Article

pubs.acs.org/IECR

Hydrogen Source Diagram: A Procedure for Minimization of

Hydrogen Demand in Petroleum Refineries
Joana L Borges,† Fernando L. P. Pessoa, and Eduardo M. Queiroz*
Departamento de Engenharia Química, Escola de Química, Universidade Federal do Rio de Janeiro, Av. Athos da Silveira Ramos, 149,
CT, Bl. E, Cidade Universitária, 21941-909, Rio de Janeiro, Brasil

ABSTRACT: The consumption of hydrogen in petroleum refineries is increasing due to its growing intensive use in treatment
processes and in new refining technologies. This trend and the reduction of hydrogen production in gasoline processing have
generated a new panorama. In this new context, target identification for minimum consumption of hydrogen in a distribution
network has become fundamental information in the definition of minimum hydrogen requirements in petroleum refineries. This
paper presents the hydrogen source diagram (HSD), a new algorithmic method for minimum hydrogen consumption target
calculation and, simultaneously, hydrogen network flowsheet synthesis. The proposed HSD procedure is described in details
using a case study, showing its wide scope. The results demonstrate the effectiveness of the procedure and show that the
generated network can be easily adapted to meet specific plant restrictions.

1. INTRODUCTION requirement. However, there are little systematic proposals to

It is uncontroversial that the oil industry, particularly oil do so present in literature.
refineries, has great potential to negatively affect aquatic Initially, the analyses of hydrogen availability have been
environments, as well as the air and soil, during its production conducted internally by oil processing companies focusing the
process. Because of this effect and the increase of societies’ application of specific technologies, rather than analyzing the
environmental conscience, the oil sector is facing a growing global distribution problem.3 From the 2000s, many studies
demand for reduction in waste generation and suitability of originally developed for water networks integration have been
products to new environmental specifications. adapted in order to optimize the use of hydrogen.
Hydrogen is an input in several refining process. It has been Many of these works are based on graphical methods,3−10
used in cracking processes for the suppression of the formation with which the minimum supply target is calculated. In the first
of heavy byproducts, and in treatment processes in order to paper of the area3 it was necessary to establish initial random
preserve the catalyst and improve the characteristics of products gas consumption targets. From those initial ideas, algebraic
(reducing the amount of sulfur, for example). Hydrogen can be methods have been developed as an adaptation of the graphical
also used as fuel, since the limitations in gas emissions ones.11−15 Also optimization methods based on superstructure
restrained the refineries use of oil to produce power, compelling or linear algebraic procedures, adding various constraints, have
them to seek alternative fuels with lower carbon content. been developed.16−25
The regulatory requirements, however, are not the only Some authors have recently studied hydrogen networks using
reason why refineries have been significantly more interested in data from real refineries and procedures based on mathematical
optimizing the use of hydrogen. The gas has been widely programming. For example, Fonseca et al.26 solved a problem
studied for changes in conventional processes and for of the Porto Refinery of GALP ENERGIA; Khajehour et al.27
development of new refining technologies aimed at the used data from the Tehran North Refinery (Iran), while Liao et
improvement of hydrocarbons conversion. al.28 used data from an existing refinery near Shanghai (China).
In addition to the advantages presented by the use of Wan Alwi et al.8 presented a procedure that performs
hydrogen in a refining scheme, if the utilities system is not simultaneously the synthesis and the targeting analysis;
included, hydrogen production and the FCC unit (fluid however, their procedure needs a graphical iteration procedure
catalytic cracking) are the most impacting processes of a which is not practical to perform in a large network.
refinery's CO2 emissions.1 Furthermore the synthesis of the network is not directly
Traditionally, there was an excess of hydrogen in refineries, achieved by the graphical procedure as claimed. They use a
since the amount produced was systematically higher than the nitrogen network as example.
amount consumed. With the growing needs of hydrogen in This paper presents the hydrogen source diagram (HSD), a
treatment processes and new refining technologies, combined process integration tool that calculates the minimal hydrogen
with the reduction of hydrogen production as a result of consumption target and simultaneously makes the synthesis of
processing low aromatic gasoline, it has become increasingly
necessary to supplement hydrogen in the network.2 Received: April 9, 2012
There is a consensus between industry and the scientific Revised: August 29, 2012
community regarding the need for effective management and Accepted: September 4, 2012
optimization of hydrogen in order to minimize the plant’s Published: September 5, 2012

