Evaluation of Thiol Raman Activities and pKa Values Using Internally

Subscriber access provided by ORTA DOGU TEKNIK UNIVERSITESI KUTUPHANESI

Article

Evaluation of Thiol Raman Activities and pKa Values using Internally-referenced Raman-based pH Titration Nuwanthi Suwandaratne, Juan Hu, Kumudu Siriwardana, Manuel Gadogbe, and Dongmao Zhang Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.5b04241 • Publication Date (Web): 08 Mar 2016 Downloaded from http://pubs.acs.org on March 15, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Evaluation of Thiol Raman Activities and pKa Values using Internallyreferenced Raman-based pH Titration Nuwanthi Suwandaratne,+ Juan Hu,# Kumudu Siriwardana, + Manuel Gadogbe, + and Dongmao Zhang,*,+ +

Department of Chemistry, Mississippi State University, Mississippi State, MS 39762 #

Department of Mathematical Sciences, DePaul University, Chicago, IL 60604

*

Corresponding author. Email: [email protected]

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ABSTRACT: Thiols, including organothiol and thiol-containing biomolecules, are among the

most important classes of chemicals that are used broadly in organic synthesis, biological chemistry, and nanosciences. Thiol pKa values are key indicators of thiol reactivity and functionality. Reported herein is an internally-referenced Raman-based pH titration method that enables reliable quantification of thiol pKa values for both mono- and di-thiols in water. The degree of thiol ionization is monitored directly using the peak intensity of the S-H stretching feature in the 2600 cm-1 region relative to an internal reference peak as a function of the titration solution’s pH. The thiol pKa values and Raman activity relative to its internal reference were then determined by curve-fitting the experimental data with equations derived on the basis of the Henderson-Hasselbalch equation. Using this Raman titration method, we determined for the first time the first and second thiol pKa values for 1,2-benzenedithiol in water. This Raman-based method is convenient to implement and its underlying theory is easy to follow. It should therefore have broad application for thiol pKa determinations and verification.

Keywords: pH titration, Raman, thiol, pKa, titration


Page 2 of 25

Page 3 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


INTRODUCTION The thiol (RS-H) functional group is ubiquitous, existing in many organic and biomolecules. Organothiols and thiolated synthetic- and bio-polymers have been used extensively in metallic nanoparticle applications including surface enhanced Raman spectroscopy, molecular electronics, and biosensing.1-7 One key indicator of a thiol’s reactivity is its pKa value. However, reliable determination of thiol pKa is challenging. Existing methods, including potentiometry, UV-vis, NMR, and isothermal titration microcalorimetry (ITC),8-11 are essentially all indirect techniques and are often susceptible to interference. For example, pinpointing thiol pKa in multifunctional thiol-containing molecules through potentiometric pH titration is almost impossible in many cases. This is because of the interference from other ionizable functional groups such as amines. The UV-vis method monitors the UV-vis spectral change as a function of solution pH. It requires the thiol-containing molecule to contain a chromophore that is sensitive only to the thiol’s ionization. This imposes a significant constraint on the general application of this method. Indeed, we recently demonstrated that the pKa values of 1,4-benzenedithiol (1,4-BDT) determined with this UV-vis method are highly erroneous.12 Only one 1,4-BDT thiol can be deprotonated even in highly concentrated (2 M) NaOH solution. This is in contrast to the first and second pKa values of 6 and 7.7 reported on the basis of UV-vis titration of 1,4-BDT.13 The more recent ITC method deduces the thiol pKa value on the basis of the kinetics of thioether formation reactions as a function of the solution pH. A relatively large number of control and sample measurements must be carefully conducted in order to get the reaction rate constants at each of the pH values. Consequently, this technique involves laborious sample preparations, measurements, and intensive data analysis.11



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

About two decades ago, Thomas et al. proposed a Raman-based method for determining thiol pKa values by monitoring the S-H intensity in a series of thiol-containing solutions that have the same concentration but different pHs.14 The thiol pKa value corresponds to the solution pH at which the S-H stretching intensity is exactly half of that for the non-ionized thiol. This Raman-based method enables direct visualization of the degree of thiol ionization. This is because the peak position of the S-H feature is around ~2600 cm-1, well-separated from other Raman peaks in Raman spectra obtained with essentially any samples. However, the experimental procedure of this original Raman method is cumbersome. It employs nitrogen in air as the external reference for correcting S-H peak intensity variations induced by measurement conditions. Although this external reference enables correction of S-H peak intensity variations induced by the fluctuation in laser excitation power and detection sensitivities, it is unable to compensate for signal variation induced by changes in photon excitation and collection efficiencies and in sample concentrations. Therefore, care must be taken to ensure the Raman sampling geometries to be exactly the same for different samples. Demonstrated herein is a Raman titration method for thiol pKa and Raman activity determination for both mono- and di-thiols in water. Instead of using nitrogen in air as the external reference, we employed internal references for determining the peak intensity ratio of the S-H peak relative to its internal reference peak. Since the internal reference can compensate for S-H signal variation induced by sample dilution, this method enables us to use Raman-based pH titration to determine the thiol pKa with a single thiol-containing solution. In addition, we have presented in this work the detailed theoretical justification of this internal referenced Raman technique for thiol pKa value determination of both monothiols and dithiols.


