Host-Guest Accelerated Photodimerisation ofAnthracene-labeled Macromolecules in Water

Frank Biedermann, Ian Ross, and Oren A. Scherman*

Melville Laboratory for Polymer Synthesis, Department of Chemistry, University of Cambridge, LensfieldRoad, Cambridge CB2 1EW, UK. Email: oas23@cam.ac.uk

S.1 Materials and methods

All starting materials were purchased from Alfa Aesar and Sigma Aldrich and used as received unless statedotherwise. CB[8] was prepared according to literature procedure (J. Kim, I.S. Jung, S.Y. Kim, E. Lee, J.K.Kang, S. Sakamoto, K. Yamaguchi, K. Kim,J. Am. Chem. Soc.2000, 122, 540-541). N3-PEG was synthe-sised in analogy to literature procedures, see for example (S. Zalipsky, Bioconjugate Chem. 1995, 6, 150-165)and reference therein.1H and13C NMR spectra were recorded on an Avance 500 BB-ATM (500 MHz) spec-trometer, UV/visible spectra on a Varian Cary 4000 UV-Vis spectrophotometer and fluorescence spectra wererecorded on a Cary Eclipse spectrofluorometer. Gel permeation chromatography (GPC) was carried out in wa-ter on a Shodex glucose column with a Shimadzu SPD-M20A prominence diode array detector. Samples werefiltered over 0.2 mm PVDF filters before injection using a 0.6 mL / min flow rate. Rheological characterisationwas performed using an Discovery HR-2 hybdrid rheometer from TA instruments fitted with a water bath set to25 C. Strain sweep measurements were performed at a frequency of 10 rad/s. Frequency sweep measurementswere performed at a 5% strain amplitude. All measurements were performed using a 20 mm parallel plategeometry and analyzed using TA Instruments TA Orchestratorsoftware. Hydroxyethyl cellulose (HEC) waspurchased from Aldrich and dried overnight in a vacuum oven at 105C.

S.1.1 Determination of solution binding constants by ITC

Titration experiments were carried out on a VIP-ITC from Microcal Inc., at 25˚C in deionised water (Millipore,18.2 MΩ·cm). The binding equilibria were studied using a cellular CB[8] concentration of typically of 0.04 mMto which the 0.5-1.0 mM solution of1 was titrated. Typically 20-30 consecutive injections of 2.4 µL each wereused. The first data point was removed from the data set prior to curve fitting. The data was analyzed utilisingthe Origin 7.0 software package.

S.1.2 Photochemical reactions

Photoirradiation experiments were performed in a UV-box from Luzchem equipped with 10 UVA light bulbsfrom Hitachi (8 W, centered around 350 nm). The UV-lamps were"pre-warmed" for 15 min prior to anyexperiments. A Spectrosil quartz cuvette (170-2,700 nm spectral range, 3.5 mL volume, 10 mm pathlength)equipped with a magnetic stir bar, a screw-cap lid and 3 mL of an aqueous solution of the analyte (typically 10µM concentration) was placed in the middle of the chamber and irradiated under vigorous stirring for a defined

S1

Electronic Supplementary Material (ESI) for Polymer Chemistry.This journal is © The Royal Society of Chemistry 2014

time. Care was taken that in each case the positioning and orientation of the cuvette is similar. The smallestirradiation time increment was chosen to be 15 s to keep manual time stopping errors to a minimum. UV/Vis andfluorescence spectra were taken by transferring the cuvetteinto the spectrometer, making sure that an identicalorientation and positioning of the cuvette was kept for eachmeasurement. Unless mentioned otherwise, thesolutions were aerated. The reproducibility of the kinetics was confirmed by carrying out multiple runs.

S.1.3 Regioisomers formed by the photoreaction

The photodimerization experiments in the presence and absence of CB[8] were carried out in H2O, followedby removal of the solvent through lyophilization. The sample prepared in theabsenceof CB[8] was thendirectly redissolved in d6-DMSO and analysed by NMR experiments. The sample prepared in thepresenceofCB[8] was uptaken in acetonitrile, which decomplexes the dimerised anthracene species from the CB[8] host,and then CB[8] was removed by filtration (CB[n] macrocycles are fully insoluble in organic solvents). Thesolvent was removed under reduced pressure, and the sample was uptaken in d6-DMSO and analysed by NMRexperiments, which confirmed that CB[8] had been fully removed. From the1H and13C NMR spectra of thephotodimerization products in comparison to the starting material1a, it is clear that a [4+4]-type photoreactionof the anthracene moieties occurred,e.g. the 9- and 10-anthryl protons and carbons shifted upfield into thealiphatic peak region upon photoirradiation (Fig. S14). Inprinciple, four different regioisomers, depicted inFig. S5, could result as racemic mixtures upon dimerisationof 1a. Analysis of the1H and13C NMR spectra(Fig. S14) revealed that an approximately equimolar mixture of two regioisomers was formed in presence ofCB[8], whereas in the absence of host all four regioisomers,unreacted starting material and possibly some otherside products were present. The attempted structural assignment of such products by NOESY and COSY NMRwas inconclusive, however, it is reasonable to assume that the NMR peaks of the -N(CH3)3 groups are moredownfield shifted for thehead-to-headthan for thehead-to-taildimers on account of charge accumulationon one face of the molecule, see Fig. S5. Under this premise, it follows from a comparison of all NMRspectra that predominately thehead-to-taildimers were produced for the CB[8] mediated photodimerisation(tail-to-tail dimers could not be observed by1H NMR, e.g. their fraction is below 5-10% ), whereas in theabsence of the templating host the reaction was much less selective. This conclusion is also supported by1HNMR spectra in D2O in the presence of CB[8], where CB[8] has been added prior (Fig. 2 in the main text)and after the photoirradiation (Fig. S13); a significant larger number of peaks assignable to the regioisomericphotodimers were observed for the sample where photoirradiation was carried out in the absence of CB[8]. Apossible explanation for the presence of all possible regioisomeric photodimers in the absence of templatingCB[8] is the high reactivity of photoexcited1a that reacts upon diffusional encounter with another equivalentof 1a relatively irrespective to the molecular orientation of encounter. In contrast, the templating effect of theCB[8] macrocycle preselects ahead-to-tailarrangement of two1a molecules in the host cavity on accountof minimised1a-1a charge repulsion. Subsequent photoexcitation and photodimerization therefore biases theproduct mixture tohead-to-tailphotodimers.

S.1.4 Synthesis of N,N-dimethylanthracen-2-amine

N,N-dimethylanthracen-2-amine was obtained via reductive amination from 2-amino-anthracene in analogy toliterature procedures: R. Borch, A. Hassid,J. Org. Chem, 1972, 1673-1674: To a stirred slurry of 2-amino-anthracene (1.0 g, 5 mmol) and 4 mL (50 mmol) of 37% aqueous paraformaldehyde in 100 mL acetonitrile,cooled in an ice bath, was added 950 mg (15 mmol) of sodium cyanoborohydride. Glacial acetic acid (0.5 mL)was added over 10 min, the reaction mixture was warmed to roomtemperature and the reaction was stirred for2 h. An additional 0.5 mL of glacial acetic acid was added, andstirring was continued for another 1 h. Thereaction mixture was poured into 150 mL of diethyl ether and then washed with three 20-ml portions of 1NKOH and one 20-mL portion of brine. The ether solution was dried over magnesium sulfate and the solvent

S2

was evaporated under reduced pressure to yield the title compound in 95% yield. The crude material was ofsufficient purity (>90% by NMR) for the following steps and not further purified.

1H NMR (400 MHz, CDCl3): δ = 8.27 (s, 1H), 8.19 (s, H), 7.93-7.88 (m, 3H), 7.40-7.31 (m, 3H) 3.11 (s,6H) ppm.13C NMR (126 MHz, CDCl3): δ = 133.25, 132.33, 129.58, 129.04, 128.13, 127.37, 126.88, 125.77,125.10, 123.51, 122.87, 118.27,52.68 ppm. HRMS: m/z calcd for [M+H]+ 222.1278, found 222.1271 Da.

S.1.5 Synthesis of N,N,N-trimethylanthracen-2-aminium chloride (1a)

To a solution of N,N-dimethylanthracen-2-amine (0.1 g, 0.5mmol) in 20 mL acetone was added methyl iodide(0.2 ml, 2.5 mmol) and the reaction mixture was stirred underlight protection for 7 days at room temperature.The resulting precipitate was collected by suction filtration and washed with diethyl ether to yield the NMR-pure product as its iodide salt. Counterion exchange to chloride was achieved through stirring and sonicatingof an aqueous solution (ca 20 mL) of the product over freshly precipitated silver chloride (20 equiv. per iodidecounter ion) for several hours. The silver salt was filtered off and the filter cake was washed with water. Thecombined aqueous solution was freeze dried to yield the title compound as a off-white solid (74 mg, 60%).

1H NMR (400 MHz, d6-DMSO): δ = 8.79 (s, 1H), 8.74 (s, 1H), 8.66 (d, 1H, 2.5 Hz) 8.38 (d, 1H, 10.2 Hz),8.18 (mc, 2H), 8.12 (dd, 1H, 2.7 Hz, 9.5 Hz), 7.63 (mc, 2H), 3.75 (s, 9H) ppm. 13C NMR (126 MHz, d6-DMSO):δ = 143.40, 132.14, 131.81, 130.90, 129.58, 129.22, 128.07, 128.01, 127.87, 126.66, 126.52, 126.26,119.21, 117.46, 55.90 ppm. HRMS: m/z calcd for [M]+ 236.1434, found 236.1427 Da.

