環 境 にやさしい 材 料 のための 融 合 領 域 研 究 会 では 従 来 の 理 学 工 学 化 学 薬 学 などと 言 っ た 枠 にとらわれることなく 幅 広 い 学 問 領 域 を 融 合 して 環 境 にやさしい 材 料 開 発 という 新 しい 科 学 技 術 の 展 開 を 目 指

環 境 にやさしい 材 料 のための 融 合 領 域 研 究 会 講 演 予 稿 集 日 程 : 平 成 22 年 8 月 4 日 ( 水 )~5 日 ( 木 ) 場 所 : 鳥 取 大 学 広 報 センター スペースC 主 催 : 後 援 : 阿 南 工 業 高 等 専 門 学 校 鳥 取 大 学 JSTイノベーションサテライト 徳 島

環 境 にやさしい 材 料 のための 融 合 領 域 研 究 会 では 従 来 の 理 学 工 学 化 学 薬 学 などと 言 っ た 枠 にとらわれることなく 幅 広 い 学 問 領 域 を 融 合 して 環 境 にやさしい 材 料 開 発 という 新 しい 科 学 技 術 の 展 開 を 目 指 しております 環 境 にやさしい 材 料 を 融 合 領 域 という 新 しい フィールドからの 発 見 と 活 用 によって 社 会 に 資 することを 目 的 としています 是 非 この 機 会 に 活 発 なご 討 議 をよろしくお 願 いいたします

目 次 8 月 4 日 ( 水 ) 第 1 部 司 会 石 井 晃 ( 鳥 取 大 ) 14:00 開 会 挨 拶 阿 南 高 専 塚 本 史 郎 学 長 挨 拶 鳥 取 大 能 勢 隆 之 来 賓 挨 拶 JST 徳 島 今 枝 正 夫 14:30 基 調 講 演 ヨードニウムイオンをキーとするヘテロ 環 合 成 資 源 小 国 日 本 が 誇 れる 資 源 ヨウ 素 の 活 用 阿 南 高 専 小 松 満 男 1 15:15 招 待 講 演 ダイヤモンドパウダ 回 収 のための 表 面 処 理 日 本 工 大 飯 塚 完 司 5 15:40 休 憩 第 2 部 司 会 片 田 直 伸 ( 鳥 取 大 ) 15:45 招 待 講 演 半 導 体 基 板 上 に 担 持 した 有 機 金 属 触 媒 阿 南 高 専 塚 本 史 郎 9 16:10 招 待 講 演 機 能 性 分 子 合 成 用 硫 黄 修 飾 金 担 持 型 パラジウム(SAPd)の 開 発 北 大 有 澤 光 弘 17 16:35 招 待 講 演 基 板 上 に 担 持 した 有 機 金 属 触 媒 の 第 一 原 理 計 算 鳥 取 大 石 井 晃 31 i

8 月 5 日 ( 木 ) 第 3 部 司 会 塚 本 史 郎 ( 阿 南 高 専 ) 10:00 特 別 講 演 糖 尿 病 薬 :メディシナルからプロセスへ 日 亜 薬 品 小 池 晴 夫 40 10:40 特 別 講 演 最 近 の 化 合 物 ライブラリーについて シグマアルドリッチジャパン 田 中 紀 子 48 11:20 招 待 講 演 原 子 の 位 置 と 触 媒 活 性 点 の 性 質 :ゼオライト 固 体 酸 触 媒 における 酸 強 度 と 結 合 距 離 の 相 関 鳥 取 大 片 田 直 伸 52 11:45 開 会 挨 拶 阿 南 高 専 塚 本 史 郎 ii

ヨードニウムイオンをキー キーとする ヘテロ 環 合 成 資 源 小 国 日 本 が 誇 れる 資 源 ヨウ 素 の 活 用 阿 南 工 業 高 等 専 門 学 校 ( 大 阪 大 学 名 誉 教 授 ) 小 松 満 男 ヨウ 素 は 資 源 小 国 日 本 にとって 数 少 ない 国 内 で 自 給 できる 資 源 であり チリに 次 いで 世 界 第 二 位 の 埋 蔵 量 と 生 産 量 を 誇 る 従 って この 海 洋 由 来 の 重 要 な 資 源 の 有 効 利 用 は わが 国 にとって 見 過 ごすことのできない 研 究 課 題 である ヨウ 素 原 子 を 活 用 する 有 機 合 成 反 応 は ヨードラクトン 化 をはじめとして 古 くから 研 究 され 合 成 反 応 におけるヨード 官 能 基 の 有 用 性 から 近 年 も 精 力 的 に 研 究 がなされており 今 後 の 発 展 が 大 いに 期 待 できる 分 野 である 一 方 ヘテロ 環 化 合 物 は ヘテロ 原 子 のもつ 大 きな 電 気 陰 性 度 と 非 共 有 電 子 対 のもたらす 特 性 が 自 由 回 転 の 制 御 や 特 異 な 共 役 系 に 基 づく 環 骨 格 の 特 性 と 相 俟 って 鎖 状 化 合 物 には みられない 機 能 を 発 現 する 従 って ヘテロ 環 骨 格 は 様 々な 機 能 性 物 質 において 多 才 な 能 力 を 示 す 基 本 骨 格 として 重 要 であり それらの 合 成 手 法 の 開 発 は 医 農 薬 や 種 々の 機 能 性 材 料 を 開 発 する 上 で 極 めて 意 義 ある 研 究 課 題 である 中 でも 窒 素 を 含 むヘテロ 環 はアルカロイド 等 の 天 然 物 中 に 幅 広 く 存 在 し また 生 体 内 において 重 要 な 役 割 を 果 たしているだけでなく 機 能 性 材 料 としても 重 用 されているものが 多 い そこで 我 々はヨウ 素 触 媒 ヨウ 素 官 能 基 あるいはヨウ 素 を 含 む 試 薬 をそれぞれ 巧 みに 活 用 し ヨードニウムイオンを 経 由 する 高 効 率 的 な 新 しいヘテロ 環 合 成 法 の 開 発 を 展 開 してきた 1.ヨウ 素 触 媒 によるヘテロ 小 員 環 合 成 -グリーンなアジリジン 化 法 我 々の 研 究 室 では ヨウ 素 を 触 媒 とし 安 価 で 取 り 扱 い 容 易 なクロラミンTを 窒 素 源 とす るアルケン 類 のアジリジン 化 法 を 見 出 している 1) 本 法 は 様 々なアルケンに 活 用 できるだけ でなく 非 常 に 穏 和 な 条 件 で 容 易 にヘテロ 小 員 環 を 構 築 することが 可 能 である ここで 1 源 であるクロラミン T の 水 溶 性 が 高 いという 特 性 を 利 用 すれば 本 反 応 の 水 系 への 展 開 が 可 2) 能 であると 考 え 環 境 に 優 しい 有 機 合 成 反 応 の 開 拓 を 検 討 した その 結 果 相 間 移 動 触 媒 あるいはシリカを 添 加 3) することにより 効 率 的 にしかも 立 体 選 択 的 にアジリジンが 水 中 で 合 成 できることを 明 らかにした I 2 (cat.) Ts Cl O PTC or silica R 1 R 2 R + S Me R 3 a O H 2 O, r.t. R 1 R 3 Chloramine-T R 1 = Ph, alkyl, or H; R 2, R 3 = alkyl or H とくに シリカを 水 に 添 加 する 系 は 新 しい 概 念 に 基 づく 有 機 合 成 の 反 応 場 として 期 待 で きる 即 ち シリカ 表 面 と 有 機 物 との 間 には 静 電 的 な 相 互 作 用 が 働 く この 状 態 に 水 が 存 在 すれば 反 応 物 質 が 系 内 のシリカ 表 面 上 に 均 一 に 分 散 するだけでなく 水 との 静 電 反 発 により シリカへの 吸 着 が 促 進 され その 表 面 上 での 反 応 が 生 起 すると 考 えている このシリカ- 水 系 という 新 しい 反 応 場 を 用 い 上 記 アジリジン 化 のみならず アジリジンの 水 溶 性 求 核 剤 に よる 開 環 反 応 の 開 発 にも 成 功 している 4) 1

R 1 R 2 I 2 Cl Ts a olefin layer water layer silica gel or MCM-41 water layer R 2 Ts R 1 I 2 Cl a surface of silica gel or MCM-41 R 1 R 2 I 2 a Cl Ts R 1 Ts R 2 Figure 1. Proposed pathway for aziridination utilizing a water-silica system. 2.ヨウ 素 官 能 基 のイオン 的 原 子 移 動 反 応 に 基 づくヘテロ 環 合 成 上 記 で 述 べたアジリジン 化 の 反 応 経 路 の 考 察 から 新 しい 環 形 成 反 応 を 設 計 した クロラミ ンTはナトリウム 塩 を 形 成 しているため それ 程 強 くはないが 窒 素 求 核 種 として 作 用 する 例 えば 1-ヨードオクタンのヨード 基 はアセトニトリル 中 室 温 でクロラミンTにより 置 換 される この 場 合 還 元 剤 を 加 えないにもかかわらず 窒 素 上 が 水 素 となったオクチルスル ホンアミドが 生 成 する これは 脱 離 したヨウ 化 ナトリウムのヨウ 化 物 イオンが 置 換 直 後 の 窒 素 上 の 塩 素 を 引 抜 き 後 処 理 の 水 によりプロトン 化 さ れて 生 成 したものと 考 えられる もし ヨードアルカンの 代 わりにヨードアルケ ンを 用 いれば イオン 的 なヨウ 素 原 子 移 動 反 応 が 起 こり 環 状 ヨードニウム 中 間 体 を 経 由 する 興 味 ある 環 化 が 進 行 すると 考 えた 事 実 5-ヨード-1- ペンテンをアセトニトリル 中 2 当 量 のクロラミ ンTで 処 理 したところ 予 期 した 通 りにヨードメ チル 基 をもつピロリジンを 91%という 高 収 率 で 合 成 することができた この 反 応 は 様 々なヨードアルケンに 適 用 可 能 で あり 結 果 の 一 部 を 表 1 に 示 す この 非 常 にシン プルな 方 法 により 種 々のアルケニルヨージドを ピロリジン 類 へと 変 換 でき とくにトランスある いはシスのアルケン 部 位 をもつ 基 質 は 完 全 に 立 体 特 異 的 に 対 応 するピロリジンへと 導 くことができ た また 様 々な 環 サイズをもつビシクロ 環 を 立 体 選 択 的 に 構 築 することにも 成 功 した 本 環 化 反 応 の 経 路 を 解 明 するために ヨードア ルケンとクロラミンTの 置 換 反 応 により 系 内 で 生 成 すると 考 えられる 窒 素 上 が 塩 素 化 された - アルケニルスルホンアミドを 次 亜 塩 素 酸 tert- ブチルを 用 いて 別 途 合 成 した この 化 合 物 にヨウ 化 ナトリウムを 作 用 させたところ 効 率 良 く 環 化 反 応 が 進 行 した 以 上 の 結 果 から ヨードアルケ Table 1 Synthesis of various pyrrolidines from 4-alkenyliodides and Chloramine-T (CT) a. 4 pentenyliodide product yield (%) Me Me Me I I I I I I I Ts Ts Ts Ts Ts Ts Ts I Me I Me I I Me I I I 91 75 81 b 77 c 50 c 83 c 64 a Reaction conditions: alkenyliodide (0.5 mmol), CT (1 mmol), MeC (3 ml), rt, 48 h. b Cis : trans = 69 : 31. c Anhydrous CT and 12 ml of MeC were used. 2

ンとクロラミンTとの 反 応 はスキーム 1 に 示 した 経 路 で 進 行 していると 考 えられる ここで 反 応 基 質 のヨウ 素 は(1)クロラミンTとの 反 応 による 脱 離 基 (2)-Cl 結 合 を 切 断 する ためのルイス 塩 基 (3)オレフィン 部 位 の 活 性 化 剤 (4) 立 体 制 御 するための 環 状 カチオ ン 種 (5) 生 成 物 中 の 官 能 基 として 多 様 な 働 きを 司 っており 非 常 にユニークな 元 素 である 一 方 このようなアミド 窒 素 が 塩 素 化 された 化 合 物 とヨウ 化 ナトリウムとの 反 応 もこれまで に 知 られておらず ヨウ 素 の 特 性 を 利 用 した 興 味 深 い 反 応 である 5) R 1 R 2 I CT leaving group R 1 R 2 Ts Cl I Lewis base activator R 1 I R 2 stereocontrolling cyclic cation Ts Ts functional group R 1 R 2 I Scheme 1 Multiple functions of the iodo group. 3. 次 亜 ヨウ 素 酸 tert-ブチルを 活 用 するヘテロ 環 合 成 上 記 項 目 2 の 反 応 の 中 間 体 である -クロロスルホンアミドの 特 異 な 反 応 挙 動 に 着 目 し 窒 素 -ヨウ 素 結 合 をもつアルケニルスルホンアミドの 形 成 を 検 討 した 即 ち 先 の 反 応 の 経 路 を 解 明 するために 次 亜 塩 素 酸 tert-ブチルを 用 いてアルケニルスルホンアミドの 窒 素 を 塩 素 化 したが これに 対 して 次 亜 ヨウ 素 酸 tert-ブチル( 次 亜 塩 素 酸 tert-ブチルとヨウ 化 ナ トリウムより 調 製 される)を 添 加 したところ アミド 窒 素 がヨウ 素 化 された 生 成 物 は 確 認 さ れず 一 挙 に 環 化 したピロリジンが 高 収 率 で 得 られることが 判 明 した( 前 ページ) そこで 種 々のアルケニルス ルホンアミドを 出 発 原 料 とし 次 亜 塩 素 酸 tert-ブチルとヨウ 化 ナトリウムを 作 用 させたとこ ろ 非 常 に 高 収 率 でしかも 高 立 体 選 択 的 に 様 々なヘテロ 環 を 合 成 することに 成 功 した この 特 異 な 環 化 の 経 路 を 明 らかにするためにアルケンと -メチルスルホンアミドの 混 合 溶 液 に 次 亜 ヨウ 素 酸 tert-ブチルを 添 加 したところ アルケンは 全 く 変 化 せず スルホンア ミドの 窒 素 がヨウ 素 化 された 化 合 物 のみが 得 られた 以 上 の 実 験 から 本 環 化 反 応 はまずアルケニルスルホン アミドの 窒 素 がヨウ 素 化 され た 後 このヨウ 素 がアルケン に 移 動 して 環 状 ヨードニウム 中 間 体 を 形 成 し つづく 分 子 内 環 化 を 経 て 生 成 物 に 至 ると 考 えている 6) Ts Me I H Ts Ts t BuOCl (1.0 eq.), a I (1.0 eq.) Me I t BuO I MeC, r.t., 5 h 93% 100% 96% 94% Ts 97% I Ts 97% I Ts Ts 27% I I Ts Ts Ts 98% I 79% I I 3

そこで 単 純 なスルホンアミドやカルボン 酸 アミドを 次 亜 ヨウ 素 酸 tert-ブチル 存 在 下 オレフィンに 作 用 させれば 分 子 間 反 応 により 複 素 環 化 が 進 行 するのではないかと 考 えた ス チレンと p-トルエンスルホンアミドをアセトニトリルに 溶 解 し t-buocl と ai を 加 え 室 温 で 5 時 間 撹 拌 したところ 95% という 高 収 率 でスチレンのアジリ ジン 化 が 進 行 した 7) さらに ス ルホンアミドに 代 えてカルボン 酸 アミドを 用 いれば アジリジンで はなく オキサゾリンを 合 成 でき ることも 明 らかにした 8) 本 反 応 は 他 のスルホンアミド やカルボン 酸 アミドおよびオレフ ィンに 適 用 でき 種 々のアジリジ ン 環 やオキサゾリン 環 の 構 築 が R 1 O O S O O R t BuOCl, ai R 2 + S H2 R MeC, r.t., 5 h R 1 R2 R 1, R 2 = alkyl, aryl R = aryl, Me, CH 2 CH 2 SiMe 3 up to 95% R 1 R 2 O H 2 入 手 容 易 な 原 料 から 簡 便 にしかも 実 用 的 に 行 える 方 法 論 である 以 上 のように ヨウ 素 の 特 性 を 巧 みに 活 用 した 簡 便 でしかも 高 効 率 的 なヘテロ 環 合 成 法 を 開 拓 することに 成 功 した いずれの 反 応 もヨードニウムイオンを 経 由 し 立 体 選 択 的 にヘテ ロ 環 を 構 築 できることから 利 用 価 値 の 高 い 合 成 法 としてその 応 用 が 期 待 できる 文 献 1) Ando, T.; Kano, D.; Minakata, S.; Ryu, I.; Komatsu, M. Tetrahedron 1998, 54, 13485. 2) Kano, D.; Minakata, S.; Komatsu, M.. J. Chem. Soc., Perkin Trans. 1 2001, 3186. 3) Minakata, S.; Kano, D.; Oderaotoshi, Y.; Komatsu, M. Angew. Chem. Int. Ed. 2004, 43, 79. 4) Minakata, S.; Komatsu, M. Chem. Rev. 2009, 109, 711. 5) Minakata, S.; Kano, D.; Oderaotoshi, Y.; Komatsu, M. Org. Lett. 2002, 4, 2097. 6) Minakata, S.; Morino, Y.; Oderaotoshi, Y.; Komatsu, M. Org. Lett. 2006, 8, 3335. 7) Minakata, S.; Morino, Y.; Oderaotoshi, Y.; Komatsu, M. Chem. Commun., 2006, 3337. 8) Minakata, S.; Morino, Y.; Ide, T.; Oderaotoshi, Y.; Komatsu, M. Chem. Commun., 2007, 3279. + R t BuOCl, ai MeC, r.t., 5 h R 1 R 1, R 2 = alkyl, aryl R = aryl, n-bu, CF R 2 3 up to 79% O R 追 記 時 間 が 許 せば ライブラリー 合 成 によく 用 いられる 固 相 合 成 法 の 環 境 への 優 しさ における 問 題 点 と それを 解 消 する 合 成 例 (ケイ 素 の 特 性 を 活 用 した 1,3- 双 極 子 付 加 反 応 )についても 紹 介 したい 4

