|
|||||||
|
|
|
|||||
|
|
|||||||
|
|||||||
|
|
|
|||||
|
|
|||||||
MICROBIOLOGY
OF COFFEE PROCESSING.
Part 2/.
TAINTS AND OFF FLAVOURS.
Ken
Calvert.
RECAPITULATION:
In the first part, we discussed how one can get a better understanding
of how coffee is processed, by looking at it in terms of microbiological action.
The race is on between the soft rot bacteria, who do the best job of
fermentation, and those yeasts and moulds, which can cause off flavours in
the finished product.(1). To encourage those beneficial bacteria, warm
water,>20oC, neutral conditions and a good supply of oxygen
are required throughout the fermenting tank. It is sometimes confusing to
the layman to think that coffee fermenting under water can have sufficient
oxygen, but we only need to think of fish and how they live, and make sure
that we wash our coffee and change the water at least once every day, to keep
up that supply of dissolved oxygen. It
must be reiterated however that the fastest action comes from allowing the
tank to drain freely and atmospheric oxygen to permeate the mass.(1).
To get the best quality coffee, normal fermentation should proceed as rapidly
as possible, around 6-8 hours. Any coffee
fermentation that takes longer than 12 hours is just not doing it right!
And, as will be detailed in the next article, the final weight of coffee will
be 1-2% less for every day that the coffee is still in contact with water.The
best way to achieve this is to both warm and recirculate ones pulping water
while the factory is working. Indeed just recirculating it will allow its
temperature to build up, and by draining the water through the bottom of the
tank and then pumping it back to the pulper, the levels of warmth, dissolved
oxygen and enzymes are the same right throughout the tank. (8)
MAKING YOUR OWN ENZYMES:
Enzymes are the biological catalysts which living things use to make biochemistry happen. However, these precise acting catalysts also need to work in precise conditions. The bacterial enzymes that break down mucilage or pectin need to work in essentially neutral conditions, (9),
whereas all of the commercial pectin enzymes such as Cofepec and Ultrazym
are made from moulds or fungi, and as can be seen in Figure 1 like to work
in acid conditions.Novo. What
this means is that packet enzymes such as Ultrazym are ideal for recovering
from a fermentation that has gone wrong, because it will still work when the
levels of acidity have risen and the bacterial enzymes have all closed down.(12).
However, used under essentially normal conditions at pH 7, and ambient
temperatures, only a fraction of its effectiveness is available.
For fermenting coffee mucilage, natural bacterial enzymes are much more suitable. Even though there are no bacterial type pectinases
on the commercial market, it is relatively easy to make your own.
While it is equally as effective to use normal pulpage for this process, it is a pity to waste good
coffee when lights and floaters are just as effective.(2). The only problem of course is how to get an adequate
supply of them. If an Aagaard is not
available, then one may be able to flood the first tank that the pulpage is
going into for long enough to rake sufficient floaters off through a flotation
notch or into a bucket. Then the bottom drain can be opened to recirculate
the water back to the pulpers again. About
a bucket of floaters are required for every large fermenting tank.
All that is required to do is to fill the bucket of floaters with water,
put a lid on it to keep the solids under the water, and stop the fruit flies,
and then leave it for five days. It
should then be possible to strain off the solids, which now must be discarded,
and have about half a bucket of clear yellow liquid which is full of potent
bacterial enzymes.(2). If the pulping water is being recirculated,
then once the required amount of floaters have been recovered for the treatment
five days down line, then the brew can just be mixed into the pulping water.
This will assist in maintaining rapid fermentation even if the water is too
cold (1).
If recirculation is not possible,
then the brew should be diluted sufficiently to spread it evenly all over
the top of the tank of fresh pulpage when pulping is completed, and then allow
it to drain down through the tank. Using
either of these methods, overnight fermentation can be achieved, allowing
the reuse of those same tanks for the next days pulping. This process can double the capacity of an
existing factory with limited tank space.
The only problems that may occur is when the pulpage is allowed to
go solid in the tank.(3).
FRUITY FLAVOUR AND SOUR COFFEE.
Even
though the beneficial soft rot or lactic acid bacteria can get an early lead
in the race towards complete digestion of the mucilage, once their food supply
is used up, then they start to slow down and the slower growing yeasts can
overtake them and redigest the soluble mucilage a stage further.(9). As in traditional brewing terms, the yeasts
go on to produce not only alcohol but the bouquet of fruity flavours, that
makes every wine different. If the coffee is washed, then the alcohol will
be removed, and a fresh supply of aerated water will keep the bacteria ahead
in the race. However, leave the coffee/fermentage
in the tank for too long, or let it solidify, and the yeats will kick in.
