Antimony and Other Inorganic Compounds
In many polymers the high concentration of halogenated organic compounds needed to impart flame retardancy adversely affects their physical properties. In practice halogen containing flame retardants are formulated with inorganic compounds that behave synergistically with the halogen. This enables formulators to use less additives without diminishing flame retardance. Indeed in many instances flame retardancy is improved when inorganic halogen synergists are used.
Antimony Trioxide. In 1979 approximately 15 900 metric tons of antimony trioxide (commonly referred to as antimony oxide) was used to impart flame retardance to a variety of plastics. Antimony trioxide is manufactured by oxidizing molten antimony sulfide ore and/or antimony metal in air at 600 800°C. Typical properties for antimony trioxide are listed in Table 1.
Antimony trioxide is a white pigment (qv). Its pigment strength is a function of the average particle size and the particle size distribution. Particle size can be controlled during its manufacture to produce either a high tint or a low tint product. The difference in the particle size and particle size distribution between high tint and low tint antimony trioxide is illustrated in Figure 1. Both grades have the same flame inhibiting efficacy but have different effects on pigmentation and physical properties.
Domestic products and trade names of antimony trioxide are summarized in Table 2.
Special grades are available at higher costs. They include White Star S15 from the Harshaw Chemical Company and ultra fine antimony oxide from PPG Industries.
Antimony Trioxide in Cellulosics. Antimony trioxide can be used as a condensed phase flame retardant in cellulosic materials. In these substrates it reacts endothermically with the hydroxyl groups and forms a variety of products. The endothermic reaction absorbs heat needed to propagate the flame. The products formed are difficult to ignite and shield the underlying cellulose from the flame minimizing pyrolytic and oxidative degradation.
Antimony Pentoxide. Antimony pentoxide is manufactured by the oxidation of antimony trioxide with nitrates or peroxides (5 9). For Sb205 the wt % of antimony is 72.8 and the specific gravity is 3.8.
When the pentoxide is heated above 380°C it disproportionates into antimony tetroxide with the evolution of oxygen.
Commercially antimony pentoxide is primarily available as a stable colloid (Nyacol Inc.) or as a redispersible powder (Nyacol Inc. PPG Industries Inc.). It is significantly more expensive than antimony trioxide and is designed primarily for highly specialized applications. Antimony pentoxide manufacturers suggest fiber and fabric treatment applications as a potential area for its use. The redispersible powder form of antimony pentoxide which is also recommended for plastics contains 88% antimony pentoxide and 12% dispersing agents. Care must be exercised when this product is incorporated into plastic since the dispersing agents can adversely affect the thermal stability and physical properties.
Sodium Antimonate. Sodium antimonate Na2OSb2O5.½H2O is a free flowing white powder made by the oxidation of antimony trioxide in a basic medium. A few of its properties are shown in Table 3.
The pigmenting strength of sodium antimonate is less than antimony trioxide. It is recommended for formulations in which deep tone colors are required. Because it contains 62 wt % antimony somewhat higher concentrations are needed to make it as effective as antimony trioxide which has 83 wt % antimony.
Mixed Metal Antimony Compounds. Recent developments in inorganic flame retardant synergists have centered on mixed products that contain antimony and other metals which reportedly give excellent performance at reduced cost.
Thermoguard CPA (M & T Chemicals Inc.) appears to be as effective as antimony trioxide in most flame retardant applications and has a significantly lower price. Although it contains a lower level of antimony compared to antimony trioxide other metals contained in the product significantly boost its flame retarding properties.
NL Industries has developed a series of antimony silico complexes under the trade name Oncor. These products contain up to 50% antimony trioxide. They are less opacifying than either high or low tint antimony oxide. Generally antimony silico complexes are less effective flame retardants than antimony trioxide. Therefore although the cost per kilogram is less than antimony trioxide the cost effectiveness of the antimony silico complexes can be higher.
Antimony Halogen Mechanisms. Antimony trioxide is used almost exclusively with heat labile halogen compounds. Most of the mechanisms proposed indicate that antimony trioxide is activated by reaction with halogens forming antimony trihalides or antimony oxyhalides.
Antimony trichloride and antimony oxychloride work primarily as flame phase flame retarders. The type of antimony halide formed depends on the concentration of the hydrogen halide and the temperature of the reaction.
In this study a typical aliphatic chlorinated paraffin containing 70 wt % chlorine (Chlorowax Diamond Shamrock) was heated alone at a rate of 20°C/min. A 67% weight loss was noted at 250 360°C (see Fig. 2). The loss is equivalent to 93 wt % of the theoretical stoichiometric quantity of hydrogen chloride.
When an equal weight of antimony trioxide was added to the chlorinated paraffin and the mixture was heated at the same rate a 76% weight loss at 310 400°C was noted (see Fig. 3). If there were no reaction the loss would have been only 37.5% since only half of the mixture was the chlorinated paraffin and antimony trioxide does not volatilize below 656°C. The higher weight loss indicates that some reaction between either the decomposition products of the chlorinated paraffin or the chlorinated paraffin itself and antimony trioxide have taken place. The gas generated by the reaction has been analyzed and identified as antimony trichloride. The weight loss is equivalent to 90% of the theoretical quantity of antimony trichloride that can be formed from the mixture. From this thermal analysis it is apparent that antimony trichloride is the predominate antimony species formed from combinations of antimony trioxide and aliphatic chlorine compounds that generate high concentrations of hydrogen chloride upon thermal degradation.
When a cyclic halogenated organic compound Dechlorane 5 10 (Hooker) that contains 77% chlorine was heated at a rate of 20°C/min 90% of its weight was lost between 280 and 400°C. It does not generate hydrogen chloride directly upon decomposition (see Fig. 4). When equal weights of Dechlorane 5 10 and antimony trioxide were heated at the same rate a different weight loss pattern was noted (see Fig. 5).
Instead of the smooth continuous decomposition pattern observed for either the chlorinated paraffin antimony trioxide mixture or Dechlorane 5 10 itself a two stage decomposition pattern was observed. There was a 45% weight loss between 305 and 410°C and another between 490 and 680°C.
It appears that antimony oxyhalides are the primary antimony compounds formed when organic halogen compounds which do not generate hydrogen chloride directly upon thermal exposure and antimony trioxide are heated together.
Antimony trihalides are the flame retarding species whether they are generated directly from the starting antimony halogen mixture or from antimony oxyhalide. They inhibit combustion by altering the manner and type of decomposition products formed by the plastic and by modifying the reactions in the flame to make them less exothermic. In the condensed phase or molten polymer just beneath the flame antimony trihalide promotes reactions that form carbonaceous chars instead of highly volatile reactive gases. The chars act as heat shields which deflect the heat of the flame and slow down the thermal and oxidative decomposition of the polymer. The chars also form a seal around the polymer preventing potentially flammable gas from escaping and entering the flame.
Once in the flame the antimony trihalides decompose into various antimony oxides and halogen compounds. The decomposition mechanism has not been completely determined.
The antimony oxides formed also participate directly in reactions with the hydrocarbons to give water and molecular hydrogen instead of flame propagating radicals. Since the formation of the nonpropagating molecules is less exothermic than the formation of flame propagating radicals less heat is generated.
Approximately 2700 metric tons of borates was used as flame retarders for poly(vinyl chloride) cellulosics and unsaturated halogenated polyesters in 1979. Zinc borate is by far the most widely used of this class of compounds (see Boron compounds). There are a variety of zinc borates available that vary in zinc boron and water content.
Manufacturers and trade names of commercially available borate flame retardants are shown in Table 4.
Zinc borate is rarely used alone. It acts synergistically with antimony oxide enabling compounders to extend antimony trioxide in some formulations.
Zinc borate is also used with high levels of alumina trihydrate in some halogenated unsaturated polyester resins.
Boric Acid Sodium Borate. Boric acid and sodium borate (borax) are two of the oldest known flame retardants. They are used primarily to flame retard cellulosics such as cotton (qv) and paper (qv). Both products are inexpensive and fairly effective in these applications. Their use is limited to products for which nondurable flame retardancy is acceptable since both are very water soluble.
Boron Mechanism. Boron compounds function as flame retardants in both the flame and condensed phases. Flame phase active boron compounds are generated from combination of borates and halogenated organic compounds. These compounds usually generate boron trihalides which have been used to reduce the flame volatility of air hexane mixtures.
Boric acid and borax are effective condensed phase flame retardants in polyhydroxyl compounds especially in cellulosic fibers. When these compounds are exposed to a flame they melt and form a glasslike coating around the fibers. Prolonged exposure causes the coating to dehydrate generating water which cools the flame and cause it to extinguish. The boron residue also reacts with the hydroxyl groups of the cellulose to generate additional quantities of water and form an inorganic char that is difficult to ignite and burn. The char is an insulator that slows down the rate of polymer degradation and fuel formation.
Boron compounds that also contain other metals are active in both phases. Although zinc borate is not used alone to flame retard PVC it does inhibit flammability in the condensed and flame phases. Upon exposure to the flame the PVC generates hydrogen chloride which can react with the zinc borate to form nonvolatile zinc compounds as well as volatile and nonvolatile boron compounds.
The nonvolatile zinc compounds and boric acid promote char reducing fuel formation and the boron trichloride and water cool and extinguish the flame.
Ammonium Fluoroborate. Ammonium fluoroborate NH4BF4 is another boron containing compound that has some utility as a flame retardant. It can decompose to yield both halogen and boron functionalities to the flame retarding process. Flame retardant plastic formulations recently published suggest that ammonium fluoroborate should be used primarily in combination with antimony trioxide. Manufacturers propose that the following reaction describes functionally what takes place when the two products are exposed to flaming conditions.
The products formed contribute to extinguishing the flame by the mechanisms proposed in proceeding paragraphs.
Approximately 159 000 metric tons of alumina trihydrate (ALTH) was used to flame retard unsaturated polyesters and foam carpet backing in 1979. ALTH is made either from bauxite by the Bayer process from recovered aluminum by the sinter process. Physical properties listed in Table 5 and principal suppliers in Table 6.
Alumina trihydrate is the only aluminum compound of commercial significance as a flame retardant. It functions as a flame retardant in both the condensed and flame phases.
When alumina trihydrate is exposed to temperatures above 250°C it forms water and alumina.
The evolution of water absorbs heat. The water cools the flame and dilutes the flammable gases and oxidant in the flame. The alumina residue an excellent heat conductor increases removal of heat from the flame zone.
Although ALTH is an inexpensive compound it is a comparatively inefficient flame retardant. High add on levels up to four times as much as the plastic itself are needed to impart acceptable flame retardance. It is used alone only in polymers in which large amounts of filler can be tolerated and increased weight (or density) is desired. The major application areas for ALTH are filled thermoset polyesters and styrene butadiene rubber latex rug backing.
Alumina trihydrate is also used as a secondary synergist to improve the flame retardance of polymer systems that already contain antimony trioxide zinc borate or some phosphorus flame retardants.
Molybdenum compounds have been used as flame retardants of cellulosics for many years. Recently they have found some use in other polymers. Molybdenum compounds appear to function as condensed phase flame retarders (32). After ignition of PVC formulations containing molybdenum oxide (MoO3) and antimony oxide 90% of the molybdenum remained in the ash and only 10% of the antimony was found.
Since most of the molybdenum remained in the ash and the formulation did have flame retardant properties molybdenum is probably a condensed phase flame retardant that promotes char. The precise mechanism of action has not been sufficiently defined to warrant further speculations.
Halogenated Flame Retardants
The development and extensive use of synthetic polymers in both old and new types of applications has intensified the concern for combustibility. Although these new polymers are not necessarily more flammable than natural polymers they are more readily used in forms eg foams electrical applications etc that can result in an increased fire control problem.
Along with the development of many synthetic polymer systems during the 1930s and 1940s a significant advance in the science of imparting flame resistance occurred ie the use of halogenated organic materials to impart ignition resistance to these new polymer systems.
In early plastics applications the small size of fabricated articles and the relative scarcity of these articles made fire retardancy a secondary consideration. Advances in plastics technology have led to increasingly large scale applications especially in the construction industry. Since many polymers have fuel values (heats of combustion) comparable to common fuels eg wood oil alcohol etc. it is readily understandable that they contribute to the burning process in a typical fire.
Commercial halogenated products used as flame retardants for plastics currently in use are mainly compounds containing high (50 85 wt %) levels of either chlorine or bromine ie decabromodiphenyl oxide chlorendic acid tetrabromophthalic anhydride etc. These materials fall into two distinct types additives and reactives. The additives have the advantage of being readily added to a polymer by mechanical means with a minimum of reformulation being required. The reactives on the other hand require the development of essentially new polymer systems.
Only massive polymer forms are considered though the materials and concepts discussed are almost similarly applicable to fibers fabrics coatings and elastomers. Halogenated phosphorus compounds are included under Flame retardants phosphorus compounds.
Principles of Developing Flame Retardant Polymers
Any discussion of the principles of developing flame retardant polymer systems must acknowledge the chaotic situation that exists at present. This situation has arisen for a variety of reasons technical economic legal and semantic.
The semantic problem is the worst in that it is at the root of most of the other problems and is caused by the fact that the term fire or flame retardant may be perceived in a variety of ways depending upon the user s viewpoint. The term as defined above means simply that some change has been made in a polymer system so that it will pass one or more of at least a hundred different flammability tests. These tests are normally designed to minimize but not eliminate the fire risk associated with the use of a polymer in some specific use or product. As a consequence a modification of a polymer that makes it suitable for one use does not necessarily make it suitable for others. There is no single fire retardant chemical or method that is applicable to all polymer systems or even to all uses of a single polymer.
It is therefore necessary that early in the development of a flame retardant polymer system the question Why? is answered before much effort is put into answering the question How? .
It is not unusual to see many compounds proposed as flame retardant chemicals that are clearly unusable in any practical sense but that allow a polymer system to pass a specific flammability test. A polymer system can be easily modified so that it can be called flame retardant by some test. It is difficult however to do so and keep a polymer system that is low cost environmentally and physiologically acceptable and also mechanically and esthetically not too dissimilar from nonfire retardant counterparts.
One of the most common approaches used to modify the burning properties of polymers at the present time is by incorporation of halogen into the polymer matrix either directly or through the use of halogenated additives. The usual rationale for the use of the halogens as flame retardants is based on the theory that they function in the gas phase as radical traps. It is generally agreed that the combustion of gaseous fuels is a high temperature process which proceeds via a free radical mechanism.
In the radical trap theory of flame inhibition it is thought that equations 6 10 effectively compete with equations 2 5 for those radical species that are critical for flame propagation ie .OH and .0. thereby slowing the rate of energy production and resulting in the extinction of the flame. Hydrogen fluoride does not significantly enter into the flame chemistry thus fully fluorinated compounds are generally considered to be ineffective as flame retardant agents. The radical trap theory of flame inhibition although attractive in that it can be adapted to any situation tends to lead to the belief that the simple inclusion of small amounts of halogen into a polymer system will render the system flame retardant.
A recent physical theory of flame suppression by the halogens although conceding that the halogens enter into flame chemistry suggests that this participationper se cannot be the primary mechanism by which the halogens function. Rather it is postulated that the halogens act by altering the physical properties ie the density and mass heat capacity of the gaseous fuel oxidant mixture so that flame propagation is effectively prevented. The physical theory is primarily based on the observations that any gaseous mixture of fuel and halogenated agent generally propagates flame when mixed with air as long as the mass fraction of halogen in the mixture is less than ca 0.7 and the relative effectiveness of the halogens is directly proportional to their atomic weights ie F CI Br I = 1.0 1.9 4.2 6.7. The halogenated agents probably act by the same basic mechanisms as the inert gases ie CO2 N2 etc and their suppressant effects are additive to those of the inert gases.
