Acetic Acid, Acetaldehyde, Acetic Anhydride, and Vinyl Acetate

---

Chapter 10

Evolving Production of the Acetyls (Acetic Acid, Acetaldehyde, Acetic Anhydride, and Vinyl Acetate): A Mirror for the Evolution of the Chemical Industry Downloaded by PENNSYLVANIA STATE UNIV on June 29, 2012 | http://pubs.acs.org Publication Date: December 31, 2008 | doi: 10.1021/bk-2009-1000.ch010

Joseph R. Zoeller Research Laboratories, Eastman Chemical Company, P.O. Box 1972, Kingsport, TN 37660

Introduction Acetic acid is one of the earliest chemical entities produced by mankind, with intentional production evident as early as 3,000 BC, although it was likely produced concurrently with the earliest attempts at wine making (ca. 10,000 BC.) The earliest applications were to preserve food by pickling, but its use in metal refining and some studies on its properties appear from ancient Greece and the period of alchemy. Given its antiquity and long history of utility to mankind, acetic acid would appear to be a natural early choice for production and utilization by the chemical industry. As a consequence, the production of acetic acid and its related products and starting materials have mirrored the development of the chemical industry, reflecting the availability of new resources, the development of new technologies, and the changing needs of the various industries dependent upon chemicals. Indeed, changes in its production methods and those for its key derivatives have often been harbingers of changes in the industry. In this chapter, we will discuss the evolution of processes for the production of the acetyl chemicals - acetic acid (AcOH), acetic anhydride ( A c 0 ) , acetaldehyde (AcH), and vinyl acetate (VA). These four materials coevolved in a series of synergistic relationships which ultimately led to the modern acetyl stream which now exceeds 6 X 10 kg/yr of acetyl (as acetic acid equivalents) per year and will be discussed in evolutionary context with each other. Opportunities and resultant innovations in the chemical industry are normally created when lower cost raw materials become available or existing products fail 1a

1a

2

9

© 2009 American Chemical Society

In Innovations in Industrial and Engineering Chemistry; Flank, W., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2008.

365

366 to meet the needs and requirements generated by new industries and products. The production and interplay of the acetyl products is a mirror of the evolution of the chemical industry as a whole and we shall discuss how each process evolved over time. (There are numerous good sources for more detailed overviews of these processes provided in the references. " ) 1

5

Downloaded by PENNSYLVANIA STATE UNIV on June 29, 2012 | http://pubs.acs.org Publication Date: December 31, 2008 | doi: 10.1021/bk-2009-1000.ch010

Acetyl Processes 1700-1910. Using What Nature Provided - Acetic Acid from Natural Sources. Like the chemical industry, the roots of acetyl production start with the production of acetic acid by natural processes. While we already discussed the ancient history of acetic acid production, larger scale production and isolation began in the 1700's with the "Orleans" (or "Slow") process which consisted of intentional fermentation of wines in a semi-batch process in which the wine was added in increments until the barrel was full. However, this was displaced by the "German" (or "Fast") process in which wine was continuously fed over charcoal or wood chips and oxidized with a continuous flow of air to increase the activity of the bacteria. (As a continuous process, the "German" process likely represents one of the earliest examples of a continuous reactor system.) These fermentation processes were reportedly capable of producing up to 10 tons/day of product. ' The fermentation processes were ultimately displaced in the late 1800 s when destructive (pyrolytic) wood distillation, which co-produced methanol, acetic acid, methyl acetate, acetone, methyl ethyl ketone, and creosote, became the common method of production for acetic acid. While these natural sources served mankind for several millennia, and research continues on fermentation processes, they were inefficient processes which generated large amounts of waste and consumed large amount of energy. Further, none of these processes produced glacial acetic acid which had to be produced by precipitation as the calcium salt and subsequent acidification with, and distillation from, sulfuric acid. However, demand for glacial acetic acid was relatively small in this period and this process sufficed until the beginning of the 20 century. 1 6

f

6

th

The Acetylene Period (1910-1950). Genesis of Acetyls as a Chemical Enterprise (1910-1920) New enterprises and rapid innovation in the chemical industry are normally created when existing products fail to adequately meet the material needs and requirements generated by new industries or products. The newly introduced product then finds numerous new, often unanticipated, applications. Slower,

In Innovations in Industrial and Engineering Chemistry; Flank, W., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2008.

367 more evolutionary, innovations occur when lower cost raw materials become available, or new, more efficient, technologies are developed. In the case of acetyls, all of these events occurred nearly simultaneously in the period 19ΙΟ­ Ι 920 and led to the parallel development of an array of successful processes. Let's begin with the advent of the new raw material and a new technology. At the dawn of the 20 century, acetylene, which was generated by heating coal and lime and subsequently hydrolyzing the calcium acetylide (also called calcium carbide), became available as the first simple building block for the chemical industry. The first fully synthetic process for generating acetic acid involved the addition of water to acetylene in the presence of a mercuric salt and sulfuric acid and subsequent air oxidation of the resultant acetaldehyde. (See equations [1] and [2].) This process was first commercialized in 1911 and provided the first readily available source of glacial acetic acid. By 1916, the acetaldehyde portion of the process was improved by the addition of iron which aided the reoxidation of mercury. The oxidation of acetaldehyde would remain the predominant means of generating acetic acid for the next 60 years, although the raw materials leading to acetaldehyde would change over time. While this potentially made large amounts of glacial acetic acid available, there was limited demand for the glacial acetic acid product and its primary outlet was to make acetone by the pyrolysis of calcium acetate.

