ARTICLE pubs.acs.org/Langmuir
Michael Addition Reactions for the Modification of Gold Nanoparticles Facilitated by Hyperbaric Conditions Kurtis D. Hartlen, Hossein Ismaili, Jun Zhu, and Mark S. Workentin* Department of Chemistry and the Centre of Advanced Materials and Biomaterials Research (CAMBR), The University of Western Ontario, London, Ontario, N6A 5B7, Canada
bS Supporting Information ABSTRACT:
The chemical interfacial modification of organic solvent soluble 2.4 ( 0.5 nm maleimide-modified monolayer protected gold nanoparticles (2-C12AuNPs) with primary or secondary amines via Michael addition reactions is demonstrated. Michael addition reactions between 2-C12AuNPs and primary or secondary amines at ambient temperature and pressure and under the conditions where the AuNP is soluble and stable are possible albeit sluggish, often taking days to weeks to go to completion. The rates and efficacies of the these same reactions are drastically increased at hyperbaric pressure conditions (11 000 atm) with no observed adverse effect to the gold nanoparticle stability. The resulting Michael addition adducts (3-C12AuNPs) formed from 2-C12AuNPs and the corresponding amines were characterized by TEM and by comparison of the 1H NMR spectra of the 3-C12AuNPs with those of model reactions of the same amines with N-dodecylmaleimide, 2. The Michael addition reactions occur more readily with 2 rather than 2-C12AuNPs, consistent with the local environment of the latter imposing additional steric or other barriers to the reaction. The use of hyperbaric conditions makes the reaction of the organic solvent soluble 2-C12AuNP via Michael addition a viable interfacial modification process that is otherwise impractical. The results also suggest that it is a useful protocol for facilitating Michael addition reactions generally in solution at low temperatures.
’ INTRODUCTION Gold nanoparticles (AuNPs) have unique chemical and physical properties which can be utilized in catalysis,1 nanoelectric devices,2 and drug delivery,3 as well as a variety of sensors.4 These applications require a ligand on gold nanoparticle surface for stability and to introduce the appropriate functionality specific to serve the requirements for the particular application. Developing various techniques to introduce functionality to the AuNPs and understand the chemistry in the unique AuNP environment remains a critical aspect for their use. Typically, functionality is introduced to thiolate stabilized, organic solvent soluble AuNP (1 8 nm) via three main approaches: the direct synthesis of AuNPs with the target ligand,5 ligand-exchange reactions,6 and postsynthesis interfacial reactions of terminal functional groups exposed on the monolayer surface of AuNPs.7 Functionalization of AuNPs via direct synthesis does not always work; for example, thiols with functional groups that are sensitive to the AuNP synthesis conditions (strong reducing agents) will not be able to be introduced to the AuNP by this method. The place exchange method is a versatile way to introduce functionality to already prepared AuNP but requires the synthesis of the thiol with the desired functionality and in large excess. An approach where the desired functionality is introduced post-AuNP synthesis r 2011 American Chemical Society
via an interfacial reaction is often more desirable. AuNPs that contain a versatile reactive functional group that can participate in a variety of reactions could be utilized as a template AuNP to introduce a wide variation of structural diversity tailored to the designated application. One such versatile functional group that could be incorporated onto a AuNP and serve as a template AuNP is the N-substituted maleimide, which contains a highly reactive symmetrical double bond that can readily act as an electrophile in a Michael addition reaction,8 as well as a dienophile in a [4 + 2]7c,9 or a dipolarophile in a [3 + 2] cycloaddition reaction.7d The versatility of the reactivity of the maleimide functional group has been demonstrated with a variety of self-assembled surfaces such as 2-D silica and noble metals,10 quantum dots,11 and AuNP.7c,d,12 14 In addition, the maleimide group is frequently used in the formation of a variety of polymers and dendrimers, often as a cross coupling agent.15 Michael addition is one the most powerful and useful bond forming reactions because it is a simple and high yielding reaction, which makes it an ideal reaction for the modification Received: September 19, 2011 Revised: November 14, 2011 Published: November 15, 2011 864
dx.doi.org/10.1021/la203662n | Langmuir 2012, 28, 864–871
Langmuir of self-assembled nanostructures using interfacial reactions. Arguably, it is the most attractive and common reaction in use for bioconjugation studies for labeling peptides, proteins, and DNA.10 Given the propensity of maleimide to react with thiols and amines through a Michael addition reaction, maleimide has been shown to react with cysteine residues in DNA or RNA, as well as amine or thiol-modified biosubstrates, ultimately for applications as bioactive labels, biosensors, and in biochip fabrication.10 Schreiber10a and Bohn10b isolated maleimide groups onto silica substrates; then after exposure to various thiolated ligands, they were able to graft biotin, steroid digoxigenin, pipecolyl α-ketoamide, tetramethylrhodamine,10a and Cytochrome b510b to the silica surfaces via Michael addition. Mrksich extended the use of maleimide grafted surfaces to self-assembled monolayers, and have developed “biochip-type” assemblies by adding peptides, carbohydrates, or cell ligands onto such SAMS by the Michael-type addition reactions.10c e More recently, the groups of Mattoussi13 and Rodriguez-Fernandez14 have demonstrated the use of the Michael addition reactions involving large (>15 nm), watersoluble, maleimide modified AuNPs for bioconjugation applications. The ability to do these Michael addition reactions on smallsized AuNP in cases where protic solvents that facilitate the reaction cannot be used because of AuNP solubility, to structurally characterize the reaction products, and to achieve reaction efficiency remain desirable goals for the application of AuNP. Oftentimes, interfacial reactions between the functional group on the monolayer moiety of the AuNPs and reagents are sluggish relative to similar reactions carried out in a typical solution phase due to the unique reaction environment created by the AuNPs (ligands bound to different facets of the nanoparticle experience different rotational environments, as well as some steric effects of neighboring ligands). Furthermore, high temperatures, strong acids and bases, as well as many catalysts that often facilitate an organic reaction can affect the stability of AuNPs, which additionally restricts the variety of possible interfacial reactions that can be used to functionalize AuNPs.7,16 Additionally, in the case of reactions involving smaller thiol stabilized AuNPs their solubility properties often limit the types of solvents or reagents that can be utilized. Consequently, finding and employing efficient interfacial reactions that are not destructive toward the AuNP and work under mild reaction conditions is an important challenge in AuNP applications. Our group showed that slow reaction rates for a number of cycloaddition reactions at the AuNP interface, imposed by the need to work under mild conditions and by the local steric environment of the AuNPs, can be overcome by conducting interfacial reactions at high-pressure conditions (11 000 atm) with no observed change to the nanoparticle size or stability.7c e We utilized hyperbaric conditions to chemically modify maleimide-modified AuNPs through postsynthesis interfacial reactions including Diels Alder7c and nitrone 1,3-dipolar cycloaddition.7d In these studies, the reactions at ambient conditions would take days to weeks to go to completion (if they reacted at all), yet under hyperbaric conditions, the same chemical reactions were shown to go to completion in less than 1 h. In this report, we now extend the versatility of the maleimide-modified AuNP as a template nanoparticle to introduce functionality through the Michael addition reaction, further demonstrating the usefulness of hyperbaric reaction conditions to increase the efficacy of suitable reactions on AuNP materials. Chechik and co-workers have investigated Michael addition reactions of acrylate-modified AuNPs with dendritic polyamines
ARTICLE
and followed the addition reactions kinetics.17 They found that the Michael addition between the polyamine and acrylatemodified AuNPs was slightly retarded when compared to a similar solution-phase Michael addition between a model acrylate and polyamine. In their case, a complicating side reaction, methanolysis of the acrylate groups, led to a more complex kinetic study. The symmetric and more stable maleimide moiety is utilized here with the aim of a better understanding of how the unique environment of the nanoparticle influences the reactivity of these interfacial reactions. The Michael addition is known to have a negative volume of activation, and therefore becomes more favorable at elevated pressures.18 In fact, this reaction has been studied at high pressures in a seminal paper by the Jenner and Maddaluno19 groups, to name a few, for typical solution phase Michael addition reactions.20,21 However, there is no report of this reaction type being studied on AuNP surfaces at high pressures. This paper utilizes a 2.4 ( 0.5 nm maleimide-modified AuNP (2-C12AuNP) as a model organic solvent soluble AuNP and demonstrates its reactivity with a variety of primary and secondary amines (a h) in CH2Cl2 to form a series of 3a h-C12AuNPs with new functionality that was introduced via the Michael addition reaction using hyperbaric pressure conditions (Scheme 1).
’ EXPERIMENTAL SECTION Commercial Reagents and Solvents Used. The compounds dodecanethiol, hydrogen tetrachloroaurate(III), tetraoctylammonium bromide, 1,12-dibromododecane, sodium borohydride, ferrocene, maleimide, potassium thioacetate potassium carbonate, benzene-d6, CDCl3, CD2Cl2, and anhydrous ethanol were used as received from the suppliers. The primary and secondary amines used were all distilled/recrystallized prior to use, except compound 2-aminoethylferrocenylmethyl ether, which was synthesized as described below. General Instrumentation. 1H NMR spectra and 13C NMR spectra were recorded at the following frequencies: 1H, 600 MHz; 13 C, 150 MHz; or 1H, 400 MHz; 13C, 100 MHz in deuterated chloroform, deuterated methylene chloride, or deuterated benzene solutions and are reported in parts per million (ppm), with the solvent resonance used as a reference, C6D6 (7.15 ppm), CD3Cl (7.26 ppm), and CD2Cl2 (5.32 ppm). All the high pressure reactions were carried out in a LECO high-pressure chemical reactor. Infrared spectra were recorded on a FTIR spectrometer and are reported in wavenumbers (cm 1). Highresolution transmission electron microscopy images were collected from a JEOL 2010F HRTEM. Cyclic voltammetry was performed using a Perkin-Elmer Par 263A potentiostat interfaced to a computer equipped with PAR 270 electrochemistry software. The working electrode was a 1-mm-diameter glassy carbon rod, Tokai, GC-20, sealed in glass tubing. The counter electrode was a Pt wire. The reference electrode was a silver wire immersed in a glass tube with a sintered end containing 0.1 M tetra-butylammonium perchlorate (TBAP) in dichloromethane, (DCM). After each experiment, it was calibrated against the ferrocene/ferrocenium couple at 0.5 V versus saturated calomel electrode (SCE). Synthesis of Dodecanethiol AuNPs (Base-C12AuNP). Hydrogen tetrachloroaurate (III) trihydrate (0.30 g, 0.77 mmol) was dissolved in 28 mL distilled water (resulting in a bright yellow solution). This was added to a stirred solution of tetraoctylammonium bromide (2.30 g, 4.2 mmol) in 70 mL toluene. The contents were vigorously stirred for 30 min at room temperature in order to facilitate the phase transfer of the hydrogen tetrachloroaurate (III) trihydrate from the aqueous layer into the organic layer, indicated by a color change in the organic layer from clear to a dark-orange color and the aqueous layer 865
dx.doi.org/10.1021/la203662n |Langmuir 2012, 28, 864–871
Langmuir
ARTICLE
Scheme 1. Synthesis of Maleimide-Modified AuNP (2-C12AuNP) and Its Use in Michael Addition Reactions with a Variety of Primary and Secondary Amines (a h) under Ambient and Hyperbaric Pressure Conditions
dry DMF. K2CO3 (0.51 g, 3.6 mmol) was added and the mixture was heated to 50 °C for 3 h. The mixture was then was partitioned between water (100 mL) and DCM (100 mL). The organic layer was washed with water to remove DMF and then dried with MgSO4 and the solvent evaporated to dryness. The crude product was then purified by liquid column chromatography with an eluent of 3:1 ethyl acetate/hexanes. The product was then dissolved in 50 mL of toluene and heated to 100 °C for 10 h to liberate the furan which yielded 2, white crystal, overall yield 64%. 1H NMR (CDCl3, 600 MHz) δ (ppm): 5.83 (s, 2H), 3.42 (t, J = 7.24 Hz, 2H), 1.57 (m, 2H), 1.21 1.43 (broad, 18H), 1.02 (t, J = 7.06 Hz, 3H) ppm. 13C NMR (CDCl3, 150 MHz) δ (ppm): 171.4, 134.0, 37.9, 31.9, 29.6, 29.5, 29.4, 29.3, 29.2, 29.1, 28.5, 26.7, 22.7, 14.1 ppm. IR (cm 1, drop cast on NaCl): 3080, 2919, 2852, 1699, 1404, 1369, 1123, 846, 691. Exact Mass (C16H27NO2) calc: 265.2042; found, 265.2046. Synthesis of 2-Aminoethylferrocenylmethylether (b). To a solution of sodium aminoethanoate, which was obtained from a mixture of (2.29 g, 10.0 mmol) sodium metal and 25 mL of ethanolamine, was added (ferrocenylmethyl)trimethylammonium iodide (3.85 g, 10.0 mmol). The suspension was heated to 100 °C for 24 h yielding a dark brown solution. The reaction was quenched with water and extracted with diethyl ether, washed with 3 50 mL of water, and dried over MgSO4. The solvent was evaporated to dryness and recrystallized from hexanes giving a dark yellow solid, 48% yield. 1H NMR (CDCl3, 600 MHz) δ(ppm): 4.17 (s, 2H), 4.12 (s, 5H), 4.10 (s, 2H), 3.63 (t, J = 4.8 Hz, 2H), 3.52 (s, 2H), 2.80 (t, J = 5.3 Hz, 2H), 1.58 (broad, 2H) ppm. 13 C NMR (CDCl3, 150 MHz) δ (ppm): 84.3, 68.4, 68.3, 67.8, 60.8, 50.6, 48.4 ppm. IR (cm 1, drop cast on NaCl): 3091, 2923, 2852, 1668, 1456, 1103, 1077, 1001, 817, 484. Exact Mass: (C13H17FeNO) calc: 259.0660; found, 259.0648.
from yellow to clear and colorless. After phase transfer, the aqueous layer was removed and the organic layer was cooled to 0 °C in an ice bath. Dodecanethiol (0.156 g, 0.19 mL, 0.77 mmol) was added to the solution via a volumetric pipet and allowed to stir for ten minutes. The addition of dodecanethiol resulted in a color change from dark-orange to clear and colorless. A fresh solution of sodium borohydride (0.33 g, 8.7 mmol) in 28 mL water was then added to the rapidly stirring toluene solution over 5 s. The solution darkened instantly, eventually becoming dark brown. The mixture was allowed to stir overnight (∼18 h) as it warmed to room temperature. The aqueous layer was removed and the toluene layer was washed with 3 20 mL distilled water and dried over MgSO4. The toluene layer was then isolated by gravity filtration and evaporated to dryness. The resulting mixture of Base-C12AuNP and tetraoctylammonium bromide was suspended in 200 mL of 95% ethanol and placed in the freezer overnight during which time the Base-C12AuNP precipitated from solution. Afterward, the supernatant was decanted and the precipitate (Base-C12AuNP) was dissolved in benzene and concentrated, resulting in the formation of a thin film. This film was washed repeatedly with 10 15 mL of 95% ethanol, resulting in pure BaseC12AuNP as judged by 1H NMR spectroscopy, which showed no signs of free dodecanethiol, dodecyldisulfide, or tetraoctylammonium bromide. The resulting Base-C12AuNP was dark brown in color.
Preparation of Maleimide-Modified AuNPs (2-C12AuNPs). The maleimide-modified AuNPs (2-C12AuNPs) were prepared according to the reference procedure.7c,d Base-C12AuNP (0.15 g) was dissolved in 50 mL toluene and degassed with nitrogen. Thiol 1 (0.8 g, 0.16 mmol) was added to the solution and stirred for 75 min. The mixture was then concentrated and remaining thiol 1 was washed away from the resulting 1-C12AuNPs with 95% ethanol. The 1-C12AuNP (0.1 g) was dissolved in 30 mL toluene and heated at 100 °C for 10 h to induce the liberation of furan. The 2-C12AuNP product was concentrated and washed with 95% ethanol. The 1H NMR spectrum showed that the retroDiels Alder reaction was quantitative. Synthesis of N-Dodecylmaleimide (2). Furan-protected maleimide (3,6-Endoxo-Δ4-tetrahydrophthalimide)7c,d (1.005 g, 6.1 mmol) and 1-bromododecane (4.44 mL, 24.4 mmol) were dissolved in 70 mL
General Procedure of the Michael Addition Reaction between 2-C12AuNP or 2 and Various Primary and Secondary Amines under Ambient and Hyperbaric Pressure Conditions. Model maleimide 2 or 2-C12AuNP (20 mg) and a 5 molar excess of the primary or secondary amine were dissolved in 2.5 mL dichloromethane (CD2Cl2). The mixture was then separated into two 866
dx.doi.org/10.1021/la203662n |Langmuir 2012, 28, 864–871
Langmuir
ARTICLE
portions. One was placed in an NMR tube at ambient conditions. The other portion was placed in a PTFE tube and sealed with brass clamps and exposed to 11 000 atm until the reaction went to completion. Both reactions were monitored by 1H NMR spectroscopy; the completion of the reaction was indicated by the disappearance of the alkene peaks from unreacted maleimide, and the appearance of new peaks corresponding to the product. Any excess amine could be removed under vacuum. In the case of the AuNP adducts, the solvent was evaporated to form a thin film and the film washed repeatedly with EtOH. The characterization data of the resulting products 3a h and 3a-h-C12AuNP are provided in the SI.
’ RESULTS AND DISCUSSION Our approach (Scheme 1) for the preparation of maleimidemodified AuNP (2-C12AuNPs), as described previously,7c,d,12 involves first synthesizing dodecanethiol AuNPs (Base-C12AuNP), followed by the incorporation of thiol 1 onto the Base-C12AuNPs via a place exchange reaction, affording 1-C12AuNPs, then utilizing the thermal retro-Diels Alder reaction to unmask the 1-C12AuNPs by removing furan resulting in 2-C12AuNPs. This protocol, involving the incorporation of maleimide onto the AuNP in its furan-protected state by use of the place exchange reaction, serves several purposes: protecting the maleimide with furan prevents the Michael addition reactions between maleimide and other thiols present during the exchange reaction; using the furan-protected maleimide thiol 1 for a place exchange reaction allows one to control the amount of maleimide moiety onto the AuNP, because a high concentration of maleimide on the AuNP limits the solubility in the solvents used; the furan-protected maleimide AuNP (1-C12AuNP) can be stored for long periods of time, and the maleimide species can be unmasked when required. Dodecanethiol-protected gold nanoparticles (Base-C12AuNP) were prepared via a Brust-Schiffrin two phase synthesis (see Experimental Section).5 The Base-C12AuNPs were characterized using TEM and 1H NMR spectroscopy. Analysis of several TEM images determined the average nanoparticle diameter to be 2.2 ( 0.3 nm. The maleimide moiety was then introduced as a furanprotected maleimide thiol 1 (Scheme 1). Stirring a solution of Base-C12AuNP and thiol 1 in toluene for 75 min afforded the 1C12AuNP.7c,d The 1H NMR spectrum of 1-C12AuNP indicates the incorporation of the thiol 1 with the emergence of new broad signals at 5.79 ppm due to the olefinic protons (e), 5.04 ppm due to the protons α to the oxygen (d), 3.57 ppm due to the protons α to nitrogen in the maleimide ring (b), and 2.24 ppm due to the protons α to the carbonyl carbons in the maleimide ring (c) (Figure 1A). The broad peak between 1.83 and 1.13 ppm is due to the methylene protons of the alkyl chains of both the dodecanethiol and 1. The smaller broad peak (a) at 0.96 ppm corresponds to the terminal CH3 of the dodecanethiol ligand. These later two are shown off-scale in order to better illustrate the protons of the functional ligands. The protecting furan was liberated from 1-C12AuNPs to give 2-C12AuNPs by heating at 100 °C for 10 h in toluene, inducing the reverse Diels Alder reaction and liberating the furan into the solution. It has been reported that AuNPs are sensitive to high temperatures.22 However, a study by Schlenoff and co-workers showed that at temperatures around 100 °C only a small, negligible amount of thiol ligands will be cleaved off a SAM surface; therefore, it is reasonable to extend such a result to AuNPs.23 This was also evident by the nanoparticle stability and solubility after exposure to such temperature in our experiments.7c,d,12 Any excess and
Figure 1. 1H NMR spectra of (A) 1-C12AuNPs in C6D6 and (B) 2C12AuNPs in C6D6. Spectrum B contains a small amount of the furan protected maleimide.
liberated ligands (dodecanethiol or 1) as well as furan were removed by concentrating the reaction mixture to form a thin film, followed by washing the film with copious 95% ethanol. It should be noted that, in a 1H NMR spectrum of AuNP, free, nonbound ligands will appears as sharply defined NMR signals, while ligands bound to the AuNP surface will appear at roughly the same chemical shift as the free ligand, but the signals will be broad, lacking the distinct peak splitting. As a result, the purity of AuNPs can be determined from the lack of sharp peaks in the 1 H NMR spectrum. The 1H NMR spectrum confirms the formation of 2-C12AuNP by showing characteristic alkene peaks associated with the maleimide (Figure 1B). The spectrum of 2-C12AuNP contains two main diagnostic peaks: the alkene protons (f) in the maleimide moiety centered at 5.91 ppm and the methylene protons (e) α to the nitrogen in maleimide at 3.40 ppm. The relative integration of the peaks at 0.96 (nonexchanged dodecanethiol ligands) and 3.44 ppm (due to maleimide ligands on the 2-C12AuNP) indicated the ratio of dodecanethiol ligands to maleimide ligands anchored to the AuNP surface. It was found that the 2-C12AuNP contained ca. 24% of maleimide ligand compared to that of dodecanethiol and can be verified by degradation of the particles and analyzing the 1H NMR spectrum of the ligands that are cleaved.14 For the reactions described below, the 2-C12AuNP were prepared fresh (furan liberated as needed). Any unprotected maleimide would be evident in the NMR and could be removed by continued heating. As previously discussed, liberation of the maleimide by release of the furan results in a small size change of the AuNP. In the present case, the 2-C12AuNP were 2.4 ( 0.5 nm.7c,d,12 The maleimide moiety is susceptible to Michael addition reactions, and a variety of nucleophile thiols and amines could be used as nucleophiles to demonstrate the Michael addition reactions with 2-C12AuNP as long as the reaction conditions are not detrimental to the nanoparticle stability. Although thiols are excellent Michael donors, they pose potential problems. Thiols may participate in Michael addition reactions with maleimide as well as undergo place exchange reactions at the AuNP surface, which can make characterization complicated. Therefore, for this proof of principle study, amines although poorer nucleophiles were selected as the model Michael donor molecules, as they will selectively participate in the Michael addition reaction. The maleimide moiety is also a useful template on AuNP because the complication of regioisomers is absent because of its symmetry. 867
dx.doi.org/10.1021/la203662n |Langmuir 2012, 28, 864–871
Langmuir
ARTICLE
Table 1. Quantitative Formation of Michael Adducts 3a h and 3a h-C12AuNP after Reaction with Primary (a d) and Secondary (e h) Amines under Ambient and Hyperbaric Pressure Conditions
To investigate the reactivity of maleimide-modified nanoparticles toward a Michael reaction with the selection of primary and secondary amines with varying steric requirements, 2-C12AuNP (20 mg) was dissolved in CD2Cl2 and mixed with 5 times molar excess of each of the corresponding amines (relative to maleimide) selected for their structural diversity (a h, Table 1). The excess amine helps to push the reaction to completion as well as acts as a base. Each reaction solution was separated into two portions, one portion was left at atmospheric pressure and temperature (1 atm, 25 °C) and the other portion was placed in a sealed Teflon tube and exposed to 11 000 atm at 25 °C in a highpressure reactor. The progress and extent of the reactions was monitored using 1H NMR spectroscopy by following the disappearance of the olefinic protons of the maleimide on 2C12AuNP (the signals at 5.79 ppm) and by the emergence of the new peaks corresponding to the Michael adducts, 3a hC12AuNP. The time taken for the Michael addition reactions to go to completion (loss of maleimide signals at 5.79 ppm) at atmospheric and hyperbaric pressure conditions is reported in Table 1. After reaction completion, the products were worked up by washing with 95% ethanol and acetonitrile to remove unreacted amine leaving only 3a h-C12AuNP. Any impurity (or excess reagent) that was not bound to the AuNP would appear as sharp signals in the 1H NMR (vide supra); this is how the purity of the 3a h-C12AuNP was assessed. Because of the characteristic broadness of the peaks in the 1H NMR of AuNPs, characterization of interfacial reactions on the AuNPs by 1H NMR
Scheme 2. Synthesis of Model Michael Adducts (3a h) under Atmospheric and Hyperbaric Conditions
spectroscopy is verified by comparison to the 1H NMR spectra of AuNPs with those of the model compounds. We prepared the model Michael adducts 3a h (Scheme 2) by carrying out the Michael addition reactions between model maleimide 2 with the same series of amines (a h) at similar reaction conditions (Table 1). With model Michael adducts in hand, we assigned the 1H NMR spectra of 3a h-C12AuNP and by comparing the chemical shifts of 3a h to those of 3a h-C12AuNP we could support that the intended Michael adduct has formed on the AuNPs. Figure 2 shows the 1H NMR spectra of 3f-C12AuNPs and 3f, products of Michael addition reaction of 2-C12AuNPs and 2 with dibenzylamine under high pressure conditions, respectively. As can be seen in Figure 2A, the 1H NMR spectrum of 3fC12AuNPs, the appearance of peaks b f, assigned to the Michael adduct, and disappearance of maleimide alkenes peak verify the occurrence of designed Michael addition on the 2-C12AuNPs. However, we can ensure this by comparing the 1H NMR spectra 868
dx.doi.org/10.1021/la203662n |Langmuir 2012, 28, 864–871
Langmuir
ARTICLE
Figure 2. 1H NMR spectra of (A) 3f-C12AuNPs in CD2Cl2 and (B) 3f in CDCl3.
of 3f-C12AuNPs with 3f, and very good spectral alignment of the chemical shifts provides strong support for the synthesis of the Michael adduct on the AuNPs. Similar comparison spectra for all of the compounds (3a h-C12AuNPs and 3a h) can be found in the Supporting Information. Furthermore, by carrying out these model and AuNP Michael addition reactions under the same conditions (1 atm or 11 000 atm), we can gain some perspective on the extent that AuNPs inhibit the reaction rate of interfacial reactions compared to that of typical solution phase model reactions (Table 1). Figure 3 shows 1H NMR spectra of 2-C12AuNPs (A) and the products obtained from reacting 2-C12AuNPs with amines b, g, and h, respectively (3b,g,h-C12AuNP), as representative examples of the Michael adducts on the AuNPs. Once more, the emerging new peaks in the 1H NMR spectra and the disappearance of maleimide alkene peak (peak a in Figure 3A) confirm the efficacy of the Michael addition reaction under high preassure conditions. It should be noted that the reactions described do contain some inherent bias: Michael addition reactions are known to be accelerated in protic solvents like methanol, and the present reactions are run in methylene chloride.24 This bias is partially the result of the solubility characteristics of these small 2-C12AuNP where the addition of methanol leads to some aggregation and insolubility. To avoid this, we ran the reactions under conditions where the 2-C12AuNP remain soluble. Table 1 shows that the Michael addition reactions conducted under these conditions at atmospheric conditions are slow, on the scale of days. However, under the same reaction conditions, except conducted at 11 000 atm, the reaction goes to completion in most cases in 30 min. The minimum reaction time that a reaction could be exposed to high pressure was 10 min (the time it takes to pressurize and depressurize the chamber); however, test reactions showed that at least 20 min was required for qualitative results for the fastest reactions, so 30 min was chosen as the minimum reaction time to ensure the reactions went to completion. Reactions that previously took days or even weeks (in the case of dibenzylamine)
can now be completed in less than one hour. Moreover, by comparing the reactions of 2 and 2-C12AuNPs at 1 atm, it is not surprising to see the amines (a h) react more efficiently with 2 than 2-C12AuNPs; this is also consistent with results previously obtained that demonstrate that interfacial biomolecular reactions are subject to additional environment/steric constraints.7 Closer examination of Table 1 also reveals that the reaction times for each amine took various lengths of time but follow a general trend of relative nucleophilicity of the amine25 (H2NCH2Ph > butylamine > t-butylamine) and where the more stericaly hindered secondary amines take longer to go to completion than the less sterically hindered primary amines. We and others have previously shown that reactions at the AuNP interface are subject to additional steric effects.7 However, the main point of consideration here is that, regardless of relative nucleophilicity and associated steric effects due to the amine or interfacial interactions, utilizing high pressures ensured that all reactions went to completion in less than an hour, and in most cases in 30 min. All of the reactions were monitored by 1H NMR spectroscopy as described above. In some cases, like 3a- and 3d-C12AuNP the characterization of the product is not as clear as the other cases (for example, like those in Figure 3). This is due to the lack of key diagnostic protons that are well separated from the signals from the C12 functionality, the breadth of the signals of the nucleophiles on the AuNP, and, in the case of 3a-C12AuNP, the reaction not going to completion. In the case of 3d-C12AuNP, the diagnostic signal was that of the added methyl groups of the t-butyl amine. These two cases highlight the additional benefit of nucleophiles like that used in the preparation of 3b-C12AuNPs (Figure 3B). Here, the nucleophile is a primary amine but also contains an electrochemical redox active moiety (ferrocene). Electrochemical methods are more sensitive than 1H NMR spectroscopy; therefore, one can potentially monitor the extent of this Michael addition reaction with a higher degree of sensitivity than by using 1H NMR spectroscopy alone (particularly valuable for many AuNPs applications where 1H NMR characterization can be difficult, or using extremely low 869
dx.doi.org/10.1021/la203662n |Langmuir 2012, 28, 864–871
Langmuir
ARTICLE
Figure 3. 1H NMR spectra of (A) 2-C12AuNP and representative spectra of the products obtained after Michael addition reactions with amines b, g, and h: (B) 3b-C12AuNP; (C) 3g-C12AuNP; (D) 3h-C12AuNP.
Figure 5. TEM images of (A) 2-C12AuNPs before exposure to hyperbaric pressure conditions (2.4 ( 0.5) and (B) after exposure to hyperbaric pressure conditions for 18 h (2.4 ( 0.5) (The scale bar for both is 20 nm).
Figure 4. Cyclic voltammetry of 0.0019 mol 3b (---), and 5 mg of 3bC12AuNP (___) in 1.8 mL CH2Cl2 containing 0.1 M TBAP as electrolyte. The CVs are referenced relative to ferrocene/ferrocenium couple at 0.5 V versus SCE.
concentrations). Cyclic voltammetery was carried out on the resulting Michael adducts 3b-C12AuNP after reacting with the ferrocenyl amine (b, Table 1) and confirmed the presence of the ferrocene moiety attached to the succinimide for both the model compound and the AuNP (Figure 4). It is also important to note that there was no observed ill-effect to the nanoparticle size or stability after exposure to hyperbaric
pressure conditions. Any aggregation or change in particle size would result in either a precipitate forming or an observable surface plasmon resonance; neither was observed. TEM images before and after exposure to hyperbaric pressure conditions confirm that there was no change in particle size (Figure 5).
’ CONCLUSION Given that Michael addition reactions are the most versatile for bioconjugation applications, the results presented here demonstrate the further versatility of maleimide functionalized gold nanoparticles, such as 2-C12AuP, as a template surface for the interfacial modification through Michael addition reactions adding to its use in Diels Alder and 1,3 dipolar cycloaddition reactions. The modification of the AuNP through the Michael addition reaction was carried out using a number of primary and secondary amines with varying steric requirements as 870
dx.doi.org/10.1021/la203662n |Langmuir 2012, 28, 864–871
Langmuir
ARTICLE
nucleophiles. Although most of these reactions could be carried out under normal 1 atm and 25 °C conditions, they are extremely sluggish, on the order of days, and really not very practical for any useful application. However, upon exposure to hyperbaric pressure conditions, 11 000 atm under otherwise mild conditions at 25 °C and no catalyst, these reactions can go to completion in less than an hour. Moreover, the high pressures conditions are not detrimental to the nanoparticle size or stability. Therefore, high pressure facilitates the use of maleimide modified AuNPs as an effective template macromolecule where further task-specific functionality can be introduced via Michael addition reactions. The next challenge is to design and prepare small (