© 2012 American Chemical Society 12877 dx.doi.org/10.1021/ie300893d | Ind. Eng. Chem. Res. 2012, 51, 12877−12885
Industrial & Engineering Chemistry Research Article

the hydrogen network. The HSD is an extension of the Water Operation. In this step, the amount of hydrogen consumed
Source Diagram (WSD) presented by Castro and co-workers,29 (reacted) in each operation, in each interval is calculated. The
and later extended by Gomes and co-workers.30 More details procedure is based on the assumption of constant hydrogen
about the analogies between the WSD and HSD can be seen in consumption independent of its concentration. The amounts
Borges.31 are registered in brackets over each arrow, into the respective
interval.
2. HYDROGEN SOURCE DIAGRAM PROCEDURE 2.6. Sixth Step: Identification of the Available H2
The hydrogen source diagram (HSD) procedure has eight Source in Each Interval. The hydrogen sources can be
steps. In this section, the steps will be briefly described, and in either primary (external) or from operations situated at
the next section a case study will be used to illustrate its previous intervals (internal). Operations classified as “consum-
application. er” in the second step are hydrogen internal sources only for
2.1. First Step: Processes Mapping and Building of themselves.
the Opportunity Table. Initially, it is necessary to map the 2.7. Seventh Step: Calculation of the Hydrogen Mass
processes that consume and produce hydrogen in the refinery Flow Rate in Each Operation in Each Interval. For the
in order to identify the mass flow rate and hydrogen calculation of the hydrogen mass flow rate in each operation in
concentration of streams which could be sources or sinks of each interval some heuristics rules are used. They are
hydrogen. With this information, the Opportunity Table is i. use of internal sources has priority over the external
built. As explained in Alves and Towler,3 in general, the sources
makeup, recycle, and purge flow rates extracted from the ii. use of less concentrated streams has priority
material balance of the network need to be manipulated to iii. use of the maximum amount of hydrogen in each interval
effectively represent the values of process’ inputs and outputs. iv. for operations covering more than one concentration
2.2. Second Step: Classification of the Operations and interval, the stream must continue to flow through the
Calculation of Their Mass Consumption of Hydrogen. In same operation until its end (this heuristic avoids
all hydrogen consuming processes there is consumption of both operations division)
hydrogen and contaminants. Gas networks have distinct v. use of hydrogen generated in situ has priority
features when compared with water networks: in water
The required mass flow rate from source q in operation k, at
networks the loss is generally by evaporation and not by
concentration interval i, can be calculated by eq 2 for external
reaction and it is always possible to consider flow rates in a base
sources and eq 3 for internal sources.
free of contaminants. In hydrogen distribution networks,
commonly the processes present hydrogen loss (consumption) N fin, i
i
Δm H2ki − ∑ j = 1 (F jki[Pfi − Pji])
and can be treated by subdividing into two virtual subprocesses: e
Fqki =
one that can be treated assuming constant total flow rate, with Pfi − Pq e (2)
the original inlet and outlet concentrations; and a second one
q i
which has no effluent. Therefore, it is sufficient to consider that Δm H2ki − ∑ j = q − 1 (F jki[Pfi − Pji])
the first operation has a flow rate similar to the original i
Fqki =
operation outlet and the second has a flow rate equal to the Pfi − Pq i (3)
difference between the original inlet and outlet flow rates.
An operation with inlet mass flow rate Fi and composition Pi where: is the required mass flow rate of the external source
Feqki
which generates an outlet mass flow rate Fo with composition q in operation k at interval i; Fiqki is the required mass flow rate
Po has its respective hydrogen consumed mass ΔmH2 calculated of the internal source q in operation k at interval i; ΔmH2ki is the
by eq 1. amount of hydrogen that must be consumed in operation k at
interval i; Fijki is the mass flow rate of source j in operation k at
Δm H2 = FP
i i − FoPo (1) interval i; Pfi is the final concentration of the interval i, Pji is the
concentration of source j used in interval i; Pqe is the external
The operations are also classified as constant flow rate, source q concentration; Nfin,i is the number of internal source
consumer, or source. The consumer operations are those available at interval i; Pqi is the internal source q concentration.
without outlet streams. The sum in both equations represent the amount of hydrogen
2.3. Third Step: Definition of the Hydrogen Concen- already consumed in operation k, at interval i, using sources
tration Ranges in the Process. The building of the HSD with priority over the considered one.
begins by writing in a line, in descending order, the If the flow rate calculated for the considered source is greater
concentrations of the external hydrogen sources, as well as than the available, use the available flow rate and go to the next
the inlet and outlet concentrations of all operations. All source, until the calculated value is equal or smaller than the
adjacent concentrations define a concentration interval, inside available one.
which calculations are made. Following this procedure, the necessary mass flow rate in
2.4. Fourth Step: Representation of the Operations in each operation in each interval is calculated, as well as its
the HSD. Each operation is represented by an arrow, starting respective source or sources.
from the inlet concentration to its outlet. The identification of 2.8. Eighth Step: Calculation of the Total Hydrogen
the operations and their mass flow rate limits are presented on Mass Flow Rate in Each Interval, Pinch Determination.
the left side of the diagram. Therefore, the HSD has a matrix The pinch of the system is determined by observing the
organization, with flow rates as lines and concentration intervals behavior of the sum of the consumed hydrogen flow rate in
as the columns. each interval (in all operations) and identifying the interval in
2.5. Fifth Step: Calculation of the Hydrogen Mass which its value decreases. The concentration in the beginning
Consumed in the Concentration Intervals in Each of this interval is the pinch of the system.
12878 dx.doi.org/10.1021/ie300893d | Ind. Eng. Chem. Res. 2012, 51, 12877−12885
Industrial & Engineering Chemistry Research Article

The pinch determination is important to identify the best Table 2. Opportunity Table with Streams Classification and
streams to be purified and also to identify the equipments by Computed Hydrogen Consumption
which individual optimization has the greatest potential to
F PE (mol PS (mol ΔmH2 (mol
debottleneck the network, because such strategies will only
operation (mol/s) % H2) % H2) classification H2/s)
bring benefits if the hydrogen availability above the pinch is
increased or the hydrogen necessity below the pinch is reduced. 1 1801.19 80.61 75.00 constant flow 101.05
1′ 693.81 80.61 consumption 559.28
3. CASE STUDY 2 138.60 78.85 75.00 constant flow 5.34
2′ 41.60 78.85 consumption 32.80
The presented study involves a set of operations already used in 3 346.50 77.57 73.00 constant flow 15.84
literature. This choice reveals itself useful especially for 3′ 207.90 77.57 consumption 161.27
comparative purposes and for demonstrating advantages of 4 457.40 75.14 70.00 constant flow 23.51
the HSD methodology. The example involves four processes 4′ 263.30 75.14 consumption 197.84
that consume hydrogen and two hydrogen producers and there SRU 623.80 93.00 source
is an external source. CRU 415.80 80.00 source
3.1. Consumption Targets with Maximum Reuse. The Import 346.50 95.00 source
example uses data presented by Alves and Towler3 in a paper
that could be seen as a precursor of a recent set of hydrogen
distribution network methodologies for setting a minimum
consumption target. The main objective is to evaluate the
hydrogen economy that could be implemented in a distribution
network only through the reuse of streams. Therefore,
significant capital investment is not considered (except for
rearrangement of pipes) as a possible option.
The HSD methodology for maximum reuse is presented
according to the steps already introduced in the previous
section.
Step 1: The original operational limits are presented in Table
1 which can be seen as the Opportunity Table of the study.

