SPE 1

Laboratory Development of a Novel, Non-Triazine-%ased Hydrogen Sulfide

Scavenger and Field Implementation in the Haynesville Shale
F. Chaudhry, SPE, G. L. Mobley, Encana Oil and Gas, (USA) Inc.; Y. H. Tsang, SPE, S. Ramachandran, SPE, V.
Jovancicevic, SPE, S. C. Braman, SPE, E. N. Rowton, A. P. McDonald, J. A. Davis, Baker Hughes Inc.

Copyright 2013, Society of Petroleum Engineers

This paper was prepared for presentation at the SPE International Symposium on Oilfield Chemistry held in The Woodlands, Texas, USA, 8–10 April 2013.

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 th e paper have not been
reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessarily reflect any position of the Society of Petroleum Engineers, its
officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written consent of the Society of Petroleum Engineers is 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 acknowledgment of SPE copyright.

Abstract

Hydrogen sulfide (H2S) is found in many shale gas fields. It is important to remove the H2S to allow for safe transport of the
produced gas and to reduce corrosion concerns. The H2S concentration present in the gas is typically low enough that it
becomes uneconomical to utilize a regenerative facility such as an amine plant. For this reason, liquid H2S scavengers are the
most common mitigation strategy, with nitrogen-based triazines being the most commonly utilized scavenging chemistry.
Hydrogen sulfide scavengers can be introduced by direct injection 2 into mixed production or separated wet gas pipelines.
They can also be reacted with H2S in scavenger flooded contact towers. Applications utilizing towers consume less H2S
scavenger; however, the use of contact towers introduces additional capital costs in the operation. Often when high pH,
triazine-based H2S scavengers are used, calcium carbonate scale deposition can occur because of neutral pH brine, requiring
the facility to be shut in and treated with mineral acids. The reaction product of triazine with H2S can be difficult to dispose
of, which increases facility operating costs (e.g. separate disposal tanks and dedicated disposal wells for spent scavenger).

A novel, fast-acting, non-triazine based H2S scavenger has been developed. This new H2S scavenger reacts quicker and has
better efficiency than other non-nitrogenous based scavengers (e.g. glyoxal). The neutral pH of the new product eliminates
the concern for calcium carbonate deposition that is often experienced with triazine-based scavengers. The new scavenger
has been tested for thermal stability and compatibility with other production chemicals. The product has the ability to
partition in oil and water phases, making it a more versatile scavenger. Finally, the new product has a better environmental
profile than other commonly used H2S scavenger chemistries.

A field trial was conducted using the new non-triazine H2S scavenger. The trial included direct injection into mixed
production and separated wet gas lines. Innovative engineering solutions were devised to apply the new H 2S scavenger. The
combination of the novel application methodology, with the new H2S scavenger chemistry, resulted in a more cost-effective
means of treating sour shale gas than is currently achieved with conventional triazine-based H2S scavengers utilizing contact
tower applications.

Introduction

The Haynesville Shale is an unconventional, natural gas shale reservoir (280°F – 360°F BHT), primarily located in
Northwest Louisiana. The Haynesville formation produces primarily natural gas and brine, with no liquid hydrocarbon, from
a total vertical depth of 10,000 to 12,000 feet. The majority of the gas produced from the Haynesville formation is sweet (<
25 ppm H2S); however, roughly 10% – 15% of the gas produced is relatively high in H2S (25 – 400 ppm). Operators are
required to sweeten this gas in order to meet pipeline specifications for sale. The concentration of H2S varies greatly based
on gas composition, geographic location, and well flow dynamics.
2 SPE 16

The specific wells discussed in this paper are located in the vicinity of Pleasant Hill, LA, which is geographically situated in
the sour region of the production area. The pipeline outlet ppm for H2S in the production area ranges from 4 – 50 ppm.
Typical, traditional sweetening methods by most companies are amine towers/flooded towers and direct injection of triazine
chemistry. These methods have generated operational issues with the brine from this formation because of the higher
calcium content of the brine resulting in high scale tendencies.

