The effect of process variables on the dispersion process with applications to paint processing

Kazembe, Callisto (2011-10)

Thesis (MScIngwet)--University of Stellenbosch, 2005.

Thesis

ENGLISH ABSTRACT: Paint manufacturing through the dispersion process is an important part of the chemical industry that relies on limited and expensive supplies of titanium dioxide pigment. The aims of the project were to: (i) identify and evaluate the factors that affect the opacity and flocculation gradient, (ii) establish and evaluate the mechanism of the process, and (iii) evaluate and apply the rheology of the dispersion process. The mechanism of the dispersion process can be broken down into the following steps: (i) wetting, (ii) particle break down, and (iii) stabilisation against agglomeration. Wetting can be further broken down into adhesional, immersional and spreading wetting, which can be evaluated in terms of the contact angle. Titanium dioxide pigment particles are held together in the agglomerate state through attractive Van der Waals forces. These forces must be broken down through shear stresses applied through the Cowles mill or the homogeniser. The sensitivity analysis that was carried out confirms that Van der Waals forces are effective only for sub-micron-sized particles. Acoustic cavitation also increases the rate of particle break down in a homogeniser and it depends on turbulence intensity. The mechanism of dispersion can be evaluated in terms of the Reynolds number. The opacity and flocculation gradient of paint were found to depend on: (i) the mean pigment particle size of titanium dioxide and extender, (ii) the particle size distribution of titanium dioxide, (iii) the pigment volume concentration of titanium dioxide and extender, and (iv) the wavelength of the incident radiation. Correlations of opacity (contrast ratio) or flocculation gradient could be set up on the basis of the above explanatory variables in terms of a multiple linear regression. However, it was found out that the methods used for measuring the contrast ratio and flocculation gradient were unreliable. In the case of the contrast ratio, there was no standard procedure for preparing pigment dispersion samples, thus resulting in values that were very high and insensitive to process changes. Samples whose contrast ratio has to be determined must be diluted with resin. Measurements of the flocculation gradient were found to be erratic with very low linear association. Samples of pigment dispersions processed through the dispersion process were found to be shear thinning. Flow curves obtained showed that the high shear rate and low shear rate behaviour of dispersions were significantly different. Of the processing variables investigated, dispersant had the greatest impact on the dispersion process. It significantly affects wetting and the particle break down rate. Samples of dispersions from the dispersion process are thixotropic and recover their structure in a reasonable time after the application of shear stress. Overall, the agitator speed and diameter did not impart a significant effect on the dispersion process.

AFRIKAANSE OPSOMMING: Die vervaardiging van verf met die dispersie proses vorm ‘n belangrike deel van die chemiese industrie wat staatmaak op die verskaffing van ‘n beperkte voorraad, duur titaniumdioksied pigment. Die doel van hierdie projek was om (i) die faktore te identifiseer en te evalueer wat die ondeurskynendheid en flokkulasie gradiënt beïnvloed, (ii) die meganisme van die proses vas te stel en te evalueer, en (iii) die reologie van die dispersie proses te evalueer. Die meganisme van die dispersie proses kan verdeel word in die volgende stappe: (i) benatting, (ii) afbreking van partikels, (iii) stabilisering teen agglomerasie. Benatting kan verder verdeel word in adhesiebenatting, onderdompelingsbenatting en spreibenatting, wat dan in terme van die hoek van kontak ge-evalueer kan word. Titaniumdioksied pigment partikels word deur Van der Waals kragte bymekaar gehou in die agglomeraattoestand en hierdie kragte moet gebreek word met behulp van skuifspanning wat deur die Cowles Meul en Homogeniseerder aangewend kan word. ‘n Sensitiwiteits analise het bevestig dat die Van der Waals kragte slegs effektief is vir sub-mikron grootte partikels. Akoestiese kavitasie verhoog die tempo waarteen partikels breek in die (homogeniser) en dit hang af van die intensiteit van die turbulensie. Die meganisme van dispersie kan in terme van die Reynolds getal ge-evalueer word. Daar is gevind dat die ondeurskynendheid en flokkulasie gradiënt van die volgende afhanklik is: (i) die gemiddelde pigment partikel grootte van die titanium dioksied en aanvuller, (ii) die partikelgrootteverspreiding , (iii) die pigment volume konsentrasie van die titanium dioksied en die aanvuller, en (iv) die golflengte van die invallende bestraling. Korrelasies van die ondeurskynendheid of flokkulasie gradiënt kan opgestel word op grond van die bogenoemde verduidelikende veranderlikes in terme van ‘n veelvuldige linêere regressie. Daar is egter gevind dat die metodes vir die meet van die ondeurskynendheid en flokkulasie gradiënt nie betroubaar is nie. In die geval van die ondeurskynendheid was daar geen standaard prosedure vir die voorbereiding van die pigment dispersie monsters nie en die gevolg was waardes wat baie hoog was en onsensitief was vir proses veranderinge. Monsters waarvan die ondeurskynendheid bepaal moet word behoort met ‘n hars verdun te word of by relatiewe lae vastestof inhoude te wees. Flokkulasie gradiënt metings was baie wisselvallig. Pigment monsters wat deur die dispersie proses geprosesseer is , is pseudoplasties. Die vloeikurwes dui daarop dat die hoë skuif tempo en die lae skuif tempo gedrag van die dispersies grootliks verskil. Van die proses veranderlikes wat ondersoek is het die dispersant die grootste invloed op die dispersie proses gehad het. Dit het die benatting en partikel afbreking grootliks beïnvloed. Dispersie monsters van die dispersie proses is thixotropies en herwin hul struktuur redelik gou na die aanwending van skuifspanning. Die roerspoed het geen betekenisvolle invloed op die dispersie proses gehad nie binne die beperkte grense in roerspoed wat ondersoek is.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/16609
This item appears in the following collections: