Chemical Bandwidth Report
Chemical Bandwidth Report
Chemical Bandwidth Report
Preface
The U.S. Department of Energy, Office of Energy Efficiency and Renewable Energy (DOE/EERE), Industrial Technologies Program (ITP) supports research and development (R&D) to improve the energy efficiency and environmental performance of industrial processes. The program's primary role is to invest in high-risk, high-value R&D projects that will reduce industrial energy requirements while stimulating economic productivity and growth. ITP's Chemicals subprogram supports R&D relevant to the chemical industries. This study, which focuses on energy efficiency in the chemical industry, was initiated in FY2003 by Dr. Dickson Ozokwelu, Lead Technology Manager, ITP Chemicals subprogram to help guide research decision-making and ensure that Federal funds are spent effectively. The study was overseen by both Dr. Ozokwelu and Dr. Joseph Rogers of the American Institute of Chemical Engineers (AIChE), with analytical studies performed by Psage Research, LLC and JVP International. The intent of the study is to apply energy and exergy analysis to selected chemical manufacturing processes to determine sources of inefficiency and to locate potential process-specific areas for energy recovery. Front-end analysis was performed by Psage Research, LLC, using various software tools developed by Psage, Jacobs Engineering of the Netherlands, and AspenTech (Aspen Plus, and the AspenPEP library, a collaboration between AspenTech and SRI's PEP program). JVP International reviewed and further analyzed the results to prepare recommendations for future research. The study provides valuable insights into potential targets for the development and adoption of advanced, energyefficient technologies in chemicals manufacture. It will be an important tool at DOE for assessing future directions in chemicals R&D conducted under the ITP Chemicals subprogram. Paul Scheihing Team Leader, Chemicals and Enabling Technologies Industrial Technologies Program Office of Energy Efficiency and Renewable Energy U.S. Department of Energy
This summary is a condensation of a much larger work, and does not contain the comprehensive data sets generated in that effort. Questions concerning the original work or this summary report can be directed to the authors shown below. Dickson Ozokwelu U.S. Department of Energy Industrial Technologies Program 1000 Independence Ave, S.W. Washington, DC 20585 Phone: 202-586-8501 Email: dickson.ozokwelu@ee.doe.gov Joseph Porcelli JVP International, Incorporated 102 Lincoln Street Staten Island, NY 10314 Phone: 917-912-9804 Email: jvpii@jvporcelli.com Peter Akinjiola Psage Research, LLC 234 Village Walk Drive Macungie, PA 10314 Phone: 610-966-7106 Email: psageresearch@msn.com
Table of Contents
Introduction
Overview of Chemical Industry Energy Use ........................................................................ 1 Objectives of the Analysis .................................................................................................... 2
Methodology
Concepts of Energy Bandwidth, Energy and Exergy Analysis............................................. 3 Selection of Chemical Processes .......................................................................................... 5 Energy and Exergy Modeling Methodology......................................................................... 5 Model Integration............................................................................................................. 5 Unit Processes and Equipment......................................................................................... 8 Model Output................................................................................................................... 9 Limitations of the Approach ................................................................................................. 10
Summary of Results
Overview of Results.............................................................................................................. 11 Chemical Bandwidth Profiles ............................................................................................... 12 Ethylene .........................................................................................................................................13 Ammonia ........................................................................................................................................14 Ethylene Oxide ...............................................................................................................................14 Propylene .......................................................................................................................................15 Terephthalic Acid ...........................................................................................................................15 Methyl Tert-Butyl-Ether (MTBE) ...................................................................................................16 Formaldehyde.................................................................................................................................16 Methanol ........................................................................................................................................17 Acrylonitrile (ACN) ........................................................................................................................18 Styrene............................................................................................................................................19 Ethylbenzene...................................................................................................................................20 Nitric Acid ......................................................................................................................................21 p-Xylene..........................................................................................................................................21 Carbon Dioxide ..............................................................................................................................22 Vinyl Chloride ...............................................................................................................................22 Acetic Acid......................................................................................................................................23 Butadiene........................................................................................................................................23 Cumene...........................................................................................................................................24 Energy and Exergy Losses in Unit Operations .................................................................... 25
Introduction
Overview of Chemical Industry Energy Use
Chemicals manufacture is the second largest energy-consuming enterprise in U.S. industry, accounting for over 6.5 quadrillion Btus (quads) of feedstock and process energy use in 2002, or nearly a third of industrial energy use [ACC 2003]. More than half of the energy used by the chemical industry is used as feedstocks (Figure 1). The other half is primarily used to provide heat, cooling, and power to manufacturing processes, with a small amount used for conditioning and lighting buildings. The chemical industry has achieved significant energy efficiency gains since the 1970s, precipitated by the Middle East oil crises and resulting pressures on Other Natural Gas 426 TBtu energy supply. Between 1974 and 1990, Coal 680 TBtu Natural Gas NGL/LPG 271 TBtu fuel and power consumed per unit 1901 TBtu 1497 TBtu output in the industry has decreased by Heavy Liquids Electricity 1132 TBtu 525 TBtu nearly 40% (see Figure 2). However, as Figure 2 illustrates, efficiency improvements have not been as impressive since the early 1990s, and have remained relatively flat over the Total Feedstock Energy Total Fuel and Power last five years. Further improvements in 3342 TBtu 3182 TBtu energy efficiency will be necessary for Figure 1. Energy Use in the U.S. Chemical Industry, 2002 [ACC 2003] the industry to maintain a competitive edge.
Fuel Oil/LPG 58 TBtu Coal 32 TBtu
The chemical industrys dependence on energy for raw materials as well as fuel and power makes it particularly vulnerable to fluctuations in energy price. High fuel and feedstock prices can have a profound effect on chemical processing, which typically requires large amounts of energy to convert raw materials into useful chemical products. Recent spikes in natural gas price, for example, caused temporary plant shutdowns of gas-based cracking facilities in some regions of the country. Petroleum and natural gas price increases continue to create price uncertainties in commodity chemical markets, and are a key driver for olefins pricing [CMR 2004].
120 100
Chemical Industry
80 60 40 20 0 1970 1974 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002
U.S. Industry
As energy prices continue to rise and supplies become more volatile, chemical companies are increasingly looking toward energy efficiency as a way to reduce production costs and improve their competitive edge. The challenge for todays chemical manufacturers is to effectively focus their resources on improving the equipment and processes that will produce the greatest benefits in energy use, productivity, and yield.
Methodology
Concepts of Energy Bandwidth, Energy and Exergy Analysis
Energy bandwidth analysis provides a snapshot of the energy losses that can potentially be recovered through improvements in technology, process design, operating practices, or other factors. Bandwidth analysis quantifies the differences between these measures: theoretical minimum energy (often called asbolute minimum) that is required for a process; practical minimum energy that is required for a process, given irreversibilities and other limitations; and current energy requirement for an individual process, based on average values in todays manufacturing environment.
The theoretical minimum energy is based solely on chemical conversion reactions. It represents the energy required to synthesize the product in its standard state, at 100% selectivity, from the raw materials in their standard states, disregarding irreversibilities. In reality, the energy consumed by a process must exceed the theoretical minimum energy due to the non-standard conditions of reactions, products, and reactants; the formation of by-products; the need to separate products; and other factors. These conditions impose limitations that make it impossible to operate at the theoretical minimum. This higher energy requirement is sometimes referred to as practical minimum energy.
Energy lost to inefficient equipment, poor design, limited heat recovery, other factors Energy lost to process irreversibilities, nonstandard conditions, byproducts
Energy based on ideal chemical reactions, standard state, 100% yield, no irreversibilities
In reality, most chemical processes use significantly more energy than the practical minimum energy requirements due to energy losses. These external energy losses are due to many factors, including inherently inefficient or outdated equipment and process design, inadequate heat recovery, poor integration of heat sources and sinks, poor conversion selectivities, and a host of other site-specific issues. Energy required under actual plant conditions, current energy, invariably exceeds the practical minimum because of these external losses.
From an energy efficiency perspective, the difference between current energy and theoretical minimum energy is an important Figure 3. Depiction of Energy Bandwidth portion of the bandwidth diagram shown in Figure 3. This difference represents the greatest potential target for reduction in energy demand or energy recovery and reuse. However, without examining the quality of the energy in this band, it is difficult to credibly determine how much of that energy it is practical to recover under realistic plant operating conditions. In addition, it is not practical or economically feasible to reduce all process irreversibility-related losses or the inefficiencies. This is where exergy analysis can significantly assist in pinpointing opportunities.
