Introduction.—The process of beating has for its object the complete breaking down of the bleached pulp to the condition of single fibres, and the further reduction of the fibres, when necessary, into smaller pieces. The disintegration of the material is essential for the production of a close even sheet of paper, and the amount of beating required varies greatly according to the nature of the raw material, and the class of paper to be produced.

The textile trade, on the other hand, depends on a raw material composed of strong fibres, or of filaments characterised by great length, and any processes of treatment which tend to reduce the length of such fibres are carefully avoided, and it is therefore obvious that fibres which are of no value for textile purposes can be appropriated for paper-making.

Condition of Fibres.—The great differences in the physical characteristics and structure of the fibres employed for paper-making suggest that the possible variations in the final product obtained by beating are very numerous. This is a well-known fact, and it is further to be noted that this mechanical operation brings about not merely alterations of a physical order, but introduces some interesting and important chemical changes.

Of the better-known materials linen, with an average fibre length of 28 mm., the structure of which lends itself to considerable alteration by beating, is in marked contrast to esparto, the fibre length of which is only 1·5 mm. If the process of beating a linen rag merely resulted in the cutting of all the fibres of 28 mm. long into short fragments of 1·5 mm., there would be nothing remarkable in it, but the changes which occur in reducing the long linen fibre to 1·5 or 2·0 mm. are of a far more important character than this.

Early Methods.—In the early days of paper-making the disintegration of the half-stuff was effected by a true “beating” process, the rags being subjected to the action of heavy stampers, which broke up the mass of tangled fibre into a uniform pulp. The fibres for the most part retained their maximum length in this operation, which was exceedingly slow and tedious, though at the same time giving a sheet of paper of remarkable strength.

The nearest imitation of these old-time rag papers is to be seen in the well-known Japanese papers, which are extraordinarily strong. Some of these the writer has examined in order to determine the length of the fibre. The sheets when held up to the light appear “cloudy” and “wild” owing to the presence of the long fibres, which have only been separated or teased out by the primitive methods of beating used, and not completely disintegrated.

Conditions of Beating.—About A.D. 1700 there began a great epoch in the history of paper-making. With the invention of the Hollander engine about A.D. 1670, the process of disintegration was greatly hastened, because it was possible to reduce the half-stuff much more readily. The substitution of the idea of plain “beating” by a principle which combined the gradual isolation of the individual fibres with a splitting up of those fibres lengthwise and crosswise was not only an advantage in point of economy of time and cost, but also a material advance in the possibilities of greater variations in the finished paper.

The conditions of the process of beating carried out with a Hollander permit of considerable alteration, so that these changes in the fibre are not surprising when properly understood. In fact, it is now conceded that a close study of the theory and practice of beating is likely to bring about still more remarkable improvements in this important department of the paper-maker's work. The quality and character of the paper made may be varied with—

(1) The origin of the raw material, e.g., rags, esparto, or wood;

(2) The condition of the material, e.g., old or new rags, green or mature esparto, mechanical or chemical wood pulp;

(3) The time occupied in beating, e.g., four hours for an ordinary rag printing and twelve hours for a rag parchment;

(4) The state of the beater knives, e.g., sharp tackle for blottings and dull tackle for cartridge papers;

(5) The speed of the beater roll, also its weight;

(6) The rate at which the beater roll is lowered on to the bedplate;

(7) The temperature of the contents of the engine.

The Beater Roll.—If the beater roll is fitted with sharp knives, and this is put down close to the bedplate quickly, the fibres are cut up short, and they do not assimilate the water. If the roll is fitted with dull knives, or “tackle,” as it is sometimes called, and it is lowered gradually, the fibres are drawn and bruised out without being greatly shortened. In this condition the stuff becomes very “wet,” or “greasy,” as it is termed. The cellulose enters into association with water when beaten for many hours, and the pulp in the beating engine changes into a curious greasy-like mass of a semi-transparent character. Rag pulp beaten for a long time produces a hard, translucent, dense sheet of paper. Flax thread beaten 48 to 60 hours is used in practice for the manufacture of gramophone horns and similar purposes.