Downloaded by PENNSYLVANIA STATE UNIV on June 29, 2012 | http://pubs.acs.org Publication Date: December 31, 2008 | doi: 10.1021/bk-2009-1000.ch010

th

HC^CH + H 0 2

A c H + Vi 0

2

H SQ4,FeS04, H2SQ4 ^ g

•

^

r

AcOH

jj

[2]

The commercial driving force which created the acetyl industry was provided by the onset of World War I and the fledgling aviation industry. While World War I temporarily increased the requirements for acetic acid to make acetone (needed to make "cordite"), the permanent change would result from requirements in aviation. At the dawn of the 20 century, outside of a few natural polymers and monomers which had limited utility, the only plastic available for films, coatings, and plastics was cellulose nitrate. While the first planes used fabric wings, they were not strong enough and were too porous, so cellulose nitrate was applied to form a lightweight wing, which today we would recognize as a fiber reinforced plastic. Given the highly flammable (sometimes explosive) nature of cellulose nitrates, there were obvious problems with their use and the early airplanes were firetraps which could ignite in midair. The advent of World War I, and the obvious unsuitability of cellulose nitrate in a war environment, led to a sudden demand for a new chemical entity, not yet commercially available, to replace cellulose nitrate. While of little use at the time of its discovery, the solution to this problem was provided by a discovery by Miles and Eichengrun , who demonstrated in 1904 that cellulose triacetate (made by sulfuric acid catalyzed acetylation of cellulose with acetic anhydride) could be made soluble (in solvents like acetone and acetic acid) by partial hydrolysis of a portion of the acetate functionality. th

7

In Innovations in Industrial and Engineering Chemistry; Flank, W., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2008.

Downloaded by PENNSYLVANIA STATE UNIV on June 29, 2012 | http://pubs.acs.org Publication Date: December 31, 2008 | doi: 10.1021/bk-2009-1000.ch010

368 Cellulose acetate was stable, not readily flammable, and water repellent, so it was a suitable substitute for cellulose nitrate. Further, in its soluble form, cellulose acetate could be applied by simply painting the fabric as had been done with cellulose nitrate. However, cellulose acetate manufacture would require large quantities of acetic anhydride, and there was no good way to produce acetic anhydride other than the existing neutralization of calcium acetate with sulfuric acid. What ensued would be one of those creative bursts that occur sporadically in the chemical industry. Expanding on the mercuric sulfate - sulfuric acid technology, it was known as early as 1912 that addition of a single equivalent of acetic acid to acetylene in the presence of mercuric sulfate and sulfuric acid resulted in the generation of vinyl acetate (a mere laboratory curiosity at the time) whereas addition of two or more equivalents of acetic acid provided ethylidene diacetate (1,1diacetoxyethane, E D A ) . E D A was known to evolve acetaldehyde upon heating in the presence of acids. Upon complete removal of a mole of acetaldehyde, an equimolar amount of acetic anhydride could then be distilled from the residue. If one balances the sequential processes as shown in equations [3] through [6], one finds that the process is in stoichiometric balance for conversion of acetylene and air to cellulose acetate without by-products. (In practice there is some net acetic acid production since cellulose is never completely dry.) HC=CH + 2 A c O H (AcO) CHCH 2

AcH + V 0 2

•

(AcO) CHCH 2

3

• Ac 0 + AcH

3

•

2

Cellulose + A c 0 2

[3]

2

[4]

AcOH

[5]

• Cellulose-Acetate + A c O H

Net Reaction: Cellulose + HC=CH + V 0 2

• Cellulose-Acetate

2

[6]

[7]

The chemistry is more complex than would appear on the surface. When acetic anhydride reacts with acetaldehyde, it forms E D A in an equilibrium that favors E D A (Equation [8]), but a subsequent disfavorable equilibrium follows which forms acetic acid and vinyl acetate (Equation [9]). These equilibrium constants indicate that E D A is the most thermodynamically favored product. In the presence of acids, these equilibria occur rapidly. Using LeChatelier's principle, if the most volatile product, acetaldehyde, is continuously removed you can shift the equilibrium until only acetic anhydride remains which can then be distilled independently. (The equilibrium between E D A and vinyl acetate would prove to be pivotal in the later development of a vinyl acetate process.) AcH + A c O ^ Z ± (AcO) CHCH 2

2

3

K

eq(14

o°c) = 25

In Innovations in Industrial and Engineering Chemistry; Flank, W., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2008.

[8]

369 (AcO) CHCH 2

AcOH+(AcO)HC=CH

3

2

K

=

eq(14

o ° c ) 0.01

[9]

While the acetylene based route to acetic anhydride would remain in practice in some locations until the 1940's, a competitive technology was developed within the same decade that would dominate production even today. In this period, chemists discovered that the highly endothermic generation of ketene could be accomplished by heating either acetone (equation [10]) or acetic acid in the presence of a phosphate catalyst (equation [11]) at very high temperature (>700°C). Subsequent reaction of ketene with glacial acetic acid gave acetic anhydride (equation [12]).