Table 1. Limiting Data (Opportunity Table)3

inlet outlet Figure 1. Result of steps 3 and 4. Operation representations on HSD.
flow rate composition flow rate composition
process (mol/s) (mol % H2) (mol/s) (mol % H2)
hydrocracker unity 2495.00 80.61 1801.19 75.00
(HCU)
straight-run naphtha 180.20 78.85 138.60 75.00
hydrotreater
(NHT)
diesel hydrotreater 554.40 77.57 346.50 73.00
(DHT)
cracked naphtha 720.70 75.14 457.40 70.00
hydrotreater
(CNHT)
steam reformer unity 623.80 93.00
(SRU)
catalytic reformer 415.80 80.00
unity (CRU)
import 346.50 95.00
Figure 2. Result of step 5. Hydrogen mass consumed in each
operation, in each interval.
Step 2: The Opportunity Table shown in Table 2 has
additional information. The consuming processes are divided,
as proposed, in a constant flow process and in a consumer one. available hydrogen sources before it, respecting the heuristic
Furthermore, the respective hydrogen consumption is calcu- rules adopted (listed on subsection 2.7).
lated (ΔmH2). Note that there is no operation in the first two concentration
Steps 3 and 4: In Table 2 the operations that will be intervals. In the third interval (i = 3) there are operations 1 and
considered in the HSD are defined as well as the respective 1′. The hydrogen available for this interval comes from sources
inlet and outlet concentrations, which define the concentration in the two previous intervals and are equal to Fe2 = 623.80 mol/s
intervals. Figure 1 shows the structure of the HSD with the at 93.00 mol % H2 and Fe1 = 346.50 mol/s at 95.00 mol % H2.
preceding data. According to the proposed heuristics, the use of the stream at
Step 5: In Figure 2 the quantities in brackets over the arrows 93.00 mol % H2 is the first one to be considered since the other
are the amount of hydrogen consumed in each interval, in each available source is external, imported from outside. Although
operation. They are calculated using eq 1. the two available sources here are externals in the point of view
Steps 6 and 7: All hydrogen flow rates necessary for each of the HSD procedure, importation versus source in an onsite
operation in each interval are calculated according to the process had to be here considered. The flow rates required to
12879 dx.doi.org/10.1021/ie300893d | Ind. Eng. Chem. Res. 2012, 51, 12877−12885
Industrial & Engineering Chemistry Research Article

Figure 3. Hydrogen source diagram: maximum reuse.

transfer the amount of hydrogen in this interval in each 80.00 mol % H2 necessary to supply the amount of hydrogen
operation are calculated by eq 3. Therefore, in the third used in this interval is equivalent to 1801.90 mol/s, as shown in
interval, the necessary flow rates for operations 1 and 1′ are eq 6.
84.52 mol/s and 32.56 mol/s, respectively, as shown by eqs 4
and 5. These flow rates are represented in the HSD of Figure 3 F80% =
20.71
= 1801.90
above the corresponding arrow of each operation. (0.8000 − 0.7885) (6)
e 10.987
F213 = = 84.52 Then this internal source will be totally used. The next source is
0.93 − 0.80 (4)
the external from the CRU process, also at 80.00 mol % H2,
e 4.232 since the other internal sources will be used in operation 1′
F21 ′3 = = 32.56
0.93 − 0.80 (5) (priority rules). Then, at 80.00 mol % H2 the available stream
In the fourth interval, the operations are also 1 and 1′. For them flow rate is only 500.32 mol/s (Fi414= 84.52 mol/s from
there are available as hydrogen source the following streams: operation 1 and Fi314 = 415.80 mol/s from the catalytic
External sources: Fe1 = 346.50 mol/s at 95.00 mol % H2. Fe2 = reformerCRU). With the internal sources from operation 1′
506.72 mol/s at 93.00 mol % H2; and Fe3 = 415.80 mol/s at tied to this operation, the necessary additional hydrogen comes
80.00 mol % H2. Internal sources: Fi4 = 84.52 mol/s at 80.00 from an external source. The one at 80.00 mol % H2 is already
mol % H2, from operation; and Fi5 = 32.56 mol/s at 80.00 mol used. The next one in the priority list is the external source at
% H2. The focus is first on operation 1. The source that has 93.00 mol % H2. The flow rate of this stream necessary to
priority is the internal one that comes from itself, which is 84.52 complete the hydrogen necessity in this interval is calculated by
mol/s at 80.00 mol % H2. However, the stream flow rate at eq 2, rewritten here with the appropriate parameters as eq 7.

e 20.71 − (84.52[0.8000 − 0.7885] + 415.80[0.8000 − 0.7885])

F214 = = 105.72
(0.9300 − 0.7885) (7)

Hence, 105.72 mol/s of the available 306.72 mol/s of this Similarly, the flow rate of a stream at 80.00 mol % H2
source are used. required to provide to operation 1′ the amount of hydrogen
12880 dx.doi.org/10.1021/ie300893d | Ind. Eng. Chem. Res. 2012, 51, 12877−12885
Industrial & Engineering Chemistry Research Article

Figure 4. Flow sheet of the hydrogen network corresponding to the HSD of Figure 3: Maximum reuse (flow rates in mol/s and concentrations in
mol % H2).