Page 4 of 25

Page 5 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


The internal reference used for the thiol pKa and Raman activity determination can be a Raman peak from the thiol-containing molecule itself, or a Raman peak from an exogenous chemical such as Na2SO4 that has a relatively intense and well-defined Raman feature in the 980 cm-1 region. The key criteria for choosing the internal reference are that 1) the reference peak can be readily identified and quantified, and 2) the Raman cross-section of the internal reference peak should be insensitive to the solution pH. These criteria are to ensure that the peak intensity ratio between the S-H stretch and its internal reference peak differs when S-H is ionized. The model thiols (Figure 1) used in this work include monothiols cysteine (Cys), glutathione (GSH), 2-mercaptoethanol (2-ME), cysteamine (CyA) and sodium 3-mercapto-1propanesufonate (3-MPS), and dithiols dithiothreitol (DTT), 1,4-benzenedithiol (1,4-BDT) and 1,2-benzenedithiol (1,2-BDT). It is noted that these thiols also differ significantly in their water solubility as non-ionized thiols, which is important in order to explore the general applicability of this method. Indeed, the non-ionized 1,2- and 1,4-BDT have very low solubility (< 1 mM) in water, but their deprotonated counterparts are soluble enough for Raman spectral acquisition. Experimentally, the thiol pKa values of the water-soluble thiols were determined by NaOH titration of non-ionized thiol dissolved in water, while the pKa of the benzenedithiol isomers were evaluated by HCl titration of thiol dissolved in 1 M NaOH.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 25

Figure 1: Structures of the model monothiols and dithiols.

THEORETICAL CONSIDERATION Monothiol pKa. This internally-referenced Raman titration method for monothiol pKa determination can be mathematically justified by using the standard Henderson-Hasselbalch equation (Eq. 1) in combination with Eq. 2 and Eq. 3. The latter two equations define the peak intensity ratios of the S-H stretching feature versus its internal reference peak for non-ionized thiols and partially ionized thiols at specified pH, respectively.

pH = pK a + log

R0 =

R pH =

[ RS− ]pH

Eq. 1

[ RS − H ]pH

0 ISH [RS − H]0 σS− H K [RS − H]0 σS− H = = 0 I ref [REF]0 σ ref K [REF]0 σ ref

pH [ RS − H ]pH σ S− H K I SH = pH I ref [ REF]0 σ ref K

=

[ RS − H ]pH σ S− H [ REF]0 σ ref


Eq. 2

Eq. 3

Page 7 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


pH = pK a + log

R 0 − R pH R pH

Eq. 4

0 I SH and I 0ref are the S-H stretching and internal reference peak intensities under

pH pH and I ref are the S-H stretching and conditions that all thiol groups remain non-ionized. I SH

internal reference peak intensities of the partially ionized thiol at specified pH. K and KpH relate to instrument and sampling parameters that affect the signal intensity as a function of the analyte concentration. These parameters include sample dilutions induced by the titrants, excitation laser powers, Raman photon collection efficiency, and the spectral integration times. [ RS − H ] pH and

[ RS − ] pH are the concentrations of the non-ionized and ionized thiols, respectively, at specified pH conditions. [ RS − H ]0 and [REF ] 0 are the concentrations of the thiol and internal reference under conditions that all thiol groups remain non-ionized. Importantly, [ RS − H ]0 and [REF ] 0 can be identical if a different Raman feature from the same thiol molecule is used as the internal reference. σ SH and σ ref are the Raman cross-sections of the thiol stretching mode and its inner reference feature peak. Eq. 4 implies that the thiol pKa is equal to the solution pH at which the intensity ratio of S-H stretching versus its internal reference is exactly half of that when all thiols are non-ionized. In this case, R pH = R 0 / 2 . However, pinpointing the “half point” is challenging. This is because R 0 can only be approximated by the peak intensity ratio of S-H versus the internal reference in solutions for which the pH is significantly lower than thiol pKa value. However, obtaining high quality Raman spectra with non-ionized thiols can be challenging because non-ionized thiols usually have lower solubility than ionized ones. Furthermore, exactly locating the “half point” is statistically impossible in realistic experimental measurements. This is due to the difficulty of



