|
Originally Published IVD Technology March 2001 Manufacturing high-quality gold solUnderstanding key engineering aspects of the production of colloidal gold can optimize the quality and stability of gold labeling components. Basab Chaudhuri and Syamal Raychaudhuri
A recent article nicely explained the place of gold in the development
of rapid diagnostic tests.4 Table
I reproduces from that article a useful comparison of the
characteristics of labels commonly employed in rapid tests. A
significant topic not mentioned in the piece, and deserving discussion,
is the role of various process parameters in determining the quality
of the gold suspension. The first step toward manufacturing a consistent gold-protein conjugate is to make a gold sol having particles of proper size and dimension. Basically, colloidal gold sols consist of small granules of this transition metal in a stable, uniform dispersion. Most preparations of colloidal gold are made up of particles varying in diameter from about 5 to around 150 nm. For the development of diagnostic assays that make use of gold conjugates, typical particle sizes in the gold sol range between 20 and 40 nm. Since these are very small particles, the surface area of the gold in the sol is remarkably high. This means that production of colloidal gold sol involves the creation of a large surface area having a very high surface energy. Any colloidal suspension with high surface energy can lose its stability if proper operating conditions are not maintained during its production. This article discusses some process engineering aspects of gold sol manufacture that have considerable influence over the quality and stability of the suspension. It highlights the physical, rather than chemical, factors that play important roles in gold sol production. But first, a look at process chemistry is in order. Basic Chemistry A variety of chemical methods can be employed to produce monodisperse colloidal gold suspensions. However, three procedures have become the most common for making particles that fall into predictable size ranges. In all three processes, tetrachloroauric acid (HAuCl4) in a 1% aqueous solution is reduced by an agent in order to produce spheroidal gold particles. The greater the power and concentration of the reducing agent, generally, the smaller the resultant gold particles in the suspension (see Table II).
To create large-particle colloidal gold dispersions, an aqueous solution of tetrachloroauric acid is treated with trisodium citrate in aqueous solution. This results in particles sized 15–150 nm, the final range depending on the concentration of the citrate used in the reduction process. Medium-sized gold particles with diameters between 6 and 15 nm and an average size of 12 nm are formed by reducing the tetrachloroauric acid solution with an aqueous sodium ascorbate solution. The smallest particles, measuring less than 5 nm in diameter, are produced by reduction with either white or yellow phosphorus in diethyl ether.3,5 The following discussion focuses mainly on the process-development aspects of making colloidal gold with 20- to 40-nm particles. Gold sols with particles in that range are the most suitable candidates for use in developing rapid diagnostic tests. It can be seen from the foregoing that the starting raw materials for production of these gold sols are an aqueous solution of tetrachloroauric acid and an aqueous solution of trisodium citrate, which are brought together to create a chemical reaction. As described below, such chemical reactions are broadly classified as either homogeneous or heterogeneous. Homogeneous Reaction. In a homogeneous reaction, all the reactants are miscible and form a homogeneous solution. The products that are formed from the reaction are also soluble; therefore, there is no phase separation at any time during the course of reaction. The rate of such a reaction depends on the concentration of the reactants and on the operating temperature. The temperature influences the rate constant of the reaction. Usually, a 10° rise in the operating temperature enhances the rate of reaction by a factor of two. This effect of operating temperature on the rate of reaction is exploited in order to increase reaction speed or achieve greater production throughput with a single reactor. For irreversible reactions—reactions that produce almost complete conversion of reactants to products—the upper temperature limit is established by the highest operating temperature of the material from which the reactor is constructed. The effect of temperature on reversible reactions is more complicated, because both the reaction rate and the equilibrium conversion are functions of temperature and both need to be taken into consideration. For a reversible endothermic reaction, the rate of reaction and the equilibrium conversion increase with an increase in temperature; therefore, the highest possible temperature is suitable for large-scale production. For a reversible exothermic reaction, however, the equilibrium conversion decreases while the rate