Downloaded by PENNSYLVANIA STATE UNIV on June 29, 2012 | http://pubs.acs.org Publication Date: December 31, 2008 | doi: 10.1021/bk-2009-1000.ch010

4

Acetone to ketene; C H C ( = 0 ) C H — • H C=C=0 + C H 3

3

Acetic acid to ketene: CH C(=0)OH 3

2

• H C=C=0 + H 0 2

Acetic anhydride from ketene: H C=C=0 + A c O H • 2

[10]

4

[11]

2

Ac 0

[12]

2

While acetone would serve as a source of ketene in a number of locations, the acetic acid dehydration would predominate and the acetic acid based ketene process is still widely practiced at the start of the 21 century. While acetone had some attractive features, particularly the generation of an inert co-product (methane) rather than reactive water which can destroy ketene, there are some sound reasons the acetic acid process predominated. First, let's recall that during the period 1910-1920, representing the time these processes were introduced, acetone was still made from calcium acetate, so acetone was obtained, at the time, in a two step process from acetic acid. The choice of acetic acid skipped two steps and eliminated wastes. Further, the acetic acid based ketene process was also aided by a chemical engineering advance. Chemical engineers demonstrated that glacial acetic acid could be obtained by dehydration with azeotroping agents and subsequent distillation. Glacial acetic acid was now available even i f the source was very wet. This had multiple impacts. First, the advent of azeotropic drying enabled facile dehydration of the wet acetic acid co-product obtained in the ketene generation. Second, recall that the useful form of cellulose acetate required partial hydrolysis of the cellulose triacetate. The addition of water to the cellulose acetate product mixture simultaneously quenched any excess acetic anhydride used in the reaction, performed the necessary hydrolysis, and removed the acetic acid by-product obtained during the initial acetylation. However, this hydrolytic step resulted in a very wet acetic acid stream that required recovery and recycle. Azeotropic distillation permitted the manufacturer to recycle this acetic acid co-product for use in acetic anhydride st

In Innovations in Industrial and Engineering Chemistry; Flank, W., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2008.

370 production rather than discarding it or selling it at a loss. (Until the 1950's, the market for acetic anhydride and cellulose acetate was much larger than that for acetic acid. Therefore, any process that generated a glut of acetic acid could not be absorbed by the market.) Lastly, this technology also reinvigorated the wood distillation sources since wood distillation could now provide glacial acetic acid and diversify available feedstocks for creating acetic anhydride. The feedstock picture further diversified in 1920 with the commercialization of ethanol dehydrogenation to generate acetaldehyde. (The process was conducted in the vapor phase at 260-290°C using copper-chromite catalysts.) While the process was known as early as 1886 , the development of adequate catalysts for the endothermic process would take nearly 35 years. Subsequent oxidation to acetic acid provided an additional source of acetic acid. These technologies would largely stay in place with only minor modification until the 1950's. A summary of the chemical routes to the various acetyls in 1920 is shown in Figure 1.

Downloaded by PENNSYLVANIA STATE UNIV on June 29, 2012 | http://pubs.acs.org Publication Date: December 31, 2008 | doi: 10.1021/bk-2009-1000.ch010

3b

EtOH

•

AcH + H

ΔΗ= +19.7kcal/mol

2

EtOH

[13]

Wood -H

Heat

2

V2

H C=CH(OAc)

o

I

2

AcH

2

•

AcOH

\ /

Λ»

AcOH

• • • · • Acetone

HC^CH

H C=O0 2

2AcOHV

^AcOH ^Ac 0 2

j (-AcOH) Cellulose acetate Figure 1. Acetyl Production Circa 1920.

Maturation of Cellulose Acetate and the Introduction of Vinyl Acetate. (1920-1950) As often happens when a new material is introduced, there was a rapid expansion into new, often unanticipated, applications. Cellulose acetate was no

In Innovations in Industrial and Engineering Chemistry; Flank, W., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2008.

Downloaded by PENNSYLVANIA STATE UNIV on June 29, 2012 | http://pubs.acs.org Publication Date: December 31, 2008 | doi: 10.1021/bk-2009-1000.ch010

371 exception. In the period 1920-1935, cellulose acetate and related esters helped usher in the era of plastics when cellulose acetate became the first thermoplastic available for molded parts. Cellulose acetate was spun into textile fibers, cast into films for packaging and photography, used to generate molded parts, and saw expanded applications in coatings and reinforced plastics. Over the next 30 years, wood distillation declined relative to ethanol dehydrogenation and acetylene based processes as a source of acetyls, but all three would contribute to the acetyl supply chain required to generate the acetic anhydride needed to meet the market demands for cellulose acetate as the product grew and matured over the period 1920-1940. However, while cellulose acetate and cellulose ester markets would grow, improvements in technology for the basic acetyl products were limited during this time period. Long term, the most significant new development in the acetyl market in this period was the introduction of vinyl acetate. While the liquid phase addition of acetic acid to acetylene in the presence of a mercuric sulfate catalyst at 60100°C to generate vinyl acetate was first discovered in 1912, it was introduced as a commercial scale process in 1925 and introduced vinyl acetate to the market as a commodity scale product. However, the mercury process was inefficient and toxic and did not last long as a commercial process. In 1921, it was discovered that Zn acetate on activated charcoal could catalyze the addition of acetic acid to acetylene in the vapor phase. By 1940, sequential improvements in the stability of activated charcoal provided Zn on carbon catalysts that were sufficiently stable to render any remaining Hg based processes untenable. By about 1950 the mercury based process was extinct and completely replaced with the vapor phase Zn on activated charcoal process which was operated at 170210°C and pressures just exceeding 1 atmosphere with an excess of acetylene. However, vinyl acetate still represented a relatively small portion of the market in acetyl related products. There were also improvements in acetaldehyde and acetic anhydride manufacture. A g based catalysts for the partial oxidation of ethanol became available around 1940. When used to oxidatively dehydrogenate ethanol [14], the conversion of ethanol to acetaldehyde was no longer equilibrium limited since the reaction was now very exothermic. Fortunately, the process still displayed excellent selectivity (ca. 93-97%) for acetaldehyde. This technology replaced the older Cu-Cr processes over the period of the 1940-1950 and made ethanol a much more attractive resource for acetaldehyde. When ethylene became available as a feedstock in the 1940's through 1950's, ethanol became cheaply available via ethylene hydration (as opposed to traditional fermentation). With ethanol now cheaply available from ethylene, the advent of the A g catalyzed oxidative dehydration to acetaldehyde rapidly accelerated the shutdown of the last remaining wood distillation units. EtOH + V 0 2

4 0 0 2

c

° '

A g

» AcH

ΔΗ = - 58 kcal/mol

In Innovations in Industrial and Engineering Chemistry; Flank, W., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2008.

