Alternative Solvents for Green Chemistry

By Francesca M. Kerton

The Royal Society of Chemistry

Copyright © 2009 Francesca M. Kerton
All rights reserved.
ISBN: 978-0-85404-163-3

Contents

Chapter 1 Introduction,
Chapter 2 ‘Solvent free’ Chemistry,
Chapter 3 Water,
Chapter 4 Supercritical Fluids,
Chapter 5 Renewable Solvents,
Chapter 6 Room Temperature Ionic Liquids and Eutectic Mixtures,
Chapter 7 Fluorous Solvents and Related Systems,
Chapter 8 Liquid Polymers,
Chapter 9 Tunable and Switchable Solvent Systems,
Chapter 10 Industrial Case Studies,
Subject Index, 218,

CHAPTER 1

Introduction

1.1 The Need for Alternative Solvents

One of the 12 principles of green chemistry asks us to ‘use safer solvents and auxiliaries’. Solvent use also impacts some of the other principles and therefore, it is not surprising that over the last 10 years, chemistry research into the use of greener, alternative solvents has grown enormously. If possible, the use of solvents should be avoided, or if they cannot be eliminated, we should try to use innocuous substances instead. In some cases, particularly in the manufacture of bulk chemicals, it is possible to use no added solvent — so-called ‘solvent free’ conditions. Yet in most cases, including specialty and pharmaceutical products, a solvent is required to assist in processing and transporting of materials. Alternative solvents suitable for green chemistry are those that have low toxicity, are easy to recycle, are inert and do not contaminate the product. There is no perfect green solvent that can apply to all situations and therefore decisions have to be made. The choices available to an environmentally concerned chemist are outlined in the following chapters. However, we must first consider the uses, hazards and properties of solvents in general.

Solvents are used in chemical processes to aid in mass and heat transfer, and to facilitate separations and purifications. They are also an important and often the primary component in cleaning agents, adhesives and coatings (paints, varnishes and stains). Solvents are often volatile organic compounds (VOCs) and are therefore a major environmental concern as they are able to form low-level ozone and smog through free radical air oxidation processes. Also, they are often highly flammable and can cause a number of adverse health effects including eye irritation, headaches and allergic skin reactions to name just three. Additionally, some VOCs are also known or suspected carcinogens. For these and many other reasons, legislation and voluntary control measures have been introduced. For example, benzene is an excellent, unreactive solvent but it is genotoxic and a human carcinogen. In Europe, prior to 2000, gasoline (petrol) contained 5% benzene by volume but now the content is <1%. Dichloromethane or methylene chloride (CHCl) is a suspected human carcinogen but is widely used in research laboratories for syntheses and extractions. It was previously used to extract caffeine from coffee, but now decaffeination is performed using supercritical carbon dioxide (scCO2). Perchoroethylene (CCl2CCl2) is also a suspected human carcinogen and is the main solvent used in dry cleaning processes (85% of all solvents). It is also found in printing inks, white-out correction fluid and shoe polish. ScCO2 and liquid carbon dioxide technologies have been developed for dry cleaning; however, such solvents could not be used in printing inks. Less toxic, renewable and biodegradable solvents such as ethyl lactate are therefore being considered by ink manufacturers.

Despite a stagnant period for the solvent industry during 1997–2002, world demand for solvents, including hydrocarbon and chlorinated types, is currently growing at approximately 2.3% per year and approaching 20 million tonnes annually. However, when the less environmentally friendly hydrocarbon and chlorinated types are excluded, market growth is around 4% per year. Therefore, it is clear that demand for hydrocarbon and chlorinated solvents is on a downward trend as a result of environmental regulations, with oxygenated and green solvents replacing them to a large extent. It should be noted that these statistics exclude in-house recycled materials and these figures therefore just represent solvents new to the market; the real amount of solvent in use worldwide is far higher. It also means that annually a vast amount of solvent is released into the environment (atmosphere, water table or soil). Nevertheless the situation is moving in a positive direction, as in the USA and Western Europe environmental concerns have increased sales of water based paints and coatings to levels almost equal to the solvent based market. Therefore, it is clear that legislation and public interests are causing real changes in the world of solvents.

The introduction of legislation by the United States Food and Drug Administration (FDA) means that some solvents, e.g. benzene, are already banned in the pharmaceutical industry and others should only be used if unavoidable, e.g. toluene and hexane. FDA-preferred solvents include water, heptane, ethyl acetate, ethanol and tert-butyl methyl ether. Hexane, which is not preferred and is a hazardous air pollutant, is used in the extraction of a wide range of natural products and vegetable oils in the USA. According to the EPA Toxic Release Inventory, more than 20 million kg of hexane are released into the atmosphere each year through these processes. It may seem straightforward to substitute hexane by its higher homologue, heptane, when looking at physical and safety data for solvents (Table 1.1). However, heptane is more expensive and has a higher boiling point than hexane, so economically and in terms of energy consumption a switch is not that simple. Therefore, it is clear that much needs to be done to encourage the development and implementation of greener solvents.

