{"id":17977,"date":"2026-09-07T10:34:43","date_gmt":"2026-09-07T08:34:43","guid":{"rendered":"https:\/\/www.chromacon.com\/?post_type=articles&#038;p=17977"},"modified":"2026-09-07T10:47:37","modified_gmt":"2026-09-07T08:47:37","slug":"reversed-phase-mcsgp-for-galnac-conjugated-oligonucleotides","status":"publish","type":"articles","link":"https:\/\/www.chromacon.com\/en\/articles\/reversed-phase-mcsgp-for-galnac-conjugated-oligonucleotides\/","title":{"rendered":"Reversed-Phase MCSGP for GalNAc-Conjugated Oligonucleotides"},"content":{"rendered":"\n<!-- Research Highlight \u2014 GalNAc Oligonucleotide RP MCSGP \u2014 v1.0 -->\n<style>\n        \/* ==========================================\n           CSS VARIABLES - Customize your fonts\/colors here\n           ========================================== *\/\n        :root {\n            \/* Colors - modify these to match your WordPress theme *\/\n            --color-primary: #1a365d;\n            --color-primary-light: #2c5282;\n            --color-accent: #3182ce;\n            --color-accent-light: #63b3ed;\n            --color-text: #2d3748;\n            --color-text-light: #718096;\n            --color-bg: #ffffff;\n            --color-bg-alt: #f7fafc;\n            --color-bg-card: #edf2f7;\n            --color-border: #e2e8f0;\n\n            \/* Fonts - modify these to match your WordPress theme *\/\n            --font-heading: inherit;\n            --font-body: inherit;\n\n            \/* Spacing *\/\n            --space-xs: 0.5rem;\n            --space-sm: 1rem;\n            --space-md: 1.5rem;\n            --space-lg: 2rem;\n            --space-xl: 3rem;\n            --space-2xl: 4rem;\n\n            \/* Layout *\/\n            --max-width: 900px;\n            --sidebar-width: 260px;\n        }\n\n        \/* ==========================================\n           BASE STYLES\n           ========================================== *\/\n        *, *::before, *::after {\n            box-sizing: border-box;\n        }\n\n        html {\n            scroll-behavior: smooth;\n        }\n\n        body {\n            margin: 0;\n            padding: 0;\n            font-family: var(--font-body);\n            color: var(--color-text);\n            background: var(--color-bg);\n            line-height: 1.7;\n            font-size: 17px;\n        }\n\n        img {\n            max-width: 100%;\n            height: auto;\n            display: block;\n        }\n\n        \/* ==========================================\n           LAYOUT\n           ========================================== *\/\n        .article-wrapper {\n            display: flex;\n            max-width: 1400px;\n            margin: 0 auto;\n            position: relative;\n        }\n\n        .article-main {\n            flex: 1;\n            max-width: var(--max-width);\n            margin: 0 auto;\n            padding: var(--space-lg);\n        }\n\n        \/* Sidebar \/ Table of Contents *\/\n        .article-sidebar {\n            width: var(--sidebar-width);\n            flex-shrink: 0;\n            padding: var(--space-lg);\n            display: none;\n        }\n\n        @media (min-width: 1200px) {\n            .article-sidebar {\n                display: block;\n            }\n\n            .article-wrapper {\n                padding-left: var(--space-lg);\n            }\n        }\n\n        .toc {\n            position: sticky;\n            top: var(--space-lg);\n        }\n\n        .toc-title {\n            font-size: 0.75rem;\n            text-transform: uppercase;\n            letter-spacing: 0.1em;\n            color: var(--color-text-light);\n            margin-bottom: var(--space-sm);\n            font-weight: 600;\n        }\n\n        .toc-list {\n            list-style: none;\n            padding: 0;\n            margin: 0;\n        }\n\n        .toc-list li {\n            margin-bottom: var(--space-xs);\n        }\n\n        .toc-list a {\n            color: var(--color-text-light);\n            text-decoration: none;\n            font-size: 0.875rem;\n            display: block;\n            padding: var(--space-xs) 0;\n            border-left: 2px solid transparent;\n            padding-left: var(--space-sm);\n            transition: all 0.2s ease;\n        }\n\n        .toc-list a:hover,\n        .toc-list a.active {\n            color: var(--color-accent);\n            border-left-color: var(--color-accent);\n        }\n\n        \/* ==========================================\n           HERO SECTION\n           ========================================== *\/\n        .article-hero {\n            background: linear-gradient(135deg, var(--color-primary) 0%, var(--color-primary-light) 100%);\n            color: white;\n            padding: var(--space-2xl) var(--space-lg);\n            margin: calc(-1 * var(--space-lg));\n            margin-bottom: var(--space-xl);\n            border-radius: 0 0 2rem 2rem;\n        }\n\n        .article-hero h2 {\n            font-family: var(--font-heading);\n            font-size: clamp(2rem, 5vw, 2.75rem);\n            margin: 0 0 var(--space-md);\n            line-height: 1.2;\n            font-weight: 700;\n            color: white;\n        }\n\n        .article-hero .lead {\n            font-size: 1.2rem;\n            color: white;\n            max-width: 700px;\n            margin: 0;\n        }\n\n        \/* ==========================================\n           TYPOGRAPHY\n           ========================================== *\/\n        h2 {\n            font-family: var(--font-heading);\n            font-size: 1.75rem;\n            color: var(--color-primary);\n            margin: var(--space-2xl) 0 var(--space-md);\n            padding-bottom: var(--space-sm);\n            border-bottom: 2px solid var(--color-border);\n            scroll-margin-top: var(--space-lg);\n        }\n\n        h3 {\n            font-family: var(--font-heading);\n            font-size: 1.35rem;\n            color: var(--color-accent);\n            margin: var(--space-xl) 0 var(--space-sm);\n        }\n\n        h4 {\n            font-family: var(--font-heading);\n            font-size: 1.15rem;\n            color: var(--color-text);\n            margin: var(--space-lg) 0 var(--space-xs);\n        }\n\n        p {\n            margin: 0 0 var(--space-md);\n        }\n\n        strong {\n            color: var(--color-primary);\n        }\n\n        \/* ==========================================\n           KEY CONCEPT CALLOUT\n           ========================================== *\/\n        .callout {\n            background: var(--color-bg-alt);\n            border-left: 4px solid var(--color-accent);\n            padding: var(--space-md) var(--space-lg);\n            margin: var(--space-lg) 0;\n            border-radius: 0 0.5rem 0.5rem 0;\n        }\n\n        .callout p {\n            margin: 0;\n            font-size: 1.15rem;\n        }\n\n        .callout strong {\n            color: var(--color-accent);\n        }\n\n        \/* ==========================================\n           LISTS - Card Style\n           ========================================== *\/\n        .card-list {\n            list-style: none;\n            padding: 0;\n            margin: var(--space-lg) 0;\n            display: grid;\n            gap: var(--space-sm);\n        }\n\n        .card-list li {\n            background: var(--color-bg);\n            padding: var(--space-md);\n            border-radius: 0.75rem;\n            border: 1px solid var(--color-border);\n            transition: transform 0.2s ease, box-shadow 0.2s ease;\n            font-size: 16px;\n        }\n\n        .card-list li:hover {\n            transform: translateY(-2px);\n            box-shadow: 0 4px 12px rgba(0,0,0,0.08);\n        }\n\n        .card-list li strong {\n            display: block;\n            margin-bottom: var(--space-xs);\n            font-size: 1.1rem;\n            color: var(--color-accent);\n        }\n\n        \/* Simple inline lists *\/\n        .simple-list {\n            padding-left: var(--space-lg);\n            margin: var(--space-md) 0;\n        }\n\n        .simple-list li {\n            margin-bottom: var(--space-xs);\n        }\n\n        \/* Nested lists *\/\n        .nested-list {\n            padding-left: var(--space-lg);\n            margin: var(--space-sm) 0;\n        }\n\n        .nested-list li {\n            margin-bottom: var(--space-xs);\n        }\n\n        \/* ==========================================\n           TWO-COLUMN GRID\n           ========================================== *\/\n        .two-col-grid {\n            display: grid;\n            grid-template-columns: 1fr;\n            gap: var(--space-md);\n            margin: var(--space-lg) 0;\n        }\n\n        @media (min-width: 640px) {\n            .two-col-grid {\n                grid-template-columns: 1fr 1fr;\n            }\n        }\n\n        .grid-card {\n            background: var(--color-bg);\n            padding: var(--space-md);\n            border-radius: 0.75rem;\n            border: 1px solid var(--color-border);\n            font-size: 16px;\n        }\n\n        .grid-card h4 {\n            margin-top: 0;\n            color: var(--color-accent);\n        }\n\n        .grid-card p:last-child {\n            margin-bottom: 0;\n        }\n\n        \/* ==========================================\n           IMAGES\n           ========================================== *\/\n        .article-image {\n            margin: var(--space-xl) 0;\n            border-radius: 1rem;\n            overflow: hidden;\n            box-shadow: 0 4px 20px rgba(0,0,0,0.1);\n        }\n\n        .article-image img {\n            width: 100%;\n        }\n\n        .article-image figcaption {\n            background: var(--color-bg-alt);\n            padding: var(--space-sm) var(--space-md);\n            font-size: 0.875rem;\n            color: var(--color-text-light);\n            text-align: center;\n        }\n\n        \/* ==========================================\n           ANALOGY BOX\n           ========================================== *\/\n        .analogy-box {\n            background: linear-gradient(135deg, var(--color-bg-card) 0%, var(--color-bg-alt) 100%);\n            border-radius: 1rem;\n            padding: var(--space-lg);\n            margin: var(--space-xl) 0;\n            border: 1px solid var(--color-border);\n        }\n\n        .analogy-box h4 {\n            margin-top: 0;\n            display: flex;\n            align-items: center;\n            gap: var(--space-xs);\n        }\n\n        .analogy-box h4::before {\n            content: \"\ud83d\udca1\";\n        }\n\n        \/* ==========================================\n           COMPONENT GRID\n           ========================================== *\/\n        .component-grid {\n            display: grid;\n            grid-template-columns: 1fr;\n            gap: var(--space-md);\n            margin: var(--space-lg) 0;\n        }\n\n        @media (min-width: 640px) {\n            .component-grid {\n                grid-template-columns: repeat(2, 1fr);\n            }\n        }\n\n        @media (min-width: 900px) {\n            .component-grid {\n                grid-template-columns: repeat(3, 1fr);\n            }\n        }\n\n        .component-card {\n            background: var(--color-bg);\n            border: 1px solid var(--color-border);\n            border-radius: 0.75rem;\n            padding: var(--space-md);\n            transition: all 0.2s ease;\n        }\n\n        .component-card:hover {\n            border-color: var(--color-accent);\n            box-shadow: 0 4px 12px rgba(0,0,0,0.08);\n        }\n\n        .component-card strong {\n            display: block;\n            font-size: 1.05rem;\n            margin-bottom: var(--space-xs);\n            color: var(--color-accent);\n        }\n\n        .component-card p {\n            font-size: 0.95rem;\n            color: var(--color-text-light);\n            margin: 0;\n        }\n\n        \/* ==========================================\n           CHROMATOGRAM FEATURES\n           ========================================== *\/\n        .feature-list {\n            list-style: none;\n            padding: 0;\n            margin: var(--space-lg) 0;\n        }\n\n        .feature-list li {\n            display: flex;\n            gap: var(--space-md);\n            padding: var(--space-md) 0;\n            border-bottom: 1px solid var(--color-border);\n        }\n\n        .feature-list li:last-child {\n            border-bottom: none;\n        }\n\n        .feature-label {\n            flex-shrink: 0;\n            width: 140px;\n            font-weight: 600;\n            color: var(--color-primary);\n        }\n\n        @media (max-width: 480px) {\n            .feature-list li {\n                flex-direction: column;\n                gap: var(--space-xs);\n            }\n\n            .feature-label {\n                width: auto;\n            }\n        }\n\n        \/* ==========================================\n           COMPARISON SECTION\n           ========================================== *\/\n        .comparison-section {\n            display: grid;\n            grid-template-columns: 1fr;\n            gap: var(--space-lg);\n            margin: var(--space-xl) 0;\n        }\n\n        @media (min-width: 768px) {\n            .comparison-section {\n                grid-template-columns: 1fr 1fr;\n            }\n        }\n\n        .comparison-card {\n            background: var(--color-bg);\n            border: 2px solid var(--color-border);\n            border-radius: 1rem;\n            overflow: hidden;\n        }\n\n        .comparison-card.analytical {\n            border-color: var(--color-accent-light);\n        }\n\n        .comparison-card.preparative {\n            border-color: var(--color-primary);\n        }\n\n        .comparison-header {\n            padding: var(--space-sm) var(--space-md);\n            font-weight: 700;\n            font-size: 1rem;\n        }\n\n        .comparison-card.analytical .comparison-header {\n            background: var(--color-accent-light);\n            color: var(--color-primary);\n        }\n\n        .comparison-card.preparative .comparison-header {\n            background: var(--color-primary);\n            color: white;\n        }\n\n        .comparison-content {\n            padding: var(--space-sm) var(--space-md);\n            font-size: 0.875rem;\n        }\n\n        .comparison-content ul {\n            list-style: none;\n            padding: 0;\n            margin: 0;\n        }\n\n        .comparison-content li {\n            padding: 0.35rem 0;\n            border-bottom: 1px solid var(--color-border);\n            display: flex;\n            gap: var(--space-xs);\n        }\n\n        .comparison-content li:last-child {\n            border-bottom: none;\n        }\n\n        .comparison-content li strong {\n            flex-shrink: 0;\n            width: 75px;\n            color: var(--color-text-light);\n            font-weight: 600;\n            font-size: 0.8rem;\n        }\n\n        \/* ==========================================\n           TABLE - Glossary\n           ========================================== *\/\n        .glossary-table {\n            width: 100%;\n            border-collapse: collapse;\n            margin: var(--space-lg) 0;\n            font-size: 0.95rem;\n        }\n\n        .glossary-table thead {\n            background: var(--color-primary);\n            color: white;\n        }\n\n        .glossary-table th {\n            padding: var(--space-md);\n            text-align: left;\n            font-weight: 600;\n            color: white !important;\n        }\n\n        .glossary-table td {\n            padding: var(--space-md);\n            border-bottom: 1px solid var(--color-border);\n            vertical-align: top;\n        }\n\n        .glossary-table tr:hover td {\n            background: var(--color-bg-alt);\n        }\n\n        .glossary-table td:first-child {\n            font-weight: 600;\n            color: var(--color-primary);\n            white-space: nowrap;\n            width: 220px;\n        }\n\n        @media (max-width: 640px) {\n            .glossary-table {\n                display: block;\n            }\n\n            .glossary-table thead {\n                display: none;\n            }\n\n            .glossary-table tbody {\n                display: block;\n            }\n\n            .glossary-table tr {\n                display: block;\n                margin-bottom: var(--space-md);\n                background: var(--color-bg-alt);\n                border-radius: 0.5rem;\n                overflow: hidden;\n            }\n\n            .glossary-table td {\n                display: block;\n                border-bottom: none;\n            }\n\n            .glossary-table td:first-child {\n                width: auto;\n                white-space: normal;\n                background: var(--color-primary);\n                color: white;\n                padding: var(--space-sm) var(--space-md);\n            }\n        }\n\n        \/* ==========================================\n           FAQ ACCORDION\n           ========================================== *\/\n        .faq-section {\n            margin: var(--space-xl) 0;\n        }\n\n        .faq-item {\n            border: 1px solid var(--color-border);\n            border-radius: 0.75rem;\n            margin-bottom: var(--space-sm);\n            overflow: hidden;\n            transition: box-shadow 0.2s ease;\n        }\n\n        .faq-item:hover {\n            box-shadow: 0 2px 8px rgba(0,0,0,0.06);\n        }\n\n        .faq-question {\n            width: 100%;\n            background: var(--color-bg);\n            border: none;\n            padding: var(--space-md) var(--space-lg);\n            text-align: left;\n            cursor: pointer;\n            font-size: 1.05rem;\n            font-weight: 600;\n            color: var(--color-primary);\n            display: flex;\n            justify-content: space-between;\n            align-items: center;\n            gap: var(--space-md);\n            font-family: inherit;\n        }\n\n        .faq-question::after {\n            content: \"+\";\n            font-size: 1.5rem;\n            font-weight: 300;\n            color: var(--color-accent);\n            transition: transform 0.2s ease;\n        }\n\n        .faq-item.open .faq-question::after {\n            transform: rotate(45deg);\n        }\n\n        .faq-answer {\n            max-height: 0;\n            overflow: hidden;\n            transition: max-height 0.3s ease;\n        }\n\n        .faq-item.open .faq-answer {\n            max-height: 1000px;\n        }\n\n        .faq-answer-inner {\n            padding: 0 var(--space-lg) var(--space-lg);\n            border-top: 1px solid var(--color-border);\n            padding-top: var(--space-md);\n        }\n\n        .faq-answer-inner p:last-child,\n        .faq-answer-inner ul:last-child {\n            margin-bottom: 0;\n        }\n\n        \/* ==========================================\n           MOBILE TABLE OF CONTENTS\n           ========================================== *\/\n        .mobile-toc {\n            display: block;\n            background: var(--color-bg-alt);\n            border-radius: 0.75rem;\n            padding: var(--space-md);\n            margin-bottom: var(--space-xl);\n        }\n\n        @media (min-width: 1200px) {\n            .mobile-toc {\n                display: none;\n            }\n        }\n\n        .mobile-toc-toggle {\n            width: 100%;\n            background: none;\n            border: none;\n            font-family: inherit;\n            font-size: 0.875rem;\n            font-weight: 600;\n            color: var(--color-primary);\n            cursor: pointer;\n            display: flex;\n            justify-content: space-between;\n            align-items: center;\n            padding: 0;\n        }\n\n        .mobile-toc-toggle::after {\n            content: \"\u25bc\";\n            font-size: 0.75rem;\n            transition: transform 0.2s ease;\n        }\n\n        .mobile-toc.open .mobile-toc-toggle::after {\n            transform: rotate(180deg);\n        }\n\n        .mobile-toc-list {\n            display: none;\n        }\n\n        .mobile-toc.open .mobile-toc-list {\n            display: block;\n        }\n\n        .mobile-toc-list ul {\n            list-style: none;\n            padding: 0;\n            margin: var(--space-sm) 0 0;\n        }\n\n        .mobile-toc-list li {\n            margin-bottom: var(--space-xs);\n        }\n\n        .mobile-toc-list a {\n            color: var(--color-text-light);\n            text-decoration: none;\n            font-size: 0.875rem;\n            display: block;\n            padding: var(--space-xs) 0;\n        }\n\n        .mobile-toc-list a:hover {\n            color: var(--color-accent);\n        }\n\n        \/* ==========================================\n           UTILITY CLASSES\n           ========================================== *\/\n        .visually-hidden {\n            position: absolute;\n            width: 1px;\n            height: 1px;\n            padding: 0;\n            margin: -1px;\n            overflow: hidden;\n            clip: rect(0, 0, 0, 0);\n            white-space: nowrap;\n            border: 0;\n        }\n\n        \/* ==========================================================\n           ADDITIONS FOR THIS ARTICLE TYPE (research highlight)\n           Everything above is the base stylesheet, unchanged.\n           Everything below is new and uses the same tokens.\n           ========================================================== *\/\n\n        :root {\n            \/* Figure colors follow the article's own convention (Figures 1 and 2):\n               product = red, weakly adsorbing impurities = blue, strongly adsorbing = green.\n               Only the product color is used here; the mass-flow figure adds neutral tones\n               for feed composition. Literal hex is repeated in the SVG because presentation\n               attributes cannot read CSS custom properties. *\/\n            --fig-product: #C0392B;\n            --fig-feed-product: #8FA3B0;\n            --fig-feed-impurity: #D8DFE5;\n\n            --color-green: #1b7a4b;        \/* green eluent \/ MCSGP *\/\n            --color-green-light: #e2efe7;\n            --color-flag: #c0392b;         \/* conventional ACN-TFA pool *\/\n            --color-flag-light: #f8e3dd;\n            --color-cyan-light: #dcedf4;   \/* weak-side recycling *\/\n            --font-mono: ui-monospace, SFMono-Regular, Menlo, Consolas, monospace;\n        }\n\n        \/* MOBILE OVERFLOW FIX\n           .article-main is a flex item, so its default min-width:auto prevents it\n           from shrinking below the intrinsic width of its widest child. The results\n           table is ~620px min-content, which pushed the whole page to 684px on a\n           360px screen. min-width:0 lets the flex item shrink; the table then\n           scrolls inside .table-scroll instead of stretching the document. *\/\n        .article-main {\n            min-width: 0;\n        }\n\n        \/* Eyebrow above the hero headline *\/\n        .article-hero .eyebrow {\n            font-size: 0.75rem;\n            text-transform: uppercase;\n            letter-spacing: 0.12em;\n            font-weight: 600;\n            color: var(--color-accent-light);\n            margin: 0 0 var(--space-sm);\n        }\n\n        \/* Headline figures *\/\n        .stats-grid {\n            display: grid;\n            grid-template-columns: repeat(2, 1fr);\n            gap: var(--space-sm);\n            margin: var(--space-lg) 0;\n        }\n\n        @media (min-width: 768px) {\n            .stats-grid {\n                grid-template-columns: repeat(4, 1fr);\n            }\n        }\n\n        .stat-card {\n            background: var(--color-bg-alt);\n            border: 1px solid var(--color-border);\n            border-radius: 0.75rem;\n            padding: var(--space-md);\n        }\n\n        .stat-value {\n            font-family: var(--font-heading);\n            font-size: 2rem;\n            font-weight: 700;\n            line-height: 1.1;\n            color: var(--color-primary);\n        }\n\n        .stat-value.is-green {\n            color: var(--color-green);\n        }\n\n        .stat-value span {\n            font-size: 1rem;\n            font-weight: 600;\n        }\n\n        .stat-label {\n            font-size: 0.8rem;\n            line-height: 1.45;\n            color: var(--color-text-light);\n            margin-top: var(--space-xs);\n        }\n\n        \/* Interactive process explorer *\/\n        .explorer {\n            background: var(--color-bg);\n            border: 1px solid var(--color-border);\n            border-radius: 1rem;\n            padding: var(--space-md);\n            margin: var(--space-lg) 0;\n        }\n\n        @media (min-width: 768px) {\n            .explorer {\n                padding: var(--space-lg);\n            }\n        }\n\n        .explorer-tabs {\n            display: flex;\n            flex-wrap: wrap;\n            gap: var(--space-xs);\n            margin-bottom: var(--space-md);\n        }\n\n        .explorer-tab {\n            flex: 1 1 180px;\n            background: var(--color-bg-alt);\n            border: 1px solid var(--color-border);\n            border-radius: 0.5rem;\n            padding: 0.6rem 0.9rem;\n            font-family: inherit;\n            font-size: 0.95rem;\n            font-weight: 600;\n            line-height: 1.25;\n            color: var(--color-text-light);\n            text-align: left;\n            cursor: pointer;\n            transition: all 0.2s ease;\n        }\n\n        .explorer-tab small {\n            display: block;\n            font-size: 0.75rem;\n            font-weight: 400;\n            margin-top: 2px;\n            opacity: 0.85;\n        }\n\n        .explorer-tab:hover {\n            border-color: var(--color-accent);\n            color: var(--color-primary);\n        }\n\n        .explorer-tab[aria-pressed=\"true\"] {\n            background: var(--color-primary);\n            border-color: var(--color-primary);\n            color: white;\n        }\n\n        .plot-scroll {\n            max-width: 100%;\n        }\n\n        .explorer-plot {\n            width: 100%;\n            height: auto;\n            display: block;\n        }\n\n        \/* On phones the 760-unit viewBox scales the 11px labels down to ~5px.