Thesis (MScEng)--University of Stellenbosch, 2002. / ENGLISH ABSTRACT: The objective of this study is the assessment of the usefulness of an industrial
machine vision system for laboratory studies of especially flotation froths and other
foams. With this in mind it was attempted to meet three objectives. The first objective
is the adjudication of the machine vision system's ability through image analysis to
differentiate between characteristics of two-phase and three-phase systems. Secondly,
through the use of image analysis the effect of varying surfactant concentration, as
well as the addition of quartz particles on the foam/froth stability was quantified and
thirdly it was attempted to define the nature of foam decay as stochastic or
deterministic. While not as important in the industrial setting, the last objective is
quite important when a laboratory system is concerned, even more so owing to the
difficulty of quantifying the collapse or decay of foams and froths.
Two experimental set-ups were used to attain above-mentioned objectives. A Leeds
flotation cell was used for the flotation of the quartz with cetyltrimethylammonium
bromide (CTAB) and octadecylamine (ODA) and as well as for the characterisation of
the CTAB and ODA foam. A glass column was used to investigate the CTAB foam
column decay. In both set-ups the foam/froth structures were monitored with an
ELMO charged coupled device (CCD). The CCD was connected to a personal
computer equipped with frame grabber, which captured and digitised the images.
The following conclusions can be drawn from the experimental results:
The machine vision system can differentiate accurately between
surfactant groups of different concentration and thus groups with unique
characteristics.
11 The stability of foam is highly correlated with all other foam variables
extracted from the image analysis system. In particular, the movement of
the foam/froth in the flotation cell has a substantial influence on the
stability values obtained for the foam/froth. III As far as the experimental system was concerned, small quartz particles (-
75 urn) stabilised the froth, while coarser particles (75-106 urn, 106-150
urn and 150+ urn) destabilised the froth.
IV Likewise, the decay of CTAB foam in a column was non-linear and
deterministic. When observed manually, the decay of long life foams
follow the same trend as short-life foams and short-life emulsions in that a
specific decay constant can be calculated for a specific surfactant system.
V The machine vision system is a viable alternative for current stability
measurements. Firstly, no data processing need to be done to calculate a
indirect stability parameter as it computes a direct stability parameter and
secondly it also computes the bubble size distribution of the sample.
The following recommendations are made:
The experimental design of the system needs to be altered to minimise the
effect the experimental environment has on results. The possibility of a
different camera set-up must be investigated. The focus of the camera on
the foam surface needs to be accurate as the computation of the stability
value depends on the quality of the image obtained.
Il The four factors that affect the stability of a flotation froth, namely solid
concentration, particle size, particle type and particle hydrophobicity
should be investigated in the same study. There exists many conflicting
studies on this subject as many studies for different particles with a
variety of sizes, hydrophobicity and concentration are performed. One
study should be conducted where all four these factors are investigated
under the same experimental conditions.
III The method with which the image analysis system deal with fine froth
need to be addressed. In laboratory systems a typical system investigated
will show a dense foam with small bubble distributions as in the case of
CTAB. The extent to which the grouping of fine froth as one bubble
influence the instability values obtained needs to be investigated . / AFRIKAANSE OPSOMMING: Die doelwit van hierdie studie is die evaluering van die toepassing van 'n industriële
masjienvisiesisteem in 'n laboratorium studie van hoofsaaklik flottasie skuim. Die
doel van hierdie toepassing is drie-ledig. Die eerste doelwit is om die
onderskeidingsvermoë van die masjienvisiesisteem tussen groepe met verskillende
eienskappe te evalueer deur intydse beeld verwerking. Tweedens is die invloed van
oppervlakmiddelkonsentrasie veranderinge asook die byvoeging van kwarts partikels
op die stabiliteit van die skuim ondersoek. Derdens is die aard van die skuim verval
ondersoek om vas te stelof die proses deterministies of stochasties is. Alhoewel
laasgenoemde doelwit nie so belangrik is in die industriële toepassing nie, is dit van
uiterse belang in die laboratorium studies, veral as gevolg van die probleme wat
ondervind word in die kwantifisering van skuimverval.