Mo is the mass fraction of oxygen in the combustion zone Hc is the net heat of combustion of the sample (J/g) r is the stoichiometric mass oxygen/fuel ratio Cp is the specific heat of the gases in the combustion zone Ts is the surface temperature of the sample (ºC) Ta is the ambient temperature (ºC) and HG is the apparent heat of gasification (J/g). The B number contains the fundamental properties of the polymeric materials. Thus the mass burning rate or burning intensity can be related to the fundamental properties of the material.
Where Mi is the mass fraction of the inert components of the mixture and ma mN mf and mo are the weights of agent nitrogen fuel and oxygen respectively. If all of the terms in the B number remain constant an increase in the mass of inert gas in the combustion zone (ma O) results in a lower oxygen mass fraction mo a lower B number and a corresponding reduction in the polymer burning rate.
When applied to liquid fuels the Spalding B number in its simplest form can be visualized as the ratio of the heat of combustion and the heat of vaporization (Hc/Hv). Table 1 shows the significance of this ratio applied to several halogen containing fuels. In Table 1 the flash and fire points are expressed both in °C as normally reported and as the weight of compound present in the gas phase over the surface of the liquid at this temperature (mg/L). The introduction of halogen has a lesser effect upon Hv per milligram of compound evaporated. The ratio Hc/ Hv decreases with added halogen indicating that less energy is available from the flame for gasification and in order to keep the flame burning additional heat from some outside source is required. Hv is the amount of heat required to vaporize the weight of fuel (latent heat of vaporization) present in the gas phase at the appropriate flash and fire points after the fuels have been raised to these temperatures by the outside source. Note the large increase in mass that must be vaporized in order to obtain sustained burning in the case of bromobenzene at least 100 times the mass that must be vaporized in the case of benzene itself.
The physical theory apparently accounts for the effects seen when halogenated agents are used as flame retardants. In view of the fact that the halogen content of a typical plastic is generally ca 1 30 wt % it is obvious that if the typical polymer were totally vaporized the gases given off would be quite capable of flame propagation.
In order to visualize the role of halogen it is necessary to examine the heat balance that occurs at the surface of the polymer. Figure 1 shows a schematic of this balance (10). Heat received by the polymer surface may arise either as a heat flux from the flame (T) or as an externally applied heat flux (E) derived from another source. Heat is lost either as the heat required for gasification (G) of the polymer or as heat lost L through radiation conduction convection dripping etc. T and G are agent dependent whereas E is obviously agent independent except in char forming systems. L may be agent dependent if the agent acts by increasing the drip rate of the burning polymer. Halogenated agents affect the heat balance through T G and L. Although phosphorus may act in the gas phase it appears to be the most important element affecting G and E through char formation.
Qualitatively the burning process involves heating of the substrate to a temperature high enough to drive off flammable vapors. When the rate of vapor evolution becomes high enough to generate a flammable mixture the mixture ignites. If the rate of vapor or gas evolution becomes sufficiently high the heat produced by the combustion process may return enough heat to the substrate so that the evolution offuel becomes self sustaining.
When a flame retardant that acts in the vapor phase is added to the system part of the vapor that distills from the polymer does not contribute to the heat of combustion but results only in a reduction in the mass fractions of the oxygen and fuel in the combustion zone. Hence there is an increase in the total mass of material that must be vaporized per unit time in order to keep the fire burning. A corresponding increase in the amount of energy must be added to the system from an external heat source (E Figure 1) in order to vaporize the extra material.
Both dripping and char formation interfere with the energy feedback cycle (T and E) and consequently cause an increase in the intensity of the external heat flux required to balance the energy fuel cycle.
Where the flame is actively spreading over the surface of a material the elemental composition of the vapor being evolved ahead of the moving flame is not necessarily the same as the elemental composition of the polymer. The composition of the vapors may vary considerably between the temperature at which the material first begins to evolve vapors and the temperature at which the rate of evolution supports the flame. With this type of dynamic burning condition changes in the substrate and the structure of the agent are more important than they are under steady state conditions.
There are five fundamental methods used to fire retard both natural and synthetic polymer systems. They are
Raise the decomposition temperature of the polymer. This is generally accomplished by increasing the cross linking density of the polymer as with ladder polymers (increase G).
Reduce the fuel content of the system. This approach generally involves halogenating the polymer backbone adding halogenated additives adding inert fillers or by resorting to inorganic systems (increase G decrease T).
Induce polymer flow by selective chain scission. This approach is generally applicable to thermoplastic polymer systems where interrupting the polymer backbone results in reduction of the viscosity of the polymer and promotes dripping (increase L).
Induce selective decomposition pathways. This method is most applicable to cellulosics where the introduction of phosphorus compounds generates phosphorus acids which catalyze the loss of water and the retention of the carbon as char (increase G decrease T).
Mechanical means include (1) bonding a nonflammable skin on the polymer (2) covering the polymer with an intumescent coating (3) design of the system and (4) the use of sprinklers (decrease E).
Antimony Halogen Synergism. Antimony oxide a commonly employed fire retardant adjunct for halogen containing polymer systems is usually employed as a means of reducing the halogen levels required to obtain a given degree of flame retardancy with the polymer system. This reduction is often desirable since the required halogen content for the system may be so high that it affects the physical properties of the system. In other cases the antimony oxide is used simply to give a more cost effective system.
Antimony halogen systems have been widely studied in attempts to explain the apparent synergistic effects obtained with this combination of elements. No completely satisfactory theory is available as yet but it is generally agreed that the active agents antimony trihalides or antimony oxyhalides act principally in the gas phase (12 13). As with the halogens it is generally postulated that the antimony halides act as radical traps.
Small scale tests show that the optimum halogen (CI Br)/antimony atom ratio in most systems is 3/1 (14) corresponding to the atom ratio found in the antimony trihalides ie SbCI3 SbBr3. On the usual weight basis this corresponds to a ratio of ca 0.9/1 for the chlorine antimony system and ca 2/1 for the bromine antimony systems.
Although the antimony halides appear to act principally in the gas phase some effect on the condensed phase chemistry cannot be ruled out. Antimony halogenflameretardant compositions usually produce a carbonaceous residue even in polymers such as polypropylene which produces none in the absence of fire retardants. The production of the carbonaceous residue probably results from the antimony trihalides strong Lewis acid catalysts which are capable of promoting the dehydrohalogenation of organic halides and coupling and rearrangement reactions in organic systems.
Phosphorus Halogen Systems. A large number of phosphorus containing compounds have been used in halogen containing polymer systems as a means of improving their ignition resistance. In many of these cases both the phosphorus and the halogen reside in the same molecule although there is little if any evidence to indicate that having both elements in the same molecule has any particular advantage. Because there is no fixed optimum phosphorus halogen ratio in contrast to the antimonyhalogen system it is frequently easier to optimize the ignition resistance when the phosphorus and halogen are adjusted separately.
Generally phosphorus appears to act as an acid precursor in the solid phase to induce selective decomposition pathways that result in a reduction in the rate of fuel formation and an increase in charring. This mode of action is most applicable to cellulosics but may also be important in other oxygen or nitrogen containing polymers such as polyesters (qv) polyamides (qv) and polyethers (qv).
In polymers such as polyolefins and polystyrene the formation of acids has little affect on the mode of polymer decomposition and much of the phosphorus may be volatilized in some cases as much as 50 99%. Even in these cases some of the phosphorus may end up as polyphosphoric acids which serve to protect the substrate from the heat produced by the burning gases. The phosphorus that volatilizes will show some beneficial flame retarding effects in that it has gas phase flame suppressing activity similar to the halogens.
The main fire retardants currently used in plastics and textiles fall into several distinct classes (1) alumina trihydrate (2) halogenated compounds usually used in combination with antimony oxide (3) borax and boric acid and (4) the phosphorus phosphorus nitrogen and phosphorus halogen compounds.
Mechanism of Action of Phosphorus Flame Retardants
The overview article presents a broad discussion of flame retardant mechanisms. The following discussion deals specifically with phosphorus flame retardants.
Condensed Phase Mechanisms. The mode of action of phosphorus based flame retardants in cellulose has been more extensively studied and is better understood than in most other polymer systems. Two alternative routes of cellulose (qv) pyrolysis are known to occur one route proceeds first to a tarry depolymerization product called levoglucosan (1) which decomposes to volatile combustible fragments the other route (catalyzed by acids) leads primarily to water and difficultly combustible char.
Although mineral acids in general catalyze the desired water and char forming pyrolysis route phosphoric acid is particularly advantageous because of its low volatility. Also when strongly heated phosphoric acid yields polyphosphoric acid which is even more effective in catalyzing the desired dehydration reaction. The flame retardant action of phosphorus compounds in cellulose is believed to proceed by way of initial phosphorylation of the cellulose. The phosphorylated cellulose then breaks down to water phosphoric acid and an unsaturated cellulose analogue eventually char by repetition of these steps. Certain nitrogenous compounds such as melamines guanidines ureas and other amides appear to catalyze the cellulose phosphate forming steps and are found to enhance or synergize the flame retardant action of phosphorus on cellulose.
In poly (ethylene terephthalate) and poly (methyl methacrylate) the mechanism of action of phosphorus based flame retardants has been shown to involve both a similar decrease in the amount of combustible volatiles and a similar increase in the amount of residue (aromatic residues and char). The char thus formed also acts as a physical barrier to heat and gases.
In rigid polyurethane foams the action of phosphorus flame retardants also appears to involve char enhancement.
The physical character of the char from rigid urethane foams was found to be affected by the retardant. The presence of a phosphorus containing flame retardant caused rigid urethane foam to produce a more coherent char possibly serving as a physical barrier to the combustion process. There is evidence that a substantial fraction of the phosphorus may be retained in the char.
In polymers such as polystyrene that do not readily undergo charring phosphorus based flame retardants tend to be less effective and such polymers are usually flame retarded by antimony halogen combinations. However even in noncharring polymers phosphorus additives exhibit some activity that suggests at least one other mechanism of action. It has been proposed and some evidence adduced that phosphorus compounds may produce a barrier layer of polyphosphoric acid on the burning polymer.
There is evidence that phosphorus containing additives can act in some cases by catalyzing thermal breakdown of the polymer melt reducing its viscosity and favoring the flow or drip of molten polymer from the combustion zone. In polystyrene tris(2 3 dibromopropyl) phosphate acts at least in part by this mechanism.
Several commercial polyester fabrics are flame retarded with low levels of phosphorus additives or reactives which cause them to melt and drip more readily than fabrics without the flame retardant. This mechanism can be counteracted or completely defeated by the presence of nonthermoplastic fibers such as cotton which can serve as wicks or by silicone oils which can form pyrolysis products capable of impeding melt flow.
Vapor Phase Mechanisms. In addition to the condensed phase mechanisms discussed above phosphorus flame retardants can exert vapor phase flame retardant action. It has been demonstrated that trimethyl phosphate retards the velocity of a methane oxygen flame with about the same molar efficiency as SbCl3. Both physical and chemical vapor phase mechanisms have been proposed for the flame retardant action of certain phosphorus compounds. Since tris(dibromopropyl) phosphate was found not to change the activation energy of thermo oxidative degradation of polypropylene although it raised the oxygen index a vaporphase physical shielding action was postulated. Possibly this action may be produced by bromine containing pyrolysis products rather than by the phosphate itself.
Triphenylphosphine oxide and triphenyl phosphate as model phosphorus flame retardants were shown by mass spectroscopy to break down in a flame to give small molecular species such as PO HPO2 PO2 and P2. The rate controlling hydrogen atom concentration in the flame was shown spectroscopically to be reduced when these phosphorus species were present. These data indicate the existence of a vapor phase mechanism however the stable volatile compounds used in this study are not typical of many of the phosphorus based flame retardants used commercially. Physical or chemical vapor phase mechanisms may be reasonably hypothesized in cases where a phosphorus flame retardant is found to be effective in a noncharring polymer and especially where the flame retardant or phosphorus containing breakdown products are capable of being vaporized at the temperature of the pyrolyzing surface. In General Electric s engineering thermoplastic Noryl which consists of a blend of a charrable poly (phenylene oxide) and a noncharrable polystyrene experimental evidence indicates that effective flame retardants such as triphenyl phosphate act in the vapor phase to suppress the flammability of the polystyrene pyrolysis products.
A comparison of a variety of phosphorus additives at equivalent phosphorus loadings was made in poly (methyl methacrylate) which can be retarded by condensed phase action but should also be subject to vapor phase inhibition because it depolymerizes to monomer. Poor flame retardancy was found with trimethylphosphine oxide a volatile stable species whereas a much larger oxygen index elevation was observed with phosphoric acid this result suggests that the condensed phase mechanism is the more efficient one in poly (methyl methacrylate).
Poly (ethylene terephthalate) exhibits a higher oxygen index with 5 wt % phosphorus incorporated in the backbone of the polymer as phenylphosphinyl groups as contrasted to 5 wt % phosphorus incorporated as a relatively volatile additive triphenylphosphine oxide. This result suggests that a condensed phase mechanism is more effective than a vapor phase mechanism in this polymer.
The question as to whether a flame retardant operates mainly by a condensedphase mechanism or mainly by a vapor phase mechanism is especially complicated in the case of the haloalkyl phosphorus esters. A number of these compounds upon thermal degradation release volatile halogenated hydrocarbons which are plausible flame inhibitors. At the same time their phosphorus content remains as relatively nonvolatile phosphorus acids which are plausible condensed phase flame retardants. There is no evidence for the formation of phosphorus halides.
Interactions With Other Flame Retardants. Some claims have been made for a phosphorus halogen synergism but unlike the firmly established antimony halogen synergism phosphorus halogen interactions are often merely additive and in some instances slightly less than additive. Cases of phosphorus halogen synergism (ie activity greater than that predicted by some additivity model) usually do not hold up to careful analysis and some supposed cases are artifacts of nonlinear response concentration relationships. Nevertheless combinations of phosphorus and halogen in separate compounds or in a single compound are often quite useful even if not truly synergistic.
Antagonism between antimony oxide and phosphorus flame retardants has been reported in several polymer systems and has been explained on the basis of phosphorus interfering with the formation or vaporization of antimony halide. This phenomenon is also not universal and some useful commercial PVC formulations have been described for antimony oxide and triaryl phosphates.
An interesting case of synergism has been described involving a bisphosphine oxide American Cyanamid s RF 699 and ammonium polyphosphate.
Phosphorus Based Flame Retardants in Commercial Use
Since the original report of ammonium phosphate as a flame retardant by Gay Lussac in 1821 and the commercial introduction of tricresyl phosphate as a flame retardant plasticizer for cellulosics early in the present century many thousands of phosphorus compounds have been described as having flame retardant utility. A broad sampling of these is covered in ref. 33. The more specialized topics of phosphorus monomers and polymers containing built in phosphorus have been reviewed. This article is confined to the much more limited groups of compounds that found commercial or semi commercial use.
Inorganic Phosphorus Compounds. Red Phosphorus. This allotropic form of phosphorus is relatively nontoxic and unlike white phosphorus is not spontaneously flammable (although easily ignited). It is a polymeric form of phosphorus with thermal stability up to ca 450°C. In finely divided form it has been found to be outstandingly effective as a flame retardant additive. In Europe it has found commercial use in molded nylon electrical parts. Handling hazards such as flammability odor partial reversion to toxic white phosphorus and the imparting of color have deterred broader usage. A product Exolit 505 available from Hoechst (FRG) consists of red phosphorus treated with caprolactam and is reported to be safer than the untreated material (38). Related products are marketed in Japan.
