SPE 59765

Effect of Carbon Dioxide on Hydrogen Sulfide Scavenging

T. Salma, SPE, Baker Petrolite

Copyright 2000, Society of Petroleum Engineers Inc. CO2 compared to a few parts per million (ppm) of H2 S), which
This paper was prepared for presentation at the 2000 SPE/CERI Gas Technology Symposium further makes removal of CO2 less attractive and more
held in Calgary, Alberta Canada, 3–5 April 2000.
uneconomical. The need for selective removal of H2 S
This paper was selected for presentation by an SPE Program Committee following review of
information contained in an abstract submitted by the author(s). Contents of the paper, as
therefore emerged as a key criterion for cost effective
presented, have not been reviewed by the Society of Petroleum Engineers and are subject to sweetening of natural gas.
correction by the author(s). The material, as presented, does not necessarily reflect any
position of the Society of Petroleum Engineers, its officers, or members. Papers presented at Triazines form an important class of non-regenerable H2 S
SPE meetings are subject to publication review by Editorial Committees of the Society of scavengers. They are produced from the reaction between
Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper
for commercial purposes without the written consent of the Society of Petroleum Engineers is formaldehyde and an alkanolamine comprising 1 to 6 carbon
prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300
words; illustrations may not be copied. The abstract must contain conspicuous atoms [1]. Various sour gas-sweetening processes have been
acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O. used for a number of years. Incentives for use of triazine based
Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.
scavengers include:
a) Selectivity towards H2 S
Abstract b) Cost effective removal of H2 S
Effective removal of hydrogen sulfide (H2 S) is one of the c) Simplicity of operation
challenging problems associated with the production of natural d) Adaptability to existing process equipment
gas. Since H2 S is often present in natural gas along with e) Low toxicity and non-hazardous waste characteristics of
varying levels of carbon dioxide (CO2 ), selective removal of the spent scavenger generated by the hydrogen sulfide
H2 S from sour gas in the presence of CO2 is very important for reaction.
cost effectiveness of a scavenging operation. For over ten Factors Affecting H2 S Scavenging. A. Y. Al-borno [4]
years, triazine based scavengers have been successfully provides a thorough review of the various factors affecting
applied in the sweetening of natural gas. However, hydrogen sulfide scavenging some of which are:
observations of the effect of CO2 on the scavenging capacity - production ratios between gas/water/condensate
of triazine based scavengers have not been very consistent. - phase distribution
Initial documentation and case histories on performance of - operating temperature and pressure
triazine based scavengers by Dillon [1,2] reported that CO2 did - H2 S levels, actual and desired
not have any effect on the H2 S scavenging capacity of triazine. - mixing efficiency
However, recent studies by GRI [3] indicate that CO2 - contact time between the scavenger and the gas
significantly reduces the H2 S removing capacity of triazine. - chemical interference (if any)
To address this discrepency, this paper critically evaluates the Early case histories of hexahydrotriazine reported high
effect of CO2 on the scavenging performance of a triazine selectivity for H2 S [2]. However, recent studies indicated that
based H2 S scavenger. Results from controlled laboratory CO2 can a exert significantly adverse effect on the scavenging
experiments, using a wide range of CO2 partial pressure and performance of triazine based H2 S scavengers [4]. The current
H2 S to CO2 ratios indicate that CO2 has very little effect on the study was designed to critically evaluate the effects of CO2 on
H2 S scavenging performance of a triazine based H2 S H2 S scavenging by triazine based scavengers.
scavenger. Field data supports the observations made in the
laboratory experiments. The H2 S scavenger selected for this paper was a
hexahydrotriazine, which is a reaction product of
Introduction formaldehyde and monoethanol amine. Controlled laboratory
H2 S and CO2 are two contaminants that naturally exist in sub- experiments were conducted to study the effect of CO2 on the
quality natural gas. Removal of H2 S from the gas is often scavenging capacity of triazine for H2 S. Performance data on
required because of safety, environmental or corrosion this triazine-based scavenger was also collected from various
considerations. However, these factors do not dictate the fields with a wide range of CO2 concentrations to evaluate the
removal of CO2 . Chemicals that react with H2 S, very often effect of CO2 on the scavenging capacity of the chemical for
also have an affinity for CO2 . The concentration of CO2 H2 S. Results from laboratory studies as well as the field
relative to H2 S in natural gas is often very large (percent levels
2 T. SALMA SPE 59765