Soft porous papers like blottings, filtering papers, heavy chromos, litho papers, antiques, light printings, are made from pulps which are beaten quickly with the roll put down close to the bedplate soon after the stuff has been filled in.

With strong, dense, hard papers, such as parchments, banks, greaseproofs and the like, the pulp is beaten slowly and the roll lowered gradually.

The nature of the pulp and the time occupied in beating are also important factors in producing these different papers, three to four hours being ample for an ordinary wood pulp printing, whereas a wood pulp parchment requires seven to eight hours.

Beating Pulps Separately.—The use of esparto and wood pulp in conjunction with one another, or blended with rag, has introduced new problems into the question of beating. Perhaps the most important of these is the advisability of beating the pulps separately and eventually passing them through a mixer of some kind before discharging into a stuff chest. The necessity for differentiating the amount of beating is already partly recognised when very dissimilar pulps, such as strong rag and esparto, are blended, but the whole subject ought to be carefully studied by the paper-maker and investigated on its merits from the standpoint of “beating effects,” apart from questions of cost and expediency. The former fully understood and exhaustively examined by practical tests would of course only be developed if proved to be advantageous.

The field of research in this direction has not yet been seriously explored. With the enormous consumption of wood pulps of varying quality made from many different species of wood by several processes, there is ample room for interesting and profitable enquiry, particularly as the types of beating engine are so numerous. The effects produced by the Hollander, the refiner, the edge runner, the stone beater roll, and other mechanisms, are all of varying kinds.

Effect of Prolonged Beating.

The importance of a knowledge of the precise effects produced by the beating of pulp cannot be emphasised too much, and any contributions to the subject along the lines of special research will be welcomed by all students of cellulose.

Some experiments were conducted by the writer in 1906 with cotton rags, in order to determine the results obtained by beating the pulp for a prolonged period under exact and specific conditions.

The cotton rags, of good quality, were boiled with caustic soda in the usual way for six or seven hours, at a pressure of 15 to 20 lbs., washed and partially broken down in the rag breaker, and finally bleached, made into half-stuff, and then transferred to a Hollander beating engine.

The particular conditions specified for the beating operation were that the beaterman should manipulate the pulp according to his usual routine for the manufacture of the paper which he was accustomed to make from these rags. In this case the routine process meant beating for eight hours, by which time the pulp was ready for the paper machine. In the ordinary course the pulp would be discharged into the stuff chest, and converted into a strong, thin, bank paper.

During the prolonged beating the pulp became very soft and “greasy,” and when made up into sheets the paper as it dried exhibited remarkable differences in shrinkage, the dry sheets obtained from pulp beaten thirty-seven hours being much smaller than those obtained from pulp beaten only four or six hours. The actual shrinkage is shown in the following table:—

If these results are plotted in the form of a curve the relation between the period of beating and the shrinkage in area is clearly shown. For the first twenty hours the shrinkage is proportional to the period of beating, after which the curve assumes an irregular shape, showing a tendency for shrinkage to proceed at a faster rate.

Weight and Substance of the Paper.—The shrinkage of the paper after prolonged beating indicates a closer and denser sheet, so that for papers of equal thickness the weight per unit area was much greater in the case of the pulp beaten for the full period. The results obtained are very interesting, and the following summary for a few of the readings obtained will serve to show the alteration effected.

Sizing and Glazing Effects.—The behaviour of the waterleaf paper after sizing and glazing gave some interesting results. In the first place, the effect of the altered density of the paper is strikingly shown by the amount of the size absorbed. Certain selected sheets were passed through a solution of ordinary gelatine in the usual way, and subsequently dried. The amount of gelatine absorbed differs in a remarkable degree, as shown in table.

