Textile Chemist and Colorist Agst 199O/Vol. 22, No. 8 Basic Principles of Vacm Slot xtraction By ARTHUR D. BROADBNT. University of Sherbrooke, Sherbrooke. Qebec, Canada acm extraction removes excess V liqid fromopen width textile material as it passes over a narrow slot in a tbe connected to a vacm pmp. t has been sed for this prpose since the beginning of the present centry. The pressre differential across the fabric forces mobile liqid from the interyarn spaces into the slot, and then the reslting airflow penetrates into the yarns and removes mch of the liqid adhering to the srface of the fibers. Althogh mechanical removal of water from textiles is mch cheaper than thermal drying, the vacm techniqe was not widely sed in the past becase of poor reprodcibility of the final water content. ABSTRACT Vacm slot extraction is becoming more poplar in textile wet processing becase of its energy efficiency and other economic benefits. The basic principles of vacm slot extraction are reviewed. The fnction of the components of a vacm extractor and the mechanism of liqid removal from the fabric are discssed. The effects of the most important variables controlling the final water content are illstrated sing data obtained with a laboratory vacm extraction nit. Areas reqiring frther research to improve the vacm finishing techniqe are described. KY TRMS Airflow Air Permeability xtraction qipment Finishing Techniqes Vacm xtraction Little thoght was given to nderstanding basic principles and the factors inflencing dewatering. When energy costs soared after 197, however, interest in vacm extraction was revitalized. Modern extractors are spplied with appropriately sized pmps, vacm level controls, slots with low friciton shaped to minimize retention of lint, and systems for effective separation, filtration and recycling of extracted chemical soltions. The chemical natre of the fibers present in the fabric and its air permeability are also considered in prediciting performance. Process optimization is vital to remaining competitive in both domestic and ' international markets. The vacm extraction techniqe aids in process optimization. t gives effective liqid removal from many textiles and yields excellent niformity of the residal liqid from sidetoside and endtoend of the piece. The prpose of this paper is to review the basic principles of textile vacm extraction. t serves as backgrond for sbseqent papers dealing with research on the specific applications and problems encontered in the se of this techniqe. qipment A simple vacm extractor can be made by ctting a narrow slot along the length of a steel pipe connected to a vacm pmp. The orifice, however, is preferably formed from two parallel plates, monted on top of the slot in the vacm tbe. The se of a material with low friciton and the actal shape of the orifice formed between the two plates are important factors for minimizing accmlation of lint scked from the cloth srface and for ensring good contact and fit of the fabric with the slot. The vacm tbe shold have a diameter large enogh to avoid any pressre drop in the air flowing throgh it. The tbe may have a plastic lining to avoid adhesion of lint. f delicate fabrics are vacmed, more spport may be reqired. n place of a linear orifice, a series of holes or narrow slots arranged obliqely or in a herringbone pattern can be sed. Comparison of the different types of orifices is difficlt becase of the many variables involved in effective dewatering. Herringbone slots, for example, give improved fabric spport bt tend to case greater accmlation of lint which can case streakiness. Under some conditions they may be somewhat more efficient than a linear slot becase they give longer exposre to the vacm and less deformation and opening of the fabric. The exposre time to the pressre differential is a key factor determining the efficiency of water removal. t is determined by the slot width, geometry and the fabric speed. The se of a slot of variable gap along its length has been recommended for controlling sidetoside variations in the pickp of dyes and chemicals (1,2). This may be sefl for prodcing or correcting transverse shading effects bt a linear orifice of fixed width sally gives 1 Table 1. Water Retention of Different Textiles After Optimal Vacm xtraction Fibers Water" Cotton 51 2 % Polyester staple 12 5 2 5/5 Polyester/Cotton 2? 2 65/5 Polyester/Cotton 25? 2 Polyester filament 5? 1 Viscose staple 72? 4 Wool 42? 5 Nylon staple 22 * 2 Acetate filament 16? 1 Polypropylene filament 2 + 1 =Percent water on a bone dry basis Vacm level 8 cm Hg Fabric speed 4 6 m/min Agst 199 1