[14]

Downloaded by PENNSYLVANIA STATE UNIV on June 29, 2012 | http://pubs.acs.org Publication Date: December 31, 2008 | doi: 10.1021/bk-2009-1000.ch010

372 Acetaldehyde oxidation was also marginally improved, especially for the manufacture of acetic anhydride. In 1935, workers at Shawingan Chemicals discovered that the oxidation of acetaldehyde, i f conducted in the presence of cobalt, copper, or better yet, a mixture of the two catalysts, yielded a mixture of acetic anhydride and acetic acid providing the water co-product was rapidly separated by azeotropic distillation, normally with a compatible material such as ethyl acetate. It would not be until the 1940's that this became widely practiced, but the process was eventually widely adopted. While experimental units produced ratios of acetic anhydride: acetic acid as high as 4:1, it appears that the commercial process normally gave a 5:4 mixture of acetic anhydride: acetic acid under normal operating conditions. The poor selectivity reflects both an inefficiency in the catalytic process and the difficulty in separating water rapidly enough to prevent hydrolysis of acetic anhydride to acetic acid on a commercial basis. However, this still represented a significant improvement since it reduced the amount of ketene that needed to be generated via the dehydration of acetic acid to ketene during the production of cellulose acetate. 3b

The Petrochemical Era (1945-1970). The dawn of the petrochemical industry is normally marked as 1940 in the United States and 1950 in Europe and Japan. Throughout this period, the chemical industry rapidly phased out coal and acetylene as its raw material base and shifted to using products obtained from petroleum cracking and fractionation. The new key building blocks would be hydrocarbons, olefins (especially ethylene and propylene) and the aromatics (benzene, toluene, and xylene) which were cheaper and cleaner to obtain. As expected, this change affected the methodology for acetyl production and a new set of processes began to emerge. However, the emergence of two technologies had as profound an affect on the production methods for acetyls as the change in feedstocks. First, the pioneering work of Rolen in cobalt catalyzed hydroformylation (the addition of CO and H to olefins to generate aldehydes discovered in 1938) and Reppe (the nickel and cobalt catalyzed carbonylation of alcohols, olefins, and acetylene to carboxylic acids during the period ca. 1940) ushered in the era of organometallic chemistry and modern transition metal based homogeneous catalysis. The advent of organometallic chemistry and transition metal homogeneous catalysis would profoundly affect the methods for the production of all acetyls. The second major event was the development of new materials of construction for reaction vessels, especially those resistant to halide induced corrosion. The carbonylation work of Reppe, while promising, used iodine and very high temperatures and pressures. In the original schemes this required silver lined reactors, but the advent of T i and a selection Ni-Cr alloys as materials of construction in the mid 1950's enabled the economic construction of 2

In Innovations in Industrial and Engineering Chemistry; Flank, W., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2008.

373 reactors using halides under high temperature and pressure conditions. While often overlooked, this was one of the most significant events in the history of acetyl chemicals. These two major events would significantly impact methods of production for all the acetyls.

Changes in Acetic Acid Manufacture While reports of the carbonylation of methanol to acetic acid (equation [15]) date back to 1913 (at B A S F ) the first full descriptions appeared in 1926 when workers at Celanese described the carbonylation of methanol over acidic heterogeneous catalysts. None of these processes looked promising at the time. However, in the 1950's, Reppe (BASF) reported a successful homogeneous cobalt catalyzed process operated in the presence of a methyl iodide co-catalyst. The process was commercialized in 1960 and leveraged cheaply available natural gas to generate synthesis gas (a mixture of CO and hydrogen sometimes referred to simply as syngas) via steam reforming or partial oxidation of methane. The synthesis gas was used to synthesize both the methanol feedstock and as a source of C O in the carbonylation.

Downloaded by PENNSYLVANIA STATE UNIV on June 29, 2012 | http://pubs.acs.org Publication Date: December 31, 2008 | doi: 10.1021/bk-2009-1000.ch010

lb

8

MeOH + C O

•

AcOH

[15]

Unfortunately, the process required very high temperatures and pressures (250°C, 680 atm.) and selectivity was not good, providing yields of 90% based on methanol and 70% based on C O . Several plants would be built using this technology between 1960 and 1966, but, at these extreme pressures and temperatures, the plants were very capital intensive and the yields were mediocre. Corresponding improvements in acetaldehyde production (described later in this section) which were also attributable to advances in homogeneous catalysis and the advent of corrosion resistant materials of construction, prevented the cobalt based methanol carbonylation process from displacing acetaldehyde oxidation as a means of producing acetic acid. Another competing acetic acid process also arose in the same time period. About 1957, Celanese began to operate a facility for the oxidation of butane to acetic acid to take advantage of low cost hydrocarbons. The oxidation, which is conducted in the presence of a Co catalyst at 180°C and 15-20 atm. of oxygen at a butane conversion of 10-20% yielded a mixture of products in the following portions: acetic acid formic acid 2-butanone propionic acid

12.5 1.25 2 1

In Innovations in Industrial and Engineering Chemistry; Flank, W., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2008.