\n           Hold a legible minimum and let the plot scroll inside its own box\n           instead; the figures below the chart carry the same numbers. *\/\n        @media (max-width: 700px) {\n            .plot-scroll {\n                overflow-x: auto;\n                -webkit-overflow-scrolling: touch;\n            }\n            .explorer-plot { min-width: 620px; }\n        }\n\n        .explorer-plot text {\n            font-family: var(--font-mono);\n            font-size: 11px;\n        }\n\n        .explorer-plot .trace {\n            fill: none;\n            stroke: var(--color-text);\n            stroke-width: 2.2;\n            stroke-linejoin: round;\n            stroke-linecap: round;\n        }\n\n        .explorer-plot .axis {\n            stroke: var(--color-text);\n            stroke-width: 1.2;\n        }\n\n        .explorer-plot .grid-line {\n            stroke: var(--color-border);\n            stroke-width: 1;\n        }\n\n        .explorer-state {\n            opacity: 0;\n            transition: opacity 0.35s ease;\n        }\n\n        .explorer-state.is-on {\n            opacity: 1;\n        }\n\n        .explorer-legend {\n            display: flex;\n            flex-wrap: wrap;\n            gap: var(--space-md);\n            list-style: none;\n            padding: 0;\n            margin: var(--space-sm) 0 0;\n            font-size: 0.8rem;\n            color: var(--color-text-light);\n        }\n\n        .explorer-legend li {\n            display: flex;\n            align-items: center;\n            gap: 0.4rem;\n        }\n\n        .explorer-legend i {\n            width: 14px;\n            height: 14px;\n            border-radius: 3px;\n            border: 1px solid rgba(0,0,0,0.12);\n        }\n\n        .explorer-readout {\n            display: grid;\n            grid-template-columns: 1fr;\n            gap: var(--space-md);\n            border-top: 1px solid var(--color-border);\n            margin-top: var(--space-md);\n            padding-top: var(--space-md);\n        }\n\n        @media (min-width: 768px) {\n            .explorer-readout {\n                grid-template-columns: 1.3fr 1fr;\n            }\n        }\n\n        .explorer-caption {\n            margin: 0;\n            font-size: 0.95rem;\n        }\n\n        .explorer-nums {\n            display: grid;\n            grid-template-columns: 1fr 1fr;\n            gap: var(--space-sm);\n            margin: 0;\n        }\n\n        .explorer-num dt {\n            font-size: 0.72rem;\n            text-transform: uppercase;\n            letter-spacing: 0.08em;\n            color: var(--color-text-light);\n        }\n\n        .explorer-num dd {\n            font-family: var(--font-heading);\n            margin: 2px 0 0;\n            font-size: 1.5rem;\n            font-weight: 700;\n            line-height: 1;\n            color: var(--color-primary);\n        }\n\n        .explorer-num dd span {\n            font-size: 0.8rem;\n            font-weight: 600;\n            color: var(--color-text-light);\n        }\n\n        @media (max-width: 420px) {\n            .explorer-nums { grid-template-columns: 1fr; }\n            .explorer-num dd { font-size: 1.35rem; }\n        }\n\n        .explorer-note {\n            font-size: 0.85rem;\n            color: var(--color-text-light);\n            margin: var(--space-sm) 0 0;\n        }\n\n        \/* Results table *\/\n        .table-scroll {\n            overflow-x: auto;\n            -webkit-overflow-scrolling: touch;\n            max-width: 100%;\n            margin: var(--space-lg) 0 var(--space-sm);\n        }\n\n        \/* Tell people the table scrolls \u2014 but only between 641px and 700px,\n           which is the one range where it does. At 640px and below the table\n           collapses into stacked cards and there is nothing to swipe. *\/\n        .scroll-hint {\n            display: none;\n            font-size: 0.85rem;\n            color: var(--color-text-light);\n            margin: var(--space-md) 0 calc(-1 * var(--space-sm));\n        }\n\n        @media (min-width: 641px) and (max-width: 700px) {\n            .scroll-hint { display: block; }\n        }\n\n        .data-table {\n            width: 100%;\n            min-width: 620px;\n            border-collapse: collapse;\n            font-size: 0.95rem;\n        }\n\n        .data-table thead {\n            background: var(--color-primary);\n            color: white;\n        }\n\n        .data-table th {\n            padding: var(--space-sm) var(--space-md);\n            text-align: right;\n            font-size: 0.85rem;\n            font-weight: 600;\n            color: white !important;\n            vertical-align: bottom;\n        }\n\n        .data-table th:first-child {\n            text-align: left;\n        }\n\n        .data-table td {\n            padding: var(--space-sm) var(--space-md);\n            text-align: right;\n            border-bottom: 1px solid var(--color-border);\n        }\n\n        .data-table td:first-child {\n            text-align: left;\n            font-weight: 600;\n            color: var(--color-primary);\n        }\n\n        .data-table tr.subhead td {\n            background: var(--color-bg-card);\n            color: var(--color-primary);\n            font-weight: 700;\n            text-align: left;\n        }\n\n        .data-table tr.subhead td span {\n            font-weight: 400;\n            font-size: 0.8rem;\n            color: var(--color-text-light);\n            margin-left: var(--space-xs);\n        }\n\n        .data-table tr.highlight td {\n            background: var(--color-green-light);\n        }\n\n        .data-table .is-up {\n            color: var(--color-green);\n            font-weight: 600;\n        }\n\n        .data-table .is-down {\n            color: var(--color-flag);\n        }\n\n        .table-note {\n            font-size: 0.85rem;\n            color: var(--color-text-light);\n            margin: 0 0 var(--space-lg);\n        }\n\n        \/* Limitations box *\/\n        .limits-box {\n            background: var(--color-bg-alt);\n            border: 1px solid var(--color-border);\n            border-left: 4px solid var(--color-text-light);\n            border-radius: 0 0.5rem 0.5rem 0;\n            padding: var(--space-md) var(--space-lg);\n            margin: var(--space-xl) 0;\n        }\n\n        .limits-box h4 {\n            margin-top: 0;\n        }\n\n        .limits-box ul {\n            margin-bottom: 0;\n        }\n\n        \/* Paper \/ citation card *\/\n        .paper-card {\n            display: grid;\n            grid-template-columns: 1fr;\n            gap: var(--space-lg);\n            background: var(--color-bg-alt);\n            border: 1px solid var(--color-border);\n            border-radius: 1rem;\n            padding: var(--space-lg);\n            margin: var(--space-lg) 0;\n        }\n\n        @media (min-width: 640px) {\n            .paper-card {\n                grid-template-columns: 180px minmax(0, 1fr);\n            }\n        }\n\n        \/* No cover image available: collapse to a single full-width column\n           so the text does not get trapped in the 180px image track. *\/\n        .paper-card.no-cover {\n            grid-template-columns: minmax(0, 1fr);\n        }\n\n        .paper-card img {\n            border-radius: 0.5rem;\n            border: 1px solid var(--color-border);\n        }\n\n        .paper-title {\n            font-family: var(--font-heading);\n            font-size: 1.25rem;\n            line-height: 1.3;\n            color: var(--color-primary);\n            margin: 0 0 var(--space-xs);\n        }\n\n        .paper-authors {\n            font-size: 0.95rem;\n            margin: 0 0 var(--space-xs);\n        }\n\n        .paper-meta {\n            font-size: 0.85rem;\n            color: var(--color-text-light);\n            margin: 0 0 var(--space-md);\n        }\n\n        .btn-row {\n            display: flex;\n            flex-wrap: wrap;\n            gap: var(--space-xs);\n            margin-bottom: var(--space-md);\n        }\n\n        .btn {\n            display: inline-block;\n            background: var(--color-accent);\n            color: white !important;\n            text-decoration: none;\n            padding: 0.6rem 1.1rem;\n            border-radius: 0.5rem;\n            font-size: 0.9rem;\n            font-weight: 600;\n            transition: background 0.2s ease;\n        }\n\n        .btn:hover {\n            background: var(--color-primary);\n        }\n\n        .btn.secondary {\n            background: transparent;\n            color: var(--color-accent) !important;\n            border: 1px solid var(--color-border);\n        }\n\n        .btn.secondary:hover {\n            background: var(--color-bg-card);\n            color: var(--color-primary) !important;\n        }\n\n        .disclosure {\n            font-size: 0.85rem;\n            color: var(--color-text-light);\n            border-top: 1px solid var(--color-border);\n            padding-top: var(--space-sm);\n            margin: 0;\n        }\n\n        .cite-block {\n            margin-top: var(--space-sm);\n        }\n\n        .cite-block summary {\n            cursor: pointer;\n            font-size: 0.85rem;\n            font-weight: 600;\n            color: var(--color-text-light);\n        }\n\n        .cite-block pre {\n            background: var(--color-bg);\n            border: 1px solid var(--color-border);\n            border-radius: 0.5rem;\n            padding: var(--space-sm);\n            margin: var(--space-xs) 0 0;\n            font-family: var(--font-mono);\n            font-size: 0.78rem;\n            line-height: 1.6;\n            white-space: pre-wrap;\n            overflow-wrap: anywhere;\n            overflow-x: auto;\n        }\n\n        \/* ==========================================================\n           MOBILE DELIVERY\n           .article-main is a flex item, so by default it will not\n           shrink below the widest thing inside it. Without min-width:0\n           a wide child pushes the whole column past the viewport\n           instead of wrapping.\n           ========================================================== *\/\n        .article-main {\n            min-width: 0;\n            overflow-wrap: break-word;\n        }\n\n        .explorer-plot,\n        .paper-card img {\n            max-width: 100%;\n        }\n\n        .stats-grid > *,\n        .component-grid > *,\n        .two-col-grid > *,\n        .comparison-section > *,\n        .explorer-nums > *,\n        .paper-card > *,\n        .explorer-tab {\n            min-width: 0;\n        }\n\n        \/* Results table: collapse into stacked cards below 640px, the\n           same treatment the stylesheet already gives .glossary-table,\n           rather than forcing a 620px sideways scroll. *\/\n        @media (max-width: 640px) {\n            .table-scroll {\n                overflow-x: visible;\n            }\n\n            .data-table {\n                display: block;\n                min-width: 0;\n                font-size: 0.9rem;\n            }\n\n            .data-table thead {\n                display: none;\n            }\n\n            .data-table tbody {\n                display: block;\n            }\n\n            .data-table tr {\n                display: block;\n                margin-bottom: var(--space-md);\n                background: var(--color-bg-alt);\n                border-radius: 0.5rem;\n                overflow: hidden;\n            }\n\n            .data-table tr.subhead {\n                background: transparent;\n                margin-bottom: var(--space-xs);\n            }\n\n            .data-table tr.subhead td {\n                border-bottom: none;\n                border-radius: 0.5rem;\n            }\n\n            .data-table tr.subhead td span {\n                display: block;\n                margin-left: 0;\n            }\n\n            .data-table td {\n                display: flex;\n                justify-content: space-between;\n                align-items: baseline;\n                gap: var(--space-sm);\n                padding: 0.6rem var(--space-md);\n                text-align: right;\n            }\n\n            .data-table td:last-child {\n                border-bottom: none;\n            }\n\n            .data-table tr:not(.subhead) td:first-child {\n                display: block;\n                text-align: left;\n                background: var(--color-primary);\n                color: white;\n                border-bottom: none;\n                padding: var(--space-sm) var(--space-md);\n            }\n\n            .data-table tr.highlight td:first-child {\n                background: var(--color-green);\n            }\n\n            .data-table td[data-label]::before {\n                content: attr(data-label);\n                font-weight: 600;\n                color: var(--color-text-light);\n                text-align: left;\n            }\n        }\n    <\/style>\n\n\n\n<div class=\"article-wrapper\">\n        <!-- Sidebar TOC (desktop only) -->\n        <aside class=\"article-sidebar\">\n            <nav class=\"toc\">\n                <p class=\"toc-title\">In this article<\/p>\n                <ul class=\"toc-list\">\n                    <li><a href=\"#why-matters\">1. Why This Study Matters<\/a><\/li>\n                    <li><a href=\"#at-a-glance\">2. The Study at a Glance<\/a><\/li>\n                    <li><a href=\"#what-yield-buys\">3. What the Yield Buys<\/a><\/li>\n                    <li><a href=\"#designing\">4. Designing the Process<\/a><\/li>\n                    <li><a href=\"#six-cycles\">5. What Six Cycles Showed<\/a><\/li>\n                    <li><a href=\"#results\">6. Full Results<\/a><\/li>\n                    <li><a href=\"#implications\">7. What It Means for Manufacturing<\/a><\/li>\n                    <li><a href=\"#glossary\">8. Key Terms<\/a><\/li>\n                    <li><a href=\"#faq\">9. FAQs<\/a><\/li>\n                <\/ul>\n            <\/nav>\n        <\/aside>\n\n        <!-- Main Content -->\n        <article class=\"article-main\">\n            <!