Optimal Process for an Isobutene Rich Feed

As noted previously, isobutene can be sourced from several process units within a large refinery complex. The two most common sources of isobutene for etherification are steam cracking units (35-50 wt% isobutene) and catalytic cracking units (10-25 wt% isobutene). This section examines the implications of processing C4 hydrocarbon from a steam cracking unit (SCU) and the effect it has on process selection.

A typical SCU product composition of 40% isobutene, 30% n-butenes and 30% butadiene, was assumed as the ETBE unit feed. Process designs were completed for a 5000 kg/hr (1000 bbl/day) ether unit using (a) conventional technology, (b) a reactive distillation column that was designed for MTBE synthesis, and (c) two designs which were specifically tailored for ETBE synthesis. For each type of system, an optimised MTBE design (with 5% excess methanol in the feed) was used as the base case and then two ETBE designs were completed, with and without a 5% stoichiometric excess of ethanol. The cases with an excess of ethanol should produce the highest isobutene conversions while the cases with only stoichiometric ethanol should produce the highest purity ETBE products. It is important to recognise that, where a stoichiometric excess of ethanol is not used, the rate of formation of DIB will be higher (Kitchaiya and Datta, 1996).

6.2.3.1 Conventional Process

This design comprises a two-stage reactor with an intercooler and a purification column, and conforms closely to the process flow diagram shown in Figure 6.1. The first reactor operates isothermally at 80°C. The intercooler cools the reactor effluent to 40°C before the second reaction stage that is operated adiabatically. This represents a reaction scheme that is close to the optimum for ETBE synthesis where the target isobutene conversion is 90%.

The effluent from the second reactor is purified by a distillation column with 30 ideal stages (the condenser is stage 1 and the reboiler is stage 30), operating with an overhead pressure of 700 kPa. This is considered to be typical of industrial MTBE processes, although a wide variation inevitably exists due to variances in product specifications and feed compositions. The feed point, stage 18, was optimised using rigorous simulations of the column. An ether product composition with 0.01% C4 was specified, based on the potential detrimental influence on the product volatility of a higher C4 content. This was achieved with a reflux ratio of 1.2. The pretreatment stage and ethanol recovery equipment were not simulated as they are essentially independent of the process design optimisation.

Pro/II (SimSci, 1994) was used for all the simulations undertaken here. Chemical equilibrium was assumed at both reactor outlets corresponding to a case with surplus catalyst (e.g. plant start-up). The ETBE reaction model of Jensen and Datta (1995) was used to provide the necessary equilibrium data. The UNIFAC model was used to predict the liquid phase non-idealities and all calculations were made in terms of component activities rather than concentrations. Other physical properties were predicted with the Soave-Redlich-Kwong equation of state. This combination of reaction data and physical property methods was used successfully in Chapter 3 and was validated against experimental data for an MTBE column.

The reactor feed rate (hydrocarbon plus ethanol) was scaled to produce exactly 5000 kg/hr of ether for the MTBE case and each of the ETBE cases (with and without an ethanol excess). The key simulation results for both cases are shown in Table 6.1. The net oxygen production rate (kg/hr of oxygen in the ether product) and consumption of raw materials per kg of ether were calculated for each case to facilitate an effective comparison between cases. These results, plus the isobutene conversion and ETBE purity have been highlighted.

The oxygen production rate includes a contribution from both the ether product and the alcohol reactant but the methanol contribution in MTBE production is normally negligible due to the high MTBE purity. For ETBE, 11-14% of the net oxygen production comes from ethanol contained in the ether product. This essentially balances the higher oxygen content of MTBE compared with ETBE so that there is no clear advantage for either ether. The net isobutene consumption is roughly equivalent in each system but MTBE production is much more economical with respect to the consumption of the alcohol reactant, The difference arises from (a) the more favourable stoichiometry of the MTBE reaction, (b) recycling methanol from the distillate in the MTBE process and (c) the relatively small concentration of methanol in the MTBE product. The ETBE reaction produces a higher mass yield of product but this is balanced by the lower conversion.

