Skip to main content
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2015 May 19.
Published in final edited form as: Synth Commun. 2014;44(16):2336–2343. doi: 10.1080/00397911.2014.894527

PRACTICAL SYNTHESIS OF AROMATIC DITHIOCARBAMATES

Panuwat Padungros 1, Alexander Wei 2
PMCID: PMC4437221  NIHMSID: NIHMS679607  PMID: 25999616

GRAPHICAL ABSTRACT

graphic file with name nihms-679607-f0001.jpg

Oxidation-sensitive N,N-diaryl dithiocarbamates (DTCs) are synthesized in good yields by the generation of metal amide salts from N-benzoyl precursors, followed by addition of CS2. para-Substituted diphenylamines are prepared by electrophilic aromatic substitution of diphenylbenzamide and saponification. Deacylation of electron-rich species such as bis(p-dimethylaminophenyl)benzamide is challenging because of the oxidative sensitivity of the anionic intermediate but could be achieved in good yield by using n-BuLi to generate a hemiaminal adduct, prior to acidification. The N,N-diaryl DTCs are stable as alkali salts and can be used to produce densely packed monolayers on gold surfaces.

Keywords: Benzamide, deprotection, dithiocarbamate, oxidation sensitive

INTRODUCTION

Dithiocarbamates (DTCs) are important precursors for a wide range of chemical applications,[1] such as intermediates in the chemical synthesis of biologically active compounds and natural products,[2,3] as ligands in coordination chemistry to stabilize various transition metals,[4] and as initiators in the vulcanization of rubber and in the synthesis of polymers by reversible addition-fragmentation (RAFT) processes.[5] Dithiocarbamates can also serve as ligands for novel applications in surface chemistry. It has been argued that DTCs can form more robust monolayers than alkylthiols on gold, aided by the favorable epitaxy between the bidentate carbodithioate unit and the Au(111) surface.[6] DTCs have been used to anchor a variety of functional molecules and polymers[7,8] onto metal surfaces, such as peptides,[6] carbohydrates,[9] DNA,[10] and cancer-targeting ligands.[11]

DTCs have particularly intriguing potential as interconnects for molecular electronics, because of their extended conjugation. Wessels et al. first demonstrated this possibility using bisDTC ligands as molecular junctions between Au nanoparticles, with large increases in overall conductivity relative to dithiol spacers.[12] Surface spectroscopy studies on DTC monolayers further established a high charge transport across aromatic DTCs assembled on Au.[13] Advances in this research area may be supported by developing a practical synthesis of DTCs attached to π-conjugated systems with tunable electronic structures.

Many DTCs can be conveniently prepared simply by combining the corresponding amine and CS2 in polar solvents.[6,14,15] However, this is not the case for diaryl DTCs; at present only a few syntheses have been reported, mostly for simple N,N-diphenyl DTCs.[16,17,18] Diaryl DTCs, particularly those with electron-donating substituents, are of strong interest for studies in molecular electronics, but to our knowledge a general synthetic procedure is lacking. This may be due partly to the oxidative sensitivity of electron-rich diarylamines, which are challenging to isolate in neutral form. We note that while para-substituted N,N-diphenylamines have been prepared by Buchwald–Hartwig coupling,[19] this approach involves the coupling of substituted anilines and phenyl halides and does not address the air sensitivity of the products.

In this article we show that para-substituted N,N-diphenyl DTCs can be synthesized by addition of CS2 to the corresponding metal amide, generated in situ from N-benzamide derivatives. This practical approach allows us to use standard electrophilic aromatic substitution to generate the desired N,N-diphenylamine derivatives and permits the isolation of the desired DTCs as air-stable alkali salts.