Tensile Strength of the Paper.—It is interesting to note that the tensile strength of the waterleaf papers appears to remain fairly constant throughout the whole period of beating. But this uniformity is greatly altered by the operations of sizing and glazing.

Percentage of Air-dry Gelatine absorbed by the Waterleaf Sheets.

These results are rather remarkable. The prolonged beating does not seem to have affected the tensile strength of the waterleaf, and the practical loss of strength which actually occurs in the more completely finished paper does not manifest itself until after the sizing process. The importance of the gelatine as a factor in the ultimate strength is thus clearly and strikingly demonstrated.

Tests for Strength on Original Waterleaf Paper.

Tests for Strength on Papers, Sized only.

Tests for Strength on Paper Sized and Glazed.

It may also be noticed that the strength of the finished paper after twenty hours' beating, as in class B, is equal to that of the paper after nine hours' beating, as in class A. This is curious, especially in view of the fact that the percentage of gelatine in the papers of class B. is only 4·4 per cent. as against 6·0 per cent. in class A.

The relation of the percentage of gelatine to the period of beating thus becomes a matter of interest, and well worth investigation. The figures are suggestive of further experimental research along definite lines.

Developments in Beating Engines.—Since the introduction of the Hollander beating engine, about A.D. 1670, other types of beater almost too numerous to mention have been devised to supersede it, but the fact remains that the principle of the original Hollander and its general design are still adhered to in the engines used by paper-makers for high-class work.

The alterations and improvements which have taken place during the last fifty years relate chiefly to the modifications naturally arising from the introduction of fibres not requiring such drastic treatment as rags.

The machines now in use for reducing half-stuff to beaten pulp ready for the paper machine may be classified as follows:—

(1) Beaters of the Hollander type, in which the circulation of the pulp in the engine and the actual beating process are both effected by the beater roll.

(2) Beaters of the circulator type, in which the movement of the pulp is maintained by a special contrivance, and the beater roll used only for beating.

(3) Beaters of the stone roll type in which the roll and bedplate are either or both composed of stone, granite, or similar non-metallic substance.

(4) Refiners, containing conical shaped beater rolls working in a conical shell fitted with stationary knives.

The Hollander.—This beating engine in its simplest form consists of an oval shaped trough, divided into two channels by a “midfeather,” which does not, however, reach completely from one end to the other.

In one of the channels the bed of the trough slopes up slightly to the place where the “bedplate” is fixed. The bedplate consists of a number of stout metal bars or knives firmly fastened into an iron frame, which lies across this channel. The beater roll, a heavy cast-iron roll provided with projecting knives or blades arranged in clumps of three around the circumference, and supported on bearings at each side of the engine, revolves above the bedplate with the knives adjusted to any required distance from it, the raising or lowering of the beater roll for this purpose being effected by the use of adjustable bearings.

The bed of the trough behind the beater roll rises sharply up from the bedplate and then falls away suddenly, as shown in the diagram, forming the “backfall.”

When the engine is in operation the mixture of water and pulp is drawn between the knives and circulated round the trough. The material is disintegrated into fibres of the required condition, discharged over the backfall, and kept in a state of continual circulation, and the beating maintained until the stuff has been sufficiently treated.

The dimensions of the engine vary according to the capacity, which is usually expressed in terms of the amount of dry pulp the beater will hold, and the following figures may be taken as giving the average sizes:—

Sundry modifications in the form and arrangement of the beater have been tried from time to time. In 1869 Granville patented the substitution of a second beater roll in place of the stationary bedplate for the purpose of hastening the operation. Repeated attempts have been made to construct a beating engine with two or more rolls, but it is evident that such a device could hardly succeed, since it would be impossible to ensure proper adjustment of the rolls, and in that case one roll might be doing all the work.

The first machine of this type was patented in 1872 by Salt. Similar beaters were devised by Forbes in 1880, Macfarlane in 1886, Pickles in 1894, who proposed to use three rolls, and Partington in 1901. Hoffman describes a beating engine which was working in America containing four rolls, as shown in the diagram.

