Hydrogen has always played an important role in oil refining, but refiners today are
finding that it is one of the most critical challenges facing
them as they plan production of clean
fuels. In addition, hydrogen management practices
significantly impact operating costs, refinery margin and
Therefore, an effective hydrogen management program must
address refinery-wide issues in a systematic, comprehensive
way. Managing hydrogen more effectively has been found to
improve refinery profitability by millions
of dollars a year, and often enables the refiner to avoid the
capital cost of new hydrogen production.
The hydrogen system consists of hydrogen producers, hydrogen
purification, hydrogen consumtand the distribution network
itself (Fig. 1). Tools and techniques are available to manage
each of these individual hydrogen network components.
However, if you consider the refinery as a whole, instead of
individual process units, much greater opportunity exists to
impact the refinery profit. The key to unlocking significant
profit improvement opportunities is to focus on hydrogen's
effect on the performance of hydroprocessing units - and
therefore gross margin.
Presented here are various components of the hydrogen
network and effective strategies for improving the network,
from simple operating adjustments to capital improvements. A
methodology for analyzing the hydrogen network as a whole is
described, and guidelines are provided to manage the network
effectively. Also discussed are hydrogen's role in refinery LP
models and some approaches for determining refinery hydrogen's
The primary sources of hydrogen in a refinery are catalytic
reformers, hydrogen plants and purchased hydrogen.
Catalytic reformers - operations. Operating
conditions of the catalytic reformers (rates and severities)
are typically set by overall refinery economics (the gasoline
pool) rather than the need for hydrogen. Hydrogen yields are
primarily a function of the properties of the feed naphtha,
severity, catalyst and operating pressure.1,2,3 If
the operating conditions are set by the Planning Department
based on refinery-wide economics, there is little opportunity
to improve hydrogen production through operating
Hydrogen plants - operations. Hydrogen
plants produce hydrogen primarily through the steam reforming
and water gas shift reactions:
CH4 + H2O 1
3H2 + CO (methane feed)
CO + H2O 1 H2 +
The reforming reaction is endothermic and equilibrium
limited. Lower pressure and higher temperature favor higher
conversion to hydrogen. Operating pressure is typically set by
a practical hydrogen delivery pressure. Maximum temperature is
constrained primarily by tube life and coking concerns. The steam-to-carbon ratio is a critical operating
variable that affects conversion and coking.
Optimum operation is unique to each hydrogen plant because
the constraints in each unit will be unique. If the refiner's
goal is to minimize the per-unit cost of hydrogen rather than
maximizing production, there will be a different optimum
temperature and steam-to-carbon ratio. Since these optimum
setpoints can change daily, as a function of rates and feed
compositions, the operator should have the tools to optimize
the reformer accordingly.4,5 All the skills required
to operate a furnace efficiently and safely are critically
important for hydrogen plant operations.
Increasing hydrogen production. In a
catalytic reformer, a number of methods are available to
increase hydrogen production. Obviously, hydrogen production
may be increased by modifying equipment to enable increased
charge rate. Also, hydrogen yields can be improved by: changing
the naphtha feed to one more favorable for hydrogen production;
decreasing pressure; or replacing the catalyst charge with one
that provides a higher hydrogen yield.6,7 Large
increases in hydrogen production can be achieved through
pressure reduction by converting from fixed bed to continuous
catalytic regeneration mode.
A low-cost (capital and operating) catalyst regeneration
process has been developed for the majority of fixed bed
reforming and stacked reactor reforming units to maximize
hydrogen yields. This type of project can be quite attractive if
the alternative is building a new hydrogen plant.
For hydrogen plants, a number of approaches exist to revamp
for higher capacity. Increases of up to 25% are common.
Debottlenecking may be achieved by mechanical modifications to
remove equipment constraints, adding pre-reforming or adding
Over-the-fence hydrogen supply. The
make-or-buy decision is a relatively straightforward economic
calculation. Non-economic issues such as reliability, long-term commitments
and operating coordination need to be considered. The option to
cogenerate electricity will also affect the economics.
An additional consideration is the availability of any
hydrogen-containing streams currently going to fuel. These may
be valuable to the hydrogen supplier either for hydrogen
recovery (purification) or as feed to a hydrogen plant. If the
refiner chooses not to recover this hydrogen himself, these
streams can potentially be sold to the hydrogen supplier.