RESULTS AND DISCUSSION

para-Substituted N,N-diphenylamines were synthesized from the diphenylbenzamide 1 (Scheme 1). Bromination of benzamide 1 followed by saponification with Claisen’s base (KOH in H2O/MeOH) provided N,N-bis(p-bromophenyl)amine 3 in 84% yield. The double nitration of benzamide 1 was less straightforward; using 70% HNO3 as the nitrating agent produced the mononitrated benzamide (N-Bz-4) as the major product.[20] N,N-Bis(p-dinitrophenyl)benzamide 5 could be obtained by treatment with a stronger Brønsted acid, namely a mixture of 70% HNO3 and TFA, accompanied by small amounts of N-Bz-4. The N-Bz-4 and benzamide 5 were saponified with Claisen’s base to yield N-(p-nitrophenyl)-N-phenylamine 4 and N,N-bis(p-nitrophenyl)amine 6 in 48% and 67% yield, respectively. Compound 5 could also be reduced by SnCl2 to bis(p-amino)phenylbenzamide,[21] followed by reductive amination to provide N,N-bis(p-dimethylaminophenyl)benzamide 7 in 94% yield.

Scheme 1.

Scheme 1

Synthesis of para-substituted N,N-diphenylamines.

To obtain N,N-bis(p-dimethylaminophenyl)amine 8, benzamide 7 was subjected to a variety of deacylation conditions (Table 1). Claisen’s base did not react with 7 at 25 °C and caused decomposition at 100 °C (Table 1, entry 1); saponifications with phenylthiolate salts were likewise unsuccessful (entries 2 and 3). We suspected that the electron-rich diarylamine was susceptible to oxidation under basic conditions[22] and attempted to remove the benzoyl group under acidic conditions at 100 °C, but this resulted in only partial conversion (entry 4). We then investigated acyl addition and cleavage under anhydrous conditions using stronger nucleophiles and were pleased to find that n-BuLi added readily to the benzamide carbonyl at 0 °C. Acidic workup (6 M HCl) produced the hydrochloride salt of 8 in good yields (entry 5). Isolation of 8 as its HCl salt was necessary to prevent product oxidation during storage and handling. We note that 1-phenylpentanone was obtained as a reaction by-product only after acidic workup, not during the course of the reaction, indicating that deacylation proceeds through a stable hemiaminal intermediate,[23,24] one that is presumably less susceptible to oxidation than the free amine (Scheme 2).

Table 1.

Synthesis of N,N-bis(p-dimethylaminophenyl)amine hydrochloride

graphic file with name nihms-679607-t0004.jpg

Entry Reaction conditions Yield
1 Claisen’s base, 100 °C Complex mixture
2 PhSK, Claisen’s base, 100 °C Complex mixture
3 PhSLi, THF, 70 °C No reaction
4 Aq. HCl, 100 °C μwave) 40%a
5 (i) n-BuLi, THF, 0°C; (ii) 6M HCl 94%b
a

Conversion determined by 1H NMR.

b

Isolated yield.

Scheme 2.

Scheme 2

Formation of 8 via hemiaminal intermediate.

With diarylamines in hand, we surveyed several basic conditions for generating aromatic DTCs (Table 2). In principle, DTCs can be prepared from their corresponding amines under mildly basic conditions, if the latter is sufficiently nucleophilic.[3,6,15] However, diphenylamine (pKa ~ 25)[25] and related aromatic amines are insufficiently reactive toward CS2, and thus require generation of their conjugate base. In the cases of diphenylamine and bis(p-bromophenyl)amine 3, deprotonation with n-BuLi or n-BuLi/DMSO (dimsyl-Li) followed by addition of CS2 was sufficient to produce lithium salts of N,N-diphenyl DTC (9) and N,N-bis (p-bromophenyl) DTC (10) respectively, which were isolated in good yields by precipitation with Et2O (Table 2, entries 1–4). The 13C NMR spectra of these diaryl DTCs revealed thiocarbonyl signals at δ 219–224 ppm, more downfield than those produced by dialkyl DTCs (δ 200–210 ppm).[26] However, DTC formation from deprotonated N-(p-nitrophenyl)-N-phenylamine 4 or N,N-bis(p-nitrophenyl)amine 6 was sluggish and did not reach completion regardless of reaction temperature or the amount of CS2 added, with only trace amounts of DTC formed based on 13C NMR analysis (entries 5 and 6). The poor reactivity was attributed to the reduced nucleophilicity of the amide anion by one or both nitro groups.

Table 2.

