By exploring strategies for process integration, such as waste-heat recovery and combined heat and power generation, plants can significantly improve their energy efficiency and reduce utility costs
In a chemical processing plant, utility costs typically account for about 10 to 20% of the total product cost . Advancements in systematic techniques for better utilization of energy, such as through process integration, had received recognition following the fuel crisis in the 1970s . Apart from economic reasons, improving energy management also brings environmental advantages, such as reductions in greenhouse-gas emissions.
Process integration can be defined as a holistic approach to design and operation that emphasizes the unity of the process [3, 4]. Since the 1970s, various process-integration tools have been developed for heat-exchanger network synthesis, as well as for the optimization of various energy-intensive processes, and are well documented in literature [5, 6]. Two such important tools showing the relationship between heat flow and temperature are the heat-recovery pinch diagram and the grand composite curve (GCC), both shown in Figure 1.
Note that in many cases, however, processes have been designed without much concern for energy utilization. Hence, the opportunities for energy optimization have never been fully explored. Through detailed inspection, many energy-saving opportunities can be easily identified, especially with regard to utilities. These activities lead to a more energy-efficient plant, with lower energy consumption and operating costs.
This article outlines the steps to evaluate an energy-intensive process to improve its energy utilization. To achieve this, example stream data were used for process integration. This example demonstrates how improvements in flowsheet design can help save energy. For instance, instead of supplying energy for process heating utilities, the process improvements identified ways that excess energy could be used for reactor integration, as well as options for combined heat and power (CHP) generation for the process.
Example — H2 production
The analysis outlined in this article relates to a hydrogen-production plant situated in a petroleum-refinery complex in Northwest England. However, the general principles will be applicable to many types of heat-exchanger networks.
At this site, hydrogen (H2) is produced for use in the refinery’s hydrotreating and hydrocracking units, and also for export to a nearby fertilizer manufacturing facility. The steam methane reforming (SMR) process is employed for the production of hydrogen. Figure 2 shows the process flow diagram (PFD) of the SMR plant.
A fresh feedstream of natural gas (NG) is mixed with a small portion of recycled hydrogen product before the mixture is pre-treated within the desulfurizer (DSU-001) and pre-reformer (R-001) units. In the latter unit, hydrogen sulfide, mercaptans and heavier hydrocarbons are removed. The desulfurized NG then enters the tubular furnace reactor (SMR-001), which is operated at 1,056°C. The NG serves two purposes in the SMR-001 unit: as supplementary fuel; and as a reactant for the steam-reforming process, along with steam. As a result, synthesis gas (syngas) is produced from the reaction between water and methane. The heat from the fluegas is recovered in the convective section of SMR-001 (down to 550ºC) before stack discharge.
Syngas effluent from SMR-001 is next cooled in a waste-heat boiler (WHB-001) before passing to a high-temperature water-gas shift (WGS) reactor (R-002) and then a low-temperature WGS reactor (R-003). In both reactors, carbon monoxide and water react to form carbon dioxide and hydrogen. The hydrogen-rich stream then undergoes air-cooling (in AC-001) in order to condense any water vapor, while its gas mixture passes through a pressure-swing adsorption (PSA) unit for the separation of hydrogen. The offgas from the PSA is sent to SMR-001 as fuel.
Figure 3(a) shows the GCC of the base case, which indicates that the minimum hot utility target (QH, min) of the process is identified as 4.83 MW, for a minimum approach temperature (ΔT) of 10°C, without any cold utility requirement (data for the heat integration study is given in Table 1). Note that this situation is termed as a “threshold problem” in process integration literature . Figure 3(b) shows the GCC for the process, following the modifications that are outlined in the following sections. These modifications reveal excess energy availability.
Energy reduction strategies
In general, the energy efficiency of a process can be improved by the following techniques :
- Reducing waste and losses
- Optimizing process operation
- Heat-recovery improvement
- Identifying process changes
- Optimizing energy-supply systems
These techniques are the underlying concepts used in developing the process-improvement strategies outlined in this article.
According to Ref. 6, heat recovery can be further increased by exploiting the process conditions that possess flexibility and freedom to be changed (within process limits). The SMR design evaluated in this article consists of several units that can be explored for improving energy efficiency, such as furnaces, boilers, reactors, stacks and so on. The following section describes four such strategies for enhancing heat recovery, leading to overall lower energy consumption. Stream data for these revised heat-integration steps are given in Table 2.
Strategy 1: Utilities. In the base case, medium-pressure steam (MPS) is fed to the deaerator (DEA) in order to remove the oxygen content of the demineralized (demin.) water before it is heated in heat exchangers HEX-001 and HEX-002, as well as the furnace convective section, in order to produce steam. However, an excess amount of steam is fed to the DEA, so it can be sold as superheated steam after further heating. Upon inspection, only 48% (12.6 ton/h) of the 26.5 ton/h of generated steam is actually required as a reactant in the plant.