Manufacturing Process
Today engineers and scientists often use enthalpy (a thermodynamic quantity equal to the amount of energy in a system) or energy balances to evaluate the performance of chemical production processes and quantify energy losses. However, this approach does not consider the quality of the energy lost or the Chemical Bandwidth Study Summary Report 3
actual energy potential associated with process streams. Using enthalpy, for example, 1,000 Btu/hour of low-pressure steam would compare equally with 1,000 Btu/hour of electricity. In reality, the amount of usable energy from the low-pressure steam is less than a third of that represented by the electricity, because the energy quality of the low-pressure steam is much lower.
Exergy.
is defined as the maximum amount of work that can be extracted from a stream as it flows toward equilibrium. This follows the 2nd Law of Thermodynamics, which states that not all heat energy can be converted to useful work. The portion that can be converted to useful work is referred to as exergy, while the remainder is called non-exergy input.
Exergy analysis provides a powerful tool for assessing the quality of energy and quantifying the portion of energy that can be practically recovered. Exergy analysis uses parameters such as temperature or pressure to determine energy quality and calculate potentially recoverable energy. Exergy, or energy quality, diminishes each time energy is used in a process. For example, a large percentage of energy content can be extracted from flowing steam at high temperatures. As the steam temperature drops (e.g., after passing through a heat exchanger), the percentage of energy that can be recovered is reduced. This drop in energy quality is referred to as a loss of exergy or energy degradation.
The concept of exergy and energy quality as applied to a chemical process is depicted in Figure 5. Total energy input (QIN) is comprised of both exergy and non-exergy input. During the process, total energy input is converted to some useful work (QW), while some is lost due to internal and external energy loss factors (QLOSS). The non-exergy component of total input energy has zero quality and is rejected (QREJECT). It is assumed that it is technically feasible to recover some portion of QLOSS (the exergy component).
Exergy
Figure 4 illustrates the change in energy quality with temperature drop. At 110oF, little recoverable energy (exergy) is available when compared with the same stream available at higher temperatures. Exergy analysis also quantifies energy that cannot practically be recovered and accounts for non-standard conditions and irreversibilities. It does not, however, take into account the economic feasibility of energy recovery.
1000oF
500oF
2 200 00oF
110oF
PROCESS
QLOSS =Non-Product Effluents (external exergy loss) + Process Irreversibilities (internal exergy loss) (Recoverable Energy) QIN = QW + QLOSS (exergy component) + QREJECT (non-exergy component)
Figure 5. Concept of Exergy in a Chemical Process Chemical Bandwidth Study Summary Report 4
EXERGY TOTAL ENERGY INPUT LOST EXERGY (RECOVERABLE AVAILABLE EXERGY) LOST EXERGY (RECOVERABLE AVAILABLE EXERGY) NONEXERGY INPUT
NONEXERGY INPUT
External exergy losses are embodied in non-product effluents such as vent gases, byproducts and waste water. Internal exergy losses are due to process irreversibilities. Energy quality is the ratio of exergy content to energy content. For this analysis, streams with a quality greater than 20% were assumed to be economically recoverable; between 5-20% some energy might be economically recoverable; below 5% energy recovery might not be likely.
Figure 6 shows qualitatively how exergy relates to the energy bandwidth shown in Figure 3. The exergy and non-exergy input shown in bar B represent all the energy inputs. C shows the breakdown of input process energy into theoretical minimum requirements, recoverable energy and non-exergy components. Bar D illustrates processes that operate with actual process energy requirements; input process energies are higher than the theoretical minimum and recoverable energies are therefore lower.
a computer program developed by Psage Research that interfaces with the AspenPlus and ExerCom models and calculates energy and exergy balances around each unit operation. Table 1. Chemicals Selected for Energy and Exergy Analysis
Chemical
Ethylene Propylene Ammonia MTBE Vinyl Chloride Carbon Dioxide Nitric Acid Ethylbenzene Styrene Terephthalic Acid Formaldehyde P-Xylene Ethylene Oxide Cumene Methanol Acetic Acid Butadiene Acrylonitrile TOTALS
% of Top 50 Chemicals
5.9 3.6 3.3 2.3 2 1.8 1.7 1.4 1.3 1.1 1.1 0.9 0.8 0.9 0.8 0.5 0.5 0.3 30.2%
A schematic of the modeling approach is shown in Figure 7. Process modeling was first accomplished using the Aspen Plus model, drawing on process modules available from the SRI Consulting/Aspen Process Economics Program (PEP) library. This library utilizes information from public sources and inhouse engineering expertise to reproduce the technology of a particular licensor, plant operator, or research organization. Currently this library only contains 15 technologies from the list of top 50 chemicals, although more are expected to be added. Processes for which AspenPlus models were not available from the SRI PEP library were developed using data from open literature sources. The ApsenPlus models provide the process energy and material balances for the selected chemical manufacturing processes.
AspenPEP Library Database of SRI process models built into AspenPlus flowsheet models AspenTech AspenPlus 11.1 version Steady-State Process Models ExerCom (Jacobs Engineering) PsageDeveloped Software
Energy/Exergy of Streams
Process Modeling
Chemical Bandwidth Study Summary Report
Figure 7. Integrated Modeling Approach Used for Exergy and Energy Analysis 6
As shown in Figure 7, after process models are developed the ExerCom model uses the output of the Aspen Plus simulation and internal databases of standard chemical exergies and enthalpies to compute the exergy and energy of all of the liquid and gaseous material streams. ExerComs internal databases contain thermodynamic data for a limited number of chemical species, requiring that some data be calculated for those that are missing. The exergies of heat, work and solid streams must also be calculated manually. In the last phase of modeling, the Psage-developed computer program interfaces directly with the AspenPlus and ExerCom results to calculate the exergies of heat, work, and solid streams around individual process units and for the overall process model. Exergies of heat streams not calculated by ExerCom are computed from enthalpies using the Carnot quality factor (c). Where model boundaries do
not include all exergy inflows (mainly refrigeration and separation units), values must be estimated based on known exergetic efficiencies of similar unit operations.
Heat Exchangers
Process Furnaces
Model Output
The integrated modeling approach produces a number of energy and exergy quantities for each chemical process and individual unit operation (see Table 4). These quantities provide a process efficiency baseline against which new or improved technologies can be compared. A key output is the potentially recoverable energy (QLOSS), which can be used to establish potential for improved efficiency. Table 4. Model Outputs
Total Process Energy Input (QIn) all energy inputs to the process regardless of quality Total Process Exergy input (TPEI) the component of input energy that can be converted to work or recovered Actual Process Exergy (QW) the component of input exergy that is converted to useful work Theoretical Minimum Process Energy (TMPE) the minimum amount of energy required for the process based on chemical reactions and ideal or standard conditions and 100% yield External Exergy Loss (EEL) potentially recoverable energy in the form of non-product effluents such as steam and wastewater Internal Exergy Loss (IEL) potentially recoverable energy lost throughf process irreversibilities Potentially Recoverable Available Energy (QLoss) the sum of recoverable energies (IEL and EEL)
A sample output of the analysis is provided in Figure 8 for the process model based on production of vinyl chloride monomer (VCM) from ethylene dichloride. The analysis illustrates that a portion (about 8%) of the energy input to this endothermic process is available downstream as recoverable energy. Analysis of the performance of individual unit operations within each process helps to pinpoint the locations of energy and exergy losses in each of the processes. For production of vinyl chloride monomer, for example, the analysis revealed that the largest source of energy and exergy losses were due to vaporizing ethylene dichloride, the endothermic furnace reactor (rapid quench), low temperature distillations, and separation of hydrochloric acid (HCl).
Chemical reactions can be either endothermic (heat absorbing) or exothermic (heat generating). For endothermic reactions, the energy converted to useful work (QW) will be shown with an arrow flowing away, indicating that it was absorbed by the process. For exothermic reactions, QW will be shown with an arrow flowing toward the process, indicating additional heat energy has been generated. Figure 8 is an example of an endothermic process. The modeling outputs for each of the chemical technologies selected can be evaluated in various ways to identify sources of inefficiency and potential improvement targets. Energy and exergy losses, for example, can be sorted and ranked across all the chemical technologies by the same common unit operations to reveal energy efficiency trends and provide further focus for targeting research.
