken packing in the still column. inlet separator.
Flooding of the still column. 2. Deposition of coke, tar formed by glycol overheating
High still pressure. and/or hydrocarbon decomposition.
3. Liquid glycol level drops exposing firetubes. Low
11.19.5 Low Glycol Circulation – Glycol level shutdown.
Pump
11.19.10 High Gas Dew Points
Check pump operation. If it is a glycol powered
pump, close lean glycol discharge valve; if pump The most obvious indication of a glycol dehydration
continues running, it needs repair. If gas or electric, malfunction is a high water content or dew point of the
check circulation by stopping glycol discharge from outgoing sales gas stream. In most cases, this is caused by
331
an inadequate glycol circulation rate or by an insufficient 3. Check the stripping gas, if applicable, to be sure
reconcentration of the glycol. These two factors can be there is a positive flow of gas. Be sure steam is
caused by a variety of contributing problems as illus- not back flowing into the accumulator from the
trated in the following sections. reboiler.
11.19.11 Cause – Inadequate Glycol 11.20.1 Cause – Operating Conditions

Circulation Rate Different from Design
1. Glycol powered pump. Close the dry discharge valve
1. Increase the absorber pressure, if needed. This
and see if the pump continues to run; if so, the pump
may require the installation of a back pressure
needs to be repaired.
valve.
2. Gas or electric driven pump. Verify adequate circu-
2. Reduce the gas temperature, if possible.
lation by shutting off the glycol discharge from the
absorber and timing the fill rate in the gauge glass. 3. Increase the glycol circulation rate, if possible.
3. Pump stroking but not pumping. Check the valves to
see if they are seating properly.
4. Check pump suction strainer for stoppage.
11.20.2 Cause – Low Gas Flow Rates
5. Open bleeder valve to eliminate ‘‘air lock’’. 1. If the absorber has access man-ways, blank off
6. Be sure surge level is sufficiently high. a portion of the bubble caps or valve trays.
2. Add external cooling to the lean glycol and
11.19.12 Low Reboiler Temperature balance the glycol circulation rate for the low gas
rate.
Inadequate size firetube and/or burner. 3. Change to a smaller absorber designed for the lower
Setting temperature controller too low. rate, if needed.
More water in inlet gas because pressure is low or
temperature high.
Temperature controller not operating correctly. 11.20.3 High Glycol Loss
Carryover of water from inlet separator.
Inaccurate reboiler thermometer. Cause – Foaming
Foaming is usually caused by contamination of glycol
with salt, hydrocarbons, dust, mud and corrosion in-
11.19.13 Flash Separator Failure hibitors. Remove the source of contamination with ef-
fective gas cleaning ahead of the absorber, improved
Check level controllers.
solids filtration, and carbon purification.
Check dump valves.
Excessive circulation.
Cause – Excessive Velocity in the Absorber
1. Decrease the gas rate.
11.20 Cause – Insufficient
2. Increase the pressure on the absorber, if possible.
Reconcentration of Glycol
1. Verify the reboiler temperature with a test ther- Cause – Trays Plugged With Mud, Sludge, and
mometer and be sure the temperature is in the Other Contaminants
recommended range of 350 F to 400 F for triethy- If the pressure drop across the absorber exceeds about
lene glycol. The temperature can be raised closer to 15 psi, the trays may be dirty and/or plugged. Plugged
400 F, if needed, to remove more water from the trays and/or downcomers usually prevent the easy flow of
glycol. gas and glycol through the absorber. If the absorber has
2. Check the glycol-to-glycol heat exchanger in the access handholes, manual cleaning can be helpful.
accumulator for leakage of wet, rich glycol into the Chemical cleaning is recommended if handholes are not
dry, lean glycol. available.
332
11.20.4 Cause – Loss of Glycol Out of 2. The wet glycol discharge line to the reboiler is re-
stricted. A pressure gauge installed on the line will
Still Column
show the restriction immediately.
1. Be sure the stripping gas valve is open and the ac- 3. The suction line is too small and an increase in
cumulator is vented to the atmosphere. temperature and pumping rate cavitates the pump.
2. Be sure the reboiler is not overloaded with free water 4. A leaky check valve, a foreign object lodged under
entering with the gas stream. a check valve, or a leaky piston.
3. Be sure excessive hydrocarbons are kept out of the 5. Look for metal chips or shavings under the pump
reboiler. D-slides.
4. Replace the tower packings in the still column, if 6. Traps in the wet glycol power piping send alternate
fouled or powdered. slugs of glycol and gas to the pump.
If the outlet gas temperature from the absorber exceeds 7. Insufficient glycol to the Main Piston D-slide ports.
the inlet gas temperature by more than 20 to 30 F, the Elevate the control valve end of the pump to correct.
lean glycol entering the top of the absorber may be too The glycol pH should be controlled to avoid equipment
hot. This could indicate a heat exchanger problem or an corrosion. Some possible causes of a low, acidic pH are:
excessive glycol circulation rate.
1. Thermal decomposition, caused by an excessive
If a glycol pump has been operating in a clean system,
reboiler temperature (over 404 F), deposits on the
no major service will probably be needed for several
fire tube, or a poor reboiler design.
years. Only a yearly replacement of packing is usually
required. Normally the pump will not stop pumping 2. Glycol oxidation, caused by getting oxygen into the
unless some internal part has been bent, worn, or broken, glycol with the incoming gas; it can be sucked in
a foreign object has fouled the pump, or the system has through a leaking pump or through unblanketed
lost its glycol. A pump which has been running without glycol storage tanks.
glycol for some time should be checked before returning 3. Acid gas pickup from the incoming gas stream.
to normal service. The pump will probably need at least
new ‘‘O’’ rings. The cylinders and piston rods may also Salt accumulations and other deposits on the fire tube
have been scored from the ‘‘dry run’’. can sometimes be detected by smelling the vapors from
The following are typical symptoms and causes for the still vent. A ‘‘burned’’ odor emitted from these vapors
a Kimray pump operation. usually indicates this type of thermal degradation. An-
other detection method is to observe the glycol color. It
Symptoms will darken quickly if the glycol degrades. These de-
tection methods may prevent a fire tube failure.
1. The pump will not operate.
In general, maintenance and production records to-
2. The pump will start and run until the glycol returns gether with the used lean and rich glycol analyses can be
from the absorber. The pump then stops or slows very useful to the troubleshooter. A history of filter el-
appreciably and will not run at its rated speed. ement, carbon, tower packing, and fire tube changeouts
3. The pump operates until the system temperature is can be shown and the frequency of pump repairs and
normal and then the pump speeds up and cavitates. chemical cleaning jobs are most helpful to the trouble-
shooter, who can quickly eliminate and thus prevent
4. The pump lopes or pumps on one side only.
costly problems.
5. Pump stops and leaks excessive gas from the wet
glycol discharge.
6. Erratic pump speed. Pump changes speed very few 11.21 Sour Water Stripping Process
minutes.
7. Broken pilot piston. Sour water streams containing hydrogen sulfide (H2S),
other organic sulfur compounds, ammonia (NH3), and
oil are collected from various refinery process units and
Causes are combined in a feed surge tank. Liquid hydrocarbons
1. One or more of the flow lines to the pump are are decanted from the water and returned to the re-
completely blocked or the system pressure is too low covery oil tank. The Sour Water Stripper (SWS)
for standard pumps. removes hydrogen sulfide and ammonia (H2S/NH3)
333
from the sour water using a stripper tower having recovery unit as feed. The stripped water is reused at the
a steam heated reboiler. crude desalters and at process units requiring wash water
A sour water stripper either uses a steam fired (e.g. for ammonia removal). Any remaining stripped
reboiler as a heat source or direct application of steam water is routed to the washwater treatment plant.
(usually low quality and low pressure). There are no
fired heaters associated with the sour water stripper. A Case Study
The purpose of this unit is to drive as much H2S and 80,000 gal/day of sour water at 100 F and 40.0 psia goes
NH3 overhead in the stripper as possible. The sizing of through a feed/effluent exchanger where it recovers heat
the unit is essential, since its capacity must equal or from the tower bottoms stream (stripper bottoms). The
exceed the normal production rates of sour water from new stream (stripper feed) at 200 F enters on tray 2 of
various sources in the refinery. Generally, refiners find an 8 stage distillation column with a reboiler and a total
their strippers undersized because of a lack of provision reflux condenser. A quality specification of 10 ppm wt.
of handling large amounts of sour water, which can ammonia on the tower bottoms (stripper bottoms) is
result from upset conditions (e.g. start-up and shut specified. The tower bottoms (stripper bottoms) ex-
down). There is often a large quantity of sour water change heat with the incoming feed and exists as
waiting to be processed in the stripper, and in the light effluent.
of the increasing importance of environmental re- The sour water feed compositions are:
strictions and regulations by governments (e.g. U.S.
EPA), the sour water stripper unit plays an important Component Mass fraction
role in mitigating the overall pollution of refiners.
Figure 11-29 shows the process flow diagram where Hydrogen sulfide (H2S) 0.007
the feed sour water is preheated by exchange with the Ammonia (NH3) 0.005
stripper bottoms stream. The reboiler is heated with low Water (H2O) 0.988
pressure steam to generate a vapor stream flowing up the
stripper column. Vaporization of water strips H2S and
NH3 from the downcoming sour water. Overheads Solution
vapors are cooled by an overhead condenser. The con- UniSim simulation software Sour-Water-Unit USC
densed water reflux is returned to the top tray in the design R 370 is used to simulate the sour water at 100 F
stripper tower. The overhead, non-condensable mate- and 40.0 psia entering the tube side of the heat exchanger
rials, primarily H2S and NH3 are routed to the sulfur with stripper bottoms entering the shell side of the
Figure 11-29 Process flow diagram of a sour water stripping unit.
334
exchanger. The stripper feed stream exits the outlet of 3. Define the column configuration as shown below.
tube side and enters the sour water stripping column at
the second stage of an 8 plate distillation column, which
In this cell Enter
is used to strip the sour components from the feed
stream. The liquid leaving the bottom of the column Connections No. of stages 8
heats the incoming feed stream in the heat exchanger. Inlet Stream Stripper
There are two basic steps in setting up the simulation Feed
process: Inlet stage 2
1. Setup. This case study uses the Sour Peng-Robinson Condenser Type Full Reflux
package and the following components: H2S, NH3 Ohd Vapor Off Gas
and H2O. Bottoms Liquid Stripper
Bottoms
2. Steady State Simulation. The case consists of an
Reboiler Energy Q-Reb
8 stage stripper, used to separate H2S and NH3 and
Stream
uses a heat exchanger to minimize heat loss.
Condenser Energy Q-Cond
The specifications are as follows: Stream
Pressure Condenser Pressure 28.7 psia
Feed Stream Profile
Reboiler Pressure 32.7 psia
Specify the feed stream as shown.
In this cell Enter 4. In the column property view, click the Design tab,
Temperature
100 F then select the Monitor page.
Pressure 40 psia In the present configuration, the column has two
Std Ideal Liq Vol Flow 80,000 barrel/day degrees of freedom, the two specifications used will
Comp. Mass Frac (H2S) 0.0070 be a quality specification and a reflux ratio.
Comp. Mass Frac. (NH3) 0.0050 5. Modify the existing specification as shown below
Comp. Mass Frac. (H2O) 0.9880
Column [SW Stripper]