-- Hero Section -->\n            <header class=\"article-hero\">\n                <p class=\"eyebrow\">Research Highlight<\/p>\n                <h2>Breaking the Yield\u2013Purity Trade-Off for GalNAc Oligonucleotides<\/h2>\n                <p class=\"lead\">A reversed-phase batch purification of a GalNAc-cluster-conjugated gapmer recovered just over half the product at specification. Run as a continuous twin-column process on the same chemistry, it recovered 91.6% \u2014 enough to shrink the oligonucleotide synthesis step by 42.5% for the same daily output.<\/p>\n            <\/header>\n\n            <!-- Article citation card \u2014 lead position, directly under the header -->\n            <div class=\"paper-card no-cover\">\n                <div>\n                    <p class=\"paper-title\">Purification of a GalNAc-cluster-conjugated oligonucleotide by reversed-phase twin-column continuous chromatography<\/p>\n                    <p class=\"paper-authors\">Richard Weldon, J\u00f6rg Lill, Martin Olbrich, Pascal Schmidt and Thomas M\u00fcller-Sp\u00e4th<\/p>\n                    <p class=\"paper-meta\">\n                        <em>Journal of Chromatography A<\/em> <strong>2022<\/strong>, 1663, 462734 \u00b7 Available online 12 December 2021<br>\n                        DOI: 10.1016\/j.chroma.2021.462734 \u00b7 Open access, CC BY-NC-ND 4.0\n                    <\/p>\n                    <div class=\"btn-row\">\n                        <a class=\"btn\" href=\"https:\/\/doi.org\/10.1016\/j.chroma.2021.462734\" target=\"_blank\" rel=\"noopener\">Read the full article<\/a>\n                        <a class=\"btn secondary\" href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0021967321008566\" target=\"_blank\" rel=\"noopener\">View on ScienceDirect<\/a>\n                    <\/div>\n                    <p class=\"disclosure\">The authors declare no known competing financial interests or personal relationships that could have appeared to influence the work. J\u00f6rg Lill, Martin Olbrich and Pascal Schmidt are at F. Hoffmann-La Roche Ltd., which funded the study; Richard Weldon and Thomas M\u00fcller-Sp\u00e4th are at YMC ChromaCon.<\/p>\n                    <details class=\"cite-block\">\n                        <summary>Citation (BibTeX)<\/summary>\n<pre>@article{weldon2022galnac,\n  title   = {Purification of a GalNAc-cluster-conjugated oligonucleotide by\n             reversed-phase twin-column continuous chromatography},\n  author  = {Weldon, Richard and Lill, J{\\\"o}rg and Olbrich, Martin\n             and Schmidt, Pascal and M{\\\"u}ller-Sp{\\\"a}th, Thomas},\n  journal = {Journal of Chromatography A},\n  volume  = {1663},\n  pages   = {462734},\n  year    = {2022},\n  doi     = {10.1016\/j.chroma.2021.462734}\n}<\/pre>\n                    <\/details>\n                <\/div>\n            <\/div>\n\n            <!-- Mobile TOC -->\n            <nav class=\"mobile-toc\" id=\"mobileToc\">\n                <button class=\"mobile-toc-toggle\" aria-expanded=\"false\" aria-controls=\"mobileTocList\">\n                    Table of Contents\n                <\/button>\n                <div class=\"mobile-toc-list\" id=\"mobileTocList\">\n                    <ul>\n                        <li><a href=\"#why-matters\">1. Why This Study Matters<\/a><\/li>\n                        <li><a href=\"#at-a-glance\">2. The Study at a Glance<\/a><\/li>\n                        <li><a href=\"#what-yield-buys\">3. What the Yield Buys<\/a><\/li>\n                        <li><a href=\"#designing\">4. Designing the Process<\/a><\/li>\n                        <li><a href=\"#six-cycles\">5. What Six Cycles Showed<\/a><\/li>\n                        <li><a href=\"#results\">6. Full Results<\/a><\/li>\n                        <li><a href=\"#implications\">7. What It Means for Manufacturing<\/a><\/li>\n                        <li><a href=\"#glossary\">8. Key Terms<\/a><\/li>\n                        <li><a href=\"#faq\">9. FAQs<\/a><\/li>\n                    <\/ul>\n                <\/div>\n            <\/nav>\n\n            <p>The problem has a familiar shape. Widen the collection window and yield goes up while purity falls; narrow it and purity rises while yield falls. For an oligonucleotide, impurities differing from the target by a single nucleotide tend to co-elute with it, which makes that trade-off an expensive one. The study reports a clinical-grade purity target of more than 90% as typically costing between 20% and 50% of the product, followed by expensive re-chromatography to claw some of it back.<\/p>\n\n            <p>That loss does not stay in the chromatography suite. Solid-phase synthesis is the major cost driver in oligonucleotide production, so whatever the purification gives up has to be made in the first place.<\/p>\n\n            <div class=\"callout\">\n                <p>This study takes a reversed-phase batch method for a <strong>GalNAc-cluster-conjugated DNA-LNA gapmer<\/strong> and converts it to twin-column MCSGP. Yield rises from 52.70% to 91.58% at the same purity \u2014 94.2 area%, from a 77.5 area% crude \u2014 with productivity and buffer consumption essentially unchanged.<\/p>\n            <\/div>\n\n            <p>It is described as the first evaluation of <a href=\"https:\/\/www.chromacon.com\/en\/technologies\/mcsgp\/\">MCSGP<\/a> for a GalNAc-cluster-conjugated gapmer in reversed-phase mode, and it carries the result through to a scale-up estimate: what that yield is worth in kilograms of synthesis per day.<\/p>\n\n            <!-- Section 1 -->\n            <h2 id=\"why-matters\">1. Why This Study Matters: Synthesis Is Where the Cost Is<\/h2>\n\n            <p>Oligonucleotide production runs in four major steps \u2014 solid-phase synthesis, chromatographic purification, desalting by diafiltration or precipitation, and isolation by freeze drying. The study is direct about where the money sits: oligonucleotide synthesis is currently the major cost driver, and any loss of product in the downstream steps has a major impact on overall manufacturing costs.<\/p>\n\n            <p>It also frames the scale problem. Demand for production capacity is expanding rapidly, but oligonucleotide therapeutics are yet to be manufactured at multi-ton scale and no compound-dedicated production facilities exist \u2014 which is why pharmaceutical companies and CMOs are actively evaluating platform technologies that will optimize production and reduce costs.<\/p>\n\n            <h3>Why the separation is hard<\/h3>\n\n            <p>Synthesis delivers roughly 60% to 80% desired product alongside a mixture of impurities, and they fall into three families:<\/p>\n\n            <ul class=\"card-list\">\n                <li>\n                    <strong>Shortmers<\/strong>\n                    Sequences missing one or more nucleotides (N-x), left behind by incomplete chemical coupling reactions during elongation.\n                <\/li>\n                <li>\n                    <strong>Full-length product impurities<\/strong>\n                    Molecules of the right length carrying a defect \u2014 cyanoethyl modification, phosphate impurity, depurination \u2014 produced by modifying side reactions.\n                <\/li>\n                <li>\n                    <strong>Longmers<\/strong>\n                    Sequences with extra material, such as lambdamers and branchmers, arising from impurities present in the starting raw materials.\n                <\/li>\n            <\/ul>\n\n            <p>N-1 and N+1 are singled out as particularly challenging: one nucleotide shorter or longer than the target, chemically almost identical, and so tending to co-elute. Even after considerable optimization, a preparative batch method leaves impurities significantly co-eluting with the product, especially for highly phosphorothiolated sequences. Meeting specification therefore means fractionating carefully and accepting that the side fractions are either discarded or re-chromatographed at cost \u2014 lower productivity, more buffer, intermediate storage, extra quality control.<\/p>\n\n            <h3>Recycling, but not the usual kind<\/h3>\n\n            <p>MCSGP is not the only way to recycle. Steady-state recycling and closed-loop recycling do it too, and the paper draws a specific distinction. In those single-column processes the chromatographic profile passes through a pump every cycle, which is possibly detrimental to product stability and limits the process to isocratic operation. In MCSGP the recycled side fractions are fully adsorbed onto the second column and then subjected to renewed gradient elution, so product and impurities pass through a pump only during the load step. The authors argue this makes MCSGP well suited to complex molecules such as oligonucleotides.<\/p>\n\n            <!-- Section 2 -->\n            <h2 id=\"at-a-glance\">2. The Study at a Glance<\/h2>\n\n            <p>The molecule is a DNA-LNA gapmer conjugated to a bislysin GalNAc cluster \u2014 the sugar ligand that transports synthetic oligonucleotides into hepatocytes by endocytosis, allowing the development of treatments targeting diseases of the liver. It was assembled by standard phosphoramidite chemistry on solid phase at a 2.0 mmol scale, then conjugated to the GalNAc cluster in solution phase. The crude entered purification at 77.5 area% by RP-HPLC.<\/p>\n\n            <div class=\"stats-grid\">\n                <div class=\"stat-card\">\n                    <div class=\"stat-value is-green\">91.6<span>%<\/span><\/div>\n                    <p class=\"stat-label\">Product yield at specification, against 52.7% for the batch benchmark<\/p>\n                <\/div>\n                <div class=\"stat-card\">\n                    <div class=\"stat-value\">94.2<\/div>\n                    <p class=\"stat-label\">Area% product purity, enriched from a 77.5 area% crude<\/p>\n                <\/div>\n                <div class=\"stat-card\">\n                    <div class=\"stat-value is-green\">42.5<span>%<\/span><\/div>\n                    <p class=\"stat-label\">Smaller oligonucleotide synthesis step for the same daily output, in the scale-up estimate<\/p>\n                <\/div>\n                <div class=\"stat-card\">\n                    <div class=\"stat-value\">1.3<\/div>\n                    <p class=\"stat-label\">g\/L<sub>resin<\/sub>\/h productivity \u2014 essentially unchanged from the batch process<\/p>\n                <\/div>\n            <\/div>\n\n            <h3>What was run, and on what<\/h3>\n\n            <div class=\"two-col-grid\">\n                <div class=\"grid-card\">\n                    <h4>The batch benchmark<\/h4>\n                    <p>A preparative reversed-phase method on a <strong>YMC Triart Prep C8-S<\/strong> column, 150 \u00d7 10 mm i.d., 10 \u00b5m, 12 nm, at 45 \u00b0C. Mobile phases were acetonitrile against 0.2 M sodium acetate. Load was 2.52 g per liter of resin; higher loads gave insufficient product purity, though those data are not shown. Elution fractions were analyzed by HPLC for purity and by Nanodrop for concentration.<\/p>\n                <\/div>\n                <div class=\"grid-card\">\n                    <h4>The continuous process<\/h4>\n                    <p>MCSGP on a <strong>Contichrom CUBE 30<\/strong> with two columns of the same C8 resin and the same 15 cm bed height, at 45 \u00b0C, with UV recorded at 300 nm after each column outlet. The method was designed from the batch chromatogram using the MCSGP Wizard and run for six cycles, giving twelve product elutions.<\/p>\n                <\/div>\n            <\/div>\n\n            <p>Two deliberate simplifications distinguish the continuous method from its batch parent. The wash after loading was dropped, because the strong recycling step effectively substitutes for it. And regeneration, which the batch process performs after every elution, was applied only at the end of the MCSGP run \u2014 with no evident negative impact on process performance.<\/p>\n\n            <!-- Section 3 -->\n            <h2 id=\"what-yield-buys\">3. What the Yield Buys Upstream<\/h2>\n\n            <p>At a matched purity of roughly 94 area%, the batch process recovered 52.70% of the product loaded and MCSGP recovered 91.58%. The interesting question is what that difference is worth once the process is scaled, and the study answers it with a simplified estimate built around a single 60 cm batch column producing 1.3 kg of product a day.<\/p>\n\n            <div class=\"explorer\">\n                <div class=\"explorer-tabs\" role=\"group\" aria-label=\"Select a scale-up scenario\">\n                    <button type=\"button\" class=\"explorer-tab\" data-state=\"a\" aria-pressed=\"true\">Batch<small>Reference process<\/small><\/button>\n                    <button type=\"button\" class=\"explorer-tab\" data-state=\"b\" aria-pressed=\"false\">MCSGP, same feed<small>More product out<\/small><\/button>\n                    <button type=\"button\" class=\"explorer-tab\" data-state=\"c\" aria-pressed=\"false\">MCSGP, same output<small>Less synthesis in<\/small><\/button>\n                <\/div>\n\n                <div class=\"plot-scroll\">\n                <svg class=\"explorer-plot\" viewBox=\"0 0 760 352\" role=\"img\"\n                     aria-label=\"Daily mass flow for three scale-up scenarios. The reference batch process takes 3,200 grams of crude per day through one 60 cm column holding 42 liters of resin and returns 1,307 grams of product. MCSGP with the same feed takes the same 3,200 grams through two 60 cm columns holding 70 liters of resin in total and returns 2,270 grams. MCSGP sized for the same output takes 1,841 grams of crude through two 45 cm columns holding 42 liters of resin, the same as the batch process, and returns 1,307 grams. The crude contains 77.5 percent product in every case.