S.1.6 Synthesis of N,N-dimethyl-N-(prop-2-yn-1-yl)anthracen-2-aminium chloride

To a solution of N,N-dimethylanthracen-2-amine (0.1 g, 0.5mmol) in 20 mL acetone was added propagylbromide (0.25 ml, 80 wt% solution in toluene, 2.5 mmol) and the reaction mixture was stirred under lightprotection for 7 days at room temperature. The resulting precipitate was collected by suction filtration andwashed with diethyl ether to yield the NMR-pure product as its bromide salt. Counterion exchange to chloridewas achieved through stirring and sonicating of an aqueous solution (ca 20 mL) of the product over freshlyprecipitated silver chloride (20 equiv. per bromide counter ion) for several hours. The silver salt was filteredoff and the filter cake was washed with water. The combined aqueous solution was freeze dried to yield the titlecompound as a yellowish solid (71 mg, 58%).

1H NMR (400 MHz, d6-DMSO): δ = 8.79 (s, 1H), 8.74 (s, 1H), 8.68 (d, 1H, 2.5 Hz), 8.39 (dd, 1H,9.7 Hz,2.7 Hz), 8.21-8.17 (m, 2H), 8.07 (dd, 1H, 9.7 Hz, 2.7 Hz), 7.63(mc, 2H) 5.09 (d, 2H, 2.5 Hz), 3.86 (t, 1H, 2.5Hz) 3.78 (s, 4H) ppm.13C NMR (126 MHz, d6-DMSO): δ = 140.87, 132.07, 131.66, 130.87, 129.45, 128.92,127.90, 127.86, 127.79, 126.60, 126.42, 126.16, 120.67, 117.16, 82.67, 72.39, 57.44, 53.11. ppm HRMS: m/zcalcd for [M]+ 260.1434, found 260.1428 Da. FTIR:ν = 3143 (H-C≡C), 2120 (-C≡C-) cm−1.

S.1.7 Synthesis of 1b

Azido functionalised poly(ethylene glycol) monomethyl ether, N3-PEG-OMe, 2,000 g/mol, (200 mg, 0.08 mmol)and N,N-dimethyl-N-(prop-2-yn-1-yl)anthracen-2-aminium chloride (34 mg, 0.12 mmol) were dissolved underlight protection in 20 mL water and degassed with nitrogen. To this solution was added a degassed and for 5min sonicated solution of CuSO4 (1.5 mg, 0.001 mmol) and sodium ascorbate (1.8 mg, 0.001 mmol) in 5 mLwater. The resulting mixture was stirred at room temperature for 24 h under light protection. The aqueoussolution was extracted with CHCl3 (3 times 50 mL) and the combined organic fractions were washed with anaqueous EDTA solution followed by a washing step with brine and with water. All work-up steps should beperformed under light protection. The organic phase was dried over magnesium sulfate and filtered. The sol-vent was removed under reduced pressure and the solid was redissolved in a minimal amount of DCM (approx5 mL). Precipitation of the product occurred upon addition of 100 mL of diethylether. The product was kept at

S3

–4 C overnight in the freezer and collected by suction filtration. Overnight drying in the vacuum oven at 30Cyielded the title compound in 80% yield as brown-yellow solid.

20.0 25.0

re

lative

Ab

s (

30

0 n

m)

time (min)22.517.5

z = 3

z = 2

z = 1

1000 1500 2000 2500

rela

tive

in

ten

sity

m/z (Da)

3000

b)a)

Figure S1: a) ESI-MS spectra of1b in acetonitrile. b) GPC elution curve for1b in THF.

1H NMR (400 MHz, CDCl3): δ = 8.59 (s, 1H), 8.54 (s, 1H), 8.50 (s, 1H), 8.48 (d, 1H, 2.3 Hz),8.37 (dd,1H, 9.5 Hz, 2.6 Hz), 8.27 (d, 1H, 9.5 Hz), 8.06-8.02 (m, 2H), 7.56 (mc, 2H) 4.43 (t, 2H, 5.1 Hz), 4.42 (s, 6H),3.96-3.35 (PEG-backbone) ppm.13C NMR (126 MHz, CDCl3): δ = 144.43, 141.69, 135.73, 133.07, 132.60,132.22, 130.09, 129.69, 129.41, 128.50, 128.24, 128.21, 127.00, 126.80, 126.68, 124.10, 120.44, 117.50,72.00-70.00 (PEG-backbone), 68.90, 63.91, 59.03, 53.85, 50.43, 36.30, 31.91, 30.02, 29.64, 29.35, 22.68,14.12 ppm. HRMS: see Fig. S1a. FTIR: N3 stretch is absent, suggesting that the azido-groups of N3-PEG werefully consumed. GPC: See Fig. S1b.

