Updated 09-XII-2018

Sir William Crookes

This article was written by fellow lamp engineer and collector Edward J. Covington, and originally appeared on his own website of biographical sketches of persons involved in the lamp industry. Following his passing in February 2017, and with kind permission of his family, Ed's words have been preserved here in the hope of maintaining access to his writings for the benefit of subsequent generations.

William Crookes in 1856 (age 24), 1889 (age 57), 1911 (age 79)2

Some persons who contributed to the efforts to develop a practical electric incandescent lamp are better known for contributions to science rather than to the incandescent lamp. However, even if that is the case, contributory efforts to lamp development should be recognized and written about. One such effort early in the development days was that by Sir William Crookes (June 17, 1832 - April 4, 1919).

William Crookes was involved in part-time efforts in the incandescent lamp field, probably between 1881 and 1889. Perhaps his most significant contribution was his ability to reduce the vacuum in vessels to lower values than other workers had achieved, and this was the impetus for some experimenters, notably Joseph Wilson Swan, to continue work in this area. Crookes' efforts were notable, although little is said about him in the incandescent history books today.

Crookes Lamps
The following descriptions appear in the Dredge book1:
"The filament is mounted in the first place upon its permanent support, and is then placed with this support in a vessel from which the air can be exhausted. Vapour is then admitted, and the filament is rendered incandescent. The lamp bulb is made in two parts, the filament being mounted on the one part, and enclosed in the chamber described in Specification 3799 of 1881, to adjust its conductivity, the terminals of conductors being joined to the fixed terminals in the chamber. The two parts of the lamp are then brought together and fused, and the globe is exhausted and sealed as usual. The lamp is then finished in a small machine, where its neck is set with plaster into a wooden cup, the conductors passing through holes in the bottom. The projection on the neck of the lamp, left where the connection with the air pump is made, is loosely enveloped with paper to prevent contact between it and the plaster, and is received in a cavity b in the cup bottom. The terminals c are then screwed down on to the conductors, which are twisted round their shanks. In Fig. 479 the cup is partly of brass. The lamp foot is screwed into a socket, its terminals coming against horn-like springs, each in the shape of a segment of, say, one-third of an annulus bent upwards from the base to which one end is fixed. Or the lamp may be suspended from a button of insulating material, provided with two metal springs connected to the leads and entering holes in the lamp terminals. The lamp may be suspended directly from conductors passing to the ceiling, and secured to the lamp terminals by small levers having cam-shaped ends working in a slot in the terminal. The conductors are laid in mouldings along the ceiling, having removable caps, and connected to pieces of metal on the side of an insulating cylindrical boss extending out from the centre of the rosette. The conductors from which the lamps hang have their ends inserted into holes in the metal pieces, and a ring or slider is drawn over them. For side wall connections, two separate metal studs are each provided with a slider securing the conductors into a hole in the side of the stud. Each circuit is completed through a switch consisting of two bent conductors fixed to a wooden base. One bent conductor is depressed, and an eye on the other part engages with an arm on the first part. When the part having the eye is depressed by means of a button, the eye is removed and the arm springs back, breaking the circuit."