ダイヤモンドパウダ 回 収 のための 表 面 処 理 飯 塚 完 司 日 本 工 業 大 学 工 学 部 創 造 システム 工 学 科 345-8501 埼 玉 県 南 埼 玉 郡 宮 代 町 学 園 台 4-1,iizuka@nit.ac.jp 1 はじめに 工 業 材 料 としてのダイヤモンドは 研 磨 材, 潤 滑 材, 光 学 材 料, 半 導 体 デバイスなど 様 々な 分 野 で 利 用 されている.この 中 で, 磁 気 ディスクの 製 造 工 程 でディスクの 表 面 をダイヤモンドパウ ダで 研 磨 する 工 程 があり, 磁 気 ディスクの 記 憶 密 度 向 上 のため に 表 面 の 粗 さを 0.1nm オーダーで 制 御 する 必 要 がある 1). 研 磨 の 際 には,このダイヤモンドパウダと 純 水 を 混 ぜたスラリーと して 使 用 される. 環 境 を 考 えるとスラリーを 回 収 し, 再 利 用 す る 必 要 がある. 研 磨 に 使 用 される 通 常 のダイヤモンドパウダは, スラリーでの 拡 散 性 を 良 くするため 酸 性 溶 液 などを 用 いて 親 水 化 処 理 されている 4).この 表 面 には, 酸 素 を 伴 う 原 子 団 が 導 入 されることが 知 られている 2-3). 一 方,ダイヤモンドパウダを 回 収 する 際 には,スラリー 内 で 素 早 く 沈 殿 させる 必 要 がある. そこで, 本 研 究 では 単 結 晶 と 多 結 晶 のダイヤモンドパウダを 用 い,パウダの 表 面 へ 酸 素 などの 原 子 を 導 入 するため, 硫 酸 と 硝 酸 で 薬 品 処 理 を 行 った.また, 超 高 真 空 中 で 加 熱 処 理 を 施 し た 時 の 結 合 状 態 変 化 を X 線 光 電 子 分 光 法 (X-ray Photoelectron Spectroscopy: XPS) を 用 いて 評 価 した.そして, 各 処 理 を 施 した ダイヤモンドパウダを 水 溶 液 中 へ 拡 散 させ, 拡 散 性 の 評 価 を 行 い 表 面 状 態 と 拡 散 性 の 関 連 性 を 調 査 した. 2 実 験 方 法 実 験 には 市 販 の 単 結 晶, 多 結 晶 のダイヤモンドパウダを 用 い (a) Fig.1 SEM images of diamond powder used for the experiment (a) single (b) crystal diamond powder (b) polycrystalline diamond powder. た.Fig.1 に 本 研 究 で 使 用 したダイヤモンドパウダの 電 子 顕 微 鏡 (Scanning Electron Microscope: SEM) 像 を 示 す.これらのダイヤ モンドパウダに 硫 酸, 硝 酸 で 2 時 間 のボイル 処 理 を 行 い, 表 面 へ 酸 素 を 導 入 した.その 後, 超 高 真 空 中 へ 導 入 し 加 熱 を 行 った. 加 熱 温 度 を 100 C から 500 C まで 100 C おきに 変 化 させ 各 温 度 で 2 時 間 保 持 した. 保 持 終 了 後, 室 温 まで 温 度 を 下 げ 表 面 の 結 合 状 態 を 真 空 チャンバで 繋 がれた XPS にて 評 価 した.X 線 源 に は Mg Kα 線 を 使 用 し, 得 られたスペクトルのエネルギー 軸 補 正 には In の 3d 5/2 ピークを 用 いた.また,パウダ 内 部 の 不 純 物 の 特 定 及 び 加 熱 処 理 による 結 晶 性 の 変 化 を 見 るため,X 線 回 折 (X-ray diffraction: XRD) にて 測 定 を 行 った. 拡 散 性 の 評 価 は, 試 験 管 に 塩 酸 を 用 いて 水 素 イオン 濃 度 指 数 (potential Hydrogen: ph) を 1 から 5.5 に 調 整 した 溶 液 を 作 製 後, 各 試 験 管 に 同 量 のダイヤモンドパウダを 加 えてよくふり 拡 散 さ せた 後 静 置 し, 時 間 の 変 化 における 拡 散 性 を 観 察 することによっ て 行 った. 3 実 験 結 果 および 検 討 3.1 ダイヤモンドパウダの 表 面 分 析 Fig.2 に 薬 品 処 理 無 し 単 結 晶 ダイヤモンドパウダの C1s XPS 測 定 結 果 を 示 す. 図 中 の 破 線 はグラファイトの C1s 値 である 284.5eV を 示 す. 全 てのスペクトルが 高 エネルギー 側 へシフト しおり, 最 大 で 実 線 で 示 した 286.8eV までシフトしているのが 確 認 できる. 高 エネルギー 側 にシフトした 理 由 としては 以 下 の 2 つが 考 えられる.1 つは,ダイヤモンドなどの 絶 縁 物 の 試 料 表 面 にX 線 を 照 射 する 際, 帯 電 によりピークが 高 エネルギー 側 にシフトするチャージアップである.もう 一 つは, 周 囲 の 状 態 などにより 原 子 の 結 合 エネルギーに 変 化 が 生 じおこるケミカル シフトによるものである. 前 者 に 関 しては, 導 電 性 であるIn3d5/2 のピークを 使 って 帯 電 補 正 を 行 っているため 今 回 のピークシフ トの 原 因 はケミカルシフトによるものと 断 定 できる.XPSのデー タブックより C1s の 場 合,ケミカルシフトによってピーク 位 置 は 281~293eV の 範 囲 で 変 化 する.また, 加 熱 温 度 が 上 昇 する につれXPSスペクトルのピークが 高 エネルギー 側 へシフトして いることが 分 かる.しかし,500 C の 加 熱 により 加 熱 前 のピー ク 位 置 である 286.3eV 付 近 へ 戻 っているのが 確 認 できる.これ は, 加 熱 により 結 合 状 態 が 変 化 しているためと 考 えられる. しかし,このままでは 具 体 的 な 結 合 状 態 を 明 らかにすること ができない.そこで,カーブフィッティングを 行 い, 各 結 合 状 態 の 強 度 を 求 めた.Fig.3 に 単 結 晶 ダイヤモンドパウダにおける C1s 結 合 状 態 の 加 熱 温 度 依 存 性 を 示 す. 全 ての 試 料 で C-O-H の アルコール 結 合 が 支 配 的 であることが 明 らかになった.これは, 5

炭 素 が 酸 素 と 結 合 した 状 態 で 存 在 しているためであると 考 えら れる.また, 薬 品 処 理 無 しの 試 料 もアルコール 結 合 であること から 表 面 へ 酸 素 が 導 入 されていることが 明 らかになった.これ は,ダイヤモンドパウダの 製 造 工 程 で 粒 子 間 の 不 純 物 を 取 り 除 く 際 に, 行 われる 薬 品 処 理 によって 導 入 されたものであると 考 えられる. 次 に,O1s の 測 定 を 行 いダイヤモンドパウダ 表 面 の 酸 素 の 結 合 状 態 を 検 討 した.Fig.4 に 単 結 晶 ダイヤモンドパウダにおける O1s 結 合 状 態 の 加 熱 温 度 依 存 性 を 示 す. (a) の 薬 品 処 理 無 しの 試 料 では 全 体 を 通 して 金 属 酸 化 物 が 支 配 的 であることが 明 らか になった.これは, 貼 り 付 けに 用 いた In からのスペクトルであ ると 考 えられる. (b) の 硫 酸 処 理 を 施 した 試 料 では 炭 酸 基 や 硫 酸 基 が 主 に 見 られた.これは, 硫 酸 での 薬 品 処 理 によりダイヤ モンドパウダ 表 面 へ 導 入 されたものであると 考 えられる.(c) 硝 酸 処 理 を 施 した 試 料 では 硝 酸 基 が 確 認 できる.こちらも, 薬 品 処 理 によって 導 入 されたものであると 考 えられる. Fig.5に 多 結 晶 ダイヤモンドパウダのC1sの 結 合 状 態 加 熱 温 度 依 存 性 を 示 す. 多 結 晶 ダイヤモンドパウダの 表 面 は C-O-H のア ルコール 結 合 やC-C 単 結 合 が 支 配 的 であることが 明 らかになっ た.また,C-C 単 結 合 は 単 結 晶 では 見 られなかった 結 合 状 態 で ある.これは, 多 結 晶 ダイヤモンドパウダ 表 面 がグラファイト 等 の 不 純 物 に 覆 われているためであると 考 えられる. Fig.6に 多 結 晶 ダイヤモンドパウダのO1sの 結 合 状 態 加 熱 温 度 依 存 性 を 示 す. 薬 品 処 理 無 しの 試 料 では 200 C までの 加 熱 では 硝 酸 基 が 支 配 的 であったが 300 C 以 降 は 金 属 酸 化 物 が 支 配 的 に なることが 明 らかになった.これは,300 C 以 降 の 加 熱 により ダイヤモンドパウダ 表 面 の 酸 素 が 脱 離 し 以 降 は, 下 地 の In の 酸 素 が 現 われているためであると 考 えられる. 硫 酸 処 理 を 施 した 試 料 では 200 C 以 下 では 硫 酸 基,300 C 以 降 では 炭 酸 基 が 支 配 的 であることが 明 らかになった.これは, 硫 酸 処 理 によって 導 入 された 酸 素 は 500 C までの 加 熱 を 行 っても 炭 酸 基 としてダイ ヤモンドパウダ 表 面 に 残 るためであると 考 えられる. 硝 酸 処 理 Intensity (arb. units) 500 C 400 C 300 C 200 C 100 C Before heating 286.8eV 292 290 288 286 284 282 Binding Energy (ev) 284.5eV Fig.2 C1s XPS spectra of untreated single-crystal diamond powder. を 施 した 試 料 では, 全 体 を 通 して 硝 酸 基 が 支 配 的 であることが 明 らかになった.これは, 硝 酸 処 理 によって 導 入 された 酸 素 は 500 C の 加 熱 を 行 っても 硝 酸 基 として 残 っているためであると 考 えられる. XPS 測 定 で 得 られた 結 果 を 基 に 薬 品 処 理 によってダイヤモン Intensity (arb. units) 0 100 200 300 400 500 0 100 200 300 400 500 0 100 200 300 400 500 Heat treat. temp. ( C) Heat treat. temp. ( C) Heat treat. temp. ( C) (a) Untreated (b) Sulfuric acid (c) itric acid Fig.3 Heat-treatment temperature dependence of the C1s combination state of single-crystal diamond powder. C-C C-O-H C-O-H H-C=O Intensity (arb. units) 0 100 200 300 400 500 0 100 200 300 400 500 0 100 200 300 400 500 Heat treat. temp. ( C) Heat treat. temp. ( C) Heat treat. temp. ( C) (a) Untreated (b) Sulfuric acid (c) itric acid Fig.4 Heat-treatment temperature dependence of the O1s combination state of single-crystal diamond powder Carbonates Metal Oxides Sulfates itrates 6

ドパウダ 表 面 へ 導 入 された 酸 素 量 の 検 討 を 行 った.Table 1 に XPS スペクトル 積 分 強 度 による 酸 素 量 の 比 較 を 示 す. 硫 酸 処 理 と 硝 酸 処 理 では, 硫 酸 処 理 の 方 が 多 くの 酸 素 が 表 面 へ 導 入 され, 加 熱 により 脱 離 していることが 明 らかになった.これは, 硫 酸 処 理 の 280 C, 硝 酸 処 理 の 180 C とそれぞれの 薬 品 の 沸 点 の 違 いにより 薬 品 処 理 時 のボイル 温 度 に 差 が 生 じたため, 導 入 され た 酸 素 の 量 に 差 が 生 じたと 考 えられる.ダイヤモンドパウダの 種 類 では, 単 結 晶 ダイヤモンドパウダの 方 が 多 くの 酸 素 が 導 入 されていることが 明 らかになった.これは, 単 結 晶 はダイヤモ ンドの 表 面 が 出 ており 多 くの 酸 素 を 導 入 できたのに 対 して, 多 結 晶 では 表 面 が 不 純 物 等 で 覆 われており 酸 素 の 導 入 が 少 なく なったためであると 考 えられる. Fig.7 に 単 結 晶, 多 結 晶 ダイヤモンドパウダの XRD 測 定 結 果 を 示 す.どちらの 試 料 も, 図 中 に 破 線 で 示 した Diamond (111) 面 の 回 折 ピークが 見 られた.しかし, 多 結 晶 ダイヤモンドパウダ のみ 図 中 の 実 線 で 示 した Graphite (110) 面 の 回 折 が 見 られた. このことから, 多 結 晶 ダイヤモンドパウダは 粒 子 内 部 や 表 面 に グラファイトが 存 在 していることが 明 らかになった.Fig.8 に 加 熱 処 理 前 後 の 単 結 晶 ダイヤモンドパウダの XRD 測 定 結 果 を 示 す.ダイヤモンドは 真 空 中 では 1300 C から 炭 素 への 再 転 移 が 始 まる.しかし, 粒 子 の 細 かいダイヤモンドパウダは 温 度 が 低 くなり 約 900 C から 再 転 位 が 始 まるという 報 告 がある 5). 今 回 の 実 験 では 超 高 真 空 中 にて 500 C の 加 熱 を 行 ったが 加 熱 前 後 に おいて XRD スペクトルに 変 化 は 見 られず, 超 高 真 空 中 での 加 熱 によるダイヤモンドの 炭 素 への 転 移 等 は 見 られなかった. 3.2 拡 散 性 の 評 価 Fig.9 に 単 結 晶 ダイヤモンドパウダの 拡 散 性 の 変 化 を 示 す.pH は 左 から 0.5 刻 みで ph1 から ph5.5 とした. 未 処 理 の 試 料 では 6 時 間 後 ph1 から ph3 までが 沈 殿 しているが, 薬 品 処 理 を 施 し た 試 料 では ph3 でも 拡 散 が 維 持 されていることが 確 認 できる. これは, 薬 品 処 理 を 施 すことよって 粒 子 表 面 に 酸 素 が 導 入 され, Intensity (arb. units) Intensity (arb. units) Table 1 Amount of oxygen estimated by integrated intensity of the Sulfuric acid itric acid XPS spectra of the diamond powder. Type Single crystal Polycrystalline Single crystal Polycrystalline Graphite (011) Single crystal diamond powder Polycrystalline diamond powder Diamond (111) 30 40 50 2θ (deg.) Fig.7 XRD spectra of single and polycrystalline diamond powder. Before heating After heating Before heating 10.9 9.3 9.3 6.4 Graphite (011) Diamond (111) 30 40 50 2θ (deg.) Fig.8 XRD spectra of single crystal diamond powder with or without heat treatment After heating 1.5 1.4 2.3 6.2 Amount of desorption 9.4 7.9 7.0 0.2 Intensity (arb. units) 0 100 200 300 400 500 0 100 200 300 400 500 0 100 200 300 400 500 Heat treat. temp. ( C) Heat treat. temp. ( C) Heat treat. temp. ( C) (a) Untreated (b) Sulfuric acid (c) itric acid Fig.5 Heat-treatment temperature dependence of the C1s combination state of polycrystalline diamond powder. C-C C-O-H C-O-H H-C=O Intensity (arb. units) Carbonates Metal Oxides Sulfates 0 100 200 300 400 500 0 100 200 300 400 500 0 100 200 300 400 500 Heat treat. temp. ( C) Heat treat. temp. ( C) Heat treat. temp. ( C) (a) Untreated (b) Sulfuric acid (c) itric acid itrates Fig.6 Heat-treatment temperature dependence of the O1s combination state of polycrystalline diamond powder. 7

溶 液 中 に 含 まれる 水 素 イオンと 反 発 する 力 が 強 くなったためで あると 考 えられる.そして, 溶 液 中 へ 拡 散 させたすべてのダイ ヤモンドパウダは 時 間 経 過 に 伴 って 徐 々に 沈 殿 していくことが 確 認 できる.これは, 時 間 が 経 過 することによって, 粒 子 表 面 に 導 入 された 酸 素 と 溶 液 中 に 含 まれる 水 素 イオンの 反 発 する 力 が 弱 くなり, 粒 子 が 凝 集 し 沈 殿 したためであると 考 えられる. 24 時 間 後 では, 薬 品 処 理 を 施 したものは ph3 でも 拡 散 が 維 持 されており, 薬 品 処 理 を 施 すことによって 未 処 理 よりも 低 い ph 値 で 拡 散 が 維 持 されることが 明 らかとなった. Fig.10 に 多 結 晶 ダイヤモンドパウダの 拡 散 性 の 変 化 を 示 す. 薬 品 処 理 無 しの 試 料 では 6 時 間 後 ph3 まで 拡 散 が 維 持 さている 6 hours 24 hours 6 hours 24 hours Untreated Sulfuric acid itric acid Fig9 Change of the diffusivity of single crystal diamond powder. Untreated Sulfuric acid itric acid Fig10 Change of the diffusivity of polycrystalline diamond powder. 6 hours 24 hours Before heating 300 C 500 C Fig.11 Change of the diffusivity of single crystal diamond powder with heat treatment. ことが 確 認 できる.また, 薬 品 処 理 を 施 すことで ph2.5 まで 拡 散 が 維 持 され, 拡 散 性 が 良 くなることが 明 らかになった.これ は, 薬 品 処 理 によってパウダ 表 面 へ 導 入 された 酸 素 の 働 きによ るものであると 考 えられる.そして,24 時 間 後 では 全 ての 試 料 が 沈 殿 するのが 明 らかになった.これは, 多 結 晶 ダイヤモンド パウダは 粒 子 に 不 純 物 が 含 まれているため 表 面 へ 導 入 される 酸 素 の 量 が 少 なく, 長 時 間 に 渡 って 拡 散 状 態 を 維 持 することがで きなかったためであると 考 えられる. Fig.11 に 加 熱 処 理 を 施 した 単 結 晶 ダイヤモンドパウダの 拡 散 性 の 変 化 を 示 す. 加 熱 前 の 場 合 では 6 時 間 後 ph1 から ph3 ま でが 沈 殿 しているに 対 し,300 C の 加 熱 を 施 した 試 料 では ph1 から ph3.5 までが 沈 殿 し,500 C では 全 ての ph で 沈 殿 した. これは, 加 熱 処 理 を 施 すことよってパウダ 表 面 に 酸 素 が 脱 離 さ れ, 水 と 混 ざりにくい 疎 水 性 へ 変 化 し, 溶 液 中 に 含 まれる 水 素 イオンと 反 発 する 力 が 弱 くなり 粒 子 が 凝 集 し 沈 殿 したためであ ると 考 えられる.そして,300 C では 時 間 経 過 に 伴 って 徐 々に 沈 殿 し,24 時 間 後 の 時 点 では ph4 の 場 合 も 沈 殿 した.このこ とより, 加 熱 温 度 を 上 昇 させることによって, 拡 散 時 間 が 短 く なることが 明 らになった. 4 まとめ 本 研 究 では 表 面 処 理 を 施 したダイヤモンドパウダの 表 面 分 析 及 び 拡 散 性 実 験 を 行 った.その 結 果, 薬 品 処 理 を 施 したダイヤ モンドパウダ 表 面 には 酸 素 が 導 入 されることが 確 認 された.そ して, 導 入 された 酸 素 は 超 高 真 空 中 での 加 熱 により 脱 離 するこ とが 明 らかになった.XRD の 測 定 結 果 から 多 結 晶 ダイヤモンド パウダ 粒 子 にはグラファイトなどの 不 純 物 が 存 在 していること が 明 らかになった.また, 加 熱 前 後 のダイヤモンドパウダをし た 結 果 回 折 ピークに 変 化 は 見 られず, 超 高 真 空 中 での 加 熱 によ る 炭 素 等 への 転 移 は 見 られなかった. 拡 散 性 を 検 討 した 結 果, 単 結 晶 ダイヤモンドパウダでは 薬 品 処 理 を 施 すことで 拡 散 性 が 向 上 し 低 い ph 値 でも 長 時 間, 粒 子 が 拡 散 した. 一 方, 多 結 晶 ダイヤモンドパウダでは 薬 品 処 理 に よる 拡 散 性 の 改 善 が 見 られたが, 長 時 間 拡 散 状 態 を 保 つことが できなかった.そして, 加 熱 処 理 を 施 したダイヤモンドパウダ では 加 熱 温 度 が 上 がるほど 拡 散 性 が 悪 くなり 疎 水 性 へと 変 化 す ることが 明 らかになった. 通 常,ハードディスクの 研 磨 に 用 いられている 単 結 晶 ダイヤ モンドパウダでは 未 処 理, 薬 品 処 理 において 拡 散 と 沈 殿 を 起 こ す ph には 違 いがあるものの,いづれの 場 合 でも ph を 0.5 程 度 変 えることで 拡 散 状 態 から 沈 殿 状 態 に 変 化 させることが 可 能 であることが 明 らかとなった.これにより 繊 維 フィルタ 等 での ダイヤモンドパウダ 除 去 の 前 に 溶 液 による 分 別 回 収 の 可 能 性 が あることが 示 された. 謝 辞 本 実 験 で 用 いたダイヤモンドパウダは, 株 式 会 社 サミットスー パーアブレーシブ 社 からの 提 供 による.また, 株 式 会 社 ディス コ 大 島 龍 司 氏 には 貴 重 なコメントをいただいた. 本 研 究 は, 当 研 究 室 の 大 学 院 修 了 生 古 川 祐 司 君 により 推 進 されたもので ある. 参 考 文 献 1) 吉 川 晶 範, 内 田 宏, 斉 藤 良 夫 : ダイヤモンド 技 術 総 覧 (GT 出 版, 東 京 都,2007) p.514. 2) T. Andou, Y. Ishii, M. Kashige and Y. Sato: Japan ew Diamond Fo rum Vol.10, o.2 (1994) 2-8. 3) T. Andou, Y. Ishii, M. Kashige and Y. Sato: Japan ew Diamond Fo rum Vol.10, o.3 (1994) 2-7. 4) R. Oshima, H. Yamanaka and S. Hosomi: J. Society of Materials Scie nce Vol.49, o.6 (2000) 602. 5) H. Makita: Japan ew Diamond Forum Vol.12, o.3 (1996) 8. 8

