You are on page 1of 5


Analysers and next

generation SRU control
Acid gas analysis has arrived and tail gas analysers are now coming into the third generation.
S. Simmonds and R. Hauer of AMETEK discuss current trends in SRU process analysers and
A. Henning of Jacobs reports on experience to date with Advanced Burner Control+ (ABC+)
which now includes sour water acid gas (SWAG) control.

he history of on-line process analyt- fied Claus process coincided with the first The life cycle of an analyser is expected
ics is a relatively short one. Devel- attempts to control the modified Claus pro- to be 15-25 years and some analysers can
opment was driven by the need for cess using an on-line analyser and closed be in the field for 30 years or more if prop-
process control of high value hydrocarbon- loop control. erly protected and maintained. The prod-
based products. The first on-line analyser The first report of an on-line tail gas uct life cycle of an analyser is of the same
applications came during the second war. analyser was a technical paper based order, 15 years or more and is mostly sub-
Rapid development came in the 1970s on a gas chromatograph installed at the ject to the obsolescence of the electronic
with the advent of the microprocessor and Dow Chemical Freeport TX (USA) facility components.
resultant chemometric techniques. Led in 1970. In 1972 Amoco Oil and Dupont
by Phillips Petroleum, Union Carbide and Process Instruments published a paper on Process gas analysers in SRUs
Dupont Chemical, amongst others, these an ultraviolet (UV) based tail gas analyser The suite of process gas analysers for a
initiatives were the antecedents of todays installed at the Amoco Whiting refinery, sulphur recovery unit (SRU) can vary from
Siemens Applied Automation, ABB Analyt- Indiana, USA. Shortly after this, in 1974 one up to as many as eight analyser tags
ics and AMETEK Process Instruments. Western Research participated in a pilot (see Fig. 1). The (H2S/SO2) SRU tail gas
Whether or not sulphur can be consid- study using a prototype UV-based tail gas analyser is the most common. Every SRU
ered a high value product, in most cases analyser at the Shell Waterton gas plant in (or enhanced SRU, Superclaus, sub dew-
the driving force for process measurement Alberta, Canada. point) will have this critical, primary ana-
and control of the SRU is largely environ- Subsequent developments by vari- lytical measurement. If the SRU has an
mental. The US EPA Clean Air Act of 1970 ous manufacturers were all based on UV associated amine based TGTU there is a
and the ground-breaking study by Alberta spectroscopy and extractive sampling tech- second critical analyser and this analyser
Environment on the capability of the modi- niques and that remains the case today. can be in one of three locations. The third

Fig 1: Simplified process flow diagram of a SRU showing 8 analyser tags

Source: Ametek stack gas

emissions monitoring
SO2 NOx (mass flow)
Claus sulphur recovery unit Tail gas treater unit unit
absorber outlet
acid gas feed forward quench column H2 H2S (COS)
hydrocarbons H2S NH3 start up combustion Claus tail gas off-gas
(H2O CO2) oxygen O2 H2S SO2 (COS CS2) H2 H2S AT6 AT7


converters & reduction gas cobalt molybdenum
H2S condensers generator reduction reactor thermal
feed oxidiser
H2S recycle
inlet reactor furnace
AT5 to Claus inlet
& waste heat
exchanger absorber
reduction reactor
sulphur pit & stripper
off-gas H2 SO2
sweep gas AT3 sulphur pit columns

