Supplementary Material - Near-infrared, surface-enhanced

fluorescence using silver nanoparticle aggregates in solution

(Plasmonics)

Michael D. Furtaw †, ‡,*, Jon P. Anderson †, Lyle R. Middendorf †, Gregory R. Bashford ‡

†LI-COR Biosciences, Inc., Lincoln, Nebraska 68504, and ‡Department of Biological Systems Engineering,

University of Nebraska, Lincoln, Nebraska 68583

E-mail: mike.furtaw@licor.com

Table of Contents

1. UNCERTAINTY IN THEORETICAL CALCULATIONS..........................................................................2

2. OPTIMAL SALT CONCENTRATION................................................................................................2

3. DATA PROCESSING.......................................................................................................................4

4. ENHANCEMENT CALCULATIONS..................................................................................................5

5. SEM IMAGES OF AgNP.................................................................................................................5

6. TEM IMAGES OF AGNP AGGREGATES..........................................................................................6

7. NIR-SEF OF VARIOUS FLUOROPHORES........................................................................................9

8. REFERENCES...............................................................................................................................10

S-1

1. UNCERTAINTY IN THEORETICAL CALCULATIONS

As mentioned in the manuscript, theoretical calculations based on commonly accepted experimental

values should be used with caution. The quality factor (Q) for Ag is shown in Fig. S1 with error bars

corresponding to those published in Johnson & Christy (1972)[1]. While many plasmonic calculations do

not require Q explicitly, they all depend on the complex permittivity values of which Q is based. The

authors are not aware of any theoretical plasmonic journal articles where this error margin has been

discussed.

400 500 600 700 800 9000

50

100

150

200

250

300

350

λ (nm)

Q

Fig. S1 Quality factor for Ag with error bars (using experimental margin of error[1])

2. OPTIMAL SALT CONCENTRATION

The optimal salt concentration (under our conditions) was found by conducting a 2-D dilution

experiment in a 384-well plate. The columns contained varied salt concentration from 200 mM down to

0.2 mM NaCl in 2-fold dilutions (2 columns per dilution) with the final 2 columns containing dH2O. The

rows contained varied IRDye 800CW-SAv concentration from 500 ng/ml down to 500 fg/ml in 10-fold

dilutions (2 rows per dilution), with the final 2 rows being background wells. The data (Fig. S2-S3) shows

S-2

an optimal salt concentration near 5 mM NaCl. The data also show that the signal is linear over the

range used of IRDye 800CW-SAv, for multiple concentrations of NaCl. This indicates that slight variation

in salt concentration should not significantly affect the ability to detect and quantitate the dye with our

NIR-SEF protocol.

1 10 100 10000.01

0.1

1

10

100

1000

5000 pg/ml 500 pg/ml50 pg/ml5 pg/ml0.5 pg/ml

NaCl (mM)

Fluo

resc

ence

(AU)

Fig. S2 Fluorescence intensity versus NaCl at multiple concentrations of IRDye 800CW-SAv

0.1 10 1000 1000000.1

1

10

100

1000

12.5 mM6.25 mM3.125 mM0.195 mM0 mM

SAv - 800 CW (pg/ml)

Fluo

resc

ence

(AU)

Fig. S3 Fluorescence intensity versus IRDye 800CW-SAv at multiple concentrations of NaCl

S-3

3. DATA PROCESSING

All fluorescence data were normalized to detector gain intensity of 11 by

I=I n∙211−n

where n is the detector gain intensity. The values for all sample wells were taken at the highest

detector gain intensity prior to detector saturation. In order to verify the linearity of the instrument

across all detector gain settings, one sample well for each data set were corrected from multiple images

(Fig. S4). This analysis also shows signal intensity remains constant over the 30 minutes of image

acquisition (approximately 5 minutes for each image at differing scan intensity). This suggests that

aggregate sedimentation is not responsible for signal enhancement by bringing more fluorophores into

the depth of field of the instrument (FWHM = 1.5 mm). If sedimentation were significant, one would

expect an increase in signal with time. The only other possibility is that sedimentation fully completed

within the 1 hour incubation period which is not plausible. Even so, if sedimentation did occur it could

only be responsible for a modest amount of enhancement as the original sample depth is only about 3

times the depth of field of the instrument.

0 2 4 6 8 10 120

50

100

150

200

250

AgNP + saltAgNPdH2O

Scan intensity

Corr

ecte

d in

tens

ity

Fig. S4 Corrected intensity for one set of wells at all detector gain settings. The first data points (intensity = 1) for ‘dH2O’ and ‘AgNP + salt’ may be too small to be considered within the linear range. The near-constant value of the rest of the corrected intensities suggests the instrument is linear across all detector gain settings

S-4

4. ENHANCEMENT CALCULATIONS

Fluorescence enhancement was calculated by

SEF=I sample−I background, sampleI reference−I background, reference

Where the sample background was AgNP added to dH2O or 5 mM NaCl and the reference background

was dH2O.

5. SEM IMAGES OF AgNP

SEM images were acquired using a Hitachi S4700 field-emission scanning electron microscope (Hitachi

High Technologies America Inc., Schaumburg, IL) at the University of Nebraska – Lincoln. The samples

were immobilized on silicon wafers following a NIST-NCL joint protocol[2]. The immobilized samples

were sputter-coated with a thin layer of nickel.

S-5

Fig. S5 SEM image of AgNP monomers. The sizes of the individual nanoparticles appear to be in good agreement with the DLS average of 21 nm

Fig. S6 SEM images of AgNP monomers in wider field-of-view

6. TEM IMAGES OF AGNP AGGREGATES

TEM images were acquired using a Hitachi H7500 transmission electron microscope (Hitachi High

Technologies America Inc., Schaumburg, IL) at the University of Nebraska – Lincoln. The samples were

immobilized on silicon monoxide films supported with Formvar on a 200 mesh copper grid (Prod #

01830, Ted Pella Inc., Redding, CA) following a NIST-NCL joint protocol[3]. It is important to point out

that it is very difficult to immobilize salt-induced aggregates without affecting the morphology due to

the inherent instability in the aggregate solution. The images presented here should be taken as

qualitative representations of the aggregates and visual proof of the aggregation process. The actual

aggregates in solution may differ from those immobilized during the TEM imaging process.

S-6

Figure S7. TEM image showing a monomer and two dimers formed by adding SAv to the AgNP solution.

S-7

Figure S8. TEM image showing trimers formed after adding SAv and salt to the AgNP solution. Notice the presence of a

triangular plate, which do not show up often in EM images, so they are expected to contribute very little, if at all, to the SEF

process.

S-8

Figure S9. TEM image showing higher-order aggregates after adding SAv and salt to the AgNP solution. It appears salt-induced

agglomeration may be occurring, which is consistent with Zhang et al[4]. It is unknown at this time what this process

contributes to the observed SEF.

7. NIR-SEF OF VARIOUS FLUOROPHORES

The technology was also tested on other NIR fluorophores conjugated to SAv with similar spectral

properties to IRDye 800CW. DyLight 800 (Thermo Fisher Scientific, Rockford, IL) was purchased pre-

conjugated to SAv, while Alex Fluor 790 (Life Technologies, Grand Island, NY) and CF790 (Biotium,

Hayward, CA) were conjugated using a common NHS-ester labeling technique[5]. The table below

summarizes the fluorophores along with their physical characteristics (per manufacturer websites).

Fluorophore λabs,max λem,max ελabs,max

Alexa Fluor 790 782 805 260,000DyLight 800 777 794 140,000

IRDye 800CW 778 794 240,000

S-9

CF790 784 806 210,000

Figure S10 shows the fluorescence enhancement for the various fluorophores. The enhancement is

very similar for all of the fluorophores tested. Some variation is expected as the fluorophores have

differing intrinsic quantum yield[6] and absorption/emission spectra. It is unclear at this time as to why

the salt-induced aggregation gave less enhancement for the CF790.

Alexa Fluor 790

DyLight 800

IRDye 800CW

CF790100

1000

10000

Fluo

resc

ence

enh

ance

men

t

Figure S10. Fluorescence enhancement for various NIR fluorophores with AgNP only (blue) and AgNP with salt (red).

8. REFERENCES

1. Johnson PB, Christy RW (1972) Optical constants of the noble metals. Phys Rev B 6 (12):4370-4379. doi:10.1103/PhysRevB.6.43702. Vladar AE, Ming B (2011) Measuring the size of colloidal gold nano-particles using high-resolution scanning electron microscopy. NIST - NCL Joint Assay Protocol. Natl. Inst. Stand. Technol., NIST (U.S.)3. Bonevich JE, Haller WK (2010) Measuring the size of nanoparticles using transmission electron microscopy (TEM). Gaithersburg, MD4. Zhang PX, Fang Y, Wang WN, Ni DH, Fu SY (1990) Influence of addition of potassium chloride to silver colloids. J Raman Spectrosc 21:127-1315. IRDye 800CW Protein Labeling Kit - High MW (2006).

S-10

6. Tynan CJ, Clarke DT, Coles BC, Rolfe DJ, Martin-Fernandez ML, Webb SE (2012) Multicolour single molecule imaging in cells with near infra-red dyes. PLoS One 7 (4):e36265

S-11