半導体基板上に担持した有機金属触媒 阿南工業高等専門学校 塚本史郎 小西智也 西脇永敏*1 東條孝志 平山基 植田有紀子 立石清 多田孝 森時秀司 木原義文 遠野竜翁 武藏美緒 立石学 石川琢馬 森本真司 三並貴大 寺岡輝記 日亜薬品工業株式会社 小池晴夫 奥山彰 杉岡智教 近藤竜二 西本遼右 大都裕希*2 鳥取大学 石井晃 横山真美 小田泰丈 物質 材料研究機構 下田正彦 吉川英樹 日本原子力研究機構 高橋正光 藤川誠司 英国ウォーリック大学 Gavin R.Bell 医 農薬などのファインケミカルやその他の化学工業分野における有機合成では 均一系有機金属触 媒が広く使用されている このプロセスは反応効率が高く有用な手法であるものの 反応後に精製 分 離の問題 触媒金属と合成原料 生成物との分離 や使用済み触媒の廃棄処理の問題がある これら の問題を解決するためには 分離が容易で再利用が可能な不均一系の環境調和型固体担持有機金属 触媒が有望であり これまで硫黄終端ＧａＡｓ表面上有機金属触媒に関しての研究を行ってきた 現在 その活性メカニズムをＳＴＭＢＥ装置 分子線エピタキシィ ＭＢＥ 成長その場走査型トンネル顕微鏡 Ｓ ＴＭ 観察装置 大型放射光施設ＳＰｒｉｎｇ８ 第一原理計算などの手法を用いて より詳細に解析を行 い 更に高活性で汎用性のある工業化に適したＧａＮを用いた新型触媒材料の開拓に取り組んでいる ナノテクノロジー 原子レベル評価 ナノ界面制御 電子材料 化合物半導体 GaAs, InAs, Ga 環境調和型担持触媒開発 ナノ材料設計 作製 高活性メカニズム解析 機能性材料 有機金属触媒 Pd(PPh3)4, Pd(OAc)2 応用分野 医薬品 有機ＥＬ 色素 STMBE 装置 分子線エピタキシィ MBE 成長その場走査型トンネル顕微鏡 STM 観察装置 STMBE system STM is placed completely inside MBE growth chamber and it is possible to perform true dynamic imaging. *1 現在 高知工科大学 *2 阿南工業高等専門学校 派遣技術員 9

JST 地 域 イノベーション 創 出 総 合 支 援 事 業 重 点 地 域 研 究 開 発 推 進 プログラム( 育 成 研 究 ) 窒 化 ガリウム 基 板 を 用 いた 固 定 型 遷 移 金 属 触 媒 の 開 発 : 研 究 代 表 者 塚 本 史 郎 2008.4-2011.3 阿 南 高 専 作 製 調 整 基 板 改 良 日 亜 薬 品 合 成 反 応 窒 化 ガリウム 基 板 上 への 遷 移 金 属 触 媒 定 着 条 件 の 探 索 設 定 確 立 *ガリウム- 硫 黄 - 触 媒 結 合 確 認 計 算 解 析 : 鳥 取 大 学 元 素 分 析 : 物 質 材 料 研 究 機 構 構 造 分 析 : 日 本 原 子 力 研 究 開 発 機 構 より 簡 便 な 基 板 の 探 索 供 給 反 応 結 果 フィードバック 触 媒 活 性 評 価 実 際 の 製 造 現 場 での 適 用 反 応 時 間 の 短 縮 と スケールアップを 考 慮 高 周 波 を 用 いた 反 応 検 討 フォーカスドライブラリー の 作 成 受 託 合 成 その 他 MS 産 経 ニュース 共 同 通 信 日 経 ビジネス 特 許 1 件 査 読 付 英 文 論 文 7 件 ( 内 3 件 投 稿 中 ) 国 際 会 議 18 件 国 内 会 議 20 件 10

Organopalladium catalyst on S-terminated GaAs 001-2 Ã 6 surface Tomoya Konishi, Takashi Toujyou, and Takuma Ishikawa Anan ational College of Technology, Tokushima 774-0017, Japan Gavin R. Bell Department of Physics, University of Warwick, Coventry CV4 7AL, United Kingdom Shiro Tsukamoto a Anan ational College of Technology, Tokushima 774-0017, Japan Received 17 February 2009; accepted 30 June 2009; published 8 September 2009 Organopalladium molecules, such as Pd CH 3 COO 2 Pd, immobilized on the S-terminated GaAs 001, termed GaAs S Pd have high catalytic activity and cycle durability in the Mizoroki Heck reaction. It is thought that the presence of Ga S bonds in the single atomic layer S-termination is essential for these catalytic properties despite the much higher thickness 100 nm of the Pd films. In this study, the authors demonstrate the retention of Ga S bonds in ultrathin GaAs S Pd by using reflection high-energy electron diffraction and scanning tunneling microscopy STM. The ultrathin GaAs S Pd was prepared by using a vapor-deposition technique. Deposited Pd was observed as 1 nm dotlike structures with STM. The adsorption rate of Pd was also investigated. 2009 American Vacuum Society. DOI: 10.1116/1.3193687 I. ITRODUCTIO An unusual, highly active and reusable heterogeneous catalyst for the Mizoroki Heck reaction 1,2 has been discovered recently, 3 5 best described as the three-component system GaAs S Pd. The Mizoroki Heck reaction is one of the important organic reactions forming a C C bond and is widely used for drug discovery sciences and pharmaceutical industries. 6 The substrate for GaAs S Pd is S-terminated GaAs 001-2 6 GaAs S, produced by immersion of GaAs 001 wafers in H 4 S x solution 3,7 or vapor deposition of S. 8,9 Pd CH 3 COO 2 Pd is dissolved in acetonitrile and deposited on the GaAs S substrates by immersion, resulting in structured Pd-containing thin films which repeatedly give nearly 100% yield in the Mizoroki Heck reaction. 4 On the other hand, deposited Pd on other surfaces has so far shown poor catalytic properties. For example, poor catalytic activity is realized on the S-terminated Si surface and poor cycle durability occurs on the S-terminated Au or nonterminated GaAs 001. 10 Although emphasis is placed on the S-termination before the immobilization of Pd for the GaAs S Pd to exhibit the fascinating catalytic property, 3,5 the surface molecular structure including Ga S bonds after immobilization of Pd has not yet been investigated. It is vital to examine further the surface molecular structure of GaAs S Pd by employing the tools of analytical surface science to begin to understand the mechanism of the high catalytic activity in this unique system. In this article we demonstrate the vapor deposition of Pd on a GaAs S surface and report surface observations with reflection highenergy electron diffraction RHEED and scanning tunneling microscopy STM in an ultrahigh vacuum UHV. a Author to whom correspondence should be addressed; electronic mail: tsukamot@anan-nct.ac.jp II. EXPERIMETAL PROCEDURES The sample preparation and the measurements were completely performed in a UHV molecular beam epitaxy MBE chamber equipped with RHEED and STM as described elsewhere. 11 Pieces 11 13 0.6 mm 3 of GaAs 001, which were cut out from a wafer, were used as substrates. First the surface of the substrate was thermally cleaned to remove the oxide layer under 1 10 4 Pa of an arsenic atmosphere in a MBE chamber. ext, GaAs buffer layer was grown on the surface by MBE until atomically smooth surface was obtained. The substrate was cooled to room temperature to form c 4 4 reconstruction. The substrate was transferred to a load lock chamber to terminate the c 4 4 surface with S. The surface was exposed to 1 10 3 Pa of S vapor for 5 min. The substrate was again transferred to the MBE chamber and was annealed at 400 C for 12 h in an UHV. The substrate was subjected to RHEED and STM observation to confirm the formation of S-terminated GaAs 001-2 6 reconstruction 9,12 and the absence of any contamination on the S layer. The vapor deposition of Pd on the GaAs S surface was then performed in the load lock chamber. For the vapor deposition of Pd we prepared a sublimation cell by using a tubular ceramic crucible wound by a tungsten wire heater. Since powdered Pd sublimates as a trimer at 155 C, 13 this cell was designed for the thermal stability in the range of 120 180 C. We confirmed the sublimation of Pd in this temperature range by Arrhenius behavior of the vapor pressure. The substrates were then vapor deposited with Pd by the cell at the partial pressure of 6 10 5 Pa for 4 and 8 s. These conditions correspond to supplying at least 1 and 2 MLof Pd, respectively, on the basis of the atomic pair density of GaAs 001. 2206 J. Vac. Sci. Technol. B 27 5, Sep/Oct 2009 1071-1023/2009/27 5 /2206/3/$25.00 2009 American Vacuum Society 2206 11

2207 Konishi et al.: Organopalladium catalyst on S-terminated GaAs 001-2Ã6 surface 2207 FIG. 1. RHEED patterns of GaAs S surface before vapor deposition of Pd. FIG. 2. RHEED patterns of GaAs S surface after vapor deposition of Pd. The GaAs S Pd sample was again transferred to the MBE chamber and was heated up to 100 C for RHEED measurement and STM observation with tip bias voltage of 2.8 V and set-point current of 0.2 na. III. RESULTS AD DISCUSSIO Figures 1 and 2 show the RHEED patterns of the substrate surface before and after the deposition of Pd, respectively. From the brightness line profile analysis, the patterns indicated 2 6 reconstruction brought by S-termination. The deposition of Pd did not add any other electron diffraction pattern to the 2 6 pattern. This may be due to the low coverage of Pd Table I or that the deposited Pd did not form any periodic structure. Moreover, the pattern did not change by the vapor deposition of Pd. Similar results were obtained when the amount deposited was increased much more. This implies that vapor deposition of Pd did not corrupt the 2 6 reconstruction. It is known that the uniform surface with 2 6 reconstruction shows better stability of Pd in GaAs S Pd than roughly S-covered surface. 3 Figure 3 shows the STM images of the substrate surface after the vapor deposition of Pd for 4 and 8 s. The dimer rows of S were clearly observed in the background. The line profile along the 1 10 direction suggested the dimer row spacing of 0.8 nm by the 2 6 reconstruction of S-terminated surface. The STM images also show scattered dotlike structures, which were not observed before the deposition, on the background S-dimer rows. The 2 6 reconstruction structure near the dotlike structure was not corrupted. As the line profile pattern across the dotlike structure is also superimposed on Fig. 3 a, the typical size of the dotlike structure was 1 nm width and 0.2 nm height. On the surface with 8 s deposition the dotlike structures were observed more with some clusters in places than that with 4 s deposition. The estimation of amount supplied, amount deposited, and adsorption rate of Pd on the GaAs S surface is presented in Table I. The amount deposited increased depending on the amount supplied. The coverage rate was estimated to be 2% for the both deposition conditions. The similar width dimension 1nm of the dotlike structures observed in Fig. 3 is found in the literature. 14 They report the STM investigation of in situ complexation of Pd by a monolayer of a bipyridine derivative at a graphite surface. In their work, a Pd molecule is known to have the dimension of 1 nm width with STM. Therefore, accounting for their number density and width dimension, the dotlike structures in Fig. 3 are very likely to be Pd molecules. The STM line profile pattern was also investigated on the Pd molecule placed on a cleavage surface of a highly oriented pyrolytic graphite HOPG piece. Since the similar profile height 0.2 nm was also measured on the Pd molecule on the HOPG surface to that on the GaAs S sur- TABLE I. Estimation of amount supplied, amount deposited, and coverage rate of Pd by vapor deposition. Deposition time s Amount supplied cm 2 Amount deposited a cm 2 Coverage rate % 4 6.0 10 14 9 10 12 2 8 1.2 10 15 2 10 13 2 a Estimated form the STM images. JVST B - Microelectronics and anometer Structures 12

2208 Konishi et al.: Organopalladium catalyst on S-terminated GaAs 001-2Ã6 surface 2208 (a) 2 nm 0.8 nm 5 nm (b) [110] [110 [110] [110 mechanism. The 2 6 reconstruction consists of five S S adatom dimers, one missing dimmer, and strong Ga S bonds, 16 satisfying the local-charge neutrality in themselves. 9 Because the electrons of S are attracted by Ga in Ga S bonds, it is probable that the binding affinities of S and Pd are moderated so that Pd should easily separate from S to participate in chemical reactions. Further discussion needs more information on the electronic structure of Ga S bonds by vibrational spectroscopic analyses or molecular orbital calculations. IV. COCLUSIO Pd CH 3 COO 2 was vapor deposited on S-terminated GaAs 001-2 6 surface. The RHEED patterns and STM images showed that Pd CH 3 COO 2 molecules were immobilized on the surface with preserving the 2 6 reconstruction structure and hence the substructural Ga S bonds. This supports the assumption that the high catalytic activity and stability are owing to the amount of Ga S bonds. The result is potentially one of the important clues to understand the mechanism of the high catalytic activity and stability of the GaAs S Pd catalyst. 5 nm ACKOWLEDGMETS This research was partially supported by Japan Science and Technology Agency JST. One of the authors T.K. would like to thank Dr. agatoshi ishiwaki for the valuable discussion on catalyst-poison effect of GaAs S. FIG. 3. Color online STM images of GaAs S surface after vapor deposition of Pd for a 4 s and b 8 s. Line profile pattern along the line marker is also superimposed. face known from the line profile in Fig. 3 a, the Pd molecules are settled on the S reconstruction layer without corrupting or substituting it. This is consistent with the RHEED results that 2 6 reconstruction pattern is remaining after the deposition of Pd. From the RHEED and STM investigations, it was concluded that the Pd molecules are immobilized on the GaAs S surface with preserving the 2 6 reconstruction structure and hence the substructural Ga S bonds. This result supports the assumption that the S-termination and Ga S bonds are essential for the high catalytic activity and stability of GaAs S Pd catalyst. 3,5 Generally S is known as a catalyst poison that significantly deactivates organometallic catalysts because of the formation of a strong metal-s bond. 15 It is interesting that GaAs S Pd keeps high catalytic activity and stability even with the presence of S. The evidence of preserving S layer after immobilization of Pd on GaAs S Pd surface may also give one of the important clues to understand the 1 T. Mizoroki, K. Mori, and A. Ozaki, Bull. Chem. Soc. Jpn. 44, 581 1971. 2 R. F. Heck and J. P. olley, J. Org. Chem. 37, 2320 1972. 3 I. Takamiya, S. Tsukamoto, M. Shimoda,. Miyashita, and M. Arisawa, Chem. Lett. 33, 1208 2004. 4 M. Arisawa, M. Hamada, I. Takamiya, M. Shimoda, S. Tsukamoto, Y. Arakawa, and A. ishida, Adv. Synth. Catal. 348, 1063 2006. 5. ishiwaki, M. Shimoda, T. Konishi, and S. Tsukamoto, Appl. Phys. Express 2, 051002 2009. 6 M. Larhed and A. Hallberg, Drug Discovery Today 6, 406 2001. 7 M. S. Carpenter, M. R. Melloch, M. S. Lundstrom, and S. P. Tobin, Appl. Phys. Lett. 52, 2157 1988. 8. Koguchi, K. Ishige, and S. Takahashi, J. Vac. Sci. Technol. B 11, 787 1993. 9 S. Tsukamoto and. Koguchi, Appl. Phys. Lett. 65, 2199 1994. 10 M. Arisawa, S. Tsukamoto, M. Shimoda, M. Pristovsek, and A. ishida, Jpn. J. Appl. Phys., Part 2 41, L1197 2002, and unpublished data. 11 S. Tsukamoto and. Koguchi, J. Cryst. Growth 201 202, 118 1999 ; 209, 258 2000. 12 J. Suda, Y. Kawakami, S. Fujita, and S. Fujita, Jpn. J. Appl. Phys., Part 2 35, L1498 1996. 13 H. Schäfer, C. Brendel, H. Rabeneck, and E. Schibilla, Z. Anorg. Allg. Chem. 518, 168 1984. 14 M. M. S. Abdel-Mottaleb,. Schuurmans, S. D. Feyter, J. V. Esch, B. L. Feringa, and F. C. D. Schryver, Chem. Commun. Cambridge 17, 1894 2002. 15 J. K. Dunleavy, Platinum Met. Rev. 50, 110 2006. 16 H. Oigawa, J.-F. Fan, Y. annichi, K. Ando, K. Saiki, and A. Koma, Jpn. J. Appl. Phys., Part 2 28, L340 1989. J. Vac. Sci. Technol. B, Vol. 27, o. 5, Sep/Oct 2009 13

Applied Physics Express 2 (2009) 051002 Green Chemical Catalyst Supported on S-Terminated Ga(0001) agatoshi ishiwaki, Masahiko Shimoda 1, Tomoya Konishi, and Shiro Tsukamoto Center for Collaborative Research, Anan ational College of Technology, Anan, Tokushima 774-0017, Japan 1 Photocatalytic Materials Center, ational Institute for Materials Science, Sengen, Tsukuba, Ibaraki 305-0047, Japan Received March 5, 2009; accepted April 10, 2009; published online May 1, 2009 A novel function of nitride-based semiconductor is successfully developed for organic synthesis, in which palladium supported on the surface of sulfur-terminated Ga(0001) serves as a unique green chemical catalyst. It efficiently catalyzes Heck reaction with simple manipulations and its catalytic activity is retained for several repeat reactions. Moreover, it is easily reused without any special treatment. A plausible mechanism for Pd adsorption is provided for the first time; the SH groups on the surface of the substrate attract Pd 2þ, and reduce to Pd 0. The presence of Pd 0 on the surface was confirmed by X-ray photoelectron spectroscopy measurements. # 2009 The Japan Society of Applied Physics DOI: 10.1143/APEX.2.051002 Sulfur termination (S-termination) on a GaAs(001) surface 1,2) has received great attention because it is associated with improved device properties, such as enhanced photoluminescence, and an increased sensitivity of the Schottky barrier height to the metal work function, because the surface states are reduced within a band gap. 3 6) On the other hand, transition-metal-catalyzed reactions have played an important role in synthetic and process chemistry. Homogeneous catalysts are certainly effective. 7,8) However, much effort and energy are required for recovering and recycling the catalyst or for disposal of waste fluid containing heavy metal since it is quite difficult to separate products and catalyst. Thus, heterogeneous catalysts (supported catalysts) are widely used in industrial chemistry because of their easier tractability, though some restrictions remain. 9) For example, palladium catalyst adsorbed on activated carbon (Pd/C) should be carefully treated as combustible material, and the reactivation of the used catalyst is somewhat troublesome. 10) Although an easily recyclable polymer-supported-catalyst has been developed recently, it cannot tolerate high temperatures and it sometimes suffers from swelling by organic solvents. 11,12) From this viewpoint, development of an easily treatable heterogeneous catalyst is desired for reducing waste of expensive rare metal. Therefore, we have reported palladium catalyst supported on S-terminated GaAs(001) undergoes Heck reaction, which is one of the important organic reactions forming a C C bond, and the catalyst is readily recycled several times with high catalytic activity. 13 16) Furthermore, the catalyst is quite stable under heated conditions and in organic solvents. Hence, the catalyst is superior to commonly used heterogeneous catalysts. However, highly toxic As is an obstacle for use in medicinal chemistry. In the present work, we studied the development of a novel palladium catalyst supported on S-terminated Ga(0001), which keeps the important bond of Ga S without As. The catalytic plate was prepared via three steps: (1) the S- termination, (2) the adsorption of palladium species (Pd adsorption), and (3) the aging (Fig. 1), and then the catalytic activity of the plate was evaluated by Heck reaction. The optimized procedure for preparing the catalytic plate (chemical method) and Heck reaction are shown as follows. The Ga substrate (11 13 mm 2 ) was heated at 60 Cin E-mail address: tsukamot@anan-nct.ac.jp 14 Table I. Study on S-termination and Pd adsorption. Substrate Sulfur termination Pd adsorption Conversion (%) Method Reagent Pd(OAc) 2 (Heck reaction) A Physical S 8 Physical 0 B Physical S 8 Chemical 18 C aþ Chemical (H 4 ) 2 S x Chemical 49 D aþ Chemical a 2 S 4 Chemical 53 E aþ Chemical a 2 S Chemical 48 F aþ Chemical ash Chemical 54 G Chemical ash Chemical 100 aþ S-Termination was performed by heating Ga substrate in an aqueous 5% solution of sulfur reagent at 60 C for 0.5 h. Then the plate was heated in CH 3 C solution of Pd(OAc) 2 at 80 C for 24 h. The plate was employed for Heck reaction without aging. 20% ash aqueous solution (3 ml) for 0.5 h, and washed with water (1 ml). To a suspension of palladium acetate [Pd(OAc) 2, 5.0 mg] in acetonitrile (CH 3 C, 3 ml), the S-terminated substrate was added, and heated at 80 C for 12 h. After washing with CH 3 C (2 ml), the plate was heated in xylene (3 ml) at 130 C for 12 h to afford catalytic plate G (see Table I). To a solution of iodobenzene (56 L, 0.5 mmol), methyl acrylate (54 L, 0.6 mmol), and triethylamine (Et 3, 90L, 0.65 mmol) in CH 3 C (3 ml), the catalytic plate was added, and the resultant mixture is heated at 100 C in a sealed tube with monitoring by high performance liquid chromatography (HPLC). The S-termination and the Pd adsorption were conducted by both physical and chemical methods. In the physical method, a Ga(0001) film grown on a sapphire(0001) was loaded into a molecular beam epitaxy (MBE) sampleintroduction chamber. 1) After pre-baking at 300 C for 10 min in the introduction chamber, the sample was transferred to the MBE growth chamber and baked at 800 C for 5 min without any atmosphere, which resulted in the formation of a Ga-rich surface. 17) Immediately after cooling to 200 C, the sample was transferred back to the introduction chamber. In this chamber, the sulfur deposition was performed by exposing the sample to S 8 vapor at 100 200 C, which elevated the chamber pressure to 2:0 10 3 Pa. By these procedures, S Ga bonds are formed in the same way as in the S-terminated GaAs(001). 1,2,15) In the chemical method, to the contrary, Ga substrates were simply heated in aqueous 051002-1 # 2009 The Japan Society of Applied Physics