Table 6.1 - Conventional Processes for ETBE and MTBE (Isobutene-Rich Feed)

MTBE with 5% Excess Methanol

ETBE with 5% Excess Ethanol

ETBE with Stoichiometric Ethanol

Reactor Feed Rate (kg/hr)

10,400

10,260

10,330

Second Stage Reactor Outlet Temperature (°C)

48

54

54

iBut Conversion: 1st Reaction Stage

88.0%

79.6%

78.0%

2nd Reaction Stage

61.4%

53.8%

48.9%

Overall

94.8%

90.0%

88.3%

Bottoms Composition (wt%)

5.4% EtOH, 0.7% DIB

Distillate Composition (wt%)

2.6% iBut, 93.4% other C4s, 3.7% MeOH

5.6% iBut, 94.3% other Qs, 0.15% EtOH

6.6% iBut, 93.2% other C4s, 0.14% EtOH

Oxygen Production Rate (kg/hr)

915

913

885

Net Alcohol Consumption (kg/kg ether)

0.365

0.526

0.511

Net Isobutene Consumption (kg/kg ether)

0.832

0.821

0.826

Reboiler Duty (MW)

1.58

1.40

1.43

Column Diameter (mm)

1200

900

900

Overall, the two most important disadvantages of the ETBE process are the low isobutene conversion and the low ETBE product purity. The isobutene conversion in the reactor is around 5% lower than would be achieved in MTBE production. This increases the consumption of both raw materials. The ETBE product purity is 5-7% lower than would be achieved in MTBE production. This further increases the ethanol consumption as ethanol contained in the column bottoms product remains with the ETBE product, even if downstream recovery/recycling equipment is present.

The ETBE process does, however, have two advantages. Firstly, the reboiler duty is 10-12% lower than required for MTBE production. This is primarily because ethanol is recovered preferentially in the bottoms product rather than the distillate. The resulting boil up rate and vapour flow up the column is, therefore, lower than the MTBE process and the column diameter can be reduced accordingly. Secondly, the distillate from ETBE synthesis contains only 0.14-0.15% ethanol.

A low concentration of ethanol in the distillate is potentially advantageous as it might allow the ethanol recovery equipment (shown downstream of the purification column in Figure 6.1) to be retired with some saving in operating expenses. The main considerations here are the possible poisoning of catalysts in downstream process units and the potential saving in raw material costs. As the ethanol concentration in the distillate decreases, both of these factors become less important. The use of recovery equipment incorporating a water wash also creates a risk of water contamination of the reactor feed following process upsets. Even during normal operation, the ethanol-water azeotrope will introduce small quantities of water into the system leading to isobutanol formation. Although the recovery equipment might feasibly be eliminated from the ETBE process, it should be considered essential for MTBE production as the distillate product contains over 3% methanol and recovery can be effected without contaminating the recycle stream with water (due to the absence of an azeotrope in the methanol-water system).

Comparing the two ETBE cases, a stoichiometric excess of ethanol increases the ethanol consumption but reduces the isobutene consumption. The optimal stoichiometric excess should be optimised with respect to the price and availability of each raw material while also considering the effect of excess ethanol on the reaction selectivity (the DIB reaction is suppressed even at low values of excess) and the acceptable level of ethanol in the ether product.

ETBE has a slightly higher value than MTBE due to its lower RVP and better blending characteristics. However, overall, ETBE production is unlikely to be competitive with MTBE unless ethanol can be obtained very cheaply.

6.2.3.2 Reactive Distillation Process

Reactive distillation columns for MTBE synthesis are now commonplace. Several sources suggest that these columns can also be used for ETBE synthesis (e.g. Petrochemical Processes, 1995). To test this hypothesis, a reactive distillation system was designed and optimised for MTBE production, and then simulated using Pro/II which now contains a reactive distillation module (version 4.01 onwards). Suggested operating conditions for producing MTBE from an isobutene-rich feed using reactive distillation, and the predicted process results are shown in Table 6.2. It can be clearly seen (by comparison with Table 6.1) that this technology performs better than the conventional process, especially with respect to the isobutene conversion obtained.