Hydrogen recovery. Look for
hydrogen-containing streams, such as hydrotreater offgases or
"excess" hydrogen streams that are currently being sent to fuel
gas or hydrogen plant feed. The cost of hydrogen recovery can
be as low as 50% of the production cost. Whereas the cost of
generation is usually fixed for a certain location (primarily a
function of energy cost), the cost of hydrogen recovery depends
on many factors, such as fuel gas pressure and composition of
the hydrogen-containing stream. In general, hydrogen recovery
should be pursued for all streams with favorable economics. In
some rare situations, additional constraints such as
minimum/maximum hydrogen content of the fuel system must be
Potential for debottlenecking purification
units. If the additional purification needed is less
than approximately 30% of the capacity of an existing hydrogen
purification unit, a revamp is often the most cost-effective
way to obtain the extra capacity. In many cases, it will be
required to sacrifice either hydrogen purity or recovery
against the increased capacity. Pressure drop over the unit may
increase significantly and must be considered. Pre- and
post-treatment equipment may also have to be revamped to cope
with increased flowrates. Often more plot space will be
Selecting purification technology. The three most
commonly used processes for hydrogen purification are pressure
swing adsorption (PSA), semipermeable membranes and cryogenic
separation (cold box). These processes have been described in
detail elsewhere.10 Important parameters to consider
when selecting the hydrogen purification technology are:
- Minimum purity of the hydrogen product. If a hydrogen
product with a purity of 99 vol% or higher is required, PSA
is the obvious choice. PSA is the only process capable of
producing a hydrogen stream containing less than 10-ppmv
carbon oxides, a typical requirement for catalytic processes.
Membranes and cold boxes can produce hydrogen in excess of
95-vol% hydrogen content.
- Pressure drop between feed gas and hydrogen. Effective
membrane separation requires that the feed stream is
available at a pressure two to five times the pressure of the
permeate product (this is the driving force needed to let
hydrogen permeate preferentially over other components). If
this pressure ratio is not available, additional compression
equipment is required. PSA and cold boxes have only minimal
pressure losses between feed and hydrogen product.
- Possibility of byproduct recovery. A cryogenic unit has
the ability to deliver separate hydrocarbon byproduct streams. The
offgas from membranes contains all the nonhydrogen components
in a single stream close to feed pressure, so further
separation is possible. Offgas from a PSA unit is at low
pressure, making by-product recovery not cost-effective.
- Delta cost between hydrogen product and feed gas. The
larger the difference in value between hydrogen and fuel, the
more capital can be expended on purification equipment. Each
unit will have its own trade-off curve between cost (capital
and operating) and recovery (ratio of hydrogen contained in
the product stream over hydrogen contained in the feed
stream). See section, "How much is hydrogen worth?," p.
- Capacity. Membranes are usually the lowest cost option
for low product rates, but are rarely cost-effective for
larger capacities, because the number of membrane elements is
proportional to the hydrogen product rate. PSA units are
built to produce from 5 to 100 MMscfd of hydrogen and more.
The cost of a PSA unit for a small capacity is higher than
that of a membrane unit, but PSAs have better economies of
scale. Cryogenic separation is also expensive for small
capacities, but has excellent economies of scale.
Purification design decisions. Once the
technology has been selected based on the criteria mentioned
above, there are further considerations. If a PSA unit is the
technology of choice, the main question is: what tail gas
pressure will the PSA operate at? The tail gas pressure is the
pressure at which the adsorbent is regenerated. Lower tail gas
pressure means higher recovery and capacity, but tail gas
compression may be needed to operate at very low pressures. In
most cases, the tail gas is discharged to the refinery fuel gas
header, unless a dedicated low-pressure burner is available.
The impact of high tail gas pressure on hydrogen recovery may
be as much as 20% (Fig. 2).
Effect of PSA
tail gas pressure at a typical feed gas
The PSA unit produces high-purity hydrogen. Operating the
PSA at lower purity will have a minimal effect on recovery. If
lower purities are acceptable during certain periods, it is
better to install a feed-to-product bypass around the PSA unit.
Blend a fraction of the feed gas in the hydrogen product, if
you can tolerate feed components in the hydrogen product. In
this way, overall hydrogen recovery can be maximized at the
lowest acceptable purity.
Although H2S does not present a problem for the
PSA adsorbent, it is recommended that H2S be removed
upstream of the PSA unit. The PSA unit will concentrate all
H2S in the tail gas stream, which is compressed and
sent to a fuel gas system. Besides environmental considerations, safety
and metallurgy must be considered.