Laboratory Development of Non-Triazine

Experimental Setup and Procedure

A new test apparatus similar to one developed by Mendez et al.1 was designed and constructed. A continuous gas flow
(CGF) test apparatus is designed using a glass pressure reactor. The test chamber contains fluid that is continuously sparged
with an H2S gas mixture. A custom built gas delivery system consisting of mass flow controllers delivers precise quantities
of gas containing H2S to the pressure vessel. The concentration of hydrogen sulfide leaving the reaction vessel is measured
using an H2S solid state electrochemical sensor based analyzer.

Laboratory Results

Three (3) scavengers were tested in the CGF apparatus. They are the commercially available triazine, the commercially
available glyoxal and the new, proprietary non-triazine hydrogen sulfide scavenger (HSS). A known amount of H2S /
nitrogen / CO2 mixed gas is continuously sparged into the test fluid at a constant rate until a steady state concentration of H2S
is achieved (approximately 3,200 ppm w/w in the gas phase in the example shown in figure 1). A known amount of the
scavenger (200 ppm) is then injected into the CGF reactor. The same amount of scavenger is used in each test. After
injection of the scavenger, H2S concentration at the outlet decreases before reaching the maximum scavenging efficiency, and
then starts to increase slowly to the baseline level as the H2S scavenger is being spent. The concentration of H2S was
continuously measured with time and recorded. Figure 1 depicts the real time H2S level for the three scavengers tested.

It can be seen that both HSS and triazine reduce the concentration of H2S significantly faster than glyoxal based on initial
decrease of H2S level. The reduction in H2S concentration is greatest with the new, non-triazine HSS. In addition, the
amount of H2S that reacts with the same mass of injected H2S scavenger is largest with the non-triazine HSS for this test,
which is derived from the area under the baseline for each curve.

Figure 1. H2S measured after injection of the new proprietary, non-triazine Hydrogen Sulfide Scavenger (HSS), commercially
available glyoxal and commercially available triazine as a function of time.
SPE 16 3

Environmental and HS&E Profile

To evaluate its environmental profile, the new scavenger was tested for biodegradability and its aquatic toxicity to
Skeletonema costatum and sheepshead minnows, and compared to other scavengers. Table 1 depicts that the new non-
triazine HSS product and the commercially available glyoxal are more environmentally friendly than the commercially
available triazine-based product.

Table 1: Screening test for the environmental profiles of HSS and commercially available glyoxal
(*21 days test / **96 hr test).

Hydrogen Sulfide
Commercially Commercially
Property Scavenger
Available Glyoxal Available Triazine
(non-triazine)

Biodegradability, 28 days 91 % 94% 100% *

Effective Concentration, 50 % , 72 hr for Skeletonema

> 100 mg/l > 200 mg/l 5.4 mg/l
Costatum

Lethal Concentration, 50 % , 72 hr for Sheepshead

> 2500 mg/l 2031 mg/l 180 mg/l **
Minnow

In terms of handling of the new product and HS&E profile, the new non-triazine HSS product is not DOT regulated and there
is no reportable quantity of the new product in the event of a spill. In addition, the commercially available triazine has a
strong amine-like odor, while the new non-triazine HSS product is essentially odorless. The spent non-triazine HSS product
can be comingled with a produced water stream for disposal, whereas spent triazine product may require a dedicated
tank/disposal well. Table 2 summarizes the various HS&E properties of the three scavengers studied here.

Table 2: Comparison of HS&E Properties of HSS, Glyoxal and Triazine.

