N A N O P R O B E S E - N E W S
Vol. 7, No. 2 February 17, 2006
For more information, see the recent book, Nanobiotechnology: Concepts, Applications and Perspectives, edited by Mirkin and Niemeyer. This contains several chapters on the applications of gold particles, including contributions from ourselves and our collaborators, who describe many applications of our Nanogold® labeling technology.
Nanobiotechnology: Concepts, Applications and Perspectives - details and ordering information
In previous reports, we have described the development of DNA-based nanowires, gold-decorated nanotubes, and the use of rigid, self-assembled DNA structures as templates for deposition of gold nanoparticles in organized patterns. This science was extended recently by Zhang and group, who report in Nano Letters the templating of gold nanoparticles into periodic arrays, with a repeat distance of 38 nm, using 2-dimensional DNA nanogrids. As shown in Figure 1, the DNA nanogrid structure was generated by utilizing a family of DNA tiles that resemble a cross structure, each composed of four four-arm DNA branch junctions, designed with appropriate complementary sequences at the end of each arm, so that hybridization resulted in the assembly of the DNA into a two-dimensional array; a single unit of the cross structure had previously been shown to self-assemble into 2D nanogrids with periodic square cavities.
The sequences for the two DNA tiles were modified from that used for similar tiles constructed previously, by incorporating a 15-base polyadenine tag (A15) into the A tile as a hybridization site for the gold particle. Custom oligonucleotides were purchased and purified by denaturing PAGE. After approximate measurement of concentration using OD260nm, each tile was assembled by mixing a stoichiometric quantity of the strands involved in the tile in 1 x TAE/Mg buffer (20 mM Tris, pH 7.6, 2 mM EDTA, 12.5 mM MgCl2). The final concentration of DNA was 1.0 x 10-6 M, and the final volume was 60 microliters. The oligo mixtures were cooled slowly from 90°C to room temperature in 2 L water, placed in a styrofoam box for 16 hours to facilitate hybridization. 5nm gold nanoparticles were conjugated to 3-thiolated DNA oligos containing 15 thymine bases (T15). An excess amount (1.5 times in molar ratio) of thiolated T15 DNA was added to the solution of Au nanoparticles, and the solution was shaken overnight at room temperature. 0.125 microliters each of 1 M NaCl and 0.1 M phosphate buffer were added and the mixture shaken at low speed for a further 24 hours at room temperature, then centrifuged on an ultracentrifuge to yield a red colored solution of nanoparticles. The concentration of gold nanoparticle-DNA conjugates was calculated by measuring the UV/visible absorption of the gold at the 520 nm wavelength, and the solution was kept at 4°C before use.
Hybridization of conjugates to DNA nanogrids was achieved by spotting a microliter drop of 1 micromolar self-assembled DNA nanogrid solution onto freshly cleaved mica and adsorbing to the surface for 1 minute. The buffer salts were then rinsed away with 100 microliters of distilled water, the sample was blown dry with compressed air, and a 4 microliter volume of 0.8 x 10-6 M DNA-gold conjugate in TAE/Mg buffer was added. This was allowed to hybridize to the DNA nanogrids for 4 minutes; the surface was then rinsed in 100 microliters of water and blown dry with compressed air. The gold-decorated nanogrids were imaged by AFM in AAC mode, using the tip on the thinner and shorter cantilever of the NP-S tips.
Upper: Design of complementary DNA tiles showing hybridization sites for gold particles in red. Lower: assembly of 2-dimensional tiles, and resulting pattern of gold nanoparticle hybridization.
Interestingly, only labeling occurred only at alternate sites, as shown in above. Although the spacing was expected to allow hybridization at each site, it was believed that the steric hindrance of the coating of oligonucleotides over the gold contributed to the lower labeling. Electrostatic repulsion between the predominantly negatively charged particles and grid - as well as between adjacent particles - was also thought to contribute to the larger than expected gold separation. Significantly, the spacing remained regular, demonstrating that these interactions may be used to program the spacing of templated particles over greater distances than expected.
Nanoprobes offers products and technologies for this and for related applications. Our Nanogold® labeling reagents are available with different specific reactivities, including thiol-reactive maleimides, amine-reactive sulfo-NHS ester, nitrilotriacetic acid (NTA)-Ni(II) chelate reagents for conjugation to polyhistidine tags, and amino-functionalized particles which may be conjugated to a variety of groups using different cross-linkers. In addition, we are working to offer larger functionalized gold particles with similar reactivities so that particles with different physical, chemical and electronic properties are available.
