4000 mtpd Ammonia plant based on

proven technology
Today’s state of the art capacity of a world scale ammonia plant is about 2000 mtpd. Driven by
economies of scale the first of the next generation of large scale plants – based on Uhde license with
a capacity of 3300 mtpd – is currently under construction for SAFCO in Al Jubail, Kingdom of Saudi
Arabia. In advance of a continued trend towards even larger plants, Uhde has already checked this
concept for capacities of 4000 mtpd and beyond and found it fully viable. Some details are discussed
here.

Joachim Rüther*, John Larsen**, Dennis Lippmann*, Detlev Claes*

*
Uhde GmbH, Dortmund, Germany
**
Uhde Corporation of America, Houston, Texas

2000 P.T. Kaltim Pasifik Amonik (Bontang, Indonesia)

Introduction 2000 mtpd - Haldor Topsoe license
2005 Burrup Fertiliser (Burrup, Australia)
2200 mtpd - KBR license
Brief capacity history of world scale ammonia 2006 SAFCO (Al-Jubail, KSA)
plants 3300 mtpd - Uhde license

A lready the first commercial scale ammonia

plant – built by Uhde in 1928 – put up a
capacity of 100 mtpd. This number was
not to be significantly exceeded until turbo compressors
were introduced into ammonia processing and plant ca-
From a more generalized point of view three market
trends can be distinguished, each one aiming at produc-
tion cost savings:
• Improvement of plant energy efficiency al-
pacities rapidly increased to about 1000 mtpd in mid to ready reached an optimum in the 1990s (e.g.
late 1960’s. From those days on a continuous rise of BASF, Antwerp, Belgium).
world scale plant capacities occurred and is still going • Relocation towards low cost natural gas sites
on. The following plants may be recognized as mile- is in full progress. Currently almost no plants
stones of the recent history: are being built in high cost gas areas.
• Progressive cost reduction by plant capacity
1988 Hydro Agri, Sluiskil E (Sluiskil, the Netherlands) scale up (economy of scale) is expected to be
1750 mtpd - Braun license the future trend. The above mentioned
1991 BASF (Antwerp, Belgium) 3300 mtpd single train plant – currently under
1800 mtpd - Uhde license construction and to be commissioned in 2006 –
illustrates this.

AMMONIA TECHNICAL MANUAL 312 2005

Process concepts for the next generation single train, which is currently under construction by
Uhde for SAFCO in Al-Jubail, KSA.
The above mentioned trend towards large scale
ammonia plants is recognized by almost all major licen- Furthermore, there are other good reasons to stay with a
sors with each one having its own process concept for conventional, externally heated front-end layout.
the next generation ammonia plant. The most popular • superior hydrogen yield
concepts are summarized below: • less duty for the CO2 removal unit
• No need for an external air separation unit or
Oxygen-fired ATR process integrated cold box technology. Add-
Recently a front-end concept originating from heavy ing an air separation unit adds cost, either capi-
fuel oil gasification with two stage HT shift and two tal investment or operating cost, if oxygen is
stage CO2 removal (Rectisol & liquid nitrogen) has supplied over the fence.
been presented to operate with an oxygen-fired auto-
thermal reformer at high pressure and natural gas as
feedstock. The process is proposed for capacities
around 4000 mtpd. Up to now, no reference plant has
been built, thus the operating conditions of the ATR are
non-proven.

