Topics in Catalysis

, Volume 55, Issue 7, pp 625–630

Synthesis of 3,3-Dimethylbutanol and 3,3-Dimethylbutanal, Important Intermediates in the Synthesis of Neotame


  • Setrak K. Tanielyan
    • Center for Applied Catalysis, Department of Chemistry and BiochemistrySeton Hall University
    • Center for Applied Catalysis, Department of Chemistry and BiochemistrySeton Hall University
Original Paper

DOI: 10.1007/s11244-012-9841-z

Cite this article as:
Tanielyan, S.K. & Augustine, R.L. Top Catal (2012) 55: 625. doi:10.1007/s11244-012-9841-z


It has been shown that some N-alkyl derivatives of Aspartame (1) are enhanced sweetening agents [1]. In particular, the 3,3-dimethylbutyl derivative, 2, (Neotame) is about 70 times sweeter than Aspartame. With this in mind a research program directed toward the synthesis of 3,3-dimethylbutanal (3) was begun. The first phase of this research involved the synthesis of 3,3-dimethylbutanol (4) by the acid catalyzed alkylation of isobutene with ethylene to give the sulfate ester [2] which was readily hydrolyzed to the alcohol, 4, in isolated yields in the 70–75 % range. It was found that the most efficient method for the conversion of 4 to the aldehyde, 3, was by a vapor phase dehydrogenation over a copper catalyst. The effect which the reaction variables have on the production of 4 will be discussed. This will include factors such as the ethylene pressure, the acid/isobutene ratio, the use of a hydrocarbon solvent, the reaction temperature and the mode of addition of the isobutene. The discussion of the dehydrogenation procedure will include the nature of the catalyst used and the reaction parameters needed to maximize the formation of 3 and keeping the amount of the over-oxidation carboxylic acid product below 1 %.


Isobutene condensationEthylene–alkene condensationAlkylsulfate hydrolysis3,3-Dimethylbutanol formationDimethylbutanol oxidationDimethylbutanol dehydrogenation

1 Introduction

Aspartame (1) is a dipeptide artificial sweetener which is about 170 times sweeter than a 2 % sucrose solution. It was found that structural modifications to 1, especially the introduction of certain N-alkyl groups, can enhance the sweetening power even more. The best of these modified species, called Neotame (2), is the N-substituted 3,3-dimethylbutyl aspartame which is about 11,000 times sweeter than sugar and about 70 times sweeter than aspartame. The presence of this N-alkyl group results in an increased stability of 2 with respect to 1. As a result, 2 can be used at somewhat higher temperatures than 1 and is also compatible with food ingredients containing carbonyl groups which can react with the nitrogen atom in 1. Neotame also has a long shelf life under dry conditions. Based on the increased sweetening capability its cost is competitive with that of sucrose [1] (Scheme 1).
Scheme 1

Neotame preparation

Since Neotame is prepared by reductive alkylation of 1 with 3,3-dimethylbutanal (3), synthetic investigations were directed toward the synthesis of this compound with the initial work centered on the preparation of the precursor, 3,3-dimethylbutanol (4). Scheme 2 shows some of the reactions screened for applicability to large scale production of 4. The results of this study led to the adoption of the acid catalyzed condensation of ethylene with isobutene which was described previously as a method for the production of the sulfate ester [2] (Eq. 1) which could be hydrolyzed to the alcohol (Eq. 2).
Scheme 2

Methods for the synthesis of 3,3-dimethylbutane

2 Results and Discussion

Table 1 lists the reaction variables and the ranges studied to find a set of reaction conditions with which a significant yield of the alkyl sulfate could be obtained.
Table 1

Reaction parameters, the ranges investigated and the “optimal” conditions used for the preparation of the alkyl sulfate




Stirring rate

Slow, medium, high


Ethylene pressure

60–120 psig

110 psig

Isobutene quantity

100–250 mmol

214 mmol/20 mL heptane

Sulfuric acid quantity

175–535 mmol

428 mmol/30 mL heptane

Sulfuric acid/isobutene ratio




−5 to 25 °C

−15 °C


n-Alkanes, pentane, hexane, heptane, decane


Solvent quantity

25–200 mL

50 mL

A drawing of the reactor used is shown in Fig. 1. The reactor (A) was a 100 mL jacketed threaded ACE glass flask attached to a circulating chiller capable of keeping the temperature of the reaction to, at least, −25 °C. The best procedure for reaction was found to place a mixture of 428 mmol (41.9 g, 22.8 mL) of sulfuric acid and 30 mL of heptane into reaction flask (A) which was cooled to −15 °C and then cover it with a gaseous phase of ethylene at 110 psig. A solution of 214 mmol (20 mL) of liquid isobutene obtained using the dip tube in the pressurized isobutene cylinder,(acid/isobutene ratio of 2:1) in 20 mL of heptane was placed in a second, 75 mL, ACE threaded flask (B). Vigorous stirring of the reactor was initiated to thoroughly mix the acid, hydrocarbon and ethylene phases. The isobutene/heptane solution was then added over a three hour period at a rate of 0.2 mL/min using an Eldex piston pump (C). After the addition was complete, the reaction mixture was stirred for an additional 30 min, the ethylene pressure was released slowly and the temperature increased to −5 °C.
Fig. 1

