TBP01x 5.2 Upstream processing
Welcome to this unit, which is dedicated to Upstream Processing, dealing with preparing
high quality feedstocks from a diversity of biobased raw materials, for further conversion
into a range of chemicals, fuels and energy.
Much of the useful technologies are related to common agro/food products, or those from
forestry such as the pulp and paper industry. Also in those cases, the raw material or
feedstock does not automatically fit into the process equipment and must be prepared to
make it fit.
The structure of 1st and 2nd generation biomass was discussed earlier in this course. The
conventional fermentation feedstocks such as sugars and starches are produced by mature
agro/food processing technologies. These open the biomass structure and recover the
sugars and starch products. The lignocellulosic residue was originally considered to be a
waste product, but is now increasingly used for the (combined) power and heat production
in the biorefinery.
Since there is often overproduction, alternative uses, such as for the production of liquid
fuels and chemicals, are emerging.
These can add potentially more value but also require more investments. Pretreatment is
needed for both types of biomass, but is more complex for 2nd generation feedstocks. Also
other products can be recovered such as oils and lipids, proteins and nutrients. Each of these
requires dedicated technologies and will not be treated here.
The structure of the Upstream Processes is rather uniform but technologies are not
identical.
The common overall structure requires mechanical steps for size reduction (or modification)
and size selection, opening the biomass structure, depolymerisation when necessary (in case
of lignocellulosics or other biopolymers), extraction of fermentable components such as
sugars, further conditioning for instance to remove inhibitors, and finally: conversion to
desired products with or without further purification.
For the first step (size reduction), there is a multitude of milling, grinding, cutting, chopping,
slicing, mulching and shredding technologies available.
There is even very specialised equipment such as the Tilby contactor that can split, scrape
and sort sugar cane in one go. Also for particle size selection, there are many methods. The
exact equipment depends strongly on feedstock and application and here we only show a
few examples of the possibilities. Besides size reduction, pelletising is another form of size
modification.
For proper performance, some additional and very practical aspects must be taken into
account. These deal with removal of undesired objects such as knives, bottles and stones.
Processing time as well as prior and subsequent storage are critical since in most cases we
deal with living and metabolising plant material. It is also subject to fungal and animal
invasion. Also, even regular drying on air can change the biomass structure strongly and
make subsequent steps even more costly and difficult.
Transport of biomass to, in and from the processing plant of is also a key aspect. Note that
broad particle size and density distributions could give rise to particle segregation (small
particle group together as well as larger particles do), leading to uneven flow, etc. A
potential solution is to standardise the particle size (distribution)  for instance by
pelletisation, since the operating range of this sort of equipment is limited. Another aspect is
the operational robustness: mechanical processes often lead to equipment wear and for
instance the most sensitive parts need to be replaceable. And last but not least, the large
scale of operation usually also gives rise to a high power consumption.
This is partially dissipated in the feedstock, leading to thermal and other related effects.
When the raw biomass is processed to more or less uniform particles, these can be treated
with high temperature, acids and bases, and chemicals such as solvents to make the
structure more permeable and facilitate depolymerisation for lignocellulosic fractions.
Important criteria are the amounts and efficiency of chemicals and energy use. This varies
greatly with the various methods. This can be measured in terms of (final) yield of sugars,
but also which and how many byproducts are formed. And last but not least, this also
depends a lot on the feedstock used, which is also subject to seasonal variations.
This slide shows a more elaborate comparison of the options mentioned on the previous
slide. It should really be seen as indicative, and certainly not absolute.
Some of the chemicals used for pretreatment, also have an effect on the depolymerisation
of the biomass, especially the acids and bases . Another route is to use the natural process
of enzymes. The latter are not only used for processing lignocellulosic feedstocks but also
when 1st generation feedstocks such as starches are used.
The next critical step in preparation of the feedstock for the conversion process, is the
efficient extraction with water of  fermentables , which are often the sugars. This can be
 regular C6 sugars such as in conventional sugar cane and sugar beet processes, but also
hydrolysate sugars which are mixtures of C5 and C6 sugars plus the byproducts mentioned
before.