Operations Tab [Page] Variable Spec Modify
1. Install and specify the heat exchanger as:
Design Ovhd Vap Rate Active ¼ uncheck
[Specs]
Reflux Ratio Active ¼ checked
Tab [Page] In this cell Enter Spec Value ¼ 10
Design Tube Side Sour H2O Molar
[Connections] Inlet Feed
Tube Side Outlet Stripper
Feed 6. Add a Component Fraction specification, and enter
Shell Side Inlet Stripper the following information in the Comp Frac Spec
Bottoms view
Shell Side Outlet Effluent
Design Heat Exchanger Exchanger Tab In this cell Enter
[Parameters] Model Design
(Weighted) Parameters Name NH3 Mass Frac.
Tube Side Delta P 10 psi (Reboiler)
Shell Side Delta P 10 psi Stage Reboiler
Worksheet Temperature 200 F Spec Value 0.000010
[Conditions] (stripper Feed) Component Ammonia
Summary Active Checked
Reflux Ratio Active
2. Install a distillation column. The column will have Spec Value 10 Molar
both a reboiler and an overhead condenser.
335
7. Click the Parameters tab, then select the Solver outer ring of aromatic compounds of lower molecular
page. Change the Fixed Damping Factor to 0.4. weight. They can precipitate when the continuous
A damping factor will speed up tower convergence nature of the surrounding ring of aromatics is broken
and reduce the effects of any oscillations in the cal- down by cracking processes.
culations (the default value is 1.0). ASTM: American Society of Testing and Materials.
8. Run the column to calculate the values by clicking Nearly all of the refinery products tests have been
the Run button. standardized by ASTM.
The results of the simulation are shown in Atmospheric Residuum: The heaviest material from
Appendix L. the distillation of crude oil in a crude distillation
column operating at a positive pressure.
Azeotrope: A constant boiling point mixture for which
Glossary the vapor and liquid are of identical composition.
Components of azeotropes cannot be separated with
Absorption: A separation process involving the transfer conventional distillation.
of a substance from a gaseous phase to a liquid phase Boiling point: Typically refers to the temperature at
through the phase boundary. which a component or mixture of components start
Acid Gases: Impurities in a gas stream usually consisting to vaporize at a given pressure. When used in petro-
of H2S, CO2, COS, RSH, and SO2. Most common in leum refining it is usually synonymous with the
natural gas are of H2S, CO2, and COS. normal boiling point (boiling point at one
Adsorption: The process by which gaseous components atmosphere).
adhere to solids because of their molecular attraction Boiling range: The lowest through highest boiling
to the solid surface. temperatures for a petroleum stream when distilled.
Alkanolamine: An organic nitrogen bearing compound Boiling ranges are often reported on a TBP (true
related to ammonia by having at least one of its hy- boiling point) basis, i.e. as normal boiling points.
drogen atoms substituted with at least one linear or Bubble point: The temperature at which a component
branched alkanol groups where only one or two could or mixture of components begins to vaporize at
also be substituted with a linear or branched alkyl a given pressure.
group (i.e. methyldiethanolamine MDEA). The Carbon number: The number of carbon atoms in one
number of hydrogen atoms substituted by alkanol or molecule of a given hydrocarbon.
alkyl groups at the amino site determines whether Characterization factor: A calculated factor used to
the alkanolamine is primary, secondary or tertiary. correlate properties for petroleum streams. It is
Amine treating: Contacting a gas or light hydrocarbon a measure of the paraffinicity of the stream and is
liquid with an aqueous solution of an amine com- defined as ([MABP**1/3]/Sg) where MABP ¼ mean
pound to remove the H2S and CO2. average boiling temperature in degrees ( R) and
API Gravity: A method for reporting the density of Sg ¼ specific gravity at 60 F (15.9 C) relative to water.
petroleum streams. It is defined as [(145.1/Sg) – Catalytic cracking: Process in which an oil feed is
131.5] where Sg is the specific gravity relative to reacted in a hot catalyst bed to produce heating oils,
water. API gravity is reported at a reference tem- gasoline, and light petroleum liquids and gases. The
perature of 60 F (15.9 C). products are stripped from the catalyst and the cat-
Aromatics: Derivatives of benzene, a six carbon atom alyst is burned in a regeneration step to remove coke
ring structure with three double bonds. Aromatics deposits resulting from the cracking reactions.
may have multiple combined benzene rings, with Catalytic polymerization (cat.poly): Process in which
side chains consisting of other chemical structures propylene and or butylene components are chemi-
attached to the carbon atoms in the ring structure. cally joined to produce gasoline. A phosphoric acid
Asphalt: Heavy tar-like residue from distillation of catalyst is usually employed in the process.
some types of crude oil. Asphalt components are Catalytic reforming (reforming): Process where low
high molecular weight derivatives of aromatic com- octane naphthas are chemically changed into high
pounds. Not all asphalt materials are suitable for use octane gasoline and light gas products over a plati-
as building agents in road pavement. num catalyst.
Asphaltenes: Highly condensed masses of high molec- Continuous Catalytic Reforming (CCR) process: is
ular weight aromatic compounds. They exist in pe- where the catalyst is circulated through the reactors
troleum residue as the center of colloidal particles or and a regeneration step, analogous to catalytic
micelles. The asphaltenes are kept in solution by an cracking processes.
336
Cetane number: A number calculated from the API D86: ASTM laboratory distillation method conducted
gravity and the D86 fifty percent distilled for a petro- at atmospheric pressure for streams boiling below
leum stock. It is used to rate the performance of a fuel approximately 700 F (371 C). The D86 is the most
in diesel engines. commonly used laboratory distillation for petroleum
Cloud Point: The temperature at which the first crys- stocks.
tals form when a petroleum stock is cooled in a test D1160: ASTM laboratory distillation method for high
apparatus. boiling streams. The D1160 is performed under
Coke: A carbonaceous deposit formed by the thermal vacuum conditions with 10 mm Hg being the most
decomposition of petroleum. common pressure used for the test. D1160 data are
Coking: Refining process in which petroleum oil is normally reported on a 760 mm Hg basis.
heated destructively such that the heaviest materials D2287: ASTM simulated distillation method for high
are converted to coke. There are two processes: boiling streams. The D2887 has an upper limit of
delayed coking and fluid coking, with delayed coking 1000 F (538 C) and the temperatures are reported
being the most widely used. vs. weight per cent distilled. A normal paraffin stan-
Condensation: Reaction in which aromatic ring struc- dard is used to convert the chromatographic results
tures combine to form ring structures larger than the to a boiling point curve.
reactants. D3710: ASTM simulated distillation method for gaso-
Conversion: The fraction of a feed stock converted to line and light naphthas. D3710 data are reported on
gasoline and lighter components. a volume basis.
Cracking: The process by which the composition of Debutanizer: A column that removes n-butane
petroleum is changed by heat induced reactions. (nC4H10) and lighter components in the top
Cracking may or may not take place in the presence product.
of a catalyst. Deethanizer: A column that removes ethane (C2H6)
Critical point: The temperature and pressure at which and lighter components in the top product.
a component or mixture of components enter Delayed coking: A refinery process in which heavy oil is
a dense phase, being neither liquid nor vapor. heated and then allowed to thermally crack and de-
Critical Pressure: The vapor pressure at the critical posit coke in a large drum that provides delay time
temperature. for the coking reaction to occur. The coke is removed
Critical Temperature: The temperature above which from the drum with a hydraulic drill that is lowered
a component cannot be liquefied. For mixtures, the into the drum from a derrick that sits on top of the
temperature above which all of the mixture cannot drum.
be liquefied. Dehydration: a process used to remove water from
Crude oil: A complex mixture of hydrocarbons ranging natural gas and natural gas liquids (NGLs), and is
from methane (CH4) to compounds boiling higher required to prevent formation of hydrates, corrosion
than 2000 F (1094 C). It is a natural resource and is and condensation of free water in processing and