S.1.8 Synthesis of N3-HEC

Hydroxyethyl cellulose, (2.0 g, Mw = 700,000 g/mol) was dissolved in 150 mL N-Methyl-2-pyrrolidone(NMP) at 110C under stirring. The solution was allowed to cool to room temperature and 1-chloro-3-isocyanatopropane (0.1 mL) and of one drop of dibutyl tin dilaurate (TDL) were added. The reaction wasstirred for 24 h at room temperature and then sodium azide (0.1 g) was added. The reaction mixture was heatedto 80C and stirred for another 24 h. Upon cooling to room temperature and addition of ten parts acetone, thepolymer precipitated and was collected by suction filtration. The solid was redissolved in a minimal amount ofwater and reprecipitated from acetone. The filter cake was washed with copious amounts of acetone and driedin the vacuum oven at 40C overnight to yield the title compound in 90% yield.

1H NMR: see Fig. S2a. FTIR:ν = 2103 cm−1 (-N3), Fig. S2b.

S.1.9 Synthesis of 1c

Azido functionalised hydroxyethyl cellulose, N3-HEC, (0.2 g, Mw = 700,000 g/mol) and N,N-dimethyl-N-(prop-2-yn-1-yl)anthracen-2-aminium chloride (40 mg) were dissolved under light protection in 200 mL waterand degassed with nitrogen. To this solution was as added a degassed and for 5 min sonicated solution of CuSO4

(1.5 mg, 0.001 mmol) and sodium ascorbate (1.8 mg, 0.001 mmol) in 5 mL water. The resulting mixture wasstirred at room temperature for 24 h under light protection.The solvent was reduced to approx 25 mL underreduced pressure and ten parts of acetone were added. All work-up steps should be performed under lightprotection. The precipitate formed was collected by suction filtration and washed with acetone. The solid wasredissolved in a minimal amount of water and dialysed (regenerated cellulose membrane from Spectrum Labswith Mw(cutoff) = 3500 Da) against a 0.1% brine solution and then against water for 48 h in the darkness. After

S4

freeze drying, the title compound was obtained in 85% yield as fluffy yellowish solid. From the absorbance at254 nm of a 60µg/mL stock solution, a polymer loading of approx. 80µmol of the anthracene side chains pergram of polymer was calculated which equals approx. 30 mg of the anthracene moiety per gram of polymer.

1H NMR: see Fig. S2a. FTIR: N3 stretch is absent, suggesting that the azido-groups of N3-HEC were fullyconsumed (Fig. S2b). GPC: See Fig. S3

8.0 4.0 2.0 0.06.0ppm

HOD HEC

NMP

HEC

N3-HEC

1c

100015002000250030003500

wavenumber (cm-1)

HEC

N3-HEC

1c

rela

tive

tra

nsm

issio

n in

ten

sity

a) b)

N3-stretch

Figure S2: a) 1H NMR (D2O) spectra and b) FTIR spectra of commercial HEC (700,000 g/mol), N3-HEC and1c.

26 28 30 32

re

lati

ve

Ab

s (2

54

nm

)

time (min)

Figure S3: GPC elution curve for1c in water.

S5

S.2 Supporting Tables

Table S1: Thermodynamic data for the binding of compounds1a and1b with CB[7] and CB[8]. [a] Meanvalues and estimated errors at 25C in water, pH 7.0. [b]Ka(ternary) =Ka(1)×Ka(2). Note thatKa(ternary)can be obtained with higher accuracy from the measured isotherms than the individual binding constantsKa(1)andKa(2), see the discussion in L. M. Heitmann, A. B. Taylor, P. J. Hart, A. R. Urbach,J. Am. Chem. Soc.2006, 128, 12574 for further details. [c] Total enthalpy contribution to ternary complex formation measured byITC. [d] Total entropy contribution to ternary complex formation calculated fromKa(ternary) and∆H(ternary)values.