The fruity flavours will establish and the alcohols will carry on to be converted to vinegar
type acids which spell 'sour coffee'. The
point to make about this is, that the onset of fruity flavour is not tied
to time, but to the completion of fermentation. If your coffee ferments in 12 hours, then you
will start to get fruity flavour at 16 hours. If fermentation takes two days, then the onset of fruity will commence
in 2.5 days. The moment that fermentation is complete, final washing should
commence. Early stage fruity flavours
can still be washed out, so there
is usually a few extra hours of grace, before even a thorough wash and soak
cannot still recover a batch of ‘overferment’ to finally produce fine parchment.
FLORAL, WINEY AND HERBY TAINTS:
Like fruity flavours, there are many specific
'bouquet' type taints that occur on the side of an alcoholic yeast fermentation.
The alcohols are more water soluble, and volatile enough to be removed
in the washing and drying processes, but the various aldehydes, ketones and
perfumery base type chemicals tend to dissolve into the essential oil of the
green bean and then reappear at roasting. (7).
ONION FLAVOUR:
There are of course many different organic chemicals which can be produced
not only by a variety of micro-organisms, but also by their supply of nutrients.
The onset of onion flavour for instance, comes about when the ratio
of soluble sugars to protopectins becomes too low.(10).
The initial quick build up of beneficial bacteria is fuelled by the
relatively high levels of sugars present within the ripened mucilage.
If however, excessive fresh water is used in pulping, or, more particularly,
in the use of an Aagaard pregrader, then most of these soluble sugars are
leached out before normal fermentation commences, and the beneficial soft
rot bacteria can be overrun later in the fermentation not only by the yeasts
but also by other bacteria which convert acetic and lactic acid to propionic
and butyric acids, the root cause of onion flavour. Once again, these faults can be minimised by
recycling the pulping and Aagaard water, should one be used. Maintaining a high level of sugars and enzymes
in the water will speed up the normal bacterial action. It must also be said
that the moment that the days pulping is completed this recycled water must
be discarded and fresh water be used to start off again the following day.
EARTHY, MUSTY AND PHENOLIC TAINTS:
The third group of micro-organisms relevant to coffee processing are the
moulds and fungi. As stated in the
previous article, if there is sufficient mucilage present to keep the bacteria
and yeasts multiplying, then the moulds do not get a chance to even germinate
, let alone to grow.(7). It is usually only after the parchment
has been washed free of mucilage and put out to dry, that mouldiness can occur.
Indeed the cleaner and the more well soaked the parchment is, the more
likely it is for moulds to develop on damp coffee. There is one theory circulating, that the recent cause of Rio flavour
is linked to the drive towards good quality soaked parchment.
The practical solution to these kind of musty, mouldy taints, including
the dreaded Rio flavour, is to get ones clean wet parchment skin dried as
rapidly as possible before any mould spores get a chance to germinate, particularly
in the centre cut. If the morning
that one intends to put coffee out to dry is wet or dull, then the coffee
should be kept soaking in the tank, under a fresh batch of water, until the
sun comes out and the surface water can be dried off as soon as possible in
one non stop operation. Mould spores
cannot germinate under water, and they cannot germinate on the surface of
dry parchment. The danger is dampness,
and that can be very insidious. Once
coffee has dried enough for the bean to shrink and separate away from the
parchment, there is a little micro climate in there under the parchment, fed
by moisture coming out of the still moist core of the bean.(5). If that drying coffee is wrapped up tight in
an impervious plastic sail, then just like ourselves, walking round in the
sun in a raincoat, instead of the moisture continuously leaving our bodies
as a dry vapour to keep us cool, it comes out as liquid drops of sweat. And
like our own sweat, that from half dry coffee has sufficient nutrients and
salts in it, to allow all sorts of microbiologicals to thrive and do a lot
of damage right underneath the parchment where we cannot see it. So, even on a wet day, every brief period that the rain stops, at least one or two sails must
be opened up, shaken and rewrapped, in short order to dispel the warm damp
air inside the sail and minimise on mould spore germination. During the course of the day, it should be
possible for each drying sail to be opened up at least once, even if only
for a few seconds.
SMALLHOLDERS AND RIO FLAVOUR:
Smallholders may think that they can avoid this problem, because they
only put their coffee out to dry when the sun is shining and it is worth the
while of someone to sit and watch it.
However, the danger area here is when half dry coffee is stored in a
impervious white plastic sack, and kept inside a nice warm house, while waiting
for the sun to shine. Every morning and
afternoon during bad weather, the bags should be taken out and the coffee
poured into another bag in order to get rid of the warm damp air. Once emptied, that bag should be well
shaken and turned inside out before
being refilled from the next one.