Synthesis of substituted diphenylamine DTCs

graphic file with name nihms-679607-t0005.jpg

Entry Amine Basea Diaryl DTC (yield)b
1 Ph2NH n-BuLi 9 (64%)
2 Ph2NH dimsyl-Li 9 (98%)
3 3 (X=Br) n-BuLi 10 (96%)
4 3 (X=Br) dimsyl-Li 10 (93%)
5 4 (X=H,NO2) dimsyl-Li Trace
6 6 (X=NO2) dimsyl-Li Trace
7 8 (X=NMe2)c n-BuLi 11 (94%)
8 8 (X=NMe2)c dimsyl-Li 11 (72%)
9 Carbazoled NaH 12 (55%)e
a

1.0–2.5 equiv.

b

Isolated yields unless stated otherwise.

c

Hydrochloride salt, neutralized with one equivalent of n-BuLi or dimsyl-Li.

d

Reaction conditions: (i) 50 °C for 5 h, then −78 °C; (ii) CS2 (5–10 equiv), warmed to rt.

e

Yield after a single precipitation.

Electron-rich diarylamine 8 was converted straightforwardly into N,N-bis (p-dimethylaminophenyl) DTC (11) by deprotonation of the HCl salt using 2 equivalents of n-BuLi or dimsyl-Li, followed by dropwise addition of CS2 and precipitation in Et2O to produce the lithium salt of DTC 11 in good isolated yield (Table 2, entries 7 and 8). The lithium salt of N-carbazolyl DTC (12) could also be obtained by deprotonating carbazole with n-BuLi or dimsyl-Li followed by CS2 addition; however, attempts to precipitate this product were not successful. Deprotonation with NaH instead produced sodium N-carbazolyl DTC, which could be isolated in satisfactory yields (entry 9). It should be mentioned that the yield of sodium N-carbazolyl DTC (12) was moderate because the instability of the product prevented us from performing a second precipitation. Converting the sodium salt in N-carbazolyl DTC (12) into a triethylammonium (Et3HN+) salt prolonged its stability and allowed collection of all spectroscopic data. We note that N-carbazolyl DTC can also be synthesized using potassium metal as a base,[27] but the potassium salt of 12 is highly air sensitive and decomposes within seconds upon contact with air.

CONCLUSION

A practical method of converting N-benzoyl diarylamines into dithiocarbamates is presented. In some cases, diarylamines can be obtained straightforwardly, then deprotonated with a strong base, treated with CS2, and precipitated from the reaction mixture to yield the desired DTCs as lithium salts. Electron-rich diaryla-mines can be prepared by adding n-BuLi to the benzamide to form a hemiaminal intermediate and isolated as their ammonium salts by acidic workup. These aromatic DTCs can be used to prepare self-assembled monolayers on Au(111) for surface spectroscopy studies, which will be reported in due course.

EXPERIMENTAL

All chemicals and reagents were obtained from commercial sources and used as received unless stated otherwise. Solvents were freshly distilled prior to use. A 0.2 M solution of dimsyl lithium (the conjugate base of dimethyl sulfoxide) was prepared by treating DMSO (1.42 mL, 20 mmol) in anhydrous THF (43.5 mL) with n-BuLi (5 mL, 10 mmol) at 0 °C, then stirred for 30 min and maintained at 0 °C prior to use. 1H and 13C NMR spectra were recorded on a Varian Inova 300, a Bruker ARX400, or a Bruker DR×500 and referenced to the solvent used (CDCl3: 7.27 and 77 ppm; CD3OD: 3.31 and 49.2 ppm; DMSO-d6: 2.50 and 39.5 ppm). DTC salts 9–12 decompose rapidly upon atmospheric exposure and are best handled under anaerobic conditions and stored as powders in vacuum-sealed ampoules.

General Procedure for Synthesis of Aromatic Dithiocarbamates

A solution of aromatic amine (2.96 mmol) was dissolved in degassed THF (6 mL), then cooled to 0 °C, and treated with dimsyl lithium solution (5.92 mmol, 2 equiv) or n-BuLi (5.92 mmol, 2 equiv). After stirring for 30 min at 0 °C, CS2 (29.6 mmol, 10 equiv) was added. The ice bath was removed and the mixture was stirred for 12 h room temperature. The crude mixture was precipitated from THF in Et2O to afford diaryl DTC product.

