Quantum Dots Stability in Aqueous Media

May 22, 2018

by Kil Ho Lee

For the past several decades, the synthesis routes for high quality QDs have been reported by a number of research groups.  Quantum Dots (QDs) can be either hydrophobic, or hydrophilic based on the type of ligands/surface coating passivating the particle surface; however, the stability of QDs tends to be higher for hydrophobic QDs. Therefore, QDs used for biological applications often require polymeric layer, which is terminated with functional groups (i.e. –COOH, or –NH2).

In the series of reports on “Toward DNA embedded Quantum Dots”, we demonstrated ligand exchange technique that allows substituting hydrophobic ligands (i.e. octadecylamine, or ODA) with hydrophilic ligands (i.e. 3-mercaptopropionic acid, or 3-MPA). The successful ligand exchange provided QDs soluble and stable in water without adding polymeric layer. Then, 3-MPA coated aqueous QDs (3-MPA coated CdSe/ZnS) were subjected to the formation of additional ZnS shell. In addition, the latest report demonstrated semi-successful DNA embedding during the formation of ZnS shell; TIRF microscopy showed both QD aggregates and individual QDs post DNA embedding procedure was performed.

So far, the experimental approach to optimize DNA embedded QDs was done using commercially available, hydrophobic CdSe/ZnS. Despite extensive optimization effort, we recently learned that the same procedure was not applicable to more recently purchased QDs (the same vendor and the same product, but different lot #). The batch to batch variation(s) of commercial QDs is a limitation that we encounter more often than desired. In fact, the same problem was acknowledged during the optimization period.

In this report, we examined the stability of QDs. More specifically, we characterized the stability and fluorescence property of aqueous QDs. These QDs are commercially available and coated with a polymer layer, which is functionalized with carboxylic acid group (-COOH). The purpose of this report is to demonstrate the effect of different processing variables (i.e. dilution, centrifugal filtration, buffer exchange, etc) on the overall quantum yield of QDs. Photoluminescence quantum yield (PLQY), or quantum yield (QY), is a ratio of the number of photons emitted to the number of photons absorbed. Hence, QY measures the efficiency of photons, or fluorescence, emitted by QDs upon excitation.

Without going through the details of every parameter tested, here we discuss the most relevant parameter (i.e. centrifugal filtration, buffer exchange) and how they affect the stability of aqueous QDs.

 

Buffer exchange

Buffer exchange is a process of dispersing QDs in a solvent to a different solvent. For example, QDs in water may require the buffer exchange to a solution of ionic salts, or to a solvent at a specific pH. Buffer exchange is often required for biomodification or conjugation.

Buffer

Vendor A

Vendor B

Borate

50 mM, pH 9

36.26 ± 0.89°

21.97 ± 3.17

Water

36.43 ± 0.50

21.35 ± 0.73°

MES

0.1 M. pH 4.7

-

13.19 ± 0.77*

°indicate the original buffer condition after the 1st filtration using centrifugal filter device [Table III]

*indicate statistically different (α = 0.05) QY compare to QY before the buffer exchange

We tested CdSe/ZnS terminated with –COOH from two different vendors. These QDs were similar in emission wavelength; hence, they were approximately the same size. QDs from vendor A were shipped and stored in Borate buffer (50 mM, pH 9), whereas QDs from vendor B were shipped and stored in water. When QDs from vendor A were dispersed in water, QY did not change and the statistical analysis suggest that QYs before and after buffer exchange were not statistically different. The same result was consistently found for QDs from vendor B. However, when QDs from vendor A were dispersed in MES buffer (0.1 M, pH 4.7), QDs almost instantaneously aggregated and precipitated to the bottom of a test tube. QDs from vendor B, however, did not show a sign of aggregation and the QY was reduced slightly. Statistical analysis suggest that QY for QDs from vendor B in MES buffer was statistically different from QY in the original solvent.

MES buffer (0.1 M, pH 4.7) is a buffer often used for bioconjugation. For exchange, conjugating a protein (i.e. antibody), or ssDNA on QD surface is done via EDC/Sulfo-NHS chemistry in MES buffer.

Loops, trains, and tails technique and DNA embedding are two approaches we have tested because, for the most part, the conjugation chemistry is not very efficient. Also, as suggested in this report, QD stability may be reduced significantly.

Centrifugal filtration

In between almost all QD modification steps, centrifugal filtration is done to purify QDs from excess monomers, proteins, DNA, etc. The same QDs tested for the buffer exchange effect were used to examine the effect of centrifugal filtration.

Filtration

Vendor A

Vendor B

Wash 1

36.26 ± 0.89

21.35 ± 0.73

Wash 2

37.13 ± 0.44

18.00 ± 1.19*

Wash 3

33.93 ± 0.76*

11.58 ± 1.66*

*indicate statistically different (α = 0.05) QY compare to QY before centrifugal filtration

Typically, at least 3 filtration (or washing) steps are done to QDs to remove excess reagents in the solution. As shown in Table 2., QY was reduced after 2-3 times of washing. For Loops, trains, and tails and DNA embedding, QDs are frequently filtered. Therefore, the reduced fluorescence intensity reported in previously can at least partially explained.