Excess air blown ATR in combination with heat

exchange reforming
This approach is based on an excess air blown ATR and
a heat exchange reformer, both of which are fed with
fresh feed/steam mixture in parallel. The excess nitro-
gen is removed in a cold box. The synthesis loop oper-
Fig. 1: Uhde Dual Pressure Process - once through
ates at low pressure and uses a ruthenium based cata-
reactor in between the synthesis gas compressor
lyst. There are references for each of the characteristic
casings
process steps, however, especially for the heat ex-
change reformer the maximum capacity that has been
Recent design experience from the world’s
built is much lower than that expected for a next gen-
largest ammonia plant
eration plant (i.e. about 4000 mtpd).
Uhde has already gained experience from the de-
Enriched air-fired ATR sign and construction of a next generation ammonia
An intermediate option between the oxygen-fired ATR plant, the above mentioned SAFCO IV plant. This re-
and the excess air blown ATR is the oxygen-enriched sults in a significantly reduced scale-up factor and
air fired ATR concept. Conventional downstream gas therefore means considerably improved ‘bankability’ to
processing may be employed, however, the extra duty the customer. After commissioning of SAFCO IV in
for the CO2 removal due to the autothermal heat supply 2006 Uhde will have the full range of experience neces-
should be considered as in the other autothermal front- sary to realize such projects in a highly efficient way.
ends. These financial considerations are mainly based on
equipment and machinery issues. However, there are a
Uhde Dual Pressure Process number of minor issues, which in combination can put
In contrast to the above mentioned concepts the Uhde time schedule, budget or plant quality at risk. These as-
Dual Pressure Process [1] focuses on the debottleneck- pects are to be discussed later on.
ing of the conventional synthesis loop instead of modi-
fying the front-end process, which is not considered to
be critical if based on Uhde technology. By insertion of
a once through synthesis reactor at an intermediate
pressure level the production capacity can be raised by
about 65% still using proven equipment. The concept is
the key to the impressive capacity of 3300 mtpd in a

2005 313 AMMONIA TECHNICAL MANUAL

Uhde Dual Pressure Process assessment
for 4250 mtpd

Overview

Based on the above mentioned experience, Uhde

carried out a detailed study to validate the SAFCO IV
process concept for capacities around 4000 mtpd. Any
possible critical items such as turbo compressors, large
liquid pumps and static equipment as well as piping and
valves, have been thoroughly examined. 4250 mtpd was
chosen as the target capacity.

Static equipment Fig. 2: Primary and secondary reformer acc. to

Uhde design (top fired radiant section, cold outlet
Primary reformer manifold, central riser pipe, ring-shaped arch)
Due to its highly modular design a primary reformer
can generally be scaled up quite easily since the design Secondary reformer
of each reformer tube and burner group can remain un- The secondary reformer will slightly exceed the dimen-
changed. However, at least the manifold system has to sions of SAFCO IV. However, much larger autothermal
be enlarged and therefore to be checked. reformers have been built, which operate under even
A primary reformer for 4250 mtpd is expected to more severe conditions. In addition it has to be kept in
consist of about 540 tubes (5 inch in diameter). Uhde’s mind that the pressure retaining wall is reliably kept
cold outlet manifold system has already been applied to cool by means of a refractory lining and a water jacket,
top fired reformers with up to 960 tubes. The following even if the inside temperature is very hot. Furthermore
list gives the basic data of some large primary reform- the span of the refractory arch of an Uhde secondary re-
ers: former is only about half that of other designs due to
the ring-shape design around the central riser pipe (see
plant prim. reformer dimensions No. of fig. 2, also).
L x W x H 1 [m x m x m] tubes
QAFCO 4 plant secondary reformer dimensions
18.1x13.7x12.2 288
2000 mtpd Ø [m] H [m]
SAFCO IV QAFCO 4
19.1x17.9x13 408 4.5 18
3300 mtpd 2000 mtpd
plant study SAFCO IV
20.2x22.1x13 540 5 20
4250 mtpd 3300 mtpd
Qafac plant study
14x49x12.6 960 5.7 22
(Methanol) 4250 mtpd
1
H: heated length of reformer tubes
Reformed gas waste heat boiler
Convection bank For technical and economic reasons the maximum ca-
For the convection bank a design with horizontal tubes pacity of a single reformed gas waste heat boiler is
has been chosen. This is well proven in refinery service. about 3800 mtpd, thus a dual flow design has to be cho-
Any single coil is designed according to API and sen for larger plants. For a 4000 mtpd plant one of these
ASME code respectively. The dimensions will be about boilers will be of the same design as the one installed in
22m x 25m x 13m. QAFCO 4, Qatar. The design of such a boiler system
for 4250 mtpd is of course well within the bounds of
feasibility as can be seen from the 3300 mtpd single
flow design of SAFCO IV. Concerning the dual flow
design, the change from a single waste heat boiler to