Ka(1) Ka(2) Ka(ternary) ∆H(ternary) –T∆S(ternary)(M−1)[a] (M−1)[a] (M−2)[b] (kJ/mol)[c] (kJ/mol)[d]

CB[7] + 1a (4.1±0.5)× 105 – – –18.1±1.0 –14.0±1.5CB[8] + 1a (4±1) × 104 (2±1)× 107 (1±0.5)× 1012 –81±4 13±2CB[7] + 1b (8.0±1.0)× 103 – – n.a. n.a.CB[8] + 1b (8±1) × 104 (3±1)× 105 (2.1±1.0)× 1010 –94±5 35±5

S6

S.3 Supporting Figures

NNH2 NaBH3CN

H2CO

yield: 95%

1) CH3I2) ion exchange Cl

N

yield: 60%

Cl

N

yield: 58%

1)2) ion exchange

BrH2C

Cl

N

+

N3

OO

n

n ≈ 45

water, 48h

Cu(I) 10 mol% N

yield: 80%end group conversion ~ 90%

N

NN

ClO

On

1a

1b

1c

N3-PEG

N3-HEC

water, 48h

Cu(I) 10 mol%

R = -H, -C2H4OH,

-C2H4O(CO)NH-C2H4N3

RO

O

H

O

H

H

ORORH

OR

m

RO

O

H

O

H

H

ORORH

OR

Figure S4: Synthetic pathway towards1a, 1b and1c.

S7

Me3N

Me3N

Me3N

NMe3NMe3

Me3N

Me3N

Me3N

anti head-to-tail syn head-to-tail

anti head-to-head syn head-to-head

Figure S5: Possible regioisomeric products of the dimerisation reaction of 1a.

S8

S.3.1 Host-guest complexation and photodimerisation of small molecule anthracene 1a

a)

b)

CB[8]CB[8]

8.0 7.0 4.0 3.05.06.0ppm 9.0

HOD

Figure S6: 1H NMR spectra (D2O) of 1a (0.5 mM) in a) the presence of 0.5 equiv. of CB[8] and b) absence ofthe host.

S9

a)

b)

CB[7] CB[7]

8.0 7.0 4.0 3.05.06.0ppm 9.0

HOD

Figure S7: 1H NMR spectra (D2O) of 1a (0.5 mM) in a) the presence of 1 equiv. of CB[7] and b) absence ofthe host.

200 250 300 350 4000

0.5

1.0

1.5

2.0

Abs (

a.u

.)

wavelength (nm)

1a + CB[8]

0.0 0.1 0.2 0.3 0.4 0.5 0.6

equiv. CB[8]

0.0 0.5 1.0 1.5

rel.

Abs (

254 n

m)

0.2

1.0

equiv. CB[8]

b)

0.0

200 250 300 350 4000

0.5

1.0

1.5

2.0

Abs (

a.u

.)

wavelength (nm)

1a + CB[7]

0.0 0.4 1.0 1.4

rel.

Abs (

254 n

m)

0.2

1.0

equiv. CB[7]

a)

0.0 0.2 0.6 0.8 1.2

Figure S8: UV/vis spectra for the titration of1a (10 µM in water) with a) CB[7] and b) CB[8].

S10

400 450 500 550 600

rela

tive

em

issio

n inte

nsity

wavelength (nm)

+ CB[7]

01.0

1.2

1.4

F/F

0

0.5 1.0 1.5ratio of CB[7]:1a

400 450 500 550 600

rela

tive

em

issio

n inte

nsity

wavelength (nm)

+ CB[8]

00.0

0.2

0.4

0.6

0.8

1.0

F/F

00.25 0.50 0.75

ratio of CB[8]:1a

a) b)

1.3

1.1

Figure S9: Emission spectra (356 nm excitation) for the titration of1a (10 µM in water) with a) CB[7] and b)CB[8]. The inset shows the normalised total fluorescence as afunction of the ratio of CB[n] to 1a.

S11

0 1.0

-10

-8

-6

-4

-2

0

-2.5

-2.0

-1.5

-1.0

-0.5

0.0

30 60 90 120time / min

μca

l/se

ckca

l/m

ole

of

inje

cta

nt

molar ratio: 1a to CB[8]

0

2.0 3.0 4.0

30 60 90 120time / min

0

0 1.0

molar ratio: 1a to CB[7]

2.0 2.50.5 1.5

-4.0

-2.0

-0.2

-0.4

0.0

μca

l/se

ckca

l/m

ole

of

inje

cta

nt

0

b)a)

s

30 60 90 120time / min

0

0 1.0

molar ratio: 1b to CB[8]

2.0 3.0 4.0

-12

-8

-4

0

-2.0

-4.0

-1.0

-3.0

0.0

μca

l/se

ckca

l/m

ole

of

inje

cta

nt

30 60 90 120time / min

0

0 1.0

molar ratio: 1b to CB[7]

2.00.5 1.5

-4.0

-2.0

-0.2

-0.4

0.0

μca

l/se

ckca

l/m

ole

of

inje

cta

nt

0.0

-0.6

-0.8

d)c)

Figure S10: Representative isotherms for titration of a solution of1a or 1b (0.5 - 1.0 mM) into an aqueoussolution of CB[7] or CB[8] (40µM) at 25C. Integrated heats are shown as black squares, which were correctedfor the heat of dilution (grey circles) prior to fitting of theisotherms to a 1:1 binding model in the case of CB[7]or a stepwise binding model in the case of CB[8].