Lithium N,N-Diphenyl Dithiocarbamate (9)

Off-white solid. 1H NMR (300 MHz, CD3OD): δ 7.37–7.20 (m, 8 H), 7.17–7.06 (m, 2 H). 13C NMR (75 MHz, CD3OD): δ 219.0, 150.4, 129.7, 129.3, 127.2. IR (thin film): ν 3354, 1592, 1491, 1452, 1343, 1313, 1223, 1110, 1048, 1012, 884, 834 cm −1. ESI-MS: m/z for C26H20LiN2S4 [2M–Li] 495.[16]

Lithium N,N-Bis(p-bromophenyl)dithiocarbamate (10)

Pale-yellow solid. 1H NMR (500 MHz, DMSO-d6): δ 7.35 (d, 4 H, J = 8.7 Hz), 7.13 (d, 4 H, J = 8.7 Hz). 13C NMR (125 MHz, DMSO-d6): δ 218.8, 148.6, 131.3, 131.0, 117.5. IR (thin film): ν 3426, 1512, 1484, 1432, 1282, 1296, 1010, 956, 887, 867, 790 cm −1. ESI-MS: m/z for C12H8Br2N [M–Li–CS2] 326.

Lithium N,N-Bis(p-dimethylaminophenyl)dithiocarbamate (11)

Off-white solid. 1H NMR (300 MHz, CD3OD): δ 7.60 (d, 4 H, J = 8.8 Hz), 7.12 (d, 4 H, J = 8.8 Hz), 3.31 (s, 12 H). 13C NMR (75 MHz, CD3OD): δ 218.2, 150.3, 141.2, 129.5, 129.2, 114.1, 113.8, 41.0. IR (thin film): ν 3401, 1634, 1607, 1516, 1445, 1331, 1281, 1221, 1183, 1153, 1127, 1050, 884, 707, 693 cm−1. ESI-MS: m/z for C17H20N3S2 [M–Li] 330.

Sodium N-Carbazolyl Dithiocarbamate (12)

Yellow solid. 1H NMR (300 MHz, CD3OD): δ 8.67 (d, 2 H, J = 8.3 Hz), 7.79 (d, 2 H, J = 7.6 Hz), 7.24 (t, 2 H, J = 7.7 Hz), 7.05 (t, 2 H, J = 7.3 Hz). 13C NMR (75 MHz, CD3OD): δ 224.4, 141.0, 126.2, 125.2, 121.9, 119.9, 116.3. Sodium N-carbazolyl dithiocarbamate 12 was converted to more stable triethylammonium salt for the characterization purpose. Triethylammonium N-carbazolyl dithiocarbamate was an orange-yellow solid. 1H NMR (300 MHz, CDCl3): δ 8.92 (d, 2 H, J = 8.4 Hz), 8.05 (d, 2 H, J = 7.6 Hz), 7.48 (t, 2 H, J = 7.0 Hz), 7.33 (m, 2 H), 3.26 (q, 6 H, J = 7.5 Hz), 1.42 (t, 9 H, J = 7.2 Hz). IR (thin film): ν 3369, 1631, 1490, 1450, 1351, 1328, 1306, 1274, 1046, 884, 707, 693 cm−1. ESI-MS: m/z for C12H8N [M–Et3NH–CS2] 166.

Supplementary Material

Suppl.

Acknowledgments

FUNDING

We gratefully acknowledge financial support from the National Science Foundation (CHE-0957738) and Grants for Development of New Faculty Staff, Ratchadaphiseksomphot Endowment Fund, Chulalongkorn University. P. P. is thankful for a start-up grant for laboratory equipment from the Department of Chemistry, Chulalongkorn University. NMR and MS analyses were also supported by the Purdue Interdepartmental NMR facility and the Purdue University Center for Cancer Research.

Footnotes

SUPPLEMENTAL MATERIAL

Supplemental data for this article can be accessed on the publisher’s website.