survey clearly indicate that H2 S scavenging capacity of the ± 0.5 min) were slightly shorter than those obtained with
triazine based scavenger is independent of the CO2 higher concentrations of CO2 (t = 16.5 ± 0.5 min).
concentration and the ratio of CO2 to H2 S in the feed gas. Set 3. The H2 S concentration was maintained at 20,000ppm
with a 50% triazine formulation charged to the tower for each
Experimental Procedure experiment. At 2% (20,000 ppm) H2 S, variation of CO2 from
Figure 1 shows the experimental setup used in this study to 10 to 98% yielded breakthrough times of t = 8.5 ± 0.5 min.
determine the effect of CO2 on the scavenging capacity of Results from sets 2 and 3 supported the observations in set 1
triazine. The contact tower, made from plexiglass was fitted that increasing the concentration of CO2 and/or increasing the
with a 70micron diffuser gasket to homogeneously disperse CO2 partial pressure does not have any adverse effect on the
the gas into the liquid scavenger solution. The contacting scavenging capacity of the triazine. These results also
section of the tower was 1.25” in diameter and 14” tall, confirmed that the selectivity for H2 S of the triazine
opening into a disengagement zone 3” diameter x 5” high to (hexhydrotriazine) formulation tested in this study is
separate the gas from the liquid. Methane (CH4 ), carbon independent of the partial pressure or concentration of CO2 in
dioxide (CO2 ) and hydrogen sulfide (H2 S) were metered into the natural gas.
the line using three individual rotameters. The first step was to
establish the flow rates of individual gas components using 3- Field Evaluation. To verify the observations made in the
way valves installed on each line. Once the flow rate of each laboratory experiment, performance data was also collected
component was stabilized, the next step was to charge a 100ml from various field locations using the same triazine scavenger
solution of known concentration of triazine into the bubble for removal of H2 S with different concentrations of CO2 in the
tower. A mixture of H2 S, CO2 and methane at known natural gas.
concentrations, 5 SCFH flow rate and 40 psia pressure was Table 3 summarizes the operating conditions encountered at
sparged through the solution. Concentration of H2 S was the different fields and the mode of application of the
monitored in the outlet gas as a function of time. An H2 S scavenger. In general, the optimum capacity of the triazine
monitor (Model HS50) was used to determine the initial based scavenger (hexahydrotriazine) used in this study is
breakthrough. Breakthrough time was defined as the time for determined to be 2 lbs of H2 S removed per gallon of scavenger
the outlet gas to reach 10ppm H2 S after passing through the in contact tower applications and ~ 1lb of H2 S removed per
scavenging solution and was used as the main performance gallon of scavenger in flow line injection. One of the primary
criteria for effective scavenging. reasons for lower performance in flow line applications is
Gas compositions were selected and blended in the laboratory mass-transfer limitations and relatively poorer mixing
to include a wide range of CO2 , H2 S and CH4 ratios ranging efficiencies than those available in contact tower applications.
from 1000 ppm H2 S, 500 ppm CO2 and the balance CH4 to Flow Line Applications. Inspection of field data shows that
20,000 ppm H2 S, 98% CO2 and no CH4 . Table 1, summarizes for CO2 partial pressures (p CO2 ) varying from 0.8 psi (Kansas)
the various compositions tested in the laboratory. to 40.7 psi (Louisiana), scavenging performance of triazine
was within the expected range for flow line applications.
Results Furthermore, increasing the ratio of CO2 to H2 S did not
Laboratory Studies. Table 2 provides a summary of the adversely affect the scavenging performance. For example, at
breakthrough times observed for various concentrations of CO2 :H2 S of 2000:1 (Michigan 1), the scavenging performance
CO2 , H2 S and CH4 . is 0.98 lb of H2 S removed per gallon of scavenger compared
Set 1. The H2 S concentration was maintained at 1000ppm to the scavenging performance of 0.94 lbs H2 S removed per
with a 5% triazine formulation charged to the tower for each gallon of scavenger with a CO2 :H2 S of 1000:1 (Louisiana).
experiment. The ratio of CO2 :H2 S was 15:1 in composition 1 Contact Tower Applications. Field data on contact tower
compared to 500:1 in composition 2. For a 5% triazine applications further illustrated the independence of scavenging
formulation, comparison of breakthrough times for performance relative to CO2 concentration in the feed gas. For
composition 1 (t = 53 min) and composition 2 (t = 60 min) example, for zero ppm CO2 (W. Virginia 1, pCO2 = 0) in the
demonstrate that there was no reduction in the H2 S scavenging feed gas, ~ 1.85 lbs of H2 S were removed by 1gallon of the
capacity at higher concentrations of CO2 . In fact, the scavenger compared to 1.63 lbs of H2 S removed per gallon for
breakthrough time was slightly higher for higher CO2 levels in a CO2 :H2 S ratio of 290:1 (W. Virginia 2, p CO2 = 19.5 psi) and
composition 2. The slight positive effect at higher only 1.0 lb H2 S removed per gallon of scavenger for CO2 :H2 S
concentration of CO2 was within the range of experimental ratio of 25:1 (Michigan 2, p CO2 = 4.6 psi).
error. Evaluation of field data, in agreement with laboratory
Set 2. The H2 S concentration was maintained at 10,000ppm observations, confirmed that the H2 S scavenging performance
with a 50% triazine formulation charged to the tower for each and capacity of the triazine formulation used in this study was
experiment. The concentration of CO2 was varied from zero independent of the carbon dioxide levels in the field gas.
to 99% CO2 in the gas mixture passing through a 50%
scavenger solution. Breakthrough times with no CO2 (t = 15.5
SPE 59765 EFFECT OF CARBON DIOXIDE ON HYDROGEN SULFIDE SCAVENGING 3