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 25

controlling the degree of thiol ionization induced by each titration. Eq. 4 can be rearranged into Eq. 5 (Derivation is shown in Supporting Information) in which the pKa and R 0 values can be directly obtained by fitting the experimental data of R pH as a function of solution pH using the standard Boltzmann function. No experimental R 0 is needed and all the titration data are utilized for determining the thiol pKa values. This curve-fitting can be performed using commercial software such as Origin Pro by choosing Boltzmann function (Eq. 6) as the fitting function and presetting the values of d and A2 in Eq. 6 as 1/2.303 and 0, respectively. We termed this data fitting procedure as constraint Boltzmann function fitting because d and A2 are preset to those values to ensure Eq. 6 is identical to Eq. 5. A simulated Raman-based pH titration curve derived according to Eq. 5 was shown in Figure 2. Apparently, one can readily pinpoint the monothiol pKa1 values graphically by monitoring thiol Raman intensity over a wide pH range. However, in practice, this visual examination method can be problematic because of the measurement errors. In contrast, the least-squares curve fitting method is much more robust because it gives the statistically most likely thiol pKa value that best explains all the titration data.

R

pH

y=

=

R0 1 + e2.303( pH − pKa )

A1 − A2 + A2 1 + e (x − x0 )/d


Eq. 5

Eq.6

Page 9 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 2: Simulated Raman-based pH titration curve plotted using Eq.5. (pKa =9, R0 =1)

While the pKa value reflects the acidity of the thiol group, one can derive the thiol Raman cross-section from R 0 . Indeed, rearranging Eq. 2 leads to Eq. 7 and further to Eq. 8 in which [thiol] represents the total concentration of non-ionized and ionized thiols. The reason Eq. 7 is equivalent to Eq. 8 is that titration and thiol ionization does not change the concentration ratio of the internal reference and the total thiols, regardless of whether the thiols are totally non-ionized, partially or completely ionized. As a result, one can simply use the thiol and internal reference sample concentration in the initial titration solution for the thiol Raman activity calculation. Eq. 8 indicates that after obtaining R 0 from the curve-fitting, one can readily determine the thiol relative Raman cross-section defined as the Raman cross-section ratio of the thiol and the internal reference peak.

σ S −H = R0

σ S −H = R0

[REF]0 σ ref

Eq. 7

[RS − H]0

[REF] pH σ ref [thiol] pH

0 or σ S − H = R

[REF]ini σ ref


[thiol]ini

Eq. 8


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 25

Dithiol pKa(s). The above theoretical consideration is for monothiol (one S-H group) pKa. For molecules that contain multiple thiols, the pKa value for individual thiols can also be resolved by using this internally-referenced Raman titration method. For the sake of simplicity, we just give the functional forms for curve-fitting determination of the first and second pKa for molecules that contain two thiols. Detailed derivation of Eq. 9 is shown in the Supporting Information.

R

pH

R 0 + α R 0 10 ( pH − pKa1 ) = 1 + 10 ( pH − pKa1 ) (1 + 10 ( pH − pKa2 ) )

Eq. 9

α is the ratio of the S-H Raman cross-section of the monothiolated versus non-ionized dithiols. This value is 0.5 as deprotonation of the first thiol has no effect on the Raman crosssection of the remaining thiol in the dithiol. R 0 and R pH are the peak intensity ratios of the S-H stretching feature versus its internal reference peak for non-ionized dithiol and partially ionized dithiol at specified pH, respectively. pKa1 and pKa2 are the first and second acid dissociation constants of the dithiol, respectively. The only experimental parameter needed for curve-fitting determination of the first and second thiol pKa values, together with α and R 0 , are the RpH values as a function of solution pH. A computer program for curve-fitting determination of these parameters was developed using R (Supporting Information), a computer software package commonly used in statistical data analysis. Importantly, pKa1 and pKa2 values can be approximated graphically or by fitting the experimental data in different pH regions using the standard Boltzmann function but different constraints (Figure 3). When solution pH is significantly smaller than pKa2 but covers the region in which the first thiol ionizes, Eq. 9 can be simplified into Eq. 10, which is equivalent to Eq. 11.


Page 11 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


The latter is equivalent to the standard Boltzmann function (Eq. 6) in which A1 is R0 for the curve-fitting and the ratio between A2/A1 is α.