Appl. Phys. Express 2 (2009) 051002. ishiwaki et al. Ga plate Catalytic plate Fig. 1. Preparative procedures for catalytic plates. I + OMe Et 3 (1.3 equiv) O CH 3 C (1.0 equiv) (1.2 equiv) 100 C, 1 d O OMe Recycle 1st 2nd 3rd 4th 5th Conversion/% 100 89 86 82 83 Fig. 2. Recycling of catalytic plate G for Heck reaction. solutions of sulfur reagents, and the substrates were heated with an CH 3 C solution of Pd(OAc) 2 for the Pd adsorption. Six catalytic plates A F were prepared by combining these methods (Table I). ote that, for plate A, we used a Pd(OAc) 2 sublimation cell 18) which was composed of a tubular ceramic crucible wound by a tungsten wire heater to give thermal stability in the range of the sublimation temperature of Pd(OAc) 2, 120 180 C. While plate B catalyzed Heck reaction, plate A did not show catalytic activity sufficiently, which indicates enough palladium was not adsorbed and/or fixed on the S- terminated surface by the physical vapor deposition. Since Pd(OAc) 2 volatilizes as a trimer at 155 C in vacuum, 19) it is considered that intact Pd(OAc) 2 physically adsorbs on the surface without strong interaction with sulfur atoms. In the chemical method, on the other hand, (H 4 ) 2 S x (H 4 þ S S x 2 S H 4 þ ), a 2 S 4 (a þ S S 2 S a þ ), and a 2 S (a þ S 2 a þ ) were usable as sulfur reagents for plates C, D, and E, respectively, revealing similar catalytic activity. Thus, the S-termination is not significantly concerned with counter cation and length of the S S chain in the sulfur reagents; only a single sulfur atom is enough for the subsequent Pd adsorption. In other words, the hydrolysis or protonation of the S S chain by water forming a thiol group ( SH) is an important process in the S-termination, which is supported by the insufficient result for plate A because of the absence of water in the MBE chamber. This consideration was confirmed by using ash which directly introduces the SH group on the surface of Ga substrate (plate F). To our expectation, plate F exhibited similar catalytic activity to other plates C E. Among sulfur reagents, ash was most suitable with regard to treatability and atom-economy. As a result of optimizing the reaction conditions at each step, plate G exhibiting high catalytic activity was successfully prepared. Moreover, as shown in Fig. 2, plate G was recycled more than 5 times without considerable loss of the catalytic activity. Our catalytic plate is easily separated from the reaction mixture by simply 15 picking up with tweezers and can be reused without special treatment. On the basis of above experimental facts, a plausible mechanism is illustrated in Fig. 3. The S-termination easily proceeds by forming strong Ga S bonds. 2) When sulfur reagents were used having multiple sulfur atoms such as S 8 and (H 4 ) 2 S x, water hydrolyzes the S S bond to afford a SH group whose attracting ability of metal species is well known. 20) Thus, Pd(OAc) 2 is also attracted to the surface of the substrate by complex formation or by their own affinity. 21) In Heck reaction, Pd 0 atoms and/or nanoclusters serve as the actual catalyst and they are recovered as Pd 0 after passing through the catalytic cycle. Therefore, it is considered the Pd 2þ on the surface is reduced to Pd 0 by adjacent SH groups during Pd adsorption and aging steps. 21,22) The presence of palladium species on the surface was confirmed by X-ray photoelectron spectroscopy (XPS) measurements using plate G. As shown in Fig. 4(a), clear peaks of the Pd 3d core-level photoemission are observed from the catalyst even after being subjected to Heck reaction. This indicates that palladium species on the surface are immobilized and stable enough to survive during the catalytic reaction. ote that the spectrum was observed with Al K radiation (h ¼ 1487 ev) and that the energy shift due to sample charging was corrected using the lowest binding energy C 1s peak (285.0 ev) as a reference. The chemical state of Pd is evaluated by comparing this spectrum with the spectra from metal Pd(0) [Fig. 4(b)] and Pd(II)(OAc) 2 [Fig. 4(c)]. The Pd 3d peaks of the catalyst appear at significantly lower binding energy than that of Pd(II)(OAc) 2 and very close to that of metal Pd(0), suggesting the valence of Pd is close to 0 rather than þ2 and the Pd exists as Pd(0) species or nanoclusters. The XPS experimental results are consistent with reduction mechanism as shown in Fig. 3 though further investigation should be continuously conducted. ote that Pd(OAc) 2 was measured separately using synchrotron radiation with the BL15XU beam line in SPring-8. 051002-2 # 2009 The Japan Society of Applied Physics

Appl. Phys. Express 2 (2009) 051002. ishiwaki et al. Fig. 3. A plausible mechanism for S-termination and Pd adsorption. Fig. 4. Pd 3d photoemission spectra from (a) the catalyst after Heck reaction, (b) metal Pd(0), and (c) Pd(II)(OAc) 2. In summary, a novel function of nitride-based semiconductors is successfully developed, which serves as a unique substrate of the catalyst for Heck reaction. The catalyst is environmentally acceptable due to the easy treatability and recyclability. The study on the application of this catalyst to other organic reactions and study on the surface structure are now in progress. Acknowledgments This research was supported by Japan Science and Technology Agency (JST). Dr. Gavin Bell (The University of Warwick) is thanked for useful discussion. We are also grateful to agao Co., Ltd. for supplying sulfur reagents. 1) S. Tsukamoto and. Koguchi: Appl. Phys. Lett. 65 (1994) 2199. 2) S. Tsukamoto, T. Ohno, and. Koguchi: J. Cryst. Growth 175 176 (1997) 1303. 3) Q.-H. Fan, Y.-M. Li, and A. S. C. Chan: Chem. Rev. 102 (2002) 3385. 4). E. Leadbeater and M. Marco: Chem. Rev. 102 (2002) 3217. 5) C. E. Song and S.-G. Lee: Chem. Rev. 102 (2002) 3495. 6) J. Dupont, R. F. de Souza, and P. A. Z. Auarez: Chem. Rev. 102 (2002) 3667. 7) G. W. Parshall and S. D. Ittle: Homogeneous Catalysis (Wiley, ew York, 1992). 8) Applied Homogeneous Catalysis by Organometallic Complexes, ed. B. Cornils and W. A. Herrmann (Verlag Chemie, Weinheim, 1996). 9) Immobilized Catalysts: Solid Phases, Immobalization and Applications, ed. A. Kirschning (Springer, Berlin, 2004). 10) S. ishimura: Handbook of Heterogeneous Catalytic Hydrogeneation for Organic Synthesis (Wiley, ew York, 2001). 11) Y. Uozumi: Top. Curr. Chem. 242 (2004) 77. 12) S. Kobayashi, H. Miyaura, R. Akiyama, and T. Ishida: J. Am. Chem. Soc. 127 (2005) 9251. 13) M. Arisawa, S. Tsukamoto, M. Shimoda, M. Pristovsek, and A. ishida: Jpn. J. Appl. Phys. 41 (2002) L1197. 14) I. Takamiya, S. Tsukamoto, M. Shimoda,. Miyashita, M. Arisawa, Y. Arakawa, and A. ishida: Chem. Lett. 33 (2004) 1208. 15) I. Takamiya, S. Tsukamoto, M. Shimoda, M. Arisawa, A. ishida, and Y. Arakawa: Jpn. J. Appl. Phys. 45 (2006) L475. 16) M. Arisawa, M. Hamada, I. Takamiya, M. Shimoda, S. Tsukamoto, Y. Arakawa, and A. ishida: Adv. Synth. Catal. 348 (2006) 1063. 17) A. R. Smith, R. M. Feenstra, D. W. Greve, M. S. Shin, M. Skowronski, J. eugebauer, and J. E. prthrup: Appl. Phys. Lett. 72 (1998) 2114. 18) T. Konishi, T. Toujyou, T. Ishikawa, G. R. Bell, and S. Tsukamoto: umpublished. 19) H. Schäfer, C. Brendel, H. Rebeneck, and E. Schibilla: Z. Anorg. Allg. Chem. 518 (1984) 168. 20) T. Kang, Y. Park, and J. Yi: Ind. Eng. Chem. Res. 43 (2004) 1478. 21) Y. egishi, H. Murayama, and T. Tsukuda: Chem. Phys. Lett. 366 (2002) 561. 22) M. Arisawa, C. Sugata, and M. Yamaguchi: Tetrahedron Lett. 46 (2005) 6097. 051002-3 # 2009 The Japan Society of Applied Physics 16

SAPd SGPd Scheme 1 Scheme 2, eq 1 Scheme 2, eq 2 SGPd SGPd SAPd SAPd SGPd SAPd SAPd SAPd SGPd SAPd 1a 2a 3a Table 1 SAPd SGPd 17

1b 1j Table 2 SAPd SGPd SAPd SAPd SAPd SAPd 3 Table 2 SAPd [1] a) K. C. icolaou, P. G. Bulger, D. Sarlah, Angew. Chem. Int. Ed. 2005, 44, 4442. b) C. Torborg, M. Beller, Adv. Synth. Catal. 2009, 351, 3027. [2] a). Miyaura, K. Yamada, A. Suzuki, Tetrahedorn Lett. 1979, 20, 3437. b). Miyaura, A. Suzuki, Chem. Rev. 1995, 95, 2457. [3], C. E. Garrett, K. Prasad, Adv. Synth. Catal. 2007, 346, 889. [4] a) D. E. De Vos, M. Dams, B. F. Sels, P. A. Jacobs, Chem. Rev. 2002, 102, 3615. b) L. Yin, J. Leibsher, Chem. Rev. 2007, 107, 133. [5] a) M. Arisawa, M. Hamada, I. Takamiya, M. Shimoda, S. Tsukamoto, Y. Arakawa, A. ishida, Adv. Synth. Catal. 2006,348, 1063. b). Hoshiya,. Isomura, M. Shimoda, H. Yoshikawa, Y. Yamashita, K. Iizuka, S. Tsukamoto, S. Shuto, M. Arisawa, ChemCatChem 2009, 1, 279. [6]. Hoshiya, M. Shimoda, H. Yoshikawa, Y. Yamashita, S. Shuto, M. Arisawa, J. Am. Chem. Soc. 2010, 132, 7270. 18

Published on Web 05/12/2010 Sulfur Modification of Au via Treatment with Piranha Solution Provides Low-Pd Releasing and Recyclable Pd Material, SAPd aoyuki Hoshiya,, Masahiko Shimoda, Hideki Yoshikawa, Yoshiyuki Yamashita, Satoshi Shuto, and Mitsuhiro Arisawa*, Faculty of Pharmaceutical Sciences, Hokkaido UniVersity, Kita 12, ishi 6, Kita-ku, Sapporo 060-0812 Japan, ational Institute for Materials Science, 1-2-1 Sengen, Tsukuba, Ibaraki 305-0047 Japan, and IMS Beamline Station at SPring-8, ational Institute for Materials Science, 1-1-1 Kouto, Sayo-cho, Sayo-gun, Hyogo 679-5148 Japan Received December 2, 2009; E-mail: arisawa@pharm.hokudai.ac.jp Research involving self-assembled monolayers (SAMs) of alkylthiols (RSH) on metal (especially gold) surfaces has rapidly expanded due to the potential applications of these materials, the ease of their preparation, and their facile connection between organic moieties and metal surfaces. 1-5 However, since studies on the thermal stability of SAMs are limited, and the role of the solvent in the formation of SAMs is not yet well understood, questions still remain concerning the mechanism of monolayer formation. 6,7 Prior to the emergence of SAMs, Piranha solution (Danger! A strong oxidizer.) had been traditionally used to clean gold surfaces. 5,8-12 In an effort to create reusuable, Pd-leach-free catalysts, many heterogeneous catalysts have been developed with Pd immobilized on supports such as activated carbon, inorganic solids, and polymers. 13,14 Yet few of these immobilized Pd catalysts for C-C bond formations can claim both high recyclability (>10 times) and low Pd-leaching (<1 ppm). People purify the product several times and sometimes they use commercially available Pd scavengers, which increase the cost. In this report, we offer new insight into the treatment of Au with Piranha solution to induce sulfur modification. In addition, we show the application of this process to the development of novel Pd materials that exhibited both high recyclability and remarkably low Pd-leaching in the Suzuki-Miyaura coupling. In the course of our research to develop an environmentally benign catalyst, 15 we unexpectedly found in an SR-HXPS (Synchrotron radiation hard X-ray photoelectron spectroscopy) analysis that Piranha treatment can place a sulfur atom on the gold surface. The sulfur is expected to be located under the Pd(dba) 2 /Pd overlayers. The usual laboratory-xps is disadvantageous in observing the underlying sulfur because of its short observation depth. SR-HXPS with the long observation depth up to 20 nm enables the high sensitivity detection of the underlying sulfur. 16 SR-HXPS measurements were performed at the beamline BL15XU in the synchrotron radiation facility SPring-8. 17 We measured the SR- HXPS spectrum of Piranha-treated Au(111)/mica and found, to our surprise, that the S 1s peak appeared near 2478 ev in the spectrum (Figure 1 Red line), although no S 1s peak was observed on Au(111)/mica before the Piranha treatment. These results indicated that the surface of Au(111)/mica had been modified by a sulfur species upon Piranha treatment. Although Piranha treatment has been frequently used to remove impurities from the gold surface, 5,8-12 our results show that this treatment could also be employed for sulfur modification of the gold surface. To the best of our Figure 1. Sulfur 1s core-level photoemission spectra. Light blue, a 2 SO 3 ; Pink, PdS; Blue, A after 10 times of Suzuki-Miyaura coupling; Green, A before Suzuki-Miyaura coupling; Red, Piranha-treated Au(111)/mica. knowledge, this is the first report that Piranha treatment attaches S atoms to the gold surface. Since the binding energy of the S 1s peak for Piranha-treated Au(111)/mica is close to that of a 2 SO 3, Hokkaido University. ational Institute for Materials Science. IMS Beamline Station at SPring-8. Furuya Metal Company Limited, Tokyo, 170-0005 Japan. Figure 2. Palladium 2p 3/2 core-level photoemission spectra. Red, Pd(meal); Green, A before Suzuki-Miyaura coupling; Blue, A after 10 times of Suzuki-Miyaura coupling; Pink, PdS. 7270 9 J. AM. CHEM. SOC. 2010, 132, 7270 7272 10.1021/ja9100084 2010 American Chemical Society 19

COMMUICATIOS Table 1. Yields of the Suzuki-Miyaura Coupling Using Pd Materials A-D a Yield of 3a (%) b Entry Pd material Support Pd source first second third fourth fifth sixth seventh eighth ninth 10th 1 A Au(111)/mica Pd(dba) 2 96 95 97 96 98 99 >99 96 96 99 2 B Au(foil) Pd(dba) 2 97 96 92 89 97 97 97 99 98 95 3 C Au(mesh) Pd(dba) 2 >99 >99 >99 >99 >99 >99 >99 >99 >99 >99 4 D (SAPd) Au(mesh) Pd(OAc) 2 >99 98 >99 >99 >99 96 >99 97 99 >99 b Determined by HPLC. a Conditions of Suzuki-Miyaura coupling: 1a (102 mg, 0.5 mmol), 2a (1.5 equiv), K 2 CO 3 (2 equiv), EtOH (3 ml), Pd material, 80 C, 12 h. Table 2. Amount of Released Pd into Reaction Mixture in the Suzuki-Miyaura Coupling of 1a and 2a Amount of Releasing Pd (ng) b Entry Pd material Pd (µg) a first second third fourth fifth sixth seventh eighth ninth 10th Total 1 B 95 90 60 50 50 40 40 30 110 50 30 550 2 C 80 60 80 70 70 30 60 20 50 30 10 480 3 D (SAPd) 38 ( 9 c 36 ( 15 26 ( 9 25 ( 11 28 ( 23 24 ( 15 28 ( 14 19 ( 8 30 ( 8 25 ( 12 23 ( 13 260 ( 110 a Amount of absorbed Pd on Au substrate. The standard deviation was calculated from 4 sets of samples. b The whole reaction mixture was directly subjected to ICP-Mass measurement. The standard deviation was calculated from 4 sets of samples. c Pd (%) based on the amount of starting material is 7.0 10-2. Table 3. Suzuki-Miyaura Coupling of Various Substrates Using Pd Material D Substrate [Ar-X] Boronic acid Yield of 3 (%) a,c Entry Ar X Y first second third fourth fifth sixth seventh eighth ninth 10th 1 b 1a C 6 HM 5 I 2a Cl 3a >99 (36) 98 (26) >99 (25) >99 (28) >99 (24) 97 (28) >99 (19) 97 (30) 99 (25) >99 (23) 2 I 2b Me 3b 94 (55) >99 92 97 99 94 90 92 >99 94 (47) 3 I 2c H 3c 92 (43) 92 92 90 91 91 85 87 91 89 (36) 4 I 2d Ac 3d 99 98 >99 98 97 99 99 98 97 99 5 1b 4-O 2 C 6 H 4 I 2a Cl 3e 96 (1307) 96 90 93 >99 96 94 >99 99 >99 (295) 6 I 2b Me 3f 93 96 99 96 88 99 94 95 >99 95 7 I 2c H 3g 97 99 90 >99 98 96 96 98 99 95 8 I 2d Ac 3h 96 >99 89 87 95 92 98 98 94 98 9 b 1c 4-MeOC 6 H 4 I 2a Cl 3i 96 (641) 97 98 96 93 97 95 97 96 92 (54) 10 I 2b Me 3j >99 >99 >99 >99 95 >99 >99 >99 >99 96 11 I 2c H 3k 99 99 96 91 98 95 >99 98 96 93 12 b I 2d Ac 3l 97 94 99 99 99 95 94 96 97 92 13 1d 4-O 2 C 6 H 4 Br 2a Cl 3e >99 (482) 94 >99 >99 >99 >99 >99 >99 >99 99 (116) 14 1e 4-Ac-C 6 H 4 Br 2a Cl 3m 95 97 99 99 99 93 81 95 86 80 15 1f 2-Me-C 6 H 4 I 2b Me 3n 96 97 93 94 92 92 91 99 95 89 16 1g 2-HO-C 6 H 4 I 2a Cl 3o 96 97 93 94 92 92 91 99 95 89 a Isolated yields. b Yields were determined by HPLC. c umbers in parentheses indicate the amount of releasing Pd(ng) in the solution. which was used as a reference sample (Light blue line), the sulfur on Au appears to be oxidized sulfur. Given that sulfur has a high affinity for Pd, Pd(dba) 2 was adsorbed on this sulfur modified Au(111)/mica in xylene (100 C, 12 h). After the Pd adsorption, the S 1s peak appears at 2470 ev (Green line), which is almost identical to the binding energy of the S 1s peak from the S-modified GaAs. 15 Significant changes in S 1s binding energies between those in the Piranha-treated Au(111)/ mica and those in the sample A indicate that the sulfur on Au was reduced and had chemically bonded with Pd during Pd adsorption. 18 In Pd 2p core-level photoemission spectra, the peaks from A appear close to the metallic palladium peak. These results suggest that, in the case of A, Pd(dba) 2 molecules react with the substrate, yielding zerovalent molecules or metallic palladium (Figure 2). 19 Single crystal Au(111)/mica is very expensive and not versatile. Consequently, we used Au foil and mesh instead of Au(111)/mica J. AM. CHEM. SOC. 9 VOL. 132, O. 21, 2010 7271 20

COMMUICATIOS to prepare B and C according to the Pd-adsorption procedure for A. When A, B, or C was subjected to the Suzuki-Miyaura coupling of iodobenzene and 4-chlorophenylboronic acid, the yields for runs 1 to 10 were excellent to quantitative in all cases (Table 1, entries 1-3). D, prepared from Au(mesh) with Pd(OAc) 2 as a Pd source, was also highly active for the Suzuki-Miyaura coupling (entry 4). These results showed that the Pd materials A, B, C, and D are highly recyclable. 20 We next measured the amount of Pd adsorbed on the Pd materials B-D with inductively coupled plasma mass spectrometry (ICP- MS) analysis. These analyses revealed that B, C, and D included 95, 80, and 38 µg of Pd, respectively. 21 We subsequently measured the amount of Pd released into the reaction mixtures of each Suzuki-Miyaura coupling run by ICP-MS. Table 2 shows that the amount of released Pd in each run was extremely low. The amount of Pd in the reaction mixture using B-D is far lower than the U.S. government-required value of <5 ppm residual metal in product streams. 22 In particular, in the case of D, the leached Pd into the reaction mixture was only 76-38 ng for 1 mmol scale preparation (12.7-6.3 ppb in 3 ml of solvent, 0.2-0.1% of Pd from D), the average being 26 ng for 10 runs. The Crudden group reported an excellent Pd material supported on mercaptopropyl-modified mesoporous silica, which has been recognized as one of the best catalytic Pd materials from the point of view of leaching: 720-30 ng for 1 mmol scale preparation (72-3 ppb in 2.5 ml of solvent, 0.006-0.13% of Pd from the Pd material), the average being 242 ng for 4 runs. 23,24 The amount of the leached Pd in the reaction of our Pd material D is similar or less, compared with that of Crudden s material. Furthermore, since Crudden s material can only be recycled 5 times, D would be considered highly recyclable, making it one of the lowest releasing Pd materials with high recyclability. Due to its extremely low Pd leaching levels, we employed D, i.e., SAPd (Sulfur-modified Au supported Pd material) for further studies. We then investigated the scope and limitation of SAPd in the Suzuki-Miyaura coupling using aryl iodides and arylboronic acids. The corresponding products were obtained in excellent yields as summarized in Table 3. It is noteworthy that isolated yields (%) of 3i between the first run to fourth run were 54, 46, 8, and 5 respectively, when the control experiment of entry 9 was carried out using SAPd without Piranha treatment. These control experiments indicate Piranha treatment is necessary and sulfur is needed to create an active catalyst or retain Pd on the surface. In summary, we have found in the SR-HXPS measurement of Piranha-treated Au(111)/mica that the gold surface underwent sulfur modification during this treatment, which was believed to have only removed impurities from the gold surface. We also successfully developed a practical Pd material, SAPd, whose Pd was immobilized on sulfur-modified Au. With the lowest Pd-releasing levels and high recyclability, this is one of the best Pd materials thus far developed. Because it leaches extremely low levels of Pd into reaction mixtures, removal of the residual Pd is unnecessary using SAPd, even in syntheses involving pharmaceutical ingredients. Acknowledgment. We are grateful to D. omoto, Dr. S. Ueda, and Dr. K. Kobayashi (IMS, SPring-8), Prof. S. Tsukamoto, Dr. T. Konishi, K. Tateishi, and T. Tojo (Anan ational College of Technology) as well as Dr. K. Iizuka, M. Yokota, and Y. Furukawa (ippon Institute of Technology) for measuring SR-HXPS. This research was partially supported by Grants-in-Aid from the Ministry of Education, Culture, Sports, Science and Technology, Japan. Supporting Information Available: Experimental procedures and full characterizations of compounds. This material is available free of charge via the Internet at http://pubs.acs.org. References (1) Review: Ulman, A. An Introduction to Ultrathin Organic Films: From Langmuir-Blodgett to Self-Assembly; Academic Press: ew York, 1991. (2) Review: Ulman, A. Chem. ReV. 1996, 96, 1533 1554. (3) Review: Flink, S.; van Veggel, F. C. J. M.; Reinhoudt, D.. AdV. Mater. 2000, 12, 1315 1328. (4) Review: Gooding, J. J.; Mearns, F.; Yang, W.; Liu, J. Electroanalysis 2003, 15, 81 96. (5) Review: Love, J. C.; Estroff, L. A.; Krebel, J. K.; uzzo, R. G.; Whitesides, G. M. Chem. ReV. 2005, 105, 1103 1169. (6) ishida,.; Hara, M.; Sasabe, H.; Knoll, W. Jpn. J. Appl. Phys. 1996, 35, 5866 5872. (7) Grönbeck, H.; Curioni, A.; Andreoni, W. J. Am. Chem. Soc. 2000, 122, 3839 3842. (8) Voicu, R.; Ellis, T. H.; Ju, H.; Leech, D. Langmur 1999, 15, 8170 8177. (9) Chen, I.-W. P.; Chen, C.-C.; Lin, S.-Y.; Chen, C.-H. J. Phys. Chem. B 2004, 108, 17497 17504. (10) Petrovykh, D. Y.; Suda, H. K.; Opdahl, A.; Richter, L. J.; Tarlov, M. J.; Whitman, L. J. Langmur 2006, 22, 2578 2587. (11) Pisarek, M.; Lewandowska, M.; Roguska, A.; Kurzydlowski, K. J.; Janik- Czachor, M. Mater. Chem. Phys. 2007, 104, 93 97. (12) Seu, K. J.; Pandey, A. P.; Haque, F.; Proctor, E. A.; Ribbe, A. E.; Hovis, J. S. Biophys. J. 2007, 92, 2445 2450. (13) Phan,. T. S.; Van Der Sluys, M.; Jones, C. W. AdV. Synth. Catal. 2006, 348, 609 679. (14) Yin, L.; Liebscher, J. Chem. ReV. 2007, 107, 133 173. (15) For example: Hoshiya,.; Isomura,.; Shimoda, M.; Yoshikawa, H.; Yamashita, Y.; Iizuka, K.; Tsukamoto, S.; Shuto, S.; Arisawa, M. ChemCatChem 2009, 1, 279 285. (16) This does not mean a sulfur atom is embedded at an 20 nm depth, and it is unclear whether the sulfur atoms are embedded at the interface between gold and Pd adlayers or not. (17) Proposal o. 2007B4600 and o. 2008A4605. For the measurements, the incident X-ray energy was 5.95 KeV and a total energy resolution of 240 mev was employed. (18) It should be noted that the sulfur for sample A remained in almost the same chemical state before (Figure 1, Green line) and after (Figure 1, Blue line) the Suzuki-Miyaura coupling. Pd-XPS was shown in Figure 2. (19) We did not use Pd 3d peaks but Pd 2p peaks, since the peaks of Pd 3d 5/2 or 3d 3/2 overlap with that of Au 4d 5/2 of substrates in the energy region. (20) D could be recycled at least 20 times. (21) Au(foil) in B is 13 11 0.127 mm 3 (325 mg). Au(mesh) in C and D is 12 12 mm 3 (100 mesh, 99 mg). D included 14 ( 9 µg ofs(n ) 4). (22) Flahive, E. J.; Ewanicki, B. L.; Sach,. W.; O eill-slawecki, S. A.; Stankovic,. S.; Yu, S.; Guinness, S. M.; Dunn, J. Org. Process Res. DeV. 2008, 12, 637 645. (23) Crudden, C. M.; Sateesh, M.; Lewis, R. J. Am. Chem. Soc. 2005, 127, 10045 10050. (24) Webb, J. D.; MacQuarrie, S.; McLeney, K.; Crudden, C. M. J. Catal. 2007, 252, 97 109. JA9100084 7272 J. AM. CHEM. SOC. 9 VOL. 132, O. 21, 2010 21

J. Med. Chem. 2010, 53, 3585 3593 3585 DOI: 10.1021/jm901848b Investigation of the Bioactive Conformation of Histamine H 3 Receptor Antagonists by the Cyclopropylic Strain-Based Conformational Restriction Strategy Mizuki Watanabe,,# Takatsugu Hirokawa,,# Takaaki Kobayashi, Akira Yoshida, Yoshihiko Ito, Shizuo Yamada, aoki Orimoto, Yasundo Yamasaki, Mitsuhiro Arisawa, and Satoshi Shuto*, ) ) Faculty of Pharmaceutical Sciences, Hokkaido University, Kita-12, ishi-6, Kita-ku, Sapporo 060-0812, Japan, Computational Biology Research Center, ational Institute of Advanced Industrial Science and Technology, Aomi, Koutou-ku, Tokyo 135-0064, Japan, Department of Pharmacokinetics and Pharmacodynammics and Global Center of Excellence (COE), School of Pharmaceutical Sciences, University of Shizuoka, Yada, Shizuoka 422-8526, Japan, and Hanno Research Center, Taiho Pharmaceutical Co. Ltd., Misugidai, Hanno 357-8527, Japan. # M.W. and T.H. contributed equally to this work. Received December 15, 2009 ) We previously identified the highly potent histamine H 3 receptor antagonists (1R,2S)-2-[2-(4-chlorobenzylamino)ethyl]-1-(1H-imidazol-4-yl)cyclopropane (1) and its enantiomer ent-1. Although the conformations of 1 and ent-1 are restricted by the central cyclopropane ring, the 2-(4-chlorobenzylamino)ethyl side chain essential for the H 3 receptor binding may somewhat freely rotate. To investigate the bioactive conformation, the 1 0 -ethyl-substituted derivatives 2a and 2b and their enantiomers ent-2a and ent-2b were designed as side chain conformation-restricted analogues of 1 and ent-1, based on the cyclopropylic strain. These compounds were synthesized, and their analysis by MR and calculations suggested that the side chain moiety was effectively restricted in a syn-form or an anti-form by the cyclopropylic strain as expected. Pharmacological evaluation and docking simulation showed that the bioactive conformations of 1 and ent-1 appear to be the syn-form and the anti-form, respectively. Thus, the cyclopropylic strain can be effectively used for conformational restriction of the side chain moiety of cyclopropane compounds. Introduction Attention has been focused on the histamine H 3 receptor, which is a G-protein-coupled receptor (GPCR a ) distributed mainly in the central nervous system. 1 Antagonists to the H 3 receptor are considered to be potential drugs for various diseases, such as Alzheimer s disease, attention-deficit/ hyperactivity disorder (ADHD), schizophrenia, depression, dementia, obesity, and epilepsy. 1b,d,e As a result, attempts to develop H 3 receptor antagonists have led to the identification of potent H 3 receptor ligands, 1b-f some of which are shown in Figure 1. GPCRs are considered to be major targets for drug development. 2 Indeed, it is estimated that over 50% of all modern drugs are targeted at GPCRs. 2a However, because of the membranous nature of these proteins and their very low natural abundance, structural analysis of GPCRs is difficult. In fact, until the most recent resolution of the adrenergic β 2 receptor structures, 3 the only high-resolution structure of a GPCR available had been that of bovine rhodopsin. 2b One obvious drawback in drug development targeting GPCRs is therefore poor structural data on these proteins. *To whom correspondence should be addressed. Phone and fax: þ81-11-706-3769. E-mail: shu@pharm.hokudai.ac.jp. a Abbreviations: ADHD, attention-deficit/hyperactivity disorder; EBA, Epstein-Barr virus nuclear antigen-1; GPCR, G-proteincoupled receptor; GPCRDB, G-protein-coupled receptors data bank; IFD, induced fit docking; OPLS-AA, optimized potentials for liquid simulations-all-atom; PGME, phenylglycine methyl ester; XP, extraprecision. Conformational restriction of neurotransmitters may improve the specific binding to one of the receptor subtypes. 4 In conformationally restricted analogues highly selectively bound to the target receptor, the functional groups essential for the receptor binding must assume a special arrangement superimposed on the bioactive conformation, in which these functional groups effectively interact with certain amino acid residues in the binding pocket of the receptor. The major problem in designing conformationally restricted analogues specific for a receptor subtype is that the conformation of the conformationally flexible lead compound that binds to the target subtype, i.e., the bioactive conformation, is often unknown. This is mainly because structural analysis of membranebound proteins is tremendously difficult, 2 compared with that of proteins soluble in blood or cytosol. Thus, a method for effectively identifying compounds targeting GPCRs, which do not involve structural data, would be highly useful in drug development. Consequently, we have devised a stereochemical diversity-oriented conformational restriction strategy to develop compounds that bind selectively to target proteins of unknown structure such as GPCRs. 5 In this strategy, the versatile chiral cyclopropane units with different stereochemistries (Figure 2) 5a are effectively used as the key tool for the design and synthesis of a series of conformationally restricted analogues with stereochemical diversity. 5 On the basis of this stereochemical diversity-oriented strategy, we recently designed and synthesized a series of conformationally restricted analogues of histamine, as shown in Figure 2, with different stereochemistries. 5 In these analogues, the imidazole r 2010 American Chemical Society Published on Web 04/16/2010 pubs.acs.org/jmc 22

3586 Journal of Medicinal Chemistry, 2010, Vol. 53, o. 9 Watanabe et al. and the amino groups are located in a variety of spatial arrangements because of the conformational restriction. Some of these analogues were shown to be potent H 3 receptor ligands, and a conformationally restricted analogue 1 with a (1R)-transcyclopropane structure and its enantiomer ent-1 (Figure 3) were identified as highly potent H 3 receptor antagonists. 5c With these results in hand, we thought that, based on the structures of 1 and ent-1, identification of the bioactive conformation for the H 3 receptor antagonists might be possible. If indeed this could be accomplished, the information obtained would be useful in designing further effective compounds. Although the conformation of 1 is restricted by the (1R)-trans-cyclopropane structure, the spatial arrangement of the basic nitrogen of the side chain, which seems to be essential for activity, would be somewhat flexible. Therefore, we decided to restrict the side chain conformation by the cyclopropylic strain-based method, 6,7 a detail of which is described below, to identify the bioactive conformation. During our study, X-ray crystallographic structures of adrenergic β 2 receptors were reported, 3 and as a consequence, the H 3 receptor model using the structural data of a β 2 receptor was constructed and used for investigating the binding conformation of the conformationally restricted analogues. In this report, we describe the design, synthesis, pharmacological effects, and receptor modeling studies of the cyclopropylic strain-based conformationally restricted analogues 2a and 2b and their enantiomers ent-2a and ent-2b (Figure 3) for the identification of their bioactive conformations. structure, cyclopropane is effective in restricting the conformation of a molecule without changing the chemical and physical properties of the lead compound. 8 A characteristic structural feature of cyclopropane is that cis-oriented adjacent substituents on the ring exert significant mutual steric repulsion because they are fixed in the eclipsed orientation, which we previously termed cyclopropylic strain. 7 Consequently, conformation of the substituents on a cyclopropane can be restricted so that the steric repulsion due to the strain is minimal, as indicated in Figure 4. Considering that the basic amino function of the H 3 receptor antagonist 1 is likely to be important for the H 3 receptor binding, the conformation of the side chain would significantly affect the activity of the compound. While the cyclopropyl-c1 0 (C2-C1 0 ) bond may freely rotate somewhat, the rotation can be restricted by the cyclopropylic strain. Therefore, the two conformers A (syn, the C-3 of the cyclopropane up /the benzylaminomethyl up ) and C (anti, the C-3 of the cyclopropane up /the benzylaminomethyl down ) would be preferable to conformer Results and Discussion Cyclopropylic Strain-Based Design of the Conformationally Restricted Analogues. Because of its small and rigid ring Figure 3. Previously synthesized H 3 receptor antagonists 1 and ent-1 and their side chain conformationally restricted analogues newly designed. Figure 1. Histamine and representative H 3 receptor ligands. Figure 4. Cyclopropylic strain-based conformational restriction. Figure 2. A series of conformationally restricted analogues of histamine with stereochemical diversity synthesized from the chiral cyclopropane units. 23

Article Journal of Medicinal Chemistry, 2010, Vol. 53, o. 9 3587 Scheme 1 Figure 5. Presumed stable conformations of 1 (a), 2a (b), and 2b (c). Figure 6. Stable structures of the conformationally restricted analogues in the gas phase obtained by the conformational search program in MacroModel: (a) the syn-conformer and the anticonformer for 1; (b) the syn-conformer for 2a and the anti-conformer for 2b. B in compound 1 because of the significant steric repulsion for the adjacent cis-proton in conformer B, as shown in Figure 5a. The conformation of 1 was analyzed by molecular mechanics calculations with MacroModel (Schr odinger, LLC). As a result, as shown in Figure 6a, two significantly stable structures were obtained, which correspond to the syn- and the anti-conformers in Figure 5a, respectively. The two conformers are nearly equally stable, while the anti-conformer is only 0.31 kcal/mol more stable than the syn-conformer. Thus, the results of calculations are in accord with the above hypothesis on the conformation of 1, which suggests that the bioactive conformation may be analogous to either the syn-conformer or the anti-conformer. We designed the 1 0 -ethyl-substituted derivatives 2a and 2b (Figure 3) as side chain conformation-restricted analogues of 1. Introducing an ethyl group into the R-position of the amino function of 1 would prevent the rotation of the side chain moiety by restricting the conformation due to the cyclopropylic strain, i.e., the steric repulsion for the adjacent eclipsed proton. Accordingly, depending on the configuration at the C1 0 position, the conformation of the compounds can be restricted; the syn-conformer would be quite stable in 2a of the 1 0 R-configuration (Figure 5b); conversely, the anticonformer would be stable in 2b of the 1 0 S-configuration (Figure 5c). Thus, while cyclopropane is very effective for conformational restriction of conformationally flexible lead compounds, the cyclopropylic strain-based conformational restriction makes the more precise conformational restriction of cyclopropane compounds possible, especially in the side chain moiety. Synthesis. Although much effort has been devoted to developing practical methods for preparing chiral cyclopropanes, synthesis of cyclopropane derivatives with the desired stereochemistry is often troublesome. 9 The chiral cyclopropane units (Figure 2), which are composed of four stereoisomeric cyclopropane derivatives bearing two adjacent carbon substituents in a cis or a trans relationship, are useful for the synthesis of various cyclopropane compounds, particularly for those with stereochemical diversity. 5 As summarized in Scheme 1, the target compounds 2a and 2b were synthesized from an imidazolylcylcopropanecarboxaldehyde 5 with the (1R,2R)-structure, which was prepared from the unit 4 by our previous method. 5a Treatment of 5 with EtMgBr in THF gave a diastereomeric mixture of the addition products, the Dess-Martin oxidation of which afforded the corresponding ketone 6. Wittig reaction of 6 with MeOCH 2 - PPh 3 Cl/a(TMS) 2 followed by acidic treatment gave the 24