"Mention must be made of the Sprengel mercury pump, invented in 1873, which was also extensively used by the classic experimenters on the electric discharge. In this, the air was carried away by a succession of falling mercury pellets, between which it was trapped. It was with an improved form of this pump, devised by Gimingham and having a number of tubes for the mercury fall instead of one, that William Crookes, one of the most skilled workers on vacuum physics in the second half of the nineteenth century, carried out, for instance, his classic work on the radiometer. He used phosphorous pentoxide to absorb water vapour, precipitated sulphur to stop mercury vapour and reduced copper to stop the sulphur vapour entering his evacuated space. He claims to have reduced the pressure to 0.4 x 10-6 atm, say 3 x 10-3 mm of mercury, and says, 'Formerly, an air-pump, which would diminish the volume of air in the receiver 1000 times, was said to produce a vacuum', adding that the Sprengel pump had already rarified air a few hundred thousand times, which in those days was very good performance.
The influence that Crookes' work with a modified Sprengel pump had on the work of Joseph Swan was pointed out in a biography of Swan3, which was written by one of his daughters and a son:
"The invention of the mercury vacuum pump by Hermann Sprengel, in 1865, had shown the way to get a vacuum far superior to any previously attainable, and following upon this invention, Mr. Crookes (afterwards Sir William Crookes) had, in 1875, astonished the world by the exhibition of his radiometer and by the description of the improved means he employed for obtaining the near approach to a perfect vacuum which the construction of the radiometer demanded. It was the publication of Crookes' researches which led Swan to resume his attempts to produce a satisfactory electric lamp by means of an incandescent carbon conductor in an evacuated glass container."
On February 5, 1889, William Crookes wrote the following to his patent agents2:
"Messrs. Carpmael & Co.
"Dear Sirs,
"I write to inform you that I have sold my three patents, 261281, 379981, and 107982, to the Anglo-American Brush Electric Light Co., Belvedere Road, Lambeth, and I shall be obliged if you will in future send notices of fees falling due to them. I have sent on your letter of the 1st inst."
Abridgments of those three patents follow:
GB2612, June 15, 1881
"Incandescent lamps.—In making carbon filaments, paper, linen or cotton thread, pith, cotton wool or other convenient form of cellular tissue or cellulose is freed if necessary from silica by the use of hydrofluoric acid as described in Specification No. 1422, A. D. 1881, and is then treated with a solution of cuprammonia so as wholly or partly to destroy the structure of the cellulose; the ammonia and water are then allowed to evaporate and the copper is extracted by nitric or other suitable acid; the substance is then washed with water and dried, either stretched out or pressed between sheets of absorbent material. The cotton or thread may be twisted, crocheted, or plaited to form a band of cylindrical, flat, or other section. The substance before treatment with cuprammonia may be saturated with ammonia to expel air from its pores. Before allowing the ammonia to evaporate after the treatment with cuprammonia the substances may be washed with alcohol. The copper may be removed after the carbonization of the filament by means of acid or by volatilizing it by passing an electric current through the filament in a vacuum or inert gas. The threads may be twisted, plaited, or crocheted after treatment and before or after the removal of the copper. According to another process, a loose form of cellulose such as cotton-wool or Swedish filter-paper is entirely or nearly dissolved in cuprammonia solution, and the liquid is poured into a shallow dish with a level bottom and allowed to evaporate until a film of cupric cellulose is left which is then treated with acid, washed, and dried under pressure between sheets of absorbent material. The cuprammonia solution used for treating paper or thread may have cellulose dissolved in it. The copper may be left in the filament near the junctions with the leading-in wires. The ends of the filaments may be electroplated with copper or other metal, or may be coated with a thick syrupy solution of cellulose in cuprammonia solution, so as to form a good junction after drying and carbonizing. If the filaments are made of sheets of paper, linen, cotton, or wholly dissolved cellulose, pieces of convenient form are cut or punched out to form filaments. Thin strips may be cut from the sheet before it is quite dry and twisted into a thread or bent into an elongated U-shape. Carbonization is effected slowly, the filaments being packed between sheets of blotting-paper and placed in a metal box with powdered charcoal. Air is excluded by admitting coal gas or other suitable gas or vapour. The resistance of the filament is reduced after carbonization by depositing carbon upon it by heating it electrically in a vacuous chamber containing a hydrocarbon which has a high boiling point and a low vapour tension at ordinary temperatures. Naphthalene, xylol, or chloroform, as stated in the subsequent Specification No. 3799, A. D. 1881, may be used. The hydrocarbon may be placed in a separate vessel connected by a tap."