R

pH

R

pH

R 0 + αR 0 10 (pH − pKa1 ) = 1 + 10 (pH − pKa1 )

Eq. 10

R 0 − αR 0 = + αR 0 2.303(pH − pKa1 ) 1+ e

Eq. 11

If the solution pH is equal to pKa1, and is significantly smaller than pKa2, Eq.10 is simplified into Eq. 12.

R

R 0 + αR 0 = 2

pKa1

Eq. 12

When the solution pH is significantly larger than pKa1 and covers the region in which the second thiol ionizes, Eq. 9 can be further simplified into Eq. 13; this is again a constrained Boltzmann function.

R

pH

=

αR 0 1 + e 2.303( pH − pKa2 )

Eq. 13

Under conditions in which the solution pH is equal to pKa2 but significantly larger than pKa1, Eq. 13 can be further simplified into Eq. 14.

R

pK a2

αR 0 = 2

Eq. 14

Figure 3 shows the series of simulated Raman based pH titration curves in which pKa1, R0, and α were kept to be 6.0, 1.0, and 0.5, respectively, but the value of pKa2 changes from 7, 8,



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 25

9, 10, and 11, respectively. Evidently, one can estimate the pKa1 and pKa2 values graphically if the both thiols can be ionized and their pKa differs by 3 pKa units or more.

Figure 3: Simulated Raman-based pH titration curve plotted using Eq. 9. ∆pKa is the difference between pKa2 and pKa1. ∆pKa was changed from 1 to5. (pKa1= 6, R0 = 1 and α = 0.5). Once R0 and α is known, the S-H Raman cross-section in the non-ionized ( σ Sintact - H ) and monothiolated ( σ Smono - H ) dithiols can be calculated using Eq. 15, and Eq. 16.

0 σ Sintact -H = R

[REF]ini σ ref [ dithiol]ini

intact σ Smono -H = α σ S -H


Eq. 15

Eq. 16

Page 13 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


EXPERIMENTAL

Chemicals and Equipment. Dithiothreitol and sodium 3-mercapto-1-propanesulfonate were purchased from Fluka and 1,4-benzenedithiol from Alfa Aesar. All other chemicals were purchased from Sigma-Aldrich. Nanopure water (18.2 MΩ cm-1) was used throughout the experiments. Normal Raman spectra were obtained with a Horiba LabRam HR 800 confocal Raman microscope system and a 633 nm Raman excitation laser with 13 mW laser power. The acquisition time for Cys, GSH, 3-MPS, and 2-ME is 20 s, and 50 s for CyA and the dithiols. The spectra shown in the figures are an average of 10 individual spectra. The RamChip slides used for the Raman spectral acquisition were obtained from Z&S Tech. LLC. The solution pHs were measured using a calibrated AB 15, Accumet pH electrode. Origin Pro was used for curve fitting.

Raman Titrations of Soluble Thiols. Known amounts or saturation quantities of water-soluble thiols were dissolved in 2 ml aliquots of 150 mM Na2SO4 solution. Excess thiol in each solution was removed through membrane filtration using a pore size of 0.2 µm in diameter. Each thiol solution was titrated with a 500 mM NaOH solution using a 10 µL pipette. The solution mixture was vortexed after each NaOH addition, followed with pH measurement and Raman acquisition.

Raman Titrations of Thiols with Poor Water Solubility. Both 1,2- and 1,4- benzenedithiols are poorly soluble in water in their non-ionized forms. However, their solubility in basic solution increases significantly. 1,4- benzendithiol was first added into 2 mL of 1 M NaOH followed with filtration removal of undissolved thiol. The saturated thiol-containing solution was then titrated using 1 M HCl solution. As 1,2- benzendithiol is a liquid, 20 µL was dissolved in 2 mL of 1 M NaOH and then titrated using 1 M HCl solution. Solutions were vortexed after each HCl addition, followed with pH measurement and Raman acquisition.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

RESULTS AND DISCUSSIONS

Thiol pKa of Monothiols. Figure 4 shows the Raman based pH titration data obtained with the model monothiols 3-MPS and Cys in which the Na2SO4 was added as the exogenous internal reference. Experimental results obtained with other monothiols are shown in the supporting information. Evidently, it is challenging to estimate the thiol pKa value for these organothiols on the basis of their pH titration curves (pH value as a function of the volume of the titrant) (Figure 4 and Figure S1, S2, and S3 in Supporting Information). It’s because the titration curve does not go beyond pH 14, thus the end point of the graph cannot be identified for the pKa determination. In contrast, one can readily determine the pKa value by Boltzmann function curve-fitting the Raman titration data that is the intensity ratio of the S-H stretching peak versus its internal reference as a function of the solution pH (Figure 4C and Figure S1(C), S2(C), and S3(C) in Supporting Information).


Page 14 of 25

Page 15 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 4: Raman-based pH titration for pKa determination of the thiol group in (top row) 3MPS and (bottom row) Cys. Representative Raman spectra of (A) 3-MPS and (D) Cys as a function of solution pH. The spectra are normalized by the sulfate peak intensity. The dotted line indicates the peak positions associated with the specified functional group. ((a) SO42-, (b) RSO3- , (c) S-H, and (d) C-H peak). Solution pH (red circles, left scale) and IS-H/ISO42-(black squares, right scale) as a function of titrant (500 mM NaOH) volume for (B) 3-MPS and (E) Cys. Intensity ratio of S-H stretching feature versus its internal reference peaks as a function of solution pH for (C) 3-MPS and (F) Cys. The red circles, black squares and blue triangles in (C) correspond to peak intensity ratio of S-H stretch versus its internal reference peak of the sulfate, C-H and sulfonic peaks, respectively. For ease of comparison the ratios were normalized to the ratio with sulfate. The solid curve, R0, and pKa shown in (C) and (F) are the Boltzmann function curve-fitting output of the experimental data with the Origin Pro. The 3MPS, Cys, and Na2SO4 concentrations in the initial solutions are 0.68 M, 0.83 M, and 0.15 M, respectively.

The thiol pKa values obtained with 3-MPS using three different internal reference peaks are highly similar (9.94 ±0.04). These internal references include the 980 cm-1 SO42- peak from the exogenous Na2SO4, the C-H stretching feature (2930 cm-1) and -SO3- from 3-MPS itself. The



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 25

fact that almost identical thiol pKa values are obtained with the different reference peaks indicates the robustness of this internal-referenced Raman titration technique. Excellent Raman spectra were acquired with Cys through a wide pH titration range (Figure 4D). The Raman spectra obtained with pH 5.2 and 7.5 solutions are highly identical, both containing significant S-H stretching feature. This result indicates that Cys thiol pKa is significantly higher than 7.5.

This, in combination with the observation that S-H peak

completely disappeared in the Raman spectra obtained with the pH 10.8 solution indicates the Cys thiol pKa must be between pH 7.5 and 10.8. Exact Raman titration determination of the Cys thiol pKa is achieved through the curve-fitting the experimental Raman intensity as a function of solution pH (Figure 4F). It is noted that it can be highly unreliable to determine the Cys thiol pKa using the pH titration curve alone (Figure 4E). This is because Cys has three ionizable groups (-COOH, -SH, and –NH3+), which makes it difficult to pinpoint and assign the inflection points in the pH titration curves for the determination of pKa values of these ionizable group. This explains the complexity of the Cys pH titration curve shown in Figure 4F. In contrast, with the Raman titration, one can directly measure the degree of S-H ionization. Another problem, which is common for all pKa determination using pH titration alone is the difficulty in pinpointing the inflection points in the pH titration curves obtained with weak acids with pKa larger than 7. These inflection points are most commonly taken as the half-titration point where the solution pH rises most rapidly as a function of titrant volume.

Since it is not practical to extend a pH

titration curve beyond pH~14, conventional pH titration method for pKa determination for weaker acid can be highly unreliable.


Page 17 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 5. Internally-referenced Raman pH titration for determination of the thiol pKa values in DTT. (A) Representative Raman spectra of DTT as a function of solution pH. ((a) SO42-, (b) SH peak) The spectra are normalized by the sulfate peak intensity. (B) Solution pH (red circles, left scale) and IS-H/ISO42-(black squares, right scale) as a function of titrant (500 mM NaOH) volume. (C) Curve-fitting to determination of the first and second thiol pKa values for DTT, with different pKa values. (Case 1: pKa1 = 8.9, pKa2 = 10.8, Case 2: pKa1 = 8.9, pKa2 = 10.3, Case 3: pKa1 = 8.4, pKa2 = 10.3). Case 2 is by least-square curve-fitting the experimental data with the Eq. 9 using the R program shown in the supporting information that has the negligible fitting error. The R0 and α are the curve-fitting outputs. The DTT and Na2SO4 concentrations in the initial titration solution are 0.65 M and 0.15 M, respectively.