pumped from underground locations. transportation.
Cut point temperature, cut points: Arbitrarily defined Dehydrogenation: A chemical reaction in which
temperatures on the laboratory boiling point curve a compound loses bonded hydrogen.
for a petroleum stock. The boiling point curve most Deisobutanizer: A column that removes isobutene
commonly used to define cut points is the true boil- (iC4H8) and lighter components in the top product.
ing point curve (TBP) at one atmosphere of pressure. Demethanizer: A column that removes methane (CH4)
Cut point ranges: A series of cut point temperatures are and lighter components in the top product.
defined for a petroleum stock. The cut point ranges Depentanizer: A column that removes n-pentane
are the temperature differences between adjacent (C5H12) and lighter components in the top
cut point temperatures. When developing petroleum product.
pseudo-components for a petroleum stock, cut point Desalting: A process that removes chlorides and other
ranges must be defined which include the total inorganic salts from crude oil by injecting water and
boiling point range of the stock. applying an electrostatic field to force the salt into
Cycle oil, cycle stock: An oil stock containing hydrogen the aqueous phase.
deficient compounds that was produced in a thermal Desuperheating zone: A section of a distilling column
or catalytic cracking operation. where a superheated vapor is cooled and some liquid
Cyclo-olefin: Unsaturated ring structure with one or is condensed. FCC main fractionators have a desu-
two double bonds in the ring. perheating zone.
337
Dew point: The temperature and pressure at which the mechanical restrictions or damage to the internals in
first drop of liquid will condense for a component or the column, etc. All columns are designed to handle
mixture of components. about 80% of the flow before flooding occurs.
Diesel fuel: A fuel produced for diesel engines with Fluid catalytic cracking (FCC): The most commonly
typical ASTM D86 boiling point ranges of 450– used catalytic cracking process. The catalyst is a fine
675 F (233–358 C). powder that is designed to form a fluidized bed in the
Distributed component: A component that appears in reactor and regenerator.
both the top and bottom products from a distillation Fluid coking: involves the feed being preheated and
column separation zone. sprayed into a reactor where it contacts a hot fluid-
Downcomer: A device to direct the liquid from a dis- ized bed of coke returning from a burner vessel. The
tillation column at a location above the bottom tray hot oil products are stripped from the coke which is
and below the top tray. Draws may be vapor or liquid circulated back to the burner vessel. Coke not
phase. returned to the reactor from the burner vessel is
Draw, side draw: A product stream withdrawn from withdrawn as a coke product.
a distillation column at a location above the bottom Fractionator: Another term for a distillation column.
tray and below the top tray. Draws may be vapor or Free energy of formation: The change in free energy
liquid phase. when a compound is formed from its elements with
Endothermic: a process that requires heat input. each substance in its standard state at 77 F (25 C).
End point (final boiling point): The highest boiling The heat of reaction at 25 C may be calculated by
point recorded for a laboratory distillation. Usually, subtracting the sum of the free energies of formation
there is some residual material in the laboratory still, of the reactants from the sum of the free energies of
and the end point is not the highest boiling point formation of the products.
material in the mixture being distilled. Free water: Condensed water that exists as a separate
Energy balance: Summation of the energy entering a pro- liquid phase. Most refinery distillation columns are
cess and the summation of the energy leaving a process. designed such that free water will not be present,
They must be equal for a steady state process. since it can result in column upsets and promote
Equilibrium K value (K value): The ratio of the mole corrosion of the metal in the column.
fraction in the vapor phase divided by the mole Fixed bed reactor: A reactor in which the catalyst is
fraction in the liquid phase for a component in the loaded into an immovable bed. The reactants enter
equilibrium state. Each K value corresponds to the top of the bed, and the products exit from the
a given temperature, pressure and mixture bottom of the bed. The process must be taken off
composition. line, and hot gases circulated through the catalyst bed
Entrainment: A non-equilibrium process by which liq- to burn off coke deposits and restore the catalyst
uids are mechanically carried into a vapor that is activity.
leaving a process vessel or contacting device. Gap: is usually based on ASTM D86 distillation tem-
Exothermic: Exothermic processes release heat as a by- peratures and is defined as the 5% distilled temper-
product. ature of a distillation column product minus the 95%
Flash calculation: Determination of the compositions distilled temperature of the next higher product in
and quantities of liquid and vapor that coexist in the column. When the difference is positive, the
a mixture under equilibrium conditions. difference is called a gap. When the difference is
Flash point: The temperature at which the vapors above negative, the difference is sometimes called an
a heated hydrocarbon mixture will ignite when overlap. The gap or overlap is a measure of the
contacted by an open flame. It is determined in sharpness of the separation between adjacent prod-
a laboratory. ucts in a distillation column.
Flash zone: The section of a distillation column containing Gas oil: is a term used for petroleum stocks with boiling
the column feed nozzle(s). The column feed separates ranges between 650–1100 F (344–594 C). Un-
or ‘‘flashes’’ into liquid and vapor as it expands through cracked gas oils are produced by distilling crude oil in
the feed nozzle(s) and enters the column. crude and vacuum columns. Cracked gas oils are
produced in refinery reaction processes, such as
Flooding: occurs in a distillation column when flows of
thermal and catalytic cracking, coking, visbreaking
liquid and/or vapor greater than the column can
and hydrocracking.
process are encountered. It can be caused by liquid
backing up in the column, vapor blowing through the Gas treating: Amine treating of light gases to remove
column and lifting the liquid off the trays, by impurities such as H2S and CO2. Molecular sieves
are also used to concentrate hydrogen streams
338
by removing inerts and light hydrocarbon Hydrotreating: a process where cracking is minimized.
contaminants. The hydrogen consumed in chemical reactions is that
Gasoline: A name for petroleum stocks that are used as required to convert sulfur and nitrogen compounds to
fuel for internal combustion engines. Retail gasoline H2S and NH3, plus hydrogen that replaces carbon/
is a blend of several refinery gasolines, which must carbon double bonds in saturation reactions. Some
meet specifications for octane, Reid vapor pressure, hydrogen is lost through solubility in the liquid
distillation boiling range, sulfur content, etc. Addi- product(s).
tives such as ethers or alcohols are used to improve Initial boiling point: the boiling temperature in a labo-
the octane rating for the blended product. ratory still at which the first drop of distilled liquid is
Grain: A unit of mass where one pound is equivalent to condensed. The initial boiling point may be higher
7000 grains and a specification of 0.25 grain of H2S than the boiling point of light components in the
per 100 scf is equivalent to an H2S concentration of sample, since these are not condensed by the
4.0 ppmv. apparatus.
Gravity: The specific gravity of a stream, often Hypothetical State: is defined as a fluid in a state that
expressed in API Gravity by petroleum refiners. The cannot actually exist, e.g. methane as a liquid at 60 F
basis is always the density of water. and 14.696 psia. Methane cannot be in its liquid
Gravity curve: The gravity of the material distilled from phase at this temperature and pressure, but such
a petroleum stock in a laboratory still. The gravity a state when defined, can be used in calculations.
curve is plotted against the percent distilled for the Isomerization: A reaction process where the molecular
stock. Gravity curves are most commonly reported structure of a compound is rearranged to a different
for true boiling point distillation data. isomer form, but the chemical formula remains
Gross Heating Value: is the total energy transferred as unchanged.
heat in an ideal combustion reaction at a standard Jet fuel: A kerosene material of typical ASTM D86