Vacm xtraction excellent sidetoside niformity of residal liqid even when the initial distribtion is neven. A woven fabric passing over the vacm slot is sally held between the pad rollers and a set of drive rollers rnning at slightly higher speed (Fig. 1 ). This ensres that the cloth is nder appropriate tension and correctly aligned with the slot. Vacm extraction is not limited to woven fabrics. Some knitted materials can be handled with appropriate tension and decrling controls. f tension indced deformation of the knitted strctre is a problem, the fabric can be spported on a mesh conveyor and vacmed in a relaxed state throgh the mesh (). n order to control the vacm level and ensre high speed airflow throgh the fabric, the slot beyond the selvages mst be sealed. This can be achieved simply by sing tape or a spring activated strip or tbe. Pnematic or photoelectric devices can be sed to atomatically detect the selvages and activate movement of sealing strips bt reqire freqent maintenance to avoid problems cased by accmlation of condensation or lint. Once the liqid has been removed from the web, it is swept down the vacm tbe by the air flx along with any lint. n simple vacm dewatering, lint shold be filtered from the airstream before it passes into the pmp. f the liqid extracted contains valable chemicals, it can be separated from the airstream, along with any lint, by decreasing the air speed. A box type separator, possibly with internal baffles, provides a simple means of doing this. xcellent pmp protection and chemical recovery yields of higher than 97% are feasible sing a cyclone separator whose design prevents reentrainment of the liqid (4). The Advantage of Recycling One of the advantages of vacm slot extraction is the ability to recycle any chemical soltion separated from the airflow. Liqid can be drained continosly from the cyclone by means of a barometer leg, if space permits (5). Alternatively, a pmp capable of working against the vacm in the cyclone separator can be sed to recirclate the extracted soltion back to the reservoir or pad bath on an intermittent or continos basis. To avoid excessive accmlation of lint in the system, recirclated soltion is sally filtered. To clean the filter dring continos operation, the soltion may be diverted throgh a bypass or a parallel filter. f a cyclone separator is sed for lint removal, it is possible to have atomatic ejection of lint and liqid withot releasing the vacm. Three Types of Pmps Three types of vacm pmps are commonly sed: rotary lobe positive displacement (PD), liqid ring (LR) and centrifgal exhast (C) pmps. The choice depends on the reqired airflow, the level of its contamination, the cost of the pmp and its operation and maintenance, the nmber of slots to be operated in parallel and the water consmption (6). The LR and PD pmps are constant volme, variable vacm pmps. They reqire water injection for cooling, effective sealing and flshing. PD pmps have lower clearance than LR pmps and are more efficient. They also have mch lower water consmption, which can have a definite impact on operating costs. n simple dewatering or extraction of solble waste chemicals, as in washing, extracted liqid may pass throgh an LR pmp operated withot a filter. The high flow rate of water necessary to maintain an effective ring of liqid may be sfficient to continosly flsh lint and chemicals from the pmp (5,7) bt this shold be careflly checked for textile applications. Centrifgal pmps are the most efficient. They rn at constant vacm, with variable power and airflow, and do not reqire water injection. A C pmp shold not be sed for textile vacm extraction nless all liqid and lint have been effectively removed from the airflow. Vacm Control For most applications, the vacm mst be controlled within reasonable limits ( k 1 centimeter Hg) to ensre niformity of residal liqid along the web. This is more critical for the extraction of chemical soltions to ensre even distribtion of the chemical remaining in the fabric. The Fig. 1. A typical indstrial pad/vacm nit with filtration and recirclation of the extracted soltion. () fabric; (2) pad rollers; () atomatic slot sealer; (4) vacm slot; (5) drive rollers; (6) chimney, mffler and drain; (7) btterfly valve; (8) vacm pmp; (9) recycling pmp; (O) filter; (1 ) cyclone separator; (1 2) vacm tbe; (1) pad bath; (14) recycled soltion. Note: This figre of a commercial nit was selected solely for illstrative and not promotional prposes. vacm level may be controlled in.a variety of ways. One of the simplest and most effective methods is by means of a btterfly or ball valve controlling