Sulphur 356 | January-February 2015 1


most common analyser tag is the emis- Fig 2: Model 888 tail gas analyser TGTU analyser
sions analyser, while not quite yet univer-
sal, in some cases (even when an end user The sample point location of the TGTU
is paying for high sulphur efficiency) the analyser varies depending on who is pro-
governing environmental authorities occa- viding the front end engineering design.
sionally do not specifically require an emis- In 2002 AMETEK developed a combined
sions analyser, the emissions analyser is ultraviolet (sulphur) and thermal conductiv-
the one measurement that confirms the ity (H2) analyser and to date have provided
recovery efficiency on a continuous basis. ~180 analysers for various amine based
After these three, the feed gas analyser TGTU processes. The measurement of the
is becoming more prevalent in front end (excess) hydrogen is the primary measure-
engineering design. ment in this application. As the H2 remains
As the industry drives to tighten control essentially unchanged after the CoMo
loops and reduce upsets in the process, reactor (with a slight increase across the
the analysis of the feed-forward acid gas quench tower due to renormalisation after
will become an industry standard. The the removal of the water) AMETEK consid-
results are clear that feed-forward control Source: AMETEK ers any of the three sample points to be
is both practical and worth the (relatively non consequential in terms of H2 measure-
small) capital investment. The industry is ment based on experience. As to the sul-
still in the education phase and needs to The first generation analysers proved phur gas(es), where and what to measure
be assured of certain practical implications the measurement to be viable, the second remains variable according to the designer,
but the need and benefits of feed-forward generation analysers improved on sample the spread of sample point locations are
control are already well established. If an handling and improved reliability to keep the as follows:
acid gas analyser has not been included analyser in cascade control, while the third l The majority (~150) are at the absorber
in the FEED, then the design should be generation analysers address the common outlet measuring H2S and H2. Of these
revised to allow for simple installation of failure modes external to the analyser. 150, ~15 have added the additional
a feed-forward analyser at a later date by The current design of the close-cou- analytical capability of COS meas-
installing the sample tap and associated pled (top of the pipe) tail gas analyser urement and it is interesting to note
heated acid gas probe, saving consider- was due for updating which presented almost all of these have been end users
able time, effort and cost in the future. the opportunity for some additional fea- upgrading an old GC and convinced of
The next most common analyser is the tures. That said, given the close-coupled the utility of the COS measurement.
pit gas analyser followed by an analyser to analyser has been installed in excess of The secondary measurement of the
measure the O2 at the outlet of the waste 1,200 units and the well proven success H2S (after the H2 measurement) (25-
heat boiler for exact stoichiometric com- of the technique, the goal of the update 500 ppm) at this location is a measure
bustion of natural gas during start-up and was to be evolutionary not revolutionary. of the efficiency of the amine absorber
shutdown. The principal improvements offered in the and the COS (5-500 ppm) is a measure
third generation are; of the CoMo reactor catalytic activity as
Tail gas analyser l Flange temperature RTD and alarm: An illustrated in the process upset in Fig.3
There is a choice of two tail gas analys- embedded RTD in the process connec- where the COS momentarily reached
ers - the top of the pipe (close coupled tion flange provides an alarm for poor 400 ppm during an upset. Both of these
model 888 no sample line analyser quality steam. This advanced warning sulphur values contribute to SO2 emis-
see Fig.2) or the extractive (sample line eliminates unexpected plugging. sions and can help to isolate the third
analyser, model 900 ADA). Over the last l Automatic flow control: This provides a contribution being sulphur pit sweep
five years (~600 tail gas analysers) ~80% method to automatically adjust the flow gas if vented to the incinerator
of tail analysers for new construction have rate of the analyser relative to process l Another ~25 analysers have been
been top of the pipe (close coupled type). pressure (especially important if there installed (further upstream) at the out-
A sample line type analyser is selected is entrained sulphur). let of the quench tower. Here the sec-
for three primary reasons; firstly for instal- l Ambient temperature rating: Improved ondary measurement of (~2.5%) H2S
lation in an analyser house for extreme thermal isolation between the oven and after the H2 measurement is a measure
climates, secondly if the sample point electronics results in an increase in the of the efficiency of the SRU as it rep-
location is compromised due to poor piping ambient temperature specification from resents all of the sulphur compounds
design, and finally if the end user wishes 50C to 60C. This makes installation (H2S. SO2, COS, CS2, Sv, S liq) leaving
to have the additional measurements of in regions like the Middle East possible the SRU. While the amount of trace
COS and CS2 for process optimisation without the need for an air conditioned COS (5-500 ppm) is the same as the
(only normally required for legacy straight shelter (with the aforementioned need absorber outlet it cannot be effectively
SRUs with no TGTU or for enhanced Claus for protection from solar gain an impor- measured at this point as it is masked
SRUs). tant detail). by the %level H2S
AMETEKs tail gas analyser has gone l Double-block isolation from the pro- l Approximately five analysers for H2/SO2
through two full generations and is about cess: This provides an option for any have been installed further upstream
to enter a third. site specific safety considerations. yet, at the inlet to the quench. The util-