necessary in this interval is equal to 693.81 mol/s. However, at mol/s. This means a reduction of about 3.5% in the
80.00 mol % H2 there is only Fi514 = 32.56 mol/s available from consumption of imported hydrogen and also that the HSD
operation 1′. With this flow rate, the amount of hydrogen procedure shows a similar performance when compared with
consumed corresponds to 0.37 mol/s. Therefore, 7.60 mol/s other methodologies.
remain to be furnished. Again, the external source with Step 8: The pinch is observed in the 70 mol % H2
concentration 93.00 mol % H2 is used and its flow rate is Fi214 concentration. The total flow rates of the hydrogen streams
= 53.74 mol/s, according to eq 2. A similar procedure is used in increase until this concentration where it begins to decrease. In
sequence in the following intervals. this case, the pinch is observed in the same concentration level
Figure 3 shows the results in the HSD for this case study. by the HSD methodology and others’ methodologies in
Note that there is reuse of hydrogen from operation 1 (white literature.
circle) into operations 1′, 2′, 3 and 3′. There is reuse from 3.2. Synthesis of the Hydrogen Distribution Network
operation 2 (white star) into operation 2′. There are reuses
with Maximum Reuse. From the HSD of Figure 3 it is
from operation 3 (white cross) into operations 3′ and 4, and
possible to easily draw a flow sheet that meets the minimum
also from operation 4 into operation 4′.
The minimal consumption of hydrogen of each source can be consumption calculated. Distinctly of other proposed methods,
easily calculated by adding the flow rates required from each with the HSD it is possible to establish simultaneously the
source in its first interval, resulting in a requirement of 268.86 minimum consumption target and to synthesize the corre-
mol/s of the imported external stream at 95.00 mol % H2. The spondent network. The mixing of streams creates a new stream
other “external” sources (streams at 80.00 and 93.00 mol % H2) that is able to promote the same hydrogen transfer with a
are fully used. concentration equal to the operational concentration or less
Other literature methods3,4,12 calculate a minimum con- than it. In the synthesis of the network all source streams
sumption of the stream at 95.00 mol % H2 of 268.82 mol/s. In required for a particular operation are mixed before entering it,
the original process the consumption of this stream is 277.20 as shown in Figure 4.
12881 dx.doi.org/10.1021/ie300893d | Ind. Eng. Chem. Res. 2012, 51, 12877−12885
Industrial & Engineering Chemistry Research Article

Figure 5. Hydrogen source diagram: reuse with flow rate constraints.

Note that in the network, operations created along the HSD imported hydrogen consumption might be greater than the
procedure as “consumers” do not appear. Actually, they are minimum calculated without constraints.
within the original operations and cause the decrease in the 3.3. Reuse with Flow Rate Constraints. Examples of flow
total flow rate. rate constraints can be linked with the need of purge inside
The process inlet conditions in the proposed flow sheet processes. In these cases, an additional heuristic must be used
(Figure 4) could not be fully satisfied in terms of operational in the seventh step. It could be, for example: vi. purge should be
pressures. For example, operation 3 has the inlet flow rate and up to 1% of the inlet stream and total reuse of a stream is
inlet concentration identical to those originally reported avoided.
(although the combination of streams is different in order to To attend this heuristic it is necessary to calculate the
reach the required concentration and flow rates). Nevertheless, minimum flow rate of purge in each operation and then
in operations 1, 2, and 4, inlet flow rates are lower and subtract it from the total flow rate in the respective operation
concentrations are higher than the original ones. These
exit, in order to know the actual flow rate available for the next
parameters could be adjusted by the manipulation of the total
operations in the downstream concentration intervals.
pressure of the stream or its concentrations, by adding an inert
In the original set of operations, the minimum purge from
stream or adjusting the internal recycle present in these
operation 1 is then 24.95 mol/s, from operation 2 is 1.80 mol/
processes.
These adjusting procedures could cause important impacts in s, from operation 3 is 5.54 mol/s, and from operation 4 is 7.21
cost analysis, but such evaluation is beyond the scope of the mol/s.
present study. Figure 5 shows the HSD for this situation, obtained following
Other restrictions may be easily satisfied by adapting the the steps already presented. The minimal consumption of
HSD procedure. An example is the presence of a minimum hydrogen is 274.88 mol/s at 95.00 mol % H2. The other
purge in all operations to avoid contaminants accumulation. sources (streams at 80.00 and 93.00 mol % H2) are fully used.
Additionally, restrictions in the reuse of some streams, based on Compared to actual operational conditions, there is again a
pressure requirements or unit positions in the plant, could also possibility of reduction in the hydrogen consumption, but with
be taken into account in the proposed procedure. However, in the present constraint the reduction is smaller than the
those situations assuming only the option of reuse, the observed without restrictions.
12882 dx.doi.org/10.1021/ie300893d | Ind. Eng. Chem. Res. 2012, 51, 12877−12885
Industrial & Engineering Chemistry Research Article