For reasons of efficiency, the extraction with water is done in a countercurrent way. In sugar
beet processing, this is in a so called  diffusion tower , whereas for sugar cane this is done in
a cascade of countercurrent mills that each act as an equilibrium stage. Starch is usually
recovered as a particle stream, since starch itself is also a (food) product, and subsequently
liquefied.
The extract stream is sometimes treated further to adjust colour and composition, but this is
very specific for the various feedstock conversion technology combinations. Let s take a look
at some examples such as sugars (sucrose) from sugar beets and sugar cane, glucose sugars
from (corn) starch and also hydrolysate sugars from bagasse or straw, and form wood such
as in pulp & paper industry. We will not deal with oil crops, but it should be noted that the
residual biomass coproduced with oil crops (palm, soy etc), also is a promising source of
lignocellulosic material. Also, we will not deal with all details given the objective of this
course, but it is important that you note that quality control of feedstocks is expensive and
requires considerable investments.
It should also be noted that most of the current processes are optimised for instance for
pure consumption sugar, and not for efficient production of a feedstock for further
conversion.
Producing sugar from beets starts with washing, slicing and water extraction in a diffusion
tower, with subsequent conditioning with lime and carbonate to remove impurities,
followed by concentrating by evaporation and a cascade of crystallisers. This process is
clearly optimised to produce crystal consumption sugar. As a fermentation feedstock, the
liming step may be replaced (thereby less lime waste), and also the capital and energy
intensive crystallisation omitted. This results in drastically lower CAPEX and OPEX.
Production of sugar from sugar cane shows a comparable lay out: the cane enters the plant
in a cascade of 5 or more mills, that crush the cane and simultaneously extract sugars in a
countercurrent manner. Note that this enormous machinery consumes roughly half of the
energy use of a sugar cane mill, and represents about half of the capital investment.
The so called  juice is stabilised and clarified, and subsequently concentrated by multi effect
evaporation. The concentrated juice is used for fermentation, often in combination with
molasses, if part of the sugar is also crystallised like in the example of sugar beets.
It is clear that this process has already seen more optimisation for use as a fermentation
feedstock.
Wet milling of corn (there is also dry milling which yields another product), is an even more
complex biorefining process that produces a range of products (germ, oil, gluten), but also
starch products.
The latter are products on their own but can also be hydrolysed to produce glucose syrups.
As glucose is a less sweet sugar than fructose, these streams are partially isomerised to
fructose and marketed as High Fructose Corn Syrup which you typically find in your
beverages. It is also clear here, that there are many optimisations possible when the sugars
serve as a fermentation feedstock.
Can you indicate at least 5?
Comparing the earlier sugar and starch production processes, you see now many similarities
with the emerging lignocellulosic processes. Depending on situation, they can be bolded on
an existing biorefinery or fully dedicated to new crops such as energy cane and others. For
instance in the case of sugar cane, the feedstock can be bagasse that is already milled, so
you can skip that step. But it requires pretreatment, hydrolysis, and subsequent
fermentation and further downstream processing. There are also many opportunities to
share facilities such as steam and thereby effectively reduce energy and water use.
Compare this scheme to the previous slide, and indicate at least 5 opportunities for sharing
facilities and saving energy and materials use.
But remember that lignocellulosic biorefineries are just emerging and that the first two only
started in the second half of 2014. Here you see the Brazilian  GranBio plant, and the US
POET DSM facility.
There is a lot of information available on these two plants, which I encourage you to look up
online!
The last example is the processing of wood to pulp for paper production.
In that case, intact cellulosic fibers are a clear target and the original processes are
optimised around that.
Essentially, the process efficiently removes lignin. Extending this process to include
production of fermentable molecules requires depolymerisation of the cellulosic fibers.
This is structurally comparable to the production of fermentables from starch, although the
details of the actual process will be different. Also the availability of a reasonably pure lignin
stream gives different opportunities, which we will not discuss here.
Summarizing, Upstream Processing is an integral and key element of the complete
technology for biobased products.
It generally requires significant capital investments and is responsible for considerable
energy and water consumption, when not  done right .
In the next units, we will explore how to integrate with conversion and further downstream
processing, so see you in the next unit!