Zhang, J.; Liu, Y.; Ke, Y., and Yan, H.: Periodic Square-Like Gold Nanoparticle Arrays Templated by Self-Assembled 2D DNA Nanogrids on a Surface. Nano Lett., 6, 248-251 (2006).
Construction of tiles:
Park, S. H.; Yin, P.; Liu, Y.; Reif, J. H.; LaBean, T. H., and Yan, H.: Programmable DNA self-assemblies for nanoscale organization of ligands and proteins. Nano Lett., 5, 729-733 (2005).
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
We have previously discussed the selection of chromatography media for the separation of Nanogold® conjugates. However, what can you do if set up your column and inject your reaction mixture, only to have the colored product crash out of solution and stick to the resin at the top of the column?
Help, my conjugate has crashed out and is stuck on the column! Can we recover it?
The place to look for a solution is in the solubility of Nanogold and its conjugates. Solvents and buffers that dissolve Nanogold, or your conjugate, effectively are more likely to re-solubilize it from column materials. If you encounter this problem, we recommend trying the following:
- Nanogold is highly soluble in dimethylsulfoxide (DMSO); dissolving it in a small quantity of DMSO before adding water will usually speed up solution and ensure an effective reaction. Because up to 20% DMSO is also well tolerated by many biological molecules, eluting the column with a buffer containing 20% DMSO may solubilize the conjugate without compromising the reactivity of the conjugate biomolecule. Collect the Nanogold-containing fractions, concentrate again, and repeat the chromatographic separation with 20% DMSO in the eluting buffer.
- Nanogold is also soluble in alcohols and alcohol-water mixtures. Isopropanol and ethanol are well-tolerated by many biological molecules are may therefore be used in up to 20% concentration in the eluting buffer to help redissolve precipitated Nanogold. Using 20% or lower concentration of organic solvent should help to ensure that the conjugate biomolecule will be intact and the conjugate will still bind to its target.
- Addition of a detergent, such as 0.1% Tween-20, may help solubilize Nanogold. Many of the non-specific interactions of Nanogold conjugates with column materials and with biological systems may be attributable to hydrophobic interactions: Tween-20 will help to break these interactions and solubilize the hydrophobic species in aqueous buffers.
- If using 20% organic solvent does not work, stronger measures may be needed. Try injecting 5% of the column volume of pure DMSO: you should see this displace the Nanogold (which will be visible as a brown stain) as the DMSO works its way down the column.
What does a good separation look like?
For an example of gel filtration, check the section on separating your conjugate in our Guide to Gold Cluster Labeling. In the example given, a Nanogold-labeled antibody is separated from excess unconjugated Nanogold. In this case, you see two bands elute: the larger Nanogold-labeled protein is eluted first, and usually appears light brown, while the unbound excess Nanogold elutes later, usually appearing darker brown. The labeling reaction mixture almost always contains some smaller molecules (cross-linkers, buffer components) which elute as a third, usually colorless peak, close to the column volume. If you are labeling a smaller molecule and use an excess of the molecule to be labeled, the Nanogold conjugate will be the first, colored peak to elute; it will be followed by a larger, less colored peak which is excess, unconjugated biomolecule.
The attempt to rescue the conjugate didn't work: now how do I clean the column?
The usual cleaning procedures that are recommended for gel filtration columns are meant for proteins and other types of biological molecules, which these columns are usually used to separate. Gold deposits, however, have different solubility properties, and therefore need different treatments. If you have persistent gold deposits after regular cleaning - particularly the purple or blue color that indicates colloidal gold particles - we suggest the following methods to remove them. Please note that these suggestions may not apply to all columns: you should not use any of these procedures if the instructions for your column caution against it.
- Use a thiol-containing compound - thiols have a high affinity for gold and break down gold deposits and remove them from the column. If you combine this with dimethylsulfoxide (DMSO), which is an excellent solvent for many gold compounds, the result is a highly effective gold removing wash.