The reactive distillation column used in the simulations was simply modified from the column used for the non-reactive distillation cases presented above, i.e. 30 ideal stages. This reflects the capacity for revamping conventional processes for reactive distillation with only limited capital input. Many of the existing reactive distillation etherification units are merely conventional processes which have been revamped to eliminate the second reactor and incorporate catalyst into the ether purification column. The simulations of the reactive distillation system were completed by fixing the column reboiler duty at the values used for the non-reactive distillation case, while allowing the reflux ratio to increase slightly due to the heat liberated by the reaction which contributes to the heat input to the column. The only change from the configuration used previously was that the feed point was shifted to stage 15 (from stage 18). This was done to increase the number of stages available for ether stripping below the feed-point to allow high ether product concentrations to be retained.

The same equipment was then simulated for ETBE synthesis, for cases with and without a stoichiometric excess of ethanol. Again the reboiler duty was maintained constant from the non-reactive distillation cases. The simulation results are included in Table 6.2 and indicate that the isobutene conversion and ether purity are much lower for ETBE. The reactive distillation process provides a slight improvement in isobutene conversion (approximately 0.5% with a 5% ethanol excess) but it actually reduces the ether purity. The oxygen production rate is increased slightly as almost all the unreacted ethanol is recovered with the ether product. The net consumption of both ethanol and isobutene remain essentially unchanged.

One significant advantage arising from the reactive distillation process for ETBE is that the ethanol concentration in the distillate is reduced to 1-2 ppm. This removes any incentive to construct ethanol recovery equipment downstream of the reactive distillation column. If ethanol poisoning of a downstream catalyst is still a consideration, an additional guard bed might be required at most. Nevertheless, there is a significant reduction in processing equipment with an associated capital saving and decrease in operating expenses.

Overall, the ETBE process is probably still uncompetitive with the MTBE process due to the high ethanol consumption and the comparatively low isobutene conversion. However, the retirement of ethanol recovery equipment could provide a partial justification for the construction of an ETBE unit if the availability of capital was tight.

Table 6.2 - Reactive Distillation Processes forETBE and MTBE (Isobutene-Rich Feed)

MTBE With 5% Excess Methanol

ETBE With 5% Excess Ethanol

ETBE With Stoichiometric Ethanol

Reactor Feed Rate (kg/hr)

9,880

10,200

10,350

Overall Isobutene Conversion

99.8%

90.5%

88.1%

Bottoms Composition (wt%)

0.01% MeOH, 0.7% DIB

91.3% ETBE, 7.1% EtOH, 1.6% DIB

6.0% EtOH, 2.3% DIB

Distillate Composition (wt%)

0.06% iBut, 97.6% other C4s, 2.3% MeOH

4.9% iBut, 95.1% other C4s, 2 ppm EtOH

5.7% iBut, 94.3% other C4s, I ppm EtOH

Oxygen Production Rate (kg/hr)

915

920

899

Net Alcohol Consumption (kg/kg ether)

0.365

0.523

0.512

Net Isobutene Consumption (kg/kg ether)

0.790

0.816

0.828

Reflux Ratio

1.88

1.79

1.77

Reboiler Duty (MW)

1.58

1.40

1.43

Increase in Flooding (%)

0%

-15%

+ 1%

6.2.3.3 Advantages of Reactive Distillation

The advantages of reactive distillation for MTBE are very clear from these simulations. The isobutene conversion approaches 100% (up from around 95% for the conventional process) and the product purity is greater than 99%. These results are consistent with the current trends away from the conventional synthesis route for MTBE to processes based around reactive distillation technology. Although the second stage reaction equipment and intercooler are made redundant by the reactive distillation process, the methanol recovery equipment cannot be eliminated as there remains a significant concentration of methanol in the reactive distillation column distillate due to the various methanol-C4 azeotropes present.

MTBE columns can be operated at stoichiometric methanol ratios. However, the isobutene conversion is reduced and methanol cannot be eliminated entirely from the distillate as the azeotropes continue to influence the composition of the column products. There is also no incentive to increase the ether purity as it is very high (99% +) for all methanol excesses while the volume of n-butenes is in excess of the azeotropic requirements. This allows for methanol excesses of up to around 40% for isobutene-rich feeds and much higher for isobutene-lean feeds. Even with sub-stoichiometric mixtures, methanol cannot be reacted to extinction because of a corresponding drop in conversion and the continued existence of methanol azeotropes. As with ETBE, reducing the stoichiometric excess of methanol will result in an increase in dimerisation of isobutene to DIB.