Next, the boiler feedwater (BFW) that was originally used as the cooling medium in HEX-001 and HEX-002 is removed. Doing this leads to a lower demin. water flowrate, as it is no longer constrained by the heat-transfer requirement of the heat exchangers. The heated demin. water stream from the DEA is treated as individual cold streams for steam generation in the reactors (see Strategy 2 below for details). An extra heat exchanger (HEX-n001) is used for heating the demin. water from 30 to 90°C before it enters the DEA. Following the data-extraction heuristics in Ref. 6, HEX-001 and HEX-002 are extracted as hot streams, while HEX-n001 is extracted as a cold stream in pinch analysis. As a result, an extra 4.98 MW of excess energy is available for heat integration. These process modifications are shown in Figure 4.
Strategy 2: Reactor steam supply. In the base-case process, the saturated steam from the steam drum (shown in red in Figure 5) is mixed with the feed of the pre-reformer. Four reactions consume steam (R-001, SMR-001, R-002 and R-003). The saturated steam is supplied to the pre-reformer in excess such that no additional makeup is required in subsequent reactors. Consequently, a large proportion of steam is present in each of the reactor streams. The excess steam is heated and cooled along with the process fluid. This incurs a large energy demand on heating utilities, which can be avoided. In the revised design, steam is supplied separately to each reactor as feedstock with the distributed steam-supply system. This results in a reduction of 1.7 MW in the heating requirements for the reactor feeds.
Strategy 3: Natural-gas fuel consumption. The offgas from the PSA system is fed as fuel into SMR-001, with a portion of desulfurized NG used to satisfy the fuel deficit. The furnace operates at 1,056°C, while the combustion air enters the burners at 150°C. The design calls for extra NG to heat the fuel-air mixture in SMR-001.
To reduce NG consumption, the air stream is heated to 850°C to allow for better heat integration with other hot process streams. In other words, better heat recovery will take place to raise the temperature of this air stream. Doing so leads to less fuel being needed in the SMR-001 furnace — a decrease from 1,732 down to 292 kg/h. An additional benefit is the reduction of the stack loss, which leads to improved thermal efficiency in the furnace . Figure 6 shows the combustion air temperatures in both the base case, and following modified heating to improve heat integration.
Strategy 4: Fluegas exhaust temperature. In the base case, the hot fluegas from the SMR reactor is discharged to the stack after it is cooled to 550°C. Since this is a hot stream in pinch analysis, its discharge temperature should be reduced in order to maximize its heat-recovery potential. In other words, additional heat will be made available to heat up other cold streams in the process. Note also that the furnace efficiency can be improved when the stack exhaust temperature is lowered . However, the temperature should be kept higher than the acid dewpoint, where condensation takes place. For this process, the fluegas is to be cooled to 200°C in order to maximize its energy recovery. This leads to an additional 8.3 MW of energy being recovered from the SMR reactor’s hot fluegas. Figure 7 shows the implementation of this process modification.
Evaluate the results
The revised PFD (after all process modifications) is shown in Figure 8, and a summary of all four strategies’ results is given in Table 3. Note that this PFD is meant for use in energy optimization with pinch analysis. Performing pinch analysis following the standard procedure resulted in the GCC shown in Figure 3(b). Instead of needing heating utility input of 4.83 MW as in the base case in Figure 3(a), the process now can generate an excess energy of 11.70 MW, indicated by the cold utility target (Qc, min), as seen in Figure 3(b). In other words, the process is transformed from being energy-intensive into one with excess energy availability. It is worth noting that the removal of 4.83 MW of hot utility consumption results in an avoidance of 7.86 million ton/yr of CO 2 emissions (assuming natural gas is used). Figure 3(b) also shows that the process has a very high pinch temperature of 1,051°C. This further shows that the excess heat of the process is rejected from high-temperature sources. These high-grade waste-heat sources have the potential to be better utilized in the plant, which is discussed in the following section.
Reactor integration and CHP
The high-grade excess heat revealed by the revised process layout has various options for usage in the process. These include the heating of endothermic reactors, steam generation or power generation through a combined heat and power (CHP) scheme. Such measures not only save energy and reduce plant operating costs, but also enhance a company’s overall sustainability efforts. For the hydrogen-production unit described in this article, Figure 9 shows the various ways of making use of high-grade waste heat, as follows:
- Heat integration for the endothermic reactor (R-001)
- Steam generation (to be used as a reactant in R-001, R-002, R-003 and DEA)
- Steam generation to be used for power generation in the CHP scheme
Two scenarios are analyzed, with cost data given in Table 4. Note that the utility cost for the revised PFD in Figure 8 was determined to be£7,678,088/yr ($10,465,672/yr), inclusive of power (1.75 MW) and energy utilities. The latter includes cooling water and NG fuel needed for VHPS generation as reactant. In Scenario 1, all waste heat is used to generate 2.99 kg/s of very high-pressure steam (VHPS) to be used for power generation in a CHP scheme. The rationale behind Scenario 1 is to maximize the cogeneration potential before other design choices. The VHPS needed as reactant in R-001, R-002 and SMR-001 was purchased externally. The high-pressure steam (HPS, 0.50 kg/s) and MPS (0.38 kg/s) produced from the CHP scheme are used as reactant in R-003 and DEA, respectively. This scenario has a cogeneration potential of 0.48 MW, calculated using the model given in Ref. 8. The total project cost was determined to be£6,930,609/yr ($9,446,815/yr), which takes into account the capital costs for a new turbine and heat-recovery steam generator, as well as cost savings from power and revenue resulting from MPS (excess generation from CHP) that is sold to nearby plants. The total savings for this scenario is determined to be£747,479/yr ($1,018,856/yr), a 9.74% reduction compared to the total cost of the revised design.