1.2 Safety Considerations, Life Cycle Assessment and Green Metrics

In recent years, efforts have been made to quantify or qualify the ‘greenness’ of a wide range of solvents; both green and common organic media were considered. In deciding which solvent to use, a wide range of factors should be considered. Some are not directly related to a specific application, such as cost and safety, and these will generally rule out some options. For example, room temperature ionic liquids (RTILs) are much more expensive than water and they are therefore more likely to find applications in high value added areas such as pharmaceuticals or electronics than in the realm of bulk or commodity chemicals. However, a more detailed assessment of additional factors should be performed including a life cycle assessment, energy requirements and waste generation.

A computer-aided method of organic solvent selection for reactions has been developed. In this collaborative study between chemical engineers and process chemists in the pharmaceutical industry, the solvents are selected using a rules based procedure where the estimated reaction-solvent properties and the solvent-environmental properties are used to guide the decision making process for organic reactions occurring in the liquid phase. These rules (Table 1.2), whether computer-aided or not, could also be more widely used by all chemists in deciding whether to use a solvent and which solvents to try first.

The technique was used in four case studies; including the replacement of dichloromethane as a solvent in oxidation reactions of alcohols, which is an important area of green chemistry. 2-Pentanone, other ketones and some esters were suggested as suitable replacement solvents. At this point, the programme was not able to assess the effects of non-organic solvents because of a lack of available data. However, this approach holds promise for reactions where a VOC could be replaced with a far less hazardous, less toxic or bio-sourced option.

1.2.1 Environmental, Health and Safety (EHS) Properties

The EHS properties of a solvent include its ozone depletion potential, biode-gradability, toxicity and flammability. Fischer and co-workers have developed a chemical (and therefore, solvent) assessment method based on EHS criteria.It is available at http://www.sust-chem.ethz.ch/tools/ehs/. They have demonstrated its use on 26 organic solvents in common use within the chemical industry. The substances were assessed based on their performance in nine categories (Table 1.3).

Using this EHS method, formaldehyde, dioxane, formic acid, acetonitrile and acetic acid have high (environmentally poor) scores (Figure 1.1). Formaldehyde has acute and chronic toxicity, dioxane is persistent and the acids are irritants. Methyl acetate, ethanol and methanol have low scores, indicating a lower hazard rating.

1.2.2 Life Cycle Assessment (LCA)

The function of life cycle assessment (LCA) is to evaluate environmental burdens of a product, process, or activity; quantify resource use and emissions; assess the environmental and human health impact; and evaluate and implement opportunities for improvements. It is important to realize that while this book focuses on solvents, VOC ‘free’ paints and other ‘green’ consumer items may not be entirely green or entirely VOC free when the whole life cycle is considered. For example, a VOC may be used in the preparation of a pigment or another paint component, which is then incorporated into the final non-VOC (e.g. aqueous) formulation. The same can also be said for many synthetic procedures which are reported to be ‘solvent free’. The reaction may be performed between neat reagents; however, a solvent is used in purifying, isolating and analysing the product. Chemists should be aware of this and avoid over-interpreting what authors are describing.

Fischer and co-workers undertook a LCA of the 26 organic solvents which they had already assessed in terms of EHS criteria (see above). They used the Eco-solvent software tool (http://www.sust-chem.ethz.ch/tools/ecosolvent/), which on the basis of industrial data considers the ‘birth’ of the solvent (its petrochemical production) and its ‘death’ by either a distillation process or treatment in a hazardous waste incineration plant. For both types of end-of-life treatment, ‘environmental credits’ were granted where appropriate, e.g. solvent recovery and reuse upon distillation. The results of this assessment are shown in Figure 1.2. From an LCA perspective, tetrahydrofuran (THF), butyl acetate, cyclohexanone and 1-propanol are not good solvents. This is primarily due to the environmental impact of their petrochemical production and their LCA would therefore be better if they came from a different source. For example, 1-propanol may one day become available through selective dehydration and hydrogenation of glycerol (a renewable feedstock). At the other end of this scale, diethyl ether, hexane and heptane are considered favourable solvents. However, the reader should already be aware that diethyl ether is extremely hazardous in terms of flammability, low flash point and explosion risk through peroxide contamination. Therefore, the results from the EHS assessment and LCA were combined in an attempt to provide the whole picture (Figure 1.3).

It is evident from Figure 1.3 that formaldehyde, dioxane, organic acids, acetonitrile and THF are not desirable solvents. THF and formaldehyde are significant outliers on this last graph because of their particularly poor performance under one of the assessment methods. Methanol, ethanol and methyl acetate are preferred solvents based on their EHS assessment. Heptane, hexane and diethyl ether are preferred based on LCA. However, it must be noted that the LCA was performed based on petrochemical production of the solvents and if the first group of solvents was bio-sourced, perhaps these three solvents would be the outright winners! Unfortunately, assessment tools used in this study could not be applied to many currently favoured alternative solvent technologies such as supercritical fluids and RTILs as there is a lack of available data at this time to quantify them fully.