The Umpherston.—A notable modification of the Hollander, having an arrangement by which the two channels of the engines are placed under one another, and one which is largely used for fibres, is the Umpherston. Several engines differing in detail, but embodying the same principle, have been built in imitation of this one.

Bedplates of large working surface were first tried in England by Cooke and Hibbert, in 1878, but in practice it has been found that no serious deviations from the narrow type of plate are of much value. As a matter of fact it is held by some paper-makers that one or two knives would be sufficient if they could be relied on to keep true and in proper adjustment.

The Circulating Type of Beater.—The addition of some device for keeping the pulp in circulation apart from the action of the roll has received considerable attention. The early experiments in this direction with the Hollander led ultimately to the construction of the engine of the circulator type mentioned in class 2.

Thus, in 1872, Nugent patented a special paddle to be used in the Hollander, by which the pulp in the trough of the beater was impelled towards the roll. Many other plans were tried for this purpose, and details can be seen in the List of Patents (see page 192).

The introduction of the beaters with special means of circulating the pulp was found to be of the greatest service in the treatment of stuff like esparto and wood pulp, since these materials did not require the drastic measures necessary with rag pulp. In 1890 several engines of this class were being adopted, amongst which may be mentioned Hemmer's, Reed's and Taylor's. The pulp discharged from the beater roll was drawn through an independent pipe or channel by means of an Archimedean screw, or a centrifugal pump.

Stone Beater Rolls.—The substitution of stone for metal in the roll and bedplate of the engine brings about some remarkable changes in the nature of the beaten stuff. The fibre is submitted to the action of rough surfaces rather than that due to the contact of sharp edges, with the result that the disintegration is much more rapid, and produces a “wet” working pulp suitable for imitation parchments and similar papers. The latest materials used for this purpose are basalt lava stone in Germany, and carborundum in America.

Care is necessary in the manipulation of these beaters to prevent fracture of the stone parts. In the Wagg Jordan engine this danger is materially reduced by the construction of the working parts.

Refiners.—In these engines the beater roll is a conical shaped drum carrying the knives, which revolve inside a conical shell completely lined with fixed knives. The fibres are thus cut up to the desired length, but before discharge from the engine they pass between two circular discs, one stationary and the other revolving in a vertical position. The effect of the discs is to tear or bruise the fibres rather than to cut them.

The refiner is best employed to clear or brush out the mass of pulp after a certain amount of preliminary treatment in the beater, as the refiner cannot produce the effects obtained by actual beating as in the Hollander.

Power Consumption.—The long treatment required to thoroughly pulp a strong material demands a great amount of power. Engines differ considerably in their power consumption, and comparisons are frequently made in terms of the power required to beat a given weight of pulp. But this is not always a true criterion of efficient work. Some types of beater are suitable for producing certain results, and the mere substitution of a beater consuming less power is worse than useless unless it can be shown that the same effects are being obtained. The efficiency of the Hollander for the beating of rag pulp, in spite of the high power consumption, is a case in point.

With this factor properly considered, the power required for beating becomes an interesting study. Many detailed experiments have been published from time to time, the most recent being those described by Beadle.

Patents taken out in Connection with Beating Engines.

1855. Park (1170).—A small steam engine was attached to the shaft of the beater roll, so that it could be driven direct.

1856. Kingsland (2828).—A form of refiner in which the pulp was beaten by a vertical disc rotating in an enclosed case.

1860. Jordan (792).—A machine devised for mixing size with pulp, made like a conical refining engine, the rubbing surface being provided with teeth or cutters.

1860. Jordan (2019).—An engine of the refiner type, constructed with a conical drum rotating in a conical casing. The knives at the larger end of the drum are placed closer together than those on the smaller end.

1863. Park (1138).—Two beaters placed side by side are driven by one steam engine placed between them, the operations being so timed that one rag engine is used for breaking while the other is finishing.