temperature and pressure in which all water formed boiling point range 400–550 F (205–288 C) used as
appears as liquid. The gross heating is an ideal gas fuel for commercial jet aircraft.
property in a hypothetical state (the water cannot all Joule Thompson expansion: The pressure of a mixture
condense to liquid because some of the water would is reduced with no heat transfer to or from the sur-
saturate the CO2 in the products). roundings. The pressure decrease typically results in
HSTD: High temperature simulated distilled. Labora- a temperature decrease with the exception for sys-
tory test designed for petroleum stocks boiling up to tems that consist largely of hydrogen gas.
1382 F (750 C). Kerosene: A material from distillation of crude oil that
Heat of reaction: The heat released or heat absorbed boils in the approximate ASTM D86 range of 400–
when a chemical reaction takes place. The heat of 550 F (205–288 C).
reaction may be calculated from the free energies of Key components: In a conventional distillation column
formation for the reacting components and the re- with two products, two components or groups of
sultant products at the standard temperature of 77 F components define the separation. Both components
(25 C). must be distributed to the top and bottoms products.
Heavy cycle oil (HCO): Gas oil produced in a FCC The light key appears in the bottoms product in
operation that boils in the approximate TBP range of a small significant quantity and the heavy key appears
400–1000 F (205–538 C). Heavy cycle oil is not in the top product in a small significant quantity.
withdrawn as a product, but rather it is recycled to Lean oil: The absorption oil entering the top tray of an
the reactor for further cracking in order to improve absorption column.
the overall conversion of the process. Light cycle oil (LCO): Gas oil produced in a catalytic
Heavy Key: A distributed component in a distillation cracking operation that boils in the approximate
section that is recovered in the (bottoms) heavy ASTM D86 range of 400–695 F (205–369 C).
product. Light key: A distributed component in a distillation
Hydrocarbon: Any organic compound that consists of section that is recovered in the top light product,
hydrocarbon and carbon atoms. with a small specified amount leaving the bottoms
Hydrocracking: A refining process where a petroleum product.
stream is cracked over a catalyst in a high hydrogen Liquefied petroleum gas (LPG): Liquefied petroleum
atmosphere. The high H2 partial pressure immedi- gas, typically mixtures of C3 and C4 hydrocarbons.
ately saturates any unsaturated compound produced Make up stream: A feed to a process to replace a com-
by the cracking reactions. ponent that reacts or is otherwise depleted in
a process.
339
Mass balance: Summation of the mass entering a pro- column, the overhead and the top product are the
cess and the summation of the mass leaving a pro- same.
cess. They must be equal for a steady state process. Paraffins: Hydrocarbons of the alkane family. Paraffins
Mixed phase: More than one phase; usually vapor and are saturated compounds, that is, hydrogen atoms are
liquid phases are present. appropriately attached to the carbon atoms such that
Molecular sieve: A separation process that usually the carbon atoms have only single bonds in the
works by gaseous diffusion. A membrane is selected molecular structure. The general formula for paraf-
through which the compounds being removed or fins is CnH2nþ2, where C ¼ carbon atoms and
purified can pass while the remaining compounds in H ¼ Hydrogen atoms.
the stream being processed cannot pass. Partial pressure: In a gaseous mixture, the pressure
Methyl tertiary butyl ether (MTBE): A gasoline addi- contribution for a particular component of the
tive used to increase octane rating. MTBE is pro- mixture.
duced by reacting methanol with isobutylene. Phase Envelope: The boundaries of an area on the P-T
Naphtha: A term that is applied to low boiling mixtures diagram for the material which encloses the region
of hydrocarbons, with typical TBP boiling ranges where both vapor and liquid co-exist.
between 150–450 F (66–233 C). Light and heavy Physical Solvent: A liquid capable of absorbing selected
naphthas are produced in the distillation of crude gas components by solubility alone without associ-
oils. Cracked naphthas are also produced by many of ated chemical reactions.
the refinery reaction processes. Polymerization: A reaction in which like molecules are
Naphthenes: Hydrocarbons of the cyclane family, joined together to form dimer and trimer compounds
sometimes called cycloalkanes. Naphthenes have no etc of the reactant(s). This most often occurs with
double bonds and are saturated ring structures with olefinic compounds.
the general formula CnH2n, where C ¼ carbon atoms Pour point: A sample tube of petroleum oil is chilled in
and H ¼ Hydrogen atoms. the pour point test. The pour point is defined as the
Net Heating Value: is the total energy transferred as temperature at which the sample will still pour
heat in an ideal combustion reaction at a standard (move) when the sample tube is tipped. The pour
temperature and pressure in which all water formed temperature is typically about 5 F (2.8 C) lower
appears as vapor. The net heating is an ideal gas than the cloud point.
property in a hypothetical state (the water cannot all ppmv: A volume concentration of a species in a bulk
remain vapor because, after the water saturates the fluid measured in parts per million.
CO2 in the products, the rest would condense). Preheat, preheat train: Heat exchanger or a net work of
Octane: A test used to measure the suitability of a gas- heat exchangers in which the feed to a process
oline as motor fuel. The octane test determines the (usually a distillation column) is heated by recovering
knocking characteristics of a gasoline in a standard heat from products being cooled.
test engine relative to a standard 2-2-4 trimethyl Pumparound: A liquid side draw from a distillation
pentane (224 TMP). 224 TMP is assigned an octane column that is pumped, cooled and returned to
number of 100. There are two octane tests. One is a higher location in the column. Pumparound recover
designated the research octane (F-1), and the second useable heat that would be lost at the condenser.
as the motor octane (F-2). Motor octane is deter- They also lower the vapor flow in a column and
mined in an engine more representative of actual reduce the required column diameter for vapor
operating conditions for automobiles and is lower loaded columns such as crude and vacuum columns.
than research octane for any gasoline stock. Purge: A stream that is removed from a recycle process
Off gas: The gas leaving a reflux drum or top of tray of to prevent buildup of one or more compounds in the
an absorber column. process streams.
Olefins: Hydrocarbon of the alkene family. Olefins have Quality: The weight fraction of vapor in a vapor-liquid
two carbon atoms in the molecular structure linked mixture.
by a double bond to satisfy the absence of two hy- Quench stream: A cooled stream that is used to cool
drogen atoms that are present in the corresponding another stream by direct contact, eg. Hydrogen
paraffin. The hydrogen deficiency is called unsatu- quench streams are used to quench the hot effluents
ration. The general formula for olefins is CnH2n, from hydrocracker reactors.
where C ¼ Carbon atoms and H ¼ Hydrogen atoms. Quench zone: A section of a distillation column where a
Overhead: Usually refers to the vapor leaving the top hot stream, usually vapor, is cooled by direct contact
tray of a distillation column. For an absorber with a stream that has been cooled, usually a liquid.
340
Rating calculations: Calculations in which a unit oper- Secondary absorber: The second absorber in a FCC gas
ation such as a column, heat exchanger, pump, etc. is plant is usually the last unit operation in the gas re-
checked for capacity restrictions. covery plant and is known as the sponge absorber.
Reactive distillation: A distillation column in which Selective Treating: Preferential removal of one acid gas
there is a section designed for chemical reaction, component, leaving at least some of the other acid
usually containing a catalyst bed. Some methyl ace- gas components in the treated stream.
tate, MTBE and TAME processes use a reactive dis- Separation zone: A section of a distillation column in
tillation column in place of a second reactor prior to which a separation between two products occurs.
the product stream. Components which are found in both products are
Rectification zone: the section of a distillation column said to be distributed components.
in which heavy components flow down the column Separator: is referred to as a drum, in which residence