the bleeding of air into the system via a chimney, which also serves as a mffler and drain. The movement of the valve may be activated by feedback from a pressre or hmidity measring system (electric signal from a pressre gage, transdcer or radiation gage). Mechanism of Vacm xtraction n discssing vacm extraction, it is important to distingish between bond and nbond water. Bond water is that absorbed by the fiber and held between the polymer chains in the amorphos zones by intermoleclar forces sch as hydrogen bonds. The amont of bond water depends on the hydrophilic or hydrophobic natreof the particlar fiber. Many textile materials can be vacm extracted down to moistre levels close to the water retention vales determined by centrifgation (8). Table shows some typical residal water contents of fabrics after efficient vacm extraction sing a laboratory scale nit. These vales represent the bond water content of the fibers pls a small contribtion from residal interfacial water accmlated at fiber intersections. Comparable reslts can be obtained nder prodction conditions with indstrial extraction systems provided the fabric has adeqate residence time at the vacm slot. Unbond water is held by srface tension in the interyarn and interfiber spaces. Sch water can migrate easily and the blk of it can be removed by vacm extraction. This involves two steps: sction of water from the yarn interstices becase r* 7" 25 Y. U 't 2. 4 15 1 26 9 Vacm (cm Hg) Fig. 2. Airflow throgh the fabric as a fnction of the pressre differential. Fabric: 5/5 polyester/cotton, 2/2 twill, 189 g/m2. Fabric speed:.7 m/min. Fabric width: 25 cm. Air permeability of the conditioned fabric: 24.6 cms"r2(18). Airflow determined at thepressre in the vacm manifold from velocity measrement singa Pitot tbe. 14 CCO Vol. 22, No. 8
of the pressre differential between the two faces of the fabric, and sbseqent elimination of most of the interfacial water by the high velocity airflow throgh the yarns. The viscos drag of the air creates shear forces sfficiently powerfl to overcome the interfacial tension holding water to the fiber srfaces. There will be some migration of water to fiber intersections dring this phase to minimize srface tension. When the pressre drop decreases as the fabric leaves the vacm slot, relaxation of residal distorted water drops reslts in retention of a small amont of nbond interfacial water, held predominantly at fiber intersections (9). ffective liqid removal depends on generating the maximm shear force by having the highest possible air speed throgh the yarns and ensring an adeqate exposre to the vacm. There is little evaporative drying cased by the airflow at ambient temperatres. Well extracted fabrics do not demonstrate any significant decrease in moistre content on repeated extraction. Airflow s The Key Vacm extraction can be considered in terms of airflow throgh a mesh screen or thin poros sheet as a conseqence of the pressre differential across the layer. Flid flow throgh sch materials follows irreglar patterns preclding exact soltions of the eqations of motion sed in flid dynamics. Correlations of experimental data based on theoretical models for flow throgh screens or thin poros layers have been pblished (4,lO12). These are generally based on eqations inclding both laminar (viscos) and trblent (inertial) flow for idealized strctres, and they cold be sefl for correlation of air flow data in vacm slot extraction. These correlations are based on q. 1 relating the pressre drop, AP, and the flid approach velocity, v. The two constants, a and @, depend on the characteristicsof the layer material (thickness, porosity) and the moving flid (viscosity, density). qations of this form can also be expressed in terms of the two dimensionless parameters, the friction factor, f (proportional to AP/v2) and the Reynolds Nmber, Re (proportional to v) and two constants, A and B (q. 2), related to the viscos and inertial resistance to flow, respectively. APlv = a + @v f = ARe + B q. 1 q. 2 As the flaw rate increases, the vale of Re increases and the friction factor approaches a constant vale of B as the inertial resistance to flow becomes dominant. The actal vales of the constants A and B depend on the theoretical model sed to represent the strctre of the fibros web. Agst 199 ccx) 1 c,x Y L,x il m o\o 5 %. 