PRODUCTION OF ORGANIC OR AMINO ACIDS BY
BIPOLAR MEMBRANE ELECTRODIALYSIS
The low cost production of many organic or amino acids is desirable for many industrial applications: this includes food acidulation, animal feed, and biodegradable plastics (such as polylactates). Some of these simple aliphatic acids, such as lactic, acetic, succinic, or citric acids, as well as glycine, can be produced by fermentation of various carbohydrate sources. In these processes, as well as in the chemical syntheses, bipolar membrane electrodialysis can play a major role to simplify the process, eliminate by-products, and avoid the use of additional chemicals.
Also, some salts of organic acids are produced as by-products of other chemical syntheses. For example, sodium formate is a by-product of the production of pentaerythritol and can be sold as is or upgraded to the organic acid by bipolar membrane electrodialysis.
BIPOLAR MEMBRANE ELECTRODIALYSIS:
THE TIME HAS FINALLY COME !
by: Bernard Gillery, Dr. Mathieu Bailly, and Daniel Bar
SUMMARY
back to top
Over the last 15 years, there have been many articles, papers, presentations, etc. about Electrodialysis with Bipolar Membranes (EDBM). This innovative electrochemical process has stimulated the imagination of many by its ability to economically reverse neutralization and eliminate salt effluents, while simultaneously recycling the acid and base values. In addition it can be used to acidify or basify process streams without adding another chemical. However, the number of installed EDBM plants worldwide has been relatively small until now. This is due to many reasons, such as high investment and operating costs, a perceived unreliability of the technology, scepticism in the industrial world, and the limited range of economical/practical applications.
We believe that the number of EDBM plants in operation should greatly increase in the next few years. Eurodia and Ameridia, its American division, have demonstrated that it is possible to design EDBM that can run continuously, reliably, and with long membrane life. The significant reduction in the operating costs leads to greatly improved economics; thus, many new applications can be considered even if, at this time, the technology is still mainly suitable for the production or recovery of specialty and fine chemicals. This presentation will focus on the operation since 1997 of an EDBM plant for the production of an organic acid supplied by Eurodia.
Also, to be applied effectively, EDBM cannot often be considered as a single process step and must be used in innovative processes combining also other separation techniques, such as conventional electrodialysis, crossflow microfiltration, ion exchange resins, etc.
1. HISTORY OF ELECTRODIALYSIS WITH BIPOLAR MEMBRANE
back to top
Over the past 15 years more than a dozen known commercial plants totaling about 2,500 m˛ of bipolar membranes have been installed throughout the world. The membranes came from two suppliers: Aqualytics and Tokuyama Corporation (Neosepta® commercialized since 2004 by ASTOM Corp.).
The following table reviews the (known) different plants installed worldwide and the corresponding estimated membrane area. The first bipolar system was delivered in 1986 to Washington Steel (USA) for the recovery of hydrofluoric and nitric acids from stainless steel pickling liquor and the most recent one has been installed in China for the production of an organic acid. In the last few years, some of the plants in operation in the USA have been closed due mainly to overall plant shutdowns, product line changes, or other economical reasons. Therefore, we know of about 1,850 m˛ of bipolar membrane currently (or soon) in operation, mainly for the production of specialty and fine chemicals, such as amino and organic acids.
The limited number of EDBM plants in operation is the result of many factors, mainly industry concerns over reliability and cost. The relatively high investment of EDBM systems limits its use so far to higher value products. However, improvement in stack and plant design, as well as overall process design, allows its use for the production of a growing number of products. Already, four new plants have been installed in the last two years.
| |
|
EDBM PLANTS INSTALLED |
|
Year
|
USA |
FAR EAST |
EUROPE |
|
1986
1994
1995
1996
1997
1998
1999
2000
2001
2002 |
Pickling liquor recovery (stainless steel)*
HF recovery (chemical industry)*
Organic acid production (agro industry)*
Inorganic acid production (chemical industry)*
Organic acid recovery (specialty chemical industry)
Organic acid production (specialty chemical industry)
Organic compound production (fine chemical industry) |
Organic acid production(specialty chemical industry)
Organic acid production(specialty chemical industry)
Amino acid production (pharmaceutical industry)
Organic acid production (China) (agro industry) |
Methane sulfonic acid production (Italy) (specialty chemical ind.)
Organic acid production (France) (agro industry)
Amino acid production (France)
Organic acid production (Czech Rep) (agro industry)
Organic acid recovery (Germany) (pharmaceutical industry) |
Total installed bipolar membrane area (estimated) |
1600 m˛
(*): closed |
650 m˛ |
700 m˛ |
|
|
|
|
|
|
2. EDBM A COST-EFFECTIVE SOLUTION FOR THE PRODUCTION OF SPECIALTY AND FINE CHEMICALS
| 2.1. Principle
back to top |
|
As a quick reminder, bipolar membranes are formed by combining an anion-exchange and a cation-exchange layer to allow for the effective splitting of water into the hydrogen and hydroxyl ions. Combined with anion and cation-exchange membranes in an electrodialysis stack, it is possible to convert aqueous salt solutions into acids and bases. In a three-compartment configuration, there are three separate loops: salt, acid, and base.