3588 Journal of Medicinal Chemistry, 2010, Vol. 53, o. 9 Watanabe et al. Figure 7. Determination of the 1 0 -configurations based on the Δδ values (DMSO-d 6, 500 MHz) of the (R)- and (S)-PGME amides. Figure 8. OE data of 2a and 2b in D 2 O. aldehyde 7 as an inseparable diastereomeric mixture (dr, 1/1.1). Reductive amination of 7 with 4-chlorobenzylamine and 2-picoline borane in AcOH/MeOH and subsequent acidic removal of the trityl group of the product gave the desired cyclopropylic strain-based conformationally restricted analogues 2a and 2b as a diastereomeric mixture. Although the diastereomers were inseparable at this stage, they were successfully separated by HPLC after protection of the imidazole and amino nitrogens with Boc groups to give the diastereomerically pure 8a and 8b, respectively. Acidic removal of the Boc groups of 8a and 8b afforded the target (1 0 R)-product 2a and the (1 0 S)- product 2b, respectively. The enantiomers ent-2a and ent-2b were similarly synthesized from ent-5. The 1 0 -configurations of the cyclopropylic strain-based conformationally restricted analogues synthesized were determined by the phenylglycine methyl ester (PGME) method 10 by converting diastereomerically pure ent-7a and ent-7b into the corresponding (R)-PGME amides ent-9a and ent-9b and (S)-PGME amides ent-10a and ent-10b, 11 respectively, as shown in Figure 7. Conformational Analysis by MR and Calculations. Stable structures of the conformationally restricted analogues 2a and 2b were investigated by OE experiments (Figure 8). Irradiations of H-1 0 of 2a or 2b gave OEs with both H-1 and H-3a oriented cis to H-1. Especially significant OEs were observed between H-1 0 and H-1 in both 2a and 2b, which show that the side chain conformation of the two compounds seems to be actually restricted by the cyclopropylic strain. When H-1 on the cyclopropane ring of the (1 0 R)-diastereomer 2a was irradiated, an OE with the terminal methyl protons of the ethyl group was observed to suggest that it is restricted to the syn-form as expected. On the other hand, during irradiation of H-1 of the (1 0 S)- diastereomer 2b, an OE was observed with the methylene proton H-2 0 a adjacent to the basic nitrogen to demonstrate that it is in the anti-form, as expected. The conformations of 2a and 2b were also examined by calculations with MacroModel. As shown in Figure 6b, the most stable structures obtained by the calculations were the syn-form for 2a and the anti-form for 2b, respectively. Table 1. Effects of Compounds on the Human H 3 Receptor Subtype a compd configuration conformation inhibition (%) b K i (nm) 2a (1R)-trans-(1 0 R) syn 95 19.8 ( 2.8 2b (1R)-trans-(1 0 S) anti 73 129 ( 6.0 1 (1R)-trans syn/anti 100 8.4 ( 1.5 c ent-2a (1S)-trans-(1 0 S) syn 96 63.0 ( 6.1 ent-2b (1S)-trans-(1 0 R) anti 88 6.7 ( 0.4 ent-1 (1S)-trans syn/anti 100 3.6 ( 0.4 c thioperamide 98 51.1 ( 3.8 c a Assays were carried out with 293-EBA cells or cell membranes expressing the human H 3 receptor subtype. b Inhibitory effect of compound (10-4 M) on the agonistic activity of histamine (10-6 M). c Data were taken from ref 5c. The calculated energy barriers for the rotation of the C2-C1 0 bond between the syn-form and the anti-form are significant, which are 5.22 kcal/mol for 2a and 5.55 kcal/mol for 2b, respectively. Thus, these conformational analyses suggested that the cyclopropylic strain-based conformational restriction seems to work effectively in 2a and 2b, and therefore, pharmacological evaluations of these compounds would help to identify the bioactive conformation. Pharmacological Effects. Effects of compounds on the H 3 receptor were investigated by luciferase reporter gene assay. The human histamine receptor subtypes were individually expressed in 293-Epstein-Barr virus nuclear antigen-1 (EB- A) cells according to the previously reported method, 5b and the function of the compounds on these receptors expressed on the cells was evaluated. one of the newly synthesized compounds 2a, 2b, ent-2a, and ent-2b showed any agonistic activity to the H 3 receptor at 10-5 M (data not shown). On the other hand, all of these compounds inhibited the agonistic effect of histamine to show that they are antagonists of the H 3 receptor as are the parent compounds 1 and ent-1, as shown in Table 1. Binding affinities of compounds 2a, 2b, ent-2a, and ent-2b for the human H 3 receptor subtype using [ 3 H] R -methylhistamine 5c were next investigated and were compared with those of their parent compounds 1 and ent-1 (Table 1). In this system, the well-known H 3 receptor antagonist thioperamide showed a K i value of 51.1 nm, and compounds 1 and ent-1 displayed much higher binding affinity for the human H 3 receptor as shown by the K i values of 8.4 and 3.6 nm, respectively. Although compound 2a, which is restricted in the synconformation, showed remarkable binding affinity for the receptor with a K i value of 19.8 nm, compound 2b, restricted in the anti-conformation, showed more than 15-fold reduction of potency (K i = 129 nm), compared with the parent compound 1. On the other hand, of the enantiomers ent-2a and ent-2b, ent-2b (K i =6.7 nm), which is restricted in the anti-conformation, showed 10-fold higher binding affinity for the human H 3 receptor than the ent-2a (K i = 63 nm), which is restricted in the syn-conformation. Docking Simulation by Homology Modeling. The above conformational analysis and pharmacological results showed that in the diastereomeric pair of 2a and 2b, the syn-restricted 2a was more potent than the anti-restricted 2b, while in their enantiomers ent-2a and ent-2b, the antirestricted ent-2b was more potent than the syn-restricted ent-2a. In order to understand the discrepancy in the conformation-activity relationship between 2a/2b and ent-2a/ ent-2b, we planned to perform a docking simulation with a 25

Journal of Medicinal Chemistry, 2010, Vol. 53, o. 9 3589 Article homology modeling of the H3 receptor to investigate the binding mode of these cyclopropylic strain-based conformationally restricted analogues in the active site of the H3 receptor. Previous studies showed that homology models of H3 receptor are useful for providing structural insight into the ligand binding mechanism, QSAR analysis, and in silico drug discovery.13 Furthermore, most recent resolutions of the ligand-binding adrenergic β2 receptor structures3 give us a chance to generate more accurate three-dimensional models for target GPCRs with a ligand using homology modeling and docking simulation.14 Thus, in this study, a three-dimensional model of the H3 receptor was constructed on the basis of a structural template from the crystal structure of the human β2-adrenergic GPCR recently reported by Cherezov and co-workers,3a and docking simulations of the compounds into the H3 receptor model were performed with ligand and receptor flexibility. We constructed a homology model based on the conformationally restricted analogue ent-2b, which is the most potent H3 receptor ligand in this series of the cyclopropanebased conformationally restricted analogues. Using this model, we performed docking simulation with a series of the cyclopropane-based conformationally restricted H3-receptor ligands having stereochemical diversity (16 compounds) synthesized previously5c,12 and also with the four cyclopropylic strain-based conformationally restricted analogues 2a, 2b, ent2a, and ent-2b synthesized in this study. Correlations between the calculated binding score and the pki were examined. As a result, as shown in Figure 9, a reliable correlation (R2 = 0.41) between binding score and pki was obtained. Consequently, the ent-2b-bound H3 receptor model was also used for further studies. In order to understand the binding modes of the newly synthesized cyclopropylic strain-based conformationally restricted analogues to the H3 receptor, docking simulations of 2a and 2b and their enantiomers ent-2b and ent-2b were carried out by using the ent-2b-bound model. Figure 10 shows the proposed binding modes of the potent H3 antagonists 2a and ent-2b to the homology model of the H3 receptor obtained by the simulation. These compounds are accommodated in the active site concavity formed by TM2, TM3, TM5, TM6, and TM7. The H3 receptor-binding conformations of 2a and ent-2b are the syn-form and the anti-form, respectively, as shown in Figures 10 and 11, which are in accord with the stable forms proposed by their conformational analysis by MR and the calculations described above. In the obtained binding models shown in Figure 10, the H of the imidazole ring of both compounds likewise serves as a hydrogen donor and forms a hydrogen bond with an oxygen atom of the side chain carboxyl group of Glu206. Furthermore, the protonated amine in both 2a and ent-2b forms a similar salt bridge with Asp114 in TM3, and the 4-chlorobenzyl group in 2a and ent-2b is observed to make a π-π interaction with the indole ring of Trp110 in TM3. Thus, the special positioning of the imidazole moiety and the 4-chlorobenzylamino moiety in 2a and ent-2b, which are likely to be essential for their H3 receptor binding, is analogous in the active site, and therefore 2a and ent-2b may have a common pharmacophore. As shown in Figure 11a, the imidazole and 4-chlorobenzylamino moieties of 2a and ent-2b can be superimposed, where the two cyclopropane rings orient oppositely, i.e., up in 2a and down in ent-2b, respectively. This would explain why 2a and ent-2b have a similar potent antagonistic effect on the H3 receptor, even though the two compounds are conformationally restricted Figure 9. Plot of binding score calculated by Glide extraprecision (XP) based on ent-2b-bound H3 receptor model versus experimental binding affinity pki for 20 conformational restriction analogues. The coefficient of determination, R2, between binding score and pki was 0.41 for 20 conformational restriction analogues. Figure 10. Proposed models for 2a (a) and ent-2b (b) binding to the homology model of the H3 receptor from docking simulation. Receptor residues around the compounds within 4 A are shown in line representation. Carbon atoms of 2a and ent-2b are shown in magenta and green, respectively. All nonpolar hydrogen atoms of receptor residues are omitted for clarify. Hydrogen bonding and salt bridge to side chain carboxyl group of Glu206 and Asp114 are depicted by red dots. 26

3590 Journal of Medicinal Chemistry, 2010, Vol. 53, o. 9 Watanabe et al. Figure 11. Superimposition of the structures of 2a (magenta) and ent-2b (green) binding to the homology model of the H 3 receptor (a). Threefeature pharmacophore model generated for 2a and ent-2b using MOE: hydrogen bond acceptor/donor (magenta feature), hydrogen bond donor/cationic atom (blue feature), and aromatic ring center/hydrophobic region (green feature) (b). Known H 3 receptor ligands are mapped onto the pharmacophore model obtained from 2a and ent-2b (c). Figure 12. Comparison of conformational changes between the stable forms shown in Figure 6 and the bioactive forms proposed by the H 3 receptor-bound model. Shown are the two stable conformations and the bioactive conformations for 1 (a), the stable and the bioactive conformations for 2a (b) and 2b (c), and superimposition of the bioactive conformations of 1 and 2a (d). Carbon atoms of stable conformations for all compounds are shown in gray, and carbon atoms of the bioactive conformations for 1, 2a, and 2b are shown in blue, magenta, and yellow, respectively. All compounds are aligned on the cyclopropane ring. differently, i.e., the syn-form and the anti-form, respectively, by the cyclopropylic strain. Figure 11b shows a common pharmacophore model for 2a and ent-2b, and the model effectively fitted in known H 3 receptor ligands, 1 asshowninfigure11c. We next examined the possible conformational differences between the stable form and the bioactive form of the compounds, which could significantly affect the binding affinity, based on the proposed receptor-bound model. In Figure 12, the stable conformations of 1 and its conformationally restricted analogues 2a and 2b based on MR and calculation analysis are superimposed on their receptorbound (bioactive) conformations by the receptor modeling simulations. As shown in Figure 12a, the bioactive conformation of 1 is in accord with the syn-form of the two stable syn- and anti-conformations. Figure 12b shows that the stable syn-form of 2a is almost identical with the bioactive conformation in 2a, which would make it highly potent. On the other hand, in 2b, the bioactive conformation is the synform similar to 1 and 2a, while 2b itself is stable in the antiform (Figure 12c). Thus, because of the entropic cost for the conformational change from the stable anti-form into the bioactive syn-form in its binding to the H 3 receptor, the binding affinity of 2b for the H 3 receptor is significantly decreased. In Figure 12d, the receptor-bound conformations of 1 and 2a were superimposed, showing that their bioactive conformations are the same. These results suggested that the cyclopropylic strain-based conformational restriction worked effectively as expected. As described, we identified the potent H 3 receptor antagonists 1 and ent-1 by the stereochemical diversity-oriented conformational restriction method with chiral cyclopropane units as shown in Figure 2. Their bioactive conformations and a pharmacophore model were further elaborated by the cyclopropylic strain-based conformational restriction method. These results showed that the combinational use of the stereochemical diversity-oriented and the cyclopropylic strain-based conformational restriction methods seems to be an effective strategy for developing significantly active compounds and also for identifying their bioactive conformations, especially in cases where the structural data of the target biomolecule are lacking or poorly documented. In these studies, simulations with homology modeling of the target biomolecule can be effective. This is due to the fact that a series of cyclopropane analogues and also cyclopropane strain-based conformationally restricted analogues are suitable for validating the homology models, since these consist of compounds having diversity not only in their conformation, i.e., three-dimensional structure, but also in their binding affinity for the target. Conclusion In order to clarify the bioactive conformation of the previously developed H 3 receptor antagonists 1 and ent-1, the 1 0 -ethyl-substituted derivatives 2a and 2b and their enantiomers 27

Article Journal of Medicinal Chemistry, 2010, Vol. 53, o. 9 3591 ent-2a and ent-2b were designed as side chain conformationrestricted analogues of 1 and ent-1, based on the cyclopropylic strain, and were synthesized from the versatile chiral cyclopropane units. Conformational analysis, pharmacological evaluation, and docking simulation of the compounds showed that the bioactive conformations of 1 and ent-1 seem to be the syn-form and the anti-form, respectively. On the basis of these results, a common pharmacophore for the compounds was obtained. These results suggest that the cyclopropylic strainbased strategy can be effectively used for precise conformational restriction of the side chain moiety and bioactive conformation analysis of cyclopropane compounds. Experimental Section Chemical shifts are reported in ppm downfield from tetramethylsilane. Thin-layer chromatography was done on Merck coated plate 60F 254. Silica gel chromatography was done on silica gel 5715 (Merck) or H silica gel (Chromatorex, Fuji Silysia Chemical Ltd.). Reactions were carried out under an argon atmosphere. Estimated purity of all of the final compounds by combustional analysis was always at least 95%. (1R,2R)-2-(1-Oxopropyl)-1-(1-triphenylmethyl-1H-imidazol- 4-yl)cyclopropane (6). To a solution of 5 5a (238 mg, 0.629 mmol) in THF (5.0 ml) was added EtMgBr (0.91 M in THF, 830 μl, 0.755 mmol) at 0 C, and the mixture was stirred at the same temperature for 1 h. After addition of aqueous saturated H 4 Cl, the solvent was evaporated, and the residue was partitioned between AcOEt and aqueous H 4 Cl. The organic layer was washed with brine, dried (a 2 SO 4 ), and evaporated. To a solution of the residue in CH 2 Cl 2 (5 ml) was added Dess-Martin periodinane (320 mg, 0.755 mmol) at room temperature, and the mixture was stirred at the same temperature for 1 h. To the reaction mixture was added a mixture of aqueous saturated ahco 3 and aqueous saturated a 2 S 2 O 3 (3/1, 12 ml) at room temperature, and the resulting mixture was vigorously stirred at the same temperature for 10 min. The resulting solution was extracted with AcOEt, and the organic layer was washed with aqueous saturated ahco 3, brine, dried (a 2 SO 4 ), and evaporated. The residue was purified by silica gel column chromatography (20-33% AcOEt in hexane) to give 6 (239 mg, 93%) as a white amorphous solid: [R] 19 D -217.3 (c 0.86, CHCl 3 ); 1 H MR (400 MHz, CDCl 3 ) δ 1.07 (3 H, t, J=7.3 Hz, CH 3 CH 2 -), 1.44 (1 H, m, H-3a), 1.50 (1 H, m, H-3b), 2.38 (2 H, m, H-1 and H-2), 2.61 (2 H, q, J=7.3 Hz, CH 3 CH 2 -), 6.64 (1 H, d, J=1.2 Hz, imidazolyl), 7.11-7.14 (6 H, m, aromatic), 7.28 (1 H, s, imidazolyl), 7.32-7.34 (9 H, m, aromatic); 13 C MR (100 MHz, CDCl 3 ) δ 7.96, 18.0, 22.5, 30.1, 37.0, 75.2, 118.0, 127.9, 129.7, 138.4, 138.4, 139.9, 142.2, 209.9; LRMS (EI) m/z 406 (M þ ); HRMS (EI) calcd for C 28 H 26 2 O 406.2045, found 406.2050 (M þ ). Anal. (C 28 H 26 2 O) C, H,. (1R,2R)-2-(1-Formylpropyl)-1-(1-triphenylmethyl-1H-imidazol-4-yl)cyclopropane (7). To a suspension of MeOCH 2 PPh 3 Cl (608 mg, 1.77 mmol) in THF (5.0 ml) was added a(si- (CH 3 ) 3 ) 2 ( 1.9 M in THF, 800 μl, 1.52 mmol) at 0 C, and the mixture was stirred at the same temperature for 15 min. To the resulting solution was added a solution of 6 (206 mg, 0.507 mmol) in CH 2 Cl 2 (2.0 ml) at 0 C, and the reaction mixture was stirred at the same temperature for 2 h. After addition of aqueous saturated H 4 Cl, the solvent was evaporated, and the residue was partitioned between AcOEt and aqueous H 4 Cl. The organic layer was washed with brine, dried (a 2 SO 4 ), and evaporated. The residue was purified by silica gel column chromatography (30% AcOEt in hexane) to give the enol ether product (151 mg) as a light-yellow solid. To a solution of the product in THF (10 ml) was added aqueous HCl (12 M, 5.0 ml), and the mixture was vigorously stirred at room temperature for 10 s. Immediately, the mixture was poured into aqueous saturated ahco 3 (100 ml). Then the resulting solution was extracted with AcOEt. The organic layer was washed with aqueous saturated ahco 3, brine, dried (a 2 SO 4 ), and evaporated. The residue was purified by silica gel column chromatography (25-40% AcOEt in hexane) to give 7 (diastereomixture, 138 mg, 65%) as a yellow amorphous solid: HRMS (EI) calcd for C 29 H 28 2 O 420.2202, found 420.2200 (M þ ). (1R,2S)-2-[1-Ethyl-2-(4-chlorobenzylamino)ethyl]-1-(1H-imidazol-4-yl)cyclopropane (2a/2b). To a solution of 7 (112 mg, 0.266 mmol) and 4-chlorobenzylamine (35 μl, 0.28 mmol) in MeOH/AcOH (10/1, 2.2 ml) was added 2-picolineborane (30 mg, 0.28 mmol) at room temperature, and the mixture was stirred at the same temperature for 12 h. After evaporation of the solvent, a solution of the residue in aqueous HCl (4 M, 4.0 ml) was stirred at 0 C for 20 min, and then the mixture was neutralized with a 2 CO 3. The mixture was partitioned between CH 2 Cl 2 and aqueous saturated ahco 3, and the organic layer was washed with brine, dried (a 2 SO 4 ), and evaporated. The residue was purified by neutral silica gel column chromatography (0-10% MeOH in CHCl 3 ) to give amine product (diastereomixture, 108 mg) as a light-yellow amorphous solid. To a solution of amine (108 mg) in EtOH (1.0 ml) was added aqueous HCl (2 M, 1.0 ml), and the resulting solution was stirred at 78 C for 2 h, and then the solvent was evaporated. The residue was partitioned between aqueous HCl (1 M) and CH 2 - Cl 2, and the aqueousueous layer was neutralized with aqueous aoh (2 M). The resulting solution was extracted with Et 2 O, and the organic layer was washed with H 2 O, brine, dried (a 2 SO 4 ), and evaporated. The residue was purified by H silica gel column chromatography (0-5% MeOH in CHCl 3 )to give the diastereomixture of 2a and 2b (56 mg, 70%) as a colorless amorphous solid: HRMS (EI) calcd for C 17 H 22 Cl 3 303.1502, found 303.1500 (M þ ). (1R,2S)-2-[(1R)-1-Ethyl-2-(4-chlorobenzylamino)ethyl]-1-(1Himidazol-4-yl)cyclopropane (2a) and (1R,2S)-2-[(1S)-1-Ethyl- 2-(4-chlorobenzylamino)ethyl]-1-(1H-imidazol-4-yl)cyclopropane (2b). A solution of the diastereomixture of 2a and 2b (56 mg, 0.15 mmol), Et 3 (83 μl, 0.60 mmol), DMAP (1.8 mg, 0.015 mmol), and (Boc) 2 O (130 mg, 0.60 mmol) in MeOH (1 ml) was stirred at room temperature for 16 h. After evaporation of the solvent, the residue was partitioned between AcOEt and H 2 O, and the organic layer was washed with brine, dried (a 2 SO 4 ), and evaporated. The residue was separated by HPLC (28% AcOEt in hexane, 13 ml/min, room temperature, 253 nm) with Mightysil Si 60 (0.25 cm 20 cm, Kanto Chemical Co.) to give 8a (24 mg, a colorless amorphous solid) and 8b (26 mg, a colorless amorphous solid). Each compound was dissolved in EtOH (1.5 ml)/aqueous HCl (4 M, 0.5 ml), and the mixture was stirred at 78 C for 2 h. After the mixture was concentrated and dried in vacuo, the residue was purified by H silica gel column chromatography (0-10% MeOH in CHCl 3 ) to give 2a (12 mg, 29% for three steps, a colorless amorphous solid) or 2b (14 mg, 33% for three steps, a colorless amorphous solid) as a free amine. 2a: 1 H MR (400 MHz, CDCl 3 ) δ 0.76 (1 H, m, H-3a), 0.87-0.96 (6 H, m, H-3b and H-1 0 and H-2 and CH 3 CH 2 -), 1.45 (1 H, m, CH 3 CH 2 -), 1.54 (1 H, m, CH 3 CH 2 -), 1.64 (1 H, m, H-1), 2.65 (2 H, dd, J=5.0, 12.0 Hz, H-2 0 ), 3.77 (2 H, s, -CH 2 Ph), 6.63 (1 H, s, imidazolyl), 7.27 (4 H, dd, J=8.0, 8.6 Hz, aromatic), 7.48 (1 H, s, imidazolyl); 13 C MR (100 MHz, CDCl 3 ) δ 12.0, 13.1, 14.6, 24.7, 26.3, 44.8, 53.8, 53.8, 128.7, 129.6, 132.8, 134.4, 139.2; HRMS (EI) calcd for C 17 H 22 Cl 3 303.1502, found 303.1501 (M þ ). The free amine 2a (12 mg) was dissolved in aqueous HCl (4 M), and the solvent was then evaporated. The residue was triturated with Et 2 O to give 2a dihydrochloride (15 mg, a white amorphous solid): [R] 21 D -20.5 (c 0.81, MeOH); 1 HMR(400 MHz, D 2 O) δ 0.89 (3 H, t, J=7.5 Hz, CH 3 CH 2 -), 0.93 (1 H, m, H- 3a), 0.98-1.06 (2 H, m, H-2 and H-3b), 1.22 (1 H, m, H-1 0 ), 1.54 (2 H, m, CH 3 CH 2 -), 1.88 (1 H, m, H-1), 3.11 (2 H, dd, J=1.8, 7.1 Hz, H-2 0 ), 4.20 (1 H, d, J=13.6 Hz, -CH 2 Ph), 4.26 (1 H, d, J=13.6 Hz, 28