374 This process still operates today, but is only viable due to the added value of the by-products and can not compete with modern methanol carbonylation for acetic acid production. There is continued interested even today in the oxidation of hydrocarbons, especially ethane, but these technologies are not competitive with modern methanol carbonylation for the generation of acetic acid.

Downloaded by PENNSYLVANIA STATE UNIV on June 29, 2012 | http://pubs.acs.org Publication Date: December 31, 2008 | doi: 10.1021/bk-2009-1000.ch010

Changes in Acetaldehyde Manufacture Developments in acetaldehyde manufacture allowed acetaldehyde oxidation to remain competitive with, or superior to, Co catalyzed methanol carbonylation and butane oxidation throughout the period 1945-1970 as a means of generating acetic acid. First, as mentioned earlier, ethanol was now available from ethylene via hydration, which lowered the cost of ethanol. Therefore, the oxidative dehydration of ethanol (equation [14]) was now more attractive than when ethanol was derived from fermentation. However, more significant was the introduction of a new one step process for the oxidation of ethylene to acetaldehyde using a PdCl -CuCl catalyst (equation [16]) sometime between 1957-1959. The process, now known as the Wacker Process, is operated at ca. 10 atm and 100-110°C, requires a large excess of chloride, and produces A c H in about 95% yield. The mechanism for this reaction is shown in figure 2. 2

H C = C H + Vi 0 2

2

•

2

AcH

2

[16]

The process displaced ethanol oxidation in new installations and is the predominant route to acetaldehyde today. The role of the various components was well understood. To understand the role of copper, we must first understand that, while thermodynamically feasible, Pd(0) did not readily react with oxygen and required a catalyst. Copper filled the catalyst role, since C u readily oxidized Pd(0) to P d in the presence of excess chloride anion (as HC1) and the resultant C u was readily oxidized by air. (See equations [17] and [18].) The chloride fulfilled two other critical roles. First, it was critical to the oxidation of Pd since the oxidation appeared to occur via a chloride bridge. Second, a large excess of chloride was necessary to maintain C u in a soluble form. (CuCl is insoluble but in the presence of excess chloride, CuCl formed soluble C u C l " , CuCl ", and CuCU " salts.) The reaction required access to corrosion resistant alloys due to the high chloride concentration and tapped into the then newly emerging understanding of homogeneous transition metal catalysis. In the best practice, the reaction was conducted in two stages in which the Pd catalyzed addition of water to ethylene is carried out in the presence of a large excess C u and the reoxidation of the resulting C u is conducted in a separate vessel. However, the reaction was reported to be carried out as a one stage process as well. 2+

2+

+

+

2

2

2 +

3

+

In Innovations in Industrial and Engineering Chemistry; Flank, W., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2008.

3

375

H C=CH 2

2

2CuCl [Cl Pd-CH CH OR] Downloaded by PENNSYLVANIA STATE UNIV on June 29, 2012 | http://pubs.acs.org Publication Date: December 31, 2008 | doi: 10.1021/bk-2009-1000.ch010

3

2

2

2

2 CuCl , 2C1 2

V HPdCl

3

H C=CH(OR)

HCl,2Cr

2

Separate vessel: CuCl - + 2 HC1 + Vi O f * 2

ifR = H CuCl + H 0 + 2 CI" 2

2

AcH

Figure 2. Mechanism for Wacker-Chemie Liquid Phase Acetaldehyde and Vinyl Acetate Process.

2 C u C l + Pd(0) + 2 CP

2 CuCl - + PdCl

2 C u C V + V 0 + HC1

2CuCl + 2Cl"+ 2 H 0

2

2

2

2

2

[17]

2

2

[18]

(For clarity, only the CuCl * is shown in Equations [17] and [18] and Figure 2. Cu(I) may also be present as CuCl *, and CuCl " as indicated above.) 2

2

3

3

4

Changes in Vinyl Acetate Manufacture Until the mid-1950's, vinyl acetate was a moderate volume product used to produce specialty polyolefins, such as cling films. However, in the 1950's, emulsion polymerization ("latex") began to emerge as a major new product line in the chemical industry. (Emulsion polymerization uses surfactants to generate water based emulsions of polyolefin polymers. These products are most familiar to the consumer as water borne paints, but are also found in a wide variety of

In Innovations in Industrial and Engineering Chemistry; Flank, W., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2008.

376

Downloaded by PENNSYLVANIA STATE UNIV on June 29, 2012 | http://pubs.acs.org Publication Date: December 31, 2008 | doi: 10.1021/bk-2009-1000.ch010

5

adhesives, textile sizings, and coatings. ) Three major monomers constituted the preferred feedstocks for this new technology: vinyl acetate, acrylic acid, and styrene. Further spurring demand for vinyl acetate was the advent of polyvinyl alcohol applications (obtained by hydrolysis of polyvinyl acetate) and polyvinyl butyral (made by adding butyraldehyde to polyvinyl alcohol), an integral component of safety glass. These new outlets would ultimately drive US production of vinyl acetate from only 59 K M T in 1956, to 356 K M T in 1970, and 850 K M T in 1980. However, the existing acetylene routes were inadequate to meet demand in the petrochemical era. As normally happens in the chemical industry, in the face of rapid changes in demand led to a proliferation of new processes as producers compete to fulfill the new market. The first new process was introduced by Celanese in the late 1950's. The Celanese vinyl acetate process took advantage of the equilibria: AcH + A c 0 2

(AcO) CHCH 2

3

±=+

(AcO) CHCH

4

A c O H + (AcO)HC=CH Κ«, MeOH

AcOH

2

Vinyl Acetate V2O2

+ AcOH -H 0> 4 MeOAc

H 0 (ketene)

2

coal

Downloaded by PENNSYLVANIA STATE UNIV on June 29, 2012 | http://pubs.acs.org Publication Date: December 31, 2008 | doi: 10.1021/bk-2009-1000.ch010

H C=CH

+ C0

Synthesis Gas

2

+ CO * • Ac 0 2

I

Cellulose Acetate H C=CH + Y 0 2

2

2

2

—•AcH Figure 6. Acetyl Production circa 1985.