by contact with a liquid reflux stream. In a conven- time is provided for a mixture of liquid and vapor to
tional distillation columns, this is the section of the separate into liquid and vapor streams. It is also
column from the tray above the feed tray to the top called a flash drum. The liquid and vapor leaving the
tray. separator are in phase equilibrium.
Recycle: A process stream that is returned to an up- Side heater (reboiler): A heat input to a distillation
stream operation. column that is located above the bottom tray of the
Reduced Pressure: The ratio of the absolute pressure to column.
the critical pressure. Side stripper: A small auxiliary column that receives
Reduced Temperature: The ratio of the absolute tem- a liquid draw product from a main distillation
perature to the critical temperature. column, for stripping out of light components. Light
Reformate: The gasoline produced in a catalytic components are stripped by a stripping steam, or
reforming operation. reboiling and returned to the main column. Liquid
Reflux: Condensed liquid that is returned to the top products are sometimes stripped in side strippers to
tray of a distillation column. Reflux helps rectify the raise the flash point.
mixture being distilled by washing heavy compo- Simulated distillation: A new laboratory method in
nents down the column. which a petroleum stream is separated into fractions
Reflux drum: A drum that receives the outlet from the with gas phase chromatography. Carbon disulfide
overhead condenser for a distillation column. The (CS2) is used as the carrying agent to dissolve the
liquid and vapor portions are separated in the reflux petroleum stream. The component fractions elute
drum. from the chromatographic column in a time se-
Reid vapor pressure (RVP): An ASTM test method to quence that is related to their boiling temperatures.
determine the vapor pressure of a light petroleum Temperatures are assigned to the fractions based on
stream. The Reid vapor pressure is very nearly equal the chromatographic separation of a normal paraffin
to the true vapor pressure for gasoline streams. There standard mixture. The simulated distillation ap-
is also a Reid vapor pressure test for crude oil. proaches a true boiling point distillation, and is
Residence Time: The period of time in which a process reported on a mass basis for streams heavier than
stream will be contained within a certain volume or gasoline. Aromatic compounds elute from the
piece of equipment, in seconds. column faster than paraffins of similar boiling points.
Hence, simulated distillations must be corrected for
Rich oil: The absorption oil leaving the bottom tray of
aromatic content when stocks contain significant
an absorption column. The rich oil contains the
absorbed light components. quantities of these compounds.
Slurry: It is bottom stream from FCC main fraction-
Saturated compounds: Hydrocarbons in which there
ators, and is referred to as slurry because it contains
are no double bonds between carbon atoms. Satu-
suspended catalyst particles.
rated compounds contain the maximum possible
number of hydrogen atoms. Sour gas: A light gas stream that contains acid gases, e.g.
sulfur compounds (e.g. hydrogen sulfide), ammonia
Saybolt Seconds Universal (SSU): The viscosity of
compounds and/or carbon dioxide.
a petroleum stock as measured by an ASTM labora-
tory test. The test value is the time in seconds for Splitter: A distillation column that separates a feed into
a measured sample of the stock to flow through an light and heavy products.
orifice at a given temperature. SSU is normally Specific gravity: by definition is the ratio of gas density
measured at temperatures of 100 and 210 F (38.1 (at the temperature and pressure of the gas) to the
and 99.2 C) density of dry air (at the air temperature and
pressure).
341
Sponge oil: Absorption oil that is fed to the top tray of Top product: For columns with condensers, the liquid
the sponge (secondary) absorber. and/or vapor streams from the reflux drum that exit
Stabilizer: A distillation column that removes light the process.
components from a liquid product. This term is used Topped crude oil: The bottom product from a crude
to describe debutanizer columns that remove C4 distillation column.
hydrocarbons from gasoline to in order control the Tray: A liquid/vapor contact device in a distillation
vapor pressure. column.
Steady state: A process where the mass and energy Threshold Limit Value: The amount of a contaminant to
flowing both into and out of the process are which a person can have repeated exposure for an
balanced. eight hour day without adverse effects.
Stoichiometric: Applied to reactors in which the reac- True boiling point distillation (TBP): A laboratory test
tants and products are defined in terms of the molar in which a petroleum oil is distilled in a column
quantities reacting, eg. 3H2 þ 2N2 5 2NH3 . The having at least 15 theoretical trays and a reflux ratio
stoichiometric coefficients are 3, 2, and 2 for H2, of 5.0. The distillate is continually removed and
N2 and NH3 respectively. further analyzed. The separation corresponds some-
Straight run: A product that has been distilled from what to a component by component separation and is
crude oil but has not been through a process in which a good measure of the true composition for the
its composition has been chemically altered. sample being distilled. As the temperatures in the
Stripper column: A loose designation applied to a dis- still increase, the pressure of the still is lowered to
tillation column in which light components are suppress thermal cracking of the sample. The mini-
stripped from a heavier liquid product. mum pressure for most TBP stills is about 38 mm
Stripping steam: Steam that is injected into the bottom Hg. This allows distillation of petroleum components
of a side stripping column or used to strip oil from boiling up to about 900–950 F (483–510 C) at
catalyst in a FCC operation. a pressure of 1 atm.
Stripping zone: The section of the column in which Two phase: A mixture consisting of one vapor in equi-
light components are stripped from a heavier liquid librium with one homogeneous liquid phase.
product. In conventional distillation columns, this is Undistributed component: A component in a distilla-
the portion of the column from the reboiler to the tion column separation zone that is totally recovered
feed tray. in only one of the products.
Sweet gas: Gas which has no more than the maximum Unsaturated compounds: Hydrocarbon compounds in
sulfur content defined by the specifications for the which some of the carbon atoms have multiple bonds
sales gas from a plant (i.e. a gas stream from which with other carbon atoms because of lack of hydrogen
the sulfur compounds have been removed). atoms to satisfy the carbon atom valencies.
Tail gas: the residue gas from an absorption column. In Vacuum distillation: A distillation column that operates
FCC gas plants, it is referred as the gas products at sub-atmospheric pressure. Vacuum distillation
from the secondary absorber column. allows the further distillation of heavy feed stocks at
Tertiary amyl methyl ether (TAME): used to enhance reduced temperatures and to minimize cracking
the octane rating of a motor gasoline pool. TAME is reactions.
produced by reacting the isoamylenes produced in Vacuum residuum: The heaviest product from
FCC processes with methanol. a vacuum distillation column.
Tetra ethyl lead (TEL): A compound added to gasoline Visbreaking: A process in which a heavy oil is thermally
to increase the octane rating. TEL has been super- cracked just enough to lower or break the viscosity. A
seded by other octane enhancers and is no longer small quantity of gas oil and lighter products is
used by refiners for motor gasoline. formed in the process.
Theoretical tray: refers to a vapor/liquid contact device Wash zone: A section in a column where the column
(e.g. distillation column) in which the liquid and vapor is washed of entrained heavy materials by
vapor leaving the device are in perfect phase equi- contact with a cooler injected liquid. A section of
librium. There are also perfect energy and mass bal- packed material is often used to promote good
ances for a theoretical tray. mixing of the liquid and vapor in the wash zone. All
Thermal cracking: The first cracking process, in which vacuum distillation columns have wash zones to
the oil was cracked by heating only. Thermal cracking remove heavy residual material that is carried up the
produces lower octane gasoline than catalytic column from the flash zone. If washing is inadequate,
processes. the heavy residual material forms petroleum coke
and plugs the column above the flash zone.
342
Nomenclature A ¼ Absorption factor, average