\">\n                    <defs>\n                        <marker id=\"galArrow\" viewBox=\"0 0 10 10\" refX=\"8\" refY=\"5\" markerWidth=\"8\" markerHeight=\"8\" orient=\"auto-start-reverse\">\n                            <path d=\"M0,0 L10,5 L0,10 z\" fill=\"#718096\"\/>\n                        <\/marker>\n                    <\/defs>\n\n                    <line x1=\"100\" y1=\"292\" x2=\"660\" y2=\"292\" stroke=\"#cbd5e0\" stroke-width=\"1\"\/>\n                    <g text-anchor=\"middle\" fill=\"#718096\">\n                        <text x=\"168\" y=\"312\">Crude in<\/text>\n                        <text x=\"592\" y=\"312\">Product out<\/text>\n                    <\/g>\n\n                    <!-- STATE A: reference batch -->\n                    <g class=\"explorer-state is-on\" id=\"gal-st-a\">\n                        <rect x=\"130\" y=\"92\"  width=\"76\" height=\"45\"  fill=\"#D8DFE5\"\/>\n                        <rect x=\"130\" y=\"137\" width=\"76\" height=\"155\" fill=\"#8FA3B0\"\/>\n                        <line x1=\"126\" y1=\"137\" x2=\"210\" y2=\"137\" stroke=\"#4A5568\" stroke-width=\"1\" stroke-dasharray=\"4 3\"\/>\n                        <text x=\"122\" y=\"134\" text-anchor=\"end\" fill=\"#4A5568\">77.5% product<\/text>\n                        <text x=\"168\" y=\"84\" text-anchor=\"middle\" fill=\"#2d3748\">3,200 g<\/text>\n                        <rect x=\"363\" y=\"182\" width=\"36\" height=\"110\" rx=\"3\" fill=\"#EDF2F7\" stroke=\"#4A5568\" stroke-width=\"1.4\"\/>\n                        <text x=\"381\" y=\"312\" text-anchor=\"middle\" fill=\"#2d3748\">1 \u00d7 60 cm<\/text>\n                        <text x=\"381\" y=\"326\" text-anchor=\"middle\" fill=\"#718096\">42 L resin<\/text>\n                        <rect x=\"554\" y=\"210\" width=\"76\" height=\"82\" fill=\"#C0392B\"\/>\n                        <text x=\"592\" y=\"202\" text-anchor=\"middle\" fill=\"#C0392B\">1,307 g<\/text>\n                    <\/g>\n\n                    <!-- STATE B: MCSGP, same feed input -->\n                    <g class=\"explorer-state\" id=\"gal-st-b\">\n                        <rect x=\"130\" y=\"92\"  width=\"76\" height=\"45\"  fill=\"#D8DFE5\"\/>\n                        <rect x=\"130\" y=\"137\" width=\"76\" height=\"155\" fill=\"#8FA3B0\"\/>\n                        <line x1=\"126\" y1=\"137\" x2=\"210\" y2=\"137\" stroke=\"#4A5568\" stroke-width=\"1\" stroke-dasharray=\"4 3\"\/>\n                        <text x=\"122\" y=\"134\" text-anchor=\"end\" fill=\"#4A5568\">77.5% product<\/text>\n                        <text x=\"168\" y=\"84\" text-anchor=\"middle\" fill=\"#2d3748\">3,200 g<\/text>\n                        <rect x=\"341\" y=\"182\" width=\"36\" height=\"110\" rx=\"3\" fill=\"#EDF2F7\" stroke=\"#4A5568\" stroke-width=\"1.4\"\/>\n                        <rect x=\"385\" y=\"182\" width=\"36\" height=\"110\" rx=\"3\" fill=\"#EDF2F7\" stroke=\"#4A5568\" stroke-width=\"1.4\"\/>\n                        <text x=\"381\" y=\"312\" text-anchor=\"middle\" fill=\"#2d3748\">2 \u00d7 60 cm<\/text>\n                        <text x=\"381\" y=\"326\" text-anchor=\"middle\" fill=\"#718096\">70 L resin<\/text>\n                        <rect x=\"554\" y=\"150\" width=\"76\" height=\"142\" fill=\"#C0392B\"\/>\n                        <text x=\"592\" y=\"142\" text-anchor=\"middle\" fill=\"#C0392B\">2,270 g<\/text>\n                    <\/g>\n\n                    <!-- STATE C: MCSGP, same product output -->\n                    <g class=\"explorer-state\" id=\"gal-st-c\">\n                        <rect x=\"130\" y=\"177\" width=\"76\" height=\"26\" fill=\"#D8DFE5\"\/>\n                        <rect x=\"130\" y=\"203\" width=\"76\" height=\"89\" fill=\"#8FA3B0\"\/>\n                        <line x1=\"126\" y1=\"203\" x2=\"210\" y2=\"203\" stroke=\"#4A5568\" stroke-width=\"1\" stroke-dasharray=\"4 3\"\/>\n                        <text x=\"122\" y=\"200\" text-anchor=\"end\" fill=\"#4A5568\">77.5% product<\/text>\n                        <text x=\"168\" y=\"169\" text-anchor=\"middle\" fill=\"#2d3748\">1,841 g<\/text>\n                        <rect x=\"350\" y=\"182\" width=\"27\" height=\"110\" rx=\"3\" fill=\"#EDF2F7\" stroke=\"#4A5568\" stroke-width=\"1.4\"\/>\n                        <rect x=\"385\" y=\"182\" width=\"27\" height=\"110\" rx=\"3\" fill=\"#EDF2F7\" stroke=\"#4A5568\" stroke-width=\"1.4\"\/>\n                        <text x=\"381\" y=\"312\" text-anchor=\"middle\" fill=\"#2d3748\">2 \u00d7 45 cm<\/text>\n                        <text x=\"381\" y=\"326\" text-anchor=\"middle\" fill=\"#718096\">42 L resin<\/text>\n                        <rect x=\"554\" y=\"210\" width=\"76\" height=\"82\" fill=\"#C0392B\"\/>\n                        <text x=\"592\" y=\"202\" text-anchor=\"middle\" fill=\"#C0392B\">1,307 g<\/text>\n                    <\/g>\n\n                    <path d=\"M214,240 L330,240\" fill=\"none\" stroke=\"#718096\" stroke-width=\"1.4\" marker-end=\"url(#galArrow)\"\/>\n                    <path d=\"M432,240 L546,240\" fill=\"none\" stroke=\"#718096\" stroke-width=\"1.4\" marker-end=\"url(#galArrow)\"\/>\n                <\/svg>\n                <\/div>\n\n                <ul class=\"explorer-legend\">\n                    <li><i style=\"background:#8FA3B0\"><\/i>Product contained in the crude<\/li>\n                    <li><i style=\"background:#D8DFE5\"><\/i>Impurities in the crude<\/li>\n                    <li><i style=\"background:#C0392B\"><\/i>Product recovered<\/li>\n                <\/ul>\n\n                <div class=\"explorer-readout\">\n                    <p class=\"explorer-caption\" id=\"galCaption\">The reference process. A single 60 cm column takes 3,200 g of crude a day and returns <strong>1,307 g<\/strong> of product. The crude is 77.5% product, so the darker band is what was available to recover \u2014 and just over half of it makes it out. The synthesis has to be sized to cover the rest.<\/p>\n                    <dl class=\"explorer-nums\">\n                        <div class=\"explorer-num\"><dt>Crude in<\/dt><dd id=\"galFeed\">3,200<span> g\/day<\/span><\/dd><\/div>\n                        <div class=\"explorer-num\"><dt>Product out<\/dt><dd id=\"galProduct\">1,307<span> g\/day<\/span><\/dd><\/div>\n                        <div class=\"explorer-num\"><dt>Resin<\/dt><dd id=\"galResin\">42<span> L<\/span><\/dd><\/div>\n                        <div class=\"explorer-num\"><dt>Buffer<\/dt><dd id=\"galBuffer\">12,534<span> L\/day<\/span><\/dd><\/div>\n                    <\/dl>\n                <\/div>\n\n                <p class=\"explorer-note\">Scenarios as reported in Table 4 of the study, drawn to a single mass scale so the bars compare directly. These are simplified scale-up estimates rather than demonstrated production runs. Red marks the product, following the color convention used in Figures 1 and 2 of the paper.<\/p>\n            <\/div>\n\n            <p>The study is even-handed about the first scenario. Two 60 cm columns hold 70 L of resin between them against 42 L for the single batch column, so the extra output is not free. The authors point out that a batch process using the same total resin volume could deliver 2,614 g\/day \u2014 but would need twice the feed, 6,400 g\/day, to do it. The comparison that isolates the yield advantage is the third: same product out, same total resin, 42.5% less crude in.<\/p>\n\n            <div class=\"analogy-box\">\n                <h4>A Simple Analogy<\/h4>\n                <p>A distiller takes a heart cut from each run and sets aside the heads and tails at either end. Cut the heart narrow and the spirit is clean but there is little of it; cut it wide and there is more, carrying off-notes with it. It is the same choice a chromatographer makes at the fraction collector.<\/p>\n                <p>Distillers do not throw the heads and tails away \u2014 they go back into the next charge. That is what the recycling phases do here. Because the side fractions return to the second column, the heart can be cut narrow for purity without the yield being written off.<\/p>\n            <\/div>\n\n            <!-- Section 4 -->\n            <h2 id=\"designing\">4. Designing the Continuous Process<\/h2>\n\n            <p>MCSGP is designed by dividing a batch gradient chromatogram into four phases that make up one switch. The batch run is not merely a benchmark in this study; it is the design template, which is why the paper notes that the batch process should already reach the target purity with good productivity and acceptable buffer consumption before conversion begins. A batch process with a yield between 20% and 80% is described as a good starting point.<\/p>\n\n            <ul class=\"card-list\">\n                <li>\n                    <strong>P1 \u2014 columns in parallel<\/strong>\n                    Weakly adsorbing impurities elute from column 1 to waste, while column 2 is regenerated ready for loading.\n                <\/li>\n                <li>\n                    <strong>P2 \u2014 columns interconnected<\/strong>\n                    The mixture of weakly adsorbing impurities and product elutes from column 1 and is recycled, with in-line dilution, directly onto column 2.\n                <\/li>\n                <li>\n                    <strong>P3 \u2014 columns in parallel<\/strong>\n                    Pure product is collected from column 1 while fresh feed is loaded onto column 2. To hold a switch-to-switch steady state, the quantity of new feed applied is in equilibrium with the product removed \u2014 a load the software calculates from the batch chromatogram and fraction analysis.\n                <\/li>\n                <li>\n                    <strong>P4 \u2014 columns interconnected<\/strong>\n                    The mixture of product and strongly adsorbing impurities is recycled onto column 2 in the same way as P2. The columns then swap positions and the sequence repeats.\n                <\/li>\n            <\/ul>\n\n            <p>Setting the boundaries is where the judgment sits. P3 was positioned to achieve the target purity of more than 94 area%, calculated by the wizard, and P2 and P4 were positioned to fully recycle the lower-purity side fractions. Because N-1 and N+1 are structurally so close to the target and correspondingly hard to remove, their abundance across the peak was checked as well, so that the product collection window could be placed to minimize them.<\/p>\n\n            <p>The wizard then calculated the in-line dilution flow rates needed to dilute the modifier concentration of the eluate back to initial loading conditions \u2014 essential if product recycled from one column is to fully re-adsorb on the other \u2014 along with the loading parameters required to establish a cyclic steady state. It produced three methods: a startup that preloads the upstream column as a batch process would, a main method of six MCSGP cycles, and a shutdown consisting of one switch with a final product elution but no feeding.<\/p>\n\n            <!-- Section 5 -->\n            <h2 id=\"six-cycles\">5. What Six Cycles Showed<\/h2>\n\n            <p>Six cycles were run, each comprising two switches, for twelve product elutions in total. The question a continuous process has to answer is whether it settles, and whether anything accumulates while it does.<\/p>\n\n            <p>The height of the UV profile inside the P3 product collection window stayed consistent from cycle 1 to cycle 6, which indicates that material entering the system was in equilibrium with material leaving it. The visible changes across early cycles were in the P2 and P4 recycling windows, where impurities accumulate and take several cycles to reach steady state.<\/p>\n\n            <div class=\"callout\">\n                <p>Accumulation in the recycling windows is expected. It matters only if impurities build toward the product pool: <strong>if they accumulate in the direction of P3, purity falls cycle to cycle<\/strong> and the product can drop below specification. Placing the phase boundaries appropriately during design is what avoids or controls that.<\/p>\n            <\/div>\n\n            <p>All six cycles were therefore analyzed, by HPLC for purity and Nanodrop for concentration. Concentration held at around 2.7 g\/L and total purity stayed consistently above 94 area% across the run, and neither N-1 nor N+1 showed accumulation. On that basis the authors conclude that the process parameters are well designed with respect to product quality. Comparing the final pooled product from all six cycles against the input material shows enrichment from 77.5 area% to 94.2 area%.<\/p>\n\n            <h3>Why the reported figures are steady-state values<\/h3>\n\n            <p>One thing to read carefully. Startup and shutdown both cost performance, because the load applied during startup contains more product than is recovered in the final shutdown, and the shutdown does not recycle its side fractions \u2014 making it analogous to a batch run in terms of product loss. With only six cycles run, the authors restricted the performance comparison to steady state, meaning process performance if the number of cycles were infinite. They describe this as the best-case scenario, and state that in practice a run of more than 20 cycles would be very similar.<\/p>\n\n            <!-- Section 6 -->\n            <h2 id=\"results\">6. Full Results<\/h2>\n\n            <p class=\"scroll-hint\">Swipe the table sideways to see the feed, batch and MCSGP columns.