increases with an increase in temperature. These are opposing effects, and the reaction is thus conducted at varying temperatures so as to reach the optimum rate of reaction and achieve the best conversion of reactants into products. The transport factors—heat transfer, mass transfer, and so on—do not play major roles in homogeneous reaction kinetics. Heterogeneous Reaction. In a heterogeneous reaction, more than one phase is present. The reaction might be gas-liquid, liquid-liquid, gas-solid, liquid-solid, gas-liquid-solid, or liquid-liquid-solid, or display some other progression of phases. Any solid involved could be a catalyst or a reactant. The relatively simple rules for controlling homogeneous reactions are not applicable to the control of heterogeneous reactions. In addition to concentrations and temperature, the physical shift of reactant from one phase to another assumes great importance in heterogeneous reactions. The chemical reaction between an aqueous solution of tetrachloroauric acid and an aqueous solution of trisodium citrate is interesting in that the reaction begins as a homogeneous one, but then, within a minute, the reaction mixture becomes heterogeneous. This phase transition from homogeneous to heterogeneous occurs very rapidly, making the sol-manufacturing process difficult to monitor or control effectively. Moreover, the reaction is completed so quickly that operators do not have much time to take any corrective action necessary to ensure reproducible product. Producing Gold Colloids Before the addition of the reducing agent, 100% gold ions exist in solution. Immediately after the reducing agent is added, gold atoms start to form in the solution, and their concentration rises rapidly until the solution reaches supersaturation. Aggregation subsequently occurs, in a process called nucleation. Central icosahedral gold cores of 11 atoms are formed at nucleation sites. The formation of nucleation sites, in response to the supersaturation of gold atoms in solution, occurs very quickly. Once it is achieved, the remaining dissolved gold atoms continue to bind to the nucleation sites under an energy-reducing gradient until all atoms are removed from solution. The number of nuclei formed initially determines how many particles finally grow in solution. At a fixed concentration of tetrachloroauric acid in solution, as the concentration of the reducing agent is increased the number of nuclei that form grows larger. The more nuclei, the smaller the gold particles produced. Finding the optimal concentration of the citrate in solution is therefore an important, even crucial, task. If manufacturing conditions are optimized, all nucleation sites will be formed instantaneously and simultaneously, resulting in formation of final gold particles of exactly the same size (monodisperse gold). This is indeed difficult to achieve. Most manufacturing methods fail to accommodate this ideal and generate irreproducible gold (gold inconsistent from batch to batch) that gives unstable gold conjugates in most situations. Gold colloids are composed of an internal core of pure gold that is surrounded by a surface layer of adsorbed AuCl–2 ions. These negatively charged ions confer a negative charge to the colloidal gold and thus, through electrostatic repulsion, prevent particle aggregation. All colloidal gold suspensions are sensitive to electrolytes. Electrolytes compress the ionic double layer and thereby reduce electrostatic repulsion. This destabilizing effect results in particle aggregation, which is accompanied by a color change and eventual sedimentation of the gold. The detrimental effect of chloride, bromide, and iodide electrolytes on the stability of the gold colloid is greatest for chlorides and least with iodides. All gold colloids display a single absorption peak in the visible range between 510 and 550 nm. With increasing particle size, the absorption maximum shifts to a longer wavelength, while the width of the absorption spectra relates to the size range. The smallest gold colloids (2–5 nm) are yellow-orange, midrange particles (10–20 nm) are wine red, and larger particles (30–64 nm) are blue-green. Smaller gold particles are basically spherical, while particles in the range of 30–80 nm show more shape eccentricity related to the ratio of major to minor axes. Researchers have observed several factors that affect the quality and stability of the gold colloid. An important consideration leading to the preparation of stable gold colloids is employment of thoroughly cleaned glass apparatus, 0.2-µm-filtered solutions, and, ideally, triple-glass-distilled water.5 The use of nanopure water is recommended. These precautions suggest the adverse effect that even trace contaminants have on the preparation of colloidal gold. Although the use of siliconized glassware is often recommended, good results have consistently been obtained without any special glassware. The effect of the order of reagent addition—that is, adding citrate solution to the tetrachloroauric acid solution or vice versa—on the