However, a more qualitative LCA approach has been used by Clark and Tavener to assess the neoteric solvents described in this book (Figure 1.4).The solvent must first be manufactured, usually from petroleum. This is relatively straightforward for simple and aromatic hydrocarbons that are obtained through cracking and distillation of crude oil. However, for other chemicals more complex synthetic routes are needed, e.g. to introduce heteroatoms such as halogens. Yet others, such as acetone, are produced as by-products in the manufacture of some chemicals. In terms of the alternative solvents described in this book, fluorous solvents and RTILs typically require multistage syntheses. Carbon dioxide and water do not need preparation but do need purification prior to use. Other renewable solvents, such as ethanol and esters, require separation or extraction and purification before use. A step often overlooked in LCA of chemicals is their distribution. Carbon dioxide and water are available globally and can therefore be sourced close to their point of use. Bioethanol would be a good solvent to use in Brazil but may not be readily available in other areas of the world. Therefore, the authors suggested a labelling system, similar to the ‘food miles’ being introduced at supermarkets, enabling chemists to find out where their compounds or solvents were manufactured.

The third primary stage in the life cycle of a solvent is its use. Solvents are used in many areas and not just as media for reactions (Table 1.4). The choice of the right solvent can have significant effects on energy consumption and the E-factor of a process. Solvent effects can lead to different reaction pathways for a number of reasons; some of these effects will be briefly discussed later in this chapter. The E-factor is the mass ratio of waste to desired product. If the wrong solvent is chosen, it can significantly affect the yield of a process (99% in the ‘right’ solvent compared to 30% in the ‘wrong’ one). For this reason, it is not surprising to find tables in journal articles showing the conversions or yields for a range of solvents. Clearly, in process development laboratories worldwide a significant amount of time and effort is spent optimizing the reaction conditions and the solvent choice to optimize this part of the LCA. Often the physical properties of the solvent play a significant role here; the boiling point and melting point, viscosity, volatility and density must all be considered alongside the safety issues such as flash point, reactivity and corrosiveness that were discussed earlier. At this stage in the process and the life cycle, biphasic systems and processes can be considered as these usually lead to reduced energy and increased efficiency. Fluorous solvents can be advantageous for this reason. However, all alternative solvents have advantages and disadvantages. Unfortunately, in the chemical literature, most authors are biased and are trying to ‘sell’ their chosen reaction medium. For example, the pressures involved with supercritical fluids are a disadvantage, but the facile removal of the fluid at the end of a process is an advantage. Therefore, Clark and Tavener used a scoring system to grade the solvents (Table 1.5) in an attempt to qualify the general level of ‘greenness’ of a range of alternative solvents. It becomes apparent that all the solvents have some drawbacks and therefore solvent free approaches should attract greater attention. If a solvent is used, water should be considered first, and then carbon dioxide. They also suggest that it is unrealistic to think that all VOCs can be replaced in every application and therefore there is a growing role for VOCs derived from renewable resources in the alternative solvent field. In all areas, we need to balance the technical advantages of a particular solvent with any environmental, cost or other disadvantages. For example, in the coatings industry, a reduction in the amount of VOC in paints may lead to a range of problems, including the stability of the formulation, longer drying times, a lower gloss and a less hard-wearing finish. However, aqueous emulsion paint has notable EHS advantages, including reduced VOC emissions, reduced user exposure and less hazardous waste production. Manufacturers and consumers need to decide if the advantages outweigh the disadvantages.

At the end of their life, solvents can often be reused or recycled by a range of recovery methods including distillation or biphasic separation. An environmental assessment of waste solvent distillation was recently reported and took into account a range of inputs and outputs including electricity consumption, cooling water, amount of recovered distillate and waste. On average per kg of waste solvent processed, 0.71 kg of solvent is recovered, 1.4kg steam, 0.03 kW h electricity, 1.5 x 10-3m3 nitrogen gas and 2.7 x 10-2m3 cooling water used. Steam is used for heating the waste solvent and nitrogen is used to avoid the formation of explosive vapour. Despite extensive recycling of solvents within the chemical industry, ultimately the solvent will likely be incinerated at the end of its life (Figure 1.4). Incineration can generate valuable energy but the exhaust gases from the incineration plant also need treating.

Unfortunately, accidents happen; solvents can leak or spill and may not make it through to the normal end of their life cycle. Therefore, this possibility of release into the environment must also be taken into consideration when performing LCA. In these end-of-life scenarios, carbon dioxide has little environmental impact but other green solvents do. Water can become contaminated and must be treated prior to release. Fluorous solvents are difficult to incinerate and may form dangerous acidic by-products, and they are also persistent in the environment. However, perfluoroalkyl ether compounds, which have many similar properties to perfluorocarbons, are more short-lived in the environment and are therefore better solvents in terms of LCA for fluorous biphasic approaches. Unsurprisingly, new RTILs are being developed that take into account this part of a LCA and they are being designed with biodegradation in mind.

(Continues…)Excerpted from Alternative Solvents for Green Chemistry by Francesca M. Kerton. Copyright © 2009 Francesca M. Kerton. Excerpted by permission of The Royal Society of Chemistry.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.