QW, Useful Work= 147 (Chemical Conversion) QIN Input = 2,671 = 975
Energy Exergy
PROCESS
QREJECT = 1,696
Figure 8 Sample Output of Energy and Exergy Analysis for Vinyl Chloride Monomer (Values in Btu/lb of VCM)
In addition, results do not reflect external factors that may influence plant performance. For example, large capital assets that could be improved may not be replaced until they reach the end of their useful life, regardless of the potential benefits. Environmental regulations or other factors (permitting, site limitations) may also have an impact on the feasibility of reducing energy and exergy losses. While potentially recoverable energy does provide a good perspective on efficiency opportunities, the analysis does not provide insight on the true economic feasibility of recovering energy. For example, economic factors such as limited funds for plant upgrades, poor markets, corporate investment philosophy and the high cost of environmental compliance could all have an impact on economic feasibility. However, the quality factor inherently takes into consideration that low quality energy is probably not economically suitable for recovery and uses this as a measure of recoverable energy. Despite the potential for discrepancies between the study results and actual plant performance, this analysis remains a powerful tool for pinpointing targets for improvement, provided the limitations are kept in mind.
10
Summary of Results
Overview of Results
An energy and exergy diagram similar to that shown in Figure 8 was developed for all 25 of the process technologies studied, including multiple technologies for some products. The results of this analysis are shown in Table 5. There is great potential for energy recovery in the chemical processes analyzed. The total potentially recoverable energy identified for the 25 processes studied is nearly 900 trillion Btus (using average values for multiple technologies when applicable). Recoverable energy is assumed to be of high enough quality to warrant recovery, regardless of economic feasibility. Table 5. Energy and Exergy Analysis Results for 25 Chemical Technologies (Btu/lb)
Process Ethylene (Braun) Ethylene (Kellogg) Ammonia Ethylene Oxide Propylene Terephthalic Acid MTBE Methanol (ICI LP) ACN From Propane Methanol (Lurgi) Formaldehyde ACN From Propylene Nitric Acid Styrene (Fina/Badger) Ethylbenzene (Lummus) Styrene (Lummus) Ethylbenzene (Mobil/Badger) p-Xylene (Isomerization) Carbon Dioxide Vinyl Chloride Acetic Acid Cumene (AlCl3 Cat.) Cumene (Zeolite Cat) Cumene (SPA Cat) Butadiene N/A * Total Energy Input QIN 8,656 8,139 4,596 7,741 4,548 1,919 8,868 4,883 5,381 2,273 698 4,364 232 3,365 1,528 4,703 1,787 3,228 2,083 2,671 1,612 1,124 Process Exergy Input 5,534 5,035 3,543 5,735 3,047 1,157 2,572 871 1,392 841 115 1,020 207 1,122 1,147 1,697 965 1,702 508 975 786 440 Actual Process Exergy QW 326 217 -351 -6,720 1,440 4,730 -135 -4,546 -13,152 -4,132 -3,209 -8,015 -1,401 369 -231 305 -236 -133 -426 147 -512 -240 Theoretical Minimum Energy 650 650 414 734 846 3,047 124 802 5,509 802 802 4,355 1,953 340 273 340 273 5 N/A 142 436 526 Recoverable Energy QLOSS 5,208 4,818 3,967 12,456 2,119 5,887 2,706 5,417 14,544 4,974 3,324 9,035 1,609 1,491 1,363 1,392 1,282 1,835 935 828 1,297 680 Ratio of QLOSS/QIN * 60% 59% 86% 161% 47% 307% 31% 111% 270% 219% 476% 207% 694% 44% 89% 30% 72% 57% 45% 31% 80% 61% Recoverable Energy (TBtu/yr) 271.3 251 115.0 98 67.4 55.3 53.6 39.5 39.3 36.3 30.9 24.4 24.0 16.4 16.2 15.3 15.3 15.2 14.5 14.3 6.2 5.2
-248 -245 55
Exothermic reaction, net chemical conversion exergy inflow A separation process without chemical reaction Ratios may be higher than 100% because the input energy does not include heat generated by exothermic reactions.
11
Trillion Btus/Yr
300 250 200 150 100 50 0
nol F orm rmald ldehy de Eth t hylb lbenz ene St yre rene
Et th hylle ene Pr ro opy lle ene T er re epht th hali lic Ac id id
p-X -Xy le lene C arb rbon D ioxid io ide Vin iny l C hlo lori ride
Am mo niia a
Figure 9. Comparison of Recoverable Energy Across Chemical Technologies A comparison of the recovery energy potential for each chemical product is shown in Figure 9, in descending order. Where multiple technologies were evaluated the average values for recoverable energy were used. Ethylene, ammonia, ethylene oxide, propylene, terephthalic acid and MTBE exhibit the largest potentials for energy recovery in terms of trillion Btus. Many of the energy losses are associated with waste emissions such as cooling water, air and purge streams, and by-product streams. However, exergy analyses has revealed that such streams may not always contain sufficient recoverable energy to justify energy recovery strategies. Exergy losses associated with waste recovery boilers and throttling can also be significant. Irreversibilities (or internal exergy losses) in the technologies studied were prevalent in furnaces, high temperature reactors, cooling of high temperature reactor effluents, refrigeration, and refrigerated separations.
Ac eti tic Ac id id
M eth tha
N itri ric
C ume ne
AC N
M TBE
Ac id id
Ethylene
Cracking of Propane This process uses light hydrocarbons such as propane or ethane derived from natural gas liquids as a feedstock. However, with the rise in natural gas prices, most new ethylene capacity is being based on cheaper naphtha or gas oil feeds. Total process energy required is about 13 times greater than the theoretical minimum. The greatest sources of energy-exergy losses include high temperature cracking, quenching of cracked products, and complex low-temperature separations of products and co-products. Heat exchangers (process exchangers, interstage coolers, quenching exchangers) and distillation columns (e.g., C2 splitter) comprise the majority of high energy-consuming equipment. Losses arise primarily from differing temperatures, compositions and pressures of various streams. Virtually all exergy losses in cracking and quenching are due to the quenching exchangers, which sequentially quench the reaction product gas. The C2 splitter contributes to exergy losses in product separation. About 40% of energy is lost to gas refrigerated cooling, and another 27% is lost to cooling water during interstage gas compressor cooling. The quality of recoverable energy is high enough to generate high-medium pressure steam.
Btu/lb 9000
8000 7000 6000 5000 4000 3000 2000 1000 0 Ethylene (Braun)
Theoretical Minimum Recoverable Energy Input Energy
Ethylene from Propane (Braun) Process Sub-Section Cracking And Quenching Compression And Deacidification Deethanization Demethanization Product Separation Heat & Refrigeration Recovery TOTALS
Energy Loss Btu/lb 425 1879 1294 1402 1296 761 7055
% 6 27 18 20 18 11
External Exergy Loss Btu/lb 38 208 231 554 255 520 1806
Internal Exergy Loss Btu/lb 1046 590 65 1106 306 287 3402
Total Exergy Loss Btu/lb 1084 798 296 1660 561 808 5208
% 21 15 6 32 11 16
Cracking of Naphtha This process is based on naphtha or gas oil, the feedstock chosen for most new plants in the U.S. today. As opposed to the Braun process, the highest energy-exergy consumption is concentrated in the front end of the process. The Kellogg process is more exothermic, and requires less input process energy but exergy losses are double in cracking and quenching due to the higher compression ratio used (525 psia versus 140 psia for Braun). Total process energy required is about 12 times greater than the theoretical minimum. Substantial losses occur in the demethanizer column due to the condenser, where the coolant is ethylene refrigerant. Another significant source of losses is cracking and quenching, mostly due to the cracking furnaces and the large towers where temperature differences create exergy losses. The cracked gas compressor interstage coolers are large sources of losses.
Btu/lb 9000
8000 7000 6000 5000 4000 3000 2000 1000 0 Ethylene (Kellogg)
Ethylene from Naphtha/Gas Oil (Kellogg) Process Sub-Section Cracking And Quenching Compression And Deacidification Demethanization Deethanization Product Separation Heat and Refrig. Recovery TOTALS
Energy Loss Btu/lb 1851 2958 1712 335 109 842 7807
% 24 38 22 4 1 11
Internal Exergy Loss Btu/lb 1678 540 359 336 184 554 3651
Total Exergy Loss Btu/lb 1861 748 974 396 201 638 4818
% 39 16 20 8 4 13
13
Ammonia
This process is based on a composite of current technologies which have been operating for many years but have seen improvements in catalysts, synthesis upgrading and energy recovery. Total process energy is about 11 times greater than the theoretical minimum energy requirement for the exothermic ammonia conversion reaction. The synthesis gas separator is the largest source of energy-exergy losses (hot exit carbon dioxide stream, exchanger cooling of MEA). The next largest source of energy loss is ammonia synthesis, occurring in the high pressure syngas compressor, syngas reactor, and cooling and refrigeration units. Much of the loss is low-quality energy due to low temperature levels. In preheating and reforming, large internal exergy losses occur in the secondary reformer and waste heat boiler downstream of the reformer. These losses occur due to large temperature gradient-driven heat transfer operations. Considerable waste heat recovery is already used.