1 26 9 Vacm (cm Y]) Fig.. Residal water retention as a fnction of pressre differential for two fabrics of differing air permeability. Fabric speed:.7 m/min. Fabric A: 18/82 polyester/cotton, 1 / 1 plain weave, 2g/m2, 5/5poiyester/cotton warp, l/9 polyester/cotton filling, air permeability: 64. ~m~s'cm~ (18). Fabric B: 1% cotton, 1 / 1 plain weave, 217 g/m2, air permeability: 19.5 cms'u" (S). An empirical relationship describing the airflow throgh a textile web as a fnction of the vacm level, the orifice area and the permeability of the fabric has been pblished (1,14). t has been called the Albany airflow eqation. ts application is difficlt becase of ncertainties in the mathematical signs and the nits of some of the variables. As with all the existing airflow correlations, it is not of general validity bt limited to the materials and conditions sed in developing it. The flow of air throgh a textile fabric is throgh a series of small orifices in parallel. Consideration of a simple orifice meter (4) allows development of some important basic principles applicable to vacm slot extraction. As the pressre differential is increased, the airflow throgh the fabric increases, as does the velocity of the air in the channels. Once the air reaches a velocity eqal to that of sond in air, its flow rate and velocity throgh the channels cannot be increased frther. This is the critical flow condition. ven if the pressre downstream can be decreased frther, this will not be registered in the actal channels becase the pressre wave in a compressible flid can only travel at sonicvelocity and the pressre dropcannot be transmitted pstream. For passage of air from an open face throgh a narrow section of fabric over a vacm slot, the critical pressre ratio, rc (Le., the ratio of the absolte pressre in the orifices in the fabric and of the entering air) is given by g., with a vale of.528 for air, which has a specific heat ratio, k. of 1.4 at 25C. rc= [2/(1 + k)lk/ck') q. Ths, ideally, maximm air velocity 1 26 9 Vacm (cm Hg) Fig. 4. Water retention of varios fabrics as a fnction of pressre differential on vacm extraction. Fabric speed:.7 m/min. Fabric A: 1% cotton, 2/2 twill, 25 g/m2. Fabric B: 5/5 polyester/cotton, 2/2 twill, 197 g/m*. Fabric C: 1% polyester, textrized filament, 2/2 twill, 2g/m2. throgh a fabric shold be reached at a vacm level of abot 6 centimeters Hg as shown in Fig. 2. Reslts from Clemson University have also demonstrated sch critical airflow for a large nmber of fabrics at abot the predicted vacm level (1 5). Fabric Permeability The permeability of the fabric plays a decisive role in determining the amont of water remaining in the cloth after vacm extraction. The airflow rate throgh the material can be considered to be the prodct of air velocity and the total area of the available channels. For textiles of relatively compact constrction, the maximm air velocity throgh the fabric is sally easily attained becase of the constricted natre of the channels. Considerable qantities of water can be removed at modest vacm levels and a relatively constant residal hmidity is established at higher levels (Figs. B and 4). The optimal vacm extraction will be obtained when the airflow reaches acostic velocity in the fabric channels, as this creates the maximm shear force for water removal from fiber srfaces. f the cloth being extracted is more open, the airflow throgh the channels will begreater bt may never attain thecritical flow condition. This is becase the pmping speed may not be adeqate to increase the pressre drop to the reqired level. Therefore it will be difficlt to strip water from the fibers. Residal hmidity decreases with increasing vacm level bt more slowly for open than for compact fabrics. With open fabrics, residal hmidity may never reach the bond water platea as in the previos case. 15
Vacm xtraction These two behaviors, are illstrated in Fig.. n removing nbond liqid from the material, it is the air speed and not the volmetric flow which conts. To achieve an air flow at sonic velocity, and ths with the greatest shear force to remove water from the fibers, the vacm level can be increased, or the area of fabric exposed to the vacm can be redced by decreasing theslot gap. Controlling Water Removal Few pblications are available which deal with the factors that inflence the final water content of vacm extracted textiles (16.1 7). The two most significant variables inflencing the qantity of residal water are the vacm level and the chemical natre of the fibers. Fig. 4 shows data obtained with a laboratory nit. n this case, the slot gap was.2 millimeters and the low fabric speeds gave residence times of abot 5 milliseconds, reslting in optimal extraction. The overall performance of the process also depends on the pmp capacity, the type of slot, the slot gap, theconstrction of the fabric and line speed. Many factors involved in fabric