Hydrogen Sulfide
Commercially Available Commercially Available
HS&E Property Scavenger
Glyoxal Triazine
(non-triazine)
Odor None None Amine - like
DOT classification Not regulated Not regulated Flammable
DOT Reportable Quantity Not applicable Not applicable 5988 gallons
Clean Air Act Section 112(b) HAPs Not listed Not listed Listed

Field Implementation

The traditional treating method for new wells is to use a triazine-based scavenger to sweeten gas using direct injection or
flooded towers to meet pipeline specifications. Triazine-based scavenger is an exceptionally cost-effective chemistry;
however, at times it presents operational challenges. For example, because of the high pH of the triazine chemistry, it is
often incompatible with high calcium brine leading to calcite scaling. Figure 2a shows an increasing ratio of triazine-based
scavenger to produced water from left to right. One can see the increasing amounts of calcium carbonate scale layer at the
bottom of the bottles. Figure 2b demonstrates that the same produced water remains clear with the non-triazine chemistry at
various ratios. Another operational challenge occurs when treating at reduced dosages for optimization purposes. The
optimization practice can often overspend the triazine chemistry which results in solids3, whereas with non-triazine HSS
product one can optimize closer to the pipeline specifications. For example, if treating 100 ppm H2S gas with triazine, where
the pipeline specification is 50 ppm, the operator must treat to ~5 ppm to prevent solid overspent product from building up in
the tower, due to the efficiency of triazine-based scavenger. When using non-triazine product, the operator would be able to
treat closer to the pipeline specification of 50 ppm, therefore reducing chemical demand by ~50% in this scenario. Lastly, the
spent triazine scavenger can begin to solidify at temperatures of 55°F or below.
4 SPE 16

The issues associated with scaling, overspending chemical, and solidification of spent chemical can be effectively mitigated
with mechanical and operational procedures, typically resulting in higher operational costs. For example, operators can
install scavenger injection downstream of primary separation to minimize the amount of high calcium produced water that
comes into contact with the triazine scavenger. However, it was found that in Haynesville Shale applications, even a small
amount of water carry-over downstream of the separator can cause scaling in the sales flowline when the brine comes into
contact with triazine chemistry. In addition, supplemental methanol and insulation of flowline and vessels have been utilized
to prevent the temperature from dropping below that at which the spent triazine begins to solidify.

Figure 2a: Ratio of triazine based scavenger (mls) to produced water (mls) increasing from left to right.

Figure 2b: Ratio of produced water (mls) to non-triazine HSS (mls) increasing from left to right.

Contact Loop Application

In an effort to mitigate the aforementioned operational issues, Encana and Baker Hughes implemented the new, neutral pH,
non-triazine sweetening scavenger in a contact loop and a pipeline application.

The new non-triazine scavenger was first implemented in a contact loop application. Figure 3 shows a simple process flow
diagram of such an application. The mixed production of natural gas and produced water from the wellhead first enters a
primary production separator (Figure 4a). The produced water then enters the produced water tank while the bulk water/free
gas enters the contact loop (Figure 4b). The contact loop is an extended flow loop that is installed immediately downstream
of the primary production separator. A scavenger injection point is installed with atomizers and pulsation dampeners at the
beginning of the contact loop. The sweetened gas then enters a secondary, spent scavenger separator where the spent
scavenger is removed from the gas phase, and the treated gas enters the sales flowline. The spent scavenger then enters the
produced water tank where it comingles with the produced water from the production separator. Unlike the triazine
SPE 16 5

chemistry, the new non-triazine spent chemical can be disposed of with the produced water and does not require special
handling, such as a dedicated disposal tank/well.

Figure 3. Process flow diagram of a contact loop application setup.