Btu/lb 5000 4500 4000 3500 3000 2500 2000 1500 1000 500 0
Theoretical Minimum Recoverable Energy Input Energy
Ammonia from Natural Gas Process Sub-Section Preheating/Reforming Shift Gas Upgrading Ammonia Synthesis Heat Recovery TOTALS
% 10 0 49 36 5
Total Exergy Loss Btu/lb 1478 164 796 1350 106 3893
% 38 4 20 35 3
Ammonia
Ethylene Oxide
This is an exothermic process based on the Shell process for direct oxidation of ethylene with oxygen (various others are used commercially). Most ethylene oxide plants also produce ethylene glycol in an integrated flowsheet. The process as modeled has half of the product ethylene oxide as an aqueous stream. This has contributed to unusually high energy and exergy losses. In addition, the process as modeled couples the upstream stripping column condenser with the purification column condenser, creating a very large condensing load at too low a temperature for energy recovery. This may not be the common practice. Total process energy is about 10 times theoretical minimum energy requirements. The ethylene oxide purification unit accounts for 91% of energy losses (19% of exergy losses). In the stripper section, high internal exergy losses are due to heat exchangers and columns in the recirculating water loop. Relatively low temperatures result in little opportunity for heat recovery. Internal losses could be reduced by increasing the areas of the heat exchangers. Large internal losses in the reactor section area due to large temperature differentials between the inlet gas and exothermic conditions in the reactor.
Btu/lb
14000 12000 10000 8000 6000 4000 2000 0 Ethylene Oxide
Theoretical Minimum Recoverable Energy Input Energy
Ethylene Oxide (Shell) Process Sub-Section Feed Pre-Heat Reactor EO Absorber EO Stripper EO Purification TOTALS
% 2 0 0 7 91
Internal Exergy Loss Btu/lb 1010 4163 885 3705 596 10360
Total Exergy Loss Btu/lb 1039 4163 885 3982 2388 12456
% 8 33 7 32 19
14
Propylene
This endothermic process is based on the Fina technology for production of propylene from light naphtha fractions (described in the patent literature but not yet commercialized). Most propylene is now produced as a co-product of ethylene in naphtha crackers, and it is uncertain if dedicated production of propylene from naphtha will ever be commercially popular. It is included here to provide a perspective on innovation. Total process energy is about 5 times greater than the theoretical minimum (ethylene production is 12-13 times greater than theoretical minimum). Most energy losses occur during production separation, mostly due to debutanizer column and coolers. Some level of energy recovery may be possible in this section. The largest exergy loss occurs in the reactor subsection, mostly occurring in the feed preheater, coolers and the reactor. Large internal losses in this section are due to wide differences in input and output stream temperatures.
Btu/lb 5000 4500 4000 3500 3000 2500 2000 1500 1000 500 0
Theoretical Minimum Recoverable Energy Input Energy
Propylene from Naphtha Process Sub-Section Reactor Product separation Product purification By-products TOTALS
Propylene
% 7 74 16 2
% 72 23 2 3
36 33 1 30
73 10 8 10
15
% 3 1 96
% 6 1 93
Formaldehyde
This exothermic process produces formaldehyde from methanol using a silver-based catalyst, and is based on BASF technology. Total process energy input is about 4 times greater than theoretical minimum energy requirements. In this relatively simple process configuration, the low temperature quench of the reactor effluent is responsible for most of the energy consumption as well as energy and exergy losses. The very large driving forces around the exothermic reactor contribute to the substantial internal exergy losses.
3500 3000 2500 2000 1500 1000 500 0 Formaldehyde
Theoretical Minimum Recoverable Energy nput Energy gy Input
Btu/lb
Formaldehyde from Methanol Process Sub-Section Feed Air/Recycle Air Mixer Feed Air Compressor Feed Heater Reactor Absorber Recycle Air Purge TOTALS
16
Methanol
ICI LP This exothermic process is based on the ICI low-pressure technology that includes steam reforming of natural gas, high pressure synthesis of methanol, and distillation for product recovery and separation. The total process energy input is about 6 times that of the theoretical minimum energy required. The sources of large exergy losses in the refining section are a primary distillation column and large heat exchanger. The reforming furnace experiences large internal exergy losses primarily due to large differences in the temperatures of inlet and effluent streams and combustion gases.
Btu/lb 6000
5000 4000 3000 2000 1000 0 Methanol (ICI LP)
Theoretical Minimum Recoverable Energy Input Energy
Methanol from Natural Gas (ICI LP) Process Sub-Section Reforming section Synthesis section Refining section TOTALS
% 2 3 95
% 52 10 38
LurgiTechnology This process varies from the ICI LP process in that it utilizes a combined reforming process with two stages of reforming in series, the second with oxygen injection. Process energy input is about 3 times greater than the theoretical minimum energy requirement. The heat recovery section exhibits the greatest external exergy losses and indicates the potential for significant energy recovery if low temperature users were available. The methanol column in the refining section also makes a large contribution to energy losses, although it is lower than the similar column in the ICI process. Condenser steam generation could reduce energy losses. Relatively large losses are also attributed to a process exchanger and combustion furnace in the reforming section, and to methanol reactors, condensers and air coolers in the synthesis section. The process exchanger is a candidate for steam generation with substantial energy recovery. The combustion furnace has a lower energy loss but the high external energy ratio suggests the possible use of a waste heat boiler to recover energy. Exergy losses occur in the reforming and heat recovery sections due to the wide range of inlet and outlet temperatures involved. Preheating the feed within the reactor system is the source of large internal exergy losses in the synthesis section due to large temperature differences.
External Energy Loss Btu/lb 608 471 164 922 2165 Internal Exergy Loss Btu/lb 2131 489 98 99 2816 Total Exergy Loss Btu/lb 2739 960 262 1021 4982
Btu/lb 5000 4500 4000 3500 3000 2500 2000 1500 1000 500 0
Methanol from Natural Gas (Lurgi) Process Sub-Section Reforming section Synthesis section Refining section Heat recovery TOTALS
% 20 25 16 39
% 55 19 5 20
Methanol (Lurgi)
17
Acrylonitrile
SOHIO-BP Propylene Ammoxidation This exothermic process is based on the SOHIO-BP fluidized bed ammoxidation process, which is now used predominantly for the production of acrylonitrile (ACN). The total process energy input is approximately the same as the theoretical minimum energy required (considerable energy is produced by the exothermic reaction). About 45% of energy losses are recoverable heat and refrigeration of process effluent streams. The largest energy and exergy losses occur in the heat and refrigeration section, primarily due to effects of refrigeration cycles needed to separate the product and byproducts at low temperatures. A large source of losses in the ammoxidation section is the quench column overhead cooler, although most exergy losses occur as internal losses in the ammoxidation reactors due to the large number of input and output steams at widely different temperatures. Increased heat exchange to increase the cold feed temperatures could reduce these irreversibilities, if economic. Most of the losses in the acrylonitrile separation are due to the HCN stripper column and condenser, which is cooled with refrigeration and is very energy-intensive.
Btu/lb 10000 9000 8000 7000 6000 5000 4000 3000 2000 1000 0
Acrylonitrile by Propylene Ammoxidation (SOHIOBP) Process Sub-Section Propylene Ammoxidation Acrylonitrile Separation Heat & Refrigeration Recovery TOTALS
% 24 31 45
% 51 22 27
SOHIO-BP Propane Ammoxidation This exothermic process has not been commercialized. The total process energy input is approximately the same as the theoretical minimum energy required (considerable energy is produced by the exothermic reaction). It is similar to the propylene ammoxidation process although reaction system performance is different, leading to changes in downstream processing. Differences in energy and exergy input and losses are related to differences in feedstock, products and byproduct yields. Energy losses are similar to those of the process above. The propane process input energy is not only higher than that for propylene, but the quality of input energy (exergy/energy) is also higher. The higher selectivity of the propylene process leads to lower exergy losses in production, separation and cooling of fewer byproducts. The propane process requires more energy for compression and byproduct separation. Being more exothermic, the propane process also offers more opportunity for steam generation, but is subject to more process irreversibilities due to conversion at higher temperatures. A key difference is in the ammoxidation reactors, where internal exergy losses of the process are almost double. This reflects the larger gas circulation rate caused by lower per pass conversion.