50 or 100 mM mercaptoethylamine hydrochloride (MEA) or mercaptoethanol in 50% DMSO/water is a good first choice. Make up about 10% of the column volume of this mixture; dissolve the thiolated compound in half the final volume of deionized water, then make up to final volume with DMSO. Note that mixing DMSO and water generates heat, so let the cleaning solution cool back down to room temperature before using. After injection, run the column as slowly as you can - the longer this solution remains in contact with the gold deposits, the more effective it will be. For maximum efficacy, set it up to run overnight. We suggest leaving the detector and recorder running (slowly!) to provide a record of when the various materials are eluted and to verify that the cleaning solution has cleared and absorption has returned to baseline.
- 1 M acetic acid in DMSO may be effective for colloidal gold, particularly the red-purple material that can remain behind after repeated Nanogold separations. To remove the more stubborn purple and blue colloidal gold deposits, use the thiol wash first.
- Cleaning procedures for many columns are available on manufacturer web sites; for example, GE (Amersham Pharmacia) and Bio-Rad provide comprehensive instructions for cleaning their gel filtration columns on the Amersham Pharmacia web site:
Cleaning GE gel filtration columns
Cleaning instructions for Bio-Rad Bioscale columns
Generally, 1 M NaOH followed by water is recommended to remove proteins, and 70% ethanol, 70% formic acid or 70% acetonitrile to remove hydrophobic contaminants.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Enzyme Metallography has proven to be a very clean and sensitive method for both in situ hybridization and immunohistochemistry (IHC). Recently, Wolfgang Fritzsche, Robert Möller and co-workers have applied it successfully to the development of an electrical detection method for biochips, and presented their results in Nano Letters. Because multiple independent electrical contacts can be fabricated and identified on a microscopic scale, electrical detection offers the potential for highly multiplexed target detection in a robust, miniaturized and highly portable format. Previous studies had demonstrated that using silver-enhanced gold could be used to form the contacts, but produced significant background staining even after treatment with passivating agents such as alkanethiols or mercaptoalcohols. Because of its high sensitivity and low background, enzyme metallography offers the potential for improved detection.
A chip design with 42 measuring spots was used for the electrical detection of DNA: the measuring spots are 10 micrometer wide microstructured electrode gaps prepared on a silicon oxide substrate covered by 5 nm of titanium and 100 nm of gold using standard photolithographic processes. The chip surface was modified with (3-glycidyloxypropyl) trimethoxysilane to immobilize a 30 base pair (bp) amino-modified oligonucleotides. Different capture DNA probes were then immobilized on the chip: a complementary probe (NS150), a probe with one mismatch (NS151), a probe with three mismatches (NS153), and a noncomplementary probe (N7), each 30 bp long. Two electrode pairs on each chip were left unmodified as controls for the background signal.
Enzyme metallography as an electrical detection method. left: the enzyme metallographic process. right: design of electrical detection system: target oligonucleotide is deposited between electrodes, detected with enzymatic probe which is then "developed" with enzyme metallography to produce a conductive connection.
For hybridization of the biotin-modified target probes on the chips, the probes were dissolved in 5 x SSC + 0.1% SDS, then incubated with the chips for at least 1 hour in a humidity chamber, then washed for 5 minutes in 2 x SSC + 0.1% SDS, then washed for a further 5 minutes in 5 x SSC. After a final wash for 5 minutes in 0.2 x SSC, the chips were again dried under a nitrogen. Detection using enzyme metallography was compared with that for silver-enhanced gold. For the gold, 100 microliters of a 1:100 dilution of streptavidin-5nm gold nanoparticle in PBS pH 7.4 with 0.1% BSA was applied to each chip and incubated for 1 h at 37°C in a humidity chamber. The chips were then washed in PBS, pH 7.4, for 5 min, briefly rinsed with distilled water to remove any excess chloride ions, then silver enhanced with 300 microliters of British Biocell SEKL 15 reagent for 2 or 4 minutes. The chips were washed in distilled water to stop the reaction and dried under nitrogen. The chips were finally measured using a customized readout device; the reader was constructed around a special socket, which holds the chip during the measurement and contacts all 42 electrode gaps on our chip for a semiparallel measurement; it also includes a multiplexed ohm meter controlled by an embedded PC. The procedure was repeated until a significant drop in the measured resistance was detected.