The benefits of reactive distillation can be assessed in three main areas: raw material consumption, energy efficiency and capital cost. The relative importance of each of the areas will depend on the current refinery economics, refinery configurations, particular product specifications and other local requirements. These areas are discussed below with respect to both MTBE and ETBE synthesis.

1. Raw Material Consumption

Reactive distillation produces a higher overall isobutene conversion for both MTBE and ETBE (provided that some excess ethanol is used) although the increase for the ETBE process is much less significant. Isobutene conversion to ETBE is actually reduced slightly if stoichiometric ethanol is used as the benefits of internal recycling of reactants in the reactive distillation environment are partially lost. Clearly, an increased isobutene conversion reduces the net consumption of isobutene for a fixed production rate of ether.

The ether product purity remains essentially unchanged for MTBE but is decreased slightly for ETBE. Unfortunately, the decreased purity increases ethanol consumption as ethanol lost in the ether product cannot be recovered. However, ethanol is also an effective oxygenate and its presence in the ether is not necessarily detrimental. The overall fuel purity (defined as the ether concentration plus the ethanol concentration in the bottoms) is around 99% for the reactive distillation process, which is comparable to the conventional process and also MTBE production.

2. Energy Efficiency

No additional reboiler duty is required for reactive distillation. However, as noted above, ihe heat released by the reaction increases the condenser duty. This cooling duty is elfcttively transferred from the intercooler, which is made redundant by the process changes, but is rejected at a higher temperature (62°C at a column overhead pressure of 700 kPa compared with 40°C from the intercooler). This has process benefits with respect to heat integration and could also result in shifting cooling load from cooling-water exchangers to air coolers.

3. Capital Cost

Transfemng the catalyst (whole or part) from the second, adiabatic, packed-bed reactor to the distillation column will incur some additional capital cost at both the fabrication and construction stages but it eliminates the intercooler and packed-bed reactor (and subsidiary equipment), with a substantially larger cost saving. There is no need to increase the column diameter for reactive distillation, provided the catalyst does not substantially change the fraction of open area present on each stage (trays or packings), as the flooding factors are similar to the non-reactive case. Additional savings can be realised from the effective elimination of ethanol from the distillate product. This is potentially a decisive advantage of reactive distillation as it removes the incentive to install ethanol recovery equipment.

Reactive distillation has clear advantages for ETBE synthesis in terms of reducing capital cost and improving energy efficiency The advantage is less clear with respect to the increased isobutene conversion that is seen with MTBE synthesis via reactive distillation and there is, in fact, a reduction in ether purity. Overall, the benefits of reactive distillation are apparently less significant than for MTBE synthesis.

6.2.3.4 ETBE-Specific Reactive Distillation Designs

The explanation for the enhanced benefits of reactive distillation for MTBE compared with ETBE is that the chemical and phase equilibria coincide more exactly for the MTBE system. This is a necessary requirement for the successful implementation of reactive distillation. The azeotrope between methanol and various C4s helps to lift methanol towards the reaction zone and maintains favourable reaction zone conditions. Since an equivalent azeotrope does not exist in the ETBE system, stripping separation must be optimised. If the stripping separation is too high, the reaction zone is starved of ethanol. If the stripping separation is too low, the boiling point of the ether product is suppressed by the inclusion of n-butene and other C4s.

Operating an ETBE column with less stripping separation effectively increases the conversion at the expense of the ether purity. This case was simulated by modelling a reactive distillation column with only 12 ideal stages. The resulting column will have a significantly lower capital cost than the 30 stage column simulated previously but its ether product will contain some C4s and will only be suitable for gasoline blending when the gasoline pool can absorb the additional light (high RVP) components. The simulation results are presented in Table 6.3 for the case with 5% stoichiometric excess of ethanol.