Membrane fibers are polymeric and can be damaged by
aromatics condensation. During the separation process, hydrogen
is preferentially drawn from the feed gas so the residue gas
enriches in hydrocarbons. Heat the membrane feed gas above its
dewpoint to avoid condensation during operation.
The fuel gas pressure of a cryogenic system determines the
maximum achievable purity of the hydrogen product, and also
fixes the purity/recovery relation. Again, a trade-off is to be
made between additional compression of the tail gas stream
going to fuel and increased hydrogen recovery. At typical
operating pressures, a 10-psi increase in tail gas pressure can
reduce recovery by 1%. Separate hydrocarbon streams with
different boiling point ranges can be obtained by adding
separators. A methane wash column is required if low boiling
components such as CO or nitrogen are to be
Designing for compressor flexibility provides greater
flexibility in utilizing purified streams. Modifications
requiring compressor flexibility include changing the makeup
composition (such as purifying recycle gas with a membrane
purifier and dropping the permeate to makeup compressor
suction) and the recycle purity (handling the change in recycle
gas molecular weight through the recycle compressor).
Revamping existing purification units.
After construction and startup, refinery
objectives and hydrogen balances often change. If the capacity
expansion is limited to 30% or less
of installed capacity, relatively minor modifications to the
existing unit can achieve the required capacity increase. There
are several options.11 By modifying the process
cycle, hydrogen recovery can be traded for hydrogen capacity. A
3% reduction in recovery can lead to a capacity increase of up
to 20%, the net result being an increase in hydrogen
Improved hydrogen recovery of 1% to 2% can be achieved by
high-performance adsorbents (depending on the installed charge
of adsorbents). If the PSA unit operates at high tail gas
pressure, lowering the tail gas pressure through the addition
of a tail gas compressor can increase both capacity and
Membrane units can be expanded by simply adding membrane
area. Recovery and purity can be maintained if the pressure
levels around the membrane unit are kept constant. If permeate
(membrane product) compression is added to enable reducing
permeate pressure, hydrogen recovery and purity can be improved
with the original membrane area. Increasing the outlet
temperature of the feed superheater increases hydrogen recovery
at the expense of lower purity. Expansion of membrane purifiers is
facilitated if the pre- and post-treatment equipment are
designed to accommodate future capacity expansion.
For cryogenic units, capacity increases can often be
obtained by adding a fuel gas compressor without modifications
to the cold box itself. The cost impact of initially designing
the cold box for this type of expansion is low.10
Hydrotreaters and hydrocrackers consume hydrogen in a series
of reactions converting organic sulfur and nitrogen compounds
to hydrogen sulfide and ammonia. The hydrogen also reacts with
the hydrocarbons in the oil, increasing the hydrogen-to-carbon ratio.
Hydrocracking reactions convert heavier oils to diesel and
naphtha range materials. All of these reactions increase the
products' value and contribute to the refinery's gross
Hydrogen partial pressure both drives these reactions and
suppresses unwanted coke formation. A minimum hydrogen partial
pressure (usually measured as reactor inlet purity or recycle
gas purity) is required to operate with a reasonable catalyst
life and reactor temperature. The minimum hydrogen partial
pressure is not a fixed value. It is a function of current
operating conditions - charge rate, feed properties, desired
product properties. Operating below minimum hydrogen
partial pressure reduces catalyst life. Operating
above the minimum hydrogen partial pressure typically
requires an increase in total hydrogen supplied to the hydrogen
Makeup purity is often confused with hydrogen partial
pressure. For a given set of operating conditions, hydrogen
partial pressure is determined by the combination of makeup
purity and purge flow. It is possible to adjust hydrogen
partial pressure without modifying makeup purity. Conversely,
it is possible to utilize a different makeup stream, with a
different purity, and maintain the same hydrogen partial
pressure. From the operator's point of view, hydrogen partial
pressure can be adjusted by modifying the purge flow or
modifying the streams used as makeup to adjust the makeup
Thinking beyond the issue of minimum hydrogen partial
pressure is critical. Operators should:
- Regularly monitor the hydrogen partial pressure in key
hydrotreaters and hydrocrackers
- Have available hydrogen partial pressure targets that
reflect current operating conditions and optimization of refinery gross margin
- Adjust hydrogen partial pressures accordingly
For any set of operating conditions, there is an
optimum hydrogen partial pressure. Since hydrogen
partial pressure drives the reactions, increasing hydrogen
partial pressure can enable increased charge rate, improved
product properties or longer catalyst life. In hydrocrackers,
it can enable improved yields, or greater conversion per pass.
