May-2014

Process improvements in the direct hydration of olefins to alcohols

Synthetic alcohols produced by direct hydration have wide application in the process industries as solvents, in the production of cosmetics and as intermediates in the manufacture of other chemicals.

Richard Kolodziej
Wood Group Mustang

Article Summary

Author has been involved with these direct hydration technologies for a number of years. Some ways to significantly increase yields and/or decrease energy consumption in these processes have been discussed.

Direct hydration for the manufacture of synthetic alcohols from olefins is a mature technology. Indeed, it has been the technology of choice for producing synthetic alcohols for the last 30 years. Typical applications include the conversion of ethylene to ethanol, propylene to isopropanol and isobutylene to butanol.

The olefin feedstock price accounts for 70-80 per cent of the cost of producing synthetic alcohol. The amortised plant costs and other operating costs contribute about 10 per cent each to the alcohol cost. So, although the olefin feed price is a major factor, capital and operating cost savings can still substantially enhance the profitability of a synthetic alcohol production unit.

Typical Process
A typical direct hydration flow sheet (shown in Figure 1) consists of:
• Reaction section
• Recovery and purification sections
• Dehydration section.

Reaction Section
The reaction section normally consists of multi-reactor trains where the olefin and steam are reacted to give the corresponding alcohol. There are other side reactions also taking place which produce impurities such as: alkanes, ethers, aldehydes, ketones, heavy alcohols and, to a much lesser extent, polymers. The conversions using traditional acid-based catalysts are in the range of 5-8 per cent of the feed olefin. This then results in the requirement for a significant recycle of unconverted olefin which, for a given unit capacity, directly impacts the size of the reactors, heat exchangers, scrubbing towers and recycle gas compressors.

The reaction section also consists of feed effluent exchangers, which heat the reactor feed by heat recovery, and a fired heater which further heats the reactor feed to the desired inlet temperature.

Process Improvements to the Reaction Section
Potential improvements to the reaction section can lead to significant energy savings, along with lower capital costs in the case of a new plant and potential increased production for an existing plant.

Catalyst systems
Many direct hydration catalyst systems use a base impregnated with phosphoric acid. Among other effects, reaction conversions for these systems are related to the acid strength on the catalyst. A 55-65 per cent acid concentration is typical for good conversion and selectivity. With the presence of water in the reaction, however, acid continuously strips off from the catalyst during normal operation and thus needs to be replaced. For many existing plants, this replacement is done by feeding phosphoric acid into the reactor inlet line, usually at the start of each shift. The reactor effluent is sampled and the amount of acid is measured. (A typical acid range for the reactor effluent is 250-350 wppm). This approach has four drawbacks:
1. The reactor conversion may drop by about 20 per cent from start to end of shift and this 1 per cent reduction in conversion per pass reduces the capacity of the unit by 20 per cent, based upon a given size of the recycle equipment;
2. Overdosing of acid is a common problem, which not only costs money (phosphoric acid usage is often twice as much as should be required), but also encourages unwanted polymeric olefinic reactions, i.e. selectivity can suffer;
3. The excess acid results in additional wastewater treatment costs; and
4. The major fluctuation in reactor yields is a main reason why multi-reactor train designs are typical. Acid dosing to the multi-reactor trains is staggered and thus the combined reactor effluent stream averages out these yield fluctuations and the combined product composition is more constant. A design without these yield swings could consider single train unit design, which would cost less and also reduce the complexity of unit operations.

Recent development has looked at a low rate, well-controlled, well-mixed, continuous acid addition system that would alleviate these drawbacks.

Thus, by using improved acid dosing systems, the activity of the catalyst can consistently be maintained at high (but not too high) levels. With a per pass conversion increase of 1 per cent, the capacity of the synthesis section can nominally be increased by 20 per cent with essentially no other modifications.

Recent promising work in zeolite catalyst technologies could result in hydration designs with much higher per pass conversions, e.g. 20 to 25 per cent. Although these catalysts still warrant further testing and evaluation, including the fact that these systems may require frequent catalyst regeneration, phosphoric acid-based catalyst systems may eventually be replaced by these more reactive zeolite catalyst systems. Significantly higher per-pass yields could be used to reduce energy costs on existing units by decreasing the recycle flow of unconverted olefins by a factor of two to four. Alternatively, a zeolite catalyst conversion could be used to more than double unit capacity. However, although the reactor systems may be adequate in size, the downstream fractionation systems would still need to double in capacity and thus would require a major debottleneck. Using some of the heat exchange and distillation tower revamp suggestions discussed elsewhere, these fractionation system debottlenecks could be done quite cost effectively as well.

In summary, a zeolite catalyst conversion design would either lower energy consumption by reducing recycle requirements and/or increase reactor charge capacity. Therefore, for either case, the overall cost of alcohol production could be reduced by at least 20 per cent.

Heat recovery
Worthwhile fuel savings are achievable by optimising heat recovery schemes in reaction synthesis, as well as in the recovery and purification sections. Commercially proven specialised exchangers can be used to enhance heat recovery. For example, if specified correctly, welded plate exchangers and/or plate type exchangers designed within pressure vessels can be used instead of shell and tube heat exchangers for some feed-effluent and reboiler services. Heat transfer coefficients can be improved by 200 to 400 per cent, thus achieving lower approach temperatures.