THE adiponitrile PROCESS

The transition metal catalysed addition of HCN to alkenes is potentially a very useful reaction in organic synthesis and it certainly would have been more widely applied in the laboratory if its attraction were not largely offset by the toxicity of HCN. Industrially the difficulties can be minimised to an acceptable level and we are not aware of major accidents. DuPont has commercialised the addition of HCN to butadiene for the production of adiponitrile [ADN, NC(CH2)4CN], a precursor to 1,6-hexanediamine, one of the components of 6,6-nylon and polyurethanes (after reaction with diisocyanates). The details of the hydrocyanation process have not been released, but a substantial amount of related basic chemistry has been published. The development of the ligand parameters % and 0 by Tolman formed part of the basic studies carried out in the Du Pont labs related to the ADN process [1], 

Addition of HCN to acetone to form the cyanohydrin is still the main route to methyl methacrylate. Hydrocyanins can be converted to amino acids as well. The nitrile group can be easily converted to amines, carboxylic acids, amides, etc. Addition to aldehydes and activated alkenes can be done with simple base, but addition to unactivated alkenes requires a transition metal catalyst. The methods of HCN addition have been discussed by Brown [2], 

When a large excess of ligand is used, as in the actual process, the system becomes more complicated and species such as (C2H4)L3Ni, HNi(CN)L3, and ML4 are also observed. The activation parameters of the reaction are AG (-40 °C)= 17 kcal/mol, AH = 9 kcaLmoF1, AS = -34 cal.mol. K 1. The negative entropy of activation is consistent with the formation of a five-coordinate species. The reason for the associative character of the reductive elimination is two-fold  

the addition of one more phosphite ligand reduces the electron density at nickel disfavouring the divalent state, and 

the resulting species after elimination will contain 14 electrons instead of only 12 which leads to a higher stability. 

---

Tolman, Druliner, and McKinney [5] were pioneers in nickel-catalyzed hydrocyanation they used monodentate phosphites mainly to understand and improve the adiponitrile process. Although bidentate ligands give better results in the adiponi-trile process [21], mechanistic studies with these systems are rare bidentate phos-phinites have been studied in the asymmetric hydrocyanation of MVN [19]. 

In asymmetric hydrocyanation reactions the desired isomers are the chiral branched products only. Good regioselectivity toward the branched product (>98%) is limited to vinylarenes. Hydrocyanation of 1,3-dienes gives a variety of mixtures depending on the catalyst and conditions 1-alkenes give the linear nitrile as major product [34]. Both are seen in the adiponitrile process in which the unwanted branched 2M3BN (hydrocyanation product from 1,3-butadiene) is isomerized to the linear product 3-pentenenitrile, which is then hydrocyanated by in-situ isomerization to 4-pentenenitrile, resulting in the linear adiponitrile. Thus vinylarenes and cyclic alkenes (mainly norbomene) are usually the substrates of choice for the asymmetric hydrocyanation. Hopefully 1,3-dienes will become feasible substrates in the near future. 

A new route to prepare nicotinic acid starts from 2-methylglutaronitrile, a major side-product in the adiponitrile process and, as such, a readily available starting-material. It is easily hydrogenated to 2-methylpentanediamine, which is then condensed to methyl piperidine and dehydrogenated to 3-picoline. The gas-phase ammoxidation of the latter to cyanopyridine is followed by hydrolysis to either nicotinamide or nicotinic acid (Scheme 20.4). The cyanopyridine route for the production of nicotinic acid has the advantage of a significantly better selectivity with respect to the direct oxidation route from 3-picoline owing to the easy decar-... 

FIGURE 26.19 Schematic diagram of the adiponitrile process flowsheet. AN = acrylonitrile, ADN = adiponi-hile, QS = quaternary ammonium salt [30] (with kind permission from Springer Science and Business Media). 