Figure 4a (left): View of primary production separator (left), inlet and outlet of contact loop, and spent separator (right). Figure 4b
(right): Contact loop and production facility overview
6 SPE 16

The length and ID of the contact loop was chosen in order to achieve optimal efficiency. The length of the loop was designed
such that the minimum contact time between high concentrations of H 2S in the gas phase and liquid scavenger droplets can
be achieved. Contact time is defined as the amount of time between the scavenger first entering the contact loop through an
atomizer or an injection quill and the moment when the scavenger is removed by the spent scavenger separator. Typically, a
minimum of 20-30 seconds of contact time (velocity ft/sec) is required to achieve optimal efficiency of the new non-triazine
HSS product. This minimal contact time can depend on several other factors, such as the amount of produced water carry-
over, the amount of mixing, etc. An appropriate loop ID was also selected in order to bring the linear gas velocity above the
critical velocity. The critical velocity is the gas velocity at which the gas-liquid flow regime is in mist flow inside the loop.
This is critical in terms of keeping the atomized scavenger droplets in a mist form without dropping out to the bottom of the
line too early. By maintaining a mist flow environment, the mass transfer between the gaseous H2S molecules and the liquid
scavenger droplets can be maximized by maintaining a minimal droplet size for the scavenger (hence a maximum surface
area). The ratio of the actual, linear gas velocity to the critical velocity (to maintain a mist flow) defines what is known as the
“mixing ratio”. Ideally one would want to maintain the mixing ratio above one (1) to achieve optimal scavenger efficiency
(or consumption rate).

Beginning in the summer of 2011, Encana and Baker Hughes began trialing and applying the new, non-triazine HSS using
the contact loop concept for over 10 wells in Haynesville shale. The average inlet H 2S entering the wellhead for these 10+
wells ranged from 56 ppm to 280 ppm. The average H2S outlet after scavenging with the new chemistry ranged from 14 ppm
to 36 ppm depending on the pipeline specification requirement. The average gas flow rate for these wells ranged from 1.2
MMscfd to almost 11 MMscfd, with an average wellhead temperature ranging from 83oF to 103oF. However, if the
conditions are achieved by applying the flow loop engineering design described above, it appears that the range of operation
(gas rate, temperature, H2S level) can be even broader. In fact, we have applied the new product in a pipeline with as much
as 22 MMscfd of gas with success (see next section). Table 3 summarizes the average production rates and conditions of the
wells that have permanent installations in the contact loop application with the new scavenger.

Table 3. Summary of average production rates and conditions for nine of the wells that have
implemented the contact loop concept with the new, non-triazine HSS in Haynesville Shale play.

Average inlet H2S Average outlet H2S Average gas rate Average temperature Average tubing
(ppm) (ppm) (MMscfd) (F) pressure (psig)
Well #1 280 36 1.2 90 996
Well #2 82 29 5.2 96 1227
Well #3 152 14 10.8 83 1030
Well #4 56 23 1.2 86 901
Well #5 182 27 8.0 103 1056
Well #6 127 22 7.0 103 1049
Well #7 120 17 8.1 98 1010
Well #8 124 20 7.3 96 986
Well #9 107 18 8.0 94 993

Effects of Contact Time and Mixing Ratio

Figure 5 below illustrates the effect of contact time and mixing ratio for three (3) of the wells that are currently using the non-
triazine scavenger in a contact loop application. The well parameters for these three (3) wells (wells 10, 11 and 12) are not
shown in Table 3, but are within the operating conditions outlined for the wells in Table 3. The amount of contact time,
mixing ratio and scavenger consumption rate are normalized using data from Well #10, which shows the best efficiency
(lowest consumption rate) of the product among the three wells. The scavenger consumption rate is defined as the volume of
scavenger required to remove one pound of H 2S by weight. The effect of mixing ratio is first examined by comparing Well
#12 to Well #10 where they have comparable contact times. By reducing the mixing ratio by 47.5% (from 100% to 52.5%),
the scavenger consumption rate is found to increase by 91.7% (from 100% to 191.7%). Because of the lowered velocity to
keep the scavenger droplets in a mist flow environment, the scavenger consumption rate increased significantly.