<\/p>\n\n            <div class=\"table-scroll\">\n                <table class=\"data-table\">\n                    <thead>\n                        <tr>\n                            <th scope=\"col\">Metric<\/th>\n                            <th scope=\"col\">Feed<\/th>\n                            <th scope=\"col\">Batch<\/th>\n                            <th scope=\"col\">MCSGP<\/th>\n                        <\/tr>\n                    <\/thead>\n                    <tbody>\n                        <tr class=\"subhead\"><td colspan=\"4\">GalNAc-conjugated DNA-LNA gapmer <span>specification &gt;94 area% \u00b7 MCSGP figures are steady-state values<\/span><\/td><\/tr>\n                        <tr>\n                            <td>Product purity (area%)<\/td>\n                            <td data-label=\"Feed\">77.48<\/td><td data-label=\"Batch\">94.22<\/td><td data-label=\"MCSGP\">94.16<\/td>\n                        <\/tr>\n                        <tr>\n                            <td>Impurity N-1 (area%)<\/td>\n                            <td data-label=\"Feed\">3.36<\/td><td data-label=\"Batch\">0.93<\/td><td data-label=\"MCSGP\">0.87<\/td>\n                        <\/tr>\n                        <tr>\n                            <td>Impurity N+1 (area%)<\/td>\n                            <td data-label=\"Feed\">4.23<\/td><td data-label=\"Batch\">1.22<\/td><td data-label=\"MCSGP\" class=\"is-down\">1.41<\/td>\n                        <\/tr>\n                        <tr class=\"highlight\">\n                            <td>Yield (%)<\/td>\n                            <td data-label=\"Feed\">\u2014<\/td><td data-label=\"Batch\">52.70<\/td><td data-label=\"MCSGP\" class=\"is-up\">91.58<\/td>\n                        <\/tr>\n                        <tr>\n                            <td>Productivity (g\/L<sub>resin<\/sub>\/h)<\/td>\n                            <td data-label=\"Feed\">\u2014<\/td><td data-label=\"Batch\">1.30<\/td><td data-label=\"MCSGP\">1.33<\/td>\n                        <\/tr>\n                        <tr>\n                            <td>Buffer consumption (L\/g)<\/td>\n                            <td data-label=\"Feed\">\u2014<\/td><td data-label=\"Batch\">9.59<\/td><td data-label=\"MCSGP\">9.22<\/td>\n                        <\/tr>\n                        <tr>\n                            <td>Product concentration (g\/L)<\/td>\n                            <td data-label=\"Feed\">\u2014<\/td><td data-label=\"Batch\">3.25<\/td><td data-label=\"MCSGP\" class=\"is-down\">2.71<\/td>\n                        <\/tr>\n                        <tr>\n                            <td>Load per cycle (g\/L resin)<\/td>\n                            <td data-label=\"Feed\">\u2014<\/td><td data-label=\"Batch\">2.52<\/td><td data-label=\"MCSGP\">1.83<\/td>\n                        <\/tr>\n                    <\/tbody>\n                <\/table>\n            <\/div>\n            <p class=\"table-note\">Values from Table 3 of the study, rearranged. Weldon, Lill, Olbrich, Schmidt and M\u00fcller-Sp\u00e4th, <em>J. Chromatogr. A<\/em> 2022, 1663, 462734, published under CC BY-NC-ND 4.0.<\/p>\n\n            <h3>Where the two processes differ, and where they do not<\/h3>\n\n            <p>Yield is the change. Almost everything else is close to unchanged: purity is matched by design at roughly 94 area%, productivity is 1.33 against 1.30 g\/L<sub>resin<\/sub>\/h, and buffer consumption is 9.22 against 9.59 L per gram of product. N-1 came out marginally lower in MCSGP and N+1 marginally higher.<\/p>\n\n            <p>Two figures moved the wrong way, and both are acknowledged. Product concentration was 17% lower in MCSGP, which the authors flag as leaving scope to reduce eluate volumes. And the MCSGP load was lower than the batch load, 1.83 against 2.52 g per liter of resin \u2014 not because that was optimal but because load optimization was not part of the study. Batch experiments qualitatively indicated that small increases in load led to further yield loss, which suggests the binding capacity of the resin was already being used to a great extent.<\/p>\n\n            <h3>Scale-up scenarios<\/h3>\n\n            <p class=\"scroll-hint\">Swipe the table sideways to see all three scenarios.<\/p>\n\n            <div class=\"table-scroll\">\n                <table class=\"data-table\">\n                    <thead>\n                        <tr>\n                            <th scope=\"col\">Parameter<\/th>\n                            <th scope=\"col\">Batch<\/th>\n                            <th scope=\"col\">MCSGP, same feed<\/th>\n                            <th scope=\"col\">MCSGP, same output<\/th>\n                        <\/tr>\n                    <\/thead>\n                    <tbody>\n                        <tr>\n                            <td>Feed input at 77.5% purity (g\/day)<\/td>\n                            <td data-label=\"Batch\">3,200<\/td><td data-label=\"MCSGP, same feed\">3,200<\/td>\n                            <td data-label=\"MCSGP, same output\" class=\"is-up\">1,841<\/td>\n                        <\/tr>\n                        <tr class=\"highlight\">\n                            <td>Production amount (g\/day)<\/td>\n                            <td data-label=\"Batch\">1,307<\/td><td data-label=\"MCSGP, same feed\" class=\"is-up\">2,270<\/td>\n                            <td data-label=\"MCSGP, same output\">1,307<\/td>\n                        <\/tr>\n                        <tr>\n                            <td>Column inner diameter<\/td>\n                            <td data-label=\"Batch\">60 cm<\/td><td data-label=\"MCSGP, same feed\">2 \u00d7 60 cm<\/td>\n                            <td data-label=\"MCSGP, same output\">2 \u00d7 45 cm<\/td>\n                        <\/tr>\n                        <tr>\n                            <td>Required resin volume<\/td>\n                            <td data-label=\"Batch\">42 L<\/td><td data-label=\"MCSGP, same feed\">2 \u00d7 35 L<\/td>\n                            <td data-label=\"MCSGP, same output\">2 \u00d7 21 L<\/td>\n                        <\/tr>\n                        <tr>\n                            <td>Pump size on skid<\/td>\n                            <td data-label=\"Batch\">15 L\/min<\/td><td data-label=\"MCSGP, same feed\">16 L\/min<\/td>\n                            <td data-label=\"MCSGP, same output\">8 L\/min<\/td>\n                        <\/tr>\n                        <tr>\n                            <td>Daily buffer demand<\/td>\n                            <td data-label=\"Batch\">12,534 L<\/td><td data-label=\"MCSGP, same feed\">20,929 L<\/td>\n                            <td data-label=\"MCSGP, same output\">12,050 L<\/td>\n                        <\/tr>\n                    <\/tbody>\n                <\/table>\n            <\/div>\n            <p class=\"table-note\">Values from Table 4 of the study, rearranged. These are simplified scale-up estimates, not demonstrated production runs.<\/p>\n\n            <!-- Section 7 -->\n            <h2 id=\"implications\">7. What It Means for Manufacturing<\/h2>\n\n            <p>The scale-up estimate is where the yield figure becomes an operational argument. Producing the same 1.3 kg of product a day from 1,841 g of crude instead of 3,200 g means the preceding steps, including oligonucleotide synthesis, can be downscaled by 42.5% \u2014 or, read the other way, the same amount of product can be made 42.5% faster.<\/p>\n\n            <p>The equipment picture is mixed. In the same-output scenario the two MCSGP columns hold the same total resin as the single batch column, at a smaller inner diameter, and the smaller diameter allows smaller pumps on the skid to reach the same linear flow rate. Because MCSGP uses similar equipment with a comparable footprint to batch chromatography, the authors consider it feasible to equip existing facilities.<\/p>\n\n            <p>Beyond yield, the paper lists three further advantages. Re-chromatography of side fractions is no longer required, which removes a step that improves batch yields but decreases overall productivity, increases buffer consumption and demands manual handling and storage space. The analytical burden can fall: a stable batch process requires a minimum of three pools to be collected and analyzed, more when product collection is less robust and more again after re-chromatography, whereas MCSGP typically generates a single product pool per cycle and it may be sufficient to pool and analyze every few cycles. And less human intervention is needed during production, so fewer personnel are required.<\/p>\n\n            <p>Against that, the paper names its own trade-off plainly: elevated equipment and validation complexity, since an MCSGP setup uses two columns and approximately doubles the amount of hardware.<\/p>\n\n            <p>An equivalent conversion in anion exchange mode, on a conjugated siRNA sense strand, is reported in <a href=\"https:\/\/www.chromacon.com\/en\/articles\/continuous-anion-exchange-purification-of-a-conjugated-sirna\/\">a companion study in this series<\/a>. For the platform view of where MCSGP fits in an oligonucleotide workflow, see <a href=\"https:\/\/www.chromacon.com\/en\/applications\/oligonucleotides\/\">oligonucleotide purification with MCSGP<\/a>.<\/p>\n\n            <div class=\"limits-box\">\n                <h4>What the study does not claim<\/h4>\n                <ul class=\"simple-list\">\n                    <li><strong>The MCSGP figures are steady-state values.<\/strong> Only six cycles were run, and the comparison excludes the performance impact of startup and shutdown. The authors describe this as the best-case scenario \u2014 process performance as if the cycle count were infinite \u2014 and state that a run of more than 20 cycles would be very similar. Reported yield, productivity and buffer consumption should be read on that basis.<\/li>\n                    <li><strong>Load was not optimized.<\/strong> MCSGP ran at 1.83 g\/L resin against 2.52 for batch. Load optimization was not part of the study, and batch experiments qualitatively suggested small load increases cost yield.<\/li>\n                    <li><strong>Product concentration was 17% lower.<\/strong> The authors identify this as scope for method improvement to reduce eluate volumes.<\/li>\n                    <li><strong>N+1 was slightly higher.<\/strong> 1.41 area% in the MCSGP pool against 1.22 in the batch pool, although N-1 was slightly lower at 0.87 against 0.93.<\/li>\n                    <li><strong>More hardware, more validation.<\/strong> The paper states the trade-off directly: two columns and approximately double the hardware components bring elevated equipment and validation complexity.<\/li>\n                    <li><strong>Scale-up is an estimate.<\/strong> The three scenarios are a simplified calculation from the measured performance, not demonstrated production runs.<\/li>\n                    <li><strong>Funding and affiliations.<\/strong> The study was funded by F. Hoffmann-La Roche Ltd., where three of the five authors work; the other two are at YMC ChromaCon, which commercializes MCSGP. The authors declare no known competing financial interests or personal relationships that could have appeared to influence the work.<\/li>\n                <\/ul>\n            <\/div>\n\n            <!-- Section 8 -->\n            <h2 id=\"glossary\">8. Key Terms in This Study (Glossary)<\/h2>\n\n            <table class=\"glossary-table\">\n                <thead>\n                    <tr>\n                        <th>Term<\/th>\n                        <th>Definition<\/th>\n                    <\/tr>\n                <\/thead>\n                <tbody>\n                    <tr>\n                        <td>Center cut<\/td>\n                        <td>The central portion of a batch elution peak that meets purity specification and is collected as product. The fractions either side of it are the side fractions.<\/td>\n                    <\/tr>\n                    <tr>\n                        <td>Closed-loop recycling (CLR)<\/td>\n                        <td>A single-column recycling technique in which the chromatographic profile passes back through a pump each cycle. Limited to isocratic operation, and the repeated pump passage may affect product stability.<\/td>\n                    <\/tr>\n                    <tr>\n                        <td>Cyclic steady state<\/td>\n                        <td>The condition in which each MCSGP cycle reproduces the previous one, with material entering the system in equilibrium with material leaving it.<\/td>\n                    <\/tr>\n                    <tr>\n                        <td>Gapmer<\/td>\n                        <td>An antisense oligonucleotide design with a central DNA segment flanked by modified nucleotides \u2014 here locked nucleic acid \u2014 which confer stability while the central gap remains active against the RNA target.<\/td>\n                    <\/tr>\n                    <tr>\n                        <td>GalNAc cluster<\/td>\n                        <td>An N-acetylgalactosamine ligand conjugated to an oligonucleotide, providing transport into hepatocytes by endocytosis and so enabling treatments targeting diseases of the liver.<\/td>\n                    <\/tr>\n                    <tr>\n                        <td>In-line dilution<\/td>\n                        <td>Dilution of the recycled stream between interconnected columns, returning the modifier concentration to initial loading conditions so the recycled product fully re-adsorbs on the downstream column.<\/td>\n                    <\/tr>\n                    <tr>\n                        <td>LNA<\/td>\n                        <td>Locked nucleic acid, a nucleotide analogue with a constrained ribose ring, used to raise the stability and target affinity of synthetic oligonucleotides.