Ammonium Phosphates. These salts were recommended for treating theater curtains in 1821. Their use in forest fire control is well established. Monoammonium phosphate and diammonium phosphate or mixtures of the two which are more water soluble and nearly neutral are still used in large amounts for nondurable flame retarding of paper textiles disposable nonwoven cellulosic fabrics and wood products. Their advantage is high efficacy and low cost. Ammonium phosphate finishes are not resistant to laundering or even to leaching by water but they are resistant to organic solvents such as dry cleaning solvents. One important advantage of ammonium phosphates as flame retardants and phosphorus flame retardants in general over borax (also used for nondurable cellulosic flame retardants) is their effectiveness in preventing afterglow.
The crystalline nature of ammonium phosphates may produce a gritty texture on the surface of some substrates. This characteristic is lessened by commercial ammonium phosphate formulations containing softening and penetrating agents.
Self cross linking acrylic latexes have been formulated with diammonium phosphate and organic phosphates to obtain flame retardant textile backcoatings and nonwoven binders with a small but useful degree of durability to laundering and dry cleaning.
Insoluble ammonium polyphosphate. When ammonium phosphates are heated with urea or by themselves under ammonia pressure relatively water insoluble ammonium polyphosphate (Phoschek P/30 Monsanto) is produced. These products are long chains having repeating units of the structure OP(O)(ONH4) . This product a finely divided solid is a principal ingredient of intumescent paints and mastics. In such formulations ammonium polyphosphate is considered to function as a catalyst. Thus when the intumescent coating is exposed to high temperature the ammonium polyphosphate yields a phosphorus acid which then interacts with an organic component such as dipentaerythritol to form a carbonaceous char. A blowing (gas generating) agent such as melamine or chlorowax is also present to impart a foamed characteristic to the char thus forming a fire resistant insulating barrier to protect the substrate. In addition the intumescent formulations typically contain resinous binders pigments and other fillers. Mastics are related but generally more viscous formulations intended to be applied in thick layers to girders trusses and decking they generally contain mineral fibers to increase their coherence.
Ammonia/P2O5 products. The reaction of ammonia gas with phosphorus pentoxide at high temperature yields an amorphous colorless solid slowly soluble in water to form a nearly neutral solution. The product consists of a mixture of ammonium salts of metaphosphorimidic acid. Analysis shows about two ammonium nitrogen atoms and one imide nitrogen atom for every two phosphorus atoms. Stauffer s Victamide is known to be a complex mixture but a typical component is believed to have structure (2).
Victamide as an aqueous solution can be applied to paper cotton cloth cotton batting and nonwovens. When dry it produces a smoother surface texture than that produced by the crystalline ammonium phosphates. Proprietary formulations have been developed affording some degree of water resistance presumably the Victamide acts therein as a phosphorylating agent. Ammoniation of Victamide by concentrated ammonia produces a product which when applied to a cellulosic substrate and heated yields a semidurable flame retardant finish that withstands several aqueous washes.
Phosphoric Acid Based Systems for Cellulosics. Semidurable flame retardant treatments for cotton can be attained by phosphorylation of cellulose. This was originally accomplished by heating cotton or paper with phosphoric acid in the presence of basic compounds such as urea at ca 145 180°C. Commercial formulations have been developed utilizing as coreactants either phosphoric acid and cyanamide or phosphoric acid and a dicyandiamide formaldehyde resin. The nitrogenous component catalyzes the phosphorylation of cellulose retards the acid degradation of the cellulose and synergizes the flame retardant action of the phosphorus. A fair degree of durability to laundering is achieved by such treatments. Typically several launderings can be tolerated and dry cleaning resistance is good. A substantial part of the decline in flame retardancy during laundering is caused by ion exchange of the protons of the phosphoric acid groups by sodium calcium and magnesium cations that suppress the flame retardant effectiveness of the phosphorus groups. Such finishes also have limited utility because of fabric damage during cure although applications have been made on draperies nonwoven fabrics and paper products.
Commercial formulations have been developed based on phosphoric acid ureaformaldehyde resins and dicyandiamide as leach resistant clear flame retardant coatings for wood.
Organic Phosphorus Flame Retardants Additive Types. Alkyl Acid Phosphates. The lower alkyl acid phosphates have found some limited use as additive flame retardants in cast thermoplastics and polyester resins. In cast poly(methyl methacrylate) methyl and haloalkyl acid phosphates are effective in combination with halogen containing additives.
Trialkyl Phosphates. Triethyl phosphate is a colorless liquid boiling at 209 218°C and containing 17 wt % phosphorus. It is manufactured from diethyl ether and phosphorus pentoxide via a metaphosphate intermediate. Triethyl phosphate has been used commercially as an additive for polyester laminates and in cellulosics. In polyester resins it functions as a viscosity depressant and as a flame retardant. The viscosity depressant effect of triethyl phosphate in polyester resin permits high loadings of alumina trihydrate a fire retardant smoke suppressant filler. Triethyl phosphate has also been employed as a flame resistant plasticizer in cellulose acetate. Because of its water solubility the use of triethyl phosphate is limited to situations where weathering resistance is unimportant. The halogenated alkyl phosphates are generally used for applications where lower volatility and greater resistance to leaching are required.
Trioctyl phosphate has been employed as a specialty flame retardant plasticizer for vinyl compositions where low temperature flexibility is critical eg in military tarpaulins. It can be included in blends with general purpose plasticizers such as phthalate esters to improve low temperature flexibility.
Dimethyl Methylphosphonate. Dimethyl methylphosphonate (DMMP) is made by molecular rearrangement of trimethyl phosphite. It contains 25 wt % phosphorus (near the maximum possible for a phosphorus ester) and it is therefore highly efficient on a weight basis as a flame retardant. DMMP is a low viscosity colorless liquid bp 185°C. Because of its volatility it has been useful mainly in thermoset systems. DMMP is an efficient viscosity depressant in polyester resins and epoxy resins. As a flame retardant it has somewhat greater efficiency than triethyl phosphate which is used in similar systems. DMMP is used commercially in mineral filled and glass reinforced polyester compounds where its viscosity depressant effect permits use of higher filler loadings. The use of alumina trihydrate as filler and DMMP in the dual role of viscosity depressant and flame retardant affords reinforced polyester resin formulations with low flame spread suitable for bathtubs and shower stalls. DMMP has been used commercially for boosting the phosphorus content of flame retardants used in rigid foams. DMMP is also used as a chemical intermediate for the manufacture of several other flame retardants.
Urea Formaldehyde Resins
Although the urea and melamine resin reactions have certain similarities they also have definite differences therefore it will be better to describe them separately. One mole of urea may be reacted with 1 or 2 moles of formaldehyde to produce different products. The reactions may be carried out under either acidic or basic conditions and again different products will be obtained. In addition the dimethylol urea may be produced from urea and formaldehyde and etherified with butanol separately or the etherification may be carried out simultaneously with the condensation by reacting the urea formaldehyde and butanol together. Figure 1 illustrates the condensation reactions between 1 mole of urea and both 1 and 2 moles of formaldehyde under acidic and basic conditions.
Under acidic conditions insoluble compounds are formed which cannot be used for coating resins. Under basic conditions the monoor dimethylol ureas are produced which can be used as intermediate products for coating resins and other purposes. The dimethylol urea is known commercially as DMU and is available as a white solid containing 88 90% DMU and 10 12% water. It is soluble in water and alcohol but it polymerizes slowly at room temperature to the insoluble stage.
The usual source of formaldehyde is formalin which is a 37% solution of formaldehyde in water. The urea is a white crystalline solid having a melting point of 133°C and soluble in water to the extent of 80 gm/100 ml it has a molecular weight of 60. For the production of DMU the correct amounts of urea and formalin are adjusted to a pH of 7 8.5 and reacted at about 50°0. Sodium hydroxide is the usual basic catalyst but others may be used including various amines. When the reaction is complete the mass is concentrated under vacuum to the desired solid content and may be tray dried spray dried or crystallized. The DMU is used in some of the non coating applications referred to earlier since it may be polymerized to the insoluble stage by heating. The structure of the insoluble polymer has not been proved but there can be no doubt that it is highly complex. It is probable that the structure contains cross linked linear polymers and six membered rings as indicated in Fig. 2 A and B respectively.
In order to form the six membered ring structure (B Fig. 2) the two –NH2 of urea may react differently with formaldehyde. One may react as a primary amine to form the Schiff s base followed by trimerization to the ring structure. The other may then react as an amide followed by elimination of water and formation of methylene linkages connecting the ring structures as indicated in Fig. 2 B. Until more positive evidence is obtained regarding the structure of these insoluble polymers the above theories afford some idea of their possibilities and their complexity.
Alkylation or Etherification. In order to change the DMU from a water soluble material to an organic solvent soluble material it must be made less polar. This is accomplished by alkylation with various alcohols. Obviously the lower alcohols with very short carbon chains or small non polar groups are less effective than the higher alcohols with longer carbon chains. For example the methoxy methylol ureas are water soluble the ethoxy products are soluble in ethanol but good solubility in organic solvents is not obtained until butyl alcohol is used. It will be shown later that better solubility and compatibility with other resins are obtained if the higher alcohols are used such as capryl or octyl. However these are more expensive and they retard the curing rate of the resin.
Butylated Urea Resin. A typical urea formaldehyde resin suitable for use in baking coatings may be prepared by dispersing the DMU in butanol which has been slightly acidified. The dispersion is heated and both etherification and polymerization reactions occur. It is essential that sufficient etherification take place before excessive polymerization occurs so that the product will have good solubility and stability. Conversely if a high degree of etherification occurs and relatively low polymerization the resin will have low viscosity and will be slower curing. These factors are controlled by the amount and type of acidic catalyst the temperature and the ratio of the components. A variety of acids may be used including phosphoric formic oxalic and phthalic. In general the ratio of combined butanol in the final resin is from 0.5 to 1.0 moles per mole of DMU but of course a considerable excess of butanol is used during the resin manufacture.
The water eliminated in the etherification and polymerization reactions together with any water with the original DMU is removed either by straight azeotropic distillation or by a continuous decantation procedure. When the desired degree of etherification and polymerization is reached as indicated by solubility and compatibility tests the resin is neutralized and concentrated. For a resin solution which is marketed as 50% resin 30% butanol and 20% xylol the original butanol solution would need to be concentrated to 62.5% resin and 37.5% butanol. When 100 parts of this solution are thinned with 25 parts of xylol the resulting product would meet the requirements indicated above. Every effort is made to remove as much water and free formaldehyde as possible because these detract from stability curing speed and gloss in the finished enamel. In general the final resin solution does not contain more than 0.5% water and somewhat less of free formaldehyde. Resins of this type may be prepared from the original ingredients without first preparing the DMU as an intermediate. In such cases the ingredients are reacted first under alkaline conditions to permit the necessary amount of condensation then finished under slightly acidic conditions as indicated above.
The simplified formulas for a butylated urea formaldehyde resin are shown in Fig. 3. This resin is based on a mole ratio of 1 mole urea 2 moles formaldehyde and 1 mole butanol.
The partially polymerized product in Fig. 3 is a highly simplified and idealized representation. The actual product would be much more complex and cross linked and would probably contain ring structures as indicated in Fig. 2. However the diagram will serve to illustrate the effect of the type of alcohol and the degree of etherification on the solubility and rate of cure of the resin.
Solubility and Compatibility. It should be apparent that the 4carbon chain alkyl group in the butoxymethylol urea serves three purposes (1) it decreases the amount of cross linking (2) it confers hydrocarbon solubility on the resin (3) it increases compatibility with alkyds and other resins. . Decreasing the possible cross linkages retards the curing rate and hydrocarbon solubility permits the use of xylol to replace part of the more expensive butanol. In the manufacture of the resin isobutanol may be used but the secondary and tertiary butanols react too slowly. If the ratio of butanol were reduced from the 1 mole shown in Fig. 3 to 2/3 mole there would be fewer butoxyl groups in the resin. This would reduce the hydrocarbon solubility and would provide another point for cross linking in the trimer illustrated. Higher ratios than 1 mole of butanol would increase the solubility but these are seldom used since they would retard the curing rate and amount of polymerization excessively.
It should be apparent that increasing the carbon chain from a 4 to a 10 or 12 carbon chain would provide another method for increasing hydrocarbon solubility without reducing the number of possible cross linkages. This may be done by using capryl octyl or other alcohols instead of butanol. However these longer chain alcohols are not good solvents for the intermediate DMU. Therefore the DMU polymerizes excessively before any appreciable amount of etherification takes place and a heterogeneous product is obtained which is not suitable for coating resins. However the higher alcohols may be incorporated into the resin by the transetherification procedure. The resin is prepared first with one of the lower alcohols such as methanol and the methylated methylol urea reacted with capryl or octyl alcohol. The transetherification takes place because the liberated methanol may be removed by distillation at a temperature low enough not to affect the higher boiling alcohol.
Mineral Spirits Tolerance. Amino resins are used frequently with the medium oil length alkyd resins which are thinned with mineral spirits instead of xylol. Since mineral spirits is not as strong a solvent as xylol it will be necessary for the amino resins to have better hydrocarbon solubility. One method for accomplishing this is the use of the higher alcohols referred to above. The long carbon chain on the resin makes it much more non polar and therefore more soluble in aliphatic hydrocarbons. The degree of solubility of the resin is referred to as its mineral spirits tolerance. This is measured by adding mineral spirits to the resin solution slowly until turbidity develops.
The mineral spirits tolerance is usually expressed as the pounds of mineral spirits tolerated by 100 lb of resin solution before turbidity develops. The mineral spirits tolerance may also be increased by (a) increasing the ratio of formaldehyde to urea (b) reducing the degree of polymerization of the resin (c) increasing the amount of alkylation. It will be evident that all these methods tend to reduce the curing rate of the resin which means a longer baking time for the finish in order to obtain the same hardness.
The variation in type of alcohol was the only composition variable considered in the preceding discussion of urea resins. However both the amino and the aldehyde components may also be changed. The only other amino resin which has achieved commercial importance to date is the melamine formaldehyde type described in the following section. . However mention should be made of thiourea since it was one of the early amino compounds investigated for use in resins.
The formula shows that the oxygen of urea has been replaced with sulfur. Resins made with thiourea have slightly better water and alkali resistance than comparable urea resins but they are not as pale in color are somewhat odorous and are inferior in exterior durability.
Formaldehyde is the most useful aldehyde for amino resins utilized in surface coatings. An aldehyde with more carbons such as acetaldehyde may be expected to increase the solubility of the resin but the curing rate color retention and film properties have been reported to be inferior (8). Parker (8) also points out that mixtures of formaldehyde and acetaldehyde are impractical because very little of the higher aldehyde is combined under such conditions. It can be removed quite readily by distillation of the resinous material. Furfural has not been used extensively in coating resins to date but it is employed in amino resins for adhesives and molding compounds.
In 1834 Justus von Liebig (9) produced a new chemical which he believed was the amine of melam and which he called melamine. Subsequent investigations have shown that his analysis was not entirely correct but the chemical has retained its original name. It is a member of the class of compounds known as triazines and may be designated 2 4 6 triamino 1 3 5 triazine. It may also be considered a trimer of cyanamide. It may be prepared from dicyandiamide by heating under pressure in the presence of a diluent such as alcohol or ammonia. The relation of melamine to cyanamide and dicyandiamide is shown in Fig. 4.