AMMONIA TECHNICAL MANUAL 314 2005

two parallel waste heat boilers may introduce problems Ammonia converters
due to flow mal-distribution resulting in changes in The Uhde Dual Pressure Process (which is implemented
temperatures and piping stresses. The design incorpo- in SAFCO IV, but not in QAFCO 4) makes it possible
rates two inlet nozzles on the steam superheater follow- to use almost unmodified equipment dimensions while
ing the waste heat boilers, and a common steam drum. raising the synthesis capacity by a factor of 1.65. For
The technical risk of applying parallel boilers is consid- instance, the diameters of different converters are given
ered to be relatively low since it can be mitigated by a in the table below. The additional once through con-
conservative design of the piping systems connections verter indeed tends to be a large item, but is far smaller
of these waste heat boilers. Furthermore, the application in diameter and volume than the low pressure loop con-
of parallel waste heat boilers downstream of the secon- verter designed by Uhde for CIL, Canada. This is still
dary reformer is standard design in other processes, and true at a production capacity of 4250 mtpd. The operat-
another process features a single steam drum for several ing pressure of the once through converter has already
boilers from different plant sections. So, the risk is as- been applied in the synthesis loop of the Puyang plant.
sessed to be acceptable and Uhde intends to employ The dimensions of the 4250 mtpd loop converters have
two parallel waste heat boilers receiving hot gas from to be slightly prorated. However, here also a significant
the secondary reformer. technological risk is not to be expected.

CO2 removal plant converter op. press. Ø

Using BASF’s aMDEA process for CO2 removal in a [bar] [m]
4250 mtpd plant admittedly results in impressive CIL 1128 mtpd loop conv. 87.5 5.45
equipment dimensions. A conventional process layout Zhong Yuan Chem.
results in absorber dimensions of Ø 6.9 m (top: 4.6 m) x loop conv. 110 3.2
Fert. 1000 mtpd
50 m. The diameters of HP and LP flash vessels are QAFCO 4 loop conv. I 207 3.01
calculated to be 6.7 m and 8.7 m respectively. 2000 mtpd loop conv. II 203 2.86
However, the aMDEA process has many references OT converter1 110 3.21
and there are no concerns from the process point of SAFCO IV
loop conv. I 207 3.01
view – acid gas removal units of similar dimensions are 3300 mtpd
loop conv. II 203.4 2.86
currently being specified and built for LNG plants. An- plant study OT converter1 110 3.4
other settling argument may be, that the step from 4250 mtpd loop conv. I 207 3.0
SAFCO IV to 4000 mtpd corresponds to an increase of loop conv. II 203.4 3.1
13 % in diameter (see table below, also). Elsewhere ca- 1
Converter of the once through synthesis, see fig. 1
pacity scale-up factors of about 2, which correspond to
a factor of 1.4 in diameter, are still considered to be Waste heat boilers (Synthesis Loop)
conservative. The critical design parameter of the synthesis gas waste
heat boilers is the maximum tube sheet temperature,
plant absorber HP flash LP flash which corresponds to tube sheet thickness and thus to
Ø x cyl. H [m²] Ø [m] Ø [m] diameter and capacity. For the desired capacity of
QAFCO 4 4250 mtpd the temperature of the tube sheet core can
5.0 (3.3) x 43.9 4.9 6.1
2000 mtpd reliably be kept away from the critical limit where ni-
SAFCO IV triding and embrittlement begins.
6.1 (4.1) x 37.3 5.6 7.6
3300 mtpd
plant study Rotating equipment
6.9 (4.6) x 38 6.7 8.7
4250 mtpd
Natural gas compressor
Nevertheless, the feasibility of the logistics for such The natural gas compressor is in general not a critical
equipment closely depends on the plant location. The component of a typical ammonia plant. Some plants
transport to sites without easy access to a waterway is at even operate without such a machine. However, in the
least considered to be difficult. However, due to the ex- common case where the natural gas pressure at battery
port orientation of these large plants, a site location limit is not sufficiently high, it is still needed. Then its
close to the sea is to be expected. operating conditions are generally similar to those of an
inline compressor of a natural gas pipeline. Since these