GB3799, August 31, 1881
"Incandescent lamps.—Relates to the manufacture of the glass bulb, and to methods of securing the leading-in wires in position, of regulating the resistance of the carbon filament, and of mounting the filament in position. In order to form the bulb, a glass cylinder is drawn into a wide neck A, Fig. 3, at one end, and, after sealing the neck and while it is still hot, the end is pressed inwards by means of a pointed two-pronged metal tool, thereby forming an inwardly-projecting portion B terminating in two pointed pieces K. The other end E of the glass cylinder is now closed and formed into a hemispherical end, and an exhausting tube C is sealed to the lower end of the neck A. When the glass is cool, the neck A is cut into two parts transversely, and the leading-in wires are sealed by means of white enamel or arsenic glass in openings in the pointed pieces K. Previous to being sealed in position, each leading-in wire may be coated with enamel and sealed within a short glass cylinder. After attaching the carbon filament to the leading-in wires, the two portions of the neck A are sealed together and the lamp is exhausted. The hollow at the base of the lamp may be filled with plaster of Paris or other like cement. A one-pronged tool having preferably a chisel-shaped end, or a tool having more than two prongs, may be used instead of the two-pronged tool. Compound wires having a core of copper or other conducting wire sheathed with platinum are preferably employed as leading-in wires. In order to regulate the resistance of carbon filaments, the ends of each filament are respectively attached to two small hollow platinum or copper terminals fixed on a bar of arsenic glass so as to keep them at a suitable distance apart. The filament is then placed in an exhausting receiver with the terminals of the filament placed upon a pair of pointed wires which are arranged in an electric circuit and are fitted in the non-conducting base-plate of the receiver, an electric current being sent through the filament while exhaustion is going on. In a modification, Fig. 8, the detached part of the neck of the bulb, to which are attached the leading-in wires and the filament, is suspended from the ends of glass-covered wires D which are fitted in a thick plate C, preferably of vulcanized fibre, at the top of the receiver. Mercury is poured over the top of the plate C to prevent access of air during exhaustion. If, on the current being passed through the filament, its resistance is found too high, hydrocarbon or chloroform vapour is admitted to the receiver to produce a deposit of carbon upon the highly-heated filament, and, when the resistance of the filament is reduced sufficiently, the current is stopped. Naphthalene, xylene, or other hydrocarbon may be used instead of chloroform, as described in Specification No. 2612, A. D. 1881. A number of filaments can be treated at one time by either of these methods. To increase the rigidity of a filament, especially one manufactured as described in Specification No. 2612, A. D. 1881, it is made flat and a twist of a quarter of a revolution is given to it in placing it in the ends of the leading-in wires."

GB1079, March 6, 1882
"Incandescent lamps - Carbon filaments are made from animal fibres or matters, such as silk, silk threads, hair, wool, silkworm-gut, horn, gelatine, or parchment, these materials being treated with cuprammonia and carbonized, although, according to the Provisional Specification, the treatment with cuprammonia may, sometimes, be dispensed with. A number of threads may be stretched in a rectangular frame of copper, nickel, &c;, and placed in a dipping-vessel containing the cuprammonia, after which they are removed and washed in dilute acid and then in water, the frame being then taken to pieces with the threads remaining on two of its bars. The threads are then shaped by lapping them round a glass rod and round four others equidistant from it, two above and two below, each thread passing from either of the bars up between the two bottom rods, out over the upper rod and down under the centre rod, while the two rods are weighted to keep the threads distended during drying. The threads are cut off from the bars and carbonized in layers, separated by powdered charcoal, in a closed-in iron or nickel box which is connected with a supply of coal gas while it is heated to white heat in a furnace. Some of the substances, such as silk, may be dissolved in cuprammonia, so that, on evaporation, a film or sheet is obtained from which threads may be cut. Or the silk thread &c.; may be passed through a saturated solution of cotton, or of any of the above substances, in cuprammonia, and then carbonized. Or an acid may be added to a solution of cellulose, silk, &c.;, in cuprammonia, a plastic substance being produced, which may be used alone, or with starch, gluten, &c.;, for making the filaments; or it may be used as a cement for thickening the filament ends or for connecting the filaments and their conducting wires and holders. Chloride of zinc or caustic alkali may be used in place of cuprammonia for treating cellulose, silk, &c.; The conducting-wires terminate in small hollow platinum cones, or in bent portions, or in twisted cup-shaped portions, &c.;, forming receptacles in which the filament ends are secured by cement composed of ground graphite, water, sugar, &c.; the graphitoidal carbon produced by decomposing a hydrocarbon gas or vapour by intense heat is preferably used. In some cases, the receptacles may consist of platinum foil in which the conductors are secured by riveting, and the cement may be made from the carbon deposit in the carbonizing-box. According to another method, the conducting-wire is bent on itself two or three times, and the filament end is inserted into it, pressed, and brushed over with cement, a glass cap being sometimes placed over the terminal. The filament ends may be electroplated and joined to the terminals by small copper, platinum, nickel, iron, &c.; tubes; a cement may sometimes be used and may consist of platinum powder and water, platinum yellow and water, or a mixture of these with the graphite cement. In depositing carbon on the filaments, to strengthen them, either of the chlorides of carbon, the bromides of carbon, or chloroform may be used, the operation being performed as described in Specifications No. 2612 and 3799, A.D. 1881; or the filaments may be heated electrically while immersed in the liquid chloride &c.; In exhausting the lamps and standardizing the resistance of the filaments, when the vacuum is good hydrocarbon vapour is admitted and a current is passed through until standard resistance is arrived at. The vapour supply is then stopped and the exhaustion is completed while the filaments and lamps are heated. The vapour may be supplied from a vessel forming part of the pump and containing india-rubber, a mixture of benzoate of lime or baryta with quicklime, or a mixture of sodium acetate and soda lime or charcoal saturated with benzene, the vessel being heated when the vapour ie required to be evolved. The resistances of several lamps may be equalized at one operation by arranging them in series and sending a current through them. An automatic cut-off, worked by a bridge galvanometer, may cut off the current from one lamp, when the required resistance is attained, or may divert it to another. A little mercury vapour may be left in the lamps to form a non-volatilizable compound with any oxygen remaining therein.