The effect of contact time can be examined by comparing Well #12 to Well #11. By reducing the contact time by 13.8%
(from 103.4% to 89.1%), the consumption rate is found to increase by 28.6% (from 191.7% to 246.6%). Notice that the
mixing ratio of Well #11 is higher than that of Well #12 (65.4% vs. 52.5%). If the mixing ratios were the same for the two
wells, one would expect their differences in scavenger consumption rates to become even greater.
SPE 16 7

Figure 5: Effect of mixing ratio and contact time on consumption rate of the new non-triazine HSS.

Pipeline Application

Learnings from the contact loop applications inspired Encana and Baker Hughes to develop methods for long-term strategy
for future development in sour production areas of the Haynesville play. It was determined that higher gas volumes could be
treated in pipeline applications. By treating at higher gas volumes through a pipeline application, the parameters necessary
for efficient treating were achieved (i.e. high mixing ratio and long contact time). In addition to the improved efficiency and
lowered chemical consumption rate, treating in a pipeline application reduced the overall facility footprint required for
sweetening the gas because the installation of a long contact loop was not required.

To determine the most efficient scavenging application for future development, a 3-well pad was evaluated for scavenging in
a pipeline. The average gas rate was approximately 22 MMscfd at the start of the pilot and has dropped to about 14 MMscfd
recently. The contact time of this system is over 5 minutes, which is well over the recommended 20-30 seconds.
Additionally, the mixing ratio for this system averages above 1. The average inlet H2S ranged from 75 to 160 ppm, and the
average outlet H2S always maintained below the pipeline specification of 50 ppm. The temperatures of the applications
ranged from 78oF to 125oF. Figure 6 depicts a simple process flow diagram of the pipeline application. The mixed
production from the three wells first enters their individual primary separators. The separated gas then comingles. The non-
triazine HSS is then injected into the comingled gas at the beginning of the pipeline. The gas and scavenger mixture then
enters a slug catcher at the end of the pipeline where the spent scavenger becomes separated and is sent to disposal with
produced water.
8 SPE 16

Figure 6: Process Flow Diagram of pipeline application.

Cost Comparison

Initially, the gas at this 3-well pad was treated with three individual flooded towers due to initially high gas volumes each
well produced. As gas production declined, treatment was accomplished through three individual contact loops. Most
recently, H2S is being treated in the comingled gas of the pipeline. A cost comparison was performed for four scenarios by
treating H2S with (1) three rental flooded towers of triazine based chemistry, (2) three purchased contact towers of triazine
based chemistry, (3) three contact loops using the new non-triazine HSS, and (4) the large 8-inch, 1.2 mile pipeline
application with the non-triazine HSS (Table 4). The pilot test scavenging results are shown below in Figure 7.

Table 4. Costs included in comparison of the four cases of application for the 3-well pad at Haynesville
Shale play (DI = Direct Injection) (*assumed 10 years of life time for towers and contact loop).

Case 3 (DI into three

Case 1 (three rental Case 2 (three purchased Case 4 (DI into pipeline
contact loops with non-
towers with triazine) towers with triazine) with non-triazine)
triazine)
Tower rental fee X
Tower capital cost* X
Flow loop capital cost* X
Methanol cost X X
Scale Inhibitor cost X X
Disposal cost X X
% average cost savings
NA 28.9 44.1 47.0
(compared to scenario 1)

In the base case where three flooded towers were used with triazine (red line), the calculated cost includes the triazine
chemistry cost, the tower rental fee, and the separated disposal cost due to incompatibility of the spent scavenger with
Haynesville brine. Additionally, supplemental scale inhibitor and methanol costs are included for mitigation of scale
precipitation in the towers and to minimize solid formation with the triazine chemistry, respectively.

In case 2 (green line), a purchased contact tower was used. The calculated costs include the same parameters outlined in case
1 minus the tower rental fee. However, the calculated costs for this second case includes the capital cost of purchasing three
SPE 16 9

towers, with each tower having an expected life of 10 years. Based on data from the 4 months depicted in Figure 7,
purchasing three towers would provide an average cost saving of about 29% comparing to three rental towers.