3592 Journal of Medicinal Chemistry, 2010, Vol. 53, o. 9 Watanabe et al. -CH 2 Ph), 7.10 (1 H, s, imidazolyl), 7.46 (4 H, dd, J=8.6, 8.6 Hz, aromatic), 8.44 (1 H, d, J=1.4 Hz, imidazolyl); LRMS (EI) m/z 303 ((M - 2HCl) þ ). Anal. (C 17 H 24 Cl 3 3 )C,H,. 2b: 1 H MR (400 MHz, CDCl 3 ) δ 0.72 (1 H, m, H-3a), 0.81 (1 H, m, H-3b), 0.93-0.99 (5 H, m, H-1 0 and H-2 and CH 3 CH 2 -), 1.47 (2 H, m, CH 3 CH 2 -), 1.64 (1 H, m, H-1), 2.64 (1 H, dd, J= 7.4, 11.4 Hz, H-2 0 ), 2.75 (1 H, dd, J=5.7, 11.4 Hz, H-2 0 ), 3.74 (1 H, d, J=13.3 Hz, -CH 2 Ph), 3.78 (1 H, d, J=13.3 Hz, -CH 2 Ph), 6.67 (1 H, s, imidazolyl), 7.22 (2 H, d, J=8.6 Hz, aromatic), 7.26 (2 H, d, J = 8.6 Hz, aromatic), 7.44 (1 H, s, imidazolyl); 13 C MR (100 MHz, CDCl 3 ) δ 11.8, 12.5, 14.6, 24.9, 26.4, 44.8, 53.8, 54.4, 128.8, 129.7, 132.9, 134.5, 139.1; LRMS (EI) m/z 303 (M þ ); HRMS (EI) calcd for C 17 H 22 Cl 3 303.1502, found 303.1509 (M þ ). The free amine 2b (14 mg) was dissolved in aqueous HCl (4 M), and the solvent was then evaporated. The residue was triturated with Et 2 O to give 2b dihydrochloride (17 mg, a white amorphous solid): [R] 21 D -64.5 (c 1.23, MeOH); 1 H MR (400 MHz, D 2 O) δ 0.92 (3 H, t, J=7.2 Hz, CH 3 CH 2 -), 1.05 (2 H, m, H-3), 1.09 (1 H, m, H-2), 1.23 (1 H, m, H-1 0 ), 1.43-1.60 (2 H, m, CH 3 CH 2 -), 1.74 (1 H, m, H-1), 3.12 (1 H, dd, J=7.4, 13.1 Hz, H-2a 0 ), 3.19 (1 H, dd, J=5.0, 13.1 Hz, H-2b 0 ), 4.13 (1 H, d, J= 13.6 Hz, -CH 2 Ph), 4.26 (1 H, d, J=13.6 Hz, -CH 2 Ph), 6.97 (1 H, s, imidazolyl), 7.39 (4 H, s, aromatic), 8.46 (1 H, d, J=1.4 Hz, imidazolyl); LRMS (EI) m/z 303 ((M-2HCl) þ ). Anal. (C 17 H 24 - Cl 3 3 )C,H,. (1S,2S)-2-(1-Oxopropyl)-1-(1-triphenylmethyl-1H-imidazol- 4-yl)cyclopropane (ent-6). Compound ent-6 (173 mg, 85%, a white amorphous solid) was prepared from ent-5 (190 mg, 0.50 mmol) as described for the preparation of 6: [R] 20 D þ232.1 (c 1.03, CHCl 3 ); HRMS (EI) calcd for C 28 H 26 2 O 406.2045, found 406.2044 (M þ ). Anal. (C 28 H 26 2 O) C, H,. (1S,2S)-2-(1-Formylpropyl)-1-(1-triphenylmethyl-1H-imidazol-4-yl)cyclopropane (ent-7). Compound ent-7 (94 mg, 60%, a white amorphous solid) was prepared from ent-6 (152 mg, 0.37 mmol) as described for the preparation of 7: HRMS (EI) calcd for C 29 H 28 2 O 420.2202, found 420.2220 (M þ ). (1S,2R)-2-[(1S)-1-Ethyl-2-(4-chlorobenzylamino)ethyl]-1-(1Himidazol-4-yl)cyclopropane (ent-2a). Compound ent-2a (10 mg, 30%, a white amorphous solid) was prepared from ent-8 (35 mg, 0.11 mmol) as described for the preparation of 2a: HRMS (EI) calcd for C 17 H 22 Cl 3 303.1502, found 303.1500 (M þ ). The free amine ent-2a (10 mg) was dissolved in aqueous HCl (4 M), and the solvent was then evaporated. The residue was triturated with Et 2 O to give ent-2a dihydrochloride (13 mg, a white amorphous solid): [R] 21 D þ19.9 (c 0.62, MeOH); LRMS (EI) m/z 303 ((M-2HCl) þ ). Anal. (C 17 H 24 Cl 3 3 )C,H,. (1S,2R)-2-[(1R)-1-Ethyl-2-(4-chlorobenzylamino)ethyl]-1-(1Himidazol-4-yl)cyclopropane (ent-2b). Compound ent-2b (10 mg, 30%, a white amorphous solid) was prepared from ent-8 (35 mg, 0.11 mmol) as described for the preparation of 2b: LRMS (EI) m/z 303 (M þ ); HRMS (EI) calcd for C 17 H 22 Cl 3 303.1502, found 303.1500 (M þ ). The free amine ent-2b (10 mg) was dissolved in aqueous HCl (4 M), and the solvent was then evaporated. The residue was triturated with Et 2 O to give ent- 2b dihydrochloride (13 mg, a white amorphous solid): [R] 21 D þ63.2 (c 1.11, MeOH); LRMS (EI) m/z 303 ((M-2HCl) þ ). Anal. (C 17 H 24 Cl 3 3 )C,H,. Binding Assay with Human Histamine Receptors. The assay was performed according to the method described previously. 5c Luciferase Reporter Gene Assay. The assay was performed according to the method described previously. 5b Homology Modeling of the H 3 Receptor. The histamine H 3 receptor sequence was aligned with the human β 2 -adrenergic receptor sequence 3a and 40 representative sequences of class A rhodopsin-like amine families (G-protein-coupled receptors data bank (GPCRDB): http://www.gpcr.org/7tm/) using the CLUS- TAL W (version 1.8) multiple alignment program. 15 The alignment was refined manually on the basis of the compatibility of the amino acid position with the corresponding structure of the β 2 -adrenergic receptor. A three-dimensional model of H 3 receptor was constructed using a homology modeling approach incorporatedintheprogramsegmod 16 of GeneMine. 17 Because it is very long and predicted to be disordered, the third intracellular loop was truncated by the five residues leading out of the helix V and the five residues leading into helix VI. The second extracellular loop was modeled without aligning them with those of the β 2 -adrenergic receptor. Docking Simulation. Initial coordinates of all compounds were constructed using the Molecular Builder module in Maestro (Schr odinger LLC.). Energy minimization of all compounds was performed using the optimized potentials for liquid simulations-all-atom (OPLS-AA) force field in the LigPrep in the Maestro (Schr odinger LLC.). The homology model of H 3 receptor was refined for docking simulations using the Protein Preparation Wizard Script within Maestro. This protein preparation procedure involves optimization of contacts by changing hydroxyl group orientations, flipping of Asn and Gln side chains, and selecting His tautomeric states, followed by constrained energy refinement using the OPLS-AA force field. Docking of the compounds into the H 3 receptor model utilized three main steps that take into account several levels of structural flexibility and scoring criteria: (1) molecular modeling of compound bound H 3 receptor model by docking the ent-2b molecule, considering both ligand and receptor flexibility, (2) rigid receptor docking of 20 cyclopropane-based conformational restriction analogues 5c into the active site of ent-2b compound bound H 3 receptor model from the previous step, (3) rescoring according to the calculated binding score by Glide extraprecision (XP) score (Schr odinger LLC). Following are the details of each step. In order to account for both compound and receptor flexibility in the first step, the Glide Induced Fit Docking (IFD) protocol (Schr odinger LLC.) was utilized, followed by iteratively combining rigid receptor docking (Glide) and protein remodeling by side chain searching and minimization (Prime) techniques. Hydrogenbonding constraints between the side chain COO - group of Asp114 and Glu206 were introduced because this hydrogenbonding formation is highly conserved in almost all known complexes of histamine receptor subfamily bound to histamine and to a wide variety of inhibitors. In the protein remodeling stage, all residues within a 14.0 A radius of each initial docked compound were refined using Prime. Compound was then redocked into the refined receptor structure using Glide in the standard precision (SP) mode. All of the docked structures were then ranked according to GlideScore. After modeling of the compound-h 3 receptor complex using the IFD protocol, grid generation and rigid receptor docking of the 20 cyclopropanebased conformational restriction analogues using Glide (SP mode) were carried out, using the hydrogen bonding constraint to connect the side chain COO - group of Asp114 and Glu206. The best orientation for each docked compound was rescored according to its binding score, which was calculated using the Glide XP Score (Schr odinger LLC). Acknowledgment. This investigation was supported by a Grant-in-Aids for Scientific Research (Grant 21390028) from the Japan Society for the Promotion of Science. We are grateful to Sanyo Fine Co., Ltd. for the gift of the chiral epichlorohydrins. Supporting Information Available: Synthesis of PGME amides ent-9a, ent-9b, ent-10a, and ent-10b; the experimental details by which the determination of the 1 0 -configurations of ent-2a and ent-2b by the PGME method was effected; the structures and pk i values of the H 3 receptor antagonists used for the docking simulations; and elemental analysis data of the final compounds. This material is available free of charge via the Internet at http://pubs.acs.org. 29