2005 315 AMMONIA TECHNICAL MANUAL

machines are sometimes very large, they can be taken driven integrally geared compressor, it is remarkable
as a reference for the natural gas compressor of a that some units with operation conditions close to those
4250 mtpd ammonia plant. This statement is also sup- needed here are currently under construction.
ported by design studies of different manufacturers. The technology exists today for the construction of
Availability of large capacity natural gas compressors is a process air compressor for the conventional process
not an issue. and there is a choice of different solutions. Most of the
compressors in similar service are found in air separa-
plant impellers power speed tion, terephthalic acid and acetic acid plants. Lately, the
[MW] [min-1] size of air separation plants has increased and reference
QAFCO 4 compressors exist for the conventional ammonia proc-
3/3 3.466 14400 ess. Acetic acid plants operate at similar pressures as an
2000 mtpd
SAFCO IV ammonia plant front-end but with somewhat lower flow
6 4.342 10102 requirements than that required for the large scale
3300 mtpd
plant study plants. Which compressor type is eventually chosen
6 5.400 10500 will depend to a large degree on the input from the cli-
4250 mtpd
ent. The plant arrangement work done so far on the
Process air compressor 4250 mtpd plant has been based on an in-line solution
The process air compressor may be one piece of equip- with 2 or 3 casings.
ment which cannot simply be scaled up, since the LP
casing of the conventional 2-casing machine would be- Synthesis gas compressor
come a really heavy item. Nevertheless there is still a The operating conditions of the synthesis gas compres-
reference for the casing size, but not for the desired sor are highly specific to ammonia plants. In no other
flow. This constraint is even more critical for processes application does a compressor set have to cope with a
which use excess air or autothermal reforming in the similar combination of flow, molecular weight and dis-
front-end because of the increased air flow required. charge pressure. Consequently, there is no design for a
However, there are several choices for the required synthesis gas compressor for 4250 mtpd with a com-
high flow air compressor. Taking SAFCO IV as the plete reference. However, some of the largest synthesis
starting point, the most obvious and straightforward so- gas compressor sets ever built operate in recent Uhde
lution may be a 3-casing compressor– with LP casings plants and from SAFCO IV (3300 mtpd) to a 4250
as used for SAFCO IV, i.e. once this is operational mtpd plant is just a reasonably small step. Additionally,
there will be a complete reference. On the other hand, the synthesis gas compressor duty of the Uhde Dual
there are still manufacturers who have the appropriate Pressure Process will be far smaller than with a conven-
technology to construct a reliable 2-casing compressor, tional synthesis concept. For instance, the compressor
although they do not have references for the required trains of QAFCO 4 (2000 mtpd, conventional synthesis
capacity. The 3-casing solution may be the least risky, loop) and SAFCO IV (3300 mtpd, Uhde dual pressure
but it seems to be technically feasible to use a 2-casing process) are very similar, since the synthesis gas vol-
compressor with the benefit of fewer parts resulting in ume flow to the high pressure loop is significantly re-
possible cost savings. The data for 4250 mtpd given be- duced by the ammonia synthesis and separation in-
low are for the 2-casing option: between the compressor casings.
Detailed technical studies in close cooperation with
proven manufacturers resulted in feasible concepts for
plant impellers power speed
the compressor and the associated steam turbine, too.
[MW] [min-1]
For reference reasons a dual flow steam turbine may be
QAFCO 4
2/2//2/4 14.923 6440/12119 the first choice. However, a single flow solution is also
2000 mtpd
available.
SAFCO IV
2/2//2/4 25.344 5135/8672 Regarding the compressor casing size and the num-
3300 mtpd
ber of impellers, there are references for the specified
4250 mtpd
2/2//3/3 31.000 4500/9000 compressor in similar applications. Synthesis gas com-
(2 casings)
pressors of comparable dimensions are operated in
methanol plants and are under construction for ammo-
An integrally geared compressor is another option. nia service.
Even though there is no reference for a steam turbine