quality of the gold colloid formed has been noted by researchers.6 However, no clear indication of how addition order might relate to methods of manufacturing colloidal gold suspensions reproducibly has been given. Researchers have not explicated the role of mixing in the formation of the suspension, nor have they mentioned the negative impact of the use of a stir bar (for laboratory-scale preparation) in a magnetically agitated system on the quality and stability of the gold sol. It must be kept in mind that, in a large-scale operation, it is not only the chemistry of the process that is important, but also its perhaps seemingly insignificant physical parameters. Small changes in process conditions can so adversely affect the quality of the product that its utility to end-users will be minimal. Batch or Continuous Process? Generally, manufacturing processes can be run either batchwise or in continuous mode. Certain considerations dictate the best choice. The scale of operation is an important factor. Small-scale production calls for batch operation. The reactor used in such a case is called a batch reactor. Reactants are added to the batch reactor at a suitable temperature, the reaction proceeds, and then, at the end of the batch time, the reactor's contents are removed. The reaction products are subsequently recovered by means of a separation process. Unconverted reactants can be reused in certain cases. Whenever the unconverted reactants are discarded, they must be disposed of in accordance with pertinent environmental and safety rules. In a batch reactor, the concentrations of reactants and products change continuously with time. Thus, the reaction can be tracked by noting carefully the fall in reactant concentration or the rise in product concentration as a function of time. The continuous mode of production is used for high-volume manufacture. In a continuous process, the reactants are fed into the reactor steadily and the products form and come out continuously. The flow pattern of fluid in a continuous reactor will take one of three forms. A mixed flow pattern is characterized by the reactor contents being completely mixed and the exit concentration being equal to the reactor concentration. In a plug-flow reactor, the concentrations of reactants and products change progressively as the materials pass through the reactor. There is no mixing of fluid in the longitudinal direction, although the radial mixing is complete. A flow pattern between the mixed- and plug-flow configurations is possible. In such a case, the fluid in the reactor is partially rather than completely mixed. Sound experimental techniques are available for determining the specific flow pattern in an unknown vessel, and modes of describing the vessel characteristics have been established. Without proper characterization of fluid flow within a reactor, it is practically impossible to predict the behavior of the reaction taking place there. Typical production volumes for colloidal gold are in the range of 1 to 100 L. Batch processing is appropriate for such production. Manufacturers producing gold suspension in the range of 10 to 100 L (production volume depends on the order size and the availability of suitable reactor) sell it to customers for subsequent conjugation with proteins. Typically, for a 100-L batch, the power consumption for stirring the liquid contents will be substantially large. The reactor needs an adequate piping arrangement for pumping reactants, washing solutions, etc., into the reactor. After every batch of production, the reactor must be cleaned thoroughly with a sodium bicarbonate and detergent solution, distilled water, and then with some volatile organic solvent, such as acetone. A further wash with nanopure water is recommended. Many diagnostic companies produce gold suspension in the range of 2 to 5 L for their in-house use (captive consumption). For such small-volume production, power consumed in stirring the reactor contents can be expected to be minimal, and it is easy to clean the reactor thoroughly after every batch by dismantling the entire assembly. Complicated piping arrangements that are necessary for large-batch production or in continuous processes do not figure in small-batch processes, making cleaning simple. The important parts of the batch reactor used for making colloidal gold are the reaction vessel, the agitation or stirring system, and a constant-temperature bath that keeps the reactor contents at a uniform and suitable operating temperature throughout the reaction. The reaction assembly is easy to set up; the only precaution that needs to be taken is that the assembly components must be cleaned scrupulously. It is best that all glassware be autoclaved before each use. Factors Affecting the Quality of the Final Gold Suspension A variety of physical parameters affect the quality of the final gold suspension that is produced by the reaction of aqueous tetrachloroauric acid with an aqueous solution of trisodium citrate. Key factors worthy of consideration are:
Concentration of Reactants. The rate of any homogeneous chemical reaction depends on the concentration of the reactants and the operating temperature. A low reactant concentration will result in low rates. A high concentration is therefore desirable for realizing high rates. Too high a concentration might yield other problems, however, particularly for competing reactions. In such cases, the desired product might not form in sufficient quantity and the reaction might produce a large amount of byproducts. Only one reaction is involved when tetrachloroauric acid and trisodium citrate are combined, but inadequate reactant concentration would result in gold particles of undesirable size and a broad distribution of particle sizes. The procedure developed by Frens is most commonly used to produce 40-nm gold particles.7 In accordance with this procedure, to 50 ml of tetrachloroauric acid in 0.01% solution (weight to volume) that is at a boil, 0.5 ml of a 1% solution of trisodium citrate is added. The solution initially has a gray color which changes to lavender and then, with continued boiling, after 1 to 3 minutes develops a red hue. The resulting particle size is 41 nm. Once the colloid is formed, neither prolonged heating nor further addition of the citrate solution will produce any change in particle diameter. The use of proportionally larger reaction volumes is accompanied by an increase of some 20% in the final particle size. A 20% difference in the amount of tetrachloroauric acid or trisodium citrate used has been observed not to affect the particle size substantially. Rates of initial nuclei formation are practically uniform throughout this range of concentrations. Mixing of Reactants. This is perhaps the most crucial physical process parameter. The reactants must be well mixed in order for nucleation to occur. Adequate mixing results in uniform concentration and temperature in every part of the reactor. If concentration varies from location to location, the rates of reaction in different places will be different, too. All chemical reactions are accompanied by either heat generation or heat absorption, which gives rise to either an increase or a decrease in temperature from the desired value. Different temperatures within the reactor cause heat- and mass-transfer gradients. With an intrinsically slow reaction, how the reactants are mixed is not going to be a cause of substantial distortion in the product. But in the gold-making process the reaction occurs in seconds, and the reactants must be brought to uniform concentration before it happens. A challenge of rapid mixing is therefore encountered here; special arrangement must be made for agitation of the liquid contents of the reactor. In small-scale processes a small stir bar is used to agitate the liquid in the reactor. The reactor is placed on a magnetic stirrer, and, in order to promote fast mixing, the stirrer is rotated at high speed. Careful observation reveals that under such a condition a vortex is formed in the reactor (see Figure 1). The formation of this vortex is counterproductive to proper mixing because liquid revolves in a narrow zone and the circulation current does not spread throughout the reactor. The larger the diameter of the reactor, the more pronounced will be such a deleterious effect. In a small reactor having a volume of, say, 500 ml, the quality of the colloidal gold produced is not too bad even when a vortex has formed. But under otherwise uniform conditions, when the scale of production is increased to 4 L, the effect of bad mixing is evident. The gold produced will have poor characteristics; particles become larger and display greater eccentricities. A magnetically stirred system is not recommended for making gold even in small-scale processes, simply because the mixing characteristics in the liquid undergo batch-to-batch variation and it is hard to achieve reproducible process conditions. Use of a mechanically agitated reactor is to be preferred. In such a reactor an agitator fitted to a motor imparts motion to the liquid and effects mixing. The reactor should be equipped with baffles in order to avoid any formation of a vortex. Baffles are basically rods attached to the wall of the vessel that promote turbulence in the reactor and stifle the tendency toward vortex creation. Achievement of reproducible mixing in the reactor depends on several important factors, including the diameter of the reactor, the diameter of the stirrer, the thickness of the baffles, and the speed of rotation or agitation of the reactor. In addition to these, the properties of the liquid contents are important. The higher the viscosity of the liquid being stirred, the more power must be applied for good mixing. Liquid viscosity is a strong function of temperature; if the process allows an increase in operating temperature, then making such an adjustment can reduce both viscosity and power consumption. For a given liquid at a given temperature, if the reactor diameter, stirrer diameter, baffle size, and speed of agitation are constant, the mixing characteristics produced