Therefore, increasing hydrogen partial pressure beyond the
minimum can increase the refinery gross margin well above the
additional hydrogen cost associated with increasing the
hydrogen partial pressure. To maximize the profitability of
these units, one must have a good understanding of the process
characteristics and refinery economics. Detailed process models
that reflect the performance of the units as a function of
hydrogen partial pressure are required.
Methods to significantly improve hydrogen partial pressure
include debottlenecking compression and adding H2S
scrubbing of the recycle gas.
HYDROGEN NETWORK IMPROVEMENTS
Hydrogen network analysis. One view of the
hydrogen network is as a collection of nodes (producers and
consumers) and the connections between them. Hydrogen pinch
analysis is a mathematical technique to analyze the hydrogen
network on this level.12 The analysis combines
hydrogen requirements (quantity and purity) of each consuming
unit, specified hydrogen production (quantity and purity) of
each hydrogen producing unit and designation of one hydrogen
producer as the swing unit, which will turn up or down to match
the needs of the consumers.
The approach is similar to energy pinch, but is different in
some key aspects. Hydrogen network pinch involves generating
source and sink composite curves based on network definition,
and a hydrogen surplus curve derived from the composite curves
(Figs. 3 - 5). Analysis of the surplus curve provides the
theoretical minimum hydrogen required from the swing producer
(such as a hydrogen plant) to meet the needs of the network,
assuming no constraints on how the units are connected.
Surplus curve -
This is a theoretical minimum hydrogen requirement.
Modifications to the real network necessary to achieve this
minimum might be as easy as opening and closing some valves, or as daunting as adding a
new multistage compressor to connect low-pressure sources to
high-pressure consumers. Intermediate modifications could be
adding cascades between the purge of one unit and the makeup of
Additional tools beyond hydrogen pinch are required to
design practical, efficient hydrogen networks.13 A
hydrogen network model can be used. This model must represent
the actual connectivity of the network, existing compressors,
and hydrogen consumption, light ends generation and solution
losses of each hydrogen consumer. With this tool, network
modifications can be tested and new hydrogen balances
This same model allows modifying operations to represent
different cases and operating modes - including summer/winter
and future operations - and adding new hydroprocessing units
and hydrogen purifiers.
Hydrogen purification analysis. Hydrogen
purification plays two roles in the hydrogen network. First, it
can be used to upgrade streams that are currently going to fuel
because they are too low in purity. Second, streams such as
catalytic reformer hydrogen - which are currently used directly
- can be purified to increase the partial pressure in reactors,
debottleneck makeup compressors and/or reduce purge to fuel
from hydroprocessing units.
Hydrogen pinch analysis requires all hydrogen purification
to be completely defined as part of the network definition.
Therefore it cannot directly define the optimum purification
scheme. However, the pinch purity does provide clear direction
as to which streams to consider for purification.
The most challenging aspect of purification analysis is
selecting the optimal purification scheme and operating
conditions. First, the purification technology (membrane, PSA, or on
rare occasion, cryogenic) must be selected. Beyond technology
selection, technical decisions - such as operating pressures,
membrane area, membrane polymer, PSA cycle and PSA adsorbent
mix - all have key impacts on performance and cost. In
the end, these factors must be considered in light of the
conflicting criteria of capital cost, product purity, hydrogen
recovery, compression requirements and operating flexibility.
Good purification process models, a thorough understanding of
purifier design and operations, and access to the hydrogen
network model are essential to effectively optimize the
Consider modifications to both new and existing purifiers.
Revamping existing equipment for increased throughput,
different feed or improved recovery is often
Results of hydrogen network analysis. Where
improvements in the hydrogen management system can be made is
different for every refinery, but profitability improvements
through better hydrogen management were identified in every one
of the 25 refineries we have worked with. Improvements
- Switching which streams are routed to the existing
- Routing low-purity hydrogen streams to the hydrogen
- Better control of partial pressure (purge rates)
- Improving pressure control to fuel
- Revamping PSA for higher capacity
- Increasing severity in a cat feed hydrotreater to
increase FCC gasoline yield
- Increasing throughput in the
One refinery identified over $6 million/year in hydrogen
savings with no capital projects. Another avoided the
capital cost of 20 MMscfd of new hydrogen plant capacity
through much smaller investment in hydrogen recovery capacity.