The chemistry historical development and methods of production of melamine are given in considerable detail by McClellan (11). Hughes (12) studied crystalline melamine and found it to be a resonance hybrid with the position of the atoms as shown in Fig. 5 (a). Ostrogovich (13) suggested both amino and imino structures in view of the possibility of tautomerism Fig. 5 (b). The amino form is generally used in the discussion of melamine in coating resins because its highmelting point and heat stability suggest this benzenoid structure. Melamine is a white crystalline powder with a melting point of 354°C. It has a molecular weight of 126 and specific gravity of 1.57 at 25°C. Its solubility in water has been reported by Chapman (14) to be 0.5% at 25°C 1.0% at 50°C 2.5% at 75°C and. 4.0% at 90ºC.
Terpene Solvents. The terpene solvents are the oldest in use by the paint industry and are obtained from pine trees. They have been replaced by lower cost aliphatic hydrocarbon solvents in many coatings. Their chemical properties make them quite valuable as raw materials for synthetic resins and other compounds. The chemical structures of important constituents of terpene solvents are shown in Fig. 1. Turpentine is the most widely used terpene solvent its principal use being in house paints and in some varnishes. The production of gum turpentine from the exudation of the pine tree is described. The production of wood turpentine dipentene and pine oil from solvent extraction of pine stumps followed by steam distillation is described.
Gum turpentine contains 60 65% a pinene and 30 35% pinene. Wood turpentine contains about 80% a pinene the remainder being dipentene terpinene and terpene alcohols. Dipentene has a higher boiling point than turpentine and excellent solvent properties. It is also used to retard the skinning of varnishes synthetic resins and enamels. The heavy fractions obtained from the production of wood turpentine are known as pine oil. These fractions consist of terpene tertiary and secondary alcohols plus varying percentages of highboiling terpene hydrocarbons. Small percentages of phenol ethers and ketones are also present. The polar non polar structure of pine oil makes it suitable for a wide range of uses. It has excellent solvent properties improves the flow of enamels retards skinning is an antifoaming agent and has some bactericidal action. A typical group of terpene alcohol solvents is given in Table 1.
p Cymene and p menthane are obtained from the catalytic disproportionation of dipentene. As a result they have a higher degree of purity and are used as chemical raw materials as well as solvents.
p Menthane is a saturated terpene and therefore not susceptible to oxidation like the unsaturated terpenes.
The specific uses for terpene solvents in varnishes resins and coatings are given in other sections of this book and in Volume II. The physical characteristics of a typical group of commercial terpene solvents are given in Table 2.
Hydrocarbon Solvents. The petroleum and coal tar hydrocarbon solvents are used extensively because of their low cost good solvent power for oils and resins and effectiveness as diluents for nitrocellulose lacquers. The petroleum solvents are the lighter fractions obtained by the distillation and fractionation of the crude oil. They include the original turpentine substitutes Varnish Makers and Painters Naphtha (VM & P Naphtha) and mineral spirits. Vast improvements have been made in large scale fractionation apparatus with the result that today many grades of hydrocarbon solvents are available as shown in Tables 3 and 4.
The coal tar hydrocarbons are obtained by distillation of the material from the coke oven by product recovery process. They include benzene (benzol) toluene (toluol) xylene (xylol) and other aromatic hydrocarbons. The tremendous wartime demand for toluene to make explosives stimulated research to produce it from other sources. Today more aromatics are produced from petroleum than from coal tar.
The hydrocarbon solvents may be classified chemically in three groups
Aliphatics straight or open chain saturated hydrocarbons.
Naphthenics cyclic saturated hydrocarbons with or without alkyl side chains.
Aromatics cyclic hydrocarbons carbons containing the benzene ring structure.
The commercial hydrocarbon solvents are usually mixtures of closely related compounds and isomers hence the range in distillation temperatures for single solvents shown in Tables 3 and 4. The structures of typical hydrocarbons are shown in Fig. 2 with the boiling points of the pure chemicals.
The boiling points increase with increase in molecular weight in a given series but the effect of molecular shape is shown by the first three compounds in Fig. 2. They are isomers of hexane and therefore have the same composition and molecular weight but the boiling point decreases with the shortening of the main carbon chain. The increase in boiling point of the normal straight chain saturated hydrocarbons or alkanes is shown in Fig. 3. The normal alkanes containing from 5 to 16 carbon atoms in the chain are liquids at room temperature. The solid paraffin wax contains from 18 to 25 carbons per chain and polyethylene contains several hundred carbons. This topic was discussed in Chapter 1 with respect to the secondary valence forces and chain length and their effect on the physical properties.
The wide range of properties in hydrocarbon solvents which are available commercially is illustrated in Table 3 with the Amsco solvents of the American Mineral Spirits Co. ASTM Designation D86 46 gives the procedure for distillation of petroleum hydrocarbons. The report usually contains the temperatures of the initial boiling point (when the first drop falls from the end of the condenser) the points at which 50% and 90% by volume have been distilled the dry point (at which the bottom of the flask becomes dry) and the end point or temperature at which the last drop is obtained. Additional heat must be applied after the bottom of the flask is dry to obtain the last drop. It is advantageous in many cases to have the spread in distillation temperatures kept as small as possible but this requires closer fractionation with an increase in cost.
In Table 3 the solvent power is indicated by the range of KB values from 34 to 37 for the regular mineral spirits type of solvent to 105 for toluol. The straight aniline point is used for the aliphatic or mineral spirits ype solvents and the mixed aniline point for the aromatic type as explained previously. The results from the nitrocellulose dilution ratio test are given for the solvents having fast enough evaporation rates and sufficient solvent power to be used as diluents. The test was run with butyl acetate as the true solvent portion. The values would be different if another solvent were used as explained previously. It will be noted that the aromatics are higher in weight per gallon than the aliphatics.
A typical set of characteristics of aromatic solvents as produced by the coal tar industry is given in Table 4.
Fire and Explosion Hazard. A comparison of the flash point autoignition temperature and explosive limits in air for a variety of hydrocarbon solvents may be obtained from Table 5. The fire hazard is related to the volume of vapor in the air the volumes of solvent vapor per gallon of solvent evaporated at 80°F and 212°F are given. Also given is the solvent vapor density in comparison with air.
The volume of solvent vapor at a given temperature may be calculated from the relationship between molecular weight and volume. For example when the molecular weight is expressed in pounds 1 lb mole of vapor occupies 400 cu ft at 80°F and 500 cu ft at 212°F. These factors are expressed in the following formula
Using this formula to calculate the concentration of toluol at 80°F one obtains about 31 cu ft of vapor per gallon evaporated. When using a factor of safety of 4 this means that 4 × 100 × 31 = 12 400 cu ft of air must be supplied for each gallon of toluol evaporated to keep the concentration safely below the explosive point.
Dyes and Dye Intermediates
Dyes are intensely colored substances that can be used to produce a significant degree of coloration when dispersed in or reacted with other materials by a process which at least temporarily destroys the crystal structure of the substances. This latter point distinguishes dyes from pigments which are almost always applied in an aggregated or crystalline insoluble from. Modern dyes are products of synthetic organic chemistry. To be of commercial interest dyes must have high color intensity and produce dyeings of some permanence. The degree of permanence required varies with the end use of the dyed material.
All molecules absorb energy over various parts of the electromagnetic spectrum. The characteristic of dye molecules is that they absorb radiation strongly in the visible region which extends from 4000 7000 angstroms. Only organic molecules of considerable complexity which contain extensive conjugation systems linked to electron withdrawing and attracting groups give sufficient absorption (tinctorial value) in the visible region to be useful as dyes. The shade and fastness of a given dye may vary depending on the substrate due to different interactions of the molecular orbitals of the dye with the substrate and the ease with which the dye may dissipate its absorbed energy to its environment without itself decomposing.
The primary use for dyes is textile coloration although substantial quantities are consumed for coloring such diverse materials as leather paper plastices petroleum products and food.
The manufacture and use of dyes is an important part of modern technology. Because of the variety of materials that must be dyed in a complete spectrum of hues manufacturers now offer many hundreds of distinctly different dyes. An understanding of the chemistry of these dyes requires that they be classified in some way. From the viewpoint of the dyer they are best classified according to application method. The dye manufacturer on the other hand prefers to classify dyes according to chemical type.
Both the dyer and the dye manufacturer must consider the properties of dyes with relation to the properties of the materials to be dyed. In general dyes must be selected and applied so that color excepted a minimum of change is produced in the properties of the substrate. It is necessary therefore to consider the chemistry of textile fibers as a background for an understanding the chemistry of dyes.
The major uses of dyes are in coloration of textile fibers and paper. The substrates can be grouped into two major classes hydrophobic and hydrophilic. Hydrophilic substances such as cotton wool silk and paper are readily swollen by water making access of the day to the substrate relatively easy. On the other hand the ease of penetration also allows easy removal in aqueous systems and special techniques must be used where a high degree of wet fastness is required.
On the other hand hydrophobic fibers such as the synthetic polyesters acrylics polyamides and polyolefin fibers are not readily swollen by water hence higher application temperatures and smaller molecules are generally required.
The polymer chemist has increased the versatility of the newer fibers by incorporating dye sites of a varying nature as needed to achieve dyeability with a predetermined class of dyes. It is now possible to have polyesters acrylics and polyamide fibers which can be dyed with positive (basic cationic) negative (acid anionic) or neutral (disperse) dyes. These recent developments have allowed the fabric designer to produce materials (textiles carpets) fabricated in patterns which can be dyed three different colors from one dyebath containing three types of dyes. This concept is called cross dyeing and is becoming increasingly popular as a low cost method of coloration.
Cotton and Rayon
Cotton and rayon (regenerated cellulose) fibers are composed of cellulose in quite pure from. Cellulose lacks significant acidic or basic properties but has a large number of alcoholic hydroxyl groups. It is hydrolyzed by hot acid and swollen by concentrated alkali. When cotton is swollen by concentrated alkali under tension so that the fibers cannot shrink lengthwise it develops a silk like luster. This process is called mercerization. The affinity of mercerized cotton for dyes is greater than that of untreated cotton.
Cotton and rayon fibers are easily wetted by water and afford ready access to dye molecules. Dyeing may takes places by adsorption occlusion or reaction with the hydroxyl groups. It is also possible to make cotton and rayon receptive to a variety of dyes by pretreatment or mordanting with a material capable of binding the dyes.
Wool and Silk
Wool and silk fibers are protein substances with both acidic and basic properties. They are destroyed by strong alkali. Strong acid causes hydrolysis but the process may be controlled to permit dyeing from acidic solutions.
Wool and silk are wetted by water and are dyed with either acid or basic dyes through formation of salt linkages. They may also be dyed with reactive dyes that from covalent bonds with available amino groups. Mordanting is sometimes used to alter the dyeability of wool and slik.
Acetylated cellulose fibers differ from cellulose fibers in that they are more hydrophobic and lack large numbers of free hydroxyl groups. The higher the degree of acetylation the more unlike cotton and rayon the acetates become. Strong acid and strong alkali degrade cellulose acetates although the initial attack is slow under moderate conditions because of the difficulty of wetting the fiber. The triacetate is the most hydrophobic and the most stable.
Dyeing of cellulose actetates is effected with dyes of low water solubility which become dissolved in the fiber or by occlusion of dyes formed in situ. Acid basic and reactive dyes cannot be used because of the lack of sites for attachment.
Polyamide fibers (nylon) are synthetic fibers possessing properties somewhat like those of wool and slik. They are more hydrophobic however with only a limited numbers of basic or acidic groups. Polyamides are degraded by strong acid but may be dyed from acidic dye baths under controlled conditions.
Polyamide fibers are dyeable near the boiling point of water with acid dyes that from salt linkages with basic sites. Dyeing by this means is limited by the availability of these sities. Dyes like those used on cellulose acetates (i.e. that dissolve in the fiber) or reactive dyes that bond to available amino groups may also be used.
Polyester fibers are synthetic fibers unlike any produced in nature. They are hydrophobic and posses good stability to acid and alkali as a result of this hydrophobicity. They are hydrolyzed under sufficiently drastic conditions however. Some polyester fibers lack functional groups others are provided with acidic groups or otherwise modified to make them more hydrophilic.
Unmodified polyester fibers are dyed by solution of dyes in the fiber or to a limited extent by occlusion of dyes formed in situ. Modified polyester fibers may be dyed in these ways or with dyes selected according to the nature of the sites introduced by the modification. Both unmodified and modified polyester fibers must be dyed under vigorous conditions often with the assistance of a swelling agent to open up the fiber.
Acrylics fibers are hydrophobic synethetic fibers with excellent chemical stability. They do not resemble any natural product. The only funcational groups pressent are those introduced for the purpose of providing sites for dyeing.
Acrylic fibers are dyed by solution of dyes in the fiber by occlusion of dyes formed in situ and by formation of salt linkages with dyes capable of attachement to sites provided for that purpose. Basic dyes are used on acrylic fibers bearing sulfonic acid groups for examples.
Vinyl polymers and copolymers make up a class of fiber forming materials that varies greatly in properties depending on constitution. Some vinyl fibers are very resistant to degradation by acids. Dyes are selected accoding to the nature of the specific polymer to be dyed.
Polyolefin fibers are formed from the products of polymerization of unsaturated compounds of carbon and hydrogen for example propylene. They do not absorb water and are chemically quite inert. They can be dyed with special disperse dyes but are colored best by introducing a colorant into the polymer before the fibers are spun. Some types of polypropylene incorporate metal ions such as Ni++ to act as dye sites for chelatable dyes.
Glass fibers are used for special purpose for example where flammable materials cannot be tolerated. They are often colored during manufacture but can be dyed by special techniques which involve the use of surface coatings that have affinity for dyes.
Paper is a nonwoven material made up primarily from cellulose of varying degress of refining (see chapter 15). Paper may be colored in the pulp as a watery fibrous slurry by either continuous or batch methods. The dyeing process takes place at ambient temperature and the dyes are adsorbed on the pulp by their affinity for the cellulose. Direct dyes are most commonly used. In continuous coloration the dye solutions are metered directly into a moving stream of pulp. In batch operations dye is added to a pulper beater or blending chest containing a given quantity of slurry.
Paper may also be colored on its surface after the inital sheet is formed pressed and partially dried. This can be done at the size press of the paper machine or color can be carried by a calender roll for heavier sheets. A wide vareiety of low cost dyes can be used for surface coloration.
The Properties of Dyes
The properties of dyes may be classified as application properties and end use properties. Application properties include solubility affinity and dyeing rate. End use properties include hue and fastness to degrading influences such as light washing heat (sublimation) and bleaching. Dyes are selected for acceptable end use properties at minimum expense. Involved application procedures are used only when necessary to acheive unusually good results.
It has become common practice to treat dyed textiles with agents designed to improve resistance to shrinking wrinkling and the like. These agents frequently alter the appearnce and fastness of dyes. Stability to after treatments must therefore be considered as an important end use property of dyes.
The amount of dye required to obtain a light shade is usually about 1 per cent of the weight of the fiber heavier shades may require as much as 8 per cent. These values are very approximate since dyes differ in colour strength and are usually sold in diluted form. These amounts of dye are not sufficient in most cases to markedly affect the properties other than color of the fiber. Care must be exercised however to apply the dye under conditions that do not cause fiber degradation.
It is obvious from the list above that many basic dyes have about 10 20 times the color value per molecule as the anthraquinone types. Unfortunately light fastness is in the reverse order the anthraquinones being used where maximum durability to light is needed. The challenge to be dye chemist or engineer is to increase the strength of the light fast dyes or to increase the fastness of the strongest dyes.