<\/td>\n                    <\/tr>\n                    <tr>\n                        <td>Longmer<\/td>\n                        <td>A sequence longer than the target, such as a lambdamer or branchmer, arising from impurities in the starting raw materials.<\/td>\n                    <\/tr>\n                    <tr>\n                        <td>MCSGP<\/td>\n                        <td>Multicolumn Countercurrent Solvent Gradient Purification: a continuous twin-column chromatography process that recycles impure side fractions internally, so that only product meeting specification leaves the system.<\/td>\n                    <\/tr>\n                    <tr>\n                        <td>N-1 and N+1<\/td>\n                        <td>Impurities one nucleotide shorter and one nucleotide longer than the target. Structurally very close to the product and correspondingly difficult to separate from it.<\/td>\n                    <\/tr>\n                    <tr>\n                        <td>P1 to P4<\/td>\n                        <td>The four phases of one MCSGP switch: weak impurities to waste, weak-side recycling, product collection with simultaneous feeding, and strong-side recycling.<\/td>\n                    <\/tr>\n                    <tr>\n                        <td>Phosphorothioate<\/td>\n                        <td>A backbone modification replacing an oxygen with sulfur to resist degradation. Highly phosphorothiolated sequences are noted as especially prone to impurities co-eluting with the product.<\/td>\n                    <\/tr>\n                    <tr>\n                        <td>Re-chromatography<\/td>\n                        <td>Re-running lower-purity side fractions to recover further product. It raises batch yield but lowers productivity, raises buffer consumption and adds handling, storage and quality control.<\/td>\n                    <\/tr>\n                    <tr>\n                        <td>Shortmer<\/td>\n                        <td>A sequence missing one or more nucleotides, denoted N-x, produced by incomplete chemical coupling during synthesis.<\/td>\n                    <\/tr>\n                    <tr>\n                        <td>Startup and shutdown<\/td>\n                        <td>The methods that begin and end an MCSGP run. Startup preloads the first column; shutdown performs a final elution without feeding and without recycling its side fractions.<\/td>\n                    <\/tr>\n                    <tr>\n                        <td>Steady-state recycling (SSR)<\/td>\n                        <td>A single-column recycling technique related to closed-loop recycling, sharing the same limitation that the profile passes through a pump each cycle.<\/td>\n                    <\/tr>\n                    <tr>\n                        <td>Switch and cycle<\/td>\n                        <td>A switch is one product elution from one column, made up of phases P1 to P4. One cycle is two switches, with one product elution from each of the twin columns.<\/td>\n                    <\/tr>\n                    <tr>\n                        <td>Yield\u2013purity trade-off<\/td>\n                        <td>The inverse relationship in batch chromatography between how much product is collected and how pure the pool is, set by separation performance and where the collection boundaries are drawn.<\/td>\n                    <\/tr>\n                <\/tbody>\n            <\/table>\n\n            <!-- Section 9 - FAQ -->\n            <h2 id=\"faq\">9. Frequently Asked Questions<\/h2>\n\n            <div class=\"faq-section\">\n                <div class=\"faq-item\">\n                    <button class=\"faq-question\" aria-expanded=\"false\">\n                        Can MCSGP purify GalNAc-conjugated oligonucleotides?\n                    <\/button>\n                    <div class=\"faq-answer\">\n                        <div class=\"faq-answer-inner\">\n                            <p>Yes. This study reports the first evaluation of MCSGP for a GalNAc-cluster-conjugated DNA-LNA gapmer in reversed-phase mode. Starting from an established batch method on the same C8 resin and the same acetonitrile and sodium acetate mobile phases, the continuous process reached 91.58% yield at 94.16 area% purity, against 52.70% yield at 94.22 area% for the batch benchmark.<\/p>\n                        <\/div>\n                    <\/div>\n                <\/div>\n\n                <div class=\"faq-item\">\n                    <button class=\"faq-question\" aria-expanded=\"false\">\n                        How much yield does MCSGP add for oligonucleotide purification?\n                    <\/button>\n                    <div class=\"faq-answer\">\n                        <div class=\"faq-answer-inner\">\n                            <p>In this case, yield rose from 52.70% to 91.58% at matched purity \u2014 a relative improvement of 73.6%. The size of the gain depends on the batch process it starts from: a batch process with a yield between 20% and 80% is described as a good starting point for MCSGP development, so a batch method that already recovers most of its product has less room to improve.<\/p>\n                        <\/div>\n                    <\/div>\n                <\/div>\n\n                <div class=\"faq-item\">\n                    <button class=\"faq-question\" aria-expanded=\"false\">\n                        What does a higher purification yield actually save?\n                    <\/button>\n                    <div class=\"faq-answer\">\n                        <div class=\"faq-answer-inner\">\n                            <p>Synthesis. Oligonucleotide synthesis is the major cost driver in production, and the scale of it is set by how much product the purification is expected to lose. In the study&#8217;s scale-up estimate, the same 1,307 g of daily product could be made from 1,841 g of crude instead of 3,200 g \u2014 allowing the preceding steps, including synthesis, to be downscaled by 42.5%. Read the other way, the same output could be produced 42.5% faster.<\/p>\n                        <\/div>\n                    <\/div>\n                <\/div>\n\n                <div class=\"faq-item\">\n                    <button class=\"faq-question\" aria-expanded=\"false\">\n                        Does the yield gain cost productivity or buffer?\n                    <\/button>\n                    <div class=\"faq-answer\">\n                        <div class=\"faq-answer-inner\">\n                            <p>Not in this study. Productivity was 1.33 g\/L<sub>resin<\/sub>\/h for MCSGP against 1.30 for batch, and buffer consumption 9.22 L\/g against 9.59 \u2014 both essentially unchanged. Two figures did move the wrong way: product concentration was 17% lower, and the N+1 impurity was marginally higher at 1.41 area% against 1.22.<\/p>\n                        <\/div>\n                    <\/div>\n                <\/div>\n\n                <div class=\"faq-item\">\n                    <button class=\"faq-question\" aria-expanded=\"false\">\n                        How is MCSGP different from recycling techniques like SSR or CLR?\n                    <\/button>\n                    <div class=\"faq-answer\">\n                        <div class=\"faq-answer-inner\">\n                            <p>In steady-state recycling and closed-loop recycling, the chromatographic profile passes through a pump every cycle, which is possibly detrimental to product stability and limits those processes to isocratic operation. In MCSGP the recycled side fractions are fully adsorbed onto a second column and then subjected to renewed gradient elution, so product and impurities pass through a pump only during the load step. The authors argue this makes MCSGP well suited to complex molecules such as oligonucleotides.<\/p>\n                        <\/div>\n                    <\/div>\n                <\/div>\n\n                <div class=\"faq-item\">\n                    <button class=\"faq-question\" aria-expanded=\"false\">\n                        How many MCSGP cycles are needed?\n                    <\/button>\n                    <div class=\"faq-answer\">\n                        <div class=\"faq-answer-inner\">\n                            <p>More than were run here, to see the full benefit. Startup and shutdown both reduce overall performance, because the startup load contains more product than the shutdown recovers and the shutdown does not recycle its side fractions. The paper states that running more than 20 cycles is ideal to minimize their relative impact. Because only six cycles were run in this study, the reported performance figures describe the steady state rather than a whole-run average.<\/p>\n                        <\/div>\n                    <\/div>\n                <\/div>\n\n                <div class=\"faq-item\">\n                    <button class=\"faq-question\" aria-expanded=\"false\">\n                        Does switching to MCSGP mean changing equipment or chemistry?\n                    <\/button>\n                    <div class=\"faq-answer\">\n                        <div class=\"faq-answer-inner\">\n                            <p>The chemistry stays the same. This study used the same C8 stationary phase and the same acetonitrile and sodium acetate mobile phases as the batch method, with two deliberate simplifications: the wash after loading was dropped, since the strong recycling step substitutes for it, and regeneration was applied only at the end of the run rather than after every elution. The equipment is the trade-off \u2014 MCSGP uses two columns and approximately doubles the hardware components, bringing elevated equipment and validation complexity, though the footprint is comparable to batch chromatography.<\/p>\n                        <\/div>\n                    <\/div>\n                <\/div>\n\n                <div class=\"faq-item\">\n                    <button class=\"faq-question\" aria-expanded=\"false\">\n                        Is the full article freely available?\n                    <\/button>\n                    <div class=\"faq-answer\">\n                        <div class=\"faq-answer-inner\">\n                            <p>Yes. It is published open access in <em>Journal of Chromatography A<\/em> and can be read and downloaded free of charge. Note that it carries a CC BY-NC-ND 4.0 license rather than the CC BY used by many open-access papers, so reuse is limited to non-commercial purposes and does not extend to derivative works. Supplementary material covering the specific batch and MCSGP method parameters and the analytical gradient is available alongside the article.<\/p>\n                        <\/div>\n                    <\/div>\n                <\/div>\n            <\/div>\n        <\/article>\n    <\/div>\n\n\n\n<!-- Structured data. Replace the two page URLs below with this page's permalink. -->\n<script type=\"application\/ld+json\">\n{\n  \"@context\": \"https:\/\/schema.org\",\n  \"@graph\": [\n    {\n      \"@type\": \"ScholarlyArticle\",\n      \"@id\": \"https:\/\/doi.org\/10.1016\/j.chroma.2021.462734\",\n      \"headline\": \"Purification of a GalNAc-cluster-conjugated oligonucleotide by reversed-phase twin-column continuous chromatography\",\n      \"name\": \"Purification of a GalNAc-cluster-conjugated oligonucleotide by reversed-phase twin-column continuous chromatography\",\n      \"datePublished\": \"2021-12-12\",\n      \"inLanguage\": \"en\",\n      \"isAccessibleForFree\": true,\n      \"license\": \"https:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/\",\n      \"url\": \"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0021967321008566\",\n      \"sameAs\": \"https:\/\/doi.org\/10.1016\/j.chroma.2021.462734\",\n      \"identifier\": [{ \"@type\": \"PropertyValue\", \"propertyID\": \"DOI\", \"value\": \"10.1016\/j.chroma.2021.462734\" }],\n      \"keywords\": [\"oligonucleotide\",\"purification\",\"MCSGP\",\"continuous manufacturing\",\"scale-up\",\"GalNAc\",\"gapmer\",\"reversed-phase chromatography\",\"twin-column chromatography\",\"yield-purity trade-off\"],\n      \"abstract\": \"MCSGP was applied to the reversed-phase purification of a GalNAc-cluster-conjugated DNA-LNA gapmer oligonucleotide therapeutic using a twin-column chromatography system. Based on a batch process as a starting point, MCSGP was designed, optimized and compared with the batch process for process performance and scale-up requirements. Product yields increased from 52.7 percent using batch chromatography to 91.5 percent using MCSGP, with purity, productivity and buffer consumption otherwise comparable. In a manufacturing scenario, use of MCSGP would allow the downscaling of oligonucleotide synthesis by 42.5 percent.\",\n      \"author\": [\n        { \"@type\": \"Person\", \"name\": \"Richard Weldon\", \"affiliation\": { \"@type\": \"Organization\", \"name\": \"ChromaCon AG\" } },\n        { \"@type\": \"Person\", \"name\": \"J\u00f6rg Lill\", \"affiliation\": { \"@type\": \"Organization\", \"name\": \"F. Hoffmann-La Roche Ltd.\" } },\n        { \"@type\": \"Person\", \"name\": \"Martin Olbrich\", \"affiliation\": { \"@type\": \"Organization\", \"name\": \"F. Hoffmann-La Roche Ltd.\" } },\n        { \"@type\": \"Person\", \"name\": \"Pascal Schmidt\", \"affiliation\": { \"@type\": \"Organization\", \"name\": \"F. Hoffmann-La Roche Ltd.