Note that the changes in operational conditions cause a shift between the flow rate of the outlet stream and the flow rate
in the pinch value, which is now 75 mol % H2. However any used in the regeneration (in this case study 102.52 mol/s).
change in the network structure is not observed. Following the HSD procedure with the observations
3.4. Regeneration and Reuse. A regeneration process presented in this section, the HSD obtained is shown in Figure
represents an option to reduce the imported hydrogen 7. The minimum consumption of imported hydrogen with the
consumption. Nevertheless, the costs involved may be too presence of a regeneration process is 197.51 mol/s, with a
high and imply no economic advantage on its use. As just discharge flow rate of 32.95 mol/s.
mentioned, in the present study the focus is only on the Foo and Manan,12 using the Hydrogen Cascade Analysis in
hydrogen consumption, and in this section the objective is to the same original set of operations and with equal regeneration
show how a regeneration process is added in the HSD performance, have calculated an hydrogen consumption almost
procedure. equal to that obtained by the HSD (difference of 0.4%).
An example of regeneration process is the membrane gas Ng et al.19 explored this same example, also used by Foo and
separation. For hydrogen it has a recovery of 95% and produces Manan,12 using an optimization approach with a objective
a stream with 98% of hydrogen.12 Many alternatives were function based in economic aspects to find the best stream to
investigated in a search of the best stream to be regenerated in be regenerated and its corresponding flow rate. To do a similar
the flow sheet shown in Figure 4 (maximum reuse), and the study using the HSD procedure, an external loop must be
CNHT exit stream, which has the lowest hydrogen employed. With the defining of an objective function
concentration, was the one that presented the best result.31 (regeneration costs and/or external hydrogen costs), the
Thus the following case study assumes the regeneration of the HSD procedure should be repeated sequentially changing the
CNHT exit stream (102.52 mol/s with 70 mol % H2), using the features of the external source correspondent to the
membrane gas separation process. Figure 6 presents a regenerated stream in order to achieve the minimum of the
schematic figure of the regeneration process. objective function. Details of this additional procedure will be
addressed in a future paper.
3.5. Results Summary. The aim of the HSD procedure is
to propose options to decrease the consumption of fresh
hydrogen. To compare its performance against other method-
ologies available in literature, a set of operations and process
conditions were used.
Results obtained with the HSD procedure are compared in
Table 3 with results of other authors that have used similar
definitions in solving analogous problems.
Figure 6. Regeneration process.
Table 3. Comparison of the Results
reduction in
A mass balance on the regeneration process is presented in procedure to find flow rate of the H2
eqs 8, 9, and 10. It generates a purge stream of 32.95 mol/s minimum imported H2 consumption
with 10.89 mol % H2, and a regenerated stream of 69.57 mol/s case consumption target (mol/s) (%)
with 98 mol % H2. original network 277.20
reuse hydrogen source 268.86 3.01
RFi,regyi,reg diagram
Fo,reg = Alves and Towler3 268.82 3.02
yo ,reg (8) reuse with flow rate hydrogen source 274.88 0.84
constraints diagram
Fpurge,reg = Fi,reg − Fo,reg (9) regeneration and hydrogen source 197.51 32.95
reuse (membrane diagram
Fi,puryi,reg (1 − R ) gas separation) Foo and Manan12 196.75 30.48
ypurge,reg =
Fpurge,reg (10)
where yo,reg is the regenerator purified concentration (0.98), R is
the recovery (0.95), yi,reg is the regenerator inlet concentration 4. CONCLUSIONS
(0.70), ypurge,reg is the purge concentration, Fo,reg is the This paper presents a new methodology for calculating the
regenerator purified flow rate, Fi,reg is the regenerator inlet minimal hydrogen consumption goal in oil refineries called
flow rate, Fpurge,reg is the purge flow rate. hydrogen source diagram (HSD). Good results were achieved
In the HSD procedure the regenerated stream is also mapped in a comparison to literature with the advantage of
as a hydrogen source with limited flow rate (step 1). In step 3, simultaneous synthesis of the network. The HSD procedure
the concentrations of the regenerated stream and of the allows a less confusing representation of operations and
regeneratiońs purge are considered in the definition of the hydrogen streams and also synthesizes simultaneously the
diagram limits. The other steps are performed in a similar way. network, being a tool easy to be understood and used by
Regarding the heuristics rules, the regenerated stream is process engineers.
treated as an external hydrogen source with the in situ Similar to many algorithmic methods using the pinch
generation priority (rule v−step 7). The only major difference concept, the HSD procedure has no iterations and all
is on the availability of the hydrogen stream at the end of the mathematic operations are simple and extremely easy to be
operation that has its outlet regenerated as an internal computationally implemented. Furthermore, the procedure
hydrogen source. The flow rate available is the difference might be easily adapted in order to attend several factors
12883 dx.doi.org/10.1021/ie300893d | Ind. Eng. Chem. Res. 2012, 51, 12877−12885
Industrial & Engineering Chemistry Research Article