Btu/lb 16000
14000 12000 10000 8000 6000 4000 2000 0 Acrylonitrile (from Propane)
Theoretical Minimum Recoverable Energy Input Energy
Acrylonitrile by Propane Ammoxidation (SOHIOBP) Process Sub-Section Propane Ammoxidation Acrylonitrile Separation Heat & Refrigeration Recovery
% 27 36 37
% 70 14 16
TOTALS
14823
3129
11415
14544
18
Styrene
Lummus/Monsanto/UOP This process is based on production of styrene via adiabatic dehydrogenation of ethylbenzene, an endothermic reaction. Total process energy input is about 13 times greater than the theoretical minimum energy requirement. The largest energy losses occur in air coolers, primarily due to inlet and outlet temperature differences. Generating low-pressure steam could reduce these losses, if an economic use for the steam could be identified. Large exergy losses are also found in the feed preheat section, where superheated steam is mixed with fresh and recycle ethylbenzene at lower temperatures. Another source of losses is the steam superheater. Other losses are found in strippers and fractionators and are due primarily to large temperature differences leading to process irreversibilities.
External Exergy Loss Btu/lb 17 0 0 408 5 21 26 0 1 478 Internal Exergy Loss Btu/lb 11 193 371 142 101 64 17 15 1 914 Total Exergy Loss Btu/lb 28 193 371 550 106 85 43 16 1 1392
Btu/lb 5000 4500 4000 3500 3000 2500 2000 1500 1000 500 0
Styrene (Lummus)
Styrene (Lummus/Monsanto/UOP) Process Sub-Section Steam Compressor Steam Super heater Feed- Preheat/Reactor Air Coolers Condensate Recovery EB/Styrene Stripper Styrene Fractionator EB Stripper Benzene/Toluene Stripper TOTALS
% 0 0 0 84 2 6 6 1 0
% 2 14 27 39 8 6 3 1 0
Styrene (Fina/Badger) This endothermic process is very similar to the Lummus process, except for modest differences in the reactor section. Total process energy input is about 10 times greater than the theoretical minimum. The largest energy losses are found in air coolers used to condense and cool the reactor effluent, although the quality of the energy lost is relatively low. The feed preheat is the source of significant exergy losses due to high temperature differences in reactor effluent exchangers and in the dehydrogenation reactors. Losses also occur in the ethylbenzene/styrene stripper column, which must be operated under vacuum, and the large condenser load is removed with cooling water at too low a temperature for heat recovery. The Lummus process uses a higher steam/ethylbenzene ratio than the Fina/Badger technology, and requires higher energy input. However, the Lummus process recovers low-temperature heat from the ethylbenzene/styrene stripper and exergy losses are lower than those in the same Fina/Badger unit operation.
Btu/lb 3500
3000 2500 2000 1500 1000 500 0 Styrene (Fina/Badger)
Theoretical Minimum Recoverable Energy Input Energy
Styrene (Fina/Badger) Process Sub-Section Steam Compressor Steam Super heater Feed- Preheat/Reactor Air Coolers Condensate Recovery EB/Styrene Stripper Styrene Fractionator Benzene/ Toluene Stripper
% 0 0 0 51 2 37 6 3
% 0 2 51 14 6 20 4 3
TOTALS
5293
410
1081
1491
19
Ethylbenzene
Mobil/Badger This exothermic process is based on production of ethylbenzene (EB) via vapor phase benzene alkylation. The total process input energy is about 8 times greater than the theoretical minimum energy. Note that nearly all EB production is integrated downstream with styrene production, and synergies between the two production units are not captured in the stand-alone model used for this analysis. The model also simplifies some of the aspects of the primary reactor and feed preheat to the secondary reactor (an energy-saving feature). The benzene fractionator is the largest source of energy and exergy losses, due to large temperature differences in cooling reactor effluent with incoming feed. There is potential for steam recovery and export in this section. Other sources of losses include the ethylbenzene fractionator and ethylbenzene reactor. The balance of the process is relatively energy-efficient.
Btu/lb 1800
1600 1400 1200 1000 800 600 400 200 0 Ethylbenzene (Mobil/Badger)
Theoretical Minimum Recoverable Energy Input Energy
Ethylbenzene (Mobil/Badger) Process Sub-Section Benzene Fractionator Primary Reactor Ethylbenzene Fractionator Poly Ethylbenzene Fractionator Secondary Reactor Pre-Fractionator
% 82 0 15 1 0 3
% 80 8 8 1 1 2
TOTALS
2262
870
412
1283
Ethylbenzene (Lummus) This exothermic process is based on production of ethylbenzene via liquid phase benzene alkylation, and the front end of the process differs considerably from the Mobil/Badger process. The reaction systems differ substantially in operating temperature as well as phase of reaction. After the reaction system the processes are very similar. Total process energy input for the Lummus process is about 6 times greater than the theoretical minimum, compared with 8 times for Mobil/Badger. The benzene fractionator accounts for most energy and exergy losses, similar to the vapor-phase technology. This column processes both fresh feed and recycle benzene, and its large condenser operates at a low temperature, inhibiting economic energy recovery. There is some opportunity for medium- to low-pressure steam export from the alkylation reactor.
Btu/lb 1600
1400 1200 1000 800 600 400 200 0 Ethylbenzene (Lummus)
Theoretical Minimum Recoverable Energy Input Energy
Ethylbenzene (Lummus) Process Sub-Section Benzene Fractionator Primary Reactor Ethylbenzene Fractionator Poly Ethylbenzene Fractionator Secondary Reactor Pre-Fractionator
% 82 0 17 0 0 0
% 46 40 8 0 4 1
TOTALS
2007
601
762
1362
20
Nitric Acid
This exothermic process is based on a composite of various licensed technologies for production of nitric acid via oxidation of ammonia to nitric oxide and ultimately nitric acid. The total process input energy is less than the theoretical minimum due to significant energy generation made possible by the exothermic reaction. The largest energy and exergy losses are in the heat recovery section, which appears to have considerable additional capacity for energy recovery. The details of this section are not included in the model. In addition, most nitric acid plants utilized a steam turbine and gas expander to drive one or more compressors, and these are not modeled. Energy losses in the reaction section are due primarily to the nitric acid absorber, which performs the absorption of nitrogen dioxide in water while reacting it to form nitric acid, generating heat in the process. The heat of reaction is taken out in the partial condenser of the absorber, usually with refrigeration. Large internal exergy losses are due mostly to extreme temperature differences between feed and effluent streams and other exchanged streams in the system (gas coolers, steam superheaters, evaporators).
Btu/lb 2000 1800 1600 1400 1200 1000 800 600 400 200 0
Theoretical Minimum Recoverable Energy Input Energy
20 31 49
73 10 409 492
59 15 27
p-Xylene
This process is based on conventional technology where p-xylene is produced from a mixture of C8 aromatic isomers (p-xylene, o-xylene, m-xylene, ethylbenzene). The isomerization reaction is endothermic, but chemical conversion is exothermic due to side reactions. P-Xylene recovery and purification is not included in the model. The total process input energy is 600 times greater than the theoretical minimum due based on isomerization of a p-xylene-depleted xylene mixture. Energy losses are comparable between isomerization and fractionation, but exergy losses are much higher in isomerization due to large temperature differentials between inlet and outlet streams to the reactors and feed preheaters. The low external exergy loss indicates little opportunity for further energy recovery in isomerization. In fractionation, the produce cooler is the largest source of losses. The process temperatures in this cooler are high enough to suggest steam generation or crossexchange would save energy.
Btu/lb 3500
3000 2500 2000 1500 1000 500 0 p-Xylene
Theoretical Minimum Recoverable Energy Input Energy
56 44
70 30
21
Carbon Dioxide
Carbon dioxide is produced by recovery from gas streams where it is a contaminant or byproduct. The majority comes from ammonia, hydrogen or ethylene oxide producing plants. The total process input energy is about 19 times greater than the theoretical minimum energy required. The most common process is absorption via a physical or chemical solvent. This analysis models a process where monoethanoalamine (MEA) to recover carbon dioxide from power plant flue gas. The carbon dioxide stripper and absorber are large sources of energy and exergy losses. A large energy loss occurs where hot flue gas is cooled to minimize water content and temperature of flue gas entering the MEA system. Energy recovery could be possible from the associated cooling water recycle, make-up and purge loop.