A" ¼ Effective absorptive factor
A ¼ actual desiccant bed area, ft2 ABi ¼ Absorption factor for each component at
ADES ¼ calculated area based on desiccant conditions of bottom tray
volume, ft2 ATi ¼ Absorption factor for each component at
Cycle ¼ drying cycle, hr conditions of top tray
DBED ¼ calculated bed diameter, ft Cp ¼ cp ¼ Specific heat, Btu/lb ( F)
E ¼ joint efficiency Ea ¼ Absorption Efficiency, or fraction absorbed
F ¼ loss factor for non-steady state heating Eo ¼ Overall tray efficiency, fraction
Flow ¼ gas flow rate, 106 std ft3 =day E, ¼ Stripping efficiency, or fraction stripped
h ¼ additional shell height for desiccant fai ¼ Fraction of vnþ1i absorbed by the liquid
supports and distributor, ft fsi ¼ Fraction of loi stripped out of the liquid
H1, H2 ¼ enthalpy of regeneration gas before and Gmi ¼ Mols individual components stripped per
after regeneration heater, Btu/lb hour
HGAS ¼ total regeneration gas requirement, lb DH ¼ Total heat of absorption of absorbed
HREG ¼ heat added to regeneration gas to components, thousand Btu/day
regenerate desiccant bed K ¼ Equilibrium constant, equals y/x, at average
HTOT ¼ heat required to regenerate dryer (ideal), tower conditions
Btu LMþ1 ¼ Mols/hour rich oil entering stripper
HT ¼ heater temperature, F LN ¼ Liquid leaving bottom absorber tray,
H2O load ¼ total water adsorbed, lb mols/hr
H2O Cont ¼ water content of gas, lb/106 std ft3 Lo ¼ Mols/hr lean oil entering absorber, or leaving
L ¼ desiccant bed length, ft stripper
MW ¼ molecular weight of regeneration gas Loi ¼ Molar flow of component ‘‘i’’ in entering
P ¼ dryer operating pressure, psig liquid to absorber
PICKUP ¼ useful design capacity of desiccant, %. M ¼ Number theoretical stages in stripper
Q ¼ actual gas flow rate at flowing conditions, m ¼ M (see above)
ft3/min N ¼ Number theoretical stages in absorber
R ¼ radius of desiccant bed, ft n ¼ N (see above)
S ¼ allowable stress, lb/in2 S, Se ¼ Stripping factor, average and effective,
SCF ¼ standard cube feet of regeneration gas respectively
t ¼ dryer shell thickness, inch S0 ¼ Edmister’s effective stripping factor
T ¼ gas flowing temperature, oR Ti ¼ Tray i, temperature, F
TBED ¼ temperature of desiccant bed, F TN ¼ Bottom tray temperature, F
VDES ¼ desiccant volume, ft3 TNþ1 ¼ Inlet rich gas temperature, F
VGAS ¼ superficial gas velocity, ft/min. To ¼ Lean oil temperature, F
WD ¼ desiccant weight, lb U ¼ Overall heat transfer coefficient between
WS ¼ steel weight of shell and heads, lb absorber outside surface and atmosphere,
WVES ¼ vessel weight, lb BUI/(ft2) ( F) (hr); usual value ¼ 3.0
Z ¼ compressibility factor V1 ¼ Mols/hr lean gas leaving absorber
rD ¼ desiccant density, lb/ft3 Vi ¼ Gas leaving tray i, mols/hr
Viþ1 ¼ Gas leaving tray, i þ 1, mols/hr
VN ¼ Vapor leaving bottom absorber tray, mols/hr
Nomenclature For Part 2, VNþ1 ¼ Mols/hr rich gas entering absorber
Absorption, Stripping and Vo ¼ Mols/hr stripping medium (steam or gas)
entering stripper
Adsorption Vii ¼ Molar gas flowrate of component ‘‘i’’ leaving
plate 1 in absorber
(Special notations, all others same as for Distillation Vnþ1i ¼ Molar gas flowrate of component ‘‘i’’ in
Performance Nomenclature, Part 1) entering gas to absorber
A0 ¼ Edmister’s effective absorption factor W ¼ Rate of flow, thousand lb/day
A" ¼ Outside surface area of absorber, ft2
343
X ¼ Number mols absorbed component or Yo* ¼ Number vapor phase mols of component in
stripped per mol lean oil entering column equilibrium with lean oil per mol of rich gas
X1 ¼ Number mols liquid phase component in entering
equilibrium with Y1
XiR ¼ Mols of a component in liquid absorbed per
mol of lean oil entering column
LXi ¼ Total mols of all liquid phase components
Subscripts
absorbed per mol of lean oil (omitting lean 1, 2, etc ¼ Components in a system
oil present in liquid phase, considered ¼ 1.0) Amb ¼ Ambient
XMþ1 ¼ Number liquid phase mols of component Avg, Av ¼ Arithmetic average
entering stripper per mol of lean oil
DG ¼ Discharge gas
Xoi ¼ Number liquid phase mols of component
e ¼ Effective
entering absorber with lean oil per mol of
i ¼ Individual components in mixture
lean oil
IG ¼ Intake gas
Y1 ¼ Number vapor phase mols of component
leaving top plate of absorber per mol rich gas Key ¼ Key component
entering absorber L ¼ Lean concentration end of column
Yi ¼ Mols component in vapor phase from tray Lo ¼ Lean oil
‘‘i’’/mol rich gas entering absorber Min ¼ Minimum condition
SYi ¼ Total mols of all vapor phase components o ¼ Operating condition
stripped per mol of stripping medium R ¼ Rich concentration end of column
YNþ1 ¼ Number vapor phase mols of component Ro ¼ Rich oil
entering absorber per mol rich gas entering S ¼ Absorbed components
344