Classification of Dyes
Dyes are classified according to application method for the convenience of the dyer. The best classification method available is that used in the color index a publication sponsored by the society of dyers and colourists (England) and the American Association of Textile Chemists and Colorists.
Acid dyes depened on the presence of one or more acidic groups for their attachment to textile fibers. These are usually sulfonic acid groups which serve to make the dye soluble in water. An example of this class is Acid yellow 36 (Metanil Yellow).
Acid dyes are used to dye fibers containing basic groups such as wool slik and polyamides. Application is usually made under acidic conditions which cause protonation of the basic cause protonation of the basic groups.
It should be noted that this process is reversible. Generally acid dyes can be removed from fibers by washing. The rate of removal depends on the rate at which the dye can diffuse through the fiber under the conditions of washing. For a given fiber the diffusion rate is determined by temperature size and shape of the dye molecules and the number and kind of linkages formed with the fiber.
Chrome dyes. A special kind of acid dye used mainly on wool they posses improved fastness when converted to chromium complexes. A suitable chromium salt is applied to the fiber (1) before the dye (2) at the same time as the dye or (3) after the dye. All these methods are staisfactory but more complicated than is desired. In recent years manufactures have made available dyes in which chromium is already a part of the molecule. These dyes are simpler to apply than the older types and as a consequence are increasing in importance.
Basic or Cationic Dyes
Cationic dyes become attached to fibers by formation of salt linkages with anionic or acidic groups in the fibers. Basic dyes are those which have a basic amino group which is protonated under the acid conditions of the dyebath. Cationic dyes can be divided into the three classes which are illustrated.
Basic brown 1 (Bismark brown) is an amino containing dye which is redily protonated under the pH 2 5 conditions of dyeing.
Crystal violet (Basic violet3) is an example of a cationic dye in which the cationic charge is delocalized by resonance and may be present at any one of the basic centers at any time. These resonance forms of almost equivalent energy are one of the reasons that crystal violet is among the strongest dyes known. This high color value (tinctorial strength) has important commercial interest in the hectograph copying system. In this system crystal violet in a wax base is transferred to the back of a typewritten copy sheet. By using paper moistened with alcohol more than 200 good copies may be made from the master.
Dry Cleaning Agents
Drycleanings of garments is done in much the same manner as laundering except that organic solvents are used in place of water. As in laundering detergents are added to the solvent to enhance its cleaning quality. Other solvent additives are used to give the textile the desired finish. This may be done merely to improve the hand or drape of the textile or chemical additives may be used to acheive water repellency insect repelleney or flame resistance.
Drycleaning washes are similar in construction to commercial laundry washers but in drycleaning provision is made for clarifying the solvent for reuse. In laundering the used wash water is discarded this cannot be done with the more expensive drycleaning solvents.
In the drycleaning system the solvent is continuously pumped through the washes and then through some type of after designed to remove all suspended soil. Provision is also made for distillation of the solvent to free it form the solvent soluble soil.
The filters also contain activated carbon to absorb dissolved dye which would otherwise build up in the solvent.
Other chemical products used in small quantities by drycleaners are formulted to remove stains by local appplicaton to the affected area of the garment.
Only two classes of solvents have proved suitable for drycleaning petroleum fractions and a few halogenated hydrocarbons. All other classes of solvents fail to meet the following eight major requirements of a drycleaning solvent.
It must not weaken dissolve or shrink the ordinary textile fibers.
It must not remove the common dyes from fibers.
It must be an acceptable solvent for fat and oils.
It must not impart an objectionable order to drycleaned textiles.
It must be sufficiently volatile to permit reclamation by distillation and to permit garments to be tried without prolonged heating at excessive temperatures.
It must be noncorrosive to metals either when dry or in the presence of water.
It must be relatively nontoxic.
It must have a flash point of 1000F or above.
The major drycleaning solvents used in the U.S. are the petroleum fraction called Stoddard solvent of which there are four types perchlorethylene and to a limited extent trichlorethylene and trichlorotrifluoroethane.
With the exceptions of the solvents the chemicals used in drycleaning are sold as brand name formulations and the only tests performed on them are the determination of the amount of detergent in the solvent and the amount of water in a solution of detergent in solvent. However these chemicals are tested to determine how well they perform the function they are designed for according to a number of procedurs developed by the National Insitiute of Drycleaning (NID).
Much of the drycleaning done in the United States employs a solvent corresponding to a petroleum fraction with a minimum flash point of 1000C.This solvent has been named Stoddard solvent for W.J. Stoddard. The first commercial standard for a drycleaning solvent CS3 28 was issued in 1928 by the National Bureau of Standards. The latest revision of this specification is CS3 41 it also became an ASTM. Table 1 summarizes the current specifications of regular Stoddard solvent.
Today three other petroleum fractions are also broadly termed as Stoddard solvents. These are the 1400F solvent the low end point solvent and the ordoless solvent.
1400F Solvent. This solvent is safer than the regular stoddard solvent. Therefore it may be used in locations where stoddard solvent is prohibited. Also building codes for plants using 1400F solvent are not so rigorous. For example explosion proof motors and other electrical fixtures are not required. Specifications of 1400F solvent which differ from the specificantions of the regular stoddard solvent are listed in Table 2.
Low End Point Solvent. This type of stoddard solvent has a dry point in the range of 330 3620F compared with 368 4090F for the regular stoddard solvent. The result is a rapid drying solvent. There is no specification covering this solvent. It is regaeded as a premium grade because of the fast drying feature.
Odorless Solvent. Whereas regular stoddard solvetn is specified to be free of objectionable odor this new class of stoddard solvent is free of all odors. This is acheived by removing or hydrogenating all aromatic compounds. The solvent also meets all requitements for a nonsmog producing solvent since smog production is related to the aromatic content.
Odorless solvent is also regarded as a premium grade of stoddard and is not conered by a separate specification.
Some comments to the specification tests for stoddard solvent are given here.
Odor. The term sweet as used in the specification means the opposite of rancid or sour. Although the usual methods of clarifying solvent in a drycleaning plant remove odors that accumulate during continued drycleaning these processes do not always remove dodrs caused by improper refining. Therefore the solvent when received from the refinery should be free from undesirable odor. There is nothing to show whether or not a solvent meets the requirement except the opinion of the examining chemist. Many samples of stoddard solvent have a rather strong odor but it is easily removed from the fabric by conventional drying methods.
Flash point. The flash point is governed by those portions of the solvent that have the lowest boiling points and are therefore the most volatile. Since these portions evaporate more rapidly than the rest of the solvent the flash point of Stoddard solvent in a drycleaning system gradualy rises with use. Soaps prespotters or other added materials sometimes contain low flash solvents (such as some alcohols) that lower the flash point of the solvent and increase the fire hazard. The introduction of even small amounts of methyl ethyl or isopropyl alcohol into a washer lowers the flash point of the solvent below normal room temperatures.
A lighted match held over Stoddard solvent at ordinary room temperatures does not ignite the solvent because the solvent is not giving off enough vapor to form a combustible mixture with the air. If the tempetature of the solvent is raised vapor pressure is increased and the air above the solvent becomes richer in solvent vapors. Finally a temperature is reached where enough solvent has vaporized to form a combustible mixture with the air If a flame is then introduced above the solvent the vapors will flash. The lowest temperature at which this occurs is called the flash point. Since the flash point of stoddard may not be under 1000F there is no danger of fire from solvent vapors until the temperature of the solvent rises to 1000F or above.
The flash point specification is frequently violated. In some cases the refinery may have set the lower limit of the distillation range too low. Such violations usually result in a solvent with a flash point of 98 or 990F. A more dangerous type of violation however results from careless handling of the solvent. For example stoddard is sometimes transported in tank trucks that were previously used for carrying gasoline and still contain small quanities of gasoline. As little as 1% of gasoline in stoddard solvent lowers its flash point considerably.
The flash point is determined by ASTM method.
Corrosive properties. Improperly refined solvent may contain traces of dissolved free sulfur which can corrode the metals of storage tanks and equipment. The corrosion test is carried out according to ASTMD 1616 60 at an elevated temperature 121ºF. Under these conditions corrosion that would be apparent after considerable use of the solvent at room temperature can be seen after only 3 hr.
Distillation Range. From the standpoint of a drycleaning solvent there are disadvantages in products containing very low or very high boiling hydrocarbons. Low boiling hydrocarbons petroleum ether and gasoline cause fires and high evaporation loss high boiling hydrocarbons such as kerosene cause excessive drying time. The distillation range for Stoddard solvent is between 300 and 4100F a range not low enough to cause undue fire hazard and evaporation loss or high enough to prolong the drying time.
The distillation range of stoddard solvent is determined according to ASTM D 86 66 and D 1078 67.
Residue. Excessive nonvolatile matter in the solvent often contributers to odors and lengthens drying time. Because of the high temperatures used a small amount of odorous residue is usually formed during the distillation test. A sample of the same solvent evaporated on a steam bath where temperature is not raised above 2120F yields a smaller and less odorous residue.
Acidity. If the solvent is given a sulfuric acid treatment in the refinery and not followed by a neutralizing treatment (such as with caustic soda) it will contain small amounts of sulfuric acid or other acidic materials. Even small amounts of sulfuric acid are undesirable in a drycleaning solvent as they corrode equipment and damage garments. Fortunately almost without exception drycleaning solvents pass the acidity test.
Sulfuric acid has a very high boiling point. The presence of this substances in the solvent will be as residue in the flask after the distillation. If a residue of 1 ml remains from distilling 100 ml of solvent any sulfuric acid present is concentarted there 100 times. Thus it is logical to test the residue from distillation for sulfuric acid.
According to ASTM D 1093 65 the solvent sample is shaken with water and one drop of a 0.1% methyl orange indicator solution is added to the aqueous layer there should be no change in the color of the indicator.
Doctor Test. Mercaptans impart to the solvent unpleasant odors which may be absorbed onto the garments during drycleaning. The doctor test is a qualitative method to determine whether the treatment for mercaptans was properly done in the refinary. Sulfur and sodium plumbite are added to the solvent in a test tube. If mercaptans are present in the solvent the reaction proceeds and the black lead sulfide formed is indicative of a positive test
Sulfuric Acid Absorption Test. This test determines if the solvent contains appreciable amounts of unsaturated hydrocarbons. These would be in the solvent if it was inadequately treated with sulfuric acid during refining. Since unsaturated hydrocarbons turn ranoid and cause undesirable odors in drycleaned garments it is imperative that they be removed before the solvent leaves the refinery.
In the test concentrated sulfuric acid is added in a graduated cylinder to the solvent and shaken. Sulfuric acid reacts with any unsaturated hydrocarbons present and most of the products of the reaction settle into the acid layer thus the volume of the solvent is decreased. Since some of the products formed from the reaction remain in the solvent layer the test does not give a quantitative measure of unsaturated hydrocarbons that is a 5% absorption of the solvent by sulfuric acid actually represents a greater percentage of unsaturated hydrocarbons in the solvent.
Variations in the strength of commerically available concentrated sulfuric acid cause variations in the sulfuric acid absorption test. Therefore the acid strength must be standardized if reproducibility is desired.
Perchloroethylene (tetrachloroethylene) became an important drycleaning solvent because of its nonflammability which permits its use in places where all types of flammable solvent are either forbidden by codes or inhibited by high insurance rates. Its general properties are given in Table 3 and the specifications proposed by the NID are listed in Table 4.
Residual Odor. Any residual odor left in a fabric after treatment in the solvent is objectionable. Detection of such odors by smelling is more sensitive if the fabric is steamed immediately prior to the test. A swatch of bleached but unfinished cotton poplin Style A 400W Testfabrics Inc. is used and subjected to the following test.
Condition the cotton at 60% relative humidity for at least 8 hr prior to use. Soak the swatch in perchlororthylene for 5 min then remove it and hang it to drain dry for about 4 hr. Tumble the swatch in a tumble dryer for 30 min at 1400F.
To test for odor grasp the swatch in the center with a forceps hold it in live steam for 5 sec and smell it immediately. Test an untreated swatch smiultaneously. There should be no discernible difference in odor between the two swatches.
Nonvolatile Residue. This test detects the presence of nonvolatile impurities in the solvent. It is determined gravimetrically by evaporating a measured quantity of solvent and weighing the residue as follows.
Dry a 4 in. diameter evaporating dish and weigh it to the nearest 0.1 mg. Place it on a steam bath in a hood and add the perchloroethylene to be tested by pipet in two 50 ml portions. Perchloroethylene has a high specific gravity. 1.62 and is difficult to handle in a 100 ml pipet. Add the second portion afted the first is partially evaporated.
After the solvent has completely evaporated on the steam bath heat the dish further in an oven at 1050C for I hr then cool it in a desiccator and weigh. The increase in weight of the dish in grams for a 100 ml sample is % nonvolatile residue.
Stability Test. Perchloroethylene is stabilized by adding traces of chemicals known to inhibit its decomposition. Loss of stabilizer or the pressure of certain impurities can lower the stability of the solvent.
Wash two strips foil 2.0×7.5×0.005 cm in concentrated hydrochloric acid. Rinse dry and weigh to the nearest 0.1 mg. Add 75 ml of the test solvent and 3 ml of water to a 300ml Soxhlet extractor. Place one copper strip in the flask and the other into the condenser of the soxhlet. Heat the Soxhlet at a rate that will cause it to empty 8 10 min.
After 24 hr remove the strips wash them again in concentrated hydrochloric acid and weigh. The combined weight loss of the two strips should not exceed 30 mg.
Note Do not fail to add the water with the solvent. The test is worthless in the absence of water.
Around 1960 du pont introduced trichlorotrifluoroethane as a drycleaning solvent under the trade name valclene. This solvent has aroused much interest because of its ideal properties but it is too volatile to be used in machines designed for perchloroethylene. Therefore its full utilization must a wait machine development. A number of companies have introduced small machines for the solvent but it will be several years before use of the solvent is widespread.
No special specifications or test methods have been developed for this solvent.
Used Drycleaning Solvents
In addition to the tests given under the specifications there are several analytical methods designed for quality control purposes in drycleaning operations. These methods are normally performed on used solvent taken from plant washers. The following tests are made routinely on used solvent.
Detergent Concentration. The method of fessler for anionic detergents is used. There is no satisfactroy method for drycleaning detergents that are all nonionic however manufacturers of niononic detergent formulations normally include some anionic surfactant in the mixture to serve as a tracer. This serves the purpose of quality control with a known product but not for analysis of an unknown mixture.
Nonvolatile Residure. Except that 10ml samples are used instead of 100 ml.
Moisture Content. The moisture content of the used solvent can be determined by the Karl Fischer method.
Acid Number. This test was originally designed to measure the buildup of fatty acids in the solvent. Its value has diminished in recent years because of the widespread use of amine sulfonate detergents. These detergents react quantitatively with the titrant giving a high value for the fatty acid content of the solvent. However the test is still useful for control purposes where proper correction can be applied for interfercence by the detergent.
In other fields acid number is defined as the mg of potassium hydroxide neccessary to neutralize l g of sample. In drycleaning. The NID has defined acid number as the mg of potassium hydroxide necessary yo neutralize 1.28 ml of solvent.
The titration is made in the usual manner using a 0.06 N alcoholic solution of potassium hydroxide and phenolphthalein indicator. It was found that 2 methyl 2 4 pentanediol is a better solvent than ethanol because of its solubility in petroleum solvents.