Maturity of the Acetyl Stream. Improvements in Methanol Carbonylation. (1985-2007). After 1985 the synthetic pathways for generating the acetyl chain would remain unchanged. The heterogeneous vinyl acetate process and liquid phase acetaldehyde process would remain virtually unchanged into the present with only minor improvements being realized (although acetaldehyde demand would become a relatively minor chemical product as a result of the successful Rh catalyzed methanol carbonylation process.) Acetic anhydride would still be made by either Rh catalyzed methyl acetate carbonylation or via acetic acid based ketene and the processes would remain unchanged after 1985. However, while the processes for acetaldehyde, vinyl acetate, and acetic anhydride remained stable and unchallenged, the methanol carbonylation would undergo some significant catalyst improvements starting in 1990. In 1990, Celanese (then Hoechst-Celanese), , began operating a low water methanol carbonylation process that included L i as a co-catalyst and hydrogen. ' The inclusion of L i stabilized the catalyst both within the reactor and during catalyst-product separation. It also accelerated the conversion of any methyl acetate that might be formed by esterification. This addressed the key flaw in the Monsanto process. The catalyst-production separation, while still being operated adiabatically, was more productive since there was less water being distilled prior to the onset of acetic acid distillation. It was later found that the L i was sufficiently stabilizing that a small amount of heat could be added to distill additional acetic acid, although this needs to be limited since it does not completely eliminate precipitation if the product is heated too long. 1 2,11

In Innovations in Industrial and Engineering Chemistry; Flank, W., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2008.

Downloaded by PENNSYLVANIA STATE UNIV on June 29, 2012 | http://pubs.acs.org Publication Date: December 31, 2008 | doi: 10.1021/bk-2009-1000.ch010

383 In 1996, British Petroleum announced an alternative methanol carbonylation based on an Ir-Ru-Mel catalyst. Like the Celanese Rh-Li catalyst, it was a low water process, but used about half the methyl iodide. The use of Ir as a catalyst was not new and had been disclosed by Monsanto contemporaneously with its disclosure of the Rh catalyst. However, it had very complex kinetics and was more difficult to operate. British Petroleum achieved this event by discovering a way to overcome a major shortcoming in the Ir process disclosed by Monsanto. To understand the British Petroleum breakthrough, it is necessary to look back at the original Monsanto work on Ir. Monsanto investigated both Ir and Rh when developing their methanol carbonylation catalyst and discovered that, whereas the oxidative addition of M e l to Ir(CO) l2~ (equation [27]) was significantly faster than the corresponding oxidative addition of M e l to Rh(CO) I ", the resultant MeIr(CO) I * was stable and, unlike MeRh(CO) I ", it did not undergo spontaneous insertion of C O to form AcIr(CO)I " (equation [28]). Instead, MeIr(CO) I " had to exchange carbon monoxide for an iodide ligand to form MeIr(CO) I (equation [29]). The MeIr(CO) I species was then able to insert C O to form AcIr(CO) I (equation [30]). Addition of iodide to AcIr(CO) I generated AcIr(CO) I " which rapidly reductively eliminated A c l to form the necessary acetyl iodide intermediate and regenerate the Ir(CO) I " catalyst (equation [31].) The dissociation of iodide from MeIr(CO) I " was rate limiting and the reaction kinetics were very complex. The process required careful optimization of the iodine levels and was tricky to operate, especially compared to the robust Rh catalyzed analog. Therefore, the Ir based carbonylation was not chosen for commercialization by Monsanto. 2

2

2

2

3

2

3

2

2

3

3

2

3

2

2

2

2

2

2

3

2

2

2

3

[27] [28] [29] [30] [31] What British Petroleum discovered was that Ru(CO) I (as well as some other metals) catalyzed the exchange of iodide for CO in the transformation of MeIr(CO) I " to MeIr(CO) I (equation [29]) allowing the Ir catalyst to operate without being slowed by the pesky, and previously rate limiting, carbon monoxide for iodide ligand exchange. Complete mechanistic details are available for the process. In the presence of Ru, Ir operated under the same conditions as the Rh based methanol carbonylation, was equally active as a catalyst, and had acceptable ease of operation. Advantages associated with the Ir process are that the Ir process used only half as much methyl iodide co-catalyst and the Ir and Ru catalyst components were (and continue to be) much less expensive than Rh. 4

2

3

3

2

2

4

In Innovations in Industrial and Engineering Chemistry; Flank, W., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2008.