Carbon Dioxide Recovery with MEA Process Sub-Section CO2 Absorber Mea-Makeup Mixer Lean MEA Cooling Exchanger Rich MEA Solution Pump Rich/Lean Mea Solution Exchanger CO2 Stripper Feed Quencher Column Quench/Makeup Water Mixer/Splitter Lean Mea/Makeup Water Mixer Quenching Water Cooling Exchanger Energy Loss Btu/lb 377 0 649 0 0 1001 0 589 94 336 External Exergy Loss Btu/lb 8 0 58 0 0 171 0 24 12 17 Internal Exergy Loss Btu/lb 478 9 1 0 24 42 40 17 33 1 Total Exergy Loss Btu/lb 486 9 59 0 24 213 40 41 45 18
Btu/lb 2500
2000 1500 1000 500 0
% 12 0 21 0 0 33 0 19 3 11
% 52 1 6 0 3 23 4 4 5 2
Carbon Dioxide
TOTALS
3045
289
646
935
Vinyl Chloride
This endothermic process is based on the Hoechst et al process for gas phase pyrolysis (dehydrochlorination) of ethylene dichloride (EDC). The total process input energy is about 19 times greater than the theoretical minimum energy required. The reaction is carried in the tubes of a fired furnace and the resulting effluent gases are at a higher temperature than input gases. The largest energy losses are in the quench section where reaction effluent is cooled from over 900oF to 120oF, a temperature too low for steam generation. EDC recovery also has high energy losses (source four distillation columns). The low-pressure HCl column with a refrigerated condenser accounts for losses in the HCl recovery section.
Btu/lb 3000
2500 2000 1500 1000 500 0 Vinyl Chloride
Theoretical Minimum Recoverable Energy Input Energy
Pre-Heater Dehydrogenation Reaction Quenching HCl Recovery VCM Recovery EDC Recovery TOTALS
0 0 42 14 10 35
0 0 69 44 10 102 225
17 43 10 8 5 16
22
Acetic Acid
This analysis looks at a new technology (Acetica Process ) developed by Chiyoda and UOP that is similar to other acetic acid facilities utilizing carbonylation of methanol. Differences are the bubble column reactor design which eliminates the agitator, and the immobilization of the catalyst onto solid particles rather than being dissolved in reaction medium. No commercial plants using this technology are currently operating. The total process input energy is about 4 times greater than the theoretical minimum energy. Acetic acid refining accounts for the largest energy and exergy losses, primarily due to the crude fractionator. The overhead temperature of the column is too low to reasonably recover the energy in condenser cooling water. Large internal exergy losses are due to large temperature, pressure and composition differences of the streams leaving the column. Large internal exergy losses are also present in the carbonylation reactor, due to large temperature, pressure and composition differences among the recycle, feed methanol and carbon monoxide streams.
Btu/lb 1800
1600 1400 1200 1000 800 600 400 200 0 Acetic Acid
Theoretical Minimum Recoverable Energy Input Energy
8 92
19 144 163
57 43
Butadiene
This analysis looks at extractive distillation with DMF solvent (Nippon Zeon process) to recover butadiene from mixed C4 streams (butane, butene, butylene, butadiene). This is strictly a separation process with no chemical reaction, so no theoretical minimum is given. Distillation columns (butadiene stripping column, butene extractive column, propyne and butadiene product columns) account for large energy and exergy losses in extractive and conventional distillation. The large internal exergy losses reflect wide differences in the composition and temperature of inlet and outlet streams. Large energy losses are due to refrigeration used for condensation in some cases. Most of the column condensers are operating at temperatures too low for energy recovery, except the acetylenes stripping column, where reuse of heat of condensation is possible.
Btu/lb 1400
1200 1000 800 600 400 200 0 Butadiene
Theoretical Minimum Recoverable Energy Input Energy
Feed Vaporizer & DMF Cooling Extractive Distillation Conventional Distillation TOTALS
28 35 37
9 13 21 44
79 230 61 369
88 243 82 413
21 59 20
23
Cumene
Solid Phosphoric Acid (SPA) Catalyzed This exothermic process is based cumene via propylene alkylation of benzene with a solid phosphoric acid (SPA) catalyst (UOP design). Total process energy input is about 1.5 times greater than the theoretical minimum. The largest energy losses occur in air coolers, primarily due to their low temperatures. Virtually all energy and exergy losses occur in cumene recovery, primarily due to three distillation columns. Additional energy recovery is possible from the cumene fractionator, but may not be economical. There may be opportunity for feed preheat in the alkylation section.
Btu/lb 900
800 700 600 500 400 300 200 100 0 Cumene (SPA catalyst)
Theoretical Minimum Recoverable Energy Input Energy
0 1170
1172
0 100
0 192
192
156 225
382
156 416
573
27 71
Zeolite Catalyzed The advantage of the zeolite process is that it is non-corrosive and enables operation at lower benzene/propylene ratios, resulting in lower energy and external exergy losses. Recovery of spent catalyst is also easier than the other two processes studied. Fouling of the zeolites, however, could lead to higher catalyst costs. Total process energy input is about 2 times greater than the theoretical minimum. Cumene recovery is again the greatest source of energy and exergy losses, due to several distillation columns, most of which are operating at condenser temperatures too low to generate low pressure steam. The cumene column is a candidate for steam generation, with an overhead temperature over 300oF.
Btu/lb 1200
1000 800 600 400 200 0 Cumene (Zeolite catalyst)
Theoretical Minimum Recoverable Energy Input Energy
17 83
64 240 304
40 60
AlCl3 Catalyzed This process is very similar to the zeolite process discussed above, except that the spent catalyst is not as easy to recover. There are additional minor energy and exergy losses in the catalyst recovery section, which is not required for the zeolite technology.
Btu/lb 1200
1000 800 600 400 200 0 Cumene (AlCl3 catalyst)
Theoretical Minimum Recoverable Energy Input Energy
28 0 72
68 0 268 337
40 0 60
24
Distillation Units
Table 6. Energy and Exergy Losses in Unit Operations Distillation is the most used separation Energy Loss Exergy Loss technology and contributes to a significant (Tbtu/yr) (Tbtu/yr) Unit Operation portion of energy/exergy losses. Most of the Endothermic Reaction 0 57 external exergy losses in distillation units Exothermic Reaction 20 130 occur in condensers, which are usually cooled Distillation 408 172 by cooling water or air. In many of the Evaporation 0 4 processes studied, a relatively few distillation Adsorption/Absorption 15 18 columns and heat exchangers are responsible Crystallization 23 1 for the bulk of energy and exergy losses. In Cooling Water 867 119 some cases a combination of low temperature Heat/Electrical/Steam 1209 690 requirements and non-condensables dictates Energy the use of refrigeration, a large source of energy use and losses. These losses could be minimized by improved heat integration such as cooling the condensers with other process streams or by using waste heat to raise steam. Another approach is development of alternative separation technologies that do not require raising products to their respective boiling points.
Exothermic Reactors
In exothermic reactors, exergy losses are due to the wide range of operating temperatures in feeds and products, and using the reactor to accomplish some portion of feed preheat. Exergy losses could be minimized by lowering operating temperatures or by using waste heat to preheat reactor feeds or to generate steam for reuse or export. Improved reaction conversions and selectivities or new reaction chemistries with higher selectivities are possible approaches for reducing these losses. Another option is minimizing reaction exergy losses through changes in process parameters. Lower temperature reaction systems that mitigate the need for quenching of products could also reduce losses in a number of exothermic processes.
Separations
Separations other than distillation, such as evaporation, adsorption or crystallization, do not contribute substantially to overall energy and exergy losses. However, implementation of more energy-efficient separation technologies could also play a large role in reducing energy/exergy losses in major operations. For example, the use of membrane separation in ethylene production technology could be a viable option to de-methanize or de-ethanize crude ethylene without the need for refrigeration and refrigerateddistillation. Another example is styrene production, where the very high temperature product effluent must be brought to a very low temperature before recovery, with substantial energy use and losses. A novel separation scheme to recover styrene could improve the energy profile.
Endothermic Reactions
There were no exergy losses in the endothermic reactions as the models are based on the effective energy input into each reactor. In practice, losses are inevitable due to energy transfer inefficiencies from the primary energy sources to the reactors.