1864. Ibotson (2913).—The pulp is passed continuously from one engine roll to another, or from one part of a beater roll to another part of the same roll through slotted plates.

1866. Roeckner (140).—A beating engine of the refiner type with conical drum and casing.

1866. Berham (3299).—A beating engine of the conical type with the beater roll rotating vertically instead of horizontally.

1867. Crompton (482).—Device for raising the bars in the beater roll as the edge of the plate wears away.

1867. Wood (914).—Modification in the form of the beater bars (of little importance).

1867. Edge (3673).—The knives of the beater roll distributed at equal distances apart all round the roll, alternated with strips of wood.

1869. Granville (1041).—Substitution of a second beater roll for the stationary bed-plate, the knives being set spirally round the roller.

1869. Newell (2905).—Weight of the beater roll counterpoised to allow of the exact regulation of the pressure on the stuff in the beating engine.

1870. Rose (997).—An intercepting plate fixed to the cover of the beating engine which causes that part of the stuff which was usually carried right round by the roll to fall back behind the backfall.

1870. Bentley and Jackson (1633).—A beater roll having the same width as the engine, and provided with a cover fitted with a pipe which conducted the material back to the front of the roll.

1871. Patton (1336).—Bottom of beating engine curved in order to prevent the stuff settling or accumulating at any portion of the machine.

1872. Salt (1901).—A beating engine of usual type, but having two beater rolls and two drum washers, one pair in each of the two channels.

1873. Gould (769).—A curious engine with horizontal shaft having a circular disc at the lower end, fitted with knives on the under-surface, which are in contact with fixed knives lying at the bottom of the vessel. The circulation of the pulp is effected by the centrifugal force generated.

1873. Martin (3751).—A beating engine with two rolls in the same trough, the first roll working in conjunction with a smooth surfaced beating roll, the other being in contact with a bedplate of the usual type, the object of the first roll being to partially disintegrate the material without danger of choking.

1874. Johnstone (3708).—A pulping engine in which the rubbing action of two grindstones one upon the other is utilised as a means of beating.

1876. Gardner (307).—A beating engine in which the beater roll is conical in shape, working vertically in contact with the bottom of the beating engine, which is also conical in shape, the engine itself being circular.

1878. Cooke and Hibbert (4068).—The bedplate constructed in the form of a circular segment with a much larger face than usual, and capable of adjustment, the beater roll itself being fixed in the bearings.

1880. Forbes (692).—A long oval shaped beating engine divided into three channels instead of two. In the two outer channels are placed beater rolls and drum washers. The stuff discharged over the backfalls from the two beating engines flows down the central channel and is circulated by a special paddle constructed in such a manner as to deliver the pulp in two equal streams into the outer channels to each of the beater rolls.

1880. Umpherston (1150).—An engine constructed with a passage below the backfall so that the stuff circulates in a trough underneath the beater roll, the object being to ensure more effective treatment and to save floor space.

1883. Aitchison (5381).—A beating engine of usual form, but with the beater roll made conical in shape with the larger circumference outwards, and the bedplate placed on an incline parallel with the knives on the beater roll.

1884. Mayfield (2028).—The backfall of the beating engine is of entirely different construction to the ordinary machine, for the purpose of improving the circulation.

1884. Hoyt (11177).—An engine resembling the Umpherston, but with a larger roll, the diameter of which is equal to the full depth of the engine, the backfall being in a line with the axis of the beater roll.

1885. Jordan (7156).—Additions to the Jordan engine for admitting water and steam to the engine as required.

1885. Korschilgen (9433).—The beater roll made of stone or of metal with a stone casing furnished with ribs or knives placed close together.

1886. Hibbert (4237).—A beating engine fitted with an ordinary beater roll, and having in addition a heavy disc rotating vertically, the disc being fitted with knives on one surface which rotate in contact with knives fixed on a stationary disc.

1886. Kron (9885).—A device for securing better circulation of the pulp, the stuff leaving the beater roll being divided into two streams which are brought together again in front of the roll.