S12

400 450 500 550

rela

tive

em

issio

n in

ten

sity

wavelength (nm)600

irradiation (350 nm) of 1a + CB[8]

0 20 40 60 80 100 120t (s)

0.0

0.2

0.4

0.6

0.8

1.0

F/F

0

400 450 500 550 600

rela

tive

em

issio

n in

ten

sity

wavelength (nm)

0 20040 80 160120t (s)

0.0

0.2

0.4

0.6

0.8

1.0

F/F

0

a) b) irradiation (350 nm) of 1a + CB[7]

Figure S11: Emission spectra (356 nm excitation) upon photoirradiation (350 nm) of1a (10 µM in water)in the presence of a) 0.5 equiv. of CB[8] and b) 1 equiv. of CB[7]. The inset shows the normalised totalfluorescence as a function of the irradiation time. The solidred line shows the best monoexponential fit of thekinetic data.

200 600 1000 1400 1800

0.5

1.0

1.5

2.0

200 250 300 350 400

wavelength (nm)

a) b)

Abs (

a.u

.)

rel.

Abs (

254 n

m)

0.2

1.0

0.0

t (s)

Irradiation (350 nm) of 1a

320 400 4800.00

0.08

Abs.

(a.u

.)

Figure S12: a) UV/vis spectra for1a (10 µM in water) upon photoirradiation (350 nm). b) Normalised ab-sorbance at 254 nm as a function of irradiation time. The solid red line shows the best monoexponential fit ofthe kinetic data.

S13

a)

b)

8.0 7.0 4.0 3.05.06.0ppm

c)

d) 1a

1a photoirradiated for 1h

1a photoirradiated for 3h

1a photoirradiated for 3h + CB[8]

CB[8]CB[8]

HOD

9.0 6.08.0 7.0

9.0 6.08.0 7.0

9.0 6.08.0 7.0

9.0 6.08.0 7.0

Figure S13: a) 1H NMR spectra (D2O) of 1a (0.5 mM) after photoirradiation for 3 h and subsequent additionof 0.5 equiv. of CB[8]. b)1H NMR spectra (D2O) of 1a (0.5 mM) after photoirradiation for 3 h. c)1H NMRspectra (D2O) of 1a (0.5 mM) after photoirradiation for 1 h. d)1H NMR spectra (D2O) of 1a (0.5 mM) priorto photoirradiation. The insets show the enlarged aromaticpeak region.

S14

ppm 51.052.0 51.555.556.0ppm 115135 125145

a)

b)

c)

-N(CH3)

3

-N(CH3)

3

9 and 10 anthracyl sp3 carbon

9

10

ppm 7.07.58.08.5

a)

b)

c)

ppm 3.403.503.603.70ppm 4.704.754.80

-N(CH3)

3

9 and 10 anthracyl protons -N(CH3)

3

Me3N

NMe3

anti head-tail

+ 3 other regioisomers

Figure S14: Top 1H and bottom13C NMR spectra (d6-DMSO) of a) anthracene-dimer that was prepared viaphotoirradiation (15 min) of a CB[8]·1a2 aqueous solution, followed by freeze drying and extractionof thephotodimer with acetonitrile, b) a product mixture that wasprepared via photoirradiation (3 h) of a1a aqueoussolution, followed by freeze drying, c) reference spectra for 1a.

600 800 1000 1200 1400 1600

[CB[8]•1a2]2+

rela

tiv

e in

ten

sity

m/z (Da)

[CB[8]•1a]+

[CB[8]•1a•Na]2+

[CB[8]2•1a

3]3+

200 600 1000 1400

[CB[8]•2a]2+

m/z (Da)200 600 1000 1400

[CB[8]•2a]2+

m/z (Da)

[2a]2+

hν (350 nm) + CH3CN

a) b) c)

Figure S15: ESI-MS spectra of a 1:2 solution of CB[8] and1a in water a) prior to and b) after photoirradiationfor 15 min. c) Acetonitrile was added to the photoirradiatedmixture to decomplex the ternary CB[8] complexand the ESI-MS spectrum was measured. Absolute charges wereassigned through analysis of the isotopicpattern.

S15

250 300 350 4000.0

0.2

0.4

0.6

prior irradiation

350 nm irradiation (5 min)

reverse: 300 nm irradiation (5 min)

reverse: 300 nm irradiation (15 min)

wavelength (nm)

Ab

s (

a.u

.)

Figure S16: UV/vis spectra for1a (10 µM in water) in the presence of 0.5 equiv. CB[8] upon photoirradiation(350 nm) and subsequent irradiation with a 300 nm light source to reverse the photodimerization.