Enzyme metallographic detection was accomplished using a streptavidin-peroxidase polymer. The substrates were incubated with a 1:1000 dilution of the streptavidin peroxidase polymer solution in PBS pH 7.4 with 0.05% Tween-20. A 100 microliter portion of this solution was applied per chip and incubated at 20°C for 1 hour. The chips were then washed 6 x 5 minutes in phosphate-buffered saline, pH 7.4, with 0.05% Tween-20 to remove any excess of bound enzyme complex, rinsed briefly with distilled water to remove any chloride ions, and dried under a stream of nitrogen. The chip was then developed with the enzyme metallographic reagent: after 5 minutes, development was stopped by washing the chips in water and then drying under nitrogen. The custom-built readout device was used to measure the resistance on the chip. A further enhancement was conducted (if necessary) using the British Biocell silver enhancement kit used for the gold nanoparticles.
Examination of the two chips using AFM revealed a significant reduction in background signal, with fewer silver deposits in the substrate areas adjacent to the spotted region; also, a highly defined separation was found between the spot region with deposited metal of about 130 nm height and the surrounding background, with no detectable metal clusters. Furthermore, the electrode regions that are situated inside the spot area also show a clearly suppressed metal deposition with the enzymatic process compared with images from the metal-catalyzed case. This lower background signal enables the enzyme-based detection scheme to be significantly more sensitive than the nanoparticle-based system: it was found to be possible to detect biotin-modified target DNA concentrations as low as 500 fM, corresponding to 50 amol of biotin-modified target DNA. A further increase in the sensitivity of the system seems feasible with further optimization of the development process. These results indicate that enzyme metallography can offer a significant improvement for electrical detection on biochips, bringing closer robust, portable, multiplexed biochip diagnostics.
Moller, R.; Powell, R. D.; Hainfeld, J. F., and Fritzsche, W.: Enzymatic control of metal deposition as key step for a low-background electrical detection for DNA chips. Nano Lett., 5, 1475-1482 (2005).
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Methods for the simultaneous detection of many intracellular and extracellular proteins within single cells are urgently needed in cell biology. Current optical methods based on fluorescence activated flow cytometry are difficult to multiplex, and an alternative approach with a great potential to fulfill this need is mass spectrometry combined with the use of "mass labels" - labels of a fixed, known molecular mass that can be used to identify the entity to which they have bound. Nanogold®, because it is a single large molecule, is ideal for this purpose because it has a consistent mass, and Ornatsky and co-workers report the results of their studies using Nanogold in the current Journal of Immunological Methods.
The authors describe a novel application of ICP-MS-linked metal-tagged immunophenotyping. Expression of intracellular oncogenic kinase BCR/Abl, myeloid cell surface antigen CD33, human stem cell factor receptor c-Kit and integrin receptor VLA-4 were investigated using model human leukemia cell lines. Antigens to which specific, non cross-reactive antibodies are available which may be distinguishably tagged can be determined simultaneously, or multiplexed. Four commercially available lanthanide tags (Au, Sm, Eu, and Tb) conjugated to different secondary antibodies were shown to enable a 4-plex assay. Results obtained by ICP-MS were compared with data from flow cytometry.
ICP-MS as an analytical detector possesses several advantages that enhance the performance of immunoassays. It has high precision, low detection limits, and a large dynamic range, both for each antigen and between antigens. It shows lower matrix effects from other sample components (contaminating proteins in the sample have no effect on elemental analysis), and lower background from plastic containers and plates (plastic containers do not cause interference on elemental detection, as they can with fluorescence). Results are independent of non-specific background and analytical response from incubation or storage times, since protein degradation does not affect analysis of an elemental tag. The multiplexing potential is very large (potentially, up to 167 isotopes; realistically, around 100 distinguishable tags), and better spectral resolution (abundance sensitivity) is obtained. Because signals from elemental tags are essentially non-overlapping, there is no need to compensate for equal intensity for each application.