1886. Horne (10237).—A long rectangular vessel with a large beater roll at one end, contrived so as to force the pulp leaving the beater roll to pass down a partition separating it from the pulp going towards the beater roll.

1886. Macfarlane (11084).—An engine fitted with two beater rolls which rotate in opposite directions, the stuff being mixed between them.

1887. Nacke (746).—A centrifugal circulating wheel rotating horizontally in the centre of the beating engine is used in combination with a parallel cutting disc.

1887. Marshall (1808).—A conical refiner having in addition at its large end a pair of grinding discs fitted with knives and rotating vertically.

1887. Voith (6174).—An alteration to the covers of the beater rolls which prevent stuff from being carried round the cylinder, and cause it to pass over the backfall freely.

1890. Hemmer (17483).—A beating engine provided with a separate return channel for the pulp, the circulation through the channel being effected by a small centrifugal pump.

1890. A. E. Reed (19107).—A beating engine in which the pulp discharged over the backfall is delivered to the front of the beater roll by a screw propeller.

1891. Karger (11564).—A beater similar to the Umpherston, but provided with a circulating roll fitted with radial projections which delivers the stuff to the front of the beater roll.

1892. Taylor (7397).—A beating engine in which the beater roll operates in a closed chamber above the vat full of pulp, the stuff being continually circulated by a centrifugal pump which draws the stock from the bottom of the vat and delivers it to the beater roll.

1892. Annandale (9173).—A conical-shaped beating engine with the beater roll rotating in a vertical position; the larger end of the cone being downwards.

1892. Umpherston (15766).—An addition to the beating engine arranged so that two fixed bedplates are used instead of one.

1892. Miller (15947).—A machine in which two fixed bedplates are used, one below the beater roll and one above, the engine being fitted with suitable baffle plates to ensure proper circulation.

1893. Pearson and Bertram (11956).—A special form of refining engine in which the pulp is subjected to the action of discs rotating vertically, the knives being arranged radially on the disc.

1893. Caldwell (15332).—A rotary beating engine in which the beating surfaces admit of accurate adjustment.

1894. Cornett (945).—An outlet is fixed to the beater roll casing close to the discharge from the bedplate, so that the roll is not impeded by the weight of the pulp, which is subsequently pumped to the front of the beater roll.

1894. Shand and Bertram (4136).—A beating engine similar to the Umpherston beater in which the beater roll is raised up out of the pulp and the circulation effected by means of a worm which delivers the pulp to the front of the beater roll.

1894. Pickles (20255).—A beating engine somewhat similar to an Umpherston, but fitted with three beater rolls and bedplates.

1894. Hibbert (25040).—A beating engine in which the pulp is beaten between two discs rotating vertically, the pulp being brought between the discs through the hollow shaft of one of the discs.

1895. Brown (1615).—An engine in which the beater roll and bedplate both revolve, but in opposite directions, and at different speeds in order to draw out the fibres.

1895. Schmidt (24730).—A device by means of which the pulp discharged from the beater roll is diverted into supplementary channels on either side which come together again in front of the beater roll.

1900. Hadfield (2468).—An adjustable baffle board passing through the cover of the beater roll which prevents the pulp being carried round by the roll, more or less.

1900. Masson and Scott (5367).—An improved form of Taylor beating engine in which the chest of the engine is vertical instead of horizontal.

1901. Partington (24654).—A continuous elliptical trough provided with two beater rolls.

1902. Picard (19635).—Improvements in the form of the propellers used for circulating the material.

1902. Pope and Mullen (22089).—Improvements in propellers for circulating the pulp.

1903. Annandale (26012).—A new form of beating engine somewhat on the principle of a steam turbine.

1905. Bertram (1727).—A beater similar to Masson's tower beater, but in which a pair of reciprocating wheels fitted with projecting knives are used instead of a centrifugal pump.

1907. Wagg's Jordan Engine (6788).—A conical refiner fitted with specially arranged metal or stone knives.