REFERENCES

  • 1.(a) Aly AA, Brown AB, Bedair TMI, Ishak EA. J. Sulfur Chem. 2012;33:605–617. [Google Scholar]; (b) Hassan EA, Zayed SE. Phosphorus, Sulfur Silicon Relat. Elem. 2013;189:300–323. [Google Scholar]
  • 2.(a) Buac D, Schmitt S, Ventro G, Kona FR, Dou QP. Mini-Rev. Med. Chem. 2012;12:1193–1201. doi: 10.2174/138955712802762040. [DOI] [PMC free article] [PubMed] [Google Scholar]; (b) Nagy EM, Ronconi L, Nardon C, Fregona D. Mini-Rev. Med. Chem. 2012;12:1216–1229. doi: 10.2174/138955712802762004. [DOI] [PubMed] [Google Scholar]
  • 3.(a) Padungros P, Alberch L, Wei A. Org. Lett. 2012;14:3380–3383. doi: 10.1021/ol301349w. [DOI] [PMC free article] [PubMed] [Google Scholar]; (b) Padungros P, Alberch L, Wei A. J. Org. Chem. 2014;79:2611–2624. doi: 10.1021/jo500032k. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.(a) Hogarth G. Prog. Inorg. Chem. 2005;53:71–561. [Google Scholar]; (b) Victoriano LI. Coord. Chem. Rev. 2000;196:383–398. [Google Scholar]
  • 5.(a) Moad G, Rizzardo E, Thang SH. Polymer. 2008;49:1079–1131. [Google Scholar]; (b) Keddie DJ, Moad G, Rizzardo E, Thang SH. Macromolecules. 2012;45:5321–5342. [Google Scholar]
  • 6.(a) Zhao Y, Pérez-Segarra W, Shi Q, Wei A. J. Am. Chem. Soc. 2005;127:7328–7329. doi: 10.1021/ja050432f. [DOI] [PMC free article] [PubMed] [Google Scholar]; (b) Zhu H, Coleman DM, Dehen CJ, Geisler IM, Zemlyanov D, Chmielewski J, Simpson GJ, Wei A. Langmuir. 2008;24:8660–8666. doi: 10.1021/la801254b. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Huff TB, Hansen MN, Zhao Y, Cheng J-X, Wei A. Langmuir. 2007;23:1596–1599. doi: 10.1021/la062642r. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.(a) Park M-H, Duan X, Ofir Y, Creran B, Patra D, Ling XY, Huskens J, Rotello VM. ACS Appl. Mater. Interfaces. 2010;2:795–799. doi: 10.1021/am9009007. [DOI] [PubMed] [Google Scholar]; (b) Leonov AP, Wei A. J. Mater. Chem. 2011;21:4371–4376. doi: 10.1039/c0jm04153j. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Adak AK, Leonov AP, Ding N, Thundimadathil J, Kularatne S, Low PS, Wei A. Bioconjugate Chem. 2010;21:2065–2075. doi: 10.1021/bc100288c. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Sharma J, Chhabra R, Yan H, Liu Y. Chem. Commun. 2008:2140–2142. doi: 10.1039/b800109j. [DOI] [PubMed] [Google Scholar]
  • 11.(a) Huff TB, Tong L, Zhao Y, Hansen MN, Cheng J-X, Wei A. Nanomedicine. 2007;2:125–132. doi: 10.2217/17435889.2.1.125. [DOI] [PMC free article] [PubMed] [Google Scholar]; (b) Tong L, Zhao Y, Huff TB, Hansen MN, Wei A, Cheng J-X. Adv. Mater. 2007;19:3136–3141. doi: 10.1002/adma.200701974. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Wessels JM, Nothofer H-G, Ford WE, von Wrochem F, Scholz F, Vossmeyer T, Schroedter A, Weller H, Yasuda A. J. Am. Chem. Soc. 2004;126:3349–3356. doi: 10.1021/ja0377605. [DOI] [PubMed] [Google Scholar]
  • 13.von Wrochem F, Gao DQ, Scholz F, Nothofer HG, Nelles G, Wessels JM. Nat. Nanotechnol. 2010;5:618–624. doi: 10.1038/nnano.2010.119. [DOI] [PubMed] [Google Scholar]