S16

S.3.2 Host-guest complexation and photodimerisation of anthracene-labeled PEG-polymer 1b

b)

200 250 300 350 400

wavelength (nm)

1b + CB[8]

a)

0

0.5

1.0

1.5

2.0

Abs (

a.u

.)

0.0 0.4 1.0 1.6

rel.

Abs (

254 n

m)

0.2

1.0

equiv. CB[8]

0.0 0.2 0.6 0.8 1.2 1.4

Figure S17: a) UV/vis spectra for the titration of1b (10 µM in water) with CB[8]. b)Normalised absorbanceat 254 nm as a function of the CB[8] equivalents.

400 450 500 550 600

rela

tive

em

issio

n inte

nsity

wavelength (nm)

+ CB[8]

0.2

0.4

0.6

0.8

1.0

rel.

inte

nsity / F

/F0 F/F

0

relative intensity (416 nm) relative intensity (500 nm)

0.0 0.5 1.0 1.25

equiv. CB[8]

0.250

a) b)

Figure S18: a) Emission spectra (356 nm excitation) for the titration of1b (10 µM in water) with CB[8]. b)Normalised total fluorescence and intensities at 416 nm (emission of monomeric1b) and 500 nm (excimeremission of1b) as a function of the added CB[8] equivalents.

S17

200 250 300 350 400wavelength (nm)

b)

0

1.5

2.0

Ab

s (

a.u

.)

450

1.0

0.5

320 400 4800.00

0.08

Ab

s. (

a.u

.)

rel.

Abs (

254 n

m)

0.2

1.0

0.0

t (s)

0 400 800 1200

200 250 300 350 400

wavelength (nm)

a)

0

0.5

1.0

Abs (

a.u

.)

rel.

Abs (

254 n

m)

0.2

1.0

0.0

Irradiation (350 nm) of 1b + CB[8]

t (s)

0 200 400 600 800 1000 1200

Irradiation (350 nm) of 1b

450

320 400 4800.00

0.08

Ab

s. (

a.u

.)

Figure S19: UV/vis spectra for1b (10 µM in water) upon photoirradiation (350 nm) in a) the presenceof0.5 equiv. CB[8] and b) in the absence of the host. The plot on the right shows the normalised absorbance at254 nm as a function of irradiation time. The solid red line shows the best monoexponential fit of the kineticdata.

400 450 500 550

rela

tive

em

issio

n in

ten

sity

wavelength (nm)

600 650

0 s 120 s 240 s 480 s 720 s

1440 s

rela

tive

em

issio

n in

ten

sity

400 450 500 550

wavelength (nm)

600 650

a) b)

0.0

0.2

0.4

0.6

0.8

1.0

F/F

0

t (s)0 200 400 600

irradiation (350 nm) of 1b + CB[8] irradiation (350 nm) of 1b

Figure S20: Emission spectra (356 nm excitation) upon photoirradiation (350 nm) of1b (10 µM in water) ina) the presence of 0.5 equiv. of CB[8] and b) the absence of thehost. The inset shows the normalised totalfluorescence as a function of the irradiation time. The solidred line shows the best monoexponential fit of thekinetic data.

S18

15 20 25 30 35 40

1b

1b + CB[8]

1b + CB[8] after photoirradiation

no

rma

lize

d A

bs

(25

4 n

m)

time (min)

Figure S21: GPC elution curves (eluent: H2O) for 1b upon addition of CB[8] and subsequent photoirradiation.The broad peak-shape is likely on account of aggregation in water and sticking to the GPC column.

a)

b)

CB[8]

CB[8]

CB[8]

CB[8]

8.0 7.0 4.0 3.05.06.0ppm

c)

HOD PEG

H3C-O-PEG

1b

CB[8]•1b2 photoirradiated

CB[8]•1b2

9.0 6.08.0 7.0

9.0 6.08.0 7.0

9.0 6.08.0 7.0

Figure S22: 1H NMR spectra (D2O) of 1b (0.5 mM) in the presence of 0.5 equiv CB[8] a) after photoirradiationfor 30 min and b) prior to photoirradiation. c)1H NMR spectra (D2O) of 1b (0.5 mM) in the absence of thehost. The insets show the enlarged aromatic peak region.

S19

a)

CB[8]

CB[8]

8.0 7.0 4.0 3.05.06.0ppm

b)

9.0 6.08.0 7.0

1b photoirradiated

9.0 6.08.0 7.0

1b photoirradiated + CB[8]

H3C-O-PEG

H3C-O-PEG

HOD PEG

Figure S23: a) 1H NMR spectra (D2O) of 1b (0.5 mM) after photoirradiation for 30 min and subsequentaddition of 0.5 equiv. of CB[8]. b)1H NMR spectra (D2O) of 1b (0.5 mM) after photoirradiation for 30 min.The insets show the enlarged aromatic peak region.