2 Sulphur 356 | January-February 2015


Fig 3: COS upset data from TGTU absorber Feed (acid) gas analyser
When considering the air demand of the
SRU process, primary air control using
Source: AMETEK feed-forward flow ratio of air to acid gas
accounts for ~90% of the combustion air.
Secondary air control using tail gas analy-
sis (2:1 H2S/SO2 ratio) controlling a trim
air valve by feedback control (or cascading
to the feed-forward controller) accounts for
the ~10% balance of the required combus-
tion air. Feedback control based on pro-
cess analysis provides the most precision,
however, it is impaired by the ~30 second
process lag time, especially if the composi-
tion of the acid gas changes rapidly.
Under steady state conditions air control
is stable, but in the case of sudden changes
in acid gas composition it is not uncommon
for air control requirements to exceed the
ity of measuring SO2 here is to alarm for or enhanced Claus SRU as it can be ability of the feedback control loop. In these
SO2 breakthrough from the CoMo reac- directly related to sulphur recovery effi- cases the excursion from the set point of a
tor before seeing a drop in the pH in the ciency and used as an optimisation tool 2:1 H2S/SO2 ratio can be extreme, and the
quench water. While this sample point . Mass emission requires the SO2 analyti- cause of a serious loss of recovery efficiency
is much less common, one advantage cal measurement be made on a hot-wet and increased emissions.
is the analyser can be placed in front basis because the SO2 and correspond- With advances in certain technology
of a diverter valve and never be dead ing velocity measurement must be on the and process control, the implementation of
legged in the case of TGTU bypass or same basis. Mass emission measure- a feed-forward analyser for real time acid
a sample switching valve for this pur- ment loses some of its utility when there gas analysis has been seen with increasing
pose if the primary sample point is at is a TGTU and typically the emissions in regularity, particularly on new process pro-
the absorber (or quench) outlet. these instances are measured on a dry jects. The advanced burner control system
basis and corrected to a zero oxygen ABC+ for feed-forward control developed by

If there is only going to be one analyser base to arrive at standardised measure- Jacobs Comprimo Sulfur Solutions has
(most FEED process packages do this ment. While CO and NOx are occasionally seen a big increase in uptake in 2014.
in the interest of cost), AMETEK recom- included as a permit measurement they ABC+ uses real time process analysis of
mends locating the analyser at the outlet are not all that common. What is becom- the acid gas to provide information to the
of the absorber. Most importantly add the ing more common is the inclusion of closed-loop control.
COS measurement (in addition to the H2 un-oxidised sulphurs (primarily H2S with AMETEK have supplied nine analysers
and H2S) at this point as the cost is very some COS and CS2, 5-15 ppm total) to for Superclaus ABC+ projects and an addi-
minimal and of great value when trying to be included in the overall SO2 emissions tional 40 analysers for straight SRUs and
source the cause of a process upset and
emission increase. Additionally, if the FEED Fig 4: IPS-4 acid gas analyser (open)
provides for measurement at two points in
the TGTU then add the H2 measurement
to the second point as the cost is again
minimal and provides a back up for the pri- ultraviolet sample cell
mary H2 measurement which is by far the spectrophotometer for ultraviolet
more important variable. AMETEK has ~20
installations where there are two TGTU
analysers as part of the TGTU design most sample cell
infrared filter
of these being H2S measurement at both for infrared
photometer analysis
the quench and absorber. Approximately
half of these take advantage of adding a
second H2 measurement.
heated sample
Emissions monitoring analyser system
An emissions analyser is normally
required as part of the operating permit.
Beyond the compliance aspect, mass
emission (kg/h SO2) is a sensible and Source: AMETEK
useful way to evaluate a straight Claus

Sulphur 356 | January-February 2015 3


Fig 5: Capture of transient HC variation results, the requirement for more stringent
operating limits, hazard exposure and the
non-continuous nature of portable grab
100 0. 25 sampling are reasons to consider a perma-
90 nent solution. A fixed system that draws a
80 H 2S % 0. 20 continuous sample during the operational
70 transition period, without intervention from