Residual Odor. The test is carried out according to the procedure given on p 608.
Color. The color of used drycleaning solvents may be due chiefly to dyes dissolved from the textiles. The balance is caused by colored soils or colloidally suspended pigments. The latter are removed by microfiltration prior to determining color. At NID color is determined on a Coleman universal spectrophotometer using a 40 mm cuvet at 500nm. The instrument is standardized against water.
Greying of Cotton. The cotton fabric used for the residual odor test is read on the reflectometer to determine the decrease in % green reflectance. Although this is called greying it is actually a measure of the amount of dye and colored impurities dissolved in the solvent because the insoluble material has been removed by microfiltration through 0.2 µm membranes.
Sizing. Many drycleaners use certain resins in the solvent as sizes or bodying agents for fabrics to replace the finishing materials removed during wear or cleaning. Natural terpene resins are widely used and the amount of resin in the solvent is determined at NID by extracting the nonvolatic residue with boiling ethanol. This reagent dissolves everything except the terpene resins. The procedure has not been validated however for all types of sizes.
Suspended Soilds. After microfiltration of a measured volume of the solvent the membrane witch has been previously weighed is oven dried at 1050C and weighed to determine the quantity of insoluble material suspemded in the solvent. The NID standard for this is 50mg\liter. Larger quantities can cause excessive greying of white fabrics and is an indication of poor solvent filtration.
Detergency in nonaqueous solvents follows much the same principles as in water particularly in the removal of insoluble soil. The major differences come in the attack on water soluble and solvent soluble soils. In aqueous detergency the major attack is on the oily soils because the water soluble soils are removed by simple solution. In drycleaning on the other hand the major attack by detergents is on the water soluble soil because the soil is removed by simple solution.
In both lanundering and drycleaning the process of emulsification and solubilization effects the removal of soil from the fiber surface. In both types of eleaning the detergents used are based on surface active agents.
Laundry detergents generally contain not more than 20% surface active agents (surfactants) the balance being various types of builders. Drycleaning detergents may consist of a single surfactant. The product may also contain a consolvent or coupling agent to enhance the capacity for dissolving or emulsifying water and a fluorescent whitening agent. Frequently two or more surfactants are mixed.
A drycleaning detergent performs three functions in the cleaning process. It acts as a dispersant or peptizing agent for insoluble soils. It not only disperses this kind of soil but also keeps it in suspension while it is being flushed out of the fabric and pumped to the filter. Insoluble soils may be dispersed to particle size in the submicron range by good detergents and while so dispersed the particles of soil are small enough to escape between the tightly packed fibers in textile yarns. In the absence of a good detergent this kind of soil is difficult to remove and readily redeposits on other fiber surfaces causing what is generally called greying a phenomenon also common in laundering particularly with polyester fibers. Thus the first two roles of a drycleaning detergent are to assist in the removal of insoluble soil and to prevent it from redepositing on other fabrics in the bath.
The third function of a drycleaning detergent is to emulsify water in the solvent and promote the removal of water soluble soil by the emulsified or solubilized water. Although the water plays the major role in detaching water soluble soil from the fiber surface the detergent itself can dissolve some of these soils within its micelles.
Progress in the formulation of drycleaning detergents is slow compared to the formulation of laundry detergents. One reason for the lack of progress has been the absence of reliable test methods for drycleaning detergents. The literature on drycleaning detergent test methods is scanty and the few methods that have been described have received little attention or use. The methods described here have been in use at the National Institute of Drycleaning and are designed to test the ability of a detergent to perform its three functions.
Methods of Analysis
The tests to be carried out on drycleaning detergents can be divided into two groups specification tests resulting in information on the properties of the detergnets and performance tests.
Physical Composition. Drycleaning detergents almost without exception are liquids so it is describle to know how much of the material consists of an active ingredient and how much is solvent or water. The determination is made on a perchloroethylene solution of known concentration of the detergent. An aliquot is evaporated to dryness as described on p. 608 for the determination of the nonvolatile residue of a solvent. The amount of water is determined on a separate sample by the karl fischer titration.
Some drycleaning detergents are diluted with mineral oil so that the nonvolatile residue is not all surfactant but it still established the upper limits of surfactant concentration.
Specific Gravity. The main purpose of this test is to establish what types of solvents are used as diluents. Most surfactants have specific gravities close to unity whereas drycleaning solvents have a specific gravity of about 0.8 (Stoddard solvent) or 1.62 (perchloroethylene). The determination can be carried out by any of the conventional methods.
pH. A drycleaning detergent should be essentially neutral because of the adverse effect of acids and alkalis on some types of dyes. The test is made by thoroughly shaking the detergent with water and determining the ph of the water phase.
Distillation Test. Since used dryeleaning solvent is reclaimed by distillation it is important that the detergent cause no problems in the still. This is checked qualitatively by distilling a 1% solution of the detergent in perchloroethylene in an all glass laboratory still. The process is observed for any signs of foaming flooding over or decomposition. The distillate should be pure perchloroethylene presence of other volatile solvents is undesirable.
Detergents intended for use in Stoddard solvent must be tested by vacuum distilling a 1% solution in this solvent.
Solubility in Dryeleaning Solvents. The purpose of this test is to ascertain that the detergent is soluble in both solvents. A simple qualitative test is sufficient.
Chemical Type. It is desirable to know whether the sufactant in the detergent is anionic cationic nonionic or a mixture of ionic and nonionic surfactants. This can be determined by studying the infraed spectrum of the sample as well as the methylene blue titration method given blow.
Detergent Concentration by Methylene Blue Titration. This method is widely used as a control test to determine the amount of a prticular detergent in a drycleaning solution. It was originally described by Fessler in 1951. The following procedure is from an NID publication.
Anionic Surfactants. Place 25ml of chloroform into a 100 ml glass stoppered graduated cylinder. Take at least a 5 ml sample of solution to be tested dilute to 100 ml and then add a proper aliquot to the chloroform. Add 25 ml of water containing 1 drop of a 0.5% methylene blue solution and shake. The methylene blue enters the chloroform layer as a result of solubilization by the surfactants. Start to add a 0.02% aqueous cetylpyridinium chloride solution in 0.5 ml increments and shake the mixture vigorously after each addition. As long as any free anionic surfactant remains in the chloroform layer its blue color will presist. Near the end point the methylene blue begins to pass into the aqueous layer. Eventually this is complete and the lower layer is colorless. A sharp and reproducible end point is the point of equal color distribution between the two phases. Prepare a calibration curve for each detergent by titrating a number of samples of known volume volume concentration over the expected range and plotting ml of titrant againt deteregent concentration.
Cationic Surfactants. Carry out the determination in a similar way but by using a standard anionic surfactant such as Aerosol OT as the titrant.
Nonionic surfactants cannot be titrated in this manner. However detergents consisting of nonionic surfactants generally contain a small amount of anionic surfactant as a tracer so the solution can be titrated to control concentration.
As early as 600 B.C. seaweed was used as a food for man but algin a component of seaweed was first discovered by British chemist E. C.C. Stanford in 1880. In 1896 A Krefting prepared a pure alginic acid. In 1929 Kelco Company began commercial production of alginates and introduced milk soluble again as an ice cream stabilizer in 1934. In 1944 propylene glycol alginate was developed.
Algin is a polysaccharide found in all brown seaweeds phaeophycea which grow on rocky shores or in ocean areas that have clean rocky bottoms. Although some species can be found at the high tide line other exist along the shore where depths are less than about 40 m (125 ft) the maximum depth to which sunlight will penetrate. (Since algae do not have true roots stems or leaves nourishment comes directly from sunlight and mineral nutrients in ocean water.)
Only a few species of brown seaweeds are used for commercial production of algin. The principal source of the world s supply of algin is the giant kelpMacrocystis pyrifera found along the coasts of North and South America New Zealnd Austrila and Africa. Other seaweeds used for algin manufacture areAscophyllum nodosum and species of Laminaria and Ecklonia.
Algin exists in the kelp cell wall as the insoluble mixed salt (calcium magnesium sodium potassium) of alginic acid. Alginic acid is a high molecular weight linear glycuronan comprising solely D mannuronic acid and L guluronic acid.
Algin is used in foods and general industrial applications because of its unique colloidal behaviour and its ability to thicken stabilize emulsify suspend form films and produce gels. These properties are discussed in greater detail in later sections of this chapter such as solution Properties and commercial Uses.
It has been in recent years only that the composition of alginic acid has become understood. Table 1 shows the composition of alginic acid whereas Table 2 shows the proportions of polymannuronic acid segments ployguluronic acid segments and alternating segments of these two uronic acids in three commercial samples of alginic acid. Figures 1 to 3 illustrate the structures of mannuronic and guluronic acids the apparent discrepancies between the date of Tables 2 and 3 are accounted for by variations between alginates derived from different species of brown algae.
Chemical derivatives The propylene glycol ester of alginic acid is the only orgainc derivative of alginic acid currently on the market. Propylene glycol alginate has improved acid stability and resists precipitation by calcium and other polyvalent metal ions.
Amine alginates can be made by reacting alginic acid with orgainc amines. Suitable amines are triethanolamine triisopropanolamine butylamine dibutylamine and dimylamine. Algin acetate and algin sulfate esters have been prepared but have no known applications. Carboxymethyl alginate can be made by treating sodium alginate with chloroacetic acid and alkali. A number of alkylene glycol esters of alginic acid have been prepared and evaluated.
Ethylene oxide can be reacted with alginic acid to form 2 hydroxyethyl alginate. Alginamides can be prepared by reacting propylene glycol alginate with primary amines such as ammonia ethanolamine ethylenediamine ethylamine propylamine isopropyl amine and butylamine. Very little reaction occurs with secondary amines.
Macrocystis pyrifera the brown seaweed that is the main source of algin grows in relatively calm waters and in large dense beds. The plant is a perennial and can be harvested on a continuing basis. Its rapid growth permits up to four cuttings per year.
Only mature beds are cut. At the time harvesting a dense mat of fronds floats on the ocean surfac. Cutting the dense mat on the surface allows light to penetrate the water and reach the immature fronds this stimulates their growth. Harvesting is actually a massive pruning of the kelp bed. Underwater blades mow the kelp approximately 3 ft below the water surface then the cut kelp is automatically conveyed into the hold of the barge by a moving belt.
Although commercial methods of producing sodium alginate from seaweed are proprietary the fundamental steps in a typical process essentially one of ion exchange are shown in Fig. 4. In the seaweed the algin is apparently present as a mixed salt of sodium and /or potassium calcium and magnesium and is a high molecular weight polymer. The exact composition varies considerably with the type of seaweed but does not affect processing.
It is possible to extact sodium alginate from seaweed with a strong solution of a sodium salt however for the production of purfied alginates the commercial processes are much more efficient. Alginic acid may also be neutralized with bases to give salts and reacted with propylene oxide to make propylene glycol alginate.
Commercially available water soluble alginates include the sodium potassium ammonium calcium and mixed ammonium calcium salts of alginic acid propylene glycol alginate and alginic acid itself. The physical properties of several of these alginates are given in Table 3.
Powdere Alginates Alginate as a hydrophilic polysaccharide absorbs moisture from the atmosphere therefore equilibrium moisture content is related to relative humidity as shown in Fig. 5. The dry storage stability of alginates is excellent at moderate temperatures 25oC (77oF) or less however they should be stored in a cool dry place. Table 4 gives datea showing the effects of storage for 1 year at 24.9oC (75oF) on typical alginates. Table 5 shows the effects of variious storage temperatures on the stablilities of alginates.
Pure alginates dissloved indistilled water from smooth solutions with long flow characteristics. The physical variables that affect the flow properties of alginate solutions are temperature shear rate polymer size concentration and the presence of solvents miscible with the distilled water. The chemical variables that affect algin solutions are pH and the presence of sequestrants monovalent salts polyvalent cations and quaternary ammonium compounds.
Rheological Properties The flow properties of sodium alginate solutions are concerntration dependent. A 25% medium viscosity sodium alginate solution is pseudoplastic over a wide range of shear rates (10 to 10 000s 1) wheras a 0.5% solution is Newtonian at low shear rates (1 to 100s 1) and pseudoplastic only at high shear rates (1000 to 10 000s 1) as shown in fig. 6.
Because of high molecular weight and molecular rigidity sodium alginate forms solutions of unusally high apperant viscosity even at low concentraions. Propylene glycol alginate solutions are shear thining over a wide range of shear rates at 3% concentrations. However at 1% or lower concentrations solutions have almost constant viscosity below shear rates of 100 s 1(fig. 7).
Figure 8 shows that viscosity shear curves of medium viscosity sodium and potassium alginates are virtually the same over the entire shear range. On the other hand in comparing low viscosity propylene glycol and sodium alginates the curves are indentical at shear rates greater than 10 000s 1 but diverge at low shear rates.
The effects of solution soilds on shear thinking are illustrated in fig. 9. The viscosity shear curves of a 2% solution of medium viscosity sodium alginate were the same as those of a 9% solution of a low viscosity sodium alginate. Measurements were taken using shear rates in the brookfield viscometer range (1 to10 000 s 1). At high shear rates such as those experienced at 100 000 s 1 measured with a capillary viscometer the curves diverge.
Figure 10 illustrates the effect of temperature on the flow of a high viscosity propylene glycol alginate. Addition of a sequestrant sodium hexametaphosphate to a medium viscosity sodium alginate (fig. 11) gives a viscosity shear curve comparable to that of a low calcium sodium alginate.
Xanthan gum can be used to modify the rhelogical behavior of sodium alginate solutions (fig. 12) As shown the curves for the 0.5% solutions of sodium alginate and xanthan differ greatly. A combination of the two gums produces flow properties intermediate between the two materials.
Latex paints illustrate the importance of rheological properties to the design of product performance. If the paint is highly pseudoplastic application will be easy and sagging will be prevented but flow and leveling will be minimal and brush marks will be left on the dried paint film. Elimination of the yield value will result in setting out of the pigments in the can. If dilatancy occurs stirring will be difficult and brush drag will be excessive.
Effect of temperature The viscosities of algin solutions decrease as temperatures increas approximately 12% for each 5.6ºC (100F) increase in temperature. The decrease is reversible if the high temperatures are not held for long periods. Table 6 shows the effect of time and temperature on solution viscosity. It is apparent that the heating of sodium alginate results in some thermal depolymerization the amount being related to both temperature and time.
Although a reducation in temperature of an alignate solution in an increase in viscosity it does not produce a gel. A sodium alginate solution can be frozen and thawed without any change in its appearance or viscosity after remelting. It is possible to form a freeze dried sodium calcium alginate gel with an absorptive capacity of more than 5000%.
Effects of Solvents Addition of increasing amounts of nonaqueous water miscible solvents such as alcohols glycols or acetone or an aqueous alginate solution increases solution viscosity and eventually causes precipitation of the aligante. Tolerance of the alignate solution to such solvents is influenced by the source of the alginate the degree of polymerization the cation type present and the solution concentration. Table 7 gives data on solvent tolerances of various types of alginates in solution.
Effect of Concentration Figure 13 shows the effect of solution concentration on selected grades of sodium ammonium postassium and propylene glycol alginates.
Effect of pH Sodium alginates with some residual calcium content increase in viscosity at a ph of 5.0 and are unstable at pH levels of about 11.0. Sodium alginates with minimal calcuim content do not show the viscosity increase until the pH reaches 3.0 to 4.0 Lower molecular weight sodium alginates are stable at a pH as low as 3.0 if calcuim is completely sequestered.