384 These two derivatives of the earlier Monsanto technology are the predominant acetic acid processes today and are equally competitive in the market place. Since the advent of the Monsanto Acetic Acid process almost all new acetic acid plants are based on methanol carbonylation and acetaldehyde oxidation has been nearly phased out as a source of acetic acid. The advances in Rh and Ir based methanol carbonylation have recently been reviewed. 15

Downloaded by PENNSYLVANIA STATE UNIV on June 29, 2012 | http://pubs.acs.org Publication Date: December 31, 2008 | doi: 10.1021/bk-2009-1000.ch010

Tomorrow - Acetyl Processes of the Future Predicting the future of chemical technology is always dangerous since markets and resources change at unpredictable rates. However, several features of the acetyl product stream assure their future in a world with diminishing access to petroleum for chemicals. The fact that acetic acid and acetic anhydride are now derived from synthesis gas which can be generated from more widely available resources such as methane, coal, and biomass, assures access to these products in the long term. Further, one of the two large outlets, cellulose acetate, represents one of the few renewable polymer resources in the chemical industry. (Cellulose acetate consumes about 1/3 of all the acetyl units generated worldwide.) The need for cellulose acetate is also growing as it is finding new applications in electronics (flat screen displays.) Therefore, in the quest for a sustainable chemical industry, the acetyls and cellulose esters should fare well and may even serve a leadership role. Work is already underway in various laboratories to replace the last remaining petroleum portion of the acetyl chain, namely the ethylene component of vinyl acetate, with syngas. These processes take advantages of the old Celanese E D A approach to vinyl acetate by generating E D A . Their primary drawback is that they used large volumes of acetic anhydride and acetic acid recycles. However, Eastman Chemical Company presented an interesting new scheme for the production of vinyl acetate from dimethyl ether which involves little or no recycle. The process includes a carbonylation of dimethyl ether (made by dehydration of methanol) to make acetic anhydride, a reactive distillation of acetic anhydride with acetaldehyde which reduces the Celanese E D A process to a single unit of operation, and hydrogenates acetic acid byproduct (from the vinyl acetate process) to produce the acetaldehyde component. This reduces the size of the plants required by a factor of two. (Inherent in this process is a means of producing acetaldehyde from synthesis gas as well via acetic acid hydrogénation.) As ethylene becomes more expensive, these processes may be competitive. Further, acetic acid and acetic anhydride, may serve as building blocks for additional key intermediates in a synthesis gas based chemical industry. Investigations have already been undertaken to use acetic acid as a building block for acrylic acid (by condensation with formaldehyde - a syngas molecule 4,16

9

155

In Innovations in Industrial and Engineering Chemistry; Flank, W., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2008.

385 obtained by oxidation of methanol) and homologation of carboxylic acids to higher homologs using synthesis gas. 17

References 1.

a) Wagner, F. S., Jr., "Acetic Acid", in Kirk-Othmer Encyclopedia of Chemical Technology, 5 edit., John Wiley and Sons, Inc., Hoboken, N J , Vol. 1, 2004, p. 115 and earlier editions and references cited therein, b) Cheung, H . , Tanke, R. S., Torrence, G. P., "Acetic Acid", in Ullmann's Encyclopedia of Industrial Chemistry, Wiley-VCH Verlag GmbH and Co., K G a A , Weinhiem, Germany, 6 edit., 1, 149 (2003) and earlier editions and references cited therein. c) McMahon, K . S. "Acetic Acid", in Encyclopedia of Chemical Processing and Design, McKetta, J. J., Cunningham, W. Α., eds., Marcel Dekker, Inc., N Y , N Y , 1, p. 216 (1976). Agreda, V . H . , Zoeller, J. R., eds., Acetic Acid and Its Derivatives, Marcel Dekker, N Y , NY (1993). a) Hagemeyer, H . J., "Acetaldehyde", in Kirk-Othmer Encyclopedia of Chemical Technology, 5 edit., John Wiley and Sons, Inc., Hoboken, N J , 1, p. 99 (2004) and earlier editions and references cited therein. b) Fleischman, G., Jiva, R., Bolt, H . M . , Golka, K . , "Acetaldehyde", in Ullmann's Encyclopedia of Industrial Chemistry, Wiley-VCH Verlag GmbH and Co., K G a A , Weinhiem, Germany, 6 edit., 1, 131 (2003) and earlier editions and references cited therein. c) Aquilo, Α., Penrod, J. D., in Encyclopedia of Chemical Processing and Design, McKetta, J. J., Cunningham, W. Α., eds., Marcel Dekker, Inc., N Y , N Y , 1, 114 (1976). d) Fanning, A . T., "Ethylene­ -and Acetylene- Based Processes", in Acetic Acid and Its Derivatives, Agreda, V . H., Zoeller, J. R., eds., Marcel Dekker, N Y , NY , 1993, p, 15. a) Wagner, F. S., Jr., "Acetic Anhydride", in Kirk-Othmer Encyclopedia of Chemical Technology, 5 edit., John Wiley and Sons, Inc., Hoboken, N J , 1, 146 (2004) and earlier editions and references cited therein. b) Held, H . , Renstl, Α., Mayer, D., "Acetic Anhydride and Mixed Fatty Acid Anhydrides", in Ullmann's Encyclopedia of Industrial Chemistry, WileyV C H Verlag GmbH and Co., K G a A , Weinhiem, Germany, 6 edit., 1, 179 (2003) and earlier editions and references cited therein, c) McMahon, K . S. "Acetic Anhydride", in Encyclopedia of Chemical Processing and Design, McKetta, J. J., Cunningham, W. Α., eds., Marcel Dekker, Inc., N Y , N Y , 1, 258 (1976). d) Cook, S. L., "Acetic Anhydride", in Acetic Acid and Its Derivatives, Agreda, V . H . , Zoeller, J. R., eds., Marcel Dekker, N Y , NY , p, 145 (1993). a) Cordiero, C. F., Petrocelli, F. P., "Vinyl Acetate Polymers", in KirkOthmer Encyclopedia of Chemical Technology, 5 edit., John Wiley and Sons, Inc., Hoboken, N J , 25, 557 (2007) and earlier editions and references

Downloaded by PENNSYLVANIA STATE UNIV on June 29, 2012 | http://pubs.acs.org Publication Date: December 31, 2008 | doi: 10.1021/bk-2009-1000.ch010

th

th

2. 3.

th

th

4.

th

th

5.

th

In Innovations in Industrial and Engineering Chemistry; Flank, W., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2008.