Table 1 provides an overview of hydrogen network improvement
THE REAL WORLD - MANAGING THE HYDROGEN
Hydrogen networks should be operated so that:
- Amount of hydrogen sent to fuel is minimized
- Profitability of each process unit is maximized
- Hydroprocessing catalysts are not exposed to hydrogen
partial pressures below safe operating levels
- Adjustments are made in response to day-to-day changes in
Of course, every hydrogen network is different, but it is
possible to offer some general guidelines that would serve
every refinery well.
If it's important, measure it. If you don't
know how much hydrogen is going to fuel, you can't minimize it.
If you don't know the recycle purity in your hydrocracker, you
can't optimize it. So, if you want to take control of your
- Make sure you have reliable meters on important
- Where practical or critical, put online analyzers on
recycle streams and around purifiers.
- Do whatever it takes to get lab samples in a timely and
reliable manner when the life of your catalyst is at stake,
or when you are reducing charge rates because you are wasting
hydrogen due to running the hydroprocessing unit vents too
If controllers are putting hydrogen to fuel, do
something. Some refineries continually dump
high-purity hydrogen to fuel through pressure control valves in order to maintain the
refinery hydrogen balance. There must be a better way!
Look for other streams to control with. Look for other
places to put the control flow besides fuel. Look for ways to
reduce the amount of control flow. Improvement may be found in
better control schemes, improved regulatory control or some
advanced control. Don't accept at face value that "We have to
waste this much hydrogen."
Get everybody thinking in dollar terms.
Operators, technical staff and management should think and see
dollars when they see the Hydrogen Network Daily Report. The
hydrogen in the fuel system should be reported in units of
dollars/day. So should the: hydrogen plant production;
opportunity cost of turning down the hydrocracker because of a
hydrogen supply constraint; and increased cetane number in the
diesel hydrotreater product due to higher partial pressure.
Convert setpoints to dollars in the operator's mind, and he
or she will control dollars. Represent hydrogen network
problems to management in dollar terms, and they will find the
resources to fix them.
Use meaningful hydrogen partial pressure
targets. Too-low hydrogen partial pressure can reduce
hydroprocessor performance (profitability) and, at worst,
damage the catalyst. Higher than optimum hydrogen partial
pressure often results in higher network hydrogen requirements
(costs). Not all hydroprocessing units are the same. Some are
relatively insensitive to recycle purity, have robust catalyst,
involve a relatively easy reaction and don't directly generate
a lot of profit. Critical hydroprocessing units are just the
If hydrogen partial pressure matters, then:
- Measure it.
- Establish a manual or automated control scheme to adjust
the makeup and/or purge to control it.
- Establish a target recycle purity.
- Modify the recycle purity target as feed properties, feed
rates or desired product specifications change.
- Let operators know how well they are controlling hydrogen
partial pressure, in terms of dollars/day.
Always consider the big picture. In most
refineries, the hydrogen network covers most operating areas.
Changes in one operating unit will affect the hydrogen
availability or purity in the other end of the plant. Operating
changes that affect the network should be made based on what is
most profitable for the refinery as a whole. Optimum use of
hydrogen cannot be achieved if each unit is operated
independently. Operators must communicate about actions that
affect the network. Tools and policies must be in place to help
optimize the network as a whole, rather than as individual
Putting it all together. Combine the
hydrogen network and process models with online data, an
understanding of refinery economics, objective function and a
nonlinear solver. The result is an online optimizer that will
recommend operator actions to protect catalyst, use hydrogen
efficiently and maximize refinery profitability.
HOW MUCH IS HYDROGEN WORTH?
Managing your hydrogen network requires understanding the
value of hydrogen to your refinery. You need this number to
evaluate hydrogen network improvement projects and to make
day-to-day hydrogen network decisions. Hydrogen's value is a
function of the refinery's current and future hydrogen
situation. Generally, the hydrogen situation falls into one of
the following categories:
- Catalytic reformers produce enough hydrogen to satisfy
all the users; there is no hydrogen plant or purchased
hydrogen. In this case, hydrogen should be valued at fuel
- Refinery is not constrained by hydrogen availability;
hydrogen production and/or purchase is required to meet the
hydrogen balance. In this case, the hydrogen's value is the
production/purchase cost of incremental hydrogen.
- Refinery modifications are being planned that will
require more hydrogen than is currently available. In this
case, hydrogen is valued not at its marginal production cost,
but at its fully loaded cost, including capital charges for a
new hydrogen plant. Hydrogen recovery projects are extremely attractive
in this situation.
- Refinery is constrained by hydrogen availability; at
least some of the time, charge rates and severities are
adjusted in hydroprocessing units due to limited hydrogen
supply. In this case, hydrogen's value is determined by the
opportunity cost of constraining the hydroprocessing units.