0.0

0.1

0.2

0.3

0.4

400 450 500 550

rela

tive

em

issio

n in

ten

sity

wavelength (nm)

600 700650250 300 350 400

wavelength (nm)

450 500

a) b)

Ab

s (

a.u

.)

1b photoirradiated for 1h 1b photoirradiated for 1h + 0.5 equiv. CB[8]

(1b + 0.5 equiv. CB[8]) photoirradiated for 20 min

1b photoirradiated for 1h 1b photoirradiated for 1h + 0.5 equiv. CB[8]

(1b + 0.5 equiv. CB[8]) photoirradiated for 20 min

Figure S24: a) UV/vis and b) Emission spectra (356 nm excitation) of photoirradiated1b (10 µM in water)prior to (solid black line) and after (dashed red line) addition of 0.5 equiv. of CB[8] in comparison to the spectraobtained after photoirradiation of1b in the presence of CB[8] (dashed green line).

S20

S.3.3 Gel-formation and photochemical crosslinking of 1c and CB[8].

0.4

0.8

1.2

200 250 300 350 4000

Ab

s (

a.u

.)

wavelength (nm)

1c + CB[8]

0.0 0.1 0.2 0.3 0.4 0.5

0.6

equiv. CB[8]

0.0 0.2 0.4 0.6

re

l. A

bs (

25

4 n

m)

0.2

1.0

equiv. CB[8]

0.0 0.50.30.2

b)a)

Figure S25: a) UV/vis spectra for the titration of1c (60µg/mL in water) with CB[8]. b) Normalised absorbanceat 254 nm as a function of the CB[8] equivalents.

S21

b)

200 250 300 350 400

wavelength (nm)

a)

0

0.3

0.6

Abs (

a.u

.)

rel.

Abs (

254 n

m)

0.2

1.0

0.0

Irradiation (350 nm) of 1c + CB[8]

t (s)

0 100 200 300 400 500 600

Irradiation (350 nm) of 1c

450

320 400 4800.00

0.08

Abs.

(a.u

.)

200 250 300 350 400

wavelength (nm)

450

1.2

0.8

0.4

Abs (

a.u

.)

320 400 4800.00

0.08

Abs.

(a.u

.)

t (s)

0 100 200 300 400 500 600

rel.

Abs (

254 n

m)

0.2

1.0

0.0

Figure S26: UV/vis spectra for1c (60 µg/mL in water) upon photoirradiation (350 nm) in a) the presence of0.5 equiv. CB[8] and b) in the absence of the host. The plot on the right shows the normalised absorbance at254 nm as a function of irradiation time. The solid red line shows the best monoexponential fit of the kineticdata.

1 10 100

10

100

1000

strain (%)

sto

rag

e m

od

ulu

s (

Pa

)

0.1

1

10

100

co

mp

lex v

isco

sity (

Pa

s)

hν

Figure S27: Oscillatory rheological analysis at 20C of the hydrogel formed upon addition of CB[8] to a1.0 wt% solution of1c in water. Storage modulus and complex viscosity obtained from a strain-amplitudesweep performed at 10 rad s−1. Changes upon UV-light exposure (15 min) are indicated by anarrow. Squaresrefer to the left axis and circles to the right axis. Red-symbols: prior to photoirradiation, black symbols: afterphotoirradiation.

S22

0.1

1

10

100

0.001

0.01

0.1

1

10

100

1 10 100

strain (%)

co

mp

lex v

isco

sity (

Pa

s)

sto

rag

e m

od

ulu

s (

Pa

)

hν

hν0.01

0.1

1

10

0.01

0.1

1

10

1 10 100

shear rate (s-1)

vis

co

sity (

Pa

s)

str

ess (

Pa

)

hν

0.01

0.1

1

10

100

0.01

0.1

1

10

100

1 10 100

frequency (rad s-1)

sto

rag

e m

od

ulu

s (

Pa

)

loss m

od

ulu

s (

Pa

)

a) b) c)

Figure S28: Rheological analysis at 20C of a 1.0 wt% solution of1c in water. Changes upon UV-lightexposure (15 min) are indicated by an arrow. Squares refer tothe left axis and circles to the right axis. Blue-symbols: prior to photoirradiation, green symbols: after photoirradiation. a) Storage modulus and complexviscosity obtained from a strain-amplitude sweep performed at 10 rad s−1. b) Storage and loss moduli obtainedfrom a frequency sweep performed at 5% strain. e) Steady-shear rheological measurements.

0 s

400 450 500 550

rela

tive

em

issio

n in

ten

sity

wavelength (nm)600

irradiation (350 nm) of 1c

60 s120 s180 s240 s360 s

Figure S29: Emission spectra (356 nm excitation) upon photoirradiation (350 nm) of1c (60 µg/mL in water)in the absence of the host.

S23