Arc lamps, electrodes for. Acid is added to a bath of cuprammonia saturated with cellulose, cotton, silk, &c.;, the precipitate obtained being washed and kneaded into a dough-like mass, which is moulded into sticks."
During the last few years in which the writer was employed at the Lamp Business Group of the General Electric Company in East Cleveland, Ohio, historical artifacts were collected from some of the buildings at Nela Park and were housed then in an available room. Among the various historical items collected was a Crookes lamp, which is believed to have been collected by H. D. Burnett, who was superintendent of the Thomson-Houston manufacturing plant in Lynn, Massachusetts from 1886 to 1893. The projected outlines of the small lamp, which has a maximum bulb diameter of 1-1/4 inches, were then traced. The image is shown below. It was covered by patents issued in 1881, 1882 and 1883, and was designed for 83 volts.

One lamp made by William Crookes exists in the historical lamp collection of William J. Hammer; it is numbered 1883-11. The lamp can be viewed if the reader goes to the write-up of the Hammer Collection on this website and clicks on the picture description: "Lamps Developed by Other Workers by 1904." The picture can be enlarged by clicking on the box that will appear when the mouse arrow is moved from outside the picture boundary to the inside (from outside the lower right hand corner boundary); the arrow must be inside the picture boundary to obtain the enlargement box. The Crookes lamp is in the front row, eighth from the right-hand end.

Two of the devices developed by William Crookes that present an opportunity for a person to observe interesting phenomena in nature are the spinthariscope7 and the radiometer8.

I am grateful for a listing of patents granted to William Crookes, as well as copies of their abridgments, which were obtained from Maria Lampert of the British Library, London.

Lamp-Related Patents
  1. GB3860 - Nov 5, 1875 - Radiometers; photometers; thermometers
  2. GB1422 - Mar 31, 1881 - Incandescent lamps; arc lamps
  3. GB2304 - May 25, 1881 - Incandescent lamps
  4. GB2612 - Jun 15, 1881 - Incandescent lamps
  5. GB3799 - Aug 31, 1881 - Incandescent lamps
  6. GB1079 - Mar 6, 1882 - Incandescent lamps
  7. GB4238 - Sep 6, 1882 - Incandescent lamps
  8. GB2185 - Apr 30, 1883 - Incandescent lamps; supporting lamps.

  1. "Electric Illumination", James Dredge, M.F. O'Reilly and H. Vivarez, John Wiley and Sons, New York, 1885, pp.431-432.
  2. "The Life of Sir William Crookes", E.E. Fournier D'Albe, T. Fisher Unwin Ltd., London, 1923.
  3. "A Memoir of Sir Joseph Wilson Swan, F.R.S.", M.E.Swan & K.R.Swan, Ernest Benn Limited, London, 1929, p.61.
  4. "William Crookes", Charles Coulston Gillispie, Dictionary of Scientific Biography, Vol 3, Charles Scribner's Sons, New York, 1980, pp.474-481.
  5. "The History of the Vacuum Pump", E.N. Da C. Andrade, History of Vacuum Science and Technology American Institute of Physics, 1984, pp.77-83; originally published in Adv. Vac. Sci. Tech., Vol.1, 1960, pp.14-20).
  6. "William Crookes and the Quest for Absolute Vacuum in the 1870s", R.K. DeKosky, History of Vacuum Science and Technology American Institute of Physics, 1984, pp.84-101; originally published in Adv. Vac. Sci. Tech., Vol.40, 1983, pp.1-8).
  7. http://www.frognet.net/~ejcov/index50.html/spinthar.html
  8. http://www.frognet.net/~ejcov/index50.html/radiometer.html