Development of a method of depositing monolayers of Photoreaction - - PowerPoint PPT Presentation

Development of a method of depositing monolayers of Photoreaction Center proteins on Gold Electrodes

Progress Report (Chem 449) Sylvester Zhang

Bizzotto Group University of British Columbia

May 13, 2020

1 / 30

Motivations

(a) Rhodobacter sphaeroides used as photosynthesis model and as source for biohybrid solar cells (b) Binding and measurement of protein on electrode (c) Determine the method of producing monolayer, and characterize the properties of a RC monolayer

2 / 30

Introduction

Figure: Structure of RC.Based on PDB 2J8C.

• 1. Embedded Membrane protein from Rhodobacter sphaeroides
• 2. Well studied as photosynthesis model
• 3. 3 peptides (H, L and M) surround photosynthesis factor cofactors1

1Jones, M. R. The Petite Purple Photosynthetic Powerpack. Biochm. Soc. Trans. 2009, 37 (2),

400–407.

3 / 30

Introduction

(a) Fate of an electron in an RC, and their relative energies. (b) Absorbance spectra of the Wild-Type RC. Labeled are the peaks corresponding to the Ha and P moieties in the RC.

◮ “P” center absorbs light ◮ Electrons cascade through cofactors “B” and “H” very quickly - energies

determined by orientations of B and H

◮ Stable charge separated P+Q- state lasts for 100-1000 µs

4 / 30

Introduction

Figure: Cartoon of the binding mode and electron path in a “DM” RC. The hypothesized path

• f an electron is from the protein’s P center, through the protein, and ultimately into the
• electrode. The oxidized protein is then reduced by hydroquinol, which is oxidized, as

hydroquinone (HQ[o])

◮ RC bound to electrode by a gold-thiol bond introduced by cysteines mutated

• n the ventral face of the RC

◮ Each RC is roughly 6x10x4nm ◮ P center excited, donates electron to electrode. Hydroquinone in solution then

reduces oxidized P , and allows the process to restart.

5 / 30

Introduction

Figure: Cartoon of how MCH creates a monolayer.

◮ Binding of RCs expected to be both specifically (gold-thiol bound) and

non-specifically (van der waals-esque) bound

◮ 6-Mercaptohexanol (MCH) displaces non-specifically bound RCs due to much

stronger gold-thiol bond between MCH and gold

◮ Only specifically bound RCs left on surface.

6 / 30

Methods

Figure: Schematic for deposition of RCs and MCH on electrode

◮ Clean electrode deposited in eppendorf containing solution of RCs in buffer for

1 hour

◮ Electrode then placed in eppendorf containing solution of MCH in buffer for 24

hours, and then briefly deposited in clean buffer solution

7 / 30

Methods

Figure: Lockin Technique

• 1. Measure electrochemical current - background signal (mA) from electrode -

redox species (hydroquinone) in solution

• 2. Light excites photoreaction center periodically at 13 hz - generate small

desired signal (nA/pA) from reaction center to electrode transfer

• 3. Measure periodic modulation in current - consider this photocurrent2

2Jun, D.; Beatty, J. T.; Bizzotto, D. Highly Sensitive Method to Isolate Photocurrent Signals from

Large Background Redox Currents on Protein-Modified Electrodes. ChemElectroChem 2019, 6 (11), 2870–2875.

8 / 30

Methods

(a) Gold bead formed by melting and slowly cooling a gold wire. (b) AFM image of gold(111) face

• f single-crystal gold bead.

100uM MCH was deposited on the surface for 24 hours.

◮ Single crystal gold bead formed ◮ Gold bead immersed in eppendorf containing solution of RCs in buffer for 1

hour

◮ Electrode then placed in eppendorf containing solution of MCH in buffer for 24

hours, and then briefly deposited in clean buffer solution

◮ Gold bead imaged by AFM in air, under tapping mode.

9 / 30

Concentration dependence of Photocurrent

Figure: Faradaic Current (a) and photocurrent (b) vs Potential (Ag/AgCl) for DM RCs at varying concentrations at 805 nm, vs Ag|AgCl.

◮ Faradaic currents - from oxidation of HQ at electrode - remain similar ◮ Photocurrents increase with concentration of deposited RCs

10 / 30

Concentration dependence of Photocurrent

Figure: Photocurrent vs Concentration (µM) of RCs deposited for 1 hour, at various potentials (Ag/AgCl).