25
Grouping specific equipment types provides a perspective on where energy losses are greatest. Total Energy As Table 7 illustrates, energy losses are Loss Btu/lb Type of Equipment No. of Items concentrated in heat exchangers and distillation Heat Exchangers 26 54,200 columns (strippers, fractionators). Condensers, Columns 20 37,600 air and product coolers, and heat and Compressors 2 1,190 Reactors 3 4,600 refrigeration recovery units account for a large Miscellaneous 5 6,420 share of heat exchanger losses. The column losses shown in Table 7 are also due primarily to heat exchange losses in condensers (not modeled separately in all cases). Overall, in all processes, heat exchanger accounts for the overwhelming majority of energy and exergy losses. Table 7. Energy Losses for Equipment (>500 Btu/lb) Table 8 provides details on the fifteen unit operations with the highest energy intensity (Btu/lb) for individual chemical processes. Energy quality indicates the potential for energy recovery. As stated earlier, an energy quality of 15%-20% is moderate and indicates some potential for economical energy recovery. Over 20% the potential for economic energy recovery becomes much greater. A number of technologies dominate the top fifteen energy consumers shown in Table 8. These include acrylonitrile, methanol, and styrene. Acrylonitrile separations require refrigeration and quenching, both large sources of energy and exergy losses, and there are significant opportunities for energy recovery (energy quality of 36-42%). Methanol has a similar profile. Styrene requires substantial cooling prior to recovery, but the energy quality is low, indicating little potential for additional energy recovery. The ethylene oxide condenser has the highest energy-intensity by far, primarily due to cooling and refrigeration requirements. This is due to the very low per pass conversion of ethylene needed to maintain selectivity, which necessitates scrubbing with water and results in a very dilute overhead stream that makes product recovery difficult. These results fortify the conclusion that separation processes not requiring distillation could be developed to greatly improve energy efficiency. Alternatively, fundamental process changes could mitigate the need for difficult separations. In some cases better heat integration (e.g., pinch analysis) can be applied to reduce exergy losses. However, it is limited to pure heat exchanger networks involving pure heat load analysis, and cannot be used for example, to improve a system with heat pumps. In such cases pinch and exergy analysis could be combined to better evaluate targets for improvement. Table 8. Highest Energy-Consuming Equipment, Ranked by Energy Loss
Chemical Technology Ethylene Oxide MTBE Methanol ICI LP Acrylonitrile (propylene) Acrylonitrile (propane) Styrene - Lummus Methanol Lurgi Acrylonitrile (propane) Ammonia Acrylonitrile (propane) Acrylonitrile (propylene) Formaldehyde Acrylonitrile (propylene) Styrene Fina Ethyl Benzene - Badger Equipment Name Condenser MTBE Column Methanol Column Heat & Refrigeration Recovery Heat & Refrigeration Recovery Air Cooler Heat Recovery C-101 Overhead Cooler Syn Gas Separator HCN Stripper C-101 Overhead Cooler Reactor HCN Stripper EB Column Benzene Fractionator Total Energy Loss (Btu/lb) 11621 8641 7775 6727 5516 3284 2958 2718 2608 2600 2443 2268 2073 1943 1850 External Energy Loss (Btu/lb) 1752 1355 1015 2435 2339 395 922 293 614 84 295 487 42 192 757 Energy Quality (%) 15 16 13 36 42 12 31 11 24 3 12 21 2 10 41
26
Research Recommendations
Ethylene
261
Large opportunities exist due to high volume production and energy-intensity of current process. Thermal cracking results in a highly reactive product mix that necessitates energy-intensive quenching and complex separation processes. R&D areas that could reduce energy intensity include: Low temperature, more selective retrofit reaction systems to replace pyrolysis and eliminate need for quenching Novel separation concepts (hybrid systems) coupled with new ways of producing ethylene Dehydrogenation or oxydehydrogenation based on ethane feedstock (dependent on price of NGLs versus petroleum) New routes to ethylene based on alternative feedstocks (ethanol, methanol, methan/syngas, higher olefins), coupled with simpler recovery and purification technologies Little incentive for R&D as market is not growing and producers are under severe economic pressure. A fair amount of energy recovery is already practiced. Improved carbon dioxide removal is one potential area for reducing energy losses. Ethylene oxide (EO) technology must operate at low per pass conversion to maintain selectivity and to control the reaction gas composition outside of the flammable region. New process concepts will be needed to lower energy consumption: Fluidized bed reactors Liquid-phase oxidations, liquid-phase processes using hydroperoxide or hydrogen peroxide Bioxidation of ethylene Processes for richer EO-containing streams to reduce large recycle Novel separations for richer EO streams, including carbonate system Currently nearly all propylene is produced as a coproduct with ethylene in naptha crackers. At present there are limited incentives to increase capacity for dedicated production of propylene. However, this could change in future, as the demand for propylene derivatives (polypropylene and propylene oxide) is beginning to outstrip demand for ethylene derivatives. Purification requirements are critical and current yields are approaching stoichiometric. Improvements could be made in purification and catalyst recovery, which are both complex and energy-intensive: Novel separation schemes for solvent recovery and dehydration and for refining/purifying PTA Process requiring less corrosive solvent Entirely new concepts for producing PTA
Ammonia
115
Ethylene Oxide
98
Propylene
67
55
27
Research Recommendations
MTBE Methanol
53 36
Demand for MTBE is declining due to legislation banning its use as a gasoline additive. No research is warranted. Expectations for building methanol plants in the U.S. are not high, unless used as a means for bringing methane to market. Innovations could include: Liquid phase processes for methanol production Better process technologies for production of synthesis gas Improved catalysts, including biocatalysts Alternative feedstocks (methane, biomass) Novel separation technologies to reduce distillation Conversion to acrylonitrile (ACN) takes place at high temperatures and requires a rapid quench of reaction gases to lower temperatures with a complex separation scheme (often with refrigeration). Novel ideas are needed to reduce energy intensity: Fluidized beds Recycle process with substitution of oxygen for air Biocatalytic production of ACN Novel concepts for difficult acetonitrile/acrylonitrile separations It is uncertain what the process of choice will be for formaldehyde production (mixed-oxide versus silver catalyst). Possible improvements: More selective, longer-life catalysts New ways to recover formaldehyde (without polymerization) Growth in nitric acid markets is stagnant, with little incentive for R&D. Current processes practice significant heat recovery. Lower temperature catalysts for ammonia oxidation could be an area for research if incentives were present. Current high temperature endothermic reaction requires preheating of feed and cooling of effluents with high energy burdens. R&D to reduce energy use: Liquid phase lower temperature process with continuous removal of hydrogen Novel separation technologies to remove hydrogen Process using diluents other than steam Alternative feedstock process Ethylbenzene is used exclusively for the production of styrene, and synergies should be considered, as well as the possibility of finding alternative feedstocks for producing styrene. The current process is relatively efficient; finding more active catalysts to lower the alkylation temperature would be a useful future research area. Considerable energy recovery is already practiced. New opportunities include: New separation technologies, e.g., removing p-xylene during isomerization Couple p-xylene process with downstream terephthalic acid process to achieve reductions in energy use (e.g., unique catalyst for oxidation) Solvents for recovering carbon dioxide are limited and expensive, and could poison recycle gases. Possible areas for research: Better solvents, especially adducts Novel separations or hybrid separations with membranes, PSA, etc
Acrylonitrile
32
Formaldehyde
31
24 16
Ethylbenzene
16
p-Xylene
15
Carbon Dioxide
15
28
Research Recommendations
Vinyl Chloride
14
Large energy use and losses are due to the need to vaporize ethylene dichloride (EDC), crack at high temperature and then quench the reaction gas (done to minimize coking). In addition, many distillation systems are needed to separate HCl coproduct and purify vinyl chloride and EDC. Concepts to reduce reaction temperature and energy for separations include: Cracking additives Low-temperature catalysts Alternative feedstocks, e.g., catalytic dehydrogenation of ethyl chloride Novel separation systems to reduce distillation Catalyst research continues to improve acetic acid production. Other research needs: Novel separations to improve carbonylation routes (e.g., separation of gases from carbonylation reaction, supplementation of distillation) Other routes to acetic acid (oxidation of butane, ethylene-based, oxidative dehydrogenation of ethane) Acetic acid from biomass via chemical or biocatalysis All cumene goes to production of phenol and acetone. Demand for phenol is not balanced with demand for acetone (often sold at distress prices). The result is a major thrust to find alternative processes to produce phenol that do not require propylene or produce acetone. Related research topics include alternative (or one-step) routes to phenol and integration with bisphenolA processes. Almost all butadiene is present in C4 (butane and derivatives) streams from refineries and steam crackers, and little dedicated production exists. Improvements could be made in methods of separating butadiene from butane/butane/butadiene mixtures (new solvents, hybrid systems, membranes, PSA).