in the reactor also will be reasonably constant. An irreproducible equipment setup is more likely than a constant configuration to result in an irreproducible product. Therefore, the reactor needs to be designed correctly and operated uniformly. Simple experiments, commonly called tracer studies, can be devised to find the mixing characteristics of a particular vessel. In these studies, a tracer, usually a known amount of dye, is injected into a known volume of water in the reactor. The concentration of the dye in water can be calculated readily when the tracer is dispersed uniformly. The time required to achieve this uniform dispersion is determined by noting the change in concentration of the dye as a function of time at different speeds of agitation and with different stirrer designs. The combination that takes the least time to reach uniform dispersion is chosen for the process. In making gold suspensions, it is first necessary to bring the two reactants to uniform concentration as soon as possible. Once this is achieved, the speed of agitation must be reduced; otherwise, the gold particles will collide with each other and form larger particles. Rapid mixing of the reactants thus is recommended, followed by slow agitation designed to impart flow to the liquid, but not much turbulence. Order of Reactant Addition. The order in which reactants are combined in the reactor is another crucial factor that determines the quality of the gold sol. Is it better to add the aqueous solution of trisodium citrate to the tetrachloroauric acid, or vice versa? There is not much apparent difference between these two modes of addition. But in reality the difference is substantial. A thought experiment may help to bring out the difference. Consider a homogeneous reaction that develops as follows: A + B = C and A + C = D. Now suppose a beaker containing only substance A to which substance B is added quantum by quantum. A will react with B to produce C, and since C will find itself in a large excess of A, D will be the predominant product to form in the beaker. On the other hand, suppose A is added quantum by quantum to a beaker containing a large excess of B. The concentration of C will rise gradually. With the continued addition of A, the concentration of C will pass through a maximum. At some point a competition will develop between B and C to react with A. This stage of competitive reactions presents a choice: If the desired product is D, then the former contacting pattern--adding B to the beaker--should be chosen. If, however, the desired product is C, then the second contacting pattern is better; substance A is added quantum by quantum, the concentration of C reaches the maximum, and then C is separated from the product mix. This is one way to engineer the chemistry so as to obtain most efficiently the product desired. The gold-making process does not involve the multiple reactions in this illustration. Still, the order of reactant addition is important. If a small quantity of trisodium citrate is introduced to the tetrachloroauric acid, it takes a long time for it to uniformly disperse in the large volume. The reactor might contain pockets of notably high or low concentration. Such a circumstance promotes unequal rates of reaction, a condition that in turn leads to unequal rates of nucleation and, hence, bad gold as the finished product. On the other hand, if the tetrachloroauric acid is quickly added to the citrate solution, the chances of pockets of varying concentration forming are rather remote. Operating Temperature. All standard textbooks prescribe adding trisodium citrate to a boiling tetrachloroauric acid solution, and continuing the boiling for about 15 minutes thereafter. The reaction usually is complete within 5 minutes, and there is no subsequent change in the particle size of the gold. How is the boiling achieved, and is it at all necessary? In small laboratories the reactor is placed on a combination magnetic stirrer and hot plate and the liquid is heated by means of the hot plate. This form of heating has been observed to be detrimental for colloidal gold. The reason is that tiny bubbles form at the bottom of the reactor, and the moment they disengage from the hot surface they briefly leave dry spots behind. Any colloidal gold particles at these dry spots become dehydrated and lose their desirable characteristics. The liquid in the reactor does not have to be boiled in order for good gold to be produced. As long as the temperature in the liquid phase is controlled at around 95°C, the gold-making reaction proceeds smoothly. Control of the temperature is what is important. Operating the reactor at a constant temperature requires the selection of a proper heating system. In order to maintain uniform temperature and avoid hot spots, hot liquid should be circulated around the reactor. The liquid could be high-molecular-weight oil. For small-scale production a thermostat employing hot oil may be used. The reactor should be placed inside the thermostat. The oil temperature can be controlled by means