STINK COFFEE:
For good quality coffee production, the factory water and equipment must
be kept clean, but as in terms of the Onion flavour and Rio flavour mechanisms,
it is not just a case of the cleaner the better. Cleanliness can be overdone! Nevertheless,
the cause of stinker beans is not from unusual micro-organisms but from excessive
over fermentation with normal ones. Each day the factory tanks and machinery must
be cleaned to make sure that old beans are not retained for several days in
cracks and crevices to come out and contaminate a later batch of coffee.6.
Such extreme 'over fermentation' of beans left in small pockets tends to germinate
the coffee seed which then rapidly dies and leaves a black spot under the
parchment. If the bean has been skinned
in the pulping, then the sprout falls out to leave a hollow pit in the end
of the bean. These dead beans then
rapidly putrify and develop a cheesy and evil smelling texture which is very
evident when one is squashed or cut. Only
one or two such stinker beans can contaminate and spoil a whole batch of good
coffee.
OTHER FLAVOURS:
For the sake of completeness one should also mention those off flavours
and taints which are not caused by micro-organisms. Green, grassy and some harsh flavours are caused by picking and
processing immature cherry.(5). The problem is not so much to do with green unripe cherry which will not pulp, because that gets nipped
and discoloured and is then easily removed by hand picking or a good colour
sorter when the coffee is dry. Late
in the season, lots of cherry looses its green colour but does not go completely
red. Called under-ripe, this cherry will pulp easily, but along with other nearly
ripe cherries earlier in the season the silverskin is full of green chlorophyll.(4). This is very easily seen in fresh wet washed
parchment which shows up the colour of the silverskin underneath. The answer here is to sun dry the coffee as
much as possible, so that the ultraviolet light can bleach out the greenness
in the silverskin. When botanical specimens
are pressed and dried in the dark they remain green. But leaves dried out
in the sun go brown. Slight greenness often fades with time, and is
no longer evident at the final destination, but a lot of
unripeness will allow chlorophyll type chemicals to be absorbed back
into the oil fraction of the final product.
Bagginess, oiliness and coal
tar taints generally come from contamination.(7). Cheaper types of jute bags have excessive
amounts of, or low grades of, 'batching oils'. These are the oils used to
lubricate the bag making machinery, and manufacturers of high quality jute
sacking always lubricate with vegetable oils to avoid this problem. Over lubrication of Asaro dryers and drying
coffee on the road side are also possible
sources of this problem. Smoky flavours
tend to come from leaky Asaro burners and earthy, dirty, foxy and similar
taints and appearance are from too much skins in the parchment, not only during
over fermentation, but also at the
time of hulling, when the green bean is screwed up against the dry but still
dirty skins and parchment hulls. (4).
WHATS NEXT.
In the third and final part of this series we will look at how microbiology
can help us to understand the problems that can occur with processed green bean
in storage, premature aging, UV
fluorescence, rancidity and other problems that derive from the overheating of
coffee, either during mechanical drying or in the huller.
-oOo-
REFERENCES:
1/. Arunga R.O., 1973.
Enzymatic Fermentation of Coffee.
Kenya Coffee. Dec. 1973.
2/. Butty M. 1973.
Rapid fermentation of Coffee.
Kenya Coffee. July 1973.
3/. Gamble F.M., Wootton A.E., 1963.
A Note on the Very Slow Fermentations.
Kenya Coffee. Dec.1963.
4/. Kulaba G.W., 1981.
Coffee Processing Research. A Review.
Kenya Coffee. December 1981.
5/. Menon S.N., 1992.
Interconnection between Post Harvest Processing and
the Intrinsic
Quality of Coffee.
Cafe Cacao The. Juil-Sept. 1992.
6/. Northmore J.M., 1969.
Overfermented Beans and Stinkers as Defectives of
Arabica Coffee.
ASIC. 4th Coll. Amster dam. 1969.
7/. Ojijo N.K.O., 1993.
Some Common Aroma Notes of Coffee and their Chemical
Origins
or Associated
Substances. A Review.
Kenya Coffee. Nov. 1993.
8/. Rolz C., Menchu J.F., etal
1982.
Biotechnology in Washed Coffee Processing.
Process Biochemistry. Mar/Apl 1982.
9/. Vincent J.C., 1992.
Personal Communication.
Montpellier Research Station. CIRAD.
10/. Wootton A.E., 1961.
Fermentation and Onion Flavour.
Kenya Coffee. April 1961.
11/. Wootton A.E., 1963.
The Fermentation of Coffee. (In four parts.)
Kenya Coffee. July-October
1963.
12/. Wootton A.E., Brownbridge J.M., 1966.
The Use of Pectic Enzyme Preparations in Coffee
Processing.
Kenya Coffee. June 1966.