Flow Chemistry & Feedstocks & Automated Manufacturing of any Compound
Next: Summary of Unfactory
Flow Chemistry
There is a trend towards miniaturization of chemical processes called flow chemistry.
In it chemical reactions are done in continuous mode rather than in batch production. Pumps move fluids through tubes and when tubes join reaction takes place. Flow chemistry is a well-established technique at large scale, but the pumps and tubes can be made smaller to operate at different scales.
One of the drivers for this has come from medical industries where the opportunity is to switch from development of common high-volume drugs to therapies treating rarer diseases with smaller patient populations. This has meant the need for small-volume manufacturing methods that still meet the good manufacturing practices imposed by regulators.
Traditionally pharmaceuticals have been made in batch mode in multi-purpose stirred tank vessels because even for blockbuster chemicals the needed production volumes are much smaller than for other chemical industries. However, batch type reactors have well-recognised limitations. For example, reactions that release a lot of energy are very difficult to implement. Also, some reactions proceed via unstable, explosive or highly toxic intermediaries.
To address this, development has progressed towards continuous small-scale flow chemistry technology during the past decades.
The base units in flow chemistry are many discrete flow components like pumps, mixers, small reactors, separation units etc. made on a on small to medium scales. Together with technologies to feed material in (feed delivery systems), flow metering and continuous separation this allows assembling specialised flow systems.
Modularity means the same components can be assembled in multiple ways and re-used again and again to synthetise different end-products. The assembled system can then be run in continuous production mode until the needed production volume is achieved.
Heating has also been moving from conductive heating to use micro-wave. Microwave works in fraction of the time compared to traditional conductive heating and reduces reaction times radically.
Through indirect measurements, it is possible to monitor how well the process is happening and adjust it.
Using small pumps, vessels, tubes, mixers that can freely be linked together, it’s possible to quickly dismantle any chemical processing flow and rebuild the setup to form a complete different processing pipeline. This in just half a day. This is lab room size chemical manufacturing.
Flow synthesis has several benefits solving the problems of traditional pharmaceuticals production. Product is formed continuously, with no downtime of production for charging and discharging the reactors, large amounts of material can be processed even in comparatively small reactors given enough time. A smaller volume means that there is less material in the reactor at any given time and even in accidents happen, the amount of material or energy released in greatly reduced. In very small scale like production of pharmaceuticals the material content in micro-reactors is often too small to cause serious damage to human health or the environment in the event of an accident.
When synthesising a chemical end-product, the process engineer (designer of solution) can choose from a wide range of different potential processes that all produce the same outcome. Processes used today are based on what produces the best return on investment in large scale. This has led to one or few chemistries dominating in different areas.
When moving towards small scale production, the dominant chemistry is likely to change as the benefits of scale are no longer valid and other factors like safety become increasingly important. So, some old process variants may become relevant again when flow chemistry becomes more common.
So as a summary:
Flow chemistry is a small-scale production model where chemical processes are made in lab-size units. Feedstocks are stored in small vessels and small pumps are used to move them through narrow glass tubes. At junctions, feedstocks are mixed together and reactions start to occur. They happen at junction and in the tube leading away where both reactants are present. If the reaction is endothermic (needs energy), its relatively easy to heat with microwaves in principle like food is heated at home. If it is exothermic (releasing energy), cooling is relatively easy to organize as only a small volume of materials react at the same time. Problems are much, much more difficult at high volumes.
The diagram below illustrates:
Feedstocks for Chemical Industries
Before going through various products that can be constructed from the generated synthetic gas or fuel, it’s good to start from an overview of what chemicals are used as raw materials. It turns out there are just a handful.
An overwhelming majority of products in chemical industries starts with one of the five major types of feedstocks:
Light olefins – ethylene (C2H4) and propylene (C3H6)
Aromatics – benzene (C6H6, in a ring formation), toluene (benzene derivate where one hydrogen is replaced with CH3 group -called methyl group), xylene (benzene derivate with two methyl groups) and a mixture of the above three chemicals called BTX that is a product of refining naphtha in an oil refinery
C4 hydrocarbons – butanes (C4H10), butenes (C4H8 also called butylene) and butadienes (C2H4)
Kerosene derived (C9 - C17) paraffins
Syngas – mixture of carbon monoxide (CO) and hydrogen (H)
Ethylene is used to make polyethylene, worlds most common plastic. Polyethylene is used to make films in packaging, carrier bags and trash liners. Ethylene is also used as a precursor feedstock (precursor) for detergents, plasticizers, synthetic lubricants, and additives.
Propene is the second most important starting product in the petrochemical industry after ethylene. Polypropylene is its main use. This plastic is used to make films, fibres, containers, packaging, and caps and closures. It’s an important material in making further chemicals like propylene oxide, acrylonitrile, cumene, butyraldehyde, and acrylic acid.
Benzene is an important feedstock where one or more of its hydrogen atoms gets replaces by some other functional group.
Simple benzene derivatives are phenol, toluene, and aniline. Phenol in turn is used to make polymers (like Bakelite), explosives, paints, pharmaceuticals (like aspirin and also many herbicides), epoxide resins (for coatings, adhesives and composite materials among others) and additives for food industries.
Toluene is predominantly used as an industrial feedstock and a solvent. p-Xylene is the principal precursor to terephthalic acid and dimethyl terephthalate, both used in the production of polyethylene terephthalate (PET) plastic bottles and polyester clothing and also used as a solvent.
Syngas can be used in the Fischer–Tropsch process to produce diesel, or converted into e.g. methane, methanol, and dimethyl ether in catalytic processes as already discussed.
Automated Manufacturing of any Compound
Artificial intelligence is being developed to plan the synthetics routes (=a series of steps to be followed in order to make a chemical compound from smaller and less complex chemicals) for compounds. The target is that AI could be used to tell what steps are needed to manufacture of any compound. When this is combined with an automated flow chemistry unit where robots assemble the production process to follow the “recipe”, we end up with an automated production facility that can produce any compound. Naturally given the limitations of available materials and flow chemistry units.
This even when there are no pre-made instructions. Just tell what molecule is needed, AI will figure out the steps and robots will assemble a local temporary “factory” to produce it.
One of the first uses will be in medicine. It will speed up new synthetic drug discovery because now the slow step is to developing the synthesis process for a potential medicinal molecule. When this is combined with crowdsourced funding for researchers for new therapeutics, the building blocks for a decentralized medicine are starting to form. Decentralized funding for teams that use decentralized synthesis facilities.
Such a facility can naturally produce other types of end products as well, no need to limit it to medicine. Its just that the need and interest is biggest there now.
To probe further:
A robotic platform for flow synthesis of organic compounds informed by AI planning.
Next: Summary of Unfactory