In case 3, the new, non-triazine chemistry is used in three direct injection contact loops (purple line). The cost includes the
scavenger costs and the capital cost of three contact loops with an expected life of 10 years. In this application, there is no
extra disposal cost because the spent scavenger can be disposed of with the produced water. In addition, there is no
supplemental scale inhibitor cost to mitigate the scale buildup. Based on data from the 4 month case study shown in Figure 7,
using the new scavenger with the contact loop concept would provide an average cost saving of about 44% compared to the
three rental towers.

In case 4 (blue line), the new, non-triazine chemistry is directly injected into the large pipeline, which the cost includes only
the scavenger cost. Based on the 4 months of data presented in Figure 7, using the new scavenger in a pipeline would
provide an average cost savings of about 47% when compared to the three rental towers.

Figure 7: Cost comparison of four scenarios to treat high volumes of gas and high inlet H2S.

Conclusions

In summary, both contact loop and pipeline applications have taken the sour treating methodology to a different level in the
Haynesville Shale play. It is important that all three of the following factors are met for these applications to be efficient: (1)
gas velocity must be above critical velocity to maintain a mist flow; (2) the contact loop needs to be sized adequately to
maintain sufficient time for the scavenging reaction process to occur; and (3) bulk produced water needs to be removed from
the gas stream via primary separation. We have demonstrated that by employing the appropriate engineering design with the
new non-triazine HSS, one can achieve substantial cost savings in the Haynesville Shale play. Table 5 summarizes the
tangible and intangible benefits of the new, non-triazine based HSS.
10 SPE 16

Table 5. Benefits of the non-triazine based HSS.

Benefits Comparison Commercially Available Triazine Non-triazine Scavenger

Environmentally friendly NO YES

Reduces H2S in produced water tank headspace NO YES

Odor YES NO

DOT reportable quantity YES NO

Requires dedicated disposal tank & well YES NO

Can increase carbonate scaling tendency YES NO

Extra scale inhibitor required YES NO

Co-inject methanol in spent tanks YES NO

Acknowledgements

The authors would like to thank Encana Oil and Gas USA, Inc. and Baker Hughes, Inc. for supporting the implementation
and success of the project. The following individuals should be acknowledged for their involvement in this project: Initial
content review was provided by Jennifer Fichter of Encana. Additional support for the project was provided by Steve Short,
P.E., Jack Kean, Bill Lofton, and Jake Wittnebert, P.E., of Encana. Laboratory support was provided by Rose Tompkins of
Baker Hughes. Field monitoring was preformed by Aaron Lee, Joshua Gilcrease, Donald Brown, Jesse Laurence, Michael
Mooney, Jeff Denney, and Brandon Mayes of Encana and Matthew O’Neal, Evan Hardy, Jeremy Descant, and Jeremy Hill of
Baker Hughes. Logistical support was provided by Cody Parker, Scottie Foster, and Glenn Lauga of Encana and Justin
Jackson of Baker Hughes. Test separation equipment was provided and monitored by Todd Bergeron and Ricky Boyd of
Express Energy Services.

References

1. C. Mendez, A.A.O. Magalhaes, P.A. Ferreiria, and E.C. Bastos, “H2S Scavengers Injection: A Novel Evaluation Protocol to
Enhance The Integrity of Production Lines”, OTC 20168, May 2009.
2. J.G.R. Eylander, H.A. Holtman, N.A. Maatschappij, T. Salma, M. Yuan, M. Callaway, and J.R. Johnstone, “The Development of
Low-Sour Gas Reserves Utilizing Direct-Injection Liquid Hydrogen Sulphide Scavengers”, SPE 71541, ATCE 2001.
3. G.N. Taylor, and R. Matherly, “Structural Elucidation of the solid Byproduct from the Use of 1,3,5-Tris(hyroxyalkyl)hexahydro-
s-triazine Based Hyrogen Sulfide Scavengers”, Ind. Eng. Chem. Res., 2011, 50, 735-740.