AMMONIA TECHNICAL MANUAL 316 2005

 neck-flanges) and the largest used nominal diameters of
plant impellers power speed different plants broken down by pressure ratings.
[MW] [min-1]
QAFCO 4 DN 150 # 300 # 600 # 900 # 1500 # 2500 #
5/4//8/1 27.331 9535
2000 mtpd
...
SAFCO IV 6"
4/4//6/1 28.600 9701
3300 mtpd 8"
plant study 10"
4/4//7/1 38.000 9000
4250 mtpd 12"
14" Qafco IV
It may be correct that the eventual limit for ammo- 16" 2000 mtpd
18"
nia plant capacity will be based on machinery issues
20"
[2], but at 4250 mtpd satisfactory compressor and tur- 22"
bine solutions for the synthesis gas compressor train are 24"
available when applying the dual pressure process. 26"
28"
Refrigeration compressor 30"
Refrigeration plants making use of ammonia as the re- 32" Safco IV
frigerant are widely used. Additionally, the operating 34" 3300 mtpd
36"
conditions of a refrigeration compressor are not that
38"
demanding, thus a scale up from 3300 to 4250 mtpd is 40"
4250 mtpd
not expected to be critical. This is also reflected by the 42"
results of studies by different manufacturers. Ammonia 44"
refrigeration compressors for these capacities are avail- 46"
able from a number of vendors. 48"
52"
Large pumps 56" standardized off standard
60"
Capacity limitations of pumps are generally not that
critical, since liquid pumps are often likely to be dou- ASME WN-Flanges
bled for reliability reasons, which can also be done for
capacity scale up. For example, this has already been It can clearly be seen, that the front-end piping (600#,
done for the semi-lean solution pump of the CO2 re- sometimes 900#) for a 4250 mtpd plant is well within
moval unit, which was realized as a 1 of 2 arrangement the limits of the ASME code. This is still true for main
for QAFCO 4 (2000 mtpd) and as a 2 of 3 system for parts of the Uhde Dual Pressure Process synthesis
SAFCO IV (3300 mtpd). Consequently, a combination piping (1500#), however, a conventional loop would be
of the pump type of QAFCO 4 and the SAFCO IV ar- limited to below 3000 mtpd. Some hot high pressure
rangement will be well suited to the flow requirements lines (2500#) are already off standard at 2000 mtpd. In
of a 4000 mtpd plant and will have sufficient refer- detail this is the piping from the gas/gas heat exchanger
ences. For 4250 mtpd only a minor scale up has to be to ammonia converters and back to the gas/gas heat
done. For lean solution pumping or boiler feed water exchanger. However, this is obviously a problem that is
service, as well, no parallel pumps will be needed. under control for conventional capacities and is not
expected to be critical at capacities discussed here.
Piping and valves Furthermore it should be kept in mind, that the
stiffness of a pipeline depends on its diameter to the
As already stated in the introduction, the synthesis fourth power, thus additional allowances have to be
loop has to be considered as the main bottleneck of the made for expansion loops in order not to exceed admis-
conventional process scheme. Besides several equip- sible nozzle loads. A conservative prorating is of great
ment items already discussed above, the availability of importance here, since unforeseen piping stresses can
appropriate piping material is another critical point . lead to severe difficulties during detailed engineering
The table given below shows the maximum standard- when the plot plan has to be changed.
ized nominal diameter of piping material (here: weld

2005 317 AMMONIA TECHNICAL MANUAL

 Several control valves, which from SAFCO IV ex- turbine drives for pumps and fans, the pipe rack dimen-
perience are known to be demanding, have also been sions will be smaller.
checked in cooperation with manufacturers. A solution
was found for every single one of these items.
Besides these major issues, there are also some mi-
nor points, which need to be considered during engi-
neering and procurement of the plant:
• With increasing line sizes operating controls
like hand-wheels move out of the operators’
reach. To ensure plant operability, additional
platforms and stagings have to be provided.
However, this equipment may in some cases
hinder the accessibility of other equipment.
This has to be born in mind when considering
plot space, not only in the detail engineering
phase. pipe rack rel. dimensions at 4250 mtpd
• For some valves a simple scale up may be un- (base case 3300 mtpd)
feasible – a change of the valve type will then section level height width
be required. However, the new type may be 2-2 1 111%
heavier by a factor of 5, for example. This ··· 2 115% 120%
should already have been considered for in- 7-7 3 118% (7-7: 100%)
stance for structure loads. 1 114%
• For accuracy, flow measurements typically re- 8-8 2 117% 100%
quire a certain diameter-related straight length 3 118%
in inlet and outlet piping – often not crucial for note: 1-1 to urea synth. (optional, dep. on urea capacity)
smaller line sizes. However, enlarging the di- 7-7 to main substation / cooling water system
ameter of the piping may result in unexpected 8-8 to offsites/utilities
problems, since expansion loops will also take
more space. The arrangement of turbo compressors and accom-
Uhde has the process concept and – based on panying condensers and intercoolers has also been
SAFCO IV – the experience to avoid such problems checked. Even with a 3-casing process air compressor
and to design a high quality plant with high reliability the compressor house of the 4250 mtpd plant has just
and operability. 20% more interior space than the 3300 mtpd plant.
The required plot space may in some cases only be a
minor cost issue, however, where space is limited, the
Plant arrangement
plant dimensions can easily become decisive for the
From a plant arrangement point of view, the follow- whole project. The comparison of ammonia plant di-
ing points need consideration: mensions below clearly shows the advantages of a sin-
• pipe rack dimensions gle train plant with respect to this point. Incidentally,
• arrangement of large turbo-compressors and the SAFCO IV project is one of these, where plot space
accompanying condensers (compressor house is tight. The plant arrangement data given below are
dimensions) based on the SAFCO IV plant layout. Therefore a tai-
• required plot area (overall plant arrangement lor-made design for a concrete project may deviate
plan) from this data.