60 0. 15 operations or analyser maintenance can

50 be easily implemented. The O2 measure-
40 0. 10 ment is based on tuneable diode laser
30 (TDL) which is non contact type of ana-
20 0. 05 lyser. The sample system is the advanced
HC %
10 sulphur removal (ASR) probe used on the
0 0 900 ADA tail gas analyser, the analyser
6/19/08 9:36 6/19/08 14:24 6/19/08 19:12 6/20/08 0:00 6/20/08 4:48 6/20/08 9:36 6/20/08 14:24 and sample system are maintained at
Source: AMETEK
time 150C so the analyser can be left on-line
as the analyser transitions from warm up
to acid gas mode.
SRU-TGTU applications. Figure 4 shows are records of an explosion occurring at
AMETEKs model IPS-4 acid gas analyser H2S values well below the 3.25% LEL. Industry survey
and Fig. 5 shows the analyser results for a AMETEK has been in discussion with The analytical instrumentation on a SRU
one time, sudden increase in total hydro- Black and Veatch on this subject. Based plant is a highly specialised field of such
carbons (HC) from 0.02 to 0.25%. on CFD modelling Black and Veatch arrive critical operational performance that
While an easy case can be made to at the conclusion H2S can exist in pock- knowledge and experience are essential
have an acid gas analyser for a gas pro- ets and stagnant areas at 16 times the to success. In a recent wide-ranging ana-
cessing SRU where variable composition mean value as measured in the sweep gas lyser survey, with questions relating to reli-
in the H2S content in the acid gas requires header meaning 2,000 ppm could consti- ability, on-line control and best practices,
feed forward control and the addition of tute a LEL condition somewhere in the pit. professionals from within the sulphur
total HC can also be utilised, there is also There is continuous analyser data to cor- recovery industry were surveyed in order
compelling reason to add a feed gas ana- roborate that where an explosion occurred to gauge the level of understanding of the
lyser for an SRU followed by a TGTU. The when there was a measured level of 4,000 needs of the analyser application and to
problem for a TGTU is not when there is a ppm H2S in the sweep gas header. Black obtain valuable user related experiences.
sudden increase in HC (resulting in a dra- and Veatch have asked if the sulphur pit Three distinct stakeholder groups were
matic rise in H2S going to the TGTU) but the (tank) can be measured at several points surveyed who are directly involved in the
resultant dramatic rise in SO2 going into and while this is not practical for several purchase and maintenance of SRU pro-
the TGTU when the HC event is over.Addi- reasons it does raise the question as to cess gas analysers:
tional work is being conducted to quantify what level of H2S constitute an alarm con- l front end engineering design and start
BTEX in amine acid gas for SRUs related to dition, to evaluate the gas flow in the head up engineers;
gas processing where these components space and for operations to fully under- l operations;
require co-firing of natural gas. stand what constitutes a hazardous condi- l end user analyser engineers and techni-
tion. Similarly, events of 3% SO2 in the pit cians.
Pit gas analyser sweep header have been recorded which
Hydrogen sulphide exists in sulphur as dis- clearly indicates a sulphur fire where the Reliability
solved H2S and chemically bound hydrogen condition prevailed for weeks without inter- The survey confirmed that the industry con-
polysulphides. The liquid sulphur produced vention, again measured values outside of siders tail gas analysers to be reliable.
from Claus SRU typically contains a total the norm must be heeded Trouble-shooting an analyser for sud-
of 200-350 ppm of dissolved H2S. Spon- den failure of an electronic problem is not
taneous or active degassing of the sul- Process O2 for start up / shut down difficult and a large refinery or gas plant
phur results in H2S accumulates in the This measurement is made during start-up complex has access to adequate skill
gas space above the liquid sulphur. H2S and shut-down, any time a sulphur recovery levels for this. Specialist help is required
becomes progressively more dangerous as unit (SRU) transitions from ambient tem- where the problem is intermittent or elec-
the levels incurred in handling and moving perature through natural gas warm up to tronic component failure is occurring on a
of the sulphur increases above toxic limits the introduction of acid gas. The measure- frequent basis.
(70 ppm), becoming lethal at 600 ppm and ment of O2 stoichiometry is critical in order The primary factor affecting reliability
reaching the lower explosive limit at ~3.25 to maintain a slight excess of oxygen. His- of the tail gas analyser is heat integrity at
%. A measurement tag for H2S and SO2 in torically operators have manually taken the process connection and the second
pit gas is sometimes required for certain samples using a portable electro-chemical leading cause is inadequate design of the
degassing licences or at locations where type O2 analyser or even mounting a sacri- analyser shelter.
there has been an explosion incident ficial analyser that is discarded after start Best practices for heat integrity include:
Where there has been an incident there up. While giving more or less satisfactory l The process nozzle should be 150mm

4 Sulphur 356 | January-February 2015


(6") or less, terminating in a 2" 150 lb domain of the FEED and EPC control engi- References
Improved Advanced Burner
raised face flange (maximum). neering team. 1. Simmonds
+ S., Cappe S. and Hauer R. (Ame-
l For top of the pipe type analysers the Reported analyser problems after start-
Control (ABC ) +
tek Process Instruments) and Cicerone D.

vendor must supply the steam jacketed up can often be related to the phenomenon (Cicerone
Since Jacobs&Comprimo
Associates): Sulphur
Sulfur recov-
ball valve. The steam needs to be of of the Claus catalyst preferentially absorb- (JCSS) first introduced ABCfront
ery on-line analysers from end design
+ in 2008, the
through life cycle ownership, Sulphur 2014,
medium pressure (nominal 7 to 10 barg ing H2S over SO2 with the result producing developments of this SRU burner control
Paris (Nov 2014).
/ 100 - 150 psig) as low pressure steam values that do not represent the true air concept and the increase in the number
is, in practice, often wet and unsuitable. control situation. The AMETEK guideline of implemented systems have been signifi-