Figure 7. Hydrogen source diagram: regeneration and reuse: regeneration using membrane gas separation.

such as safety, distance, operability, controllability, and NHT = straight-run naphtha hydrotreater
economics. The procedure can also be extended to other SRU = steam reformer unity
applications in which hydrogen is present, as the synthesis of WSD = water source diagram

■
ammonia, the treatment of metals, and other energy purposes.
In all these scenarios the advantages of the HSD methodology REFERENCES
remain the same.

■
(1) Moore, I. Reducing CO2 emissions. Pet. Technol. Q. 2005, Q2, 1
http://www.eptq.com/view_article.aspx?intAID=211.
AUTHOR INFORMATION (2) Bealing, C.; Hutton, D. Hydrogen-pinch analysis. Chem. Eng.
Corresponding Author 2002, 109 (5), 56.
*E-mail: [email protected]. Tel: (55)(21)2562-7603. Fax: (55) (3) Alves, J. J.; Towler, G. P. Analysis of refinery hydrogen
(21)2562-7567. distribution systems. Ind. Eng. Chem. Res. 2002, 41, 5759.
(4) El-Halwagi, M. M.; Gabriel, F.; Harell., D. Rigorous graphical
Present Address
† targeting for resource conservation via material recycle/reuse net-
Universidade Federal do Rio de Janeiro, Programa de works. Ind. Eng. Chem. Res. 2003, 42, 4319.
́
Engenharia Quimica, Av. Athos da Silveira Ramos, 149, CT, (5) Agrawal, V.; Shenoy, U. V. Unified conceptual approach to
Bl. G, sala 115, Cidade Universitária, 21941-909, Rio de Janeiro, targeting and design of water and hydrogen networks. AIChE J. 2006,
Brasil. 52, 1071.
Notes (6) Bandyopadhyay, S. Source composite curve for waste reduction.
The authors declare no competing financial interest. Chem. Eng. J. 2006, 125, 99.

■
(7) Zhao., Z.; Liu., G.; Feng., X. New graphical method for the
integration of hydrogen distribution systems. Ind. Eng. Chem. Res.
ACKNOWLEDGMENTS
2006, 45, 6512.
The authors thank CNPq for the research grant.

■
(8) Wan Alwi, S. R. W.; Aripin, A.; Manan, Z. A. A generic graphical
approach for simultaneous targeting and design of a gas network.
ABBREVIATIONS Resour. Conserv. Recycl. 2009, 53, 588.
CNHT = cracked naphtha hydrotreater (9) Tan, R. R.; Ng, D. K. S.; Foo, D. C. Y. Graphical approach to
minimum flowrate targeting for partitioning water pre treatment units.
CRU = catalytic reformer unity Chem. Eng. Res. Des. 2010, 88, 393.
DHT = diesel hydrotreater (10) Zhang, Q.; Feng, X.; Liu, G.; Chu, K. H. A novel graphical
FCC = fluid catalytic cracking method for the integration of hydrogen distribution systems with
HCA = hydrogen cascade analysis purification reuse. Chem. Eng. Sci. 2011, 66, 797.
HCU = hydrocracker unity (11) Almutlaq, A.; Kazantzi, M.; El-Halwagi, M. M. An algebraic
HSD = hydrogen source diagram approach to targeting waste discharge and impure fresh usage via