For Scenario 2, the generated VHPS is also utilized for the integration of the endothermic reactor (R-001), as well as for VHPS used as reactant in reactors R-001 (0.31 kg/s) and R-002 (0.66 kg/s). This results in a much smaller cogeneration potential, of 0.164 MW (calculated using the model in Ref. 8), but also much lower NG fuel expenditure for the VHPS generation for the reactors. Note that the HPS and MPS produced from the CHP scheme are used as reactant in R-003 and DEA, as in Scenario 1. For this case, much higher total savings are obtained —£1,416,981/yr ($1,931,426/yr) — almost double that in Scenario 1. Hence, a significantly lower project cost of£6,261,107/yr ($8,534,246/yr) is achieved. Scenario 2 clearly demonstrates better economic performance among the two scenarios analyzed. The economic parameters of the scenarios are summarized in Table 5.
It is clear from these analyses that energy usage for a process plant can be further optimized in order to explore its additional benefits. Figure 10 displays the respective GCCs for Scenarios 1 and 2, illustrating the heat-flow effects of the various process modifications. In most cases, the excess heat may be used for power or steam generation, for instance through a CHP scheme. Process modifications may be carried out to improve energy efficiency of the plant and to lower its utility consumptions. As shown in this article, the application of a CHP system in the hydrogen-production plant brings an additional cost savings of£1,416,981 (over $1.9 million).
Edited by Mary Page Bailey
1. Peters, M. S., Timmerhaus, K. D. and West, R. E., “Plant Design and Economics for Chemical Engineers,” 5th ed., McGraw-Hill, New York, 2004.
2. Rossiter, A. P., “Energy Management and Efficiency for the Process Industries,” Wiley-AIChE, 2015.
3. El-Halwagi, M. M., “Sustainable Design Through Process Integration,” Butterworth-Heinemann. Kidlington, Oxford, U.K., 2012.
4. El-Halwagi, M. M. and Foo, D. C. Y., “Kirk-Othmer Encyclopedia of Chemical Technology,” John Wiley & Sons. Hoboken, N.J., 2014.
5. Linnhoff, B. and Boland, D., “A user guide on process integration for the efficient use of energy,” The Institution of Chemical Engineers (IChemE), Rugby, Warks, U.K., 1982.
6. Smith, R., “Chemical Process Design and Integration,” 2nd ed., John Wiley & Sons, Ltd., Chichester, West Sussex, U.K., 2017.
7. Zhu, F., “Energy and Process Optimization for the Process Industries,” Wiley-AIChE. Hoboken, N.J,, 2013.
8. Ng, R. T. L., others, Targeting for cogeneration potential and steam allocation for steam distribution network, Applied Thermal Engineering, Vol. 113, pp. 1,610–1,621, 2017.
Yee Heng Ang (Phone: +6012-4853450; E-mail: email@example.com) is a process engineer working at Schott Glass Malaysia specializing in the ultrasonic cleaning of glass wafers and thin glass. He obtained his integrated master’s degree in chemical and environmental engineering at University of Nottingham Malaysia. His research experience includes work on the development of thermoelectric materials for sustainable energy harvesting from recycled carbon fiber and annealed copper-zinc-tin-sulfide. He is also an Associate Member of the Institution of Chemical Engineers (AMIChemE).
Dominic C. Y. Foo is a professor of process design and integration at the University of Nottingham Malaysia (Broga Road, 43500 Semenyih, Selangor, Malaysia; Phone: +60(3)-8924-8130; Fax:+60(3)-8924-8017; Email: firstname.lastname@example.org). He is a professional engineer registered with the Board of Engineers Malaysia, a chartered engineer registered with the Engineering Council of the U.K. and a Fellow for the Academy of Science Malaysia. He developed various process-integration techniques for resource conservation, CO2 reduction and production planning. He has won several industry awards and has published more than 160 papers in a variety of engineering journals. He obtained his B.Eng., M.Eng. and Ph.D., all in chemical engineering, from Universiti Teknologi Malaysia.