Acetic Acid
Cumene
Butadiene
29
References
ACC 2003 Ahern 1980 Al-Jarallah et al 1988 ANL 1991 Aspen 2001 Aspen 2001a C&E 1996 CMR 2004 Cremer 1980 Guide to the Business of Chemistry 2003, American Chemistry Council, Washington, DC Ahern, J. E. The Exergy Method of Energy System Analysis, John Wiley & sons, New York, 1980. Al-Jarallah, A. M. and A. K. Lee, Economics of New MTBE Design, Hydrocarbon Processing, July 1988. Argonne National Laboratory, Energy Conservation Potentials in the Chemical Industry Aspen Technology, Inc. What is New: Aspen Engineering suite Version 11.1, October 2001. Aspen Technology, Inc. Aspen Engineering Suite 11.1 Documentation, October 2001. Chemical & Engineering News, 6/24/1996 R. Brown, Energy Cost and Demand Issues are Key to Petrochemicals in 2004, Chemical Marketing Reporter, Vol. 265, No. 1, January 5, 2004 (p 1) Cremer, H. Thermodynamics Balance and Analysis of a Synthesis Gas and Ammonia Plant-Exergy Analysis. In Thermodynamics: Second law Analysis,, ed R.A. Gaggioli, ACS Symposium Series 122, American Chemical Society Washington , D.C., 1980, pp111-130. Energy and Environmental Profile of the U.S. Chemical Industry, U.S. Department of Energy, Office of Industrial Technologies, May 2000. Gaggioli, R.A. and P.J. Petit, Use the Second Law First, Chemtech, Vol. 7, 1977, pp 496-506. Jacobs Engineering Netherlands B.V. Exercom (Version2) Manual for Aspen Plus Version 10.2, September 2001 Moran, M. J. Availability Analysis A Guide to Efficient Energy Use Prentice-Hall, Englewood Cliffs, N.J, 1982 Process Economic Program, SRI International, Menlo Park, California 2002. Rosen, M.A. and D.S. Scott, A Thermodynamic Investigation of a Process for the Production of Ammonia from Natural Gas. In Analysis of Energy Systems Design and Applications, ed M.J. Moran, S.S. Stecco and G.M. Reistad, American Society of Mechanical Engineers, New York, 1987, pp 95100.
EI 2000 Gaggiloi et al 1977 Jacob 2001 Moran 1982 Pep 2002 Rosen et al 1987
30
Rosen et al 1985
Rosen, M.A. and D.S. Scott, The Enhancement of a Process Simulator for Complete Energy-Exergy Analysis. In Analysis of Energy Systems Design and Operations, ed. R.A. Gaggioli, American Society of Mechanical Engineers, New York, 1985, pp 71-80. Shah, V. A. and J. McFarland, Low Cost Ammonia and C02 Recovery, Hydrocarbon Processing. 43-46 March 1988. SRI PEP Process Module Documentation of 10/18/99, page 2. SRI PEP Report 43 Szargut, J., D.R. Morris, and F.R. Steward, Exergy Analysis of Thermal, Chemical and Metallurgical process, Hemisphere Publishing Corp., New York, 1988 Technology Vision 2020: The US Chemical Industry, American Chemical Society, December 1996.
Vision 2020
31
32
Table A.1 Input, Actual Process, Minimum Theoretical and Recoverable Energies Theoretical Minimum Process Energy Btu/lb External Exergy Losses as % ( Total Exergy Input)
28% 28% 23% 72% 52% 17% 33% 23% 31% 13% 24% 44% 35% 22% 15%
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Styrene (Lummus) 4,703 1,697 305 Styrene (Fina/Badger) 3,365 1,122 369 Vinyl Chloride 2,671 975 147 Ethylbenzene (Mobil/Badger) 1,787 965 (236) Ethylbenzene (Lummus) 1,528 1,147 (231) Ethylene Oxide 7,741 5,735 (6,720) Ethylene (Braun) 8,656 5,534 326 Ethylene (Kellogg) 8,139 5,035 217 Carbon Dioxide 2,083 508 (426) Acetic Acid 1,612 786 (512) Methanol (ICI LP) 4,883 871 (4,546) Methanol (Lurgi) 2,273 841 (4,132) ACN From Propylene 4,364 1,020 (8,015) ACN From Propane 5,381 1,392 (13,152) Formaldehyde 698 115 (3,209) () Exothermic Reaction, net chemical conversion exergy inflow N/A A separation process without chemical reaction
340 340 142 273 273 734 650 650 N/A 436 802 802 4,355 5,509 802
478 410 225 870 601 2,096 1,806 1,167 289 163 1,289 2,165 3,191 3,129 491
914 1,081 603 412 762 10,360 3,402 3,651 646 1,134 4,128 2,808 5,844 11,415 2,833
1,392 1,491 828 1,282 1,363 12,456 5,208 4,818 935 1,297 5,417 4,974 9,035 14,544 3,324
30% 44% 31% 72% 89% 161% 60% 59% 45%* 80% 111% 219% 207% 270% 476%*
33
Process
Table A.1 Input, Actual Process, Minimum Theoretical, and Recoverable Energies (continued) Theoretical Minimum Process Energy Btu/lb External Exergy Losses as % ( Total Exergy Input)
124% 9% 7% 32% 31% 30% 52% 34% 49% 51%
16 Terephthalic Acid 1,919 1,157 4,730 3,047 17 Butadiene 1,382 468 55 N/A 18 Propylene 4,548 3,047 1,440 846 19 p-Xylene (Isomerization) 3,228 1,702 (133) 5 20 Nitric Acid 232 207 (1,401) 1,953 21 Ammonia 4,596 3,543 (351) 414 22 MTBE 8,868 2,572 (135) 124 23 Cumene (SPA Cat) 812 328 (245) 526 24 Cumene (Zeolite Cat) 1,061 375 (248) 526 25 Cumene (AlCl3 Cat.) 1,124 440 (240) 526 () Exothermic Reaction, net chemical conversion exergy inflow N/A A separation process without chemical reaction
4,447 369 1,892 1,249 1,117 2,797 1,299 382 319 337
5,887 413 2,119 1,835 1,609 3,967 2,706 574 623 680
307% 30% 47% 57% 694% 86% 31% 71% 59% 61%
34
Process
Table A-2. Energy and Exergy Losses at the Unit Operation Level, Btu/lb Reactions (Btu/lb) Evaporation Exothermic Distillation Process Endothermic
Separations (Btu/lb) Crystallization Cooling Water Ion-exchange Adsorption/ Absorption Membrane Extraction
1 2 3 4 5 6 7 8 9 10 11 12
Styrene (Lummus) Styrene (Fina/Badger) Vinyl Chloride Ethyl Benzene (Mobil/Badger) Ethylbenzene (Lummus) Ethylene Oxide Ethylene (Braun) Ethylene (Kellogg) Carbon Dioxide Acetic A Methanol (ICI-LP) Methanol (Lurgi)
Energy Exergy Energy Exergy Energy Exergy Energy Exergy Energy Exergy Energy Exergy Energy Exergy Energy Exergy Energy Exergy Energy Exergy Energy Exergy Energy Exergy
0 69 0 248 0 360 0 3
0 106 0 492
0 54 0 5
295 129 2,120 380 1,180 239 2,126 877 1,841 687 441 516 642 1,112 1,071 1,087 1,001 253 1,394 507 8,017 3,768 1,897 716
Filtration
Drying
0 0
0 903 0 0
377 486
0 23 0 1
569 50 5,193 397 1,200 162 2,262 870 166 40 14,630 2,579 2,522 345 4,830 382 2,299 286 1,332 138 9,601 1,245 1,984 262
35
Table A-2. Energy and Exergy Losses at the Unit Operation Level, Btu/lb (continued) Reactions Separations (Btu/lb) (Btu/lb) Crystallization Evaporation Adsorption/ Absorption Exothermic Distillation Membrane Extraction Process Ion-exchange Endothermic
13 14 15 16 17 18 19
Energy Exergy Energy Exergy Energy Formaldehyde Exergy Energy Terephthalic Acid Exergy Energy Butadiene Exergy Energy Propylene Exergy Energy p-Xylene Exergy
0 1,292 0 1,269
92 80 31 151
0 851
247 115
5,488 814 7,583 1,097 2,268 487 1,985 487 712 30 3,376 310 462 25
36
Cooling Water
Filtration
Drying
Table A-1. Energy and Exergy Losses at the Unit Operation Level, Btu/lb (continued) Reactions Separations Crystallization Distillation Membrane Extraction Process Ion-exchange Endothermic Evaporation Adsorption/ Absorption Exothermic
20 21 22 23 24 25
Nitric Acid Ammonia MTBE Cumene (SPA Cat) Cumene (Zeolite Cat) Cumene (AlCl3 Cat) TOTALS
Energy Exergy Energy Exergy Energy Exergy Energy Exergy Energy Exergy Energy Exergy Energy Exergy
0 187
0 2,376
66 568 117 908 261 8705 118 2522 0 1,032 146 370 174 1,019 230 331 305 922 249 365 1,694 43,826 24,819 16,096
574 214
Filtration
Drying
0 2,561
0 0
1,074 1,858
0 0
19 12
0 0
0 0
490 232 47 207 1,723 4,596 133 3,543 8,835 8,868 1,384 2,572 834 812 63 328 797 1,061 120 375 1,043 1,124 190 440 247 82,182 87,654 115 11,941 42,039
37