Propylene glycol alginates do not gel until the pH is below 3.0 but they do saponify at pH levels above 7.0 The long term stability of sodium alignate solutions is poor when the pH reaches 10.0. At even higher pH values there is depolymerization with an accompanying viscosity loss. Figure 14 illustrates the effect of pH on viscosity for several types of alignates in solutions.
Gelation Algin polymers will react with most polyvalent cations (magnesium excepted) to form crosslinkages. As the content of polyvalent ion increases the algin solutions thicknes then gels and finally there is precipition. The proposed structure of an alignate gel in which the calcuim ions are bound between the associated segments of the polymer chain is shown in fig. 15.
All alginate gets are the result of interactions between the alginate molecules which produce a three dimensional structure controlling the mobility of the water molecules. They are not thermally reversible. By the proper selection of gelling agent gel structure and rigidity are controlled. Loss of water to the atomsphere and resulatant shrinkage is very slow in algin gels.
Metallic polyvalent ions e.g. zinc aluminum copper and silver from complexes with aliginates in the presence of excess ammonium hydroxide. When the ammonia is driven from the system the insoluble metal alignatic is formed. Calcium is the polyvalent cation most often used to change the rheological properties and get characteristics of algin solutions. Calcium is also used to form insoluble aliginate filaments and films.
The method of calicum addition to an alginate system greatly influcences the properties of the final gel. If calcium is added too rapidly the result is spot gelation and a discontinuous gel structure. The rate of calcium addition can be controlled by use of a slow dissloving calcium salt or by the addition of a sequestrant such as tetrasodium pyrophosphate or sodium hexametaphosphate.
Effect of sequestrants The purpose of squestrants in alginate solutions can be either to prevent the alginatic from reacting with polyvalent ions present in the solution or to sequester the calcium inherent in the alginate. Polyvalent ion contaminants can come from water chemicals pigments or various natural origin materials. Figure 16 and 17 show viscosity concentration relationships for two types of alginates which and without sodium hexametaphosphate as the sequestrant.
In fig. 16 a low calcium sodium alginate shows a very small viscosity change up addition of the ployphosphate sequestrant to the solution. In contrast fig. 17 should that a sodium calcium alginate solution has a major change in viscosity when the sequestrant is added. Sequestered alginate solutions are more Newtonian in behavior than are those with some available calcium.
Effect of Monovalent Salts Monovalent salts depress the viscosities of dilute sodium alginate solutions. The maximum effect on viscosity is attained at a salt level of 0.1N in the solution. Except for alginates high in calcium an increase in alginates concentration decreases the effect of the monovalent electrolyte.
Figure 18 shows the effect of sodium chloride on the viscosity of several kinds of alginate solutions whereas Table 8 shows the effects of 1% and 5% sodium chloride concentrations over a 210 day period at temperatures of 4.4 23.9 and 48.90C (40 75 and 1200F). The effect of a salt on an alginate solution will vary with the source of the alginate as well as with its degree of polymerization the concentration of alignate in the solution and the type of salt.
Insolubilization Normally insoluble adducts result when sodium alginate reacts with cationic organic ammonium compounds. This insolubilization can be prevented by adding an electroylte e.g. NaCl to suppress the activity of the cation. Salt concentrations needed to solubilze insoluble alginate adducts are listed in Table 9.
Alginates in solution have compatibility with a wide variety of materials including other thickeners synthetic resins latices sugar oils fats waxes pigments various surfactants and alkali metal solutions. Incompatibilities are generally the result of a reaction with divalent cations (except magnesium) or other heavy metal ions cationic quaternary amines or chemicals that cause alkaline degradation or acid precipitation. In many cases the incompatibility can be avoided by sequestration of the metal ion or by careful control of the solution pH.
Table 10 lists materials that were tested for compatibility with solutions of a medium viscosity purified sodium alginate.
Preservaties Alginates have compatibility with most commonly used preservatives except quaternary ammonium compounds. The polysaccharide is quite resistant to the common enzyme systems produced by bacteria however since the solutions will support microbiological growth a preservative should be used if alginate solutions are to be stored for any considerable period of time. The preservatives listed in Table 10 exhibit good compatibility.
Sodium benzoate can be used to protect against bacterial action in acid systems. For additional protection against yeast and mold potassium sorbate or calcium or sodium propionate can be effective.
Thickeners The alginates show compatibility with most commercially available thickeners both synthetic and natural. With some thickeners a synergistic viscosity increase may be noticed. If the residual polyvalent ion content of a natural gum causes gelation of an algin solution the gelation can be controlled by the proper use of a sequestrant.
Water Soluble Resins The compatibility of the alginates with most water soluble resins is excellent. Polyvinyl alcohol exhibits definite synergism with sodium alginate in the formulation of grease resistant films.
Latices Those latices normally used in the formulation of paints paper coatings and adhesives have compatibility with the alginates. However latex emulsions with pH of 4.0 or less will cause gelation of the alginate. The apparent incompatibility may be overcome by proper buffering. High viscosity ammonium alginate may be used as a creaming agent for natural rubber latex and for several types of synthetic latex.
Organic Solvents As shown in Table 10 alginate solutions will tolerate up to 30% water miscible solutions. However viscosity increases may occur with long term storage. To prevent localized gelations it is necessary that there is good agitation of the solution at the time the organic solvent is added.
Enzymes Enzymes commonly encountered as by products or as commercially available products e.g. protease cellulase amylase galactomannanase have no effect on the alginate molecule. Storage test data is given in Table 11 for representative enzymes.
Surfactants Although alginate solutions have compatibility with anionic non ionic and amphoteric surfactants high concentrations of surfactants will result in a loss of viscosity and eventually the aliginate will out salt of solutions.
Non ionic surfactants can be used at concentrations higher than those allowable for the anionics or amphoterics. Some cationic surfactants may be used if approximately 2.5% of a soluble salt such as sodium chloride is added to the system. The exact salt level required depends upon the particular cationic material in the system.
Plasticizers Plasticizers such as glycols or glycerol may be used to improve the flexibility of alignate films. Data for a number of plasticizers and their effects on alginate solutions are given in Table 10.
Inorganic Salts The compatibility of alginate solutions with inorganic salts is limited to ammonium magnestium or the alkali metal salts. Divalent or higher valence cationic salts will unless sequestered cause gelation or precipitation of the alginate. The alginates will also be precipitated by molar solutions of monovalent salts.
Salts which are sightly acidic may produce large viscosity increase after prolonged storage. Sequenstrants in many cases will improve the salt compatibility and stability of the sodium alginate. Mixed alginate salts (sodium/calcium alginate) are much more salt sensitive than are the alkali metal alginates.
Cellulose a large volume renewable agricultural raw material is transformed into hundreds of products affecting every phase of daily life. Its use and versatility are exploited by the chemical industry much as the meat industry exploits its raw materials using everything but the squeal.
The production of water soluble cellulose derivatives in contrast to that of polymers based on petrochemical resources starts with a preformed polymer backbone of either wood or cotton cellulose instead of a monomer. Cellulose is a linear polymer of anydroglucose with the O glucopyranosyl structure shown below
The properties of a specific cellulose ether depend on the type distribution and uniformity of the substituent groups. For each O glucopyranosyl ring there are three hydroxyl groups available for the nucleophilic substitution reaction. Reactions at these sites can occur either on a one to one basis or with formation of side chains depending on choice of reagent employed to modify the cellulose. In the former case the term degree of substitution (DS) is used to identify the average number of sites reacted per ring. The maximum value is 3 corresponding to the number of hydroxyls available for reaction. When side chain formation is possible the term molar substitution (MS) is used and the value can exceed 3.
The water soluble cellulose ethers possess a range of multifunctional properties resulting in a broad spectrum of end uses.
This family of commercial water soluble cellulose ethers comprises methylcellulose (MC) and the methylcellulose derivatives hydroxypropylmethylcellulose (HPMC) hydroxyethylmethyl cellulose (HEMC) also possesses many properties and end uses in common with the methylcellulose products and is included in this section.
Methylcellulose and hydroxypropylmethylcellulose are two examples of this versatile class of water soluble hydrocolloids derived from the etherification of cellulose. MC and HPMC are polymers having the useful properties of thickening thermal gelation surfactancy film formation and adhesion. Those characteristics earn them application in areas such as foods cosmetics paints construction pharmaceuticals tobacco products agriculture adhesives textiles and paper. Additionally to tailor a product for a specific end use the properties of MC and HPMC may be modified by changing the molecular weight or the relative amounts of etherifying ragents.
Commercial MC products have an average degree of substitution (DS) ranging from 1.5 to 2.0 hence one half to two thirds of the available hydroxyl units are substituted with methyl groups (Table 1). In commercial HPMC products the DS for methyl groups ranges from 0.9 to 1.8 and the substitution (MS) of hydroxypropyl groups range from 0.1 to 1.0.
MC and HPMC possess the rather unusual property of solubility in cold water and insolubility in hot water so that when a solution is heated a three dimensional gel structure is formed. By modifying production techniques and by altering the ratios of methyl and hydroxypropyl substitutions it is possible to produce products whose thermal gelation temperature ranges from 50 to 900C (122 to 1940F) and whose gel texture ranges from firm to rather mushy.
Altering the amounts of methyl and hydroxypropyl substitution also affects the solubility properties of the cellulose ether. Decreasing the substituent groups below a DS of 1.4 gives products whose solubility in water decreases. Concentrations of 2 to 8% sodium hydroxide are required for solubility as the level of substitution decreases. Increasing the substitution above an MS of 2.0 improves solubilty in polar organic solvents.
Kalle & Co. A. G. of west Germany also produces hydroxyethylmethylcellulose. The small amount of hydroxyethyl substitution increases the solubility of the polymer and raises the thermal gel point from about 55oC to about 70oC. The more polar nature of the hydroxyethyl group versus the hydroxypropyl group allows for the formation of a slightly stiffer gel than is possible with an HPMC material of comparable gelation temperature.
A third product ethylhydroxyethylcellulose is similar in many properties to MC and HPMC. The small amout of hydroxyethyl substitution raises the thermal gel point from about 55ºC (131ºF) to about 70ºC (158ºF). The more polar nature of the hydroxyethyl group allows for the formation of a slightly stiffer gel than is possible with an HPMC material of comparable gelation temperature.
In many respects the properties and uses of EHEC are also very similar to those of methylcellulose and hydroxypropylmethylcellulose. The product has the characteristic properties of thickening surfactancy film forming binding solubility in cold water and insolubility in hot water plus a broad range of solubility in many organic solvents.
The properties of EHEC are quite dependent upon the relative amounts of ethyl and hydroxyethyl substitution. By varying the ratio of substituents the gelation temperature the gel characteristics the solubility properties in different solvents and the surfactancy can be modified. Increasing the amount of ethyl substitution increases the solubility in organic media and the tendency to form a firm gel while increasing the hydroxyethyl substitution improves the water solubility reduces the tendency to from a gel on heating and improves the brine tolerance of the polymer in various salt solutions.
Methylcellulose was first produced commercially in the United States in 1938 by The Dow Chemical Co. under the registered trademark of Methocel. Hydroxypropylmethylcellulose achieved commerical significance in the early fifties. In additon to The Dow Chemical Co. other suppliers of these products are Shin Etsu Chemical Products Ltd. (Metolose) of Japan British Celanese Ltd. (Celacol) of Great Britain and Kalle & Co. A G. (Tylose) Henkel and Cie GmbH (Culminal) and Wolff A. G. of Germany.
The worldwide capacity of MC and HPMC in 1979 is estimated to be about 159 million pounds per year and is growing. Water soluble ethylhydroxyethylcellulose is produced by Berol Kemi AB (formerly Modokemi AB) of Sweden.
The relative amounts of methyl and hydroxypropyl substitution are controlled by the weight ratio and concentration of sodium hydroxide and the weight ratios of methyl chloride and propylene oxide per unit weight of cellulose.
EHEC is prepared by reacting dissolving grade wood pulp with aqueous sodium hydroxide and then with ethyl chloride and ethylene oxide as schematically illustrated below
The amount of ethylation is controlled by the amount of caustic used in the formation of alkali cellulose and the amount of hydroxyethylation is controlled mainly by the amount of ethylene oxide added to the reactor. The addition is stepwise since ethylene oxide is far more reactive than ethylene chloride and hence reacts with the cellulose first.
There are three main steps used in the manufacture of MC and HPMC.
Preparation of Alkali Cellulose Alkali cellulose is prepared by contacting cellulose and 35 to 60% aqueous caustic according to several procedures that include dipping a cellulose sheet in a caustic solution spraying the caustic onto agitated cellulose flock slurrying the cellulose in aqueous caustic and removing the excess or mixing the cellulose and aqueous caustic in an inert diluent.
Viscosity control of the final product is obtained by choice of pulp by aging the alkali cellulose in warm air and by controlling the amount of oxygen left in the reactor during methylation. For high viscosity products the higher molecular weight cotton linters are used with minimum aging. Since alkali cellulose is susceptible to oxidative degradation exposing it to air for varying time peroids is an effective method for viscosity reduction.
Reaction The alkali cellulose methyl chloride and (if required propylene oxide are loaded into a jacketed nickel clad agitated vessel and heated under controlled conditions to a maximum pressure of 1.38 MPa (200 psig.) The heat of reaction is removed by condensation of the solvents. In additon to controlling the substitution levels variations in the amounts of methyl chloride and changes in the reaction profile will affect the properties of the final product.
Purification Since MC and HPMC are insoluble in hot water the reaction by products are removed by slurrying the crude product in water heated to above 900C (1940F) and then filtering. The purified wet product in then dried ground to > 95% through 40 mesh screen and commonly packaged in 22.68 kg (50 lb) bags.
Toxicity and Handling
Commercial MC and HPMC products have been used by the food pharmaceutical and cosmetic industries for many years. They are odorless. Tasteless powders and are considered to be physiologically insert.
MC products are listed in the United States Pharmacopeia XIX and Food Chemicals Codex and are listed by the FDA as Generally Recognized as Safe (GRAS).
HPMC compounds whose methoxyl substitution ranges from 19 to 30% and hydroxypropyl substitution ranges from 4 to 12% are also listed in the United States Pharmacopeia XIX and Food Chemicals Codex. Both products can meet the requirements of Food Additive Regulations 182.1480 and 172.874 as a miscellaneous and/or general purpose food additive for nonstandardized foods.
While a gross exposure to MC or HPMC can conceivably cause temporary mechanical irritation to skin and eyes exposure to normal amounts presents no significant health hazards from either contact or inhalation.
In storage good housekeeping is suggested to prevent dusts from building up to possibly explosive levels. All the cellulose ethers are organic materials that will burn under the right conditions of heat and oxygen supply. Fires can be extinguished by conventional means. Gross powder spills should be swept up to avoid accidents caused by slippery floors or equipment and the trace residual product can be flushed to a sewer. These products showed no biochemical oxygen demand (BOD) with the standard 5 day test. However radioassay tests with activated sludge showed breakdown over a 15 to 20 day period. These products should provide no ecological hazard. The products may be disposed of by either landfill or incineration.
When the viscosity is known at one concentration the viscosity can be calculated for any other concentration by using Eq. (2) to first calculate K for the sample at the concentration for which the viscosity is known and then using Eq. (2) again to calculate the viscosity at the new concentration knowing the value of k.