Conclusions Hexahydrotriazine”, paper presented at the 1992 GRI 4th

1. Based on laboratory and field data, it is concluded that the Sulfur Recovery Conference, Austin, Texas, October 4-6.
scavenging performance of the triazine formulation is 3. K. S. Fisher, S. J. Killion, J. E. Lundeen, “GRI Field
independent of the concentration and partial pressure of Testing of Direct-Injection H2 S Scavenging”, presented at
carbon dioxide in the feed gas. the 78th Annual GPA Convention, Nashville, Tennesse
2. Increasing the ratio of CO2 to H2 S has no adverse effect March 1-3, 1999.
on the scavenging capacity of the triazine based 4. Al-borno, A. Y., R. Ryel, “ Solid Formation in H2 S
scavenger. Scavenging and Recent Development”, paper presented at
the 1995 GRI 7th Sulfur Recovery Conference, Austin,
References Texas, September 24-27.
1. E. T. Dillon, “Composition and Method for Sweetening
Hydrocarbons”, United States Patent No. 4,987,512, (Dec
1990).
2. E. T. Dillon, “Gas Sweetening with a Novel and Selective

Figure 1: Experimental Setup for Measurement of Breakthrough Times for Triazine Solutions

Pressure Gauge
@ 40 psia

3” ID Plexiglass
~ 5” Tall to serve as
Disengagement Zone

Gas Outlet

H2 S Monitor

1.25” ID Plexiglass
~ 14” Tall

Diffuser Gasket
70 microns

Gas In

CH4 CO2
H2 S
4 T. SALMA SPE 59765

Table 1

Molar Composition of Gas Mixtures Tested in the Laboratory

Composition Concentration of Pressure CO2 CH4 H2 S H2 S

No. Triazine Solution (psia) mole% mole% mole% (ppm)
(%)
SET 1
1 5 40 0.05 99.4 0.1 1000
2 5 40 50 49.9 0.1 1000
SET 2
3 50 40 0 99 1 10,000
4 50 40 10 89 1 10,000
5 50 40 20 79 1 10,000
6 50 40 50 49 1 10,000
7 50 40 99 - 1 10,000
SET 3
8 50 40 10 88 2 20,000
9 50 40 19 79 2 20,000
10 50 40 98 0 2 20,000

Table 2
Breakthrough Times For Different CO2 :H2 S and Partial Pressures of CO2

Composition CO2 Partial Pressure Ratio of CO2 :H2 S H2 S Breakthrough Time ± 0.25
No. (psia) (ppm) (min)
SET 1
1 0.02 15:1 1000 53
2 20 500:1 1000 60
SET 2
3 0 - 10,000 15.5
4 4 10:1 10,000 15.8
5 8 20:1 10,000 16.0
6 20 50:1 10,000 17.0
7 39 99:1 10,000 17.0
SET 3
8 4 5:1 20,000 8.5
9 8 10:1 20,000 9.0
10 39 49:1 20,000 8.5
SPE 59765 EFFECT OF CARBON DIOXIDE ON HYDROGEN SULFIDE SCAVENGING 5

Table 3
Field Locations – Operating Conditions

Location Flow Rate Operating Operating H2 Sin H2 Sout CO2

(MMSCFD) Pressure Temperature (ppm) (ppm) (ppm)
(Psia) (o F)
Flow Line Injection
Kansas 0.20 40 75 29 4.5 20,000
Michigan (1) 150 740 80 10 3 20,000
Louisiana 30 1135 120 35 2.0 35,000
Contact Tower Application
W. Virginia (1) 1.2 30 30 200 15 0
Michigan (2) 1.0 915 120 200 15 5,000
Alberta, Canada 12.7 595 82 21 6 5,000
W. Virginia (2) 3.1 150 70 450 0 130,000

Table 4
Scavenger Performance in the Field for Varying CO2 to H2 S Ratios

Location CO2 Partial Pressure CO2 :H2 S Pounds of H2 S Removed per

(psia) Ratio gal of scavenger
Flow Line Injection
Kansas 0.80 690:1 1.20
Michigan 1 14.8 2000:1 0.98
Louisiana 40.7 1000:1 0.94
Contact Tower Application
W. Virginia 1 0 0 1.85
Michigan 2 4.6 25:1 1.00
Alberta, Canada 3.0 238:1 1.30
W. Virginia 2 19.5 290:1 1.63