Thiol pKa values of dithiols. The Raman titration determination of the thiol pKa values of water-soluble dithiol was demonstrated with DTT as the model dithiol (Figure 5). Only one single S-H peak is observed for the non-ionized and mono-ionized species, indicating that the two thiol groups are almost identical with no strong coupling. The first and second DTT thiol pKa values, R0, and α are 8.9, 10.3, 1.83, and 0.49, respectively. These values were obtained by least squares curve-fitting the Raman titration data in Figure 5(C) with Eq. 9 using the R program developed in this work. The pKa values are in excellent agreement with the literature reported DTT thiol pKa values (9.2, and 10.1 for pKa1 and pKa2, respectively).15 Equally important is that this least-square curve-fitting method is very effective in pinpointing the correct pKa values. Large errors appear even when only one of the two thiol pKa values deviates from



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 25

the least-square value by as small as by 0.5 pKa (Figure 5C). It is also important to note that the pKa values obtained with the peak intensity ratio are very similar to those obtained with curvefitting the integrated peak area ratios (Figure S5, Supporting Information). Since the thiol pKa values were determined by least-square fitting of many Raman titration data points, the effect of random errors in the individual peak intensity ratios on the thiol pKa values is insignificant. The operational procedure of this internally-referenced Raman method is very straightforward for the water-soluble thiol as shown with Figures 4 and 5. However, many organothiols have low water solubility in their non-ionized form. Direct Raman titration of nonionized thiol in water with NaOH can be difficult because the initial amount of thiol that can be dissolved in solution is limited. In this case, it can be advantageous to use a reverse titration strategy in which the thiol is first dissolved in NaOH solution followed by HCl titration. This is because thiolates have higher solubility in water than non-ionized thiols. The pKa values of 1,4- and 1,2-BDT were explored with this reverse titration strategy (Figure 6) in which both BDT isomers were first dissolved in 1 M NaOH followed by HCl titration. Figure 6 (A) and (D) show the representative Raman spectra for 1,4- and 1,2- BDT, respectively, both as a function of solution pH. The fact that there is a relatively intense and constant S-H feature in the 1,4-BDT solution when its pH changed from 10 to 13.6 indicates only one 1,4-BDT thiol is deprotonated even when the solution pH is as high as ~14. This result argues against the belief that the two 1,4-BDT thiols have similar pKa values (pKa1=6, pKa2=7.7),13 but it is consistent with our recent work conducted with a series of para-aryl dithiols (HS-(C6H5)n-SH, n=1, 2, and 3). The latter work showed that only one PADT thiol can be ionized even when PADT is reacted with molten NaOH at elevated temperature.12 The fact that


Page 19 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1,4-BDT remains almost exclusively as monothiolate in the pH~13.6 solution indicates that the pKa for the second 1,4-BDT thiol must be significantly higher than 14 in water. Both non-ionized and monothiolated 1,4-BDT appear in the pH ~8.2 solution, indicating the first pKa is close to pH value. However, Raman pinpointing of the first pKa value for 1,4BDT thiol is not possible because of the following reasons. Obtaining Raman spectra for 1,4BDT at a solution pH lower than 8 is difficult because the 1,4-BDT solubility at neutral and acidic pH is too low. The titration solutions at these pHs become very cloudy, making reliable Raman acquisition impossible. The S-H stretching peak Raman shift in the non-ionized and the monothiolated 1,4-BDT in water are at 2580 cm-1, and 2480 cm-1, respectively. This is determined on the basis of the 1,4BDT Raman spectrum acquired with pH~8.2 sample. This data shows that the vibrational frequency of S-H stretching in the monothiolated 1,4-BDT is significantly lower than that in its non-ionized counterpart. It indicates that deprotonation of one thiol in the non-ionized 1,4-BDT has significant impact on the Raman feature of the remaining S-H bond. This red-shift of the S-H feature in the monothiolated 1,4-BDT is consistent with a previously proposed hypothesis that the remaining S-H in monothiolated 1,4-BDT is an effective hydrogen bond acceptor, and the latter is responsible for the excellent dispersion stability of AuNPs (gold nanoparticles) functionalized with 1,4-BDT.12 This conclusion is also consistent with the observation that hydrogen bonding induces red-shift of the O-H stretch in water.16,17 The remarkable differences between the 1,4-BDT pKa values obtained with Raman titration method described in this work and that obtained with the UV-vis titration method in ref. 13 highlights the advantage of this Raman titration method for thiol pKa determination and verification. As discussed in the introduction section, the UV-vis titration method is applicable