Ever since Monsanto commercialized their adiponitrile process in 1965 by electrolytic reductive coupling of acrylonitrile, a number of attempts at the commercialization of electrochemical reductions have been made. None of these attempts has succeeded in producing the tonnages involved in the adiponitrile process of approximately 100 000 tonne p.a. on plants both in the U.K and the U.S.A. However a number of significant tonnage production operations have been built and are outlined below. 

Ion exchange membranes are also being utilized in electrodialysis and electrochemistry. Salt production and desalination are the main applications of electrodialysis. The chlor-aikali process and the adiponitrile process are examples of highly successful ion exchange membrane applications. 

Addition of HCN to unsaturated compounds is often the easiest and most economical method of making organonitnles. An early synthesis of acrylonitrile involved the addition of HCN to acetylene. The addition of HCN to aldehydes and ketones is readily accompHshed with simple base catalysis, as is the addition of HCN to activated olefins (Michael addition). However, the addition of HCN to unactivated olefins and the regioselective addition to dienes is best accompHshed with a transition-metal catalyst, as illustrated by DuPont s adiponitrile process (6—9). 

Adiponitrile is made commercially by several different processes utilizing different feedstocks. The original process, utilizing adipic acid (qv) as a feedstock, was first commercialized by DuPont in the late 1930s and was the basis for a number of adiponitrile plants. However, the adipic acid process was abandoned by DuPont in favor of two processes based on butadiene (qv). During the 1960s, Monsanto and Asahi developed routes to adiponitrile by the electrodimerization of acrylonitrile (qv). 

The Monsanto adiponitrile process, first commercialized in 1965 (65—67), involves the dimerization of acrylonitrile at the cathode in an electrolytic cell (eq. 7) ... 

Pentenenitnles are produced as intermediates and by-products in DuPont s adiponitrile process. 3-Pentenenitrile [4635-87-4] is the principal product isolated from the isomerisation of 2-methyl-3-butenenitrile (see eq. 4). It is entirely used to make adiponitrile. i7j -2-Pentenenitrile [25899-50-7] is a by-product isolated from the second hydrocyanation step. Some physical properties are Hsted in Table 13. 

A.sahi Chemical EHD Processes. In the late 1960s, Asahi Chemical Industries in Japan developed an alternative electrolyte system for the electroreductive coupling of acrylonitrile. The catholyte in the Asahi divided cell process consisted of an emulsion of acrylonitrile and electrolysis products in a 10% aqueous solution of tetraethyl ammonium sulfate. The concentration of acrylonitrile in the aqueous phase for the original Monsanto process was 15—20 wt %, but the Asahi process uses only about 2 wt %. Asahi claims simpler separation and purification of the adiponitrile from the catholyte. A cation-exchange membrane is employed with dilute sulfuric acid in the anode compartment. The cathode is lead containing 6% antimony, and the anode is the same alloy but also contains 0.7% silver (45). The current efficiency is of 88—89%, with an adiponitrile selectivity of 91%. This process, started by Asahi in 1971, at Nobeoka City, Japan, is also operated by the RhcJ)ne Poulenc subsidiary, Rhodia, in Bra2il under Hcense from Asahi. 

In a typical process adiponitrile is formed by the interaction of adipic acid and gaseous ammonia in the presence of a boron phosphate catalyst at 305-350°C. The adiponitrile is purified and then subjected to continuous hydrogenation at 130°C and 4000 Ibf/in (28 MPa) pressure in the presence of excess ammonia and a cobalt catalyst. By-products such as hexamethyleneimine are formed but the quantity produced is minimized by the use of excess ammonia. Pure hexamethylenediamine (boiling point 90-92°C at 14mmHg pressure, melting point 39°C) is obtained by distillation, Hexamethylenediamine is also prepared commercially from butadience. The butadiene feedstock is of relatively low cost but it does use substantial quantities of hydrogen cyanide. The process developed by Du Pont may be given schematically as ... 

Hydrocyanation represents a reaction of considerable economic importance largely due to the value of the DuPont process involving HCN addition to butadiene to afford adiponitrile.61,62 The mechanism is well known, and involves (i) oxidative addition of H-CN to a coordinatively unsaturated metal complex, (ii) coordination of an alkene to the H-M-CN species, (iii) migratory... 