Lanthanide-labeled (Eu, Tb, Sm) affinity reagents were purchased from Perkin Elmer Life Sciences (Turku, Finland): Eu-N1-anti-rabbit (DELFIAR #AD0082), Tb-N1-Streptavidin (StrAv-Tb; DELFIAR #AD0047); Sm-N1-Streptavidin (StrAv-Sm; DELFIAR# AD0049). Human monocyte cell lines MBA-1 and MBA-4 were derived from Mo7e by retroviral induction of the p210 BCR/Abl cDNA; expression of the BCR/Abl oncogenic kinase confers growth factor independence. These cell lines serve as model systems for studying human megakaryocytic leukemia and drug inhibition of the BCR/Abl signal transduction pathways. Cells were propagated in alpha-MEM, supplemented with 10% FBS and 2 mM l-glutamine, in a humidified incubator at 37°C and 5% CO2. Cells were split every 34 days and viability checked with trypan blue (90% viable). Antibodies used were anti-CD33 (mouse monoclonal), biotinylated anti-human SCFR/c-Kit antibody (CD117), biotinylated anti-human Integrin alpha4 (CD49d or VLA-4) mouse monoclonal antibody; anti-BCR antibody raised in rabbit (Cell Signaling Tech) which detects total levels of endogenous BCR and the p210 kDa BCR/Abl fusion protein. Anti-IgG1 kappa mouse, and anti-IgG1 rabbit (BD PharMingen) were used for negative controls. Antibodies were diluted in PBS/10% FBS.
Cells in logarithmic growth phase were collected by centrifugation (1200 rpm, 10 min), washed once in PBS/10% FBS and counted in a hemocytometer. The cells were then distributed into round-bottom 5 mL flow cytometry tubes in triplicates for staining. Cells were Incubated with primary antibodies on ice for 30 minutes, permeabilized for 30 minutes at room temperature, then washed and non-specific antigen sites were blocked with PBS/10% FBS for 15 minutes at room temperature; cells were then stained for intracellular antigens. After secondary antibody incubation (30 minutes on ice), cells for ICP-MS analysis were washed twice, centrifuged and pellets were dissolved in concentrated HCl (75 microliters per tube) for 10 minutes at room temperature. Nanogold anti-mouse-IgG was prepared in three different dilutions (1:10, 1:25, and 1:50), and added to the washed cell pellets for another 30 minutes on ice. All cells were washed twice with PBS/FBS and left as a pellet after the second wash. Pellets were then frozen overnight and dissolved in 34% HCl next day and run on ICP-MS. As an internal standard for ICP-MS measurements, 75 microliters of 1 ppb Ir (Iridium) solution was added to each tube. Experimental measurements were made on a commercial ICP-MS instrument ELAN DRCPlusk (Perkin Elmer SCIEX).
An immunophenotyping experiment was set up to simultaneously identify two cell membrane antigens (CD33 and c-Kit) and the intracellular proteins BCR and p210BCR/Abl. Cells were processed as described above, permeabilized for 30 minutes, blocked 15 minutes with PBS/10% FBS, then incubated with a mix of primary antibodies (anti-CD33, biotinylated-anti-c-Kit and anti-BCR) for 30 minutes at room temperature. After one wash with PBS/10% FBS, metal-conjugated reagents (Nanogold anti-mouse, streptavidin-Tb, and anti-rabbit-Eu) were applied to the cells. Comparable signals were obtained whether the secondary tags were added all at once, or separately, indicating that there is no interference from each metal-tag when used in a multiplex assay. Finally, for comparison with flow cytometry, a multiplex four-antigen identification by ICP-MS and was conducted. Live cells were stained with anti-CD33 and anti-c-Kit antibodies, permeabilized; stained with a mix of anti-BCR and Nanogold anti-mouse, and followed by streptavidin-Sm and anti-rabbit-Eu. The final incubation included anti-VLA-4, followed by StrAv-Tb for ICP-MS.
The results of these experiments provide proof-of-principle that ICP-MS can be used in the multiplexed molecular analysis of human leukemia cell lines. Both cell surface and intracellular proteins were clearly detected, and it was clearly established that four proteins could be quantitatively analyzed, proving the multiplex potential of this approach. Although multiplexing using metal-conjugated reagents is in a very early stage of research and feasibility studies, it is already apparent that more than four antigens could be accurately detected simultaneously using the ICP-MS instrument.
Ornatsky, O.; Baranov, V. I.; Bandura, D. R.; Tanner, S. D., and Dick, J.: Multiple cellular antigen detection by ICP-MS. J. Immunol. Methods., 308, 68-76. (2006).