Schematic of reactor system used for alkene condensation

The most efficient method for hydrolyzing the sulfate ester formed in this reaction was to add 60 g of crushed ice to the reaction mixture immediately after the condensation was finished. The ice was added at such a rate that the temperature did not rise above 0 °C. The heptane layer was separated and the aqueous acid layer was subjected to reactive distillation using a Dean–Stark receiver on a condenser. The product alcohol, 4, was collected from the receiver and distilled. The isolated yield of 4 ranged between 65 and 70 % based on isobutene and 60–65 % based on ethylene, an improvement over the reported [2] yields of 50–55 % based on isobutene and 10–15 % based on ethylene.
With the alcohol, 4, in hand, the next step was the preparation of the aldehyde, 3, (Eq. 3). Of the various possible by-products formed during this conversion (Fig. 2) the only one which can interfere with the reductive alkylation reaction is the carboxylic acid, 5, which can protonate the N atom of 1 and, thus, inhibit the alkylation step. The formation of the others will lower the yield of the aldehyde so their formation must also be kept to a minimum. Since Neotame will be used in foodstuffs, stoichiometric oxidations using chromium or manganese oxides are precluded. Further, a catalytic process is preferred to keep the formation of undesired waste material as low as possible.
Fig. 2

Potential by-products formed in the oxidation of the alcohol, 4

The first reaction examined was a sodium hypochlorite oxidation catalyzed by TEMPO which was based on the procedure described by Anelli [3, 4]. The Anelli procedure, however, involved the use of large quantities of methylene chloride as a solvent and the use of KBr as a co-oxidant, two materials which are not environmentally acceptable. This procedure was modified with the substitution of sodium tetraborate (Borax) as the co-oxidant, elimination of the solvent altogether by running the reaction with the neat alcohol and minimizing the amount of bleach and buffer solutions used [5, 6]. The composition of the mixture at the end of the reaction was primarily a concentrated NaCl solution from which the product aldehyde, 3, was easily separated in 83–85 % yield. The aldehyde obtained by a phase separation on a commercial scale was sufficiently pure to be used in the production of Neotame by the N-alkylation of Aspartame.

However, the disposal of this concentrated brine may present a problem in certain locations. To address this factor a vapor phase catalytic dehydrogenation of the alcohol, 4, using a commercial supported copper catalyst was investigated. This reaction has been reported [7] but under the conditions used the catalyst deactivated rather quickly with a loss of conversion noted after only a few days. The maximum run time reported was only about 6 days. Given the fact that the catalytic dehydrogenation would be preferred over the solution reactions since there is virtually no waste material formed, a more detailed examination of this reaction was initiated with the results depicted in Fig. 3. The reaction was run with a fixed bed reactor (0.5″ od, 0.4″ id) connected to an on-line GC for automatic reaction data acquisition. The catalyst, Cu 0330 from Engelhard Industries, was the same as that used in the previous report [7]. The reactor was loaded with 17.84 g of the catalyst which was first reduced in a stream of hydrogen at 300 °C for 1 h. Dehydrogenation of 4 over this catalyst was run at 300 °C with an Ar flow of 100 cc/min and a 0.05 cc/min flow of the alcohol. Section “X” in Fig. 3 depicts the product details observed during this reaction. As observed previously [7], the conversion decreased during the 24 h run. The productivity (g product/g catalyst/h) was also low. By-product formation was generally low but the amount of acid produced was higher than desired.
Fig. 3

Reaction data related to changes in reaction temperature, gas flow rate and liquid flow rate. a Reaction parameter changes, b selectivity and conversion, c by-product formation, and d productivity

At this point the catalyst was calcined at 400 °C for 1 h at 50 cc/min air flow and then reduced in a 50 cc/min stream of hydrogen for 1 h more. The resulting catalyst was used to investigate the effect of reaction temperature on the dehydrogenation with the results summarized in the TMP Region of Fig. 3. At 300 °C the conversion and productivity were even lower than those observed for the uncalcined catalyst. The by-product formation, though, was almost non-existent. Increasing the temperature resulted in a significant increase in the conversion and productivity of the reaction. It also promoted an increase in the amount of the by-products formed but the alkene increase was most significant. The amount of acid formed remained low.

The effect of changes in the argon flow rate (GFR region in Fig. 3) was investigated keeping the temperature at 330 °C and the alcohol flow at 0.05 cc/min. Higher gas flows resulted in lowering the conversion while with the lower gas flows alkene formation increased. Changes in the liquid flow rate (LFR region in Fig. 3) were studied at 340 °C and an argon flow rate of 90 cc/min. Increasing the LFR resulted in a slight decrease in conversion, a significant decrease in by-product formation and an increase in productivity. Interestingly, the reaction selectivity remained constantly high regardless of the reaction conditions used.

The next step was to use a sample of the calcined catalyst for a long time run to determine its stability. The reactor was filled with 17.84 g of the Cu catalyst which was then calcined and reduced as described above. The catalyst was heated to 340 °C and the alcohol introduced into the reactor at a flow rate of 0.04 cc/min in an Ar flow of 70 cc/min. As the data in Fig. 4 show, the reaction proceeded continuously with no indication of deactivation for 31 days. The conversion of 4 was near 90 % throughout with about 94 % selectivity to 3 and virtually no over-oxidation to the acid, 5, during the reaction. The other by-products were also present in minimal amounts. The aldehyde formed in this reaction has been used directly in the N-alkylation of Aspartame to produce Neotame.
Fig. 4

Reaction data obtained from a thirty-one day catalytic dehydrogenation of 4. a Reaction temperature, gas flow rate and liquid flow rate, b selectivity and conversion, c by-product formation, and d productivity

3 Conclusions

The aldehyde, 3,3-dimethylbutanal (3) required for the synthesis of Neotame was produced by the oxidation or dehydrogenation of 3,3-dimethylbutanol (4) which was formed by the acid catalyzed condensation of ethylene and isobutene and hydrolysis of the intermediate sulfate ester.

Copyright information

© Springer Science+Business Media, LLC 2012