Cdu 3

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Cdu 3

Uploaded by

Copyright:

Available Formats

Chapter 11

LPG and gas Refinery fuel gas

Vacuum gas oil Heating oils

Treating and blending

Figure 11-1 A flow diagram of a refinery distillation unit.

sections are employed to characterize crude oils as po-

11.1.2 Gravity, API Indonesia Minas 0.845

where K 10 indicates an aromatic material whereas K ¼ 15

11.1.10 Pour Point ( F or C)

hydrocarbon yields, (C1–C5), and a boiling point curve

Petroleum, Complex-Mixture Fractionation, Gas Processing, Dehydration

Wide Cut Data

Wide Cut Data

RONC 56.27 39.53 46.37 18.56 - - - - - -

Petroleum, Complex-Mixture Fractionation, Gas Processing, Dehydration

wt% lv% wt% lv% wt% lv%

Total C6s 3.36 4.13 Total to C6 3.44 4.23

C7 iso paraffins 1.18 1.44

Total C7s 3.91 4.55 Total to C7 7.35 8.78

C8 iso paraffins 1.39 1.64

Total C8s 4.31 4.86 Total to C8 11.66 13.64

C9 iso paraffins 1.30 1.51

Petroleum, Complex-Mixture Fractionation, Gas Processing, Dehydration

C10 iso paraffins 1.37 1.54

Total C10s 4.24 4.56 Total to C10 20.95 23.90

C11 iso paraffins 1.25 1.39

Total C11s 4.23 4.47 Total to C11 25.19 28.37

Crude sulfur % wt 1.320 Crude Sg Sg 0.8410 Crude API 36.7000

Sample Date: 01/08/93 Analysis Date:

Petroleum, Complex-Mixture Fractionation, Gas Processing, Dehydration

Wide Cut Data

Cloud Point C - 82.2 - 58.6 15.20 9.0.0 29.2 38.5 - -

Wide Cut Data

Petroleum, Complex-Mixture Fractionation, Gas Processing, Dehydration

wt% lv% wt% lv% wt% lv %

Total C6s 2.25 2.94 Total to C6 2.30 3.01

C7 iso paraffins 0.80 1.03

Total C7s 2.46 3.06 Total to C7 4.76 6.07

C8 iso paraffins 0.89 1.11

Total C8s 2.68 3.21 Total to C8 7.44 9.28

C9 aromatics 0.22 0.22

Table 11-4 Middle East: Arabian Heavy 27.3 (Cont’d)

wt% lv% wt% lv% wt% lv %

Petroleum, Complex-Mixture Fractionation, Gas Processing, Dehydration

Total C10s 4.24 3.15 Total to C10 13.59 16.48

C11 iso paraffins 0.84 0.99

Total C11s 2.92 3.26 Total to C11 16.51 19.74

Crude sulfur % wt 2.900 Crude Sg Sg 0.8905 Crude API 27.3000

11.5 Laboratory Testing

Table 11-5 Laboratory Distillation Tests

Test name Reference Main applicability

EFV (atmospheric, superatmospheric, or Nelsony Crude oil and petroleum fractions

Table 11-6 Temperature Corrections for Barometric Pressure

Laboratory pressure, mmHg

For use with ASTM D86 Distillation

11.5.3 Boiling Points

The use of Figure 11-5 is as follows:

11.5.5 Determination of ASTM IBP, 10%,

11.5.6 ASTM 10-90% Points 11.5.8 ASTM End Point of Blend

11.5.11 Smoke Point of Kerosenes Vapor

(11-8) Natural gasoline 18.4 20.1 1.092

Light gasoline 12.0 12.9 1.07

Nigeria Nigerian 0.26 <20

½1:0 þ 0:03264 cSt

SelenolÒ: (licensed by Norton Company) uses dime- Low corrosion/carbon steel.

Silica Gel SorbeadÒ-R Beads 49 5 8 mesh 0.25 60 F dew point

Silica Gel SorbeadÒ -H Beads 45 5 8 mesh 0.25 60 F dew point

Silica Gel Sorbead($)-WS Beads 45 5 8 mesh 0.25 60 F dew point