386 cited therein. b) Roscher, G., "Vinyl Esters", in Ullmann's Encyclopedia of Industrial Chemistry, Wiley-VCH Verlag GmbH and Co., K G a A , Weinhiem, Germany, 6 edit., 38, 59 (2003) and earlier editions and references cited therein. c) Sumner, C. E., Zoeller, J. R., "Vinyl Acetate", in Acetic Acid and Its Derivatives, Agreda, V . H . , Zoeller, J. R., eds., Marcel Dekker, N Y , NY , p, 145 (1993). Partin, L . R. , Heise, W. H . , "Bioderived Acetic Acid" in Acetic Acid and Its Derivatives, Agreda, V . H., Zoeller, J. R., eds., Marcel Dekker, N Y , N Y , p. 3 (1993). a) Treece, L . C., Johnson, G. I., "Cellulose Acetate" in Acetic Acid and Its Derivatives, Agreda, V . H . , Zoeller, J. R., eds., Marcel Dekker, N Y , N Y , p. 3 (1993). b) Gedon, S., Fengi, R., "Cellulose Ester, Organic Esters", in Kirk-Othmer Encyclopedia of Chemical Technology, 5 edit., John Wiley and Sons, Inc., Hoboken, NJ, 5, 412 (2007) and earlier editions and references cited therein. b) Balser, K., Hoppe, L., Eicher, T., Wandel, M., Astheimer, H.-J., Steinmeir, H . , "Cellulose Esters", in Ullmann's Encyclopedia of Industrial Chemistry, Wiley-VCH Verlag GmbH and Co., K G a A , Weinhiem, Germany, 6 edit., 6, 647 (2003) and earlier editions and references cited therein a) British Pat. No. 283,989 (to Celanese.) b) British Pat. No. 317,867 (to Celanese.) c) British Pat. No. 343,947 (to Celanese.) d) British Pat. No. 320,457 (to Celanese.) Anon., Hydrocarbon Process., 44, 287 (1965). Kumar, D. Chen, M . S., Goodman, D. W., Catalysis Today, 2007, 123, 77. a) Howard, M . J., Jones, M . D., Roberts, M . S., Taylor, S.A., Catalysis, 1993, 18, 325. b) Zoeller, J. R., "Manufacture via Methanol Carbonylation", in Acetic Acid and Its Derivatives, Agreda, V . H . , Zoeller, J. R., eds., Marcel Dekker, N Y , N Y , p. 35 (1993). c) Gauthier-Lafaye, J., Perron, R., Methanol and Carbonylation, (English Translation), Editions Technip, Paris, France (1987), chap. 6, p. 119. a) Zoeller, J. R., Cloyd, J. D., Lafferty, N. L., Nicely, V . Α., Polichnowski, S. W., and Cook, S. L., Advances in Chemistry Series, 1992, 230, 377 b) Agreda, V . H . , Pond, D. M., and Zoeller, J. R., Chemtech, 1992, 22, 172. c) Gauthier-Lafaye, J., Perron, R., Methanol and Carbonylation, (English Translation), Editions Technip, Paris, France (1987), chap. 7, p. 150. Agreda, V . H . , Partin, L . R., Heise, W. H . , Chem. Eng. Prog., 1990, 86(2), 40. Haynes, Α., Maitlis, P. M., Morris, G. E., Sunley, G. J., Adams, H . , Badger, P. W., Bowers, C. M . , Cook, D. B., Elliott, P. I. Ghaffer, T., Green, H . , Griffin, T. R., Payne, M . , Pearson, J. M . , Taylor, M . J., Vickers, P. W. , Watt, R. J., J. Amer. Chem. Soc., 2004, 126, 2847. Haynes, Α., Topics in Organometallic Chemistry, 2006, 18, 179. th

Downloaded by PENNSYLVANIA STATE UNIV on June 29, 2012 | http://pubs.acs.org Publication Date: December 31, 2008 | doi: 10.1021/bk-2009-1000.ch010

6.

7.

th

th

8.

9. 10. 11.

12.

13. 14.

15.

In Innovations in Industrial and Engineering Chemistry; Flank, W., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2008.

387

Downloaded by PENNSYLVANIA STATE UNIV on June 29, 2012 | http://pubs.acs.org Publication Date: December 31, 2008 | doi: 10.1021/bk-2009-1000.ch010

16. a.) Gauthier-Lafaye, J., Perron, R., Methanol and Carbonylation, (English Translation), Editions Technip, Paris, France (1987), chap. 8, p. 201. b) Tustin, G. C., Colberg, R. D., Zoeller, J. R., Catalysis Today, 2000, 58, 281 17. Hawkins, J. Α., Zoeller, J. R., "Chain Growth: Acrylics and Other Carboxylic Acids", in Acetic Acid and Its Derivatives, Agreda, V . H . , Zoeller, J. R., eds., Marcel Dekker, N Y , N Y , p. 345 (1993).

In Innovations in Industrial and Engineering Chemistry; Flank, W., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2008.