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 25

only to chromophore-containing thiols in which only thiol protonation/deprotonation can induce the observed UV-vis spectral changes. The fact that UV-vis titration gave unreliable thiol pKa values for 1,4-BDT is likely due to the presence of competing processes that contributed to 1,4BDT UV-vis spectral change as a function of solution pH. In contrast, this Raman titration method enables direct visualization of the degree of the deprotonating of the thiol group as the functional of solution pH, regardless of pH-dependent physical and chemical changes that may occur in another molecular moiety in the thiol-containing molecules. The two thiols in 1,2-BDT are both deprotonated in the 1 M NaOH solution (Figure 6(F)). The first and second thiol pKa values for 1,2-BDT are 6.4 and 10.9, respectively. This is in sharp contrast to 1,4-BDT for which only the first thiol can be ionized in water with a pKa ~ 8.2. Indeed, the first thiol pKa value of 1,4-BDT must be higher than that for 1,2-BDT. Otherwise, no significant non-ionized 1,4-BDT Raman feature should appear in the Raman spectrum obtained with the pH ~8.2 solution. The pKa values for 1,2-BDT in water have, to our knowledge not been reported before. Only one 1,2-BDT thiol pKa value (6.8) was reported and that is for 1,2BDT dissolved in dimethyl sulfoxide.18 It is unclear whether this literature thiol pKa refers to the average thiol pKa for the two BDT thiols or for the first thiol.


Page 21 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 6: Internally-referenced Raman-based pH titration for thiol pKa determination in (top row) 1,4- and (bottom row)1,2-BDT. Representative Raman spectra of (A)1,4-BDT, and (D) 1,2-BDT as a function of solution pH. ((a) S-H, (b) C-H peak). The spectra are normalized by the C-H peak intensity. Solution pH (red circles, left scale) and IS-H/IC-H (black squares, right scale) as a function of titrant (1 M HCl) volume for (B) 1,4-BDT and (E) 1,2-BDT, respectively. The relative S-H peak intensity as a function of solution pH for (C) 1,4-BDT and (F) 1,2-BDT, respectively. The solid curve and the data shown in (F) is the least-squares curve-fitting output of the experimental data with the Eq. 9 using the R program (Supporting information).

Table I: Thiol pKa values and Raman activity Thiols Thiol pKa This work Literature value 3-MPS Cys GSH CyA 2-ME DTT 1,2-BDT 1,4-BDT

9.9±0.04 8.6±0.04 9.2±0.3 8.4±0.05 9.8±0.2 8.9 (pKa1) 10.3 (pKa2) 6.5 (pKa1) 10.9 (pKa2) < 8.2 (pKa1) >14 (pKa2)

Relative thiol Raman activity (σS-H/σSO42- )

Not available 8.3± 0.219 9.2 ± 0.1520 8.3521 9.722 9.2 (pKa1)15 10.1(pKa2)15 6.8 in DMSO18

0.21 0.22 0.23 0.16 0.23 0.42 (for non-ionized DTT), 0.21 for monothiolated DTT Not available

6 (pKa1)13 7.7 (pKa2)13

Not available



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 25

Table I summarizes the thiol pKa values and their relative Raman activity for the monoand di-thiols explored in this work. The relative Raman activities are represented by the Raman cross-section ratio of thiol versus SO42- peak at 980 cm-1. The thiol Raman activity in 1,2-and 1,4-BDT are not available because of the difficulty in determining the two BDT concentrations in their respective titration solution. The pKa values of the water-soluble thiols determined with this method are very similar to the literature reported values. However, there is a large difference in the thiol pKa values determined in this work versus those reported for both of the two BDT isomers. This highlights the challenges in reliable thiol pKa determination. Fruitfully, the Raman based method allows direct visualization of the degree of thiol ionization on the basis of the characteristic S-H feature. Therefore, it provides a simple means for one to evaluate the reliability of the literature thiol pKa values. Except for the thiol in CyA, all the thiols in the water-soluble thiols have similar Raman activities in which their relative Raman cross-section ratio of thiol versus SO42- is ~0.22. This result is not surprising because the thiol groups are all directly linked to a saturated carbon. The reason that the CyA thiol has the lowest pKa value and relative Raman activity among the tested water-soluble thiols is likely due to the presence of a positively charged primary amine group in CyA. The intramolecular electrostatic interaction facilitates the S-H ionization that explains the lower CyA thiol pKa in comparison to the other thiols. This electrostatic interaction should also reduce the electron density surrounding the S-H group, leading to a reduced polarizability of this functional group and its lower Raman activity. CONCLUSIONS In summary, we have presented in this work both the theoretical background and the experimental procedure including data analysis for Raman titration determination of the thiol


Page 23 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


pKa values of both mono- and di-thiols in aqueous solutions. The application of this technique for water-soluble thiols is straightforward by direct titration of the thiols dissolved in water using NaOH. pKa determination of poorly water soluble organothiols can be conducted by first dissolving the organothiols in NaOH followed by HCl titration. This internally-referenced Raman titration method is simple to implement and its underlying theory is easy to follow. It should have broad applications for thiol pKa determination and verification of literature thiol pKa values.