As shown in detail below, the pipe rack dimensions

of a 4250 mtpd plant are considerable, of course. How-
ever, even in the case of numerous redundant drive sys-
tems with electric motor and steam turbine in parallel,
as discussed here, the dimensions remain feasible. For a
more common driver concept, which does not utilize

AMMONIA TECHNICAL MANUAL 318 2005

plant plot width plot length area as shown above, there is no need to leave the safe evo-
[m] [m] [m²] lutionary path of development. Based on the Uhde Dual
QAFCO 4 Pressure Process the technological achievements of the
95 222 21090 past can be projected forward, thus combining the ad-
2000 mtpd
SAFCO IV vantages of progressive cost reduction and proven tech-
105 222 23310 nology and maintaining the safety and reliability of pre-
3300 mtpd
plant study vious plants at 4000 mtpd and above.
120 270 32400
4250 mtpd
(2x2000 mtpd) (42180)
Conclusion
The current trend to larger scale ammonia plants
Economic evaluation with capacities in the range of 4000 mtpd is obvious
It is in the nature of chemical plant construction and well-founded in plant economics. The technical
that a detailed cost estimation needs to be based on sev- feasibility of such a plant – concerning static and rotat-
eral very project specific pieces of information and ing equipment as well as piping and arrangement – was
therefore cannot be done here. However, based on ac- found fully viable. This is even true for turbo-
tual data taken from executed contracts and the SAFCO compressors, which formerly have been regarded as the
4 project, an indication can be given on how a capacity limiting equipment. Considering economics, the ex-
scale up effects cost. pected economy of scale was confirmed. Furthermore,
Starting with capital expenditure (CAPEX) and set- the 4000 mtpd capacity can be reached with Uhde tech-
ting the specific cost per tonne (i.e. cost / capacity) of a nology on the basis of long-term experience and a new
2000 mtpd plant to 100% the specific cost of a single 3300 mtpd reference plant.
train 4000 mtpd plant will be around 86%. This figure Uhde is convinced that large scale plants using the
corresponds to a cost degression exponent of 0.78. Dual Pressure Process at present constitute the best
Comparing with the degression exponent reached by trade-off between plant safety and economic risk on the
smaller plants it can be seen that some equipment has to one hand and economic benefit by economy of scale on
be paralleled (see reformed gas waste heat boiler), or the other.
may need some modification to make a scale up feasi-
ble. On the other hand the relatively small plot space References
will result in further cost reduction on owner’s side.
Concerning operating costs (OPEX) it can be [1] D. Lippmann, J. Larsen: Expanding Ammonia
stated, that compared to conventional synthesis tech- Plant Capacity Limits with Proven Technology.
nology the higher energy efficiency of the Uhde Dual Nitrogen Conference 2002, Doha, Qatar, 2002
Pressure Process reduces operating costs by about 4%.
Considering a capacity of 4000 mtpd this corresponds [2] S.E. Nielsen: Ammonia Plant Capacity Considera-
to about 1.6 million US dollars/year. Further savings tions. 46th Annual Safety in Ammonia Plants and
on owner’s side may result from reductions of insur- Related Facilities Symposium, Montreal, Canada,
ance fees, maintenance cost and personnel cost. 2001

Safety
Finally making reference to the focus of the sympo-
sium – safety in ammonia plants – it is obvious that the
present high plant safety level is a result of prolonged
evolution of equipment and safety measures. The intro-
duction of new equipment and especially new process
schemes involves new risks and has to be done ex-
tremely carefully. In this context, a scale up factor of 2
and above as well as referencing a new process scheme
on a component basis seems to be a high risk. However,

2005 319 AMMONIA TECHNICAL MANUAL