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The run of Nano-W papers continues... in the last issue, we reported on how Kirkham and colleagues identified invasive serotype 1 pneumococcal isolates that express nonhemolytic pneumolysin, and used their insights to construct and characterize a novel nontoxic protective pneumolysin mutant which provides important advantages for vaccines, while, Chow and group used negative stain transmission electron microscopy with Nano-W, in conjunction with light scattering, to study the formation and dissociation of aggregates during the unfolding and re-folding of apomyoglobin. This month, Bukreyev and co-workers continue the streak, using negative stain electron microscopy to help show that highly effective immunity against severe systemic infections, such as those caused by hemorrhagic fever agents Ebola virus (EV), recombinant human parainfluenza virus type 3 (HPIV3) could be achieved with a single dose using intranasal immunization with live vectored vaccines based on recombinant respiratory viruses. Virus particles from cultures of Vero E6 cells infected with recombinant HPIV3 or EV were used to generate negative-stained images. Formvar-carbon-coated nickel grids (300 mesh) were floated onto drops of tissue culture fluid for 10 min; the adsorbed particles were stained with 2% methylamine tungstate, pH 6.8 (Nano-W). Stained viruses were then visualized and images digitally recorded using a transmission electron microscope operating at 100 to 120 kV.
Bukreyev, A.; Yang L.; Zaki S. R.; Shieh W. J.; Rollin P. E.; Murphy B. R.; Collins P. L., and Sanchez A.: A Single Intranasal Inoculation with a Paramyxovirus-Vectored Vaccine Protects Guinea Pigs against a Lethal-Dose Ebola Virus Challenge. J. Virol., 80, 2267-2279 (2006).
Want nanoparticles of cobalt rather than gold? Check the method of Salgueiriño-Maceira and group in their recent Langmuir article. Low size polydisperse cobalt spheres with an average diameter of 95 nm were synthesized by using a borohydride reduction method: 0.1 mL (0.4 M, 4 x 10-5 mol) of aqueous cobalt chloride hexahydrate was added to an aqueous solution (100 mL) of sodium borohydride (4.4 mM, 4.4 x 10-4 mol) and citric acid monohydrate (2 x 10-6 M, 2 x 10-7 mol) under mechanical stirring. Immediately following the cobalt reduction, the particles were coated with a thin layer of silica using 400 mL of an ethanolic solution containing 15 microliters of tetraethoxysilane (98%) was added. After 15 minutes, the solution was centrifuged, and the precipitate was redispersed in ethanol (40 mL). The large uniform cobalt spheres were found to consist of smaller metallic cobalt clusters, explaining their superparamagnetic behavior.
Salgueirino-Maceira, V.; Correa-Duarte, M. A.; Farle, M.; Lopez-Quintela, M. A.; Sieradzki, K., and Diaz, R.: Synthesis and Characterization of Large Colloidal Cobalt Particles. Langmuir, 22, 1455-1458 (2006).
Krichevsky and colleagues contribute a new method for preparing silver-gold nanowires in their recent article in Langmuir article. Thin, long gold/silver nanowires were grown on substrates in thin surfactant solution films; the growth process occurred exclusively in thinning aqueous films as the water evaporated, forming elongated surfactant template structures. The nanowire growth solution was prepared by combining Au(I) and Ag(I) solutions with Au:Ag molar ratio 1:1, in the presence of a high surfactant concentration, reducing agent (ascorbic acid), and sodium hydroxide addition. The standard growth solution contained 10 mL of 0.05M cetyltrimethylammonium bromide (CTAB) with 2.5 micromol of HAuCl4 (0.5 mL of 5 mM solution), 2.5 micromol of AgNO3 (125 microliters of 20 mM solution), and 55 micromol ascorbic acid. To this solution was added 40 microliters of 1 M NaOH (40 micromol) while stirring, raising the pH of the solution to a value of ~5 and initiating the growth process. Modifications of this process were also tested: addition of 25 microliters of 3-4 nm gold seed nanocrystal solution after base addition; variation of the Ag:Au mole ratio to the extremes where no gold or no silver was present; and leaving out OH- addition. Nanowire growth required the presence of a relatively high concentration of silver ions (typical Ag:Au mole ratio of 1:1), and tuning the pH value to about 5 in the growth solution. EDS analysis indicated nanowire composition of 85-90% gold and 15-10% silver.
Krichevski, O.; Tirosh, E., and Markovich, G.: Formation of gold-silver nanowires in thin surfactant solution films. Langmuir, 22, 867-870 (2006).
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .