Measurement of elastic light scattering from twooptically trapped microspheres and

red blood cells in a transparent mediumMatti Kinnunen,1,* Antti Kauppila,1 Artashes Karmenyan,2 and Risto Myllylä1

1Optoelectronics and Measurement Techniques Laboratory, University of Oulu, 90014, Oulu, Finland2Institute of Biophotonics, National Yang-Ming University, 11221 Taipei, Taiwan

*Corresponding author: matti.kinnunen@ee.oulu.fi

Received April 20, 2011; revised August 3, 2011; accepted August 12, 2011;posted August 15, 2011 (Doc. ID 146313); published September 8, 2011

Optical tweezers can be used to manipulate small objects and cells. A trap can be used to fix the position of a particleduring light scattering measurements. The places of two separately trapped particles can also be changed. In thisLetter we present elastic light scattering measurements as a function of scattering angle when two trapped spheresare illuminated with a He–Ne laser. This setup is suitable for trapping noncharged homogeneous spheres. We alsodemonstrate measurement of light scattering patterns from two separately trapped red blood cells. Two differentillumination schemes are used for both samples. © 2011 Optical Society of AmericaOCIS codes: 350.4855, 120.4640, 290.4210.

The light scattering properties of turbid samples, i.e.,averaged information over large sample volumes, canbe measured with different optical methods, such as dif-fuse reflectance spectroscopy and optical coherencetomography [1,2]. Phantom and tissue measurementsin vitro are used to verify theoretical models andexperimental methods. Many of the currently publishedexperimental results are limited to transmittance andsmall-angle scattering cases (e.g., Refs. [3,4]). Knowledgeof the optical properties of single particles is of greatimportance when developing models on a micro scale.Measurement of these values is important, especiallyin biological applications, and will result in more exactanalytical methods for noninvasive diagnostics. Thismeans light scattering properties need to be measuredat the single-particle level [5]. Theoretical work has beendone to understand light scattering from single particles[6], from clusters of spheres [7,8], and also from multiplespheres [9]. It is important that these models can be ver-ified experimentally. Earlier measurements of the opticalproperties of single particles and cells have been donewith flow cytometry devices and optical tweezers com-bined with a light scattering measurement instrument[5,10–13]. An electrodynamic levitator trap has been usedto keep two-sphere clusters still during the measure-ments [14], but the drawback of this approach is the needto use charged particles. Another method has been a mi-crowave analog technique, where one or several particleshave been illuminated with microwaves [15,16]. In thismethod, the particle size must be in the range of severalmillimeters. An earlier work showed that dependent scat-tering is significant and needs to be taken into accountwhen the distance between particles is less than tentimes their diameter [17].This Letter depicts the optical tweezers technique [18]

with a double-beam extension [19] to keep polystyrenespheres stable during light scattering measurements.Two-sphere scattering can be measured at differentparticle positions in relation to an incoming He–Nelaser beam by changing the trap positions. We alsodescribe the trapping of two red blood cells (RBCs)

simultaneously for light scattering experiments at twodifferent positions.

An IR laser (ILM-L3IF-300 diode module, LedLightTechnology, Taiwan) was used for trapping. We used adouble-beam configuration, where polarizing cubic beamsplitters are used to separate two different beams. Half-wave plates were used to control the optical power of thebeams. A telescope in one of the beams, with a movablelens holder aligned perpendicularly to the propagationdirection of the beam, was used to move one trap posi-tion. The other trap position was controlled by using amirror for beam steering. The setup for trapping is de-scribed in more detail in Ref. [19].

The setup for the light scattering measurement con-sisted of a He–Ne laser (Melles Griot 05-LHP-151), an op-tical chopper to modulate the beam, a focusing lens(þ200mm), and an aperture. The optical power of theHe–Ne laser before the sample cuvette was 2:8mW.The output diameter of the laser beam was 2mm, andthe beam’s focal diameter under the objective was calcu-lated to be about 200 μm, which is enough for smoothillumination of a trapped particle. An amplified photo-multiplier tube (Thorlabs PMM02) was used on the de-tecting side. The detector was moved around thesample in a horizontal plane with a motorized rotationstage (Standa 8MR190-2-28). The measurement schemeis shown in Fig. 1. The rotation stage had a 0:01° resolu-tion. An angle step of 0:5° was used for the measure-ments. Labview 8.2 software was used for rotationstage movement control and data acquisition. The

Fig. 1. (Color online) Measurement configuration.

3554 OPTICS LETTERS / Vol. 36, No. 18 / September 15, 2011

0146-9592/11/183554-03$15.00/0 © 2011 Optical Society of America

scattering measurement setup and alignment procedureare described in more detail in Ref. [13].A cylindrical cuvette with an inner diameter of

22:6mm (Hellma, shortened version of 540.115) was usedin the experiments. The cuvette was carefully cleanedand filled with distilled water. The water was filtered witha filter pore size of 0:2 μm (Pall Acrodisc 13 GHP) fourtimes. A small amount of polystyrene spheres (6:0 μm,Bangs Laboratories, Inc.) were added to the filteredwater in a microcentrifuge tube, and the sample was al-lowed to sedimentate. The upper part of the suspensionwas replaced with new filtered water and mixed togetherwith the sedimentated microspheres. Approximately10 μl of the suspension was added to the bottom of thecylindrical cuvette.Fresh RBCs were collected with the finger prick meth-

od and diluted with filtered phosphate-buffered saline(PBS, Sigma, Sigma-Aldrich). The PBS was filtered witha filter pore size of 0:2 μm (Pall Acrodisc syringe filterwith a Supor membrane) three times. The cleaned cuv-ette was filled with filtered PBS. A small amount(∼10 μl) of cells was injected at the bottom of the cuvetteand allowed to sedimentate. All the samples were pre-pared and measured at room temperature.Figure 2 shows particle alignment, a near-field scatter-

ing intensity image, and light scattering distributionsfrom two trapped spheres in the cases of particles inthe direction of the incident beam [Fig. 2(a)] and parti-cles perpendicular to the direction of the incident beam[Fig. 2(b)]. We plotted the curves of two measurementsto show their stability and repeatability. Measurementwith a single sphere was done to evaluate the perfor-mance of the system. This measurement was verifiedwith theoretical modeling (not shown here). In the case

presented in Fig. 2(a), the second bead is in the shadowof the first bead, resulting in the same scattering crosssection toward the incident laser beam as in the single-sphere case. However, the scattering pattern differsconsiderably from the case of a single sphere. Mutualshielding of the first sphere by the other sphere de-creases the detected scattered intensity, and the latterparticle experiences an inhomogeneous field and not aplane wave [20]. The shielding effect needs to be takeninto account for interparticle distances up to severaltimes the particle’s diameter [20].

In the case in Fig. 2(b), the scattering cross section isdoubled in relation to the incident laser beam, which in-creases scattering intensity. Scattering peaks in the rangeof 10°–50° differ significantly in comparison to the single-particle case and have more than doubled intensity.Also, the minimums in the scattering pattern are deeper.Because of the small distance between the particles (lessthan their diameter), the near-field effects need to be ta-ken into account when interpreting the results [20]. Thisresult differs from the conclusion in Ref. [17], where theauthors note that mutual interaction is minimal when twoparticles are aligned perpendicularly to the incidentbeam. However, detailed analysis of the measurementresults is beyond the scope of this Letter.

When compared with the microwave analog technique,where particles of size 5mm and their aggregates [16] canbe used, the size parameter (x ¼ 2πr=λ) is in the range of0.845 to 14.05. In our case, when using 6:0 μm particles,the corresponding value is 29.8. We have a fixed laserwavelength, and hence the size parameter can be chan-ged by changing the size of the particle. In our systemparticles can be kept in a fixed position without contactduring the measurements. The trap positions can be gra-dually changed in our setup, allowing the distance of thespheres to be changed.

Biological cells can also be used as samples in oursetup. We can use PBS as an immersion medium andmeasure light scattering from real RBCs, which is a clearadvantage in comparison with the microwave analogtechnique. The only requirement is full transparency ofthe immersion medium.

Figure 3 shows the results from elastic light scatteringexperiments with two RBCs. The nonspherical shape ofthe cells induces different light scattering patterns thanare obtained from polystyrene spheres. Because of thesmaller relative refractive index of RBCs and PBS incomparison with the polystyrene spheres and distilledwater, the scattering intensity might also be weaker. Itis clearly seen from the figures that the RBCs are alignedaccording to the polarization of the point tweezers.Because our double-beam trapping system is made by se-parating two beams with cubic beam splitters, the com-bined beams have a 90° shift in the orientation of theirpolarization.

Simulation results obtained by He et al. [21] showedthat when two RBCs are in face-on orientation, but at dif-ferent distances from each other along the direction of anincident light beam, multiple scattering is important andneeds to be taken into account. Our measurements inFig. 3 differ slightly from that case due to the differentpolarization directions of the trapping laser beams.The left cell is not in face-on orientation. However, the

Fig. 2. (Color online) (a) Elastic light scattering from twospheres in the direction of an incident beam and (b) elastic lightscattering from two spheres perpendicular to the incidentbeam. The small figures show the orientation of the objectsin relation to the laser beam and a CCD camera image of thenear-field light scattering distribution. Vertical polarizationwas used. The scale bar is 10 μm.

September 15, 2011 / Vol. 36, No. 18 / OPTICS LETTERS 3555

intensity image reveals a form that is very different fromscattering from a single cell [19]. Hammer et al. [22]found that the conditions of single scattering are not ful-filled in whole blood. When the distance of the scatterersis of the same order as their diameters, they cannotbe treated as independent scatterers. The effects of inter-ference need to be taken into account. When comparingthe results with those from a single sphere and cell[5,12,13,19], implementation of the second beam in thetrap opens new possibilities for studying light–matterinteraction.In conclusion, we have demonstrated an optical

tweezer measurement system for studying light–matterinteractions of two-particle systems. Although the cur-rent detection system needs to be improved, we expectthat this method will be beneficial in verifying both sin-gle- and multiple-scattering theoretical models for rigidparticles in the visible wavelength range. This systemis also applicable in measuring elastic light scatteringfrom RBCs in a PBS solution.

This work is part of a joint project of the Academy ofFinland and the Russian Foundation for Basic Research,and it has been funded by the Academy of Finland (grant124176). M. Kinnunen is thankful to the Academy ofFinland for his personal grant (128073).

References

1. S. L. Jacques and B. W. Pogue, J. Biomed. Opt. 13,041302 (2008).

2. D. J. Faber, F. J. van der Meer, C. G. Aalders, andT. G. van Leeuwen, Opt. Express 12, 4353 (2004).

3. I. Turcu, V. L. Pop, and S. Neamtu, Appl. Opt. 45,1964 (2006).

4. N. Ghosh, P. Buddiwant, A. Uppal, S. K. Majumder, H. S.Patel, and P. K. Gupta, Appl. Phys. Lett. 88, 084101(2006).

5. R. M. P. Doornbos, M. Schaeffer, A. G. Hoekstra,P. M. A. Sloot, B. G. de Grooth, and J. Greve, Appl. Opt.35, 729 (1996).

6. C. F. Bohren and D. R. Huffman, Absorption and Scatteringof Light by Small Particles (Wiley, 1983), pp. 3–11, 384–389.

7. G. O. Olaofe, Appl. Opt. 9, 429 (1970).8. P. J. Flatau, K. A. Fueller, and D. W. Mackowski, Appl. Opt.

32, 3302 (1993).9. F. Voit, J. Schäfer, and A. Kienle, Opt. Lett. 34, 2593 (2009).10. D. H. Tycko, M. H. Metz, E. A. Epstein, and A. Grinbaum,

Appl. Opt. 24, 1355 (1985).11. K. A. Sem’yanov, P. A. Tarasov, J. T. Soini, A. K. Petrov, and

V. P. Maltsev, Appl. Opt. 39, 5884 (2000).12. D. Watson, N. Hagen, J. Diver, P. Marchand, and

M. Chachisvilis, Biophys. J. 87, 1298 (2004).13. M. Collins, A. Kauppila, A. Karmenyan, L. Gajewski,

K. Szewczyk, M. Kinnunen, and R. Myllylä, Proc. SPIE7376, 737619 (2010).

14. S. Arnold, A. Ghaemi, P. Hendrie, and K. A. Fuller, Opt. Lett.19, 156 (1994).

15. B. Å. S. Gustafson, J. Quant. Spectrosc. Radiat. Transfer 55,663 (1996).

16. O. Merchiers, J.-M. Geffrin, R. Vaillon, P. Sabouroux, andB. Lacroix, Appl. Phys. Lett. 94, 181107 (2009).

17. R. T. Wang, J. M. Greenberg, and D. W. Schuerman,Opt. Lett. 6, 543 (1981).

18. A. Ashkin, J. M. Dziedzic, J. E. Bjorkholm, and S. Chu,Opt. Lett. 11, 288 (1986).

19. M. Kinnunen, A. Kauppila, A. Karmenyan, and R. Myllylä,Biomed. Opt. Express 2, 1803 (2011).

20. V. P. Tishkovets, E. V. Petrova, and M. I. Mishchenko, J.Quant. Spectrosc. Radiat. Transfer 112, 2095 (2011).

21. J. He, A. Karlsson, J. Swartling, and S. Andersson-Engels,J. Opt. Soc. Am. A 21, 1953 (2004).

22. M. Hammer, D. Schweitzer, B. Michel, E. Thamm, andA. Kolb, Appl. Opt. 37, 7410 (1998).

Fig. 3. (Color online) Light scattering from two RBCs in(a) a parallel and (b) a perpendicular position in relation tothe incident laser beam. The small figures show the orientationof the cells and a CCD camera image of the near-field lightscattering distribution. Polarization of 22° from the verticaldirection was used in the RBC measurements. The scale baris 10 μm.

3556 OPTICS LETTERS / Vol. 36, No. 18 / September 15, 2011