Twee eksperimentele opstellings is gebruik om bogenoemde doelwitte te bereik. 'n
Leeds flottasiesel is gebruik vir die flottasie van kwarts met CTAB en ODA asook vir
die karakterisering van CTAB en ODA skuimeienskappe. 'n Glaskolom is gebruik vir
die ondersoek van die aard van skuimverval. In beide opstellings is die beelde
gemonitor deur 'n ELMO CCD. Die CCD is verbind met 'n rekenaar wat toegerus is
met 'n raam-vanger wat die beelde vang en digitiseer.
Die volgende gevolgtrekkings kan uit die eksperimentele resultate gemaak word:
Die masjienvisiesisteem differensieer akkuraat tussen oppervlakmiddel
groepe van verskillende konsentrasies en dus ook tussen groepe met
unieke eienskappe.
11 Die skuim stabiliteit korreleer met die ander skuim eienskappe naamlik
gemiddelde area, gemiddelde omtrek, aantal borrels, fynheid van die
beeld (SNE), beweging van die skuim en kleur van die skuim. Veral die
beweging van die skuim in die flottasiesel het 'n beduidende invloed op
die skuimstabiliteit. 111 In die genoemde eksperimentele sisteem IS gevind dat klein
kwartspartikels (-75 urn) skuim stabiliseer terwyl groter partikels (75-106
urn, 106-150 urn en 150+ urn) skuim destabiliseer.
IV Die verval van CTAB skuim in 'n kolom is nie-lineêr en deterministies. In
die geval waar die verval van die CTAB met die hand gemeet is, is gevind
dat die verval van 'n lang-lewe skuim dieselfde tendens as kort-lewe
skuim en kort-lewe emulsies volg. 'n Spesifieke verval konstante kan
bereken word vir 'n spesifieke oppervlakmiddel sisteem.
v Die masjienvisiesisteem IS 'n werkbare alternatief in die meting van
skuimstabiliteit. Anders as in bestaande stabiliteitsmetings waar 'n
indirekte stabiliteitsparameter bereken moet word, bereken die sisteem 'n
direkte stabiliteitsparameter asook die borrelverspreiding van die skuim
monster.
Die volgende aanbevelings kan gemaak word:
Die eksperimentele ontwerp van die sisteem moet aangepas word om
sodoende die invloed van eksperimentele kondisies op die resultate te
verminder. Die moontlikheid van 'n alternatiewe kamera opstelling moet
ook ondersoek word, aangesien die verkrygde stabiliteitswaardes
afhanklik is van die kwaliteit van die skuimbeelde.
11 Daar is talle teenstrydige studies ten opsigte van die vier faktore wat die
stabiliteit van flotasie skuim beinvloed naamlik erts tipe, partikel grootte,
partikel hidrofobisiteit en vastestof konsentrasie. Hierdie faktore moet
gelyktydig in 'n studie ondersoek word onder dieselfde eksperimentele
kondisies.
111 Die masjienvisiesisteem groepeer fyn skuim gedeeltes saam as een borrel.
Hierdie wyse van fyn skuim hantering moet aangespreek word aangesien
'n tipiese oppervlakmiddel sisteem in die laboratorium, soos in die geval van CTAB, digte skuim met 'n fyn borrelverspreiding toon. Die mate
waarin die onvermoë van die sisteem om fyn borrels te herken die
stabiliteitswaardes beïnvloed, moet ondersoek word.
| Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:sun/oai:scholar.sun.ac.za:10019.1/52972 |
| Date | 03 1900 |
| Creators | Ellis, Rene |
| Contributors | Aldrich, C., Stellenbosch University. Faculty of Engineering. Dept. of Process Engineering. |
| Publisher | Stellenbosch : Stellenbosch University |
| Source Sets | South African National ETD Portal |
| Language | en_ZA |
| Detected Language | English |
| Type | Thesis |
| Format | 243 p. : ill. |
| Rights | Stellenbosch University |
Page generated in 0.0127 seconds