EHEC may be dissolved in cold water to yield clear smooth solutions. Commercial products range in viscosity from 0.050 to 12.000Pa.s (50 to 12 000 cP) at 2% concentration. The solutions are pseudoplastic in that the apparent viscosity decreases with increasing rate of shear the solutions are not thixotropic unless they are gelled. (see fig. 1)
Rheology Solutions of MC and HPMC generally show pseudoplastic nonthixotropic flow properties at 200C (680F) that is not a function of substitution within the range of available commercial products and whose deviation from Newtonian character increases with increasing molecular weight (figs. 2 and 3). Dilute solutions of low viscosity products (fig. 4) do closely approach Newtonian flow but increasing the concentration of the gum to over 5% may give a solution showing same thixotropy due to weak chain to chain interactions. Since flow properties are dependent on the molecular weight and the molecular weight distribution of the polymer a blend of high and low molecular weight polymers can have different flow properties than a polymer having the same solution viscosity as the blend but having a narrow molecular weight distribution (fig. 5). This effect is generally not important for dilute solutions of higher viscosity materials (fig. 6) but can be significant when applied to solutions of over 5% of the low viscosity derivatives.
Heating a solution of MC or HPMC shows the normal effect of lower viscosity until the gelation temperature is reached at that point the viscosity of the solution increases rapidly and highly thixotropic flow is observed.
The normal effect of temperature in the range of 0 to 450C (32 to 1130F) is roughly a 3% reduction in viscosity for every degree celsius increase in the temperature of the solution (when applied to aqueous solutions containing no added solutes and showing no evidence of gelation).
Thermal Gelation MC and HPMC solutions show the unusual property of forming a structured gel when heated.
In solution these polymers exist as aggregates of long colloidal molecules. These molecules are highly hydrated with the solvent water in layers that are held through hydrogen bonding thereby giving the chains some lubricity and smooth flow. As the temperature is raised the hydrogen bonding between the water molecules weakens and the interactions between chains become significant eventually leading to the formation of a structured gel. Unlike many chemical gels those made from MC and HPMC are primarily a result of phase separation and are susceptible to shear thinning (a mechanical breaking up of the gel without affecting the molecular weight). With cooling this process is reversible and the gel reverts back to a solution whose flow properties are not changed.
The gelation temperature is dependent on the relative amounts of methyl and hydroxypropyl substitution and may be used as an indication of the relative hydrophilicity of the derivative. In general the more highly substituted derivatives have lower gelation temperatures and will be less compatible with added solutes or electrolytes. The gelation temperature of products currently produced varies from about 50 to 850C (122 to 1850F) with the resultant gels ranging from firm to rather mushy in consistency (when determined by heating a 2% solution of the gum in pure water). The gelation temperature of a product is affected by the concentration of the gum and more importantly by other dissolved solutes. Presence of salts (Table 3) will lower the gel point addition of ethanol or propylene glycol can raise the gel point as much as 200C (36OF).
Upon reaching the thermal gelation temperature EHEC will separate out of solution as either a floc or a gel depending upon molecular weight and the concentration (see Table 4).
Surface Activity MC and HPMC reduce the surface tension and interfacial tension of aqueous sustems to values of 41 to 55 dyn/cm and 18 to 28 dyn/cm respectively (depending on chemical structure) thereby functioning as moderate emulsifiers for two phase mixtures. Since they are polymeric materials they are active surfactants at very low use levels ranging from 0.001 to 1.0%. Their status as approved additives in foods makes them useful as edible surfactants. EHEC also behaves as a moderate surfactant lowering the surface tension of water to 47 to 52 dyn/cm.
Moderate foaming is usually encountered. This can be controlled if desired by use of commercially available defoamers Polyglycol P 1200 (The Dow Chemical Co.) Antifoam A AF B or FG (Dow corning corp.) Nopco KFS (Nopco Chemical Co.) or tril n butylphosphates.
Sodium Carboxy Methyl Cellulose
Sodium carboxy methyl cellulose is a water soluble anionic linear polymer. It is universally known as CMC and will sometimes be so designated here. In the food pharmaceutical and cosmetic industries the highly purified types required are referred to as cellulose gum. The United States Food and Drug Administration (FDA) has defined cellulose gum (see section on toxicological properties) also the Food Chemicals Codex and the Food and Agriculture Organization (FAO) of the United Nations have established specifications for identity and purity of sodium carboxymethylcellulose for food uses worldwide.
Purified sodium carboxy methyl cellulose is a white to buff colored tasteless odorless free flowing powder. Less purified grades contain the reaction salts (sodium chloride and sodium glycolate) and can be off white to a light brown for the low assay types (50% purity).
Sodium carboxy methyl cellulose is probably used in more varied applications worldwide than any other water soluble polymer known today. Applications vary from the large worldwide detergent use to the specialized barium sulfate suspension for medical diagnosis.
Worldwide applications of CMC in order of size of estimated end use are given in Table 1. These estimated usages demonstrate the versatility throughout the world for this modified natural long chain water soluble polymer.
It is estimated that over 250 types of sodium carboxymethylcellulose are manufactured throughout the world by over 50 producers with outputs ranging from as little as 200 metric tons to over 35 000 metric tons per year.
The growth of CMC was accelerated by the world conflict in the early 1940s when fatty acids usage was drastically shifted from civilian soap manufacture to wartime manufacture of explosives. Even though CMC was developed shortly after World War I as a possible replacement for some gelatin uses the major growth in the use of CMC began after it was discovered that it improved the efficiency of synthetic detergents. Usage during the early 1940s was primarily for detergent systems although many new applications were developed on a laboratory scale where control of water movement was important.
With the end of the world conflict in 1945 and with the huge demand for consumer products CMC backed with several years of laboratory studies began finding uses in all types of areas requiring water control in systems with various levels of soluble and insoluble solids.
In the United States a landmark in the growth of purified sodium carboxymethylcellulose (cellulose gum) was the approval by the FDA for its use as an intentional food additive (see section on toxicological properties for details). Following this was the definition in the United States Pharmacopoeia for subsequent use in pharmaceutical applications.
Cellulose is a linear polymer of anhydroglucose units. Each anhydroglucose unit contains three hydroxyl groups.
The extent of the reaction of cellulose hydroxyls to form a derivative is called the degree of substitution (DS) and is defined as the average number of the three hydroxyl groups in the anhydroglucose unit which have reacted. Thus if only one of the three hydroxyl groups has been carboxymethylated the DS is 1.0. Commercial products have DS values ranging from 0.4 to about 1.4. The most common grade has a DS of 0.7 to 0.8 and if the DS is not specifically mentioned it can be assumed to be in this range. CMC is commercially available in several different viscosity grades ranging from 4.5 Pa.s (4500 cP) in 1% solution to 0.010 Pa.s (10 cP) in 2% solution. The various viscosity grades correspond to products having molecular weights from about 1 000 000 to 40 000. Table 2 shows that 19 different DS viscosity combinations are available from one producer.
CMC is a salt of a carboxylic acid having approximately the same acid strength as acetic acid. The pK varies somewhat with degree of substitution. The pure commercial product of DS 0.8 has a pK value of 4.4 the corresponding value of K the ionization constant is 4 × 10–5. A dilute solution of such a product has a pH of about 7 and has over 99% of its carboxylic acid groups in the sodium salt form and very few in the free acid form.
CMC forms soluble salts with alkali metal and ammonium ions. Calcium ion present in concentrations normally found in hard water prevents CMC from developing its full viscosity and thus its dispersions are hazy. At much higher concentrations calcium ions precipitate CMC from solution.
Magnesium and ferrous ion have a similar effect on CMC dispersions. Heavy metal ions like silver barium chromium lead and zirconium precipitate CMC from solution. Quaternary salts attached to a long hydrocarbon chain such as dimethylbenzylcetylammonium chloride also precipitate CMC from solution.
CMC is precipitated from solution by the polyvalent cations Al3+ Cr3+ or Fe3+. If the ion concentration is carefully controlled for example by the presence of a chelating agent such as citric acid it is possible to form more viscous solutions soft gels or very rigid gels. In these instances the polyvalent ion functions as a crosslinking agent.
CMC reacts with certain proteins. For example soy protein which is insoluble in its isoelectric range can be solubilized by CMC. Thus the solubility can be extended over a wider pH range. CMC has a similar solubilizing effect on casein but in the case of gelatin which is a more soluble protein the reaction with CMC manifests itself as a rise in solution viscosity.
Like all polymers CMC may be salted out of solution. However CMC being a very hydrophilic polymer is more tolerant of alkali metal salts than many other water soluble polymers. Its salts compatibility is much greater if the salt is dissolved in the CMC solution than if the CMC is dissolved in the salt solution (see Figure 1). Such behavior relates to the fact that CMC of DS 0.7 is aggregated in solution. This is discussed later in connection with its rheological properties.
Commercial grades of CMC have most of their carboxyl groups in the sodium salt form. Such products may be converted into the free acid form e.g. by passing a solution through a suitable ion exchange resin. If the resulting free acid is freed from water by drying a film or precipitating with alcohol and drying the product is no longer water soluble. It may be dissolved however in aqueous NaOH that is by re forming a soluble salt.
The general physical properties of CMC are summarized in Table 3. Other physical properties follow.
Equilibrium Moisture Content CMC is a very hydrophilic polymer whose equilibrium moisture increases with DS. Figure 2 gives the equilibrium moisture content for products of DS 0.4 0.7 and 1.2 at different humidities. These data were obtained on dry commercial samples by conditioning the powder to constant weight at 25°C.
Molecular Weights The molecular weights shown in Table 4 were calculated from intrinsic viscosity measurements in 0.1% NaCl at 25°C using the relationship [M] = 2.9 × 10–4 M0.78 where M is the weight average molecular weight.
Solubility The only good common solvent for CMC is water. The degree of dispersion in water varies with the DS and the molecular weight. CMC with a DS of 0.7 may be dissolved in glycerin particularly in the presence of a slight amount of water by heating with good agitation. Aqueous solutions of CMC will tolerate considerable quantities of water miscible organic solvents such as methanol ethanol and acetone. For example a 1% solution of the high viscosity grade will tolerate 1.6 volumes of ethanol per volume of CMC solution before it becomes hazy and precipitates. Low viscosity grades will tolerate as much as 3.5 volumes. Aqueous solutions of CMC will tolerate large amounts of alkali metal salts and small amounts of calcium and magnesium salts. Heavy metals and multivalent salts precipitate CMC as discussed earlier under Chemical Nature.
Film Properties Table 5 gives mechanical properties of 0.508 mm (2 mil) films containing about 18% moisture for three different viscosity grades of CMC with a DS of 0.7. It is evident that the strength and flexibility are greater for the types which have high viscosities or molecular weights. Films may be insolubilized by crosslinking at the hydroxyl groups using suitable water soluble resins such as Hercules Kymene 917 and Kymene 754 Resin or Aerotex M 3. The crosslinks are formed by reaction of cellulosic hydroxyls with the aldehyde functionality of the resins. A film is cast from an aqueous solution of the resin and CMC. Upon drying and further curing the film becomes insoluble. The degree of insolubilization depends on the extent of the curing treatment. Dry CMC films may be insolubilized by treatment with aqueous solutions of aluminum salts.
The manufacture of CMC involves treatment of cellulose with aqueous sodium hydroxide followed by reaction with sodium chloroacetate
Cellulose is a fibrous solid. Chemical cellulose which is used for the manufacture of CMC is derived from cotton linters or wood pulp. To obtain uniform reaction it is essential that all the fibers be wetted out with the aqueous NaOH. One process for accomplishing this is to steep sheeted cellulose in aqueous NaOH and then press out the excess. The sheets are then shredded and the sodium chloroacetate is added. Reactions are generally conducted at 50 to 70°C. In some cases a greater amount of NaOH is added and the monochloroacetic acid is added as such the sodium salt being formed in the presence of the cellulose.
In an alternate process the steeping and pressing steps are eliminated by conducting the reaction in the presence of an inert water miscible diluent such as tertiary butyl alcohol or isopropanol. At the end of the reaction the excess alkali is neutralized and the crude product which contains sodium chloride and sodium glycolate is purified or partially purified (see Table 6). There are many variations of these processes depending on the DS level and the quality of the product desired.
Water soluble cellulose derivatives are all subject to microbiological attack under certain conditions. The magnitude of the microbiological degradation is influenced by a number of factors which include contaminants present temperature pH of system oxygen available and concentration. The first sign of biological degradation is usually loss of viscosity (i.e. chain length). This loss can be rapid under extreme conditions or very slow under much less severe conditions. Biological attack is greater in solution systems than on the dry form of sodium carboxymethylcellulose. Usually the moisture content of sodium carboxymethylcellulose is from 5 to 10% and biological degradation is generally not severe under normal dry storage conditions at this moisture level.
Manufacturing conditions and subsequent packaging systems can greatly influence the microbiological stability of the final product. In some commercial manufacturing systems using solvents the final packaged sodium carboxymethylcellulose is essentially aseptic. In the dry process production of CMC there is a greater possibility of residual biological contamination. Generally however the caustic present in the system necessary for the alkali cellulose stage is detrimental to any microorganisms present.
Packaging environment is also extremely important. Clean containers and air free of microbiological organisms are necessary in the packaging of CMC. Generally speaking the product as produced and packaged is relatively free of microbiological organisms which would promote degradation of the cellulose chain. It has been found in actual application that most biological organisms causing CMC degradation have been introduced from outside sources (other than in the CMC). Each manufacturer of sodium CMC usually runs periodic bacteriological examinations. An example of these analyses is as follows
Other specific bacteriological testing has been done on the example purified sodium carboxymethylcellulose testing for the presence of coliforms thermophilic anaerobic spores pathogenic staphylococci beta hemolytic streptococci Salmonella species and Pseudomonas aeruginosa.
It is stressed that most individual manufacturers of CMC throughout the world have had bacteriological examination of their product and will have data comparable to the preceding. Varying manufacturing processes (i.e. different solvents and dry process techniques) will yield different biological analyses but generally speaking most known processing does not promote or support bacterial growth.
The major bacteriological problems are generally caused when the sodium carboxymethylcellulose is used in solution. Contaminating spores can be introduced in the system from influent water used for solution preparation as well as from the air surrounding the mixing and makeup vessels. This is especially true in warm humid climates which are supportive of bacteriological spores. Care should always be taken in handling containers of sodium carboxymethylcellulose that are stored in locations exposed to warm humid air.
Introduction of bacteriological contamination has been observed in plant operations where the vessels piping pumps etc. have not been cleaned after each use. In many fluid handling systems the possibility exists for areas which could retain residual product. In these areas under the proper conditions microbiological growth can take place at a rapid rate thus contaminating subsequent batches pumped or handled in the same system. Industrial experience has shown that this is a major source of microbiological contamination. Simple clean out measures in any solution handling system where CMC is used are all that are necessary to prevent subsequent contamination of other batches of product.
In the United States toxicological information on sodium carboxymethylcellulose has primarily been developed on the food additive grade or cellulose gum which is of 99.5% purity. Extensive testing has been performed on this purified grade of sodium carboxymethylcellulose to assess its safety in foods as an additive. Details on testing and results follow. However a definition of food grade sodium carboxymethylcellulose is necessary to assure safe use. The United States Food and Drug Administration defines cellulose gum as the sodium salt of carboxymethylcellulose not less than 99.5% on a dry weight basis with a maximum substitution of 0.95 carboxymethyl groups per anhydroglucose unit and with a minimum viscosity of 0.025 Pa.s (25 cP) in a 2% (dry weight) aqueous solution at 25°C.
Sodium carboxymethylcellulose (cellulose gum) is classified under Substances That Are Generally Recognized As Safe (GRAS) by Title 21 Section 182.1745 (formerly 121.101) of the Code of Federal Regulations (U.S.A.).