constrction inflence its air permeability, which determines whether critical flow conditionscan beachieved (Fig. ). Correlations of airflow in vacm extraction and fabric air permeability have been established (1,1.5). t shold be pointed ot, however, that the air permeability measrements in sch correlations refer to airflow at a pressre differential of.5 inches of water for conditioned fabrics (18), whereas in vacm extraction the airflow is throgh the wet fabric with a mch higher pressre drop. Fiber Swelling Recent work (12) has clearly demonstrated that fiber swelling on water absorption decreases the air permeability of a material. This can be qite prononced. For example, a conditioned cotton fabric (1/1 plain weave, 167 g/m2) with an air permeability of 45 ~ m~slcm~ (18) gave a vale of only 17 cms cm* after padding with water to a final retention of 74% H2. There is little data on air permeabilities either at high pressre differentials or when the fibers are in a swollen condition. Airflow throgh textile fabrics tends to prodce mch larger pressre drops than their relatively high porosities wold sggest. t wold be sefl todevelop frther a mathematical model for high speed airflow throgh a fibros web inclding the effects of the real volme available for the 16 Table. Water Content of a Vacm xtracted Fabric as a Fnction of its Speed Residence Fabric Time Speed at Slot Water 1.8 m/mm a7 ms 41.6%.6 4 44. 5.5 28 45.5 7. 21 46.6 9.1 17 47.6 Fabric: 5/5 polyester/cotton. 2/2 twill, 225 g/m2. Vacm level: 28 cm Hg. Slot gap: 2.6 mm. nitial hmidity of fabric: 9 & 2%. flow and of energy absorption by displacement of individal fibers (1 9). Sch information and modeling will be valable to correlate data on vacm extraction and to develop the capability of predicting the dependence of water content on vacm level and airflow. Residence Time The qantity of liqid removed from a fabric dring extraction depends on the residence time at the slot. For a linear orifice the qantity of liqid removed is governed by the fabric speed and the slot gap. A minimm residence time of abot 1 milliseconds (longer with heavy fabrics) is sally reqired for effective dewatering. Complete eqilibrim is not generally attained in vacm extraction and the water retention after extraction decreases with decrease in fabric speed (Table 11). The effect is not sally very prononced. Wider slots expose a greater area of fabric to the vacm and increase the residence time bt reslt in decreased air velocity nless a higher capacity pmp is sed. Provided that a minimm dwell time is exceeded, the fabric speed and slot gap do not greatly inflence the final hmidity, provided that sonic air speed can be attained. Under indstrial conditions, with prodction speeds approaching 1 meters per minte, optimal extraction down to the bond water level may not always be possible becase residence times are shorter (two to five milliseconds). n sch cases, a combination of two slots can improve efficiency provided the pmping capacity is adeqate. n many cases, the final hmidity after extraction decreases only slightly on increasing the vacm level above 2 centimeters Hg (Fig. 4), and it may often be more economical to extract slightly less water at this lower vacm level than at a higher vacm reqiring a more powerfl pmp. For given conditions, the residal moistre is a minimm at an intermediate fabric tension allowing good contact with thevacm slot, minimal deformation and maximm air speed. Too high a tension may distort the material and ths decrease the air speed throgh it. This is partic larly tre for some knitted materials. The fabric may lose contact with the slot if the alignment is not exact. Too low a tension cases more fabric to be scked into the slot, increasing the exposed area and redcing the air speed. The initial water content of the fabric before extraction does not significantly inflence the qantity of water remaining after extraction provided the residence timeofthe fabricattheslot isadeqate. At high fabric speeds or low vacm levels, the retention of water after extraction increases slightly with an increase in the initial retention (Fig. 5). The nflence of Viscosity Little work has been pblished on the inflence of viscosity on the efficiency of vacm extraction. xtraction of dilte xanthan gm soltions of viscosity from.25 to.915 kg m s cased little variation in residal pickp of a 1% cotton fabric, indicating that viscosity may not have a prononced effect on extraction in the absence of other inflences (1.5). xtraction of NaOH soltion is qite difficlt and more complex. The chemical combination of the alkali with