◮ Steady rise in photocurrent with concentration of RCs observed at positive

potentials relative to OCP

◮ Plateaus after 2.5uM RC deposition conditions - expect to reach some kind of

saturation point.

11 / 30

Impact of MCH on Photocurrent

Figure: Faradaic current vs potential (a) and Photocurrent (nA) vs potential (b) (V) vs Ag|AgCl, for 5uM RCs deposited for 24 hours, with 1 hour of MCH treatment (pink), and without any MCH treatment (red).

◮ MCH - lowers the faradaic current consistently ◮ Photocurrent increases with more RCs, but when there’s too much RCs, it

decreases again

12 / 30

Impact of MCH on Photocurrent response time

Figure: Current (A) vs Time (s) for 5uM RCs deposited on a gold electrode without MCH treatment, believed to be a multilayer situation (left) and for a 5um RCs deposited on a gold electrode with MCH treatment, a monolayer situation (right)

◮ Light turned on at T=0 and left on, before being turned off ◮ Current rises slowly versus time for the multilayer ◮ Current does not really rise with time for the MCH-treated monolayer, and

sharp peaks - show - maybe charging?

13 / 30

Diffusion control for Photocurrent response time in multilayers

(a) Cartoon of the hypothesized mechanisms of mediator replenishment in the monolayer (left) and the multilayer (right) scenarios. In a multilayer, presumably mediator must diffuse through the multilayer to replenish RCs. (b) Faradaic current vs potential (a) and Photocurrent (nA) vs potential (b) (V) vs Ag|AgCl, for 5uM RCs deposited for 24 hours, with 1 hour of MCH treatment (pink), and without any MCH treatment (red).

◮ Cotrell equation: i ∝

√

D

√

t

◮ Linear with respect to

1

√

time(s) - possibly diffusion controlled?

14 / 30

MCH is critical for forming monolayers

Figure: Cartoon of how MCH creates a monolayer.

◮ Multilayers form initially from non-specific adsorption ◮ MCH competitively displaces non-specifically adsorbed RCs ◮ MCH is needed to obtain monolayers

15 / 30

Atomic Force Microscopy counting of RCs

(a) Single crystal gold (111) surface with 2.5uM RCs deposited. (b) Height distribution of bumps

16 / 30

Atomic Force Microscopy counting of RCs

Figure: Raw (a), and segmented (b) image of RCs on gold electrode

◮ Bumps are about 6nm tall - as expected for an RC lying on its side. ◮ Can count RCs ˜300/µm2 due to the smoothness of underlying gold/MCH. ◮ Photocurrent per RC ≈ 10 electrons per second - 3.71cm2area of electrode,

and ≈ 210 nA current

17 / 30

Atomic Force Microscopy counting of RCs

(a) AFM image of a RCs deposited for 24 hours, on gold (111) surface. Z scale ranges 30nm (b) Single crystal gold (111) surface with 2.5uM RCs deposited. Z scale ranges 20nm (c) AFM image of gold(111) face of single-crystal gold bead with 100uM MCH was deposited on the surface for 24 hours. Z scale ranges 3nm

◮ Multilayer very rough - no gold layer can be seen by AFM ◮ Compare to bare gold, and monolayer, where gold can be seen, and RCs can

be counted.

18 / 30

Future Work

(d) Marcus theory cartoon. Reorganization energy is the intersection of the parabolas. (e) Marcus theory vs MHC theory. Adapted from (3).

◮ Marcus theory - reorganization energy is intersection of two parabolas - energ

• f donor, and the acceptor

◮ Marcus-Hush-Chidsey - the acceptor energy states are the entire band of

electronic states in the gold electrode.

◮ Key prediction: No “Marcus inverted region”. 3

3Zeng, Y, Smith, R. B., Bai, P

. and Bazant, M. Z., "Simple formula for Marcus-Hush-Chidsey kinetics", Journal of Electroanalytical Chemistry 735 (2014), pp. 77-83.

19 / 30

Future Work

Figure: Marcus-Hush-Chidsey kinetics fits for the multilayer (5uM RCs for 24 hours, and no MCH), and monolayer (5uM RCs for 1 hour, and 100uM MCH for 24 hours)

◮ Marcus-Hush-Chidsey re-organization energy: 0.4eV ◮ Compare to 0.5eV calculated for re-organization energy of P+Q− in implicit

solvent and membrane lipids 4

◮ Reorganization energy of 1eV - typically a very tightly packed material, lower

values - easier to reorganize.