12884 dx.doi.org/10.1021/ie300893d | Ind. Eng. Chem. Res. 2012, 51, 12877−12885

Industrial & Engineering Chemistry Research Article

material recycle/reuse networks. J. Clean Technol. Environ. Policy 2005,

7, 294.
(12) Foo, D. C. Y.; Manan, Z. A. Setting the minimum utility gas
flowrate targets using cascade analysis technique. Ind. Eng. Chem. Res.
2006, 45, 5986.
(13) Nelson, A. M.; Liu, Y. A. Hydrogen-pinch analysis made easy.
Chem. Eng. 2008, 115, 56.
(14) Ng, D. K. S.; Foo, D. C. Y.; Tan, R. R.; Pau, C. H.; Tan, Y. L.
Automatic targeting for conventional and bilateral property-based
resource conservation network. Chem. Eng. J. 2009, 149, 87.
(15) Pillai, H. K.; Bandyopadhyay, S. A rigorous targeting algorithm
for resource allocation networks. Chem. Eng. Sci. 2007, 62, 6212.
(16) Hallale, N.; Liu., F. Refinery hydrogen management for clean
fuels production. Adv. Environ. Res. 2001, 6, 81.
(17) Imran, A. M.; Megan, J.; Nan, Z. Multi-period hydrogen
management. Chem. Eng. Trans. 2009, 18, 743.
(18) Ng, D. K. S.; Foo, D. C. Y.; Tan, R. R. Automated Targeting
Technique for Single-Impurity Resource Conservation Networks. Part
1: Direct Reuse/Recycle. Ind. Eng. Chem. Res. 2009, 48, 7637.
(19) Ng, D. K. S.; Foo, D. C. Y.; Tan, R. R. Automated Targeting
Technique for Single-Impurity Resource Conservation Networks. Part
2: Single-Pass and Partitioning Waste-Interception Systems. Ind. Eng.
Chem. Res. 2009, 48 (16), 7647.
(20) Ahmad, M. I.; Zhang, N.; Jobson, M. Modelling and
optimization for design of hydrogen networks for multi-period
operation. J. Clean. Prod. 2010, 18, 889.
(21) Kumar, A.; Gautami, G.; Khanam, S. Hydrogen distribution in
the refinery using mathematical modeling. Energy 2010, 35, 3763.
(22) Liao, Z.; Wang, J.; Yang, Y.; Rong, G. Integrating purifiers in
refinery hydrogen networks: A retrofit case study. J. Clean. Prod. 2010,
18, 233.
(23) Ding, Y.; Feng, X.; Chu, K. H. Optimization of hydrogen
distribution systems with pressure constraints. J. Clean. Prod. 2011, 19,
204.
(24) Jia, N.; Zhang, N. Multi-component optimization for refinery
hydrogen networks. Energy 2011, 36, 4663.
(25) Jiao, Y.; SU, H.; Liao, Z.; Hou, W. Modeling and Multi-objective
Optimization of Refinery Hydrogen Network. Chin. J. Chem. Eng.
2011, 19 (6), 990.
(26) Fonseca., A.; Sá., V.; Bento., H.; Tavares., M. L. C.; Pinto., G.;
Gomes., L. A. C. N. Hydrogen distribution network optimization: A
refinery case study. J. Clean. Prod. 2007, 16, 1.
(27) Khajehpour, M.; Farhadi, F.; Pishvaie, M. R. Reduced
superstructure solution of MINLP problem in refinery hydrogen
management. Int. J. Hydrogen Energy 2009, 34, 9233.
(28) Liao, Z. W.; Rong, G.; Wang, J. D.; Yang, Y. R. Rigorous
algorithmic targeting methods for hydrogen networksPart I:
Systems with no hydrogen purification. Chem. Eng. Sci. 2011, 66, 813.
(29) Castro, P.; Matos, H.; Fernandes, M. C.; Pedro Nunes, C.
Improvements for mass-exchange networks design. Chem. Eng. Sci.
1999, 54, 1649.
(30) Gomes, J. F. S.; Queiroz, E. M.; Pessoa, F. L. P. Design
procedure for water/wastewater minimization: Single contaminant. J.
Clean. Prod. 2007, 15, 474.
(31) Borges, J. L. Diagrama de Fontes de Hidrogênio (Hydrogen
Source Diagram). M.Sc. Thesis, Universidade Federal do Rio de
Janeiro (Federal University of Rio de Janeiro), August 2009.

12885 dx.doi.org/10.1021/ie300893d | Ind. Eng. Chem. Res. 2012, 51, 12877−12885