The original process used aqueous tetraethylammonium ethylsulfate as the electrolyte, a lead cathode, and a lead-silver alloy anode. The Mark II process, commercialized in the mid-1970s, uses an emulsion of acrylonitrile in aqueous sodium phosphate containing a salt of the hexamethylene-bis-(ethyldibutylammonium) cation. The process was invented in 1959 by M. M. Baizer at Monsanto Corporation, St. Louis, MO. It was commercialized in 1965 and has been continuously improved ever since. The process is also operated in Japan by Asahi Chemical Industry Company. In 1990, the world production of adiponitrile by this process was over 200,000 tonnes per year. 

UCB-MCI [Union Chimique—Chemische Bedrijven and Ministry of Chemical Industry for the USSR] An EHD process for making adiponitrile, differing from the Monsanto process in using an emulsion of acrylonitrile and in not using a membrane. 

Danly DE (1979) Discovery, development and commercialization of the electrochemical adiponitrile process, Amstrong Lecture, Part 2 Chem and Ind (London) (13)439 Client Abstr 92 (1980) 42379k... 

Isomerisation is also an important step in the DuPont process for making adiponitrile (Chapter 11) in which internal pentenenitriles must be converted to the terminal alkene. The catalyst is the same as that used for the hydrocyanation reaction, namely nickel(II) hydrides containing phosphite ligands. 

Development of the industrial process for electrochemical conversion of acrylonitrile to adiponitrile led to extensive investigation into the mechanism of the dimerization process. Reactions of acrylonitrile radical-anion are too fast for investigation but the dimerization step, for a number of more amenable substrates, has been investigated in aprotic solvents by electrochemical techniques. Pulse-radiolysis methods have also been used to study reactions in aqueous media. 

Byproducts of large industrial-scale processes are valorized for instance, in the DuPont process for adiponitrile, the byproduct a-methylglutaronitrile is upgraded to p-picoline and further to niacinamide. 

The same process sometimes can be performed efficiently in cells either with or without diaphragms. Figures 19.16(e) and (f) are for making adiponitrile by reduction of acetonitrile. In the newer design, Figure 19.16(f), the flow rate of the electrolyte is high... 

Figure 19.16. Basic designs of electrolytic cells, (a) Basic type of two-compartment cell used when mixing of anolyte and catholyte is to be minimized the partition may be a porous diaphragm or an ion exchange membrane that allows only selected ions to pass, (b) Mercury cell for brine electrolysis. The released Na dissolves in the Hg and is withdrawn to another zone where it forms salt-free NaOH with water, (c) Monopolar electrical connections each cell is connected separately to the power supply so they are in parallel at low voltage, (d) Bipolar electrical connections 50 or more cells may be series and may require supply at several hundred volts, (e) Bipolar-connected cells for the Monsanto adiponitrile process. Spacings between electrodes and membrane are 0.8-3.2 mm. (f) New type of cell for the Monsanto adiponitrile process, without partitions the stack consists of 50-200 steel plates with 0.0-0.2 ram coating of Cd. Electrolyte velocity of l-2 m/sec sweeps out generated Oz.

In this article, we will discuss the chemistry behind the du Pont adiponitrile process from a mechanistic viewpoint (10). It is not intended to be a comprehensive review of the hydrocyanation literature. We will restrict ourselves rather to homogeneous nickel-catalyzed hydrocyanation of olefins and will depend primarily on du Pont studies. Reviews which explore hydrocyanation in a more general way include those of Brown (77), Hubert and Puentes (72), and James (73). A general review of low-valent organo-nickel chemistry has been published by Jolly and Wilke (14). 

Pentenenitriles are produced as intermediates and by-products in DuPont s adiponitrile process. 3-Pentenenilrile is the principal product isolated from the isomerization of 2-mcthyl-3-butcnenitrilc. 

---