  • 14.Selected examples: Li R-T, Ding P-Y, Han M, Cai M-S. Synth. Commun. 1998;28:295–300. Cui J-L, Ge Z-M, Cheng T-M, Li R. Synth. Commun. 2003;33:1969–1976. Chaturvedi D, Ray S. Tetrahedron Lett. 2006;47:1307–1309. Azizi N, Aryanasab F, Torkiyan L, Ziyaei A, Saidi MR. J. Org. Chem. 2006;71:3634–3635. doi: 10.1021/jo060048g. Azizi N, Aryanasab F, Saidi MR. Org. Lett. 2006;8:5275–5277. doi: 10.1021/ol0620141. Azizi N, Aryanasab F, Tourkian L, Saidi MR. Synth. Commun. 2010;41:94–99.
  • 15.Selected examples: Tan Y, Jin S, Hamers RJ. ACS Appl. Mater. Interfaces. 2013;5:12975–12983. doi: 10.1021/am403744g. Frederick MT, Weiss EA. ACS Nano. 2010;4:3195–3200. doi: 10.1021/nn1007435. Morf P, Raimondi F, Nothofer H-G, Schnyder B, Yasuda A, Wessels JM, Jung TA. Langmuir. 2005;22:658–663. doi: 10.1021/la052952u. Morf P, Ballav N, Putero M, von Wrochem F, Wessels JM, Jung TA. J. Phys. Chem. Lett. 2010;1:813–816.
  • 16.(a) Moloy KG. Inorg. Chem. 1988;27:677–681. [Google Scholar]; (b) Ma H, Wang G, Yee GT, Petersen JL, Jensen MP. Inorg. Chim. Acta. 2009;362:4563–4569. [Google Scholar]
  • 17.Kumar S, Roy NN. Indian J. Eng. Mater. Sci. 1994;1:292–294. [Google Scholar]
  • 18.(a) Krugmann T, Krugmann M. GB Patent 877358. 1959 Aug 14;; (b) Shaw JT. U.S. Patent 3198829. 1965 Aug 3;; (c) Van Duijn S, Durville PFM, Van Rosmalen GM. EP Patent 116988. 1984 Aug 29;
  • 19.Pratt DA, DiLabio GA, Valgimigli L, Pedulli GF, Ingold KU. J. Am. Chem. Soc. 2002;124:11085–11092. doi: 10.1021/ja026289x. [DOI] [PubMed] [Google Scholar]; (b) Hanthorn JJ, Valgimigli L, Pratt DA. J. Am. Chem. Soc. 2012;134:8306–8309. doi: 10.1021/ja300086z. [DOI] [PubMed] [Google Scholar]; (c) Hanthorn JJ, Valgimigli L, Pratt DA. J. Org. Chem. 2012;77:6908–6916. doi: 10.1021/jo301013c. [DOI] [PubMed] [Google Scholar]
  • 20.See Table S1 in the supporting information for details.
  • 21.(a) Bellamy FD, Ou K. Tetrahedron Lett. 1984;25:839–842. [Google Scholar]; (b) Bellamy FD, Ou K. Tetrahedron Lett. 1985;26:1362. [Google Scholar]
  • 22.(a) Shashin SS, Emanuel ON, Skibida IP. Russ. Chem. Bull. 1994;43:1646–1650. [Google Scholar]; (b) Shashin SS, Emanuel ON, Skibida IP. Russ. Chem. Bull. 1994;43:1651–1656. [Google Scholar]
  • 23.Szostak M, Yao L, Aubé J. J. Am. Chem. Soc. 2010;132:2078–2084. doi: 10.1021/ja909792h. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.The hemiaminal intermediate may also exist as a dimer. See Adler M, Marsch M, Nudelman NS, Boche G. Angew. Chem. Int. Ed. 1999;38:1261–1263. doi: 10.1002/(SICI)1521-3773(19990503)38:9<1261::AID-ANIE1261>3.0.CO;2-E.
  • 25.Bordwell FG, Branca JC, Hughes DL, Olmstead WN. J. Org. Chem. 1980;45:3305–3313. [Google Scholar]
  • 26.Van Gaal HLM, Diesveld JW, Pijpers FW, Van der Linden JGM. Inorg. Chem. 1979;18:3251–3260. [Google Scholar]
  • 27.Bereman RD, Nalewajek D. Inorg. Chem. 1978;17:1085–1087. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Suppl.

RESOURCES