of a temperature controller. Liquid Head in the Reactor. The liquid head is a consideration important only in large-scale operation. Suppose a batch reactor is employed to make 100 L of gold. In such a situation the depth of liquid in the reactor is substantial. The reaction involved in making gold suspension takes place at atmospheric pressure, which is the pressure at the top of the liquid in the reactor. But at the bottom of the reactor, the pressure is the sum of the atmospheric pressure and the pressure due to the liquid head. Pressure at the bottom of the reactor is therefore higher than that of the atmosphere. Consequently, the boiling point of the liquid at the bottom of the reactor is higher than it would be at atmospheric pressure. Exposure to this high-boiling-point temperature could destroy the gold suspension. With rapid mixing, the temperature of the liquid should be the same throughout the reactor. But after nucleation, when slower agitation is necessary, an unusually high temperature gradient can develop. In addition, dry spots might form on the inner surface of the reactor. The gold suspension becomes dehydrated where it comes into contact with these dry spots. When the process is complete, suspended particulates appearing at the surface of the liquid will suggest that the quality of the gold produced is not likely to be good. Careful prior consideration of the liquid head in the reactor is therefore essential to any attempt to make gold suspension in large batches. Material of Construction. Finally, the possible effect of the materials from which the reactor and the agitation assembly are made on the quality and stability of the gold sol must be carefully considered. In small-scale batch production, small Teflon-coated stir bars are employed for agitation. The use of Teflon is supposed to ensure a clean system. However, the shear between the stir bar and the reactor generated during rotation and arising at the point of contact between them exposes the inner metallic region of the stir bar, imperfections that close observation will reveal. The colloidal gold interacts with the exposed metal at these tiny flaws, and particles form as a result. It must be kept in mind that any contaminant can destroy the gold sol. Therefore, an all-glass assembly is recommended. If steel reactors are used, they should be provided with a glass lining. The lining should be tested from time to time to ensure that no crack has developed. The material of construction of the agitator requires similar consideration. Agitators generally should be made of Teflon, which reacts adversely with virtually no chemical. Conclusion The manufacture of gold suspensions involves combining simple chemistry with some difficult process engineering. Chemists and biotechnologists need to understand the engineering considerations discussed in this article in order to make gold sols of dependable quality. An understanding of the effects of different process parameters on the quality of the gold suspension is a great help in troubleshooting as well. The problems discussed above are actual challenges encountered by the authors. It is hoped that small manufacturers heeding the lessons learned and reported here will see improvement in the quality of the gold suspensions they produce.
References 1. R Zsigmondy, Zur Erkenntnis der Kolloide (Jena, Germany, 1905). 2. WP Faulk and GM Taylor, "An Immunocolloid Method for the Electron Microscope," Immunochemistry 8 (1971): 10811983. 3. GT Hermanson, Bioconjugate Techniques (San Diego: Academic Press, 1996). 4. J Chandler, T Gurmin, and N Robinson, "The Place of Gold in Rapid Tests," IVD Technology 6, no. 2 (2000): 3749. 5. MA Hayat, ed., Colloidal Gold: Principles, Methods, and Applications, vol. 1 (San Diego: Academic Press, 1989). 6. DA Handley, "Methods for Synthesis of Colloidal Gold," in Colloidal Gold: Principles, Methods, and Applications, vol. 1, ed. MA Hayat (San Diego: Academic Press, 1989), 22. 7. G Frens, "Controlled Nucleation for the Regulation of Particle Size in Monodisperse Gold Solutions," Nature Physical Science 20 (1973): 241.
Copyright ©2001 IVD Technology |
As
early as the first decade of the twentieth century, colloidal
gold sols containing particles smaller than 10 nm were being produced
by chemical methods.1 However, these inorganic suspensions
were not applied to protein labeling until 1971, when Faulk and
Taylor invented the immunogold staining procedure.2
Since that time, the labeling of targeting molecules, especially
proteins, with gold nanoparticles has revolutionized the visualization
of cellular and tissue components by electron microscopy. The
silver enhancement method has extended the range of application
of gold labeling to include light microscopy. The electron-dense
and visually dense nature of gold labels also facilitates detection
in such techniques as blotting, flow cytometry, and hybridization
assays. Double- and triple-labeling systems involving immunogold
methods have been used successfully to detect more than one antigen
at the same time.3 