Table 6.3 - ETBE-Specific Reactive Distillation Designs (Isobutene-Rich Feed)

Low Capital Cost Process

High Conversion, High Purity Process

Reactor Feed Rate (kg/hr)

9,760

9,320

Overall Isobutene Conversion

94.6%

99.0%

Bottoms Composition (wt%)

5.1% EtOH, 0.8% DIB, 0.32% C4s

2.8% EtOH, 0.6% DIB, 0.06% C4s

Distillate Composition (wt%)

3.2% iBut, 96.6% other C4s, 0.13% EtOH, 0.08% ETBE

0.54% iBut, 99.1% other C4s, 0.36% EtOH, 0.04% ETBE

Oxygen Production Rate (kg/hr)

880

835

Net Alcohol Consumption (kg/kg etherj

0.501

0.478

Net Isobutene Consumption (kg/kg ether)

0.781

0.745

Number of Ideal Stages

12

18

Reflux Ratio

1.98

5.0

Reboiler Duty (MW)

1.40

2.64

Column Diameter (mm)

1050

1350

Another consequence of the reduced separation in the low capital cost reactive distillation column is an increased ethanol concentration in the distillate. Under some circumstances, this may require the addition of suitable recovery equipment, negating the capital savings gained through the reduced column height. However, if the loss of ethanol in the distillate is acceptable (eg. the ethanol eventually ends up in the gasoline pool and downstream catalysts are not affected), an overall decrease in the capital cost, and increase in value added by the process, might be realised by the shorter column. The operating costs remain essentially constant as no increase in reboiler duty was specified.

Clearly, the best reactive distillation column design for ETBE synthesis would produce both a high isobutene conversion and a high ether purity. This cannot generally be realised with the same equipment used for MTBE production, unless the MTBE design was particularly serendipitous. The ideal ETBE column design produces a moderate separation but uses a high reflux ratio to continuously recycle isobutene to the reaction zone. This is not required for MTBE as the methanol azeotropes create good reaction conditions under most operating environments. The simulation results for a column with 18 ideal stages and a high reflux ratio are also shown in Table 6.3 (High Conversion, High Purity Process). Table 6.4 summarises the results from the main process alternatives (ETBE synthesis via the conventional route, with a modified MTBE reactive distillation process and with an ideal ETBE reactive distillation process) for the case with a stoichiometric excess of ethanol.

The benefits offered by a specially designed ETBE process are significant: up to 10% higher conversion and 3-6% higher ether purity. This results in 9% lower ethanol consumption and 9% lower isobutene consumption. However, a larger diameter column (1350 mm 0 compared with 900 mm 0 for the other processes) is required and the operating costs will rise accordingly. The capital cost of the ideal ETBE column would probably be higher than the base case MTBE column, even though there are 40% less separation stages, due to the increased column diameter required. However, a very low capital cost option (Table 6.3: 12 stages, no increase in reboiler size and no recovery equipment) remains feasible. The ideal ETBE design is flexible enough to be used for MTBE production, but some reduction in performance should be anticipated (slightly reduced isobutene conversion).

Table 6.4 - C omparison of ETBE Processes hor Isobutene-Rich Feeds

(Dual Reactors)

Modified MTBE Reactive Distillation Process

Ideal ETBE

Reactive Distillation Process

Reactor Feed Rate (kg/hr)

10,260

10,200

9,320

Overall Isobutene Conversion

90.0%

90.5%

99.0%

Bottoms Composition (wt%)

6.8% EtOH, 0.7% DIB, 0.01% C4s

7.1% EtOH, 1.6% DIB, 0.01% C4s

2 8% EtOH, 0.6% DIB, 0.06% C4s

Distillate Composition (wt%)

5.6% iBut, 94.3% other C4s, 0.15% EtOH, nil ETBE

4.9% iBut, 95.1% other C,s, 2 ppm EtOH, nil ETBE

0.54% iBut, 99.1% other C4s, 0.36% EtOH, 0.04% ETBE

Oxygen Production Rate (kg/hr)

913

920

835

Net Alcohol Consumption (kg/kg ether)

0.526

0.523

0.478

Net Isobutene Consumption (kg/kg ether)

0.821

0.816

0.745

Number of Ideal Stages

30

30

18

Reflux Ratio

1.20

1.79

5.0

Reboiler Duty (MW)

1.40

1.40

2.64

Column Diameter (mm)

900

900

1350

Was this article helpful?

0 0

Post a comment