This can be calculated by adjusting the hydrogen constraint
in the refinery LP model and determining the hydrogen
marginal value. In this situation, hydrogen usually has much
greater value than in the other cases.
The opportunities to make process improvements through
optimized hydrogen partial pressure can impact hydrogen
valuation, so valuation may not be quite this simple.
IS YOUR LP LEADING YOU TO BAD
If the refinery is hydrogen-limited, planning tools (LP)
must accurately reflect the impact of crude properties and unit
operating conditions on hydrogen consumption, vent flows and
purities. Inaccurate modeling of hydrogen production,
consumption and vents can lead to incorrect valuation of crudes
or refinery operating modes. For example, different crudes and
cutpoints will affect the chemical hydrogen consumption in a
diesel hydrotreater or the hydrogen yield in a catalytic
Equally important, the purge flow on this hydrotreater may
have to be increased to maintain the desired hydrogen partial
pressure. The increased purge results in increased makeup flow,
which also needs to be represented in the LP hydrogen balance.
If the real-life hydrogen constraints are different than those
modeled in the LP, then actual process unit throughputs and
product qualities will also be different from planned.
1 Dachos, N., A. Kelly, D. Felch and E. Reis,
"UOP Platforming Process," Handbook of Petroleum
Processes, Robert A. Meyers, ed. 2nd edition,
McGraw-Hill, New York, 1997.
2 Wier, M.J., J. Utley, J. Elstein and D.
Schwake, "Strategies for Maximizing Profits from Catalytic
Reforming Units," NPRA Annual Meeting, March 1998.
3 Rachford, R. H. and N. j. Gilsdorf, "The
Platforming Process in the Reformulated Gasoline Era," NPRA
Annual Meeting, March 1992.
4 Broadhurst, P. V. and P. E. J. Abbott,
"Improving Hydrogen Plant Performance, Part I," Petroleum
Technology Quarterly, Summer 2002, p. 130.
5 Broadhurst, P. V. and P. E. J. Abbott,
"Improving Hydrogen Plant Performance, Part II,"
Petroleum Technology Quarterly, Autumn 2002, p.
6 Fecteau, D. J., "Revamp Engineering Update,"
7 Beshears, D. R., "CCR Platforming Catalyst
Selection Improves Unit Flexibility and Profitability," NPRA
Annual Meeting, March 2000.
8 Cromarty, B., C. W. Hooper and K. Chlapik,
"Cost-Effective Uprating of Existing Hydrogen Production
Units," NPRA Annual Meeting, March 1994.
9 Fleshman, J.D., "Cost Efficient Revamps in
Hydrogen Plants," Petroleum Technology Quarterly,
Summer 2001, p. 83.
10 Miller, G.Q. and J. Stoecker, "Selection of
a Hydrogen Separation Process," NPRA Annual Meeting, March
11 Picioccio, K. and E. Reyes, "Breaking the
Barrier with PSA Revamps," Petroleum Technology Quarterly, Spring
12 Alves, J.J., "Analysis and Design of Refinery Hydrogen Distribution
Systems," PhD thesis, UMIST, Manchester, UK, 1999.
13 Zagoria, A., G. P. Towler, B. M. Wood and F.
M. Hibbs, "If You Are Burning H2, You Are Burning
Money," NPRA Annual Meeting, March 1999.
14 Ciolek, W.H., "Hydrogen Management - It's
For Every Refiner," AIChE Spring National Meeting, April
Alan Zagoria is a senior
engineering consultant with the Solutions and Services
Department of UOP LLC in Des Plaines, Illinois. For the
past five years, he has led UOP's Hydrogen Management
Group, assisting refiners around the world in tackling
the challenges of meeting clean fuels requirements.
Mr. Zagoria has worked for UOP/Union Carbide for more
than 25 years. Most of his work has been in the area of
hydrogen purification. Within that specialty he has
lead UOP's efforts in process development, process
design, control system design and field services. He
earned a BS in chemical engineering from Northwestern
University. He can be reached at
Rudolf Huycke joined UOP in 1991
and is manager of the Process Engineering Department of
UOP N.V. in Antwerp, Belgium. He is accountable for
optimizing hydrogen purification schemes, process
design and development, control systems programming and
startup activities of hydrogen purification equipment
for the refining and petrochemical industry. Mr.
Huycke holds an MS in chemical engineering from the
University of Ghent, Belgium, and can be reached at