ASSOCIATED CONTENT Supporting Information. Derivation of Eq. 5 in the manuscript, derivation of Eq. 9 in the manuscript, R code for least squares curve-fitting with Eq. 9, Raman-based pH titration for the thiol pKa determination for Glutathione, Raman-based pH titration for the thiol pKa determination for 2-Mercaptoethanol, Raman-based pH titration for the thiol pKa determination for Cystamine, comparison of the thiol pKa determination for DTT using the peak intensity ratio and integrated peak area ratio. This material is available free of charge via the Internet at http://pubs.acs.org.

ACKNOWLEDGEMENT This work was supported by NSF funds (CHE 1151057 and EPS-0903787) provided to D.Z.

REFERENCES (1) Vangala, K.; Ameer, F.; Salomon, G.; Le, V.; Lewis, E.; Yu, L.; Liu, D.; Zhang, D.: J. Phys. Chem. C 2012, 116, 3645-3652. (2) Lee, K. J.; Nallathamby, P. D.; Browning, L. M.; Osgood, C. J.; Xu, X.-H. N.: ACS nano 2007, 1, 133-143. (3) Nie, Z.; Petukhova, A.; Kumacheva, E.: Nat. Nanotechnol. 2010, 5, 15-25.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 25

(4) Graham, D.; Thompson, D. G.; Smith, W. E.; Faulds, K.: Nat. Nanotechnol. 2008, 3, 548-551. (5) Erathodiyil, N.; Ying, J. Y.: Acc. Chem. Res. 2011, 44, 925-935. (6) Orendorff, C. J.; Gole, A.; Sau, T. K.; Murphy, C. J.: Anal. Chem. 2005, 77, 32613266. (7) Thompson, D. G.; Faulds, K.; Smith, W. E.; Graham, D.: J. Phys. Chem. C 2010, 114, 7384-7389. (8) DeCollo, T. V.; Lees, W. J.: J. Org. Chem. 2001, 66, 4244-4249. (9) Mössner, E.; Iwai, H.; Glockshuber, R.: FEBS Lett 2000, 477, 21-26. (10) Van Horn, J. D.; Bulaj, G.; Goldenberg, D. P.; Burrows, C. J.: J. Biol. Inorg. Chem. 2003, 8, 601-610. (11) Tajc, S. G.; Tolbert, B. S.; Basavappa, R.; Miller, B. L.: J. Amer. Chem. Soc. 2004, 126, 10508-10509. (12) Gadogbe, M.; Chen, M.; Zhao, X.; Saebo, S.; Beard, D. J.; Zhang, D.: J. Phys. Chem. C 2015, 119, 6626-6633. (13) Armstrong, D. A.; Sun, Q.; Tripathi, G. N. R.; Schuler, R. H.; McKinnon, D.: J. Phys. Chem. 1993, 97, 5611-5617. (14) Li, H.; Hanson, C.; Fuchs, J. A.; Woodward, C.; Thomas, G. J.: Biochemistry 1993, 32, 5800-5808. (15) Szajewski, R. P.; Whitesides, G. M.: J. Am. Chem. Soc. 1980, 102, 2011-2026. (16) Auer, B.; Kumar, R.; Schmidt, J.; Skinner, J.: Proc Natl Acad Sci 2007, 104, 14215-14220. (17) Liu, Y.; Ojamäe, L.: J. Mol. Model. 2014, 20, 1-8. (18) Yu, H.-Z.; Yang, Y.-M.; Zhang, L.; Dang, Z.-M.; Hu, G.-H.: J. Phys. Chem. C 2014, 118, 606-622. (19) Krekel, F.; Samland, A. K.; Macheroux, P.; Amrhein, N.; Evans, J. N. S.: Biochemistry 2000, 39, 12671-12677. (20) Tang, S.-S.; Chang, G.-G.: J. Biochem. 1996, 119, 1182-1188. (21) Connett, P.; Wetterhahn, K.: J. Am. Chem. Soc. 1986, 108, 1842-1847. (22) Jencks, W. P.; Salvesen, K.: J. Am. Chem. Soc. 1971, 93, 4433-4436.


Page 25 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


For Table of Contents Only:


Evaluation of Thiol Raman Activities and pKa Values Using Internally

Recommend Documents