Article Journal of Medicinal Chemistry, 2010, Vol. 53, o. 9 3593 References (1) (a) Arrang, J.-M.; Garbarg, M.; Schwartz, J.-C. Auto-inhibition of brain histamine release mediated by a novel class (H 3 ) of histamine receptor. ature 1983, 302, 832 837. (b) Leurs, R., Timmerman, H., Eds. The Histamine H3 Receptor, a Target for ew Drugs; Elsevier: Amsterdam, 1998. (c) Leurs, R.; Bakker, R. A.; Timmerman, H.; de Esch, I. J. P. The histamine H 3 receptor: from gene cloning to H 3 receptor drugs. at. Rev. Drug Discovery 2005, 4, 107 120. (d) Celanire, S.; Wijtmans, M.; Talaga, P.; Leurs, R.; de Esch, I. J. P. Histamine H 3 receptor antagonists reach for the clinic. Drug Discovery Today 2005, 10, 1613 1627. (e) Esbenshade, T. A.; Fox, G. B.; Cowart, M. D. Histamine H 3 receptor antagonists: preclinical promise for treating obesity and cognitive disorders. Mol. Interventions 2006, 6, 77 88. (f) Ali, S. M.; Tedford, C. E.; Gregory, R.; Handley, M. K.; Yates, S. L.; Hirth, W. W.; Phillips, J. G. Design, synthesis, and structure-activity relationships of acetylene-based histamine H 3 receptor antagonists. J. Med. Chem. 1999, 42, 903 909. (2) (a) Klabunde, T.; Hessler, G. Drug design strategies for targeting G-protein-coupled receptors. ChemBioChem 2002, 3, 928 944. (b) Palczewski, K.; Kumasaka, T.; Hori, T.; Behnke, C. A.; Motoshima, H.; Fox, B. A.; Le Trong, I.; Teller, D. C.; Okada, T.; Stenkamp, R. E.; Yamamoto, M.; Miyano, M. Crystal structure of rhodopsin: a G proteincoupled receptor. Science 2000, 289, 739 745. (c) Sarramegna, V.; Talmont, F.; Demange, P.; Milon, A. Heterologous expression of G-protein-coupled receptors: comparison of expression systems from the standpoint of large-scale production and purification. Cell. Mol. Life Sci. 2003, 60, 1529 1546. (d) Schlyer, S.; Horuk, R. I want a new drug: G-protein-coupled receptors in drug development. Drug Discovery Today 2006, 11, 481 493 and references therein. (3) (a) Cherezov, V.; Rosenbaum, D. M.; Hanson, M. A.; Rasmussen, S. G.; Thian, F. S.; Kobilka, T. S.; Choi, H. J.; Kuhn, P.; Weis, W. I.; Kobilka, B. K.; Stevens, R. C. High-resolution crystal structure of an engineered human β 2 -adrenergic G protein-coupled receptor. Science 2007, 318, 1258 1265. (b) Rosenbaum, D. M.; Cherezov, V.; Hanson, M. A.; Rasmussen, S. G.; Thian, F. S.; Kobilka, T. S.; Choi, H. J.; Yao, X. J.; Weis, W. I.; Stevens, R. C.; Kobilka, B. K. GPCR engineering yields high-resolution structural insights into β 2 -adrenergic receptor function. Science 2007, 318, 1266 1273. (c) Rasmussen, S. G.; Choi, H. J.; Rosenbaum, D. M.; Kobilka, T. S.; Thian, F. S.; Edwards, P. C.; Burghammer, M.; Ratnala, V. R.; Sanishvili, R.; Fischetti, R. F.; Schertler, G. F.; Weis, W. I.; Kobilka, B. K. Crystal structure of the human β 2 adrenergic G-protein-coupled receptor. ature 2007, 450, 383 387. (4) (a) Kier, L. B. Molecular orbital calculations of the preferred conformations of histamine and a theory on its dual activity. J. Med. Chem. 1968, 11, 441 445. (b) Silverman, R. B. The Organic Chemistry of Drug Design and Drug Action; Academic Press; San Diego, CA, 2004. (5) (a) Kazuta, Y.; Matsuda, A.; Shuto, S. Development of versatile cis- and trans-dicarbon-substituted chiral cyclopropane units: synthesis of (1S,2R)- and (1R,2R)-2-aminomethyl-1-(1H-imidazol-4-yl)cyclopropanes and their enantiomers as conformationally restricted analogs of histamine. J. Org. Chem. 2002, 67, 1669 1677. (b) Kazuta, Y.; Hirano, K.; atsume, K.; Yamada, S.; Kimura, R.; Matsumoto, S.; Furuichi, K.; Matsuda, A.; Shuto, S. (1S,2S)-2-(2- Aminoethyl)-1-(1H-imidazol-4-yl)cyclopropane, a highly selective agonist for the histamine H 3 receptor, having a cis-cyclopropane structure. J. Med. Chem. 2003, 46, 1980 1988. (c) Watanabe, M.; Kazuta, Y.; Hayashi, H.; Yamada, S.; Matsuda, A.; Shuto, S. The stereochemical diversity-oriented conformational restriction strategy. Development of potent histamine H 3 and/or H 4 receptor antagonists with an imidazolylcyclopropane structure. J. Med. Chem. 2006, 49, 5787 5796. (6) (a) Shuto, S.; Ono, S.; Hase, Y.; Kamiyama,.; Takada, H.; Yamashita, K.; Matsuda, A. Conformational restriction by repulsion between adjacent substituents of a cyclopropane ring: design and enantioselective synthesis of 1-phenyl-2-(1-aminoalkyl)-,diethylcyclopropanecarboxamides as potent MDA receptor antagonists. J. Org. Chem. 1996, 61, 915 923. (b) Shuto, S.; Ono, S.; Hase, Y.; Ueno, Y.; oguchi, T.; Yoshii, K.; Matsuda, A. Synthesis and biological activity of conformationally restricted analogs of milnacipran: (1S,1R)-1-phenyl-2-[(S)-1-aminopropyl]-,-diethylcyclopropanecarboxamide, an efficient noncompetitive -methyl-d-aspartic acid receptor antagonist. J. Med. Chem. 1996, 39, 4844 4852. (c) Shuto, S.; Ono, S.; Imoto, H.; Yoshii, K.; Matsuda, A. Synthesis and biological activity of conformationally restricted analogs of milnacipran: (1S,2R)-1-phenyl-2-[(R)-1-amino-2-propynyl]-,-diethylcyclopropanecarboxamide is a novel class of MDA receptor channel blocker. J. Med. Chem. 1998, 41, 3507 3514. (d) Ono, S.; Ogawa, K.; Yamashita, K.; Yamamoto, T.; Kazuta, Y.; Matsuda, A.; Shuto, S. Conformational analysis of the MDA receptor antagonist (1S,2R)-1- phenyl-2-[(s)-1-aminopropyl]-,-diethylcyclopropanecarboxamide (PPDC) designed by a novel conformational restriction method based on the structural feature of cyclopropane ring. Chem. Pharm. Bull. 2002, 50, 966 968 and references therein. (7) Ohmori, Y.; Yamashita, A.; Tsujita, R.; Yamamoto, T.; Taniuchi, K.; Matsuda, A.; Shuto, S. A method for designing conformationally restricted analogs based on allylic strain: synthesis of a novel class of noncompetitive MDA receptor antagonists having the acrylamide structure. J. Med. Chem. 2003, 46, 5326 5333. (8) For examples, see the following: (a) Armstrong, P. D.; Cannon, J. G.; Long, J. P. Conformationally rigid analogs of acetylcholine. ature 1698, 220, 65 66. (b) Shimamoto, K.; Ofune, Y. Syntheses and conformational analyses of glutamate analogs: 2-(2-carboxy-3-substituted-cyclopropyl)glycines as useful probes for excitatory amino acid receptors. J. Med. Chem. 1996, 39, 407 423. (c) Stammer, S. H. Cyclopropane amino acids; 2,3- and 3,4-methanoamino acids. Tetrahedron 1990, 46, 2231 2254. (d) Martin, S. F.; Dwyer, M. P.; Hartmann, B.; Knight, K. S. Cyclopropane-derived peptidomimetics. Design, synthesis, and evaluation of novel enkephalin analogs. J. Org. Chem. 2000, 65, 1305 1318. (e) Sekiyama, T.; Hatsuya, S.; Tanaka, Y.; Uchiyama, M.; Ono,.; Iwayama, S.; Oikawa, M.; Suzuki, K.; Okunishi, M.; Tsuji, T. Synthesis and antiviral activity of novel acyclic nucleosides: discovery of a cyclopropyl nucleoside with potent inhibitory activity against herpes viruses. J. Med. Chem. 1998, 41, 1284 1298. (9) (a) Wong, H.. C.; Hon, M.-Y.; Tse, C.-Y.; Yip, Y.-C. Use of cyclopropanes and their derivatives in organic synthesis. Chem. Rev. 1989, 89, 165 198. (b) Singh, V. K.; DattaGupta, A.; Sekar, G. Catalytic enantioselective cyclopropanation of olefins using carbenoid chemistry. Synthesis 1997, 137 149. (c) Doyle, M. P.; Protopopova, M.. ew aspects of catalytic asymmetric cyclopropanation. Tetrahedron 1998, 54, 7919 7946. (d) Cossy, J.; Blanchard,.; Meyer, C. Stereoselective synthesis of cyclopropanes bearing adjacent stereocenters. Synthesis 1999, 1063 1075. (e) Small Ring Compounds in Organic Synthesis VI; de Meijere, A., et al., Eds.; Topics in Current Chemistry 207; Springer: Berlin, 1999. (f) Lebel, H.; Marcoux, J.-F.; Molinaro, C.; Charette., A. B. Stereoselective cyclopropanation reactions. Chem. Rev. 2003, 103, 977 1050. (g) Garcia, P.; Diez, D.; Anton, A. B.; Garrido,. M.; Marcos, I. S.; Basabe, P.; Urones, J. G. Stereoselective synthesis of cyclopropanols. Mini-Rev. Org. Chem. 2006, 3, 291 314. (h) Muller, P.; Allenbach, Y. F.; Chappellet, S.; Ghanem, A. Asymmetric cyclopropanations and cycloadditions of dioxocarbenes. Synthesis 2006, 10, 1689 1696. (10) Yabuuchi, T.; Kusumi, T. Phenylglycine methyl ester, a useful tool for absolute configuration determination of various chiral carboxylic acids. J. Org. Chem. 2000, 65, 397 404. (11) For detail of the determination of the 1 0 -configuration, see Supporting Information. (12) The structures and pk i values for H 3 receptor of these antagonists are summarized in Table S1 in Supporting Information. (13) (a) Rai, B. K.; Tawa, G. J.; Katz, A. H.; Humblet, C. Modeling G protein-coupled receptors for structure-based drug discovery using low-frequency normal modes for refinement of homology models: application to H 3 antagonists. Proteins 2010, 78, 457 473. (b) Dastmalchi, S.; Hamzeh-Mivehroud, M.; Ghafourian, T.; Hamzeiy, H. Molecular modeling of histamine H 3 receptor and QSAR studies on arylbenzofuran derived H 3 antagonists. J. Mol. Graph.ics Modell. 2008, 26, 834 844. (c) Axe, F. U.; Bembenek, S. D.; Szalma, S. Three-dimensional models of histamine H 3 receptor antagonist complexes and their pharmacophore. J. Mol. Graphics Model. 2006, 24, 456 464. (14) (a) Mobarec, J. C.; Sanchez, R.; Filizola, M. Modern homology modeling of G-protein coupled receptors: which structural template to use? J. Med. Chem. 2009, 52, 5207 5216. (b) Senderowitz, H.; Marantz, Y. G Protein-coupled receptors: target-based in silico screening. Curr. Pharm. Des. 2009, 15, 4049 4068. (c) de Graaf, C.; Rognan, D. Customizing G protein-coupled receptor models for structure-based virtual screening. Curr. Pharm. Des. 2009, 15, 4026 4048. (d) Costanzi, S. On the applicability of GPCR homology models to computer-aided drug discovery: a comparison between in silico and crystal structures of the β 2 -adrenergic receptor. J. Med. Chem. 2008, 51, 2907 2914. (15) Thompson, J. D.; Higgins, D. G.; Gibson, T. J. CLUSTAL W. Improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. ucleic Acids Res. 1994, 22, 4673 4680. (16) Levitt, M. Accurate modeling of protein conformation by automatic segment matching. J. Mol. Biol. 1992, 226, 507 533. (17) Lee, C.; Irizarry, K. The GeneMine system for genome/proteome annotation and collaborative data mining. IBM Syst. J. 2001, 40, 592 603. 30

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 pss-header will be provided by the publisher Density Functional Theory for Green Chemical Catalyst Supported on S- Terminated Ga(0001) Mami Yokoyama *,1, Shirou Tsukamoto **,2, and Akira Ishii 1 1 Tottori University, Koyama-minami 4-101, 680-8552, Japan 2 Anan ational College of Technology,265 Aoki Minobayashi, Anan, Japan Published online ZZZ (Dates will be provided by the publisher.) Keywords (Density functional calculation; Catalysis; Gallium nitride; Palladium * Corresponding author: e-mail ishii@damp.tottori-u.ac.jp, Phone: +81 857 31 5629, Fax: +81 857 31 5629 A novel function of nitried-based semiconductor is successfully developed for organic synthesis, in which palladium supported on the surface of S-terminated Ga(0001) serves as a unique green chemical catalyst. In this study we determined the structure of Pd-catalyst supported on Sterminated Ga(0001) surface by means 1 ITRODUCTIO Transition metal catalyzed reactions have played an important role in synthetic and process chemistry. The homogeneous catalyst surely causes effective reaction, however, a lot of efforts and energy are required for reusing the catalyst. From this viewpoint, development of the easily treatable heterogeneous catalyst is strongly desired for reducing the waste of expensive rare metal. Although the polymer supported catalyst has been developed, it cannot tolerate under severe conditions such as high temperature and it is restricted to use in organic solvents. Recently, palladium acetate (Pd(OAc)2) molecules immobilized on the S-terminated GaAs(001) has high catalytic activity and stability for Heck reaction [1]. However, the GaAs substrates including toxic As, it is not suitable for mass productions. To solve this problem, a new type catalyst was reported recently [2]. Which transition metals were supported on the S-terminated Ga(0001). For developing more effective catalysts, it is very important to know its detail structures. Therefore, in this paper, we try to determine the atomic structure of this catalyst, Pd on the S-terminated Ga(0001), by the density functional theory (DFT) calculation with the plane wave expansion and the pseudo potential methods using VASP [3]. Review copy not for distribution (pss-logo will be inserted here by the publisher) of the density functional theory (DFT) within a Local Density Approximation (LDA). The important role of S on the case of Ga substrate is to make the number of the valence electron to be close to 0, in contrast to the case of GaAs. Copyright line will be provided by the publisher 2 MODEL All our calculations are carried out using a program package VASP (Vienna Ab-initio Simulation Package). Calculations are based on the density functional theory (DFT), within a Local Density Approximation (LDA). In the experiment [2], the catalytic plate was prepared via three steps: (1) the S-termination, (2) the adsorption of Pd species (Pd adsorption), and (3) the aging (fig. 1), and then the catalytic activity of the plate was evaluated by Heck reaction. The stable structure of this catalyst is determined for developing effective catalysts. [3]. In addition, we calculated the number of the valence electron for Pd which is Pd catalyst supported on S-terminated Ga(0001), and GaAs(001) as well. In this study, we prepared for the two models of GaAs; As dimmer model and S dimmer model for comparison [4]. The calculation was performed using VASP with cutoff energy 400 ev. The used pseudo potential is LDA with PAW and d-electrons are included for Ga. The Ga(0001) surface is simulated by a repeated slab model, in which 10 atomic layer slabs are separated by vacuum region of 15. The positions of the atoms in the bottom 4 layers of the slab model are fixed. The backsides of the Ga slabs are terminated by the peudohydrogen technique. For the calculation about the number of the valence electron, we used barder analysis [5]. Copyright line will be provided by the publisher 31

2 Author, Author, and Author: Short title 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 3 RESULTS AD DISCUSSIO We calculated number of valence electron of Pd for Pd-S-Ga(0001) using stable structures[5]. First, we calculated the adsorption of Pd atom on the clean Ga(0001) surface as reference. The number of valence electron of Pd on clean Ga(0001) is +0.21 in fig.1. The number of valence electron of Pd on S-terminated Ga(0001) is -0.09 and the number of valence electron of S is +0.35 in this case. The number of valence electron of Pd and S for the S adsorbed Pd adatom on Ga(0001) are -0.06 and +0.16 In fig. 10. The structure of S-Pd-Ga is more stable than structure of Pd-S-Ga. A calculated atomic structure of Pd atoms on the S-terminated Ga(0001), indicating that the stable position of Pd atom is between S and Ga atoms of the S- terminated Ga(0001). The binding energy of Pd on Ga(0001) is stronger than that of S. It means that the S adatom is easier to be desorbed from the Ga(0001) substrate than Pd during catalysis reaction. Even if other study which is about Pd catalyst on S-terminated GaAs(001) have concluded that the role of S is to make bonding between GaAs substrate and Pd [?], we found that S does not play the same role in the case of Ga. Pd transfers slightly the valence electron to S. The number of valence electron of Pd is close to 0 when it use S atom as a role of bonding between Pd and Ga(0001) substrate. As well as the result of Pd on S-terminated Ga(0001),we calculated the number of valence electron of Pd on S-terminated GaAs(001). For the calculation, we prepared for the two models of GaAs; As dimmer model and S dimmer model for comparison. In this article, we will show the figure of them in fig.?. The calculation results of the valence electron of two models are show in fig?. In this case of As dimmer model, the valence electron of Pd is +0.08. In S dimmer model case, it is 0.23. The experimental data showed us that the valence of Pd is close to 0 rather than +2 and Pd exists as Pd(0) species or ano clusters. They share certain similarities in that the valence electron number of Pd is very close to 0. This result means that it is good agreement with experiment data [?]. We determined that the most likely structure of Pd catalyst on S-terminated GaAs(001) is As dimmer model rather than S dimmer. As well as the result of Pd on S-terminated GaAs(001), we have found that the number of valence electron of Pd on S-terminated Ga(0001) have the common point which is close to 0, also. Thus, the important role of S on the case of Ga substrate is to make the number of the valence electron to be close to 0. 4 COCLUSIO The density functional calculation is performed to determine the important role of S for Pd catalyst on S-terminated Ga(0001) surface. The valence electron numbers of Pd on the S terminated Ga(0001) are very close to 0. Similarly, that of Pd on the S-terminated GaAs(001) is also close to 0. In addition, we determined that the most likely structure of Pd catalyst on S terminated GaAs is As dimer model rather than S dimer. Acknowledgements This research was supported by Japan Science and Technology Agency (JST). References [1] M. Arisawa, M. Hamada, I. Takamiya, S. Tsukamoto, Y. A- rakawa, A. ishida, Adv. Synth.Catal. 348, 1063 (2006). [2]. agatoshi, M. Shimoda, T. Konishi and S. Tsukamoto, Applied Physics Express 2 (2009) 051002 [3] G. Kresse and J. Hafner, Phys. Rev. B 47, RC558(1993) [4] A. Ishii, H. Asano, M. Yokoyama, S. Tsukamoto, S. Shuto and M. Arisawa, Phys. Status Solidi C 7, o. 2, 359-361(2010) [5] M. Yokoyama, A. Ishii, S. Tsukamoto, S. Shuto and M. Arisawa, to be published Copyright line will be provided by the publisher 32

33

34

2 I. ITRODUCTIO Transition metal catalyzed reactions have played an important role in synthetic and process chemistry. The homogeneous catalyst surely causes effective reaction, however, a lot of efforts and energy are required for reusing the catalyst. From this viewpoint, development of the easily treatable heterogeneous catalyst is strongly desired for reducing the waste of expensive rare metal. Although the polymer supported catalyst has been developed, it cannot tolerate under severe conditions such as high temperature and it is restricted to use in organic solvents. Recently, palladium acetate (Pd(OAc)2) molecules immobilized on the S-terminated GaAs(001) has high catalytic activity and stability for Heck reaction [1]. However, the GaAs substrates including toxic As, it is not suitable for mass productions. To solve this problem, a new type catalyst was reported recently [2]. Which transition metals were supported on the S-terminated Ga(0001). For developing more effective catalysts, it is very important to know its detail structures. Therefore, in this paper, we try to determine the atomic structure of this catalyst, Pd on the S-terminated Ga(0001), by the density functional theory (DFT) calculation with the plane wave expansion and the pseudo potential methods using VASP [3]. II. MODEL All our calculations are carried out using a program package VASP (Vienna Ab-initio Simulation Package). Calculations are based on the density functional theory (DFT), within a Local Density Approximation (LDA). In the experiment [2], the catalytic plate was prepared via three steps: (1) the S-termination, (2) the adsorption of Pd species (Pd adsorption), and (3) the aging (fig. 1), and then the catalytic activity of the plate was evaluated by Heck reaction. Therefore, in this study, we consider 4 type of models which are S-Ga, Pd-Ga, Pd-S-Ga and S-Pd-Ga as shown In fig. 2, fig. 3, fig. 4 and fig. 5. In addition, we prepared for the two models of GaAs; As dimmer model and S dimmer model for comparison [4]. In this article, we will show the energy contour map of the Pd adatom on S-terminated Ga(0001) and the S adsorbed Pd adatom on Ga(0001) in fig. 6 and fig. 7 The calculation was performed using VASP with cutoff energy 400 ev. The used pseudopotential is LDA with PAW and d-electrons are included for Ga. The Ga(0001) surface is simulated by a repeated slab model, in which 10 atomic layer slabs are separated by vacuum region of 15. The positions of the atoms in the bottom 4 layers of the slab model are fixed. The backsides of the Ga slabs are terminated by the peudohydrogen technique. 35

3 FIG. 1: Preparative procedures for catalytic plates. FIG. 2: The calculation model for S-Ga. III. RESULTS AD DISCUSSIO First, we calculated the adsorption of S atom on the clean Ga(0001) surface and Pd atom on the clean Ga(0001) surface as references. The calculated results of the adsorption energy contour map of the S adatom on Ga and Pd adatom on Ga are shown In fig. 6 and fig. 9. The binding energy between S and Ga(0001) substrate is 2.95eV and the binding energy between Pd and Ga(0001) substrate is 3.19eV. The calculated migration barrier energies are also shown in the figure. The adsorption energy contour map for Pd adatom on the S-terminated Ga(0001) surface is fig. 6 and the adsorption energy contour map for S adsorbed Pd adatom on Ga(0001) surface is fig. 5. The FIG. 3: The calculation model for Pd-Ga. 36

4 FIG. 4: The calculation model for Pd-S-Ga. FIG. 5: The calculation model for S-Pd-Ga. binding energy for the Pd adatom on this surface is 3.04eV and the binding energy for the S adsorbed Pd adatom on Ga(000) is 3.13eV. The number of valence electron of Pd on clean Ga(0001) is +0.21. The number of valence electron of Pd on S-terminated Ga(0001) is -0.09 and the number of valence electron of S is +0.35 in this case. The number of valence electron of Pd and S for the S adsorbed Pd adatom on Ga(0001) are -0.06 and +0.16 In fig. 10. The structure of S-Pd-Ga is more stable than structure of Pd-S-Ga. A calculated atomic structure of Pd atoms on the S-terminated Ga(0001), indicating that the stable position of Pd atom is between S and Ga atoms of the S-terminated Ga(0001). The binding energy of Pd on Ga(0001) is stronger than that of S. It means that the S adatom is easier to be desorbed from the Ga(0001) substrate than Pd during catalysis reaction. Even if other study which is about Pd catalyst on S-terminated GaAs(001) have concluded that the role of S is to make bonding between GaAs substrate and Pd, we found that S does not play the same role in the case of Ga. Pd transfers slightly the valence electron to S. The number of valence electron of Pd is close to 0 when it use S 37

5 FIG. 6: The calculated adsorption energy contour map for the adsorption of Pd on S- tour map for the adsorption of S adsorbed FIG. 7: The calculated adsorption energy con- Pd terminated Ga(0001)-(1x1) surface. adatom on Ga(0001)-(1x1) surface. FIG. 8: The calculated adsorption energy contour map for the adsorption of S on Ga(0001)- tour map for the adsorption of Pd on FIG. 9: The calculated adsorption energy con- Ga(0001)- (1x1) surface. (1x1) surface. atom as a role of bonding between Pd and Ga(0001) substrate. As well as the result of Pd on S- terminated GaAs(001) [5], we have found that the number of valence electron of Pd on S-terminated Ga(0001) have the common point which is close to 0, also. Thus, the important role of S on the case of Ga substrate is to make the number of the valence electron to be close to 0. IV. COCLUSIO The density functional calculation is performed to determine the structure of the Pd catalyst on S-terminated Ga(0001) surface. The S atom is easier to be desorbed from the Ga(0001) substrate than the Pd atom during catalyst reaction. The valence electron numbers of Pd on the S terminated 38

6 FIG. 10: The schematic illustration of the adsorption of S adsorbed Pd adatom on Ga(0001).The number of electron of Pd is -0.06. Ga(0001) are very close to 0. Similarly, that of Pd on the S-terminated GaAs(001) is also close to 0. In addition, we determined that the most likely structure of Pd catalyst on S terminated GaAs is As dimer model rather than S dimer. Acknowledgments This research was supported by Japan Science and Technology Agency (JST). [1] M. Arisawa, M. Hamada, I. Takamiya, S. Tsukamoto, Y. Arakawa, A. ishida, Adv. Synth.Catal. 348, 1063 (2006). [2]. agatoshi, M. Shimoda, T. Konishi and S. Tsukamoto, Applied Physics Express 2 (2009) 051002 [3] G. Kresse and J. Hafner, Phys. Rev. B 47, RC558(1993) [4] A. Ishii, H. Asano, M. Yokoyama, S. Tsukamoto, S. Shuto and M. Arisawa, Phys. Status Solidi C 7, o. 2, 359-361(2010) [5] M. Yokoyama, A. Ishii, S. Tsukamoto, S. Shuto and M. Arisawa, to be published 39