20 / 30

Future Work

(a) Faradaic current (a) and Photocurrent (b) vs potential for 5.0uM RCs excited at various wavelengths of light (b) Absorbance spectra of the Wild-Type RC. Labeled are the peaks corresponding to the Ha and P moieties in the RC.

◮ Wavelength of exciting light has impact on photocurrent detected ◮ Unclear if its due to absorbance - it is not known what a monolayer absorbance

spectra is - absorbance spectra in solution may not reflect absorbance of a monolayer adsorbed on gold

◮ Must determine the monolayer absorbance, and activity spectrum (action

spectrum)

21 / 30

Acknowledgements

• Prof. Daniel Bizzotto
• Dept. Microbiology and Immunology
• Ms. Tianxiao Ma
• Prof. Thomas Beatty
• Mr. Adrian Grzedowski
• Dr. Daniel Jun
• Ms. Jessica Shi
• Ms. Amita Mahej

22 / 30

Control - Photocurrent Signal Origins

(c) Excited electron donates to electrode (d) Photoreaction center flexes when excited, resulting in a 13 Hz change in background signal

◮ Electron transfer from protein to electrode ◮ Conformational flexing of reaction center changing effective concentration of

mediators near electrode

◮ 13 hz signal from excited mediator to electrode ◮ Anything else that may occur at 13hz

23 / 30

Control - RCs needed for Photocurrent

Figure: Faradaic current vs electrode potential Ag|AgCl (a) and photocurrent (b) of an electrode, one without RCs deposited(red), and one with 5.0 deposited (pink).

◮ RCs present - photocurrent detected ◮ RCs are crucial for the detection of Photocurrent

24 / 30

Control - Importance of the “P” pair

(a) Faradaic Current (a) and Photo-current (b) vs Potential (Ag/AgCl) for DM P- RCs at varying concentrations at 865 nm. (b) Cartoon of the binding mode and electron path in a “DM” RC. The hypothesized path of an electron is from the protein’s P center, through the protein, and ultimately into the electrode.

◮ P-Center removed ◮ No photocurrent ◮ Photocurrent probably from RCs - using electron pathway

25 / 30

Control - Is the Photocurrent coming from Flexing of RC?

Figure: Faradaic current (a) and photocurrent (b), of an electrode coated with MCH and RCs, with either hydroquinone (blue) or potassium ferrocyanide (red) as the mediator vs. Ag|AgCl,

◮ Faradaic currents very prominent ◮ No photocurrents for Pottasium Ferrocyanide mediator ◮ Strong photocurrent for Hydroquinone Mediator ◮ Modulating area of bare gold - not likely.

26 / 30

Diffusion Control

Figure: Current (A) vs 1

√

time(s) for 5uM RCs deposited on a gold electrode without MCH treatment - a multilayer situation

◮ Linear vs

1

√

time

◮ Diffusion Controlled

27 / 30

Open Circuit potential - monolayer vs Multilayer

Figure: Faradaic current vs potential (a) and Photocurrent (nA) vs potential (b) (V) vs Ag|AgCl, for 5uM RCs deposited for 24 hours, with 1 hour of MCH treatment (pink), and without any MCH treatment (red).

◮ OCP of monolayer: 0.1V = 0.4(SHE)-0.23 ≈ 0.15V ◮ OCP of multilayer: -0.13V - not expected. ◮ Possibly potential changes as electrons cascade through multilayer of RCs

28 / 30

Calculation of Electrons/Second ◮

qe = 1.6022e − 19C 1nA = 6.242e + 9/s 1nA = 6.242e + 9/s Ipc = 1.31082e + 12/s Aelectrode = 3.71e + 8µm2

[RCs] ≈ 300/µm2

Irc ≈ 11.77/s

◮ Detected photocurrent may have a lockin frequency dependence (using 13hz,

vs some other number)

29 / 30

Counting RCs

(a) Flat regions seen by line scanning (b) flat gold (111) with MCH

◮ Apply threshold at gold’s height (where the bulk of the flat is) ◮ Watershed filter applied ◮ Peaks are then interpreted as RCs.

30 / 30