the celllose reslts in a mch higher retention of liqid, both before and after the vacm slot, and the extracted NaOH soltion is more dilte than either that initially applied from the bath or that remaining on the cotton. Althogh NaOH soltions increase in viscosity with increasing concentration, the inflence of viscosity on their vacm extraction is probably not very significant. Applications The major applications of vacm extraction dring textile finishing are: 5 > W Lc 4 44 4c c V L aj m oh? o A 75 1 125 2 Water Content Before Vacm Fig. 5. nflence on the initial water content of the fabric on the water content after vacm extraction. Fabric: 5/5 polyester/cotton, 2/2 twill, 225 g/m*. Fabric speed: 4.6 m/min. Vacm level: crve A: 26 cm Hg; crve B: 9 cm Hg. CCO Vol. 22. No. 8
Dewatering of fabrics beforcdrying. Dwatering before \betonct pad application of chemicals. to eliminate exchange of the water in the incoming fabric with the bath soltion. Removal of nwanted chemicals in washing dring preparation and after dyeing. Low addon finishing with good penetration. Lint removal. The pad/vacm techniqe is widely sed as a low leaveon chemical application method in textile finishing. Vacm extraction is also recognized as a sefl means of improving the efficiency of continos washing processes in preparation and after dyeing. Despite its wide se there is little pblished information on the inflence of vacm extraction on the amonts of chemicals retained by or removed from the textile in relation to the chemical's sbstantivity for the fibers or its physical form. The frther development of the vacm techniqe will depend on research of these aspects. Vacm extraction is not of niversal applicatioq and has some specific limitations. t is not sitable for removal of liqid from fabrics with very compact or very open constrctions. Neither type of material permits high speed airflow. Knitted fabrics reqire special handling, with appropriate tension control and decrlers. n manycases, they are preferably vacmed throgh a mesh spport in a tensionless condition. There may be occasional difficlties with excessive accmlation of lint in the separator or filters, and, with some finishing mixes, foaming or cracking of dispersions may occr on extraction. The applications and problems of vacm extraction in textile finishing will be discssed in a sbseqent paper. Conclsions Vacm extraction is a techniqe with considerable merit. This paper has dealt with the basic principles of the techniqe. ts se reslts in considerable savings in drying and in recovered chemicals. t can play a key role in improving the efficiency of washing operations. The sccess of the vacm techniqe in dewatering and in finishing has led to its almost niversal se in North American finishing mills. More research is reqired to flly evalate the following relationships: fabric permeability and air velocity; the removal of nfixed chemicals and of improved mass transfer in washing; and the effects of the varios pertrbations sch as affinity and evaporation whrch can case changes of the concentrations of soltions recycled to the pad bath. Acknowledgements The work on vacm extraction was generosly spported by VAC nc. of Spar tanbrp, S.C., Dominion Textile nc. (Beaharnoisand Magog tinishing plants) and the Natral Sciences and ngineering Research Concil of Canada. m References () Faile. T..\.. Bod i!/ Ppcvy. loy6.4.47cc' lnri~rnrinl Coirjrrmce & shihirirtn. :\ilanta. p2hh. (2) O\tervold. J. Lnd (\ Robertaon..4iiwricii Dwsrii,!f Rrprriv. Vl. 75, ho. 9, September 1986. p5. () Mickler. P.. 8ook /. Pprri. YXY,4.47CC' lnrerntionl Conference &.rhibi/in. Philadelphid. ~25. (4) Perry, R. H, D. W. Green and J.. Malone). Perr),'.s Chemical ngineers' Handbook. Sixth dition, McCrawHill lnc.,new York. 1984. (5) Bhntia. M. V. and P. h. CheremiainoR. Process yiptnen/series. Yo/.,.4ir.L/oremenr and Vaciini rs. Technoinic Pbliahing Co.. Weatport. Conn., 1981. (6).Adnmczyk. P. L. Trxriie Chrtwsr rd Colorisr.Vol. S,No.ll.Sovemberl98.p1. (7) Ryans, J L And D. L. Roper. Process Viiin y WRNKL RCOVRY WRNKL RCOVRY RPLCAS This series of dimensional plastic replicas has been developed lor se in evalating the appearance of textile fabrics after indced wrinkling by means 1 the AATCC Wrinkle Tester (AATCC Test Method 128) By sing these replicas. vales of 5.4,,2 or 1 may be assigned in rating a fabric The set 1 five replicas. in a protective case. is now available from the AATCC Technical Center lor $255 (Order No 89) AMRCAN ASSOCATON OF TXTL CHMSTS AND COLORSTS P.O. BOX 12215, RSARCH TRANGL PARK, N.C. 2779 TLPHON: 91915498141 FAX: 919/54989 Agst 199 CO 17