\" } },\n        { \"@type\": \"Person\", \"name\": \"Thomas M\u00fcller-Sp\u00e4th\", \"affiliation\": { \"@type\": \"Organization\", \"name\": \"ChromaCon AG\" } }\n      ],\n      \"publisher\": { \"@type\": \"Organization\", \"name\": \"Elsevier\" },\n      \"isPartOf\": {\n        \"@type\": \"PublicationVolume\",\n        \"volumeNumber\": \"1663\",\n        \"isPartOf\": { \"@type\": \"Periodical\", \"name\": \"Journal of Chromatography A\", \"issn\": \"0021-9673\" }\n      },\n      \"pagination\": \"462734\"\n    },\n    {\n      \"@type\": \"WebPage\",\n      \"@id\": \"https:\/\/www.chromacon.com\/en\/articles\/galnac-oligonucleotide-purification-mcsgp\/\",\n      \"url\": \"https:\/\/www.chromacon.com\/en\/articles\/galnac-oligonucleotide-purification-mcsgp\/\",\n      \"name\": \"Breaking the Yield\u2013Purity Trade-Off for GalNAc Oligonucleotides\",\n      \"description\": \"How a reversed-phase batch purification of a GalNAc-cluster-conjugated gapmer was converted to twin-column MCSGP, raising yield from 52.7 to 91.6 percent at unchanged purity.\",\n      \"mainEntity\": { \"@id\": \"https:\/\/doi.org\/10.1016\/j.chroma.2021.462734\" }\n    },\n    {\n      \"@type\": \"FAQPage\",\n      \"mainEntity\": [\n        { \"@type\": \"Question\", \"name\": \"Can MCSGP purify GalNAc-conjugated oligonucleotides?\", \"acceptedAnswer\": { \"@type\": \"Answer\", \"text\": \"Yes. This study reports the first evaluation of MCSGP for a GalNAc-cluster-conjugated DNA-LNA gapmer in reversed-phase mode. Starting from an established batch method on the same C8 resin and the same acetonitrile and sodium acetate mobile phases, the continuous process reached 91.58% yield at 94.16 area% purity, against 52.70% yield at 94.22 area% for the batch benchmark.\" } },\n        { \"@type\": \"Question\", \"name\": \"How much yield does MCSGP add for oligonucleotide purification?\", \"acceptedAnswer\": { \"@type\": \"Answer\", \"text\": \"In this case, yield rose from 52.70% to 91.58% at matched purity \u2014 a relative improvement of 73.6%. The size of the gain depends on the batch process it starts from: a batch process with a yield between 20% and 80% is described as a good starting point for MCSGP development, so a batch method that already recovers most of its product has less room to improve.\" } },\n        { \"@type\": \"Question\", \"name\": \"What does a higher purification yield actually save?\", \"acceptedAnswer\": { \"@type\": \"Answer\", \"text\": \"Synthesis. Oligonucleotide synthesis is the major cost driver in production, and the scale of it is set by how much product the purification is expected to lose. In the study's scale-up estimate, the same 1,307 g of daily product could be made from 1,841 g of crude instead of 3,200 g \u2014 allowing the preceding steps, including synthesis, to be downscaled by 42.5%. Read the other way, the same output could be produced 42.5% faster.\" } },\n        { \"@type\": \"Question\", \"name\": \"Does the yield gain cost productivity or buffer?\", \"acceptedAnswer\": { \"@type\": \"Answer\", \"text\": \"Not in this study. Productivity was 1.33 g\/L resin\/h for MCSGP against 1.30 for batch, and buffer consumption 9.22 L\/g against 9.59 \u2014 both essentially unchanged. Two figures did move the wrong way: product concentration was 17% lower, and the N+1 impurity was marginally higher at 1.41 area% against 1.22.\" } },\n        { \"@type\": \"Question\", \"name\": \"How is MCSGP different from recycling techniques like SSR or CLR?\", \"acceptedAnswer\": { \"@type\": \"Answer\", \"text\": \"In steady-state recycling and closed-loop recycling, the chromatographic profile passes through a pump every cycle, which is possibly detrimental to product stability and limits those processes to isocratic operation. In MCSGP the recycled side fractions are fully adsorbed onto a second column and then subjected to renewed gradient elution, so product and impurities pass through a pump only during the load step. The authors argue this makes MCSGP well suited to complex molecules such as oligonucleotides.\" } },\n        { \"@type\": \"Question\", \"name\": \"How many MCSGP cycles are needed?\", \"acceptedAnswer\": { \"@type\": \"Answer\", \"text\": \"More than were run here, to see the full benefit. Startup and shutdown both reduce overall performance, because the startup load contains more product than the shutdown recovers and the shutdown does not recycle its side fractions. The paper states that running more than 20 cycles is ideal to minimize their relative impact. Because only six cycles were run in this study, the reported performance figures describe the steady state rather than a whole-run average.\" } },\n        { \"@type\": \"Question\", \"name\": \"Does switching to MCSGP mean changing equipment or chemistry?\", \"acceptedAnswer\": { \"@type\": \"Answer\", \"text\": \"The chemistry stays the same. This study used the same C8 stationary phase and the same acetonitrile and sodium acetate mobile phases as the batch method, with two deliberate simplifications: the wash after loading was dropped, since the strong recycling step substitutes for it, and regeneration was applied only at the end of the run rather than after every elution. The equipment is the trade-off \u2014 MCSGP uses two columns and approximately doubles the hardware components, bringing elevated equipment and validation complexity, though the footprint is comparable to batch chromatography.\" } },\n        { \"@type\": \"Question\", \"name\": \"Is the full article freely available?\", \"acceptedAnswer\": { \"@type\": \"Answer\", \"text\": \"Yes. It is published open access in Journal of Chromatography A and can be read and downloaded free of charge. Note that it carries a CC BY-NC-ND 4.0 license rather than the CC BY used by many open-access papers, so reuse is limited to non-commercial purposes and does not extend to derivative works. Supplementary material covering the specific batch and MCSGP method parameters and the analytical gradient is available alongside the article.\" } }\n      ]\n    },\n    {\n      \"@type\": \"BreadcrumbList\",\n      \"itemListElement\": [\n        {\"@type\":\"ListItem\",\"position\":1,\"name\":\"Research Highlights\",\"item\":\"https:\/\/www.chromacon.com\/en\/research-highlights\/\"},\n        {\"@type\":\"ListItem\",\"position\":2,\"name\":\"Breaking the Yield\u2013Purity Trade-Off for GalNAc Oligonucleotides\",\"item\":\"https:\/\/www.chromacon.com\/en\/articles\/galnac-oligonucleotide-purification-mcsgp\/\"}\n      ]\n    }\n  ]\n}\n<\/script>\n\n\n\n<script>\n        \/\/ FAQ Accordion functionality\n        document.querySelectorAll('.faq-question').forEach(button => {\n            button.addEventListener('click', () => {\n                const faqItem = button.parentElement;\n                const isOpen = faqItem.classList.contains('open');\n\n                \/\/ Close all other items\n                document.querySelectorAll('.faq-item').forEach(item => {\n                    item.classList.remove('open');\n                    item.querySelector('.faq-question').setAttribute('aria-expanded', 'false');\n                });\n\n                \/\/ Toggle current item\n                if (!isOpen) {\n                    faqItem.classList.add('open');\n                    button.setAttribute('aria-expanded', 'true');\n                }\n            });\n        });\n\n        \/\/ Mobile TOC toggle\n        const mobileToc = document.getElementById('mobileToc');\n        const mobileTocToggle = mobileToc.querySelector('.mobile-toc-toggle');\n\n        mobileTocToggle.addEventListener('click', () => {\n            const isOpen = mobileToc.classList.contains('open');\n            mobileToc.classList.toggle('open');\n            mobileTocToggle.setAttribute('aria-expanded', !isOpen);\n        });\n\n        \/\/ Close mobile TOC when clicking a link\n        document.querySelectorAll('.mobile-toc-list a').forEach(link => {\n            link.addEventListener('click', () => {\n                mobileToc.classList.remove('open');\n                mobileTocToggle.setAttribute('aria-expanded', 'false');\n            });\n        });\n\n        \/\/ Highlight active TOC item on scroll (desktop)\n        const observerOptions = {\n            rootMargin: '-100px 0px -70% 0px',\n            threshold: 0\n        };\n\n        const observer = new IntersectionObserver(entries => {\n            entries.forEach(entry => {\n                const id = entry.target.getAttribute('id');\n                const tocLink = document.querySelector(`.toc-list a[href=\"#${id}\"]`);\n\n                if (tocLink) {\n                    if (entry.isIntersecting) {\n                        document.querySelectorAll('.toc-list a').forEach(link => {\n                            link.classList.remove('active');\n                        });\n                        tocLink.classList.add('active');\n                    }\n                }\n            });\n        }, observerOptions);\n\n        document.querySelectorAll('h2[id]').forEach(section => {\n            observer.observe(section);\n        });\n\n        \/\/ Scale-up scenario explorer (Section 3)\n        (function () {\n            const DATA = {\n                a: {\n                    feed: '3,200<span> g\/day<\/span>', product: '1,307<span> g\/day<\/span>',\n                    resin: '42<span> L<\/span>', buffer: '12,534<span> L\/day<\/span>',\n                    caption: 'The reference process. A single 60 cm column takes 3,200 g of crude a day and ' +\n                             'returns <strong>1,307 g<\/strong> of product. The crude is 77.5% product, so the ' +\n                             'darker band is what was available to recover \u2014 and just over half of it makes it ' +\n                             'out. The synthesis has to be sized to cover the rest.'\n                },\n                b: {\n                    feed: '3,200<span> g\/day<\/span>', product: '2,270<span> g\/day<\/span>',\n                    resin: '70<span> L \u00b7 2 \u00d7 35<\/span>', buffer: '20,929<span> L\/day<\/span>',\n                    caption: 'Same crude, two columns. Recovering 91.58% instead of 52.70% turns the same 3,200 g ' +\n                             'of daily crude into <strong>2,270 g<\/strong> of product, a relative gain of 73.6%. ' +\n                             'It is not free: two 60 cm columns hold 70 L of resin against 42 L for the batch ' +\n                             'column, and daily buffer demand rises accordingly.'\n                },\n                c: {\n                    feed: '1,841<span> g\/day<\/span>', product: '1,307<span> g\/day<\/span>',\n                    resin: '42<span> L \u00b7 2 \u00d7 21<\/span>', buffer: '12,050<span> L\/day<\/span>',\n                    caption: 'Same product, less crude. Two 45 cm columns holding the same total resin as the ' +\n                             'single batch column produce the same <strong>1,307 g<\/strong> a day from 1,841 g ' +\n                             'of crude \u2014 <strong>42.5% less<\/strong>, which is 42.5% less oligonucleotide ' +\n                             'synthesis. The smaller diameter also allows smaller pumps on the skid.'\n                }\n            };\n\n            const tabs = document.querySelectorAll('.explorer-tab');\n            if (!tabs.length) return;\n\n            const caption = document.getElementById('galCaption');\n            const fields = {\n                feed:    document.getElementById('galFeed'),\n                product: document.getElementById('galProduct'),\n                resin:   document.getElementById('galResin'),\n                buffer:  document.getElementById('galBuffer')\n            };\n\n            function show(key) {\n                const d = DATA[key];\n                if (!d) return;\n\n                ['a', 'b', 'c'].forEach(k => {\n                    const g = document.getElementById('gal-st-' + k);\n                    if (g) g.classList.toggle('is-on', k === key);\n                });\n\n                Object.keys(fields).forEach(f => {\n                    if (fields[f]) fields[f].innerHTML = d[f];\n                });\n\n                if (caption) caption.innerHTML = d.caption;\n\n                tabs.forEach(t => t.setAttribute('aria-pressed', String(t.dataset.state === key)));\n            }\n\n            tabs.forEach(t => t.addEventListener('click', () => show(t.dataset.state)));\n        })();\n    <\/script>\n","protected":false},"excerpt":{"rendered":"<p>Purification is where oligonucleotide yield is lost, and the synthesis has to be sized to cover it. Converting a reversed-phase batch method for a GalNAc-conjugated gapmer to twin-column MCSGP lifted recovery from 52.7% to 91.6% at the same purity \u2014 enough to shrink the synthesis step by 42.5% for the same daily output.<\/p>\n","protected":false},"featured_media":6907,"comment_status":"closed","ping_status":"closed","template":"","article_category":[300],"class_list":["post-17977","articles","type-articles","status-publish","has-post-thumbnail","","article_category-oligo-purification"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.chromacon.com\/en\/wp-json\/wp\/v2\/articles\/17977","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.chromacon.com\/en\/wp-json\/wp\/v2\/articles"}],"about":[{"href":"https:\/\/www.chromacon.com\/en\/wp-json\/wp\/v2\/types\/articles"}],"replies":[{"embeddable":true,"href":"https:\/\/www.chromacon.com\/en\/wp-json\/wp\/v2\/comments?post=17977"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.chromacon.com\/en\/wp-json\/wp\/v2\/media\/6907"}],"wp:attachment":[{"href":"https:\/\/www.chromacon.com\/en\/wp-json\/wp\/v2\/media?parent=17977"}],"wp:term":[{"taxonomy":"article_category","embeddable":true,"href":"https:\/\/www.chromacon.com\/en\/wp-json\/wp\/v2\/article_category?post=17977"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}