The figure below presents the principle of EDBM operating in the two-compartment configuration with bipolar and cation exchange membranes. This configuration is the most appropriate to cost-effectively convert a weakly dissociated organic acid salt into its acid form without addition of a strong acid. Indeed, this configuration allows limiting the product losses to a maximum of 2%. Furthermore, a co-product of base constituted by the counter ion of the salt is also generated. Consequently, at the end of the EDBM stack, two streams are recovered without any environmental discharge.
|
|
|
 |
.
|
2.2. Criteria for the application of EDBM
back to top |
|
The effective utilization of EDBM requires the following conditions:
· Total multivalent cations content in the treated fluid < 1 – 5 ppm
| Indeed, the multivalent cations may precipitate by association with OH- ions in the stack. This precipitation may occur in the cation exchange membrane during the ionic transfer from the acid to the base compartment, thus destroying this membrane. It is to be noted that the bipolar membrane is not affected by this problem. |
·
Feed concentration in organic acid salt > 1 eq/l
| This specification allows a reduction of the membrane area required for the conversion because the current density can remain high enough during most of the conversion. This leads to a decrease of both the investment and operating costs. |
·
Operating temperature < 40°C
·
No oxidizing compounds
These requirements are related to the materials used to manufacture the membranes.
|
3. A SUCCESSFUL EDBM INDUSTRIAL PLANT
back to top
Eurodia Industrie delivered an EDBM plant in 1997 in France for the production of an organic acid from its sodium salt. The acid, produced by fermentation, must remain confidential as well as the name of the customer. The plant was initially designed to produce 2,600 mT/year of acid (100%) over 8,000 hours of operation. Two EUR20-240 stacks with 81 m2 of effective cell area were initially installed in the two-compartment configuration (see above) with Neosepta® CMB cation-exchange and BP1 bipolar membranes. The acid conversion rate (purity) is 98 % with a concentration of 390 g/l: this is equivalent to a final conductivity of 3 mS/cm. The NaOH was initially produced at a concentration of 6 w% to be reused in the fermentation (most microorganisms do not like acidic conditions). The plant initially operated in a batch mode to consistently meet the customer requirements. The initial battery-limit investment was 840,000 Euros (~$820,000).
After several years of operation a membrane life of 20,000 hours was reached for the bipolar membranes, while the cation-exchange membranes have been replaced after 18,000 hours. The electrodes have not been changed until now. The customer has increased the capacity in several steps by adding several EUR20 and EUR40 stacks and an additional capacity expansion is being considered. By changing the operation from batch mode to feed & bleed, the base concentration could be increased from 6 w% to 8 w% with the same current efficiency (for a new plant, the NaOH could be produced at 10 w%). The main operating costs are power at 0.88 kWh/kg of produced acid and membrane (electrodes) replacement at $0.09/kg. These results show that properly designed and operated EDBM
plants can be very attractive.
4. EDBM INCLUDED IN COMPLETE PROCESSES
back to top
As previously mentioned, the efficiency of EDBM, as a single unit operation, has been demonstrated for the recovery/production of specialty chemicals. However, to propose overall cost-effective solutions, EDBM has frequently to be considered as one step of a complete process.
As an example, the process below could be proposed for the production of an organic acid, such as lactic acid starting from sodium lactate produced by fermentation. The heart of the process is the conversion step performed by EDBM. To meet the feed requirements of EDBM (see above), the fermentation broth is first clarified by microfiltration. Then, the divalent cations content is lowered to a range of 1 to 5 ppm. The target species, i.e. sodium lactate, are concentrated and purified by conventional electrodialysis. After the conversion of sodium lactate into lactic acid, the product is purified by ion exchange resins.
|
|
 |
It is to be noted that the environmental impact is greatly reduced compared to traditional precipitation processes. Indeed, recycling can be optimized to the extent that the only environmental discharge is due to the use of ion exchange resins. Furthermore, the effluent generated by the divalent cation removal step can be reduced by coupling both ion exchange and membrane techniques.
CONCLUSIONS
back to top
In summary, the results of fifteen years of commercialization are:
- A dozen commercial plants in operation but growing demand due to improved economics resulting from better design for ED stack and operating conditions.
- In the future, EDBM must be considered as a step of a complete process and not as a single unit operation.
- Promising future for the production or recovery of organic and amino acids, as well as other specialty / fine chemicals.
|