l Use Contra-Trace or Contra-Heat to (provided by Sulphur Experts test crews) cant with three ABC+ references running,
overcome any shortcomings. Do not allow is that the analyser should only be used six more due for start-up and several more
the practice of wrapping steam trace tub- for control some eight hours after start- in different design phases.
ing and insulation around a problem and up, perhaps more depending on individual Following the success of ABC+ for feed
thereby masking that problem. plant conditions. gas analysis of amine acid gas, Jacobs
l For sample line analysers, if possible took the next step to include both amine
install the analyser above the sample Acid gas analysis feed-forward control acid gas (AAG) and sour water acid gas
point and in all cases request the ana- From the survey it was clear that all (SWAG) control.
lyser vendor confirm the length from respondents saw the benefits of feed- Sour water acid gas differs from other
on-site measurements and/or isomet- forward control but there were also many acid gases mainly because it contains
ric drawings. When sample lines are misconceptions that should be addressed. ammonia. Further, the hydrocarbon con-
estimated and then the numbers flow Firstly, respondents were unclear tent is often higher and more diverse than
through more than one set of hands, whether feed-forward control can be used in amine acid gas. Luckily, ammonia can
the probability the lines will be too long, in the cascade control loop but from field be rapidly measured by the combined UV/
is almost a certainty. experience this is exactly the intent. Rapid IR unit that is part of ABC+. Together with
compositional analysis of the hydrogen sul- the molecular weight determination (that
Best practices for analyser shelters phide, and a total hydrocarbon measure- comes as part of the ABC+ analytics), a
include: ment provide the air demand components robust control is offered.
l Provide an adequate shelter that with the addition of a carbon dioxide and
shades the analyser from direct sun water measurement (and in the case of Conventional SRU control and ABC
at all times. The shelter is no different sour water acid gas, ammonia) provide full
than what one would see over a local acid gas composition details. Both ABC and conventional SRU control
panel, quite inexpensive and almost Secondly, there was clear concern as manage feed gas flow rate variations via
always neglected at the EPC detailed to whether the information is fast enough feed-forward ratio control, but not in exactly
design stage. for feed forward control, but again this is the same way. Conventional SRU control
l Involve the analyser vendor at an early exactly the intent of these control designs. uses the main and trim air valves fully inde-
stage preferably at frontend engineer- With the use of real time spectroscopic pendently; the main air valve is controlled
ing design. As end-user insist upon technology (a combination of both ultra- in direct ratio to the SRU feed gas flow,
it and continue the dialogue into the violet and infrared) the system is able to while the trim air valve is only controlled
detailed design stage. Be very wary of provide useful data rapidly. by the tail gas quality controller. The down
surrendering all analyser design to a sides of this strategy, such as slow and
systems integrator. Survey conclusions inaccurate feed gas ratio control via the
The key conclusions of the survey were: main air valve and limited tail gas quality
Sample point selection for both tail gas l For large scale project where a systems control capacity via the trim air valve, are
and feed gas analysers is more often than integrator is involved, ensure close overcome by the different strategy which
not compromised at the piping engineering management of the project from field ABC applies. ABC controls the main and
stage. It is the lack of analyser specialists proven end-user analyser technicians, trim air valves simultaneously and utilises
at the detailed engineering stage that is in combination with close communica- each valves specific strength in the best
the root cause. A constricted and compro- tion to the key SRU analyser vendor. manner; it provides rapid and accurate
mised sample point location can affect l Ensure that the SRU analyser vendor is total air response to any change in sour
analyser operability which in itself is bad involved in the entire project process gas feed flow rate combined with enough
enough. The problem can extend to safety from the early stages of the front end tail gas quality control capacity. While
concerns at HazOp review, requiring an engineering design (FEED) right through the total air flow controller manipulates
expensive piping change to make it right. the detailed design phase. the small (thus rapid) and accurate trim
l A successful SRU analyser start-up air valve in response to any air demand
Start-up and turndown requires careful thought and planning. change, the main air valve is continuously
Slow response is sometimes attributed to The end-user technicians (or alterna- adjusted to ensure that the trim air valve
the analyser at turndown of the SRU when tively the contractor analyser techni- stays within its operating capacity. Shortly
in fact it is process lag time. Although cians) need to be factory-trained and after that response has been provided, the
asked frequently to help configure control authorised to work on the respective trim air valve returns to its optimal position
loops this is typically not a core compe- analysers, alternatively the analyser to wait for new air demand changes.
tency of analyser companies and more the vendor must